Integrated Science Assessment for Lead Appendix 3: Nervous System Effects


_	EPA/600/R-23/375
£% United States

Environmental Protection	January 2024

m m Agency	www.epa.gov/isa

Integrated Science
Assessment for Lead

Appendix 3: Nervous System Effects

January 2024

Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Enviromnental Protection Agency

-------
DISCLAIMER

This document has been reviewed in accordance with the U.S. Environmental Protection Agency
policy and approved for publication. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

3-ii

-------
DOCUMENT GUIDE

This Document Guide is intended to orient readers to the organization of the Lead (Pb) Integrated
Science Assessment (ISA) in its entirety and to the sub-section of the ISA at hand (indicated in bold). The
ISA consists of the Front Matter (list of authors, contributors, reviewers, and acronyms), Executive
Summary, Integrated Synthesis, and 12 appendices, which can all be found at https://assessments.epa.gov/
isa/ document/&deid=3 59536.

Front Matter

Executive Summary

Integrated Synthesis

Appendix 1. Lead Source to Concentration

Appendix 2. Exposure, Toxicokinetics, and Biomarkers

Appendix 3. Nervous System Effects

Appendix 4. Cardiovascular Effects

Appendix 5. Renal Effects

Appendix 6. Immune System Effects

Appendix 7. Hematological Effects

Appendix 8. Reproductive and Developmental Effects

Appendix 9. Effects on Other Organ Systems and Mortality

Appendix 10. Cancer

Appendix 11. Effects of Lead in Terrestrial and Aquatic Ecosystems
Appendix 12. Process for Developing the Pb Integrated Science Assessment

3-iii

-------
CONTENTS

DOCUMENT GUIDE	3-iii

LIST OF TABLES 	3-v

LIST OF FIGURES 	3-vii

ACRONYMS AND ABBREVIATIONS	3-viii

APPENDIX 3 NERVOUS SYSTEM EFFECTS	3-1

3.1	Introduction	 3-2

3.2	Scope	 3-2

3.3	Biological Plausibility	 3-4

3.4	Overt Nervous System Toxicity 	3-14

3.4.1	Epidemiologic Studies of Brain Structure and Function	3-14

3.4.2	Experimental Animal Studies of Brain Structure and Function	3-18

3.4.3	Integrated Summary of Overt Nervous System Toxicity 	3-25

3.5	Nervous System Effects Ascertained during Childhood, Adolescent, and Young Adult
Lifestages	 3-26

3.5.1	Cognitive Function in Children	3-26

3.5.2	Externalizing Behaviors: Attention, Impulsivity, and Hyperactivity in Children	3-87

3.5.3	Externalizing Behaviors: Conduct Disorders, Aggression, and Criminal Behavior in
Children, Adolescents, and Young Adults	3-114

3.5.4	Internalizing Behaviors: Anxiety and Depression in Children	3-126

3.5.5	Motor Function in Children	3-139

3.5.6	Sensory Organ Function in Children 	3-153

3.5.7	Social Cognition and Behavior in Children	3-162

3.6	Nervous System Effects Ascertained during Adult Lifestages	3-175

3.6.1	Cognitive Function in Adults	3-175

3.6.2	Psychopathological Effects in Adults	3-194

3.6.3	Sensory Organ Function in Adults	3-204

3.6.4	Neurodegenerative Diseases	3-212

3.7	Evidence Inventories - Data Tables to Summarize Study Details	3-228

3.8	References	3-502

3-iv

-------
LIST OF TABLES

Table 3-1	Statistics associated with the international pooled analysis of data from seven cohort

studies

3-29

Table 3-2	Summary of evidence Indicating a causal relationship between Pb exposure and cognitive

effects in children	3-84

Table 3-3

Table 3-4

Table 3-5

Table 3-6

Summary of evidence indicating a causal relationship of Pb exposure with attention,

impulsivity, and hyperactivity	3-110

Summary of evidence for a likely to be causal association between Pb exposure and
conduct disorders, aggression, and criminal behavior in children and adolescents	

Summary of evidence for a likely to be causal relationship between Pb exposure and
internalizing behaviors in children	

3-124

3-137

Summary of evidence indicating a likely to be causal relationship between Pb exposure

and motor function in children	3-151

Table 3-7	Evidence that is suggestive of, but not sufficient to infer, a causal relationship between Pb

exposure and sensory organ function in children	3-160

Table 3-8	Evidence that is suggestive of, but not sufficient to infer, a causal relationship between Pb

exposure and social cognition and behavior in children	3-173

Table 3-9	Summary of evidence for a causal relationship between Pb exposure and cognitive effects

in adults	3-192

Table 3-10

Table 3-11

Table 3-12

Table 3-1E
Table 3-1T
Table 3-2E
Table 3-3E
Table 3-4E

Table 3-4T
Table 3-5E
Table 3-6E

Summary of evidence for a likely to be causal relationship between Pb exposure and
psychopathological effects in adults	

Summary of the evidence that is suggestive of, but not sufficient to infer, a causal
relationship between sensory function in adults	

Epidemiologic studies of Pb exposure and overt nervous system toxicity
Animal toxicological studies of Pb exposure and brain function	

Epidemiologic studies of Pb exposure and full-scale intelligence quotient_
Epidemiologic studies of Pb exposure and infant development	

Animal toxicological studies of Pb exposure and cognitive function _

Epidemiologic studies of Pb exposure, academic performance, and achievement

3-203

3-210

Summary of evidence that is suggestive of, but not sufficient to infer, a causal relationship
between Pb exposure and neurodegenerative diseases in adults	3-226

3-228
3-233
3-258
3-276

Epidemiologic studies of Pb exposure and performance on neuropsychological tests of
cognitive function, i.e., learning, memory, and executive function	3-283

3-293
3-321

Epidemiologic studies of Pb exposure and cognitive effects: population or group mean

blood Pb levels >5 [jg/dL	3-329

3-v

-------
Table 3-7E Epidemiologic studies of Pb exposure and performance on neuropsychological tests of
attention, impulsivity, and hyperactivity, ADHD-related behaviors, and clinical ADHD in
children	3-347

Table 3-7T Animal toxicological studies of Pb exposure and externalizing and internalizing behaviors	3-365

Table 3-8E Epidemiologic studies of Pb exposure and performance on neuropsychological tests of
attention, impulsivity, and hyperactivity, attention deficit/hyperactivity disorder-related
behaviors, and clinical attention deficit/hyperactivity disorder in children; group or
population mean blood Pb level >5 [jg/dL, any study design	3-373

Table 3-9E Epidemiologic studies of Pb exposure and externalizing behaviors including conduct

disorders, aggression, and criminal behavior in children and adolescents	3-384

Table 3-10E Epidemiologic studies of Pb exposure and internalizing behaviors in children	3-398

Table 3-11E Epidemiologic studies of Pb exposure and motor function in children	3-408

Table 3-11T Animal toxicological studies of Pb exposure and motor function 	3-424

Table 3-12E Epidemiologic studies of Pb exposure and sensory organ function in children 	3-432

Table 3-13E Epidemiologic studies of Pb exposure, social cognition, and behavior in children	3-441

Table 3-14E Epidemiologic studies of exposure to Pb and cognitive function in adults	3-451

Table 3-15E Epidemiologic studies of Pb exposure and psychopathological effects in adults	3-461

Table 3-16E Epidemiologic studies of Pb exposure and sensory organ function in adults	3-468

Table 3-16T Animal toxicological studies of Pb exposure and sensory organ function	3-476

Table 3-17E Epidemiologic studies of exposure to Pb and neurodegenerative disease in adults	3-478

Table 3-17T Animal toxicological studies of Pb exposure and neurodegeneration	3-498

3-vi

-------
LIST OF FIGURES

Figure 3-1	Potential biological pathways for nervous system effects following developmental

exposure to Pb.	 3-5

Figure 3-2	Potential biological pathways for nervous system effects following postweaning exposure

to Pb.	 3-6

Figure 3-3	The relationship between blood Pb level at age 11 and brain outcomes in adulthood. 	3-16

Figure 3-4	Associations between blood Pb levels and full-scale intelligence quotient in children. 	3-32

Figure 3-5	Associations between biomarkers of Pb exposure and Bayley Score of Infant

Development Mental Development Index. 	3-39

Figure 3-6 Association of blood Pb level with reading and math scores among North Carolina school
children (average across all grades). Left panel displays impact of blood Pb level on math
test score. Right panel displays impact of blood Pb level on reading test score.	3-61

Figure 3-7	Relationship between concurrent blood Pb level and intelligence quotient among Italian

adolescents using a cubic spline fit.	3-63

Figure 3-8	Relationship between log-transformed blood Pb level and intelligence quotient using an

ordinary least squares fit. 	3-64

Figure 3-9	Two distributions of intelligence test scores demonstrating the consequence in a small

shift in the mean score.	3-71

Figure 3-10 Scatter plots and regression lines of blood Pb level and 18-month Mental Developmental

Index among children in manganese (A) quintiles 1-4 and (B) quintile 5.	3-75

Figure 3-11 Mean ± standard deviation behavior performance in the Go/No-Go task according to

quartiles of exposure for (A and B) cord blood Pb and (C) childhood blood Pb level at age
11 years.	 3-93

Figure 3-12 Associations of monthly airborne Pb exposure levels from birth to age 12 with scores for

anxiety and depression behaviors on the Behavior Assessment System for Children. 	3-129

Figure 3-13 Associations between biomarkers of Pb exposure and Bayley Score of Infant

Development Psychomotor Developmental Index.	3-141

Figure 3-14 Differences in mean difference tooth Pb levels for autism spectrum disorder in discordant
twin pairs versus (A) non-autism spectrum disorder twin pairs or (B) autism spectrum
disorder concordant twin pairs.	3-163

Figure 3-15 Hazard rate ratios for Alzheimer's disease mortality by blood Pb level including the lower

95% confidence interval.	3-222

3-vii

-------
ACRONYMS AND ABBREVIATIONS

AA	atomic absorption	BPAQ

AAS	atomic absorption spectrometry	BrainAGE

Ap	amyloid beta	BRIEF

ABR	auditory brainstem response

AD	Alzheimer's disease	BRIEF-A

ADD	attention deficit disorder

BRIEF-P

ADHD	attention deficit/hyperactivity disorder

ADHD-RS	ADHD rating scale

ADOS	Autism Diagnostic Observation	BRS

Schedule	ggj

ADRA2A	alpha-2A-adrenergic receptor	BSID

ALAD	aminolevulinic acid dehydratase

ALS	Amyotrophic Lateral Sclerosis	BSID-IIS

ALSPAC	Avon Longitudinal Study of Parents

and Children	BT20+

AOR	adjusted odds ratio	Ca2+

APP	amyloid precursor protein	CANTAB

AQCD	Air Quality Criteria Document

As	arsenic	CAR

avg	average	CARES

ASD	autism spectrum disorder

CARS

ASQ:I	Ages and Stages Questionnaire

Inventory	CAT

ASSQ	Autism Spectrum Screening	CBCL

Questionnaire	CBLI

ATP	adenosine triphosphate	CCAAPS

BAARS	Barkley Adult ADHD-IV Rating Scale

BACE1	beta-secretase 1	CCEI

BAEP	brainstem auditory evoked potential	Cd

BASC	Behavior Assessment System for	CDIIT

Children

BASC-2	Behavior Assessment System for	CDK5

Children, second revision	Ce

BBB	blood-brain barrier	CEM

BDI	Beck Depression Inventory	CERAD

BDNF	brain-derived neurotrophic factor

BKMR	Bayesian kernel machine regression	CHECK

BKT	Binet Kamat T est

BLL	blood lead level	CHEER

BMD	benchmark dose	^TT, r„

CHMS

BMDL	benchmark dose lower 95% confidence

limit	U

BMI	body mass index

BMS	Baltimore Memory Study

CKJJ

BNT	Boston Naming Test

CKiD

BPA	bisphenol A

Buss-Perry Aggression Questionnaire

Brain Age Gap Estimation

Behavior Rating Inventory of
Executive Functions

Behavior Rating Inventory of
Executive Functions for Adults
Behavior Rating Inventory of
Executive Functions for Preschool
Children

behavioral rating scale
Behavioral Symptoms Index
Bay ley Scales of Infant and Toddler
Development

Bay ley Scales of Infant and Toddler
Development - Spanish Version

Birth to Twenty Plus

calcium ion(s)

Cambridge Neuropsychological Test
Automated Battery

Cortisol awakening response
Communities Actively Researching
Exposure Study

Childhood Autism Rating Scale
catalase

Child Behavior Check List

cumulative blood lead index

Cincinnati Childhood Allergy and Air
Pollution Study

Crown-Crisp Experiential Index
cadmium

Comprehensive Developmental
Inventory for Infants and Toddlers
cyclin-dependent kinase 5
cesium

Coarsened Exact Matching
Consortium to Establish a Registry for
Alzheimer's Disease

Children's Health and Environmental
Chemicals in Korea

Children's Health and Environmental
Research

Child Health Monitoring System

confidence interval

Composite International Diagnostic
Interview

chronic kidney disease

Chronic Kidney Disease in Children

Cincinnati Lead Study

3-viii

-------
CNS	central nervous system

Co	cobalt

C-P	central-to-peripheral

cpd	cycles per degree

CPR	Conditioned Position Responding

CPRS	Conners' Parent Rating Scale

CPRS-R	Conners' Parent Rating Scale-Revised

CPT	Continuous Performance Test

C-R	concentration-response

CR	chromium

CREB	cyclic adenosine 3',4'-monophosphate
response element binding protein

CRISYS-R	Crisis in Family Systems-Revised

CRP	C-reactive protein

CRS	Conners' Rating Scale

CRS-R	Conners' Rating Scale-Revised

CRT	Combined Raven's Test

CSF	cerebrospinal fluid

C-TRF	Caregiver-T eacher Report F orm

CTRS	Conners'Teacher Rating Scale

CTRS-R	Conners'Teacher Rating Scale-
Revised

C-V R2	cross validated R-square

CVA	cerebrovascular accident

CVD	cardiovascular disease

CVLT	California Verbal Learning Test

CVLT-C	California Verbal Learning Test-
Children's Version

d	day(s)

DAT1	dopamine transporter

DBD	Disruptive Behavior Disorder

DDE	dichlorodiphenyldichloroethylene

DI	deionized

DISCI	Disrupted-in-Schizophrenia-1

DMTS	Delayed Matching-to-Sample

DQ	development quotient

DRD2	Dopamine Receptor D2

DNAm	DNA methylation

DSC	Digit Symbol Coding

DSM	Diagnostic and Statistical Manual of
Mental Disorders

DSST	Digit Symbol Substitution Test

DTI	Diffusion Tensor Imaging

ECAT	elemental carbon attributable to traffic

ECDI	Early Child Development Inventory

EE	effect estimate

EEG	electroencephalogram

ELEMENT	Early Life Exposure in Mexico to
Environmental Toxicants

EMOCI	emotional regulation

EOG	end of grade

EPM	elevated plus maze

EPN	early postnatal

EPSC	excitatory postsynaptic currents

ERG	electroretinography

ERP	event-related potential

ETS	environmental tobacco smoke

F	female

F#	filial generation

FA	fractional anisotropy

FBB-ADHS	Fremdbeurteilungsbogen fur

Aufmerksamkeitsdefizit/Hyperaktivitat
storungen

Fe	iron

FFQ	Food Frequency Questionnaire

FI	fixed interval

FLEHS	Flemish Environment and Health Study

FR	fixed ratio

FSIQ	full-scale intelligence quotient

FST	forced swim test

GABA	gamma-aminobutyric acid

GCNT1	glucosaminyl (N-acetyl) transferase 1

GD	gestational day

GDS	Gesell Developmental Schedules

GFAAS	graphite furnace atomic absorption

spectrometry

GMR	geometric mean ratio

GRIN	glutamate ionotropic receptor N-methyl
D-aspartate-type subunit

GSH	glutathione

GSI	Global Severity Index

GST	glutathione S-transferase

HCB	hexachlorobenzene

HDL	high-density lipoprotein

HFE	hemochromatosis gene

Hg	mercury

Hgb	hemoglobin

HHANES	Hispanic Health and Nutrition

Examination Survey

HI	hyperactivity and impulsivity

HNES	Home Nurture Environment Scale

HNRS	Heinz Nixdorf Recall Study

HOME	Health Outcomes and Measures of the
Environment

HPA	hypothalamic pituitary adrenal

hr	hour(s)

HR	hazard ratio

HR-ICP-MS	high resolution inductively coupled
plasma mass spectrometry

3-ix

-------
HRR	hazard rate ratio

HRT	hormone replacement therapy

ICD	International Classification of Diseases

ICP-DRC-MS dynamic reaction cell for inductively

coupled plasma mass spectrometry
ICP-MS	inductively coupled plasma mass

spectrometry

ICP-OES	inductively coupled plasma optical

emission spectroscopy

ICP-SFMS	inductively coupled plasma sector field

mass spectrometry

INMA	INfancia y Medio Ambiente

IQ	intelligence quotient

IQR	interquartile range

ISA	Integrated Science Assessment

ISAT	Illinois Standard Achievement Test

K6	Kessler Psychological Distress Scale

K-ABC	Kaufman Assessment Battery for

Children

K-ARS	Korean ADHD Rating Scale

K-CBCL	Korean Child Behavior Check List

KEDI	Korean Educational Development

Institute

KiTAP	Test of Attentional Performance for

Children

KNHANES	Korea National Health and Nutrition

Examination Survey

K-SADS	Kiddie Schedule for Affective

Disorders and Schizophrenia
K-SADS-PL-K Kiddie Schedule for Affective

Disorders and Schizophrenia Present
and Lifetime - Korean Version
K-XRF	K-shell X-ray fluorescence

LASSO	least absolute shrinkage and selection

operator

In	natural log

LOD	limit of detection

LTP	long-term potentiation

LURF	Land Use Random Forest

M	male

Mat	maternal

MAT	Metropolitan Achievement Test

MCU	mitochondrial Ca2+ uniporter

MDAT	Malawi Development Assessment Tool

MDI	Mental Development Index

mDISCl	mouse Disrupted-in-Schizophrenia-1

ME	maternal exposure

MEAP	Michigan Educational Assessment
Program

MeHg	methyl mercury

MHI-5	Mental Health Index 5-item

MIREC	Maternal-Infant Research on

Environmental Chemicals

MMSE	Mini Mental State Examination

Mn	manganese

mo	month(s)

MOCEH	Mothers' and Children's

Environmental Health
MRI	magnetic resonance imaging

MrOS	Osteoporotic Fractures in Men Study

MRS	magnetic resonance spectroscopy

MSCA	McCarthy Scales of Children's

Abilities

NaAc	sodium acetate

NAS	Normative Aging Study

NBAS	Neonatal Behavioral Assessment

Scales

NBNA	Neonatal Behavioral Neurological

Assessment

NCDS	Nunavik Child Development Study

NEI	National Emissions Inventory

NHANES	National Health and Nutrition

Examination Survey

NHBCS	New Hampshire Birth Cohort Study

NHS	Nurses' Health Study

NMDAR	N-methyl-D-aspartate receptor

NPR	Norwegian Patient Registry

NR	not reported

NS	no stress

OD/CD	oppositional defiant and conduct
disorder

OFT	open-field test

OLS	ordinary least squares

OR	odds ratio

ORIEN	orientation/engagement

OTB	operant test battery

Pb	lead

PbO	lead oxide

PC	primary caregiver

PCBs	polychlorinated biphenyls

PCNA	proliferating cell nuclear antigen

PD	Parkinson's disease

PDI	Psychomotor Developmental Index

PECOS	Population, Exposure, Comparison,

Outcome, and Study Design
PEG	Parkinson's Environment and Genes

PERI	perinatal

PHDCN	Project on Human Development in

Chicago Neighborhoods
PIQ	Performance Intelligence Quotient

PIR	poverty-income ratio

3-x

-------
PM2.5	fine particulate matter

PND	postnatal day

PPI	Psychopathic Personality Inventory

PR	prevalence ratio

PROGRESS Programming Research in Obesity,
Growth, Environment and Social
Stressors

PRP	post-reinforcement pause

PTA	pure-tone average

pts	points

p-tau	phosphorylated tau

PTSD	post-traumatic stress disorder

PW	postweaning

Q	quartile

RNS	reactive nitrogen species

RO DI	reverse osmosis deionized

ROS	reactive oxygen species

RR	relative risk

RSEI	Risk Screening Environmental

Indicators

SCN	suprachiasmatic nucleus

SCWT	Stroop Color-Word Test

SD	standard deviation

SDQ	Strengths and Difficulties

Questionnaire

Se	selenium

SE	standard error

SES	socioeconomic status

SGA	small for gestational age

SGPD	System Genomics of Parkinson's

Disease

SMBCS	Sheyang Mini Birth Cohort Study

SMS	Social Maturity Scale

SOD	superoxide dismutase

Sp	specificity protein

SPHERL	Study for Promotion of Health in

Recycling Lead

SPM	Standard Progressive Matrix

SQ	social quotient

SRP	self-report of personality

SRS	Social Responsiveness Scale

SWAN	Strengths and Weaknesses of ADHD

Symptoms and Normal Behavior Scale
T#	trimester #

TBD	to be determined

TBPS	Taiwan Birth Panel Study

TEACh	Test of Everyday Attention for

Children

TMT	Trail Making Test

TOKS	tin-ore kilns and smelters

TRD	Temporal-Response Differentiation

TRF	Teacher Report Form

TSCD	Tohoku Study of Child Development

TST	Tail Suspension Test

TUNEL	terminal deoxynucleotidyl transferase

dUTP nick end labeling
U.S. EPA	United States Environmental Protection

Agency

USV	ultrasonic vocalizations

VA	visual acuity

VDR	vitamin D receptor

VEP	visual evoked potential

VIF	variance inflation factor

VIQ	Verbal Intelligence Quotient

VMI	visual-motor integration

WAIS	Weschler Adult Intelligence Scale

WASI	Wechsler Abbreviated Scale of

Intelligence

wk	week(s)

WHO	World Health Organization

WIAT	Wechsler Individual Achievement Test

WISC	Wechsler Intelligence Scale for

Children

WJTA	Woodcock-Johnson Test of

Achievement
WMC	working memory capacity

WMH	white matter hyperintensities

WMS	Weschler Memory Scale

WPPSI	Wechsler Preschool and Primary Scale

of Intelligence

WRAML	Wide Range Assessment of Memory

and Learning

WRAT	Wide Range Achievement Test

XRF	X-ray fluorescence

yr	year(s)

YSR	youth self-report

Zn	zinc

3-xi

-------
APPENDIX 3 NERVOUS SYSTEM EFFECTS

Summary of Causality Determinations for Pb Exposure and Nervous System Effects

This appendix characterizes the scientific evidence that supports causality determinations for
lead (Pb) exposure and nervous system effects. The types of studies evaluated within this appendix are
consistent with the overall scope of the ISA as detailed in the Process Appendix (see Section 12.4). In
assessing the overall evidence, the strengths and limitations of individual studies were evaluated based
on scientific considerations detailed in Table 12-5 of the Process Appendix (Section 12.6.1). More
details on the causal framework used to reach these conclusions are included in the Preamble to the ISA
(U.S. EPA. 2015). The evidence presented throughout this appendix supports the follow ing causalitv
conclusions:

Outcome Group

Causality Determination

Nervous System Effects Ascertained during Childhood, Adolescent, and Young Adult Lifestages

Cognitive Effects

Causal

Attention, Impulsivity and Hyperactivity

Causal

Conduct Disorders, Aggression, and Criminal
Behavior

Likely to be causal

Internalizing Behaviors

Likely to be causal

Motor Function

Likely to be causal

Sensory Function

Suggestive of, but not sufficient to infer, a causal
relationship

Social Cognition and Behavior

Suggestive of, but not sufficient to infer, a causal
relationship

Nervous System Effects Ascertained during Adult Lifestages

Cognitive Effects

Causal

Psychopathological Effects

Likely to be causal

Sensory Function

Suggestive of, but not sufficient to infer, a causal
relationship

Neurodegenerative Disease

Suggestive of, but not sufficient to infer, a causal
relationship

The Executive Summary, Integrated Synthesis, and all other appendices of this Pb ISA can be found at
https://assessments.epa.aov/isa/document/&deid=359536



3-1

-------
3.1 Introduction

While Pb affects nearly every organ system, the nervous system appears to be one of the most
sensitive targets. The sections that follow provide an evaluation of the most policy-relevant scientific
evidence relating to the effects of lead (Pb) exposure on the nervous system. To maximize transparency
regarding the studies included in the appendix, the scope is defined in Section 3.2. Section 3.3, Biological
Plausibility, provides an overview of the biological pathways that potentially underlie the nervous system
effects discussed in subsequent sections of the appendix. Section 3.4 summarizes overt nervous system
toxicity, including changes in brain structure and function. There is no causality determination in this
section; rather, data presented in the section may be referenced in the outcome-specific "Summary and
Causality Determination" discussions in later sections if they provide support for the conclusions.

Sections 3.5 and 3.6 describe the epidemiologic and experimental animal evidence that pertains to
specific endpoints or outcome groupings, which are organized by the lifestage at which they are
ascertained (i.e., childhood, adolescence, and young adult [Section 3.5] and adult [Section 3.6] lifestages).

The strongest and most policy-relevant evidence within each section is discussed first. Within
Section 3.5, which focuses on exposures and outcomes ascertained during childhood lifestages, including
adolescence and early adulthood, the strongest evidence that is best substantiated at the lowest exposure
levels relates to Cognitive Effects (Section 3.5.1) and Attention, Impulsivity, and Hyperactivity (Section
3.5.2) in children. Conduct Disorders are discussed in Section 3.5.3, followed by Anxiety and Depression
(Section 3.5.4), Motor Function (Section 3.5.5), Sensory Organ Function (Section 3.5.6), and Social
Cognition and Behavior (Section 3.5.7). The next section (Section 3.6) includes endpoints that are
ascertained during adult lifestages. The section begins with an assessment of the evidence pertaining to
Cognitive Effects in Adults (Section 3.6.1) followed by sections on Anxiety, Depression, and
Psychopathological Effects (Section 3.6.2), Sensory Function (Section 3.6.3), and Neurodegenerative
Diseases (Section 3.6.4). Within each section, the collective body of evidence is integrated within and
across scientific disciplines, and issues relevant for interpreting the scientific evidence as well as the
rationale for the causality determination are outlined for relevant endpoints or outcome groupings.

3.2 Scope

The scope of this appendix is defined by Population, Exposure, Comparison, Outcome, and Study
Design (PECOS) statements. The PECOS statements define the objectives of the review and establish
study inclusion criteria, thereby facilitating identification of the most relevant literature to inform the

3-2

-------
Lead Integrated Science Assessment (Pb ISA).1 In order to identify the most relevant literature, the body
of evidence from the 2013 Pb ISA was considered in the development of the PECOS statements for this
appendix. Specifically, well-established areas of research; gaps in the literature; and inherent uncertainties
in specific populations, exposure metrics, comparison groups, and study designs identified in the 2013 Pb
ISA inform the scope of this appendix. The 2013 Pb ISA used different inclusion criteria than the current
ISA, and the studies referenced therein often do not meet the current PECOS criteria (e.g., due to higher
or unreported biomarker levels). Studies that were included in the 2013 Pb ISA, including many that do
not meet the current PECOS criteria, are discussed in this appendix to establish the state of the evidence
prior to this assessment. Except for supporting evidence used to demonstrate the biological plausibility of
Pb-associated nervous system effects, recent studies evaluated and subsequently discussed within this
appendix were only included if they satisfied all components of the following discipline-specific PECOS
statements:

Epidemiologic Studies:

Population: Any human population, including specific populations or lifestages that might be at

increased risk of a health effect;

Exposure: Exposure to Pb2 as indicated by biological measurements of Pb in the body, with a
specific focus on Pb in blood, bone, and teeth; validated environmental indicators of Pb
exposure,3 or intervention groups in randomized trials and quasi-experimental studies;
Comparison: Populations, population subgroups, or individuals with relatively higher versus
lower levels of the exposure metric (e.g., per unit or log unit increase in the exposure metric,
or categorical comparisons between different exposure metric quantiles);

Outcome: Nervous system effects including but not limited to cognitive function (e.g.,
intelligence quotient [IQ] decrement), externalizing and internalizing behaviors,
psychopathological effects, sensory organ function, motor function, and neurodegenerative
diseases; and

Study Design: Epidemiologic studies consisting of longitudinal and retrospective cohort studies,
case-control studies, cross-sectional studies with appropriate timing of exposure for the health

'The following types of publications are generally considered to fall outside the scope and are not included in the
ISA: review articfes (which typically present summaries or interpretations of existing studies rather than bringing

forward new information in the form of original research or new analyses), Pb poisoning studies or clinical reports

(e.g., involving accidental exposures to very high amounts of Pb described in clinical reports that may be extremely

unlikely to be experienced under ambient air exposure conditions), and risk or benefits analyses (e.g., that apply

concentration-response functions or effect estimates to exposure estimates for differing cases).

2Recent studies of occupational exposure to Pb were considered insofar as they addressed a topic area that was of
particular relevance to the National Ambient Air Quality Standards review (e.g., longitudinal studies designed to

examine recent versus historical Pb exposure).

3Studies that estimate Pb exposure by measuring Pb concentrations in particulate matter with a nominal mean
aerodynamic diameter less man or equal to 10 |im3 (PMio) and particulate matter with a nominal mean aerodynamic

diameter less than or equal to 2.5 (im3 (PM2.5) ambient air samples are only considered for inclusion if they also

include a relevant biomarker of exposure. Given that size distribution data for Pb-PM are fairly limited, it is

difficult to assess the representativeness of these concentrations to population exposure (Section 2.5.3 (U.S. EPA.

2013)). Moreover, data illustrating the relationships of Pb-PMio and Pb-PNfc.s with blood Pb levels (BLLs) are

lacking.

3-3

-------
endpoint of interest, randomized trials and quasi-experimental studies examining
interventions to reduce exposures.

Experimental Studies:

Population: Laboratory nonhuman mammalian animal species (e.g., mouse, rat, guinea pig,
minipig, rabbit, cat, dog) of any lifestage (including preconception, in utero, lactation,
peripubertal, and adult stages);

Exposure: Oral, inhalation, or intravenous routes administered to a whole animal (in vivo) that
results in a BLL of 30 (ig/dL or below;4,5

Comparators: A concurrent control group exposed to vehicle-only treatment or untreated
control;

Outcome: Nervous System effects; and

Study Design: Controlled exposure studies of animals in vivo.

3.3 Biological Plausibility

This section describes biological pathways that potentially underlie nervous system effects
resulting from exposure to Pb. Timing of exposure is important for the health effects for Pb. Exposures
during development can lead to improper formation and maturation of the nervous system and exposures
to the mature nervous system can lead to neurodegeneration. Figure 3-1 and Figure 3-2 graphically depict
these proposed pathways for health effects resulting from developmental exposure to Pb and later life
exposures, respectively. Proposed pathways are presented as a continuum of responses, connected by
arrows, which may ultimately lead to the apical nervous system health effects associated with exposures
to Pb at concentrations observed in epidemiologic studies. This discussion of "how" exposure to Pb may
lead to effects on the nervous system contributes to an understanding of the biological plausibility of
epidemiologic results evaluated throughout this appendix. Most of the studies cited in this subsection are
discussed in greater detail elsewhere in this appendix. The biological plausibility for Pb-induced effects
on the nervous system is supported by evidence from the 2013 Pb ISA and by recent evidence. Note that
the structure of the biological plausibility sections and the role of biological plausibility in contributing to
the weight-of-evidence analysis used in the current ISA are discussed in Section IS.7.2.

4Pb mixture studies are included if they employ an experimental arm that involves exposure to Pb alone.

5This level represents an order of magnitude above the upper end of the distribution of U.S. young children's BLL.
The 95th percentile of the 2011-2016 National Health and Nutrition Examination Survey distribution of BLL in
children (1-5 years; n = 2,321) is 2.66 (ig/dL (Eganet al.. 2021) and the proportion of individuals with BLL that
exceed this concentration varies depending on factors including (but not limited to) housing age, geographic region,
and a child's age, sex, and nutritional status.

3-4

-------
Altered
Transcriptional
Regulation

Altered
Neurodevelopmental
Processes

Impaired Blood
Brain & Blood
Cerebrospinal
Fluid Barrier
Permeability

Neurodevelopmental
Disorders

Impaired Behavior/
Cognitive/ Function

•i

Mood Disorders

Altered
Neurotransmitter
Signaling

Pb Exposure

Impaired sensory
function

Neurodegenerative
Diseases

Note: The boxes above represent the effects for which there is experimental or epidemiologic evidence related to Pb exposure, and the arrows indicate a proposed relationship
between those effects. Solid arrows denote evidence of essentiality as provided, for example, by an inhibitor of the pathway used in an experimental study involving Pb exposure.
Dotted arrows denote a possible relationship between effects. Shading around multiple boxes is used to denote a grouping of these effects. Arrows may connect individual boxes,
groupings of boxes, and individual boxes within groupings of boxes. Progression of effects is generally depicted from left to right and color coded (white, exposure; green, initial effect;
blue, intermediate effect; orange, effect at the population level or a key clinical effect). Here, population-level effects generally reflect results of epidemiologic studies. When there are
gaps in the evidence, there are complementary gaps in the figure and the accompanying text below. The structure of the biological plausibility sections and the role of biological
plausibility in contributing to the weight-of-evidence analysis used in the 2022 Pb ISA are discussed in Section IS.7.2. Source: (Shadbeqian et al.. 2019).

Figure 3-1 Potential biological pathways for nervous system effects following developmental exposure to
Pb.

3-5

-------
Altered Neurotransmitter
Signaling

Altered Calcium Signaling

Pb Exposure
A

Neurodegenerative
Diseases

Impaired
Cognitive/Behavioral
Function

Mood Disorders

Impaired Sensory
Function

Impaired Blood
Brain & Blood
Cerebrospinal
Fluid Barrier
Permeability

Figure 3-2

Potential biological pathways for nervous system effects following postweaning exposure to Pb.

3-6

-------
Plausible pathways connecting Pb exposure to apical events resulting from developmental and
later life exposures to Pb are proposed in Figure 3-1 and Figure 3-2, respectively. The proposed pathways
supported by the strongest evidence include the direct actions of Pb on cellular protein function and
subsequent initiation of oxidative stress-mediated pathways.

When Pb accumulates in the CNS, it can interfere with coordination of metal ions, which is
essential for the structure and function of many cellular proteins. Pb ions can compete with and displace
physiologically relevant ions (including Fe, Zn, Ca, and others) within proteins, leading to both altered
protein structure and function. As described in the 2013 Pb ISA, there is evidence that this ionic mimicry
and imbalance occurs in multiple organ systems, including the brain, and in proteins that perform diverse
functions including metabolism, inflammation, and oxidative stress responses. For example, Pb treatment
can disrupt Ca2+ signaling through interactions with calmodulin, voltage-gated Ca2+ channels, and various
adenosine triphosphate (ATP)ases (U.S. EPA, 2013). There is also evidence that Pb can replace Zn ions
in Zn finger-binding motifs, which are present in several transcription regulating proteins (U.S. EPA,
2013). Some research supports an interactive effect between Fe status and Pb exposure due to shared
metabolic and physiological profiles. Lifetime exposure to Pb in rats has been shown to affect Fe status
by increasing Fe content in the cortex and hippocampus of adult and aged animals and altering the
expression of divalent metal transporters (such as divalent metal transporter 1 and ferroportin) in the brain
(Zhu et al., 2013), suggesting that Pb may interfere with Fe trafficking in the brain. Often the effect of Pb
can be reduced with exogenous supplementation of biologically relevant metals. Recent studies support
the protective role of supplementation of Ca2+ (Basha and Reddy, 2015; Gottipolu and Davuljigari, 2014),
Zn (Pedroso et al., 2017), Fe (Liu et al„ 2013c) or essential metal mixtures (Basha et al„ 2014) on
neurologic alterations from Pb. These data support the hypothesis that direct competition of Pb with
metals can cause neurologic effects.

The brain has the highest energy demand and metabolism of any organ. Because of this fact,
energy homeostasis is critical and energy imbalance can increase the brain's susceptibility to stressors and
cell death. Pb-induced alterations in energy production and metabolism have been measured in several
ways. As discussed in the 2013 Pb ISA, Pb exposure can alter many aspects of energy metabolism, with
animal models demonstrating effects following both developmental and adult exposures to Pb (discussed
in Section 3.4.2.1). In recent studies of developmental Pb exposure, Pb-induced impairments in energy
production throughout the body have been measured as reductions in the activity of glucose and glycogen
metabolizing enzymes (Baranowska-Bosiacka et al., 2017) and alterations in the number and structure of
mitochondria (Ouvang et al., 2019; Gassowska et al., 2016a). Studies of Pb exposure in postweaning
animals showed similar reductions of metabolizing enzyme activity (Yun et al„ 2019; Verma et al„ 2005;
Yun and Hover, 2000; Sterling et al„ 1982), altered mitochondrial structure (Ouvang et al., 2019;
Dabrowska et al„ 2015; Sun et al., 2014), and ATPase activity (Thangarajan et al., 2018), suggesting
alteration of energy metabolism may occur regardless of the timing of Pb exposure.

3-7

-------
Energy production involves the formation of reactive intermediate species including reactive
oxygen (ROS) and nitrogen species (RNS). Disruptions in the mitochondria and energy metabolism result
in increased levels of ROS and RNS. While ROS are a part of normal cellular functioning, uncontrolled
production or reduced elimination of ROS by antioxidant systems can result in oxidative stress and
cellular damage (for example, DNA damage, oxidization of cellular components). Evidence reviewed in
the 2013 Pb ISA suggests that Pb may exert toxicity by disrupting cellular metabolism, increasing
ROS/RNS concentrations, and depleting antioxidant capacity (U.S. EPA, 2013). Numerous recent studies
have reported dysregulation of oxidative stress concurrent with altered mitochondrial function
(Ahmad et al., 2020; Karri et al., 2018; Maiti et al„ 2017; Kumar and Muralidhara, 2014; Baranowska-
Bosiacka et al., 2011), adding to the body of evidence. A study by Yang et al. (2014) found that Pb
downregulated the mitochondrial Ca2+ uniporter (MCU), resulting in increased ROS production in both
SH-SY5Y cells and in newborn rats. Yang and colleagues found that in vitro activation or overexpression
of MCU prevented Pb-induced oxidative stress whereas MCU inhibition or knockdown potentiated the
effects suggesting that alterations in mitochondrial function were responsible for Pb-induced ROS
production. In a similar manner, Pb has been shown to upregulate cyclophilin D, a protein that regulates
mitochondrial membrane potential, and in vitro knockdown or inhibition of cyclophilin D prevents the
Pb-induced loss of mitochondrial membrane potential (Ye et al., 2020; Ye et al., 2016a). Mitochondrial
function is also thought to be dependent on a dynamic balance between mitochondrial fission and fusion.
In a recent study, Pb reduced energy production and respiration while increasing mitochondrial ROS and
altering the expression of genes involved with mitochondrial dynamics both in vitro and in vivo
(Dabrowska et al„ 2015). In this study, knockdown of the transcription factor peroxisome proliferator-
activated receptor-y coactivator la, which protects the mitochondrial fusion and fission balance, increased
in vitro ROS production in response to Pb, further suggesting that altered mitochondrial activity results in
ROS production in response to Pb. Together, these data provide evidence that mitochondrial dysfunction
and altered energy metabolism is a source of oxidative stress.

Given their reactive nature, ROS and RNS can damage cellular proteins, lipids, and nucleic acids,
which can lead to functional and downstream signaling impairment. As discussed in the 2013 Pb ISA, Pb
exposure leads to elevated levels of ROS in neurons and other brain cells of exposed animals. Levels of
oxidative species have also been assessed indirectly by the presence of oxidative damage to DNA and
proteins as well as peroxidation of lipids. Studies assessed in the 2013 Pb ISA showed that Pb exposure
increased signs of oxidative damage in the brains of a variety of animal species (U.S. EPA, 2013; Wu et
al„ 2008). Since publication of the 2013 Pb ISA, more recent studies have demonstrated Pb-induced
increases in ROS production and oxidative damage in the brain both during development (Hossain et al„
2016; Lu et al„ 2013) and postweaning (Singh et al„ 2019; Liu et al., 2018a; Thangarajan et al„ 2018;
Singh et al., 2017; Kumar and Muralidhara, 2014; Flora et al., 2012). Proper regulation of oxidative stress
requires a balance between the presence of oxidative species (i.e., ROS and RNS) and levels of
antioxidant defense proteins (e.g., glutathione [GSH], catalase [CAT], and SOD). Along with increased
ROS/RNS production, depletion of antioxidant proteins or reductions in antioxidant enzyme activity
could contribute to an overall increase in oxidative stress. As discussed in the 2013 Pb ISA, animal

3-8

-------
studies and human panel studies have shown that BLL is associated with an increased ratio of oxidized to
unoxidized GSH (Mohammad et al., 2008; Diouf et al., 2006; Ercal et al., 1996; Sandhir and Gill, 1995).
More recent studies showed similar impairment of antioxidant defenses in animal models of Pb exposure
during developmental (Lu et al., 2013) and postweaning (Singh et al., 2019; Thangarajan et al., 2018;
Singh et al., 2017; Flora et al., 2012) Pb exposures. Furthermore, changes in antioxidant status and
oxidative stress can contribute to mitochondrial dysfunction, as described above. There is strong evidence
that Pb exposure across the lifespan disrupts multiple aspects of energy metabolism and oxidative stress
regulation.

Cellular damage caused by oxidative insults can trigger inflammation and vice versa; thus, it is
often difficult to disentangle which occurs first. Given the interrelated nature of these factors, they are
combined within the same gray box in the blood Pb diagrams (Figure 3-1 and Figure 3-2). Inflammation
is a hallmark of many neurological conditions and neurodegenerative diseases. Inflammation can be
triggered by the production of inflammatory mediators (e.g., cytokines) in response to cell or protein
damage. As discussed in the 2013 Pb ISA, Pb exposure results in signs of inflammation including
activation of inflammatory signaling pathways, inflammatory mediator production, and microglia cell
activation (U.S. EPA, 2013). Several studies have observed increased inflammatory mediator levels and
activation of inflammatory signaling pathways (for example, tumor necrosis factor-alpha) in the brains of
animals exposed to Pb during development (Chibowska et al„ 2020; Hossain et al„ 2016; Ashok et al„
2015) and postweaning (Yang et al„ 2019; Liu et al„ 2018a). Proinflammatory markers interact with, and
in some cases infiltrate, the BBB, initiating neuroinflammation, as indicated by altered gene expression,
increased apoptosis, lipid and protein oxidation, and microglial activation (Saleh et al., 2018; Shvachiy et
al„ 2018; Sobin et al., 2013). Similarly, histologic and immunohistochemical signs of neuroinflammation
in the dentate gyrus have been reported in rats exposed to Pb continuously from 7 days postconception to
28 weeks of age, which corresponded to behavioral changes (Shvachiy et al., 2018). In the same study, a
similar neuroinflammatory phenotype was observed in mice that were given an 8-week Pb abstinence
period between 12-week and 8-week Pb exposures (Shvachiy et al., 2018). In sum, recent evidence
supports the plausibility of inflammation as an intermediate event in the development of neurological
health effects regardless of the timing of Pb exposure.

While a robust immune response can protect the brain from certain insults, prolonged
neuroinflammation is associated with several neurological and neurodegenerative diseases. AD,
characterized by the accumulation of A|3 and p-tau, has been associated with increased markers of
neuroinflammation. While neurodegenerative diseases are associated with old age, studies of
developmental exposures to Pb have shown that early life exposures are associated with Alzheimer" s-like
pathology in adult animals. As discussed in the 2013 Pb ISA, Pb exposure in juvenile animals resulted in
the increased production of APP and higher levels of p-tau in offspring. Similarly, early life Pb exposure
of nonhuman primates led to Alzheimer"s-like pathology later in adulthood (Wu et al., 2008). Studies
published since the last ISA support and extend the findings that developmental exposures to Pb can lead
to increased levels of misfolded proteins (e.g., abnormal APP processing, A|3, tau protein) and

3-9

-------
Alzheimer"s-like pathologies (e.g., p-tau accumulation) (Ashok et al., 2015; Bihaqi and Zaw ia. 2013).
Evidence from exposures during development suggests that early life may represent a sensitive window
for insults associated with neurodegenerative disease (Liu et al.. 2014a). Some studies with exposure of
postweaning animals to Pb have shown increased inflammation associated with AD markers (Yang et al..
2019; Liu et al., 2018a; Zhang et al., 2012). In postweaning studies, treatment with molecules with anti-
inflammatory and antioxidative properties were able to prevent A|3 accumulation and reversed cognitive
and behavioral alterations in Pb-exposed mice (Liu et al., 2020; Yang et al., 2019; Liu et al., 2018a).
Because of this, there is a solid line connecting the box containing inflammation, oxidative stress, and
altered energy metabolism to the accumulation of A|3 in Figure 3-2. Evidence of effects of Pb on other
neurodegenerative diseases are more limited. A recent study showed that exposure of postweaning rats to
Pb resulted in increased accumulation of a-synuclein, a protein associated with PD, in the hippocampus
that correlated with impaired learning and memory (Zhang et al., 2012). Overall, new data support the
previous findings that Pb exposure can affect the development and progression of neurodegenerative
pathologies in developmentally and postweaning exposed animals.

Beyond Pb's ability to produce neuroinflammation and oxidative stress and increase expression
of disease-related proteins, excessive damage to cellular proteins or DNA can trigger cell death. While
cell death and neuronal population loss in adulthood contribute to brain pathology, the developing brain is
far more sensitive to disruption. Cell migration, differentiation, and pruning are all essential
neurodevelopmental processes that need to be carefully timed and orchestrated. Thus, increased or
aberrant cell loss results in improper nervous system development that could be responsible for the altered
mood, sensory, or cognitive functions observed in Pb-exposed children and animals. The 2013 Pb ISA
and Section 3.4.2.1 present several animal studies showing upregulation of apoptotic markers in various
regions of the brain following Pb treatment, at various lifestages, which was supported by similar findings
in in vitro experiments (U.S. EPA, 2013). Recent studies also reported activation of pro-apoptotic
pathways in response to developmental Pb exposure (Ebrahimzadeh-Bideskan et al., 2016; Hossain et al.,
2016; Su et al., 2016; Lu et al., 2013), supporting and extending the experiments reviewed in the 2013 Pb
ISA. Another study showed histologic changes in the brain concomitant with increased markers of protein
and lipid damage (Saleh et al„ 2019), suggesting a relationship between cell death and structural changes
in the brain with oxidative damage. Developmental Pb exposure caused dysregulated myelination in the
brains of rats, which could be rescued with cotreatment with antioxidants (Nam et al., 2020; Nam et al.,
2019a). Myelination is an essential step in nervous system development, as myelin sheaths facilitate quick
and efficient electrical transmission along nerve cells to preserve nervous system function and
connectivity. Several studies have also demonstrated that treatment with compounds with antioxidant
capacity reduced the apoptotic signaling (Nam et al., 2018b; Ebrahimzadeh-Bideskan et al., 2016).
Together, these data provide the justification for a solid line from the gray box containing oxidative stress
and inflammation to the box containing cell injury/death in Figure 3-2.

Widespread cell loss in the mature nervous system can also lead to functional and structural
changes that can contribute to behavioral and cognitive changes. Cell death is also a common element in

3-10

-------
many neurodegenerative diseases. The animal studies showing upregulation of apoptotic markers in
various regions of the brain following Pb treatment discussed in the 2013 Pb ISA are strengthened by
similar findings in several new studies (Amedu and Omotoso. 2020; Liu et al., 2020; Abubakar ct al..
2019; Singh et al., 2019; Yang et al., 2019; Liu et al., 2018a; Thangarajan et al., 2018; Maiti et al., 2017;
Singh et al., 2016; Flora et al„ 2012). In vitro exposure of neuronal cell lines to Pb resulted in reduced
cell viability and increased apoptosis (Ye et al., 2020; Liu et al., 2017; Neelima et al„ 2017; Meng et al„
2016; Su et al., 2016; Ye et al„ 2016b; Ahmed et al„ 2013). Additional discussion of apoptotic markers
and brain structural changes following Pb exposure are discussed in Section 3.4.2.1. Like the
developmental exposure studies, demyelination was observed in the spinal cord following postweaning
exposure to Pb (da Silva et al., 2020; Villa-Cedillo et al„ 2019). These data provide plausibility that adult
Pb exposures contribute to cognitive and behavioral changes. Some studies therapeutically targeted RNS
production in the mitochondria by treatment with fisetin, a polyphenolic compound with antioxidant
properties to ameliorate the activation of pro-apoptotic signaling (Yang et al„ 2019; Maiti et al., 2017).
Their results suggest a role for oxidative stress in triggering the apoptotic cascade. In vitro treatment with
the antioxidant genistein also protected against cell death (Su et al., 2016). Thangarajan et al. (2018)
showed that treatment with an anti-inflammatory and antioxidative compound, morin, was able to largely
restore proper brain architecture after Pb exposure. Together, these data provide the justification for a
solid line from the gray box containing oxidative stress and inflammation to the box containing cell
injury/death in Figure 3-2.

Inflammation and oxidative stress can also affect the integrity of the BBB, which provides a
selective barrier for entry from the circulation to the brain and spinal cord. As discussed in the 2013 Pb
ISA and 2006 Pb AQCD, Pb exposure in rodents was shown to increase permeability of the BBB and the
blood-CSF barrier. Interestingly, Pb alone does not have a large effect on BBB integrity but can prolong
BBB permeability in response to other stimuli (U.S. EPA, 2013, 2006). The effect of Pb on the BBB is
also selective in that the permeability of all solutes is not affected equally (U.S. EPA, 2013, 2006).
Disruption of the BBB could potentially promote increased Pb accumulation in the brain with prolonged
or repeated exposure. Two new studies assessed the integrity of the BBB following Pb exposure and
observed disruption of brain permeability with reduced levels of tight junction proteins and other
important capillary proteins (Wu et al., 2020a; Song et al., 2014). In adult rats, 8 weeks of Pb dosing
reduced expression of the tight junction proteins occludin and zonula occludens-1 at the BBB (Song et
al., 2014). Pb has been implicated in alteration of the CSF barrier in rats (Zheng et al., 1996). The CSF
can carry hormone signals important for brain development; thus, disruption of the cerebrospinal barrier
could affect proper hormone signaling for brain development. Indeed, Pb exposure was reported to
decrease transthyretin levels in the CSF, suggesting altered cerebrospinal barrier integrity (Zheng ct al..
1996). There is likely interplay between Pb effects on endocrine and nervous system development. In
conclusion, there is a potential for Pb to affect the BBB and blood spinal cord barrier, which could alter
Pb availability and uptake into the nervous system.

3-11

-------
While most data suggest that Pb acts through a mode of action involving oxidative stress and
inflammation, additional signaling pathways are also affected by Pb exposure. Pb exposure can alter ion
balance, which has particular importance with regard to the effect on Ca2+ signaling. Calcium signaling is
vital for many fundamental neurological processes including membrane excitability, neurotransmitter
release, synaptogenesis, transmission, and other processes. Of particular relevance for this review,
neurotransmitter signaling is intimately connected with Ca2+ signaling. As discussed in the 2013 Pb ISA,
developmental exposure to Pb interferes with the evoked release of neurotransmitters by inhibiting Ca2+
transport through voltage-gated ion channels (Cooper and Manalis, 1984; Suszkiw et al., 1984). Ca2+ is
also a ubiquitous second messenger, which can regulate many neuronal physiologic processes like gene
expression, membrane excitability, and dendrite development (Kawamoto et al., 2012). Interestingly, in
the absence of stimulation, Pb has some Ca2+ mimetic activity that increases baseline neurotransmitter
release (Cooper and Manalis, 1984; Suszkiw et al., 1984). In general, Pb exposure increased Ach levels,
increased dopaminergic signaling, and reduced NMDAR expression (U.S. EPA, 2013). Animal models
suggest that Pb exposure during development leads to inhibition of acetylcholinesterase (AchE), thereby
increasing the levels of Ach and causing lasting neurodevelopmental changes that persist into adulthood
(Basha and Reddy, 2015). The authors found that addition of Ca2+ restored cholinergic signaling (Basha
and Reddy, 2015). These data help to justify the solid line from ionic mimicry to altered neurochemical
signaling in Figure 3-1. Similar studies in animals postweaning have shown similar decreases in AchE
activity (Galal et al., 2019; Okesola et al., 2019; Thangarajan et al„ 2018; Andrade et al„ 2017; Ferlemi et
al„ 2014; Phyu and Tangpong, 2013). Recent literature also supports altered dopaminergic signaling
following postweaning Pb exposure (Sobolewski et al., 2020; Yousef et al„ 2019; Amos-Kroohs et al..
2016; Stansfield et al., 2015; Basha et al„ 2014; Weston et al., 2014; Cory-Slechta et al„ 2012; Graham et
al„ 2011). Ca2+ gradients are also responsible for generating action potentials. Alteration of intracellular
Ca2+ levels in neurons could cause deleterious effects on action potential generation and repolarization.
Recent evidence shows that hippocampal slices from 50-day old rats exposed to Pb both pre and
postnatally had enhanced pared pulse facilitation, suggesting Pb-induced dysregulation of Ca2+ signaling
(Zhang et al., 2015b). This data support findings from a limited number of human MRI and MRS studies
that provide evidence of physical and physiological changes in the brain corresponding to increased blood
Pb that were discussed in the 2013 Pb ISA and 2006 Pb AQCD. Together there is evidence that Pb can
alter neurotransmitter release and signal potentiation, which in turn could contribute to the changes in
brain activity seen in various behavioral and cognitive diseases.

Beyond the actions of Pb discussed thus far, there is growing evidence of the effect of changes to
the epigenome in mediating toxicity. Epigenetic changes refer to alterations in the mechanisms that
regulate gene expression without altering DNA sequence. Epigenetic programming is a fundamental
developmental process, and the complex relationships between the genome, epigenome, and environment
can shape the health of present and future generations. Epigenetic alterations are often measured as
changes in histone and DNA methylation patterns as well as the levels of the enzymes (e.g.,
methyltransferases, acetylases, deacetylases) responsible for regulating epigenetic modification in situ.
Transcriptional regulators like microRNAs and long noncoding RNAs are also considered epigenetic

3-12

-------
modifiers. As discussed in the 2013 Pb ISA, developmental Pb treatment in mice and monkeys decreased
the activity of some DNA methyltransferases. Therapeutic treatment with a methyl donor improved Pb-
induced decrements in LTP and Morris water maze performance (Cao et al.. 2008). Gestational and
postnatal exposure to Pb in rats increased histone acetylation in the hippocampus, which corresponded to
a hyperactivity phenotype (Luo et al.. 2014). The authors suggested this was due to upregulation of
histone acetyltransferases, including p300. However, these changes occurred at BLLs in excess of 50
(ig/dL; thus, the relevance of these findings to ambient exposure in humans is questionable. Other studies
have reported changes in the expression of DNA methyltransferases (Schneider et al.. 2013) and increased
hypermethylation, especially in the hippocampus of female mice (Sanchez-Martin et al.. 2015). The
window of exposure, prenatal stress, and sex can all play a role in determining the epigenetic
modifications (Sobolewski et al.. 2018). While differences in epigenetic modifications and the effects of
Pb on epigenetic enzymes have been reported and linked to behavioral effects in animals, there remains
little evidence to connect epigenetic changes to alterations in specific pathways that have the potential to
cause neurobehavior effects. As a result, the arrow for epigenetic changes in is represented as a dotted
line when connecting to the box for neurodevelopmental disorders. Future research may elucidate a role
for epigenetic modification in the etiology of neurological diseases. Most of the Pb literature on
epigenetic changes has focused on heritable epigenetic changes during development; however, the effect
of postweaning exposure to Pb on epigenetic mechanisms is not well known. Very few studies have
evaluated epigenetic changes throughout the lifetime or with later life exposures. Individual studies have
found changes in the expression of a long noncoding RNA (Nan et al.. 2016) and a methyltransferase
(Schneider et al.. 2012). which might suggest that epigenetic modification could be affected during adult
exposures; however, there are too few studies to draw reliable conclusions. The conclusion of epigenetic
modification resulting in neurological effects is only plausible for developmental exposures with the
present available data.

In summary, Pb exposure can result in a range of neurocognitive and behavioral health effects
through a myriad of complex biological pathways. The pathways described here provide biological
plausibility for associations between Pb exposure and nervous system effects in in children and adults.
The developmental timing, sex, and presence of other stressors or enrichments alongside of Pb exposure
can affect the resulting health effects. The identified pathways share many common features, including Pb
interactions with cellular proteins, competing with and displacing other biologically relevant cations,
increased oxidative stress, and inflammation, which can have widespread effects on brain structure and
function. There is also evidence for disruptions of Ca2+ signaling, which can result in altered
neurotransmitter signaling and contribute to the development of neurological health effects. Epigenetic
modifications resulting from Pb developmental exposure have been reported but are still an area of active
investigation. The role of these epigenetic changes in the progression of neurological health effects with
later Pb exposures is unclear. Together the proposed pathways provide biological plausibility for
epidemiologic evidence of neurological effects and were used to inform causality determinations
throughout this appendix.

3-13

-------
3.4

Overt Nervous System Toxicity

Overt nervous system toxicity refers to a diverse group of endpoints that inform brain structure
and function, including brain histopathology changes, brain weight, electrophysiology,
neuroinflammation, and neurotransmitter analyses. The collective body of epidemiologic and
experimental animal studies assessed in the 2013 Pb ISA demonstrated the effects of Pb exposure on an
array of nervous system outcomes. The evidence, including uncertainties, is summarized in Section 3.4.3.
Study details that supplement the information provided in the text are in the evidence inventories (Table
3-IE and Table 3-IT in Section 3.7). Previous Pb assessments reviewed epidemiologic studies that found
associations of Pb biomarkers with electrophysiologic or physical changes in the brains of adults assessed
by imaging technologies. Biological plausibility for the effects of Pb on overt nervous system toxicity
was provided by a small number of experimental animal study findings with dietary and lactational Pb
exposure, with some evidence at BLLs relevant to humans. Recent epidemiologic studies support and
extend the evidence pertaining to the association of lead exposure during childhood with brain structure
and function in adolescence or adulthood. A smaller set of cross-sectional studies also report associations
between childhood BLLs and overt nervous system outcomes. Multiple experimental animal studies
report changes in the brains of rats and mice following exposure to Pb. These include changes in
histology, neurotransmitter measures, brain weight, and electrophysiology measures, providing coherence
for the epidemiologic studies that show associations with decrements in cognition, neurodegeneration, or
increased behavioral problems. There is no causality determination for this section; rather, evidence in
this section may be referenced in the outcome-specific "Summary and Causality Determination"
discussions of Sections 3.5 and 3.6 if they provide biological plausibility or coherence for the
observations in the epidemiologic studies.

3.4.1 Epidemiologic Studies of Brain Structure and Function

Previous Pb assessments (U.S. EPA, 2013, 2006) reviewed a small body of epidemiologic
studies that found associations of Pb biomarkers with electrophysiologic and physical changes in the
brains of young adults as assessed by magnetic resonance imaging (MRI) or spectroscopy (MRS). The
implications of findings from most studies assessed in the 2006 Pb AQCD were limited by the small
sample sizes (n = 12 to 45) and inadequate consideration of potential confounding. However, analyses
cohort of adults (ages 20-23 years) reviewed in the 2013 Pb ISA included larger sample sizes and aimed
to characterize potentially important lifestages of Pb exposures (Yuan et al„ 2006), thus expanding the
evidence pertaining to potential links between physiologic brain changes and functional
neurodevelopmental effects. Overall, the small number of studies in a limited number of populations
assessed in the 2013 Pb ISA showed physical and physiologic changes in areas of the brain associated
with neurodevelopmental function, providing biological plausibility for the associations

3-14

-------
observed between Pb biomarker levels and cognitive function decrements and behavioral problems.

Several recent longitudinal studies add to the evidence characterizing the association of Pb
exposure during childhood with brain structure and function during adolescence or adulthood. A smaller
number of studies evaluated the cross-sectional association of childhood BLL with brain structure or
function in childhood. These studies are summarized below, and key information from the studies is
included in Section 3.7, Table 3-IE.

Reuben et al. (2020) conducted a study to examine the effect of childhood BLL, measured at age
11, on lower structural integrity of the brain at age 45. These investigators used data from the Dunedin
Study in New Zealand, which enrolled participants beginning in 1972 and 1973 and followed them
through April 2019. The mean early childhood BLL for the study participants was 10.99 (ig/dL. MRI was
used to assess multiple endpoints related to gray matter (cortical thickness, surface area, and hippocampal
volume), white matter (white matter hyperintensities, fractional anisotropy [FA]), and the gap between
chronological age and estimated brain age. In addition, cognitive function was estimated using the
Wechsler Adult Intelligence Scale (WAIS)-IV, self-reports, and informant reports (see Section 3.6.1). A
total of 564 of the original 1037 infants enrolled at birth were included in the analysis. Findings from the
study are depicted in Figure 3-3. In models adjusted for sex, maternal IQ, and socioeconomic status
(SES), associations were observed with cortical surface area, hippocampal volume, global FA, and the
gap between each study member's chronological age at imaging and their MRI-predicted age, but not
with all the MRI metrics assessed.

Two analyses of the Cincinnati Lead Study (CLS) have been conducted since the 2013 Pb ISA.
Confounders considered in these analyses included child characteristics, Home Observation for the
Measurement of Environment (HOME) score, maternal IQ, and SES (see Section 3.7, Table 3-IE for
study-specific confounders). Cecil (2011) examined the association of childhood BLL (childhood [3-28
months] average) with volumetric MRI, MRS, diffusion tensor imaging (DTI), and functional MRI
outcomes ascertained between ages 19 and 24 years old. This study found that childhood BLL was
associated with decreased gray matter volume in several regions (i.e., medial and superior frontal gyri,
inferior parietal lobule and cerebellar hemispheres). Higher childhood BLL was also associated with
lower metabolite concentrations in several brain regions (white matter, left basal ganglia, left cerebellar
hemisphere, and vermis). DTI and functional MRI findings also suggested injury and compensatory
activity in specific brain regions. Overall, structural, organizational, and functional changes in the brain
regions responsible for regulating behavior were indicated by this study. In another study of participants
enrolled in the CLS, Beckwith et al. (2021) examined the relationships between childhood BLL (at 78
months), structural brain volume, and adult criminality. BLLs were associated with MRI-derived
decreases in white and gray matter volumes in the frontal parietal and temporal lobes. Decreased gray
matter volume in brain regions responsible for cognition and emotional regulation was also associated

3-15

-------
with criminal arrests, potentially supporting associations observed between Pb exposure and conduct
disorders that are described in Section 3.5.3.

|~A | Cortical thickness (n = 508)

I

~ 2.50-

• •

• • ,

. 1 1 !•«!«•

S • I I • !!*••» '
j 111 j 11 j; j i; |; ;. • • •
!:':!' • *«' •

b = -0.004; 95% CI, -0.012 to 0.004; P = .39

	1	1	1

10	20	3(

Blood lead level at age 11 y, pg/dL

fi

T] Hippocampal volume (n = 508)

3ZZh

M:

IXLLLLi *!:. : .

•.I! !!jii!•

•	i •. •	• •

B

b = -0.10; 95% CI, -0.17 to -0.03; P = .006

10	20

Blood lead level at age 11 y, pg/dL

[F] Fractional anisotropy (n = 503)

10	20

Blood lead level at age 11 y, pg/dL

| B Cortical surface area (n = 508)

X

HI:

«.•••• • •
tj!!ii i|j i"

i I ! . I ! • I

0

b = -1.19; 95% CI, -2.35 to -0.02; P = .05

10	20

Blood lead level at age 11 y, pg/dL

[d] WMH volume (n = 502)

E 6-

i.

iili'il.i

i«s j; • .4—»

5 = 0.05; 95% CI,-0.02 to 0.13; P = . 17

10	20

Blood lead level at age 11 y, pg/dL

|T| BrainAGE index (n = 512)

	 . • 1 •

s#

^ i ' L!_i

•*

i

I



! I i i! * ¦

r?
;

n

urn

•• :







*





b

0.77; 95% CI, 0.02-1.51; P = .05

10	20

Blood lead level at age 11 y, pg/dL

BrainAGE = Brain Age Gap Estimation; CI = confidence interval; MRI = magnetic resonance imaging; SES = socioeconomic status.
WMH = white matter hyperintensities; yr = year(s).

Regression lines and their 95% CIs are plotted. The box plots show the distribution of BLLs and brain outcomes.

Beta coefficients shown are for an incremental increase of 5 |jg/dL in childhood BLL. Source: Reuben et al. (2020).

Figure 3-3 The relationship between blood Pb level at age 11 and brain
outcomes in adulthood.

3-16

-------
Lamoureux-Tremblav et al. (2021) examined the association between pre- and postnatal
(measured concurrently to the MRI) BLLs and MRI findings in adolescents (mean age 18.3 years)
enrolled in a longitudinal study of Inuit from Northern Quebec exposed to Pb, mercury (Hg), and
polychlorinated biphenyls (PCBs). Functional MRI data were collected during fear conditioning and
extinction tasks with the aim of understanding emotional dysregulation that could lead to anxiety
disorders. These authors found higher differential activation in the right dorsolateral prefrontal cortex in
association with higher postnatal BLL. Differential effects in the high Pb exposure group were observed
during the fear extinction phase and maintained discrimination between the safety and threatening signals
(CS+ > CS-). Activation of the dorsolateral prefrontal cortex has been associated with cognitive processes
for regulating the affective state. The mean cord blood Pb was 4.56 (ig/dL and the mean concurrent BLL
was 1.78 (ig/dL in this study. In an earlier study of this Inuit population, Ethier et al. (2012) measured
visual evoked potentials (VEPs) using electrodes on the scalp to provide a direct measure of brain
functions related to sensory function (i.e., visual contrast sensitivity and spatial vision) at age 5. The mean
cord BLL in this study was 4.6 (ig/dL. Amplitude and latency for standard VEP components were
measured (i.e., N75 [negative deflection at approximately 75 ms], P100 [positive deflection at
approximately 100 ms], and N150 [negative deflection at approximately 150 ms]) in this longitudinal
study. Multiple tests were conducted, and associations were reported with significance levels. Cord Pb
level was associated with a delay of the N150 component (e.g., |3 = 0.06 [95% CI: 0.01, 0.10]) and other
latency metrics, which may indicate a deficit in early visual processing in Pb-exposed children.
Confounders including child characteristics, maternal education, SES, drug and alcohol use, and other
metals were considered in the analyses of Inuit children (see Table 3-IE for study-specific confounders
considered). A subset of two separate studies was combined for this study, and participation rates based
on the original number of participants enrolled in the study were not reported.

In a cross-sectional analysis of children, Kim et al. (2018a) examined the interaction between
dopamine receptor D2 (DRD2) and BLL on the cortical thickness of 12 regions of the frontal lobe
ascertained via MRI. The D2 receptor is located in the prefrontal cortex of the brain and may contribute to
the pathology of attention deficit/hyperactivity disorder (ADHD). The authors relied on an age- and sex-
matched sample of children ages 6 to 17 years old with and without confirmed ADHD for the analysis.
This study found an interaction effect between a variant of DRD2 and BLL on reduced cortical thickness
of several regions in the frontal lobe in the ADHD group, but not in the healthy controls in regression
analyses adjusted for age, intracranial volume, and sex. A correlation between reduced cortical thickness
and poorer inattention score on the parent-reported ADHD rating scale was also reported, supporting a
link between Pb exposure, MRI findings, and functional decrements in attention. Results for Kim et al.
(2018a) are found in Section 3.7, Table 3-IE.

3-17

-------
3.4.1.1 Summary

Overall, multiple studies, including prospective studies following children through adolescence
and adulthood, found associations between BLL and physiologic changes in regions of the brain
responsible for cognition and behavior. The prospective studies considered important confounders
including the HOME score (CLS only), SES, drug and alcohol use, and maternal IQ. A smaller number of
studies indicated Pb-related changes in brain function or physiologic changes in children. Associations
between Pb exposure and a large number of metrics and brain regions were evaluated in the
epidemiologic studies, raising the likelihood of chance findings.

3.4.2 Experimental Animal Studies of Brain Structure and Function

Animal studies can offer insights into Pb-induced effects on brain structure and function through
investigations that cannot be conducted on human subjects. Experimental animal studies provide evidence
that Pb can cause alterations in brain development. The 2013 Pb ISA reviewed evidence that Pb exposure
produced increased levels of oxidative stress and inflammatory response markers (U.S. EPA, 2013).

These effects were observed in many regions of the brain and were associated with changes in neuronal
and glial cell morphology, neurotransmitter levels, and brain electrophysiology. Additional discussion of
the biological pathways that potentially underlie these nervous system effects are discussed in Section 3.3.

Recent studies (see Figure 3-3 and Table 3- IT) support the results summarized in the 2013 Pb
ISA, showing increases in inflammatory responses and markers of oxidative stress in various regions of
the brain. Most studies evaluated oral dosing of Pb via drinking water or gavage with exposure durations
ranging from 14 days to 701 days depending on the study, with studies in both adult and developing
animals. The specific dosing regimen varied by study, but the present review focuses on studies that
resulted in a measured BLL <30 (ig/dL. In studies with multiple time points, the magnitude or severity of
effects generally increased with exposure duration. In addition, studies with developmental Pb exposure
identified pregnancy and early development as sensitive windows for Pb toxicity. Several studies have
examined brain architecture using histological methods following Pb exposure. In general, the magnitude
and severity of the effects increased with longer exposure durations, and some studies found Pb-induced
effects could be ameliorated by co-exposures with antioxidants (such as vitamins and specific lipids) or
by environmental conditions (e.g., rearing condition or enrichment).

3.4.2.1 Histopathology

The nervous system is made up of the central and peripheral nervous systems, which include a
diversity of cell types broadly grouped into neurons and glial cells. Neurons are the functional electrically
excitable cells in the brain. Glial cells can be further divided into microglia and macroglia (astrocytes,
oligodendrocytes, ependymal cells, Schwann cells, satellite cells, radial glia, and enteric glia), which are

3-18

-------
neuronal support cells with diverse functions including innate immunity, phagocytosis, myelination,
synaptic regulation, neuronal activity, and blood-brain barrier (BBB) integrity (Rea. 2015). In the central
nervous system (CNS), astrocytes, oligodendrocytes, and microglia are the major types of glial cells.
Normal brain development relies on coordination of all cell types across time. Pathology methods can be
used to study alterations in brain cell morphology, function, and composition.

Short-term studies (<30 days) in adults evaluating Pb exposure and brain morphology after oral
exposures in rats have observed abnormalities in treated animals and their offspring, including decreased
numbers of neurons or synapses (Nam et al.. 2018a; Saleh et al.. 2018; Gassowska et al.. 2016a; Han et
al.. 2014; Rahman et al.. 2012b). disorganized cells and lack of characteristic layering (Saleh et al.. 2019;
Saleh et al.. 2018; Zhou et al.. 2018). increased vacuolization (Saleh et al.. 2019). effects on dendritic
spines (Xiao et al.. 2020; Saleh et al.. 2018; Wang et al.. 2016; Du et al.. 2015; Rahman et al.. 2012b).
increased numbers of apoptotic cells (Saleh et al.. 2019; Meng et al.. 2016). and increased expression of
various proteins and biochemical parameters related to oxidative stress (Saleh et al.. 2018; Singh et al..
2017; Zhu et al.. 2013) in multiple brain regions including the cerebellum, cerebral cortex, and
hippocampus. In aggregate, this evidence suggests that Pb has the potential to disrupt the integrity of
single neurons and populations, which may contribute to overt toxicity. Pb exposure has also been shown
to interfere with the homeostasis of other essential metal ions, such as iron (Fe), in the brain (Zhu et al..
2013) and to inhibit various enzymes involved in energy production or glucose uptake (Zhao et al.. 2021)
and metabolism (reviewed in (ATSDR. 2020)). For all these studies, BLLs were below 30 (ig/dL, and
many were below 20 (ig/dL (see Evidence Inventory Table 3- IT). Further discussion of these mechanisms
is provided in Section 3.3.

Studies of long-term Pb exposure (>30 days) have also assessed brain morphology in adult mice
and rats following oral dosing and observed neuronal damage including irregular shape, vacuolization,
and cell degeneration in the cerebellum, hippocampus, and cerebral cortex (Liu et al.. 2022c; Saleh et al..
2019; Singh et al.. 2019; Saleh et al.. 2018; Sun et al.. 2014). Other histopathological lesions, including
karyopyknosis (pre-apoptotic chromatin condensation of cell nuclei) (Nan et al.. 2016) and swollen and
distorted mitochondria (Ouvang et al.. 2019; Gassowska et al.. 2016a; Sun et al.. 2014) were observed in
other studies. Several biochemical parameters related to oxidative stress were assessed including
apoptosis using immunohistochemical methods like terminal deoxynucleotidyl transferase dUTP nick end
labeling (TUNEL staining) (Singh et al.. 2019; Baranowska-Bosiacka et al.. 2017; Meng et al.. 2016; Nan
et al.. 2016; Su et al.. 2016; Baranowska-Bosiacka et al.. 2013) as well as caspase-3 (Wang et al.. 2021a)
and cell replication using proliferating cell nuclear antigen (PCNA) (Singh et al.. 2019). In a 12-week rat
dietary study, the authors reported neuronal damage and cognitive deficits accompanied by decreased
levels of synaptic proteins as well as decreased levels of receptors and proteins related to synaptic
plasticity regulation (N-methyl D-aspartate receptor [NMDAR], cyclic adenosine 3",4"-monophosphate
response element binding protein [CREB], brain-derived neurotrophic factor [BDNF]) (Liu et al.. 2022c).
BLLs in all of these studies were below 30 (ig/dL and many were below 20 (ig/dL (see Evidence
Inventory, Table 3-IT).

3-19

-------
In the single available inhalation study, adult mice were dosed for 6 weeks via whole body
inhalation of Pb oxide (PbO) nanoparticles. Hippocampal damage was observed including shrunken and
damaged neurons, as well as increased Pb content in the brain (Dumkova et al.. 2017). The study reported
BLLs of 13.99 (ig/dL. However, exposure to PbO nanoparticles via inhalation did not impact markers of
cell proliferation (PCNA) or cellular apoptosis (TUNEL) in the hippocampus (Dumkova et al.. 2017).
Despite the limited evidence available for inhaled Pb, continuity of the effects (i.e., neuronal damage) has
been reported across different routes of exposure. Toxicology studies of Pb inhalation remain a data gap
that limits the evaluation of nervous system effects from inhaled Pb.

Brain morphology has also been assessed in developing animals with a variety of exposure
contexts including pre-mating, during gestation, and during lactation. Pb is transferred across the placenta
and through lactation (Silbergeld. 1991; Bhattacharvva. 1983). Some studies included cross fostering
experiments to assess unique sensitivities at specific developmental windows, with some evidence
indicating that the postnatal period is particularly sensitive to Pb neurotoxicity (Barkur and Bairv. 2016).
Given the altricial nature of rodents, the postnatal period is roughly analogous to the third trimester of
human brain development. Similar to results in adults, studies of developmental Pb exposures observed
alterations in brain morphology including reduced numbers of neurons in the forebrain, hippocampus,
hypothalamus, and amygdala (Long et al.. 2022; Vigucras-Villascnor et al.. 2021; Wang et al.. 2021a;
Nam et al.. 2018a; Shvachiv et al.. 2018; Xiao et al.. 2014). damaged neurons (Long et al.. 2022; Wang et
al.. 2021a; Zhu et al.. 2013). reduced numbers of glial cells (Dominguez et al.. 2019; Sobin et al.. 2013).
changes in synapses (Sadeghi et al.. 2021; Wang et al.. 2021b; Gassowska et al.. 2016a; Gassowska et al..
2016b; Zhang et al.. 2015b; Xiao et al.. 2014). reduced numbers of mitochondria (Zhang et al.. 2015b).
swollen and shrunken mitochondria (Ouvang et al.. 2019; Gassowska et al.. 2016a; Baranowska-Bosiacka
et al.. 2013). altered levels of glycoconjugates (constituents of synaptic and neural membranes) (Sadeghi
et al.. 2021). and chromatin abnormalities (Ouvang et al.. 2019; Baranowska-Bosiacka et al.. 2013). BLLs
in all of these studies were below 30 (ig/dL and many were below 20 (ig/dL. In a rat developmental study
in which animals were dosed throughout pregnancy and lactation (gestational day [GD] 1 to postnatal day
[PND] 21) with BLLs of 6.86 (ig/dL, the authors reported pathological changes in synapses, including
swelling of nerve endings, thickened synaptic cleft structure, and abnormalities in synaptic vesicle density
(Gassowska et al.. 2016a). These changes in synapse morphology were accompanied by decreases in key
synaptic proteins, as well as BDNF, a key neurotrophic factor that supports the differentiation,
maturation, and survival of neurons in both development and adulthood (Gassowska et al.. 2016a).
Additionally, studies in adult humans suggested that decreases in BDNF are associated with
neurodegenerative diseases (Bathina and Das. 2015). These changes in synapses can result in synapse
dysfunction, which would contribute to altered neurotransmission. Adverse changes in neuronal dendrite
morphology were reported following developmental Pb exposures in multiple studies, including loss of
dendritic spines, reduced spine density, decreased spine length, and impaired spine maturity and
morphology at multiple developmental stages and brain regions (hippocampus, medial prefrontal cortex,
dentate gyrus) (Xiao et al.. 2020; Saleh et al.. 2018; Zhao et al.. 2018; Sepehri and Ganji. 2016; Wang et
al.. 2016; Du et al.. 2015; Rahman et al.. 2012b). Dendritic spines are the morphological and structural

3-20

-------
basis for synaptic plasticity, learning, and memory (Frank et al.. 2018). providing biological plausibility
for altered learning and memory, as reported in Section 3.5.1.3.2.

The brain and CNS are separated from the blood by the BBB and the blood-cerebrospinal fluid
(CSF) barrier, which allows for selective transport of materials into the CNS. The BBB is formed by
several cell types including endothelial cells, astrocytes, pericytes, and microglia and plays an important
immunological role in protecting the brain from circulating pathogens and toxic substances. Two studies
assessed the integrity of the BBB following Pb exposure and observed disruption of brain permeability
with reduced levels of tight junction proteins and other important capillary proteins (Wu et al.. 2020a;
Song et al.. 2014). Increased permeability of the BBB may exacerbate neurotoxicity, as more toxicants
can penetrate the brain with repeated or continuous exposure. Additional research is needed to fully
elucidate the effects of Pb exposures on the BBB and the potential implications for CNS function and
disease.

Neuroinflammation is a complex response that involves microglia and astrocyte activation, as
well as other signaling proteins and cells (such as cytokines, reactive oxygen species, decreased
antioxidant activity). Inflammation is protective against pathogens but can result in neuronal injury or
neuronal loss in the CNS. Several studies in rodents evaluated and observed an effect of Pb exposure on
markers of neuroinflammation, including microglia and astrocyte activation, and the promotion of cellular
reactivity and inflammation (Wu et al.. 2020a; Saleh et al.. 2018; Shvachiv et al.. 2018). Numerous
studies also reported decreased numbers of microglia, which are glial cells that function primarily as
immune cells with macrophage activity, clearing cellular debris and dead neurons from nervous system
tissue (Dominguez et al.. 2019; Sobin et al.. 2013). Reduced numbers of microglia could indicate reduced
capacity for clearing cellular debris and responding to pathogens, which could contribute to functional
and morphological brain changes.

Experimental animal studies of rodents have also shown that Pb exposures affect measures of
brain metabolism, including reduced glycogen concentrations in various brain regions (such as the
forebrain, hippocampus, and cerebellum) (Baranowska-Bosiacka et al.. 2017) and reduced rates of
metabolism, which could indicate reduced glucose availability and poor metabolic cooperation between
neurons and astrocytes (Baranowska-Bosiacka et al.. 2017). A recent study directly measured decreased
hippocampal glucose metabolism following developmental Pb exposure (pre-mating through PND 10)
through reductions in glucose transporters (Zhao et al.. 2021). Glucose cannot be synthesized or stored in
neurons, hence glucose supply and transport are essential for neurophysiological processes with high
glucose demands (such as learning and memory). Because the brain has the highest energy demand and
metabolism of any organ, energy homeostasis is critical. In the above study, effects on glucose
metabolism persisted at PND 30 when the blood Pb concentration had returned to control levels (BLL
was 11.4 (ig/dL at PND 10 and 1.8 (ig/dL BLL at PND 30).

Taken together, animal studies provide strong evidence that Pb exposure impacts brain structure
and function. Altered brain morphology, increased brain inflammation, oxidative stress, and associated

3-21

-------
mitochondrial damage have all been consistently reported following Pb exposures. These effects were
observed across multiple brain regions, on different levels of brain organization, across a variety of
lifestages, and in both sexes.

3.4.2.2 Neurotransmitter Analysis

Neurotransmitters are molecules involved in the transmission of chemical signals between
neurons and target cells and are involved in controlling a wide variety of brain functions, including motor
function, learning, memory, metabolism, behavior, and hormone production. These neurochemical
systems have been implicated in the initiation and maintenance of some brain diseases and disorders, e.g.,
Parkinson's disease (PD), depression, aggression, and dementia (Monday et al.. 2018; Chichinadze et al..
2011; Haden and Scarpa. 2007; Webster. 2001). As described in the 2013 Pb ISA, exposures to Pb can
induce changes in brain neurochemistry and signaling that vary by brain region, neurotransmitter type,
and the sex of the animal. Pb can compete with calcium ions (Ca2+) for common binding sites and second
messenger system activation. When Pb activates a Ca2+-dependent system in the nervous system, it can
contribute to spurious neurotransmitter regulation and release because this system intimately relies on
Ca2+ signaling for its homeostasis. Pb-related alterations in neurotransmission are discussed in further
detail below.

A variety of neurotransmitters and their metabolites were evaluated in experimental animal
studies of rodents, across multiple brain regions (hypothalamus, cerebral cortex, nucleus accumbens,
frontal cortex, striatum, hippocampus, olfactory bulb, midbrain, cerebellum) and time points, including
serotonin and its metabolite 5-hydroxylindolacetic acid (Weston et al.. 2014; Mansouri et al.. 2013;
Graham et al.. 2011). norepinephrine and its metabolite methoxyhydroxyphenylglycol (Long et al.. 2022;
Basha et al.. 2014; Weston et al.. 2014; Biioor et al.. 2012; Graham et al.. 2011). dopamine and its
metabolites dihydroxyphenylacetic acid and homovanillic acid (Sobolewski et al.. 2020; Amos-Kroohs et
al.. 2016; Stansfield et al.. 2015; Basha et al.. 2014; Weston et al.. 2014; Corv-Slechta et al.. 2012;
Graham et al.. 2011). acetylcholine (Long et al.. 2022; Mansouri et al.. 2013). epinephrine (Basha et al..
2014). glutamate and its precursor glutamine (Long et al.. 2022). The direction of changes depended on
the brain tissue analyzed, time point, sex, and specific neurotransmitter assessed. However, multiple
studies found significant effects of Pb exposure on the dopamine system (Sobolewski et al.. 2020; Amos-
Kroohs et al.. 2016; Stansfield et al.. 2015; Basha et al.. 2014; Weston et al.. 2014; Corv-Slechta et al..
2012; Graham et al.. 2011). Some studies also reported Pb-induced changes in enzymes involved in
neurotransmitter turnover and cycling, including monoamine oxidase (Basha et al.. 2014) tyrosine
hydroxylase (Sobolewski et al.. 2020). glutamine synthase, and adenylate cyclase (Long et al.. 2022).

Pb exposure has demonstrated effects on several neurotransmitters, which are important signaling
molecules that control multiple brain functions. These effects were observed across multiple brain
regions, across a variety of lifestages, and in both sexes. Altered neurotransmitter signaling can contribute

3-22

-------
to multiple brain dysfunctions and disorders, providing biological plausibility for the health effects
discussed in subsequent sections.

3.4.2.3 Brain Weight

Organ weights are a frequently assessed in experimental animal studies as they can easily be
measured during animal necropsy. Importantly, brain weight is an indication of severe toxicity as the
body goes to great lengths to spare the brain at the expense of other bodily systems. Brain weights were
not reviewed in the 2013 Pb ISA. Several recent studies of rodents assessed brain weight following short
and long-term Pb exposures, and most of these studies reported nonsignificant findings (Vigucras-
Villasenor et al.. 2021; Mani et al.. 2020; Wu et al.. 2020a; Singh et al.. 2019; Rahman et al.. 2018; Saleh
et al.. 2018; Zhou et al.. 2018; Singh et al.. 2017; Barkur and Bairv. 2015a; Wang et al.. 2013; Rahman et
al.. 2012b). However, two studies did find significant decreases in brain weight (16%—21% decrease): one
reported a decrease after a 90-day oral exposure to Pb in juvenile rats, which resulted in a BLL of 28.4
(ig/dL (Singh et al.. 2019) and the other reported an 18% decrease in cerebellum weight in treated dams
(27.7 (ig/dL BLL), as well as reduced fetal brain weight at parturition following gestational exposure to
Pb in drinking water (GD 1 to GD 20) (Saleh et al.. 2018). In addition to gross measures of brain size and
morphology (e.g., wet weight), studies using newer anatomical imaging methods have been conducted
since the 2013 Pb ISA. Three-dimensional imaging technologies, such as MRI and ultrasound methods
are being used for the analysis of neuroactivity and phenotypes in rodent toxicology studies (Turnbull and
Mori. 2007). Brain volume and MRI morphometry were assessed in a single study of developing mice
following dietary exposure to Pb (Abazvan et al.. 2014). The authors reported no changes in lateral
ventricle volume but did observe sex-specific changes in morphology including enlarged lateral
ventricles. As the technologies improve, imaging technologies offer promising results for evaluating
physiological functions like neural activity in whole intact animals. Study details are provided in Table
3-1T.

3.4.2.4 Electrophysiology

The effects of Pb exposure on brain electrophysiology were not reviewed in the 2013 Pb ISA.
Several new studies in rodents have found that Pb exposure affects measures of brain electrophysiology,
including long-term potentiation (LTP) and evoked excitatory postsynaptic currents (EPSCs). LTP is the
process of signal transmission by which synaptic connections between neurons are activated and
strengthened and may be one of the mechanisms underlying learning and memory processes. Recording
of LTP is a recognized model for the study of memory (Lynch etal.. 1990).

Presynaptic plasticity can be assessed using paired-pulse stimulation, wherein two stimuli occur
in close succession. Hippocampal slices from rats were subjected to LTP induction and high-frequency

3-23

-------
tetanic stimulations, and the magnitudes of EPSCs were measured. In developmentally Pb-exposed rats,
the magnitudes of EPSCs were lower at PND 10, suggesting impaired hippocampal induction (Zhao ct al..
2018). Other studies have also assessed EPSCs with a longer exposure duration and different conditions
and found the ratios of EPSC responses between paired-pulse stimuli were significantly greater in
hippocampal slices from Pb-exposed rats (Zhang et al.. 2015b). These changes in EPSC responses were
accompanied by inhibition of synaptic vesicular release (Zhang et al.. 2015b). Depressed LTP following
Pb exposures was measured in (Zhou et al.. 2020a; Wang et al.. 2016; Liu et al.. 2012). These studies
additionally reported increased neuronal free Ca2+ concentration and inhibition of various signaling
proteins (Ca2+ calmodulin dependent protein kinase II and CREB), which were mediated by upregulation
of the ryanodine receptor. Ryanodine receptors are ion channels that are critical for maintaining
intracellular Ca2+ homeostasis. Changes in electrophysiological parameters affect neurotransmitter release
and cell signaling, which can affect brain function (described in Section 3.4.2.2).

In addition to these effects, (Zhu et al.. 2019a) reported alterations in cardiac sympathetic nerve
activity in rats while evaluating nerve discharge as a potential contributor to other health effects discussed
in the cardiovascular toxicity section (Appendix 4). The authors reported enhanced cervical sympathetic
nerve discharge 1 year after Pb exposure ended, suggesting that Pb-induced alterations to autonomic
nervous dysfunction can have lasting effects. This growing area of research recognizes the potential effect
of Pb on electrophysiology in the nervous system.

3.4.2.5 Circadian Rhythms

Two recent studies evaluated the effects of Pb exposure on circadian rhythms using rodent
models. The suprachiasmatic nucleus (SCN) of the hypothalamus is the primary regulator of circadian
physiological processes and is synchronized daily by signals of light. Vigueras-Villasenor et al. (2021)
subjected male rats to chronic Pb exposure from conception to euthanasia. In these adult rats, under a
standard 12:12-hour light-dark cycle, the authors observed daily delays in the nocturnal onset of
locomotor activity. With a 6-hour photoperiod delay, the activity rhythms of Pb-exposed rats entrained to
a new cycle faster than controls, and Pb treatment showed no significant effects when the photoperiod
was advanced by 6 hours. Histochemical analyses of the hypothalamus in light-pulsed Pb-treated animals
displayed decreases compared with controls in both photo-stimulated neurons (immunoreactivity to c-
Fos) and the neuronal population in the SCN. Hsu et al. (2021) assessed disturbances in rodent sleep
homeostasis by using electroencephalography and electromyography to score the sleep wake architecture
of sleep cycles and found that adult rats with chronic Pb exposure showed disturbances in sleep patterns
that were accompanied by altered clock gene expression and changes in the hypothalamus. These
alterations in behavior, sleep cycles, brain structure, neuronal function, and gene expression warrant
further investigations into the effects of Pb on the rhythm of vital circadian processes. The above studies
reinforce the importance of considering the time of day in studies measuring the effects of Pb.

3-24

-------
3.4.2.6

Summary

In conclusion, multiple studies measured a variety of nervous system endpoints in brains of rats
and mice following exposure to Pb including histology, neurotransmitter analysis, brain weight, and
electrophysiology measures. Histological analyses revealed reduced neuron counts, altered synapse
morphology, and increased apoptosis, as well as oxidative damage in several brain regions, including the
hippocampus, frontal cortex, and cerebellum. These regions were also found to have damaged
mitochondria, vacuolization, and morphological changes. Pb concentrations ranged from 4.7 (ig/dL to
28.4 (ig/dL in these studies, which were conducted in a variety of animal models, sexes, and lifestages.
These endpoints provide biological plausibility for effects on cognitive behavioral changes and diseases
described in subsequent sections.

3.4.3 Integrated Summary of Overt Nervous System Toxicity

Overt nervous system toxicity refers to a diverse group of endpoints that inform brain structure
and function, including brain histopathology changes, brain weight, electrophysiology,
neuroinflammation, and neurotransmitter analyses. As described in Section 3.1, there are no causality
determinations for this endpoint grouping. Instead, the evidence is considered supporting information that
informs the health determinations in Sections 3.5 and 3.6.

Multiple studies measured nervous system endpoints in rats and mice following exposure to Pb
including histological changes in brain structure and morphology, neuroinflammation, neurotransmitter
analysis, brain weight, and electrophysiology measures. These findings that Pb exposures affect these
endpoints provide biological plausibility for Pb to elicit human cognitive behavioral changes and diseases
described in subsequent sections and are generally coherent with the epidemiologic studies of overt
nervous system effects described in this section. The lack of toxicology studies examining Pb inhalation
remains a data gap that limits the evaluation of nervous system effects from inhaled Pb.

Multiple epidemiologic studies, including prospective studies following children through
adolescence and adulthood and a smaller number of cross-sectional studies of children, found associations
between BLL and physiologic changes in regions of the brain responsible for cognition and behavior.
These prospective epidemiologic studies considered important confounders; however, many Pb exposure
metrics and brain regions were evaluated in the epidemiologic studies, raising the possibility of chance
findings.

3-25

-------
3.5 Nervous System Effects Ascertained during Childhood,
Adolescent, and Young Adult Lifestages

The collective body of epidemiologic and experimental animal studies assessed in the 2013 Pb
ISA demonstrated the effects of Pb exposure on an array of nervous system outcomes. Overall, the largest
body of evidence assessed in the 2013 Pb ISA, as well as the outcome that was best substantiated to occur
at the lowest Pb exposure levels, was related to cognitive effects in children. Multiple prospective studies
conducted in diverse populations consistently demonstrated associations of higher blood and tooth Pb
levels with lower full-scale IQ (FSIQ), executive function, and academic performance and achievement.
The blood Pb biomarkers used in these studies reflect exposures during prenatal, postnatal and childhood
lifestages. Tooth Pb generally reflects prenatal and early childhood exposure (or exposure up to the time
that the tooth is shed depending on the specific tooth layers analyzed) (see Section 2.3.4.1.) Most studies
examined representative populations and had moderate to high follow-up participation with no indication
of selective participation among children with higher BLLs and lower cognitive function. Associations
between BLL and cognitive function decrements were found with adjustment for several potential
confounding factors, most commonly SES, parental IQ, parental education, and parental caregiving
quality. In children aged 4-11 years, associations were found with prenatal, early childhood, childhood
average, and concurrent BLLs in populations with mean or group BLLs in the range of 2-8 (ig/dL.
Although examined less extensively than cognitive effects, a strong body of evidence also indicated Pb-
associated decrements in attention and increased hyperactivity.

3.5.1 Cognitive Function in Children

The evidence evaluated in the 2013 Pb ISA was sufficient to conclude that there is a "causal
relationship" between Pb exposure and decrements in cognitive function in children (U.S. EPA, 2013).
Multiple prospective studies conducted in diverse populations consistently demonstrated associations of
higher blood and tooth Pb levels with lower FSIQ, executive function, and academic performance and
achievement. As noted above, these biomarkers reflect exposures during prenatal, postnatal and childhood
lifestages (tooth Pb concentration may reflect exposure up to the time a tooth is shed depending on the
layer analyzed [see Section 2.3.4.1]). Most studies examined representative populations and had moderate
to high follow-up participation with no indication of selective participation among children with higher
BLLs and lower cognitive function. Associations between BLL and cognitive function decrements were
found with adjustment for several potential confounding factors, most commonly SES, parental IQ,
parental education, and parental caregiving quality. In children aged 4-11 years, associations were found
with prenatal (i.e., maternal or cord BLLs), early childhood, childhood average, and concurrent BLLs in
populations with mean or group BLLs in the range of 2-8 (ig/dL. No critical lifestage or specific duration
of Pb exposure within childhood was uniquely associated with cognitive function decrements based on
consideration of evidence from epidemiologic and toxicological studies. Several epidemiologic studies

3-26

-------
found a supralinear concentration-response (C-R) relationship (i.e., larger incremental effect at lower
BLLs). A threshold for cognitive function decrements was not discernable from the available evidence
(i.e., examination of early childhood blood Pb or concurrent [with peak <10 |ig/dL| blood Pb in the range
of <1 to 10 (ig/dL). Evidence in children was clearly supported by observations of Pb-induced
impairments in learning and memory in juvenile animals. Several studies in animals indicated learning
impairments with prenatal, lactational, post-lactational, and lifetime (with or without prenatal) Pb
exposures that resulted in BLLs of 10-25 (ig/dL. Biological plausibility for Pb-associated cognitive
function decrements was supported by observations of Pb-induced impairments in neurogenesis,
synaptogenesis and synaptic pruning, LTP, and neurotransmitter function in the hippocampus, prefrontal
cortex, and nucleus accumbens.

The structure of the current assessment of cognitive effects in children is similar to that in the
2013 Pb ISA. Although the above measures of cognitive function are interrelated, the evidence for each of
these categories of outcomes (i.e., FSIQ, Bayley Scales of Infant Development [BSID],
neuropsychological tests of learning, memory, and executive function, and academic performance) was
assessed separately, to the extent possible, in the order of strength of evidence. Studies assessing
cognitive function of school-age children using instruments that measure FSIQ are described in Section
3.5.1.1, and studies assessing cognitive development in infants using the BSID and other instruments are
described in Section 3.5.1.2. Studies examining the associations of Pb exposure with outcomes on
neuropsychological tests of learning and memory and executive function in children as well as analogous
endpoints in animals are discussed in Sections 3.5.1.3 and 3.5.1.4, respectively. These sections are
followed by a discussion of studies that examine the association of Pb exposure with academic
achievement and performance in Section 3.5.1.5. The final sections discuss issues relevant for the
interpretation of the evidence base (Section 3.5.1.6) and the summary and causality determination
(Section 3.5.1.7).

Because the conclusion from the 2013 Pb ISA was "causal," the PECOS statement for studies of
cognitive effects in children (see Section 3.2) was refined to emphasize recent studies that examined
lower BLLs more similar to those of current U.S. children (i.e., <5 (ig/dL). Details of these studies are
extracted into the evidence inventories (Section 3.7, Table 3-2E [FSIQ], Table 3-3E [Infant
Development], Table 3-4E [Learning, Memory, and Executive Function], and Table 3-5E [Academic
Achievement and Performance]). Studies with central tendency blood Pb concentrations that exceed 5
(ig/dL are extracted into Table 3-6E of Section 3.7. In addition to refining the PECOS statement to focus
on lower exposure levels, studies of younger children whose BLLs were less influenced by higher past Pb
exposures are considered particularly informative. Controls for important potential confounders identified
in the 2013 Pb ISA such as SES, parental education, quality of parental caregiving (often measured as the
HOME score), nutritional status, and birth weight in studies of postnatal Pb exposure were considered
attributes of high-quality studies (see section 4.3.13 of the 2013 Pb ISA (U.S. EPA. 2013)). A summary
of the recent evidence, which is interpreted in the context of the entire body of evidence, is provided in

3-27

-------
the subsequent sections. Overall, recent studies add to the evidence generally supporting the findings from
the 2013 Pb ISA.

3.5.1.1 Full-Scale IQ in Children

A large number of studies evaluated in the 2013 Pb ISA found a consistent pattern of associations
between higher BLL and lower FSIQ in children aged 4-17 years (see Figure 4-2 and Table 4-3 (U.S.
EPA. 2013)). FSIQ has strong psychometric properties (i.e., reliability, consistency, validity), is among
the most rigorously standardized cognitive function measures, is relatively stable in school-age children,
and has been demonstrated to be predictive of educational achievement and life success. The strongest
evidence was provided by prospective studies with analyses of the association of blood Pb levels
measured in early childhood or tooth Pb level that generally reflect the early childhood Pb exposure (i.e.,
prospective studies where Pb exposure preceded the assessment of FSIQ). These prospective studies
typically considered potential confounding by maternal IQ and education, SES, birth weight, smoking
exposure, parental caregiving quality, and in a few cases, other birth outcomes and nutritional factors.
Associations were found in diverse populations (e.g., Boston, MA; Cincinnati, OH; Rochester, NY;
Cleveland, OH; Mexico City, Mexico; Port Pirie, Australia; and Kosovo, formerly of Yugoslavia) in
studies that examined children recruited from prenatal clinics, hospital maternity departments, or schools.
Studies generally reported high follow-up participation, which was supported by evidence that selection
bias did not explain the associations observed. The few studies reporting weak or null associations (i.e.,
Cleveland, Sydney cohorts) were not stronger with respect to methodology or control for potential
confounding and did not weaken the far larger body of supporting evidence (U.S. EPA, 2013).

The blood Pb-FSIQ association in children was further substantiated by an international pooled
analysis of seven prospective cohorts (Lanphear et al„ 2019, 2005) as well as multiple meta-analyses that
combined results across various prospective and cross-sectional studies (Pocock et al„ 1994; Schwartz,
1994a; Needleman and Gatsonis, 1990). Schwartz (1994a) additionally demonstrated the robustness of
evidence to potential publication bias. The pooled analysis (Lanphear et al., 2019, 2005) examined several
BLL metrics and demonstrated that early childhood and concurrent childhood BLLs explained more
variation in FSIQ compared with the other blood Pb metrics, as indicated by the R-square values. The
coefficient for concurrent BLL had a smaller standard error (SE) than the coefficient for early childhood
BLL (Lanphear et al., 2019). Across studies, no clear indication that Pb exposure during one critical
lifestage or time period within childhood was uniquely or more strongly associated with FSIQ (see
Section 3.5.1.6.3). Blood Pb-associated FSIQ decrements at ages 4-17 years were found with concurrent,
prenatal (maternal or cord), early childhood (e.g., age 2 or 4 years), multiple-year average, or lifetime
average BLLs. Associations were also found with tooth Pb levels.

Key statistics associated with the international pooled analyses of seven cohort studies are
presented in Table 3-1. The C-R function was nonlinear, with a larger incremental effect of Pb on IQ at
lower blood Pb concentrations (Lanphear et al., 2019, 2005). The log-linear model coefficient (i.e., |3

3-28

-------
coefficient) for concurrent BLL was -2.65 (95% confidence interval [CI]: -3.69, -1.61) per unit change
in natural log transformed BLL. The linear association observed for a subset of 103 children with peak
BLLs <7.5 (mean concurrent BLL = 3.2 (ig/dL) was -2.53 (95% CI: -4.48, -0.58). Linear coefficients for
higher BLLs and using a peak BLL cutoff point of 10 (ig/dL are included in Table 3-1, as are other key
statistics, including R-square values for various models.

Table 3-1 Statistics associated with the international pooled analysis of data
from seven cohort studies

Main Finding from Analyses of the Pooled Dataset

Quantitative Result3

Log-linearb model coefficient for blood Pb metrics and IQ, adjusted for
site, HOME score, birth weight, maternal IQ, and maternal education

Early childhood: -2.21 (-3.38, -1.04)
Peak: -2.86 (-4.10, -1.61)

Lifetime average: -3.14 (-4.39, -1.88)
Concurrent: -2.65 (-3.69, —1.61 )c

IQ decrement over different concurrent blood Pb ranges based on the
log-linear model

2.4 to 30 |jg/dL: 6.7 IQ pts (4.1-9.3)
2.4 to 10 |jg/dL: 3.8 IQ pts (2.3-5.3)
10 to 20 |jg/dL: 1.8 IQ pts (1.1-2.6)
20 to 30 |jg/dL: 1.1 IQ pts (0.7-1.5)

Linear coefficient,11 sample size (n) and concurrent BLL measurements
(mean, minimum, 5th and 95th percentiles, and maximum) for subset
with peak BLLs <7.5 |jg/dL

-2.53 (-4.48, -0.58 )
n = 118e

(3.3, 0.9, 1.1, 6.7, 7.4 pg/dL)

Linear coefficient,11 sample size (n) and concurrent blood Pb
measurements (mean, minimum, 5th and 95th percentiles, and
maximum) for subset with peak BLLs >7.5 |jg/dL

-0.15 (-0.19, -0.11)
n = 1215

(13.0, 0.1, 3.7, 34.2, 71.7)

Linear coefficient,11 sample size (n) and concurrent blood Pb d
measurements (mean, minimum, 5th and 95th percentiles and
maximum) for subset with peak blood Pb <10 |jg/dL

-0.77 (-1.65, 0.12)
n = 258

(4.4, 0.1, 1.4, 8.0, 9.8)

Linear coefficient,11 sample size (n) and concurrent blood Pb
measurements (mean, minimum, 5th and 95th percentiles, and
maximum) for subset with peak BLLs >10 pg/dL)

-0.13 (-0.22, -0.04)
n = 1075

(14.0, 0.1,4.4, 35.5, 71.7)

Blood Pb metric with the largest R2 for the relationship with IQ in the
log-linear models

Early childhood R2: 0.6433 = largest
Peak R2: 0.6401
Lifetime average R2: 0.6411
Concurrent R2: 0.6414

BLL = blood lead level; HOME = Health Outcomes and Measures of the Environment; IQ = intelligence quotient; Pb = lead; pts =
points.

aResults reported by Lanphear et al. (20191 and/or Crump et al. (20131 and confirmed by Kirrane and Patel (20141.
bCoefficients are not standardized, i.e., coefficients indicate the decrement in full scale IQ per unit of natural log transformed blood
Pb. Standardized estimates (i.e., standardized to a 1 unit increase for the 10th—90th percentile interval of the biomarker level and
assumed to be linear within this interval) for this study are found in Evidence Inventory (Section 3.7, Table 3-2E).

°Slopes ranged from -2.36 to -2.94 in sensitivity analyses of concurrent blood Pb-IQ association, which omitted one cohort at a time.
dLinear coefficients are standardized to a 1 |jg/dL increase in blood Pb

eThe number of children from Boston cohort with peak BLLs <7.5 |jg/dL was 28 after errors were corrected (Lanphear et al.. 20191.

3-29

-------
Several studies that conducted analyses stratified by BLL provide additional support for the
findings of Lanphear et al. (2005) and Lanphear et al. (2019). These studies comprise a compelling body
of evidence demonstrating a nonlinear C-R function (i.e., larger decrement in cognitive function per unit
increase in blood Pb level in children in the lower range of the study population blood Pb distribution) for
the association between BLL and intelligence. This evidence is described in the 2013 Pb ISA (Section
4.3.12, Figure 4-15, and Table 4-16 of the U.S. EPA (2013)). Particularly compelling evidence was
provided by analyses that examined prenatal or early childhood blood Pb levels or considered peak blood
Pb levels in school-aged children (Schnaas et al., 2006; Bellinger and Needleman, 2003; Canfield et al.,
2003a). The geometric mean Pb level in Schnaas et al. (2006) was 7.8 (ig/dL and the mean blood Pb
levels in Bellinger and Needleman (2003) and Canfield et al. (2003a) were 3.8 (ig/dL and 3.3 (ig/dL,
respectively.

In a recent analysis, Crump et al. (2013) examined the shape of the C-R function for the pooled
data using an alternative modeling strategy. Rather than model the natural log of BLL as was done in the
original analysis (Lanphear et al., 2019, 2005), Crump et al. (2013) modeled the natural log (In) of blood
Pb + 1, which has the property of equaling zero when untransformed BLL equals zero. The authors
applied F-tests to nested models containing both In (BLL + 1) and non-transformed BLL and found that
the linear coefficient did not improve the prediction of the model, indicating that In (BLL + 1) was a
better predictor across the full range of the data (e.g., 2.5-33.2, as 5th to 95th percentile concurrent
BLLs). In addition, Crump et al. (2013) considered confounding by additional covariates, which were
defined as site-specific in their final models. Despite the aforementioned differences in modeling
approach, the Crump et al. (2013) analysis corroborated the findings of the original analysis, providing
strong evidence in support of the nonlinear C-R function and the causal association between Pb exposure
and cognitive effects in children. The coefficient for the Crump et al. (2013) log-linear association
between concurrent BLL and IQ was -3.32 (95% CI: -4.55, -2.08), somewhat larger than that reported
by Lanphear et al. (2019) (see Table 3-1).

Notably, the international pooled analysis (Lanphear et al„ 2019, 2005) included data from seven
longitudinal cohorts that were initiated before 1995. The median concurrent BLL was 9.7 (ig/dL (5th and
95th percentiles: 2.5-33.2 (ig/dL) with included studies reporting limits of detection of 1 (ig/dL. Several
other longitudinal and cross-sectional studies included in the 2013 Pb ISA, however, conducted analyses
of children with mean BLLs <5 (ig/dL, collectively providing strong evidence of an association between
Pb exposure and FSIQ at lower BLLs (Kim et al., 2009; Chiodo et al„ 2007; Surkan et al„ 2007;

Bellinger and Needleman, 2003; Canfield et al„ 2003a) (see Figure 3-4).

Van Landingham et al. (2020) extended the analyses of the international pooled data described
above that were first examined by Lanphear et al. (2005), Lanphear et al. (2019) and later by Crump et al.
(2013). These authors identified "highly likely" confounders (HOME score, maternal education and
maternal IQ) using a combination of correlation analysis and backward selection with a criteria of p =
0.15 to retain covariates in the model. Van Landingham et al. (2020) also included interaction terms

3-30

-------
between BLL and each covariate. The results of their analyses were comparable to previous analyses that
did not consider interactions. Specifically, the beta coefficient for concurrent blood Pb in the loglinear
(blood Pb level +1) model with interaction terms was -4.945. Coefficients for the interactions of maternal
IQ, maternal education and HOME score with Pb were negligible (i.e., -0.0003, -0.0051, and 0.0437,
respectively). The authors also fit predictive models and calculated the IQ decrement for various
increments of BLL, and across levels of each of the covariates. The interpretation of the covariates and
the predicted IQ decrements at various combinations of BLL decrement and covariate levels that are
presented in the paper, however, potentially conflate the interpretation of direct versus total effect
estimates and may not inform causality (i.e., the table 2 fallacy) (Westreich and Greenland. 2013). As
noted previously, there is a lack of data in the international pooled dataset for children with BLLs <1
microgram per dL and the data is sparse below <5 (ig/dL; hence, uncertainty remains regarding the C-R
function associated with this dataset in the range where the data are sparse and this uncertainty is not
addressed by Van Landingham et al. (2020).

Several recent longitudinal studies add to the evidence informing the relationship between BLL
and IQ in children. Heterogeneity in the magnitude and direction of the associations, which was
potentially explained by race/ethnicity, sex, and modeling choices such as adjustment for other metals or
chemicals was present. This heterogeneity did not weaken the larger body of supporting evidence.
Overall, recent studies generally corroborated previous epidemiologic observations of associations
between Pb exposure and IQ in children with relatively low blood Pb concentrations (<5 (ig/dL) (see
Figure 3-4 and Evidence Inventory Table 3-2E).

3-31

-------
Study	Location	Blood Pb

Prospective Studies

fTatsuta et al. 2020	Tohuku district, Japan Prenatal

Prenatal

fTayloretal. 2017	Multi-center, United Kingdom Prenatal

Prenatal

tZhou et al. 2020	Jiangsu Province, China Prenatal

Prenatal
Prenatal
Prenatal

tLee et al. 2021

Bellinger et al. 2003
tlglesias et al. 2011
tLee et al. 2021
tLee et al. 2021

8 hospitals, S Korea

Boston area, MA
Northern Chile
8 hospitals, S Korea
8 hospitals, S Korea

Lanphear et al. 2005, 2019 International
Canfield et al. 2003	Rochester, NY

Kim et al. 2009	4 Cities, Korea

tTatsuta et al. 2020

tRuebner et al. 2019
tlglesias et al. 2011

Tohuku district, Japan

Multi-center U.S.
Northern Chile

Cross-sectional Studies

tDantzer et al. 2020
tMartin et al. 2021

Cincinnati, OH
East Liverpool, OH

tHong et al. 2015	5 regions, S Korea

tMenezes-Filho et al. 2018 Bahia, Brazil
tLucchini et al. 2012	Brescia, Italy

Mean

(pg/dL)

0.8

3.67
1.59

1.32

Boys
Girls

All

Boys

Girls

Early child: 2 yrs

Early child: 3-6 yrs	10.8

Early child: 4 yrs	1.41 (med)

Early child: 6 yrs	1 44(med)

Concurrent
Concurrent
Concurrent

Concurrent
Concurrent
Concurrent
Concurrent

Concurrent
Concurrent
Concurrent
Concurrent
Concurrent
Concurrent

3.2

3.3
1.7

1.2

3.5

0.57
1.13

1.8

1.64

1.71

Pre-site closure

<7.5 pg/dL
<10 pg/dL
Low Mn
High Mn

Boys
Girls

CKD patients

FSIQ Adjusted for:

Age (yrs)

Hg, child Pb

Mn, Cd
Mn, Cd
Mn, Cd
Mn, Hg, Cd

10

10-13
5-15

5-15

4-10
5

6-10

1-18

Post site closure 7-16

High Hair
Low Hair Mn

12

7-9

8-11
7-12
11-14

Hg, prenatal Pb *

-10.00

-8.00

-6.00	-4.00	-2.00	0.00	2.00

Beta values (95% CI) per 1 ug/dL increase in blood Pb

Note: Effect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb. If the Pb biomarker is log-transformed, effect estimates are standardized to
the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated interval. Categorical effect
estimates are not standardized. The "adjusted for" column indicates covariates that are not typically considered in multivariate models. The exhaustive list of confounders for individual
studies is found Table 3-2E.

tStudies published since the 2013 Integrated Science Assessment for Lead.

Figure 3-4 Associations between blood Pb levels and full-scale intelligence quotient in children.

3-32

-------
Several recent prospective epidemiologic studies were also conducted that examined the
associations of prenatal or postnatal BLLs and effect measure modification by sex. Taylor et al. (2017)
used data from the Avon Longitudinal Study of Parents and Children (ALSPAC) to study the association
of maternal BLL and child IQ at ages 4 and 8 (i.e., Wechsler Preschool and Primary Scales of Intelligence
[WPPSI] and Wechsler Intelligence Scale for Children [WISC]-III, respectively). Little evidence of an
association with IQ was observed. The change in score on the WPPSI associated with maternal BLL was
-0.32 (95% CI: -1.32. 0.68) per (ig/dL and the change in score on the WISC-III was 0.26 (95% CI:
-0.21, 0.73) per (ig/dL. Sex-stratified analyses, however, indicated an association between maternal BLL
and IQ decrement in boys (-0.29 [95% CI: -1.02, 0.44]), but not in girls, among whom positive
associations were observed (0.73 [95% CI: 0.39, 1.33]). Models were adjusted for covariates including
maternal education and indicators of SES. The mean prenatal and postnatal BLLs were 3.67 (ig/dL and
4.22 (ig/dL, respectively.

Tatsuta et al. (2020) examined the association of both cord BLL and postnatal BLL (at age 12)
with IQ (WISC-IV) among boys and girls enrolled in the Tohoku Study of Child Development, a
prospective birth cohort. In addition, the Boston Naming Test (BNT) was administered to assess language
abilities. This study found decrements in FSIQ score in association with postnatal BLL [|3 = —9.88 (95%
CI: -18.98, -0.78] among boys and a less precise association with IQ decrement that included the null
value among girls [|3 = -4.41 (95% CI: -15.94, 7.13]). Confounders considered in the analysis included
maternal IQ, parental SES (i.e., income) and Hg concentration in cord blood. Prenatal BLL was
associated with a relatively weak and imprecise decrease in FSIQ score among boys [|3 = -3.68 (95% CI:
-10.71, 3.35], The association of prenatal BLL with FSIQ was slightly positive but with the CI including
the null value among girls (|3 = 1.46 (95% CI: (-2.91, 5.83)]. Lower BNT scores (with cues) were
associated with both prenatal and postnatal BLL among boys. The associations of pre- and postnatal
BLLs with BNT were relatively weak or null in girls. The median postnatal BLL was 0.7 (ig/dL and the
median cord BLL was 0.8 (ig/dL in this study.

Desrochers-Couture et al. (2018) studied the association between cord, maternal, and childhood
(3-4 years old) BLLs with cognitive function (WPPSI-III at 3-4 years of age) among children enrolled in
the Maternal-Infant Research on Environmental Chemicals (MIREC) Study. The analysis included
mothers who participated in MIREC Chemical Study Plus (n = 610), which was conducted when the child
reached the age of 3-4 years old. The cohort from which the study participants were drawn comprised
Canadian preschoolers from mainly middle- to upper-middle SES families with low Pb exposure. The
geometric mean concurrent blood Pb concentration was 0.70 (ig/dL, and the geometric mean cord blood
Pb concentration was 0.76 (ig/d. Outcomes included FSIQ, verbal IQ, performance IQ, and a general
language composite. The authors report standardized regression coefficients for the associations of cord
blood Pb level and child blood Pb level with FSIQ. An association between cord blood Pb level and FSIQ
was observed [|3 = -0.07 (95% CI: -0.143, 0.003)], while the association of childhood concurrent BLLs
with FSIQ effectively null (|3 = 0.014 [95% CI: -0.071, 0.098], These results describe a SD change in
FSIQ (i.e., 13.5 points) per SD change in Log2-transformed blood Pb level (SD of Log2-transformed

3-33

-------
BLLs not reported; results not depicted in Figure 3-4). The cord blood model adjusted for child age, sex,
maternal education, evaluation site, and cord blood Hg, while the postnatal model adjusted for child age,
sex, evaluation site, marital status, income, HOME score, Parenting Stress Index, and cord BLL. No
pattern of Pb-associated cognitive function decrements emerged with the verbal or performance
components of IQ or with the general language composite. An association (non-standardized beta
coefficients) was observed between cord BLL and performance IQ in boys (|3 = -5.69 [95% CI: -9.97,
-1.41] but not in girls (|3 = 0.29 [95% CI: -3.79, 4.36]). The study adjusted for several important
confounders, including SES. Although maternal education was included in the model, maternal IQ, which
is a strong predictor of child IQ, was not accounted for in the analysis.

Zhou et al. (2020b) conducted a study to examine the association of cord blood concentrations of
trace elements, including Pb, manganese (Mn), and cadmium (Cd), with FSIQ, verbal IQ, and
performance IQ components among children from an agricultural region in China who were enrolled in a
prospective birth cohort. In models including each of these elements, cord BLL was not associated with
FSIQ (|3 = 0.67 [95% CI: -0.51, 1.85]). A similar lack of association was observed with performance and
verbal IQ and in sex-stratified analyses. Models were adjusted for covariates including maternal education
and family income. Each of the trace elements were included in the models but interactions between the
elements were not examined. The mean cord BLL among the children was 1.59 (ig/dL. Liu et al. (2015)
developed a predictive model to examine the association of Pb, Cd, and Hg in serum with FSIQ at age 5,
dropping variables based on the variance inflation factor (VIF >10). The final model for FSIQ (i.e.,
WPPSI) did not include cord serum Pb level; thus, no results pertaining to the association of Pb
concentration in serum with FSIQ were presented. Wang et al. (2022) investigated associations between
cord and concurrent venous blood concentrations of Pb, selenium (Se), As, Cu, Mn, and Cr and FSIQ
among children (6-8 years old) born in a hospital in Wujiang, Jiangsu Province. The geometric mean
concentrations of cord and venous blood Pb were 2.83 (ig/dL and 3.30 (ig/dL, respectively. Cord blood Pb
was weakly associated with reduced performance IQ (|3 = -0.11 [95% CI: -0.25, 0.03]) in boys, and
concurrent venous blood Pb was associated with reduced verbal IQ in girls (|3 = -0.49 [95% CI: -0.86,
-0.12]).

Lee et al. (2021) studied the association of Pb and other metals (i.e., Cd, Hg, and Mn) among
mother and infant pairs from eight hospitals in South Korea. In multivariable models including each of
the metals as well as covariates, imprecise negative associations of prenatal Pb exposure (|3 = -1.20 [95%
CI: -4.87, 2.01]), child BLL at age 4 (|3 = -1.83 [95% CI: -4.66, 1.01]), and child BLL at age 6 (|3 =
-2.61 [95% CI: -5.62, 0.40]) with FSIQ were observed. In another study of exposure to multiple trace
metals conducted in Wujiang, China, imprecise associations of maternal cord and early childhood venous
blood were observed with FSIQ (e.g., -4.77 [95% CI: -14.34, 4.79] comparing the upper quartile of
venous child BLL with the reference quartile) (Wang et al.. 2022). The geometric mean blood Pb
concentration was 2.30 (ig/dL (interquartile range [IQR]: 1.83-3.30 (.ig/dL). and the study included 113
children.

3-34

-------
The recent body of evidence also includes a which evaluated the association of BLL in early
childhood (3-6 years) and BLL later in childhood (10-13 years) when IQ was also measured. The BLLs
that were measured later in childhood corresponded to the period when a major source of Pb exposure
was eliminated (i.e., following the closure of a Pb storage facility) (Iglesias et al.. 2011). The early
childhood mean BLL was 10.8 (ig/dL, and concurrent mean BLL was 3.5 (ig/dL. This study found an
FSIQ decrement associated with concurrent BLL (|3 = -0.94 [95% CI: (-1.77, -0.11)]) and a weaker, less
precise association with early childhood BLL (|3 = -0.14 [95% CI: -0.45, 0.16]). Verbal IQ was more
strongly associated with concurrent BLL than performance IQ. These associations were adjusted for
important confounders including maternal IQ and education, HOME score, and SES; however,
participation was moderately low with approximately 43 percent of the children with early childhood
BLLs participating in the IQ assessment.

Braun et al. (2018) conducted a study to determine whether residential exposure interventions
would reduce BLL in children and further result in improvements in the IQ score assessed using the
WPPSI at 5 to 8 years of age. Eligible women were randomly assigned to either a Pb exposure reduction
or injury prevention group. The geometric mean BLLs for children from 1 to 8 years of age was 1.6
(ig/dL in the Pb exposure intervention group and 1.7 (ig/dL in the control group. Dust Pb loadings were
lower following the intervention, but no differences in BLL (i.e., risk of having blood Pb concentration
>2.5 or 5 (ig/dL) were observed among the children. The effect of the intervention on BLL differed
depending on race/ethnicity, however. Specifically, the relative risk (RR) of having an elevated BLL
(>2.5 (ig/dL) indicated a protective effect of the intervention among non-Hispanic black children who
received the Pb intervention (RR: 0.6 [95% CI, 0.4-1.0]), but not among non-Hispanic white children
who received the Pb intervention (RR: 1.0; 95% CI, 0.5-1.9; race/ethnicity x intervention p-value = 0.06).
No improvement in FSIQ was observed among children who received the Pb intervention, nor did
race/ethnicity modify the effect of the intervention on FSIQ.

Several cross-sectional analyses were also conducted. Dantzer et al. (2020) analyzed data drawn
from the Cincinnati Childhood Allergy and Air Pollution Study (CCAAPS), a longitudinal study that
followed children beginning at age 1 and included their caregivers. Children were assessed using the
WISC-IV at their age-12 study visit. BLL, toenail Pb concentration, and information on covariates were
also ascertained at the age-12 visit. A strong but imprecise association between BLL at age 12,
concurrently ascertained IQ -10.87 [95% CI: -16.89, -4.85]), and a relatively smaller association with
toenail Pb concentration (-1.70 [95% CI: -4.27, -0.86]) were observed after adjustment for caregiver IQ,
SES, BMI (sex was considered as a potential confounder). Toenail Pb concentration reflects blood Pb
concentration approximately months to a year before concurrent BLL due to the time it takes toenails to
grow. The concurrent BLL in this study was 0.57 (ig/dL.

Martin et al. (2021) found interactions between blood Pb and blood Mn level with IQ decrement
among children (n = 57-62 depending on the analysis) enrolled in the Communities Actively Researching
Exposure Study (CARES) cohort in East Liverpool, Ohio. BLL was measured and FSIQ ascertained at

3-35

-------
the first clinic visit, which occurred when the child was between 7 and 9 years of age. Stronger
associations between BLL and FSIQ were observed with increasing Mn concentrations in hair and
toenails. For example, the association of blood Pb with FSIQ ranged from 1.69 (95% CI: -3.04, 6.41)
when In hair Mn equaled 5 ng/g to -10.60 (95% CI: -17.17, -4.02) when In hair Mn equaled 7 ng/g. In
contrast, relatively imprecise associations between BLL and FSIQ at varying levels of blood Mn were
observed. The mean BLL among children in this study was 1.13 (ig/dL (range: 0.30-6.64). Havnes et al.
(2015) examined the association of BLL and FSIQ among the same cohort of children. The primary
objective of this study was to examine the effect of Mn on child intelligence. Associations between BLL
and FSIQ were not reported, although a 1 (ig/dL increase in blood Pb was associated with lower
processing speed (|3 = -3.53 [95% CI: -6.95, -0.12]).

Ruebner et al. (2019) evaluated the association between BLLs and FSIQ among children with
chronic kidney disease (CKD). FSIQ was assessed using several instruments depending on the child's age
(i.e., Mullen Scales of Early Learning [age 12-29 months], WPPSI [30 months-5 years], and Wechsler
Abbreviated Scale of Intelligence [WASI; 6-18 years]). Concurrent BLL assessment was associated with
FSIQ decrement (|3 = -2.1 [95% CI: -3.9, -0.2]). Covariates considered as potential confounders
included race, poverty, maternal education, and factors related to CKD (i.e., CKD stage, duration,
glomerular versus non-glomerular diagnosis, hypertension, proteinuria, and anemia). The median BLL in
this study was 1.2 (ig/dL.

Hong et al. (2015) found an association between blood Pb concentration and lower FSIQ (-2.12
[95% CI: -3.79, -0.45] in across-sectional analysis of Korean school children from 8 to 11 years old.
This association persisted in models adjusted for paternal education and income, ADHD rating scale
score, Mn, and Hg (-1.95 [95% CI: -3.61, -0.29] per 10-fold increase). The mean BLL in this study was
1.80 (ig/dL.

Menezes-Filho et al. (2018) examined the association of concurrent BLL with intelligence
(WASI) among children from 7 to 12 years old. Mn in hair and toenails was also measured and the
interaction between metals evaluated. Child IQ was associated with BLL in this study (-2.78 [95% CI:
-4.66, -0.89]) in adjusted models. The mean BLL of children in this study was 1.64 (ig/dL. The effect of
BLL on child IQ was greater among children with higher toenail Mn concentrations.

Lucchini et al. (2012) conducted a cross-sectional analysis of children between the ages of 11 and
14 to examine the relationship between concurrent BLL and FSIQ as well as the potential interactions
with Mn and the aminolevulinic acid dehydratase (ALAD) genotype. A decrement in FSIQ score was
observed in association with BLL after adjustment for covariates including SES and maternal education
(|3 = -2.24 [95% CI: -4.10, -0.37]). No interaction with Mn or ALAD was found. The mean BLL was
1.71 (ig/dL in this study.

3-36

-------
3.5.1.1.1 Summary

A large number of studies evaluated in the 2013 Pb ISA found a consistent pattern of associations
between higher BLL and lower FSIQ in children aged 4-17 years (U.S. EPA, 2013). Multiple recent
longitudinal studies add to the evidence informing the relationship between BLL and IQ in children.
Heterogeneity in the magnitude and direction of the associations was present across studies. In a study of
Canadian preschool children with low blood Pb levels, an association between cord blood Pb level and
FSIQ was observed, while the association of childhood concurrent BLLs with FSIQ effectively null
(Desrochers-Couture et al„ 2018). In addition, associations were observed in boys but not in girls in
several studies (Tatsuta et al., 2020; Taylor et al., 2017). There was some indication that the heterogeneity
across studies could be explained by modeling choices such as confounder adjustment for other metals.
For example, cross-sectional analyses found evidence that exposure to Mn may modify the association
between Pb exposure and IQ in some populations (Martin et al., 2021; Menezes-Filho et al„ 2018).
However, studies that adjusted for multiple metals (e.g., Mn, Hg, Cd, and Pb) in regression models,
without examining the interaction between metals, found little evidence of an association between cord or
postnatal BLL and IQ (Zhou et al., 2020b; Liu et al., 2015), imprecise associations only in boys (Tatsuta
et al., 2020), or large IQ decrements after adjustment for Mn, Hg, and ADHD rating score (Hong et al.,
2015). Overall, recent studies generally corroborated the epidemiologic observations of associations
between Pb exposure and IQ in children with relatively low blood Pb concentrations (<5 (ig/dL) among
some populations of children (see Figure 3-4 and Evidence Inventory Table 3-2E). Consistent with
findings from the 2013 Pb ISA, individual studies continue to report associations of FSIQ with prenatal
BLL (maternal and cord blood Pb) and postnatal BLLs measured at various childhood lifestages. The
heterogeneity in the observations across studies did not weaken the larger body of evidence supporting
the association of Pb exposure with cognitive effects in children at BLLs <5 (ig/dL.

3.5.1.2 Infant Development

The BSID is an assessment instrument that was developed to identify children with
developmental delays. The early versions (e.g., (Bavlev. 1969)) have been expanded and refined, with
subsequent versions incorporating three domains of development (i.e., cognitive, language, and motor),
and parent-reported subtests that reflect social, emotional, and adaptive behaviors (Albers and Grieve.
2007). The current version of the BSID, the BSID-IV, retains the same number of domains but includes
fewer questions within each domain and requires less time to complete (Balasundaram and Avulakunta.
2021).

This section focuses on the Mental Development Index (MDI) and the cognitive and language
scales of later versions of the BSID. The MDI and cognitive/language scales are reliable indicators of the
current development and cognitive function of infants, integrating cognitive skills such as sensory and
perceptual acuities, discriminations, and response; acquisition of object constancy; memory learning and

3-37

-------
problem-solving; vocalization and beginning of verbal communication; and basis of abstract thinking
(McCall et al., 1972). However, the MDI test is not an intelligence test, and MDI scores, particularly
before ages 2-3 years, are not necessarily strongly correlated with later measurements of FSIQ in children
with normal development (U.S. EPA, 2013).

In the review of the MDI evidence in the 2013 Pb ISA, emphasis was placed on results from
examinations at ages 2-3 years, which incorporate test items more similar to those in school-age IQ tests.
Most of the prospective studies reviewed in previous ISAs (U.S. EPA, 2013, 2006) found associations of
higher prenatal (cord and maternal BLL), earlier infancy, and concurrent BLL with lower MDI scores in
children aged 2 to 3 years (see Table 4-4 of the 2013 Pb ISA). These blood Pb-associated decrements in
MDI were observed in populations with mean BLLs of 1.3 to 7.1 (ig/dL. Studies typically recruited
participants before or at birth without consideration of Pb exposure or maternal IQ and reported high to
moderate follow-up participation as well as nondifferential loss-to-follow-up. Most studies adjusted for
birth outcomes, maternal IQ, and education. Cord BLLs were associated with MDI, with additional
adjustment for SES and HOME score in the Boston cohort (Bellinger et al„ 1987) and for HOME score in
the Yugoslavia cohort (Wasserman et al., 1992). Some studies found a stronger association of MDI with
prenatal BLLs than child postnatal BLLs (Hu et al., 2006; Gomaaet al„ 2002; Bellinger et al„ 1987).

Among the studies assessed in the 2013 Pb ISA, several included children with mean BLLs less
than 5 (ig/dL (Henn et al., 2012; Jedrychowski et al„ 2009b; Hu et al„ 2006; Bellinger et al„ 1987).
Recent longitudinal epidemiologic studies of populations or including groups with maternal, cord, or
postnatal mean BLLs less than 5 (ig/dL add to the overall body of evidence (see Section 3.7, Table 3-3E).
These studies are presented in Figure 3-5.

3-38

-------
Study

Location

Blood Pb

Mean

Age at Outcome

Strata

Prospective Studies





Ojg/dL)

(months)



Bellinger et al. 1987

Boston, MA

Prenatal (cord)

6.5

24

Pb 6-7 vs. *10 pg/dL (ref mean: 14.6)*





Prenatal (cord)

1.8

24

Pb <3 vs. sio pg/dL (ref mean: 14.6) *

Jedrychowski et al. 2009

Krakow, Poland

Prenatal (cord)

1.23 (med)

24







Prenatal (cord)

1.23 (med)

36



Hu et al. 2006

Mexico City, Mexico

Prenatal (T1)

7.07

24







Prenatal (T3)

6.86

24







Prenatal (cord)

6.2

24



TKimet al.2013

3 Cities, S Korea

Prenatal (early pregnancy)

1.4 (GM)

6







Prenatal (late pregnancy)

1.3 (GM)

6







Prenatal (early pregnancy)

1.4 (GM)

6

Cd <1.47 pg/L





Prenatal (late pregnancy)

1.3 (GM)

6

Cd >1.47 pg/L





Prenatal (early pregnancy)

1.4 (GM)

6

Cd <1.51 pg/L





Prenatal (late pregnancy)

1.3 (GM)

6

Cd >1.51 pg/L *

tValeri etal.2017

2 Districts, Bangladesh

Prenatal (cord)

1.8

20-40

Pabna





Prenatal (cord)

6

20-40

Sirajdikhan

Huet al. 2006

Mexico City, Mexico

Concurrent

4.79

24



Claus Henn et al. 2012

Mexico City, Mexico

Concurrent (12 months)

5.1

12-36







Concurrent (24 months)

5

12-36







Concurrent (12 months)

5.1

12-36

Mn <2 pg/dL





Concurrent (24 months)

5

12-36

Mn <2 pg/dL

	1	1	1	

-2.00	0.00	2.00

Beta values (95% CI) per 1 ug/dL increase in blood Pb

Note: Effect estimates are standardized to a 1 [jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb. If the Pb biomarker is log-transformed, effect estimates are standardized to
the specified unit increase for the 10th -90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated interval. Categorical effect
estimates are not standardized.

tStudies published since the 2013 Integrated Science Assessment for Lead.

Figure 3-5 Associations between biomarkers of Pb exposure and Bayley Score of Infant Development
Mental Development Index.

3-39

-------
Several studies were conducted using data from Mexico City birth cohorts that enrolled low and
middle-income women seeking prenatal care at maternity hospitals belonging to the Mexican Institute of
Social Security (Y Ortiz et al.. 2017; Henn et al.. 2012; Hu et al.. 2006). Hu et al. (2006) and Sanchez et
al. (2011) were designed to elucidate the time window during pregnancy when the effect of Pb exposure
on neurodevelopment is most pronounced and are discussed in Section 3.5.1.6.3 and included in Table
3-6E, which includes studies with central tendency BLLs >5 (ig/dL. Y Ortiz et al. (2017) examined the
modification of the Pb-neurodevelopment association by prenatal stress using the Crisis in Family
Systems-Revised (CRISYS-R) questionnaire, which assesses negative life events across several domains
(i.e., financial, legal, career, relationships, community and home violence, medical problems, other home
issues, discrimination or prejudice, and difficulty with authority). Using structural equation models, this
study found that 3rd trimester maternal BLL (|3 = -6.60 [95% CI: -13.49, 0.29] per unit of log-
transformed BLL) and the quadratic term for stress (|3 = -0.23 [95% CI: -0.45, -0.01] per unit of log-
transformed BLL) were associated with lower scores on the cognitive component of the BSID. A weak
more than multiplicative interaction between 3rd trimester maternal BLL and stress was also observed (|3
= 1.02 [95% CI: -0.78, 2.82]). Approximately 67% of the mother-infant pairs had complete information
for covariates, which included maternal education, IQ, and HOME score. Henn et al. (2012) studied the
interaction between postnatal blood Mn and Pb levels (age 12 and 24 months) and MDI score at five
different time points between 12 and 36 months of age among the Mexico City mother-infant pairs. The
coefficients for the association between BLL at 12 and 24 months with MDI score were -0.07 (95% CI: -
0.39, 0.25) and-0.08 (95% CI: -0.46, 0.30), respectively. Interactions between the highest quintile of Mn
and continuous BLL at 12 months were observed (|3 = -1.27 [95% CI: -2.18, -0.37]). The model was
adjusted for covariates including hemoglobin, maternal IQ, and maternal education.

Kim et al. (2013b. 2013c) studied the combined effect of prenatal exposure to Pb and Cd on
infant cognitive development at 6 months of age among participants in the Mothers" and Children's
Environmental Health (MOCEH) study, which enrolled infant-mother pairs from maternity clinics in
three Korean cities. Higher maternal BLL in late pregnancy was associated with lower MDI scores (|3 =
-1.74 [95% CI: -3.37, -0.12]), while maternal BLL in early pregnancy was not (|3 = 0.02 [95% CI:
-1.20, 1.24] per (.ig/dL). This association was found after adjustment for Cd and other covariates
including maternal education and SES. A larger decrement in MDI was associated with late pregnancy
maternal BLL among those with Cd levels above the median (|3 = -3.20 [95% CI: -5.35, -1.06])
compared with the decrement observed among those with Cd levels below the median (|3 = -0.29 [95%
CI: -2.88, 2.30]). Further, an increase in MDI was observed in association with early pregnancy maternal
BLL among those with Cd levels below the median (|3 = 2.44 [95% CI: 0.04, 4.83]), indicating the
potential for random error, differential confounding, or other forms of bias to influence findings. In
another study of mother-infant pairs in Korea, Kim et al. (2018b) evaluated the associations between MDI
and various chemicals and metals, including Pb, in perinatal maternal whole blood and umbilical cord
blood. The median maternal and cord blood Pb concentrations were 2.7 (ig/dL and 1.2 (ig/dL,
respectively. Associations of blood Pb concentrations and MDI were assessed but not reported because
they lacked statistical significance.

3-40

-------
Valeri et al. (2017) examined the combined effect of cord blood concentrations of Pb, arsenic
(As), and Mn with cognitive and languages scores on the BSID. This study enrolled infant-mother pairs
from two birth cohorts in Bangladesh, which differed substantially regarding metal profiles and maternal
characteristics including maternal education. This study presents results from multiple regression modes
and also applied Bayesian kernel machine regression (BKMR) in a prospective analysis that considered
covariates including maternal IQ, education, and HOME score. A weak association between increasing
cord Pb level and decreasing cognitive score was observed in the group with lower Mn and As
concentrations in cord blood (|3 = -0.01 [95% CI: -0.02, 0.00]) but not in the group with higher
concentrations of these metals (|3 = 0.01 [95% CI: -0.05, 0.07]).

Koshv et al. (2020) analyzed data from a birth cohort following children living in a slum in
Vellore, India. Blood Pb concentration at 15 and 24 months was averaged to determine the association
with raw cognition score on the BSID at age 2 (|3 = -0.2 [95% CI: -0.2, -0.03]). These results were
adjusted for covariates including SES, maternal IQ, and iron level. In another study, Shekhawat et al.
(2021) obtained cord blood Pb data and BSID-III scores at 6.5 months on average in a prospective cohort
study of mother-child pairs in western Rajasthan, India. The linear regression models showed no
significant associations of Pb levels and cognitive or language scores.

Paraiuli et al. (2015a) and Paraiuli et al. (2015b) assessed the association of cord BLLs with MDI
at 24 and 36 months of age, respectively, in a birth cohort of mother-child pairs recruited from a general
hospital in Bharatpur, Nepal. The median blood Pb concentration was 2.06 (ig/dL. Adjusting for in utero
Pb, As, and zinc (Zn) levels, HOME score, mother's age, parity, mother's education level, family income,
mother's body mass index (BMI) just before delivery, weight of the infant at birth and 24 months after
birth, gestational age, and infant age at the time of BSID-II assessment, no association was observed
between cord blood Pb and 24-month MDI (|3 = -4.21 [95% CI: -13.62, 5.20] per log-transformed BLL)
or 36-month MDI (|3 = 4.05 [95%CI: -3.21, 11.31] per log-transformed BLL).

Several recent studies assessed neurodevelopment using other validated instruments (Nozadi et
al.. 2021; Nvanza et al.. 2021; Zhou et al.. 2017; Vigeh et al.. 2014; Lin et al.. 2013). Zhou et al. (2017)
assessed 139 mother-child pairs from the Shanghai Stress Birth Cohort. Maternal whole blood and
maternal prenatal stress levels were assessed at 28-36 weeks of gestation, and the Gesell Developmental
Schedules (GDS) adapted for a Chinese population were administered to children at 24-36 months of age
in the study. This instrument measures development quotients (DQs) in five domains (gross motor, fine
motor, adaptive behavior, language, and social behavior) and has been validated for children 0-84 months
old. For this section on neurodevelopment, only the language domain is relevant. The Symptom
Checklist-90-Revised was used to produce a Global Severity Index (GSI) for evaluating overall maternal
emotional stress. After controlling for child sex, age, maternal age, gestational week, birth weight,
maternal education, and family monthly income, there was no association between prenatal maternal BLL
and child cognitive development. However, the authors observed interaction effects such that high
maternal stress appeared to exacerbate the effect of prenatal Pb exposure in several domains, including

3-41

-------
language (|3 = -33.82 [95% CI: -60.04, -7.59] per log-10 transformed unit of BLL), while low maternal
stress did not (|3 = -1.76 [95% CI: -13.03, 9.51] per log-10 transformed unit of BLL).

Vigeh et al. (2014) evaluated 174 children in Tehran, Iran up to 36 months postpartum in eight
developmental areas (social, self-help, gross motor, fine motor, expressive language, language
comprehension, letters, and numbers) using Harold Ireton's Early Child Development Inventory (ECDI).
Items for these areas were combined to generate a general development ECDI score, with higher scores
representing better development. This parent-reported measure is meant for use with children 15 months
to 6 years old and includes 60 age-discriminating items from the Minnesota Child Development
Inventory. To assess Pb exposure, three maternal whole blood samples and one umbilical cord blood
sample were collected from each mother-child pair in the first, second, and third trimesters and at
delivery, respectively. The authors observed increased odds (odds ratio [OR] = 1.74 [95% CI: 1.18, 2.57])
of a low ECDI score (<20% lower than expected for the children's age and sex) in the first trimester (BLL
= 4.15 (ig/dL), adjusting for hematocrit, maternal education, BMI, family income, gestational age, birth
weight, and first born.

Lin et al. (2013) measured Pb and other metals (i.e., Mn, As, and Hg) in cord blood samples from
230 mother-infant pairs from the Taiwan Birth Panel Study (TBPS) and assessed development in
cognition, language, motor, social, and self-care skills among 2-year-old children with the Comprehensive
Developmental Inventory for Infants and Toddlers (CDIIT), which has been standardized for children 3 to
71 months old. The CDIIT uses DQs, and a score of 100 represents normal development. After adjusting
for maternal age, maternal education, fish intake >2 times/week during pregnancy, infant gender,
environmental tobacco smoke during pregnancy and after delivery, and HOME Inventory score, the linear
regression models showed that highly Pb-exposed (>75th percentile: 1.65 (ig/dL) children had lower
cognitive DQs (|3 = -5.35 [95% CI: -9.64, -1.06]) compared with those in the low-exposure (<75th
percentile) group. The authors also observed an interaction with Mn such that children who were highly
exposed to both Mn and Pb had larger deficits in cognitive (|3 = —8.19 [95% CI: -14.40, -1.98]) and
language (|3 = -6.81 [95% CI: -12.16, -1.46]) DQs compared with those with low exposure to just one or
both of these metals.

Nozadi et al. (2021) collected blood samples from pregnant mothers at the 36-week visit or at
time of delivery and administered the Ages and Stages Questionnaire Inventory (ASQ:I) at 10-13 months
of age to evaluate neurodevelopment. Trained staff scored children on five 65-70 item developmental
domains: communication, gross motor, fine motor, problem-solving, and personal-social. A 1 (ig/dL
increase in prenatal blood Pb was associated with small, imprecise decreases in problem-solving (|3 =
-0.67 [95% CI: -1.54, 0.20]) scores.

Nvanza et al. (2021) collected dried blood spots from a finger prick to measure Pb (in addition to
Hg, Cd, and As) concentrations in pregnant mothers at 16-27 weeks of gestation from the Mining and
Health study in Northern Tanzania. The authors used the Malawi Developmental Assessment Tool
(MDAT) translated into Kiswahili to assess several functional domains, including social development, in

3-42

-------
children between 6 and 12 months old. MDAT has been validated for children 0-6 years old in rural sub-
Saharan Africa. Covariates in the Poisson regression model included maternal age, maternal education,
maternal and parental occupation, number of under-5-year-old siblings at home, family socioeconomic
wealth quintile, infant sex, infant age, birth weight, and height and weight at the time MDAT was
administered. Concentrations of Pb were low (median: 2.72 |ig/dL). and the German Environmental
Survey for Children reference level of 3.5 (ig/dL was used to dichotomize Pb exposure groups into low
and high exposure groups. The authors did not observe significant associations between high Pb exposure
and language impairment. However, children highly exposed to both Hg (>0.08 (ig/dL) and Pb were more
likely to have global neurodevelopmental impairment (prevalence ratio [PR =1.4 [95% CI: 0.9, 2.1]).

3.5.1.2.1 Summary

Most of the prospective studies reviewed in previous ISAs (U.S. EPA, 2013, 2006) found
associations of higher prenatal (cord and maternal BLL), earlier infancy, and concurrent BLL with lower
MDI score in children aged 2 to 3 years (see Table 4-4 of the 2013 Pb ISA). These blood Pb-associated
decrements in MDI were observed in populations with mean BLLs of 1.3 to 7.1 (ig/dL. Studies typically
recruited participants before or at birth without consideration of Pb exposure or maternal IQ and reported
high to moderate follow-up participation and nondifferential loss-to-follow-up. Recent studies continue to
support associations between Pb exposure (i.e., maternal (Y Ortiz et al., 2017; Vigeh et al„ 2014; Kim et
al„ 2013b, c), cord (Valeri et al„ 2017), and postnatal exposure (Lin et al., 2013)) and poorer performance
on tests of neurodevelopment among mothers and infants with mean BLLs <5 (ig/dL (see Figure 3-5).
Although Zhou et al. (2017) found no association overall, this study reported decrements in several
domains of the GDS among infants of mothers reporting high maternal stress. Similarly, Y Ortiz et al.
(2017) found some evidence of interaction between Pb exposure and maternal stress. Several studies
found interactions between Pb, Mn, or Mn and As (Valeri et al„ 2017; Lin et al., 2013; Henn et al., 2012)
or Cd exposure (Kim et al., 2013b, c). The direction of the interaction was not consistent across studies.
Overall, recent studies support findings from the 2013 Pb ISA and extend the evidence pertaining to
modification of the association between Pb exposure and infant neurodevelopment by maternal stress and
exposure to other metals.

3.5.1.3 Learning and Memory

The 2013 Pb ISA included many studies examining the associations of blood Pb levels with
neuropsychological tests of memory and learning. These domains of cognitive function are related to
intelligence, and several were evaluated in the subtests of FSIQ. Further, indices of memory and learning
are comparable to endpoints examined in experimental animal studies.

3-43

-------
3.5.1.3.1

Epidemiologic Studies of Learning and Memory in Children

The studies evaluated in the 2006 Pb AQCD and the 2013 Pb ISA did not clearly indicate
associations between higher BLL and poorer performance on neuropsychological tests of memory or
learning (i.e., acquisition of new information) in children 4-17 years of age (see Table 4-5 (U.S. EPA,
2013)). The studies used various tests (e.g., spatial span total errors on the Cambridge
Neuropsychological Test Automated Battery [CANTAB], digit span or learning factor score on the
WISC, Kaufman Assessment Battery for Children [K-ABC], memory score on the McCarthy Scale of
Children's Abilities, California Verbal Learning Test [CVLT], and working memory on the Wide Range
Assessment of Memory and Learning [WRAML]) to assess learning and memory, which may account for
some of the heterogeneity observed in the findings. Notably, evidence for both memory and learning from
prospective analyses of several established cohorts (i.e., Rochester, Boston, and Cincinnati) was mixed
(Canfield et al., 2004; Ris et al., 2004; Stiles and Bellinger, 1993; Bellinger et al., 1991; Dietrich et al.,
1991). These prospective studies examined blood Pb metrics including early childhood, lifetime average.
Cross-sectional studies included in the previous ISA, however, generally found associations between
higher concurrent BLLs and poorer learning and memory, including the large (n = 4,853) study of
children aged 5-16 years who participated in the National Health and Nutrition Examination Survey
(NHANES) III (Lanphear et al„ 2000). Associations of higher concurrent BLL and poorer memory in
children aged 5-16 years were also observed by Krieg et al. (2010) and Froehlich et al. (2007); however,
some studies reporting such associations had limited implications because they lacked consideration for
potential confounding (Counter et al„ 2008; Min et al., 2007). Several studies included in the 2013 Pb
ISA were conducted in populations with mean BLLs <5 (ig/dL (Krieg et al., 2010; Surkan et al„ 2007;
Lanphear et al., 2000) and reported associations between increasing concurrent childhood blood Pb
concentration and lower performance on tests of learning and memory.

A small number of recent studies examined the association of Pb exposure with children's
performance on neuropsychological tests of learning and memory (see Section 3.7, Table 3-4E). Several
such studies examined the association between Pb exposure and performance on tests of learning and
memory in models that adjusted for several important confounders plus co-exposure to other metals or
chemicals. Yorifuji et al. (2011) evaluated the association of cord BLL with several components of IQ at
age 7 and age 14 in a Faroese birth cohort also exposed to methyl mercury (MeHg). IQ components
including attention and working memory, language, visuospatial reasoning, and memory were assessed
using the WISC-R and the children's version of the CVLT. The association of cord BLL with
neuropsychological tests of cognition was reported without adjustment for cord Hg, with adjustment for
cord Hg, and with a term for the interaction of cord blood Pb and cord Hg concentration. Poorer
performance on the digit span components of the WISC-R, which measure short-term memory, were most
consistently observed in association with cord BLL. The results for associations with performance on
some of the tests indicated that the interaction between Pb and methyl Hg (MeHg) may be less than
additive (i.e., the associations of cord blood with the neuropsychological test outcomes were most

3-44

-------
discernable among children with hair Hg concentrations below 2.61 |ig/g and among those with the
lowest cord Hg concentrations (e.g., |3 = -0.27 (-0.42, -0.11) at age 14).

Another recent study by Tatsuta et al. (2014) also examined exposure to multiple chemicals
including Pb, PCBs, and MeHg. The outcome in this study was performance on the K-ABC at 42 months
of age. No associations with sequential processing speed score (-2.14 [95% CI: -12.80, 8.53]) or mental
processing score (-3.32 [95% CI: -12.41, 5.77]) were observed after adjustment for variables including
other chemicals, maternal IQ, and family income (associations per unit of log [base not reported]
transformed BLL). Similarly, Oppenheimer et al. (2022) examined the association of cord BLL with
working memory assessed using the WRAML among children (13-17 years old) living near a superfund
site and thus exposed to multiple metals. Regression models were adjusted for prenatal concentrations of
dichlorodiphenyldichloroethylene (DDE), hexachlorobenzene (HCB), PCBs, Pb, and Mn as well as other
important confounders including HOME score and maternal IQ. The associations between verbal working
memory, symbolic working memory and working memory index differences were 0.12 (95% CI: -0.20,
0.45), 0.09 (95% CI: -0.25, 0.42), and 0.59 (95% CI: -0.97, 2.15) respectively. The interaction of Pb
exposure and sex was examined but no statistical evidence of the interaction was observed.

Summary

The studies evaluated in the 2006 Pb AQCD and the 2013 Pb ISA did not clearly indicate
associations between higher BLL and poorer performance on neuropsychological tests of memory or
learning (U.S. EPA, 2013). A small number of recent studies of children with mean BLLs <5 (ig/dL add
to the evidence informing the association of Pb exposure with performance on tests of memory and
learning; however, the results from these recent studies do not enhance the consistency of the evidence as
a whole. Some of the available studies consider co-exposure to other chemicals and metals as confounders
(Tatsuta et al., 2014) although there is evidence that such co-exposures may interact with or modify the
association between Pb and the outcome (Yorifuji et al., 2011). The evidence regarding the effect of Pb
exposure on specific tests of learning and memory lacks consistency, overall.

3.5.1.3.2 Experimental Animal Studies of Learning and Memory

As described in the preceding sections, BLLs are consistently associated with decrements in FSIQ
in children but show variable associations with performance on tests of learning and memory. A
relationship between Pb exposure and cognitive function deficits is further supported by evidence for Pb-
induced impairments in memory and learning in animal models. Critical evidence for the association of
Pb with cognitive impairment comes from a series of studies describing the effects of lifetime Pb
exposure on nonhuman primates (Rice, 1992; Rice and Gilbert, 1990a; Rice, 1990; Rice and Karpinski.
1988). Cynomolgus monkeys {Macaca fascicidaris) were dosed continuously from birth and tested
repeatedly throughout their lifetime. While these exposures yielded BLLs beyond values considered

3-45

-------
relevant for the current assessment (>30 |ig/dL). they provide key evidence of Pb-induced cognitive
impairments in a translationally relevant species.

Learning and Memory - Morris Water Maze

In rodents, spatial learning and memory have been evaluated using several paradigms, including
the Morris water maze. Typically, the Morris water maze task is separated into two distinct phases.

During the training phase, spatial learning is assessed by measuring the time or distance required for a
rodent to swim to a submerged platform using visual cues beyond the maze (e.g., basic shapes). Slower
decreases in time to escape from the maze across training trials (i.e., escape latency) can be indicative of
impaired spatial learning. After the animals have learned the location of the hidden platform (confirmed
by steadily decreasing escape latencies across training trials), memory for the location of the platform is
assessed in the probe phase by removing the platform and measuring the time each animal spends in that
area of the maze. Decreased time spent or distance swam in the target zone can be indicative of a deficit
in spatial memory. Although performance in both phases is primarily a function of learning and memory,
other impairments, such as decreased motivation, motor deficits, or altered perceptual function may also
influence the results. The impact of these factors is difficult to completely characterize, but some studies
may include additional controls or tests (e.g., baseline swimming activity) to reduce this uncertainty.
Because this ISA focuses on low exposure levels which typically do not cause overt toxicity, the impact
of these factors is less likely to play a major role in the interpretation of these results.

The 2013 Pb ISA (U.S. EPA, 2013) reviewed the evidence suggesting that exposure to Pb
produced learning and memory impairments in laboratory rodents using the Morris water maze. Several
studies involved exposures to Pb of varying durations and across different developmental periods.
Significant impairments in both learning and memory were reported for developmental exposures
resulting in BLLs ranging from 23 to 70 (ig/dL. For example, Kuhlmann et al. (1997) compared the
effects of Pb exposure during various lifestages and reported impaired learning and long-term memory in
adult Long-Evans rats exposed during gestation and lactation (via maternal diet) or over a lifetime from
gestation through adulthood. Each of the exposure periods examined produced peak BLLs of 59 (ig/dL.
Exposure during adolescence only, which produced BLLs of 23 (ig/dL, did not affect memory. In contrast
with Kuhlmann, other studies reviewed in the previous ISA reported that postweaning Pb exposure (8
weeks via drinking water) in Sprague Dawley rats resulted in significant deficits in both learning and
memory using the Morris water maze (Fan et al., 2010; Fan et al„ 2009). Recent studies (see evidence
inventory Table 3-4T) provide consistent evidence for Pb-induced impairments in learning and memory
following developmental exposures with lower BLLs than covered in the previous ISA (<30 (.ig/dL).

Evidence reviewed in the previous ISA indicated that development (i.e., preconception, during
gestation, lactation) may be a critical window for Pb exposure to cause cognitive dysfunction later in life.
Several recent studies examined the effects of long-term Pb exposure that began during development and
continued into adulthood. In the study with the longest exposure duration that was relevant to this ISA,

3-46

-------
Ouvang et al. (2019) developmentally exposed Sprague Dawley rats to Pb (0.05% Pb acetate in maternal
drinking water) beginning on GD 0. After weaning, animals were maintained on drinking water
containing (0.01% Pb acetate) until PND 679 (701 total days of exposure). This exposure resulted in a
final mean BLL of 22 (ig/dL. When assessed immediately following the end of exposure, exposed
animals displayed both impaired learning and memory in the Morris water maze task (31% fewer
crossings in the target zone during the probe trial compared with controls). Zhu et al. (2019b) exposed
rats to Pb (0.5 g/L Pb acetate in maternal drinking water) for 387 days beginning at conception, which
resulted in a final mean BLL of 29 (ig/dL. In the Morris water maze, exposed animals showed
significantly increased escape latencies on the last day of training only, suggestive of slightly impaired
spatial learning. In the probe trial, exposed animals made 45% fewer crossings into the target zone than
controls, strongly suggestive of impaired spatial memory. In another long-term study, Zhou et al. (2020a)
developmentally exposed Sprague Dawley rats to Pb through maternal drinking water beginning at
conception and continuing through lactation. After weaning, animals were maintained on Pb in drinking
water (386 days). While this study examined several doses of Pb, only the lowest dose (0.5 g/L in water)
produced BLLs that were relevant to this ISA. Learning and memory were tested via the Morris water
maze during exposure at PND 21 (mean BLL 10 (ig/dL) and later immediately following the end of
exposure at PND 364 (mean BLL 15 (ig/dL). At both time points, Pb-exposed animals took significantly
more time to escape the maze during training and spent less time in the target zone during the probe trial,
indicative of impaired learning and memory. The effect of Pb was slightly more pronounced at the earlier
timepoint (number of crossings in target zone was 36% lower than that of controls at PND 21, compared
with a 26% difference at PND 364), which may be due to improvement on the task with age.

In a study by Tartaglione et al. (2020). male and female Wistar rats were developmentally and
lactationally exposed to Pb beginning 4 weeks prior to conception (GD -28) to PND 23 (50 mg/L in
maternal drinking water), resulting in a final BLL of 26 (ig/dL, and displayed increased escape latencies
compared with controls. This effect was not sex-specific. Exposed animals showed mild memory deficits
in the form of increased latency to target zone and increased distance to target zone relative to controls
(no effect observed on crossings or time spent in target zone). While effects on memory were not sex-
specific, the authors reported that Pb significantly decreased path efficiency (ratio of the shortest possible
path length to the observed path length) in females only, which may indicate a sex-specific effect of Pb on
the processes that govern spatial integration.

Xiao et al. (2014) compared the effects of Pb on two separate developmental windows in Wistar
rats: one beginning prior to conception (2 mM Pb in maternal drinking water from GD -21 to PND 21,57
days total) and the other beginning in adolescence (2 mM Pb in drinking water from PND 21 to 84, 63
days). The gestational exposure yielded a final BLL of 10 (ig/dL at PND 21, while the adolescent
exposure led to a final BLL of 4 (ig/dL. Animals from both exposure groups were tested in the Morris
Water Maze on PND 85. Exposed animals had significantly increased escape latencies and decreased
target zone time relative to control animals, indicating cognitive dysfunction. No difference was observed
between the exposure time frames, suggesting both may be similarly vulnerable to the cognitive effects of

3-47

-------
Pb assessed by the Morris water maze. Similarly, Barkur and Bairv (2015b) employed a study design that
examined multiple different time frames of exposure: pregestational, gestational, and combined gestation
and lactation. All but the combined gestation and lactation group yielded BLLs that were relevant to this
ISA. Pb exposure during each period of development had a significant negative effect on memory
compared with the control groups (learning data were not reported). The gestation and lactation groups
exhibited similar magnitudes of effects, with the pregestational group showing the smallest difference
compared with the control. This study suggests that Pb affects memory following developmental
exposure and that the periods of gestation and lactation may be more sensitive than the pregestational
period alone.

Wang et al. (2021a) exposed Sprague Dawley rats to 0.05 and 0.1% Pb in drinking water from the
beginning of gestation through the end of lactation. BLLs assessed on PND 21 resulted in BLLs of 24.9
and 30.4 (ig/dL for the two dose groups, respectively. On PND 21, Pb-exposed rats displayed
significantly increased escape latencies during acquisition, indicative of impaired learning. During the
probe trial, only animals in the highest Pb concentration had significantly fewer crossings in the target
zone. Betharia and Maher (2012) exposed Sprague Dawley rats starting at conception (10 |ig/mL in
maternal drinking water, GD 0 to PND 20) and assessed cognitive function via the Morris water maze at
two points: end of exposure (PND 21, BLLs of 0.98 (ig/dL) and later (PND 56, BLLs of 0.03 (.ig/dL). This
study reported the lowest BLLs for animals tested in the Morris water maze paradigm. In contrast to many
of the recent studies reviewed here, when assessed immediately following the end of exposure, no effects
on learning or memory were observed. At the later time point, exposed females displayed significantly
impaired memory relative to untreated controls. This minor discrepancy could be due to the lower dose
used in the study. It is also possible that repeated experience with the paradigm across two sessions could
"unmask" a subtle effect on learning and memory produced by low-level exposure to Pb, though data on
cognitive function in animals with BLLs <1 (ig/dL remain limited.

In Anderson et al. (2012). rats were exposed starting prior to conception and then continuing
through lactation (GD -10 to PND 21) to a range of doses and assessed for learning after the end of
exposure. Only the lowest Pb concentration (250 ppm in drinking water) yielded BLLs relevant to this
ISA, with final levels of 19 (ig/dL in males and 18 (ig/dL in females. During the training phase, Pb
exposure had no effect on escape latency in either sex; however, Pb-exposed females displayed
significantly decreased path efficiency compared with untreated controls. While no effect on spatial
memory was observed during the probe trials, exposed females once again exhibited lower path efficiency
scores compared with untreated controls, suggesting that, in females particularly, Pb may influence
pathfinding processes. This study also determined that Pb partially blunted the positive effects of an
enriched environment on spatial learning, which may be relevant when considering how environmental
factors (e.g., SES) may interact with Pb exposure in humans.

In Zhao et al. (2018). rats were developmentally and lactationally exposed to multiple doses of Pb
from GD -14 to PND 10, resulting in final BLLs of 1 (ig/dL for 0.005% Pb and 1.5 (ig/dL for 0.01% Pb

3-48

-------
on PND 30. The highest dose group (0.02% Pb in drinking water) yielded BLLs higher than relevant for
this ISA. At the two relevant doses, Pb exposure led to significant impairments in both learning and
memory. Additional recent studies provided evidence that developmental exposure to Pb resulted in
learning and memory deficits that persisted later into adolescence and adulthood (Xiao et al.. 2020; Li et
al.. 2016a; Zhang et al.. 2014; Rahman et al.. 2012b; Zhang et al.. 2012) In one discrepant study, Wang et
al. (2021b) exposed Sprague Dawley rats to Pb in drinking water (0.05-0.2% Pb) from 4 weeks prior to
conception to PND 21. Only the lowest exposure concentration (0.05%) resulted in a mean BLL relevant
to this ISA (21.1 (ig/dL at PND 21). Learning and memory were also assessed via the Morris Water Maze
on PND 21; the authors reported no significant effects of Pb during acquisition or testing in the 0.05%
exposure group, though some effects on memory were seen at higher concentrations.

One recent study investigated the effects of Pb exposure on the cognitive function of adolescent
rodents. Liu et al. (2022c) exposed 4-week-old Sprague Dawley rats to 0.2% Pb for 12 weeks, which
yielded a mean BLL of 17.3 (ig/dL. The authors reported no effect of Pb on learning during the
acquisition phase; however, during the probe trial, Pb-exposed animals exhibited significantly fewer
crossings relative to untreated controls, suggestive of memory impairment. These recent studies provide
broadly consistent evidence that Pb produces learning and memory impairments, with developmental
periods potentially representing a more sensitive window for exposure.

Learning and Memory - Novel Objection Recognition

Another commonly applied measure of long-term memory in animal models is the novel object
recognition task. Following habituation to an empty arena, animals are placed in the arena with two
identical mundane objects and allowed to explore freely. During the testing phase (-24 hours after
training), animals are returned to the arena with one object from the first day and one novel object and
allowed to explore. The time spent examining each object is recorded. Because rodents tend to explore
unfamiliar objects, these durations can be used to calculate a recognition index, which serves as a measure
of memory. Decreased recognition indices (less time spent with the novel object) suggest impaired
memory. The previous ISA did not incorporate novel object recognition data, but one recent study used
the paradigm to assess long-term memory following Pb exposure relevant to the current assessment.

Tartaglione et al. (2020) observed that long-term exposure to Pb via maternal drinking water (GD
-28 to PND 23), which yielded relatively high BLLs of -26 (ig/dL, caused a significant decrease in novel
object recognition index in females, but not males, when tested at PND 60-72. This study did not report
results from the pre-test phase (i.e., habituation and familiarization), so the potential influence of activity
differences or inherent place preference cannot be determined. However, the result of this single study is
generally consistent with the pattern of memory impairment observed following developmental Pb
exposure, though evidence from the novel object recognition paradigm remains limited. Sex, exposure
timing, and behavioral history may also influence effects on long-term memory, yet the contribution of
each of these factors remains unclear.

3-49

-------
Learning and Memory-Y Maze

Another measure of spatial memory in rodents is the Y maze, which relies on the natural
inclination of rodents to explore new areas rather than revisit previously explored areas. After the animal
in placed in the Y-shaped maze, spontaneous alterations (i.e., entries into an arm different than the most
recently visited arm) and total arm entries are recorded. Total arm entries reflect locomotor activity, and
re-entries into the most recently visited arm from the center of the maze (decreased spontaneous
alteration) may indicate dysfunction in working spatial memory. The previous ISA reviewed evidence
from only one study that utilized the Y maze: (Niu et al.. 2009) reported that Wistar rats exposed to Pb
from lactation up to 12 weeks of age displayed learning impairments starting at 8 weeks of age. These
exposures resulted in BLLs of 17 (ig/dL, which are relevant to the current assessment.

Three recent studies utilized the Y maze to assess spatial memory following Pb exposures that
produced comparable BLLs (Table 3-4T), and the results were inconsistent. Xiao et al. (2020) reported
that female Sprague Dawley rats with long-term developmental exposure to Pb (125 ppm in drinking
water from GD -7 to PND 68) displayed a significant decrease in spontaneous alterations compared with
control females (70 versus 55%), which suggests a deficit in spatial working memory independent of
locomotor function. In contrast, Tartaglionc et al. (2020) did not observe any changes in spontaneous
alterations following a shorter exposure in male and female Wistar rats (50 mg/L in drinking water from
GD -28 to PND 23). Tartaglionc et al. (2020) did report a significant decrease in arm entries made by
exposed rats, which may indicate a Pb-induced alteration in exploratory behavior rather than an effect on
memory. No sex effects were observed in this study. Similarly, Abazvan et al. (2014) conducted a dietary
exposure to Pb from conception to adulthood (approximately 6 months), which yielded BLLs of 26 (ig/dL
in males and 35 (ig/dL in females (not relevant to this ISA). The authors reported no significant effect of
Pb on alterations in the Y maze in either sex. While these recent studies were focused on assessing
memory rather than learning in the Y maze, the effects of Pb on Y maze performance and the influence of
the developmental window remain unclear. Further investigation may be needed.

Learning and Memory - Fear Conditioning

Another measure of learning and memory, fear conditioning, is a task in which animals are
trained to associate a particular conditioned stimulus (e.g., auditory tone) with an aversive unconditioned
stimulus (e.g., mild foot shock). After repeated pairings of the conditioned and unconditioned stimuli
(acquisition), animals are exposed to the conditioned stimulus and the conditioned response (e.g.,
freezing, defined as lack of non-respiratory movement) is recorded. Decreases in freezing behavior may
indicate memory deficits, as the animal is no longer associating the tone with the aversive stimulus.
Several variations on this procedure may be employed to interrogate different brain regions and
processes, such as "trace" fear conditioning, wherein an interval occurs between the tone and the aversive
stimulus. Though fear conditioning data were not incorporated in the previous ISA, four recent studies

3-50

-------
examined the effects of Pb exposures that produced BLLs relevant to the current assessment on
associative memory.

To assess the influence of exposure window on Pb-induced cognitive impairment, Anderson et al.
(2016) exposed male and female Long-Evans rats to Pb (150, 375, and 750 ppm in chow) during three
separate exposure windows (perinatal [GD -10 to PND 21], early postnatal [PND 0 to PND 21], and
long-term postnatal [PND 0 to PND 55]), all of which resulted in BLLs <10 (ig/dL (summarized in Table
3-4T). The authors used a "trace" fear conditioning paradigm with memory testing at 1,2, and 10 days
after conditioning. Anderson et al. (2016) reported significant effects of Pb that differed by sex, exposure
window and dose. In females, learning impairments were observed only in the highest dose group of the
perinatally exposed animals. Some memory deficits were noted in the early postnatal exposure group but
only at lower doses. Interestingly, following long-term postnatal exposure, females only displayed
memory problems at the lowest doses of Pb. This result is not easily explained by variation in BLL (i.e.,
the lowest dose group did not have higher BLLs than the other dose groups). In males, minor learning
deficits were noted in the early postnatal and long-term exposure groups. Memory impairment was noted
in perinatally exposed males at the lowest and highest doses only. The results of this experiment suggest
that both sex and exposure window influence the effects of Pb on learning and memory, and that these
effects may not follow the traditional dose-response relationship reported using other paradigms.

In a subsequent study by the same group, Verma and Schneider (2017) compared the effects of
Pb on associative memory in two different rat strains using a design similar to the previous study
(Anderson et al.. 2016) to examine the influence of exposure window. There were no significant
differences in BLL between Sprague Dawley and Long-Evans rats (Table 3-4T). The authors reported no
Pb effects on acquisition across sex, indicating no effect on learning within the fear conditioning
paradigm. In Long-Evans females, animals exposed during the early postnatal period showed a marked
decrease in percent time freezing during the memory tests (day 1, control: 90% time freezing versus
treated: 68% time freezing). Consistent with the previous study from this group, the effect was more
pronounced after the initial acquisition trials (day 10, control: 70% time spent freezing versus treated:
29% time freezing). Conversely, in Long-Evans males, there was no effect in the postnatal exposure
group, yet significant impairments were detected in the perinatal exposure group starting on the 2nd day
after acquisition (day 2, control: 68% freezing versus treated: 48% freezing). Once again, the effect was
more pronounced later in the experiment (day 10, control: 75% freezing versus treated: 39% freezing).
Interestingly, no significant effect of Pb on learning or memory was observed in Sprague Dawley rats of
either sex with BLLs of approximately 5 (ig/dL.

Wang et al. (2016) exposed male Sprague Dawley rats to 100 ppm Pb in drinking water from
PND 24 to 56 and then assessed memory using a context-dependent fear conditioning paradigm in which
the rats were returned to the same test chambers without atone or shock 24 hours after acquisition, and
freezing was recorded. In this version of the test, environmental context serves as a cue that the animals
associate with the aversive stimulus. The authors reported a dramatic decrease in % time spent freezing in

3-51

-------
treated animals during the memory test 24 hours later (64% time spent freezing in controls compared with
only 8% time spent freezing in treated animals). This long-term adolescent exposure, which produced
BLLs of 13 |ig/dL. resulted in significant memory dysfunction. The divergent results between Verma and
Schneider (2017) and Wang et al. (2016) may be explained by variations in the paradigms used and the
duration, timing of the exposures, and resulting BLLs.

Abazvan et al. (2014) conducted dietary exposure to Pb from conception to adulthood
(approximately 6 months) in mutant (double transgenic) Disrupted-in-Schizophrenia-1 (DISCI; a genetic
risk factor for schizophrenia) mice, which yielded BLLs of 26 (ig/dL in males and 35 (ig/dL in females
(outside PECOS). Single transgenic mice that possessed the mouse DISC 1 (mDISCl) transgene but did
not express mDISCl served as controls. Using a contextual fear conditioning paradigm, these authors
reported no effect of Pb on fear extinction in mutant or control mice. These studies suggest developmental
Pb may adversely affect learning and memory within the fear conditioning paradigm, but these effects
may be sensitive to factors such as sex, strain or genetics, dose, and timing of exposure.

Learning and Memory - Avoidance

Another measure of learning and memory in animal models is the avoidance paradigm, which is a
fear-aggravated test that relies on animals learning to avoid environments where they experienced an
aversive stimulus. In the passive, "step-through" variation of the test, animals are placed in an arena with
at least two compartments, separated by gates that allow passage between compartments. During training,
animals will receive an aversive stimulus (foot shock) in the darkened chamber. The animals are later
placed in an illuminated chamber and the time that elapses before the animals enter the dark chamber is
recorded (entry latency). Shorter entry latencies are associated with impaired memory. The previous ISA
did not incorporate any passive avoidance studies. Four recent studies examined passive avoidance
behavior following Pb exposure.

Barkur and Bairv (2015b) compared the effects of Pb (0.2% in maternal drinking water) on
associative learning in male Wistar rats across several different developmental exposure periods and
durations, all but the longest of which produced BLLs <30 (ig/dL. All exposed animals, except for the
pregestation group (GD -30 to GD 1), displayed decreased entry latencies relative to controls, indicative
of impaired memory. These effects persisted out to 48 hours after the initial exploration trial. Similarly,
Barkur et al. (2011) observed that male Wistar rats exposed via maternal drinking water (0.2%) from GD
0 to PND 21 had significantly shorter entry latencies when assessed at PND 25 and again at PND 120. It
should be noted that BLLs were >30 (ig/dL when measured on PND 25 but the levels decreased to -0.5
(ig/dL by PND 120. Following long-term exposure to Pb via drinking water (50 ppm, GD 0 to PND 45),
Biioor et al. (2012) reported that male and female offspring displayed significantly shorter entry latencies
than their untreated counterparts.

3-52

-------
One study utilized a "step-down" version of the test, wherein animals are placed on a platform
above a grid that delivers a mild electric shock. Over the course of training, animals should learn to
associate the grid with the aversive stimulus and avoid stepping down off the platform. During testing,
both latency (time elapsed before stepping down) and errors (number of times the animal stepped down
onto the grid) are recorded to assess memory. Following long-term developmental exposure to Pb (0.4%
in maternal drinking water from GD 0 to PND 21, BLLs of -14 (.ig/dL). Kunming mice displayed
significantly decreased step-down latency and increased errors relative to controls (Zhang et al.. 2014).
suggestive of both learning and memory impairment. These studies suggest that exposure to Pb during
development results in negative effects on associative learning and memory that may persist into
adulthood and that these effects are influenced by the developmental window during which exposure
occurs.

Learning and Memory with Stress

The paradigm of combined Pb and stress exposure experienced by a laboratory animal has been
examined by the Cory-Slechta laboratory with a focus on the common pathway of an altered
hypothalamic pituitary adrenal (HPA) axis and brain neurotransmitter levels. Effects on learning varied,
depending on the timing of stress, Pb exposure concentration, and sex of the animal. Pb-stress interactions
were found with dietary Pb exposures that resulted in BLLs relevant to this ISA. The evidence
additionally indicated that associations of Pb exposure and stress with learning deficits (multiple
schedules of repeated learning and performance in females) may be related to aberrations in
corticosterone and dopamine. Several recent studies with Pb exposures relevant to the current ISA
included stress components in their experimental designs, providing further evidence that supports an
interaction between stress experience and the effects of Pb on cognitive function.

The Cory-Slechta laboratory expanded upon their previous research by investigating the
interaction between Pb and prenatal stress in males, with additional comparisons between maternal-only
and lifelong exposure (Cory-Slechta et al.. 2012). In contrast to previous reports on females, prenatal
stress with Pb exposure was reported to enhance learning accuracy with a repeated learning and
performance schedule. The authors postulated that this effect may be due to increases in the response rate,
which have been observed in both Pb and stress independently. Thus, this seemingly positive result may
reflect an increase in response rate or impulsivity.

In (Anderson et al.. 2012). researchers examined the influence of differential rearing conditions
(enriched or barren) on Morris water maze performance in Sprague Dawley rats exposed to Pb. The
authors reported a significant positive effect of the environment on learning (decreased escape latencies)
in males regardless of Pb exposure, though this effect was only present during the first two acquisition
trials. The same trend was observed in female rats, though Pb was shown to dull the advantage provided
by enrichment, with complete negation of the advantage observed at the high dose in females. While no
consistent effect was noted on time spent in the target quadrant, Pb significantly impaired the path

3-53

-------
efficiency relative to controls in both sexes, and this effect was ameliorated by an enriched environmental
status. While these studies on the influence of Pb and stress on learning in rodents produced variable
results, they provide evidence supporting the interaction between Pb and stress and suggest that these
effects are further influenced by sex, age, and timing of exposure.

Summary

Several recent studies in laboratory animal models with exposures resulting in mean BLLs <30
(ig/dL add to the substantial body of evidence indicating that Pb exposures can impair learning and
memory. Compared with studies in the 2013 Pb ISA, more recent studies demonstrate that the
relationship between Pb exposures and learning and memory impairments is present at lower BLLs.
Recent studies provided evidence that early-life exposures were associated with learning and memory
impairments later in adulthood, indicating that development is a critical window for the effects of Pb on
cognitive function. Additionally, new evidence suggests that longer durations of Pb exposure (especially
those encompassing developmental windows) produced greater learning and memory impairments. The
few studies reporting weak or null effects were not stronger with respect to the design or methodology
and did not weaken the much larger body of supporting evidence.

3.5.1.4 Executive Function in Children

The executive function domain of cognitive function is related to intelligence. Indices of
executive function are generally comparable to endpoints examined in experimental animal studies.

3.5.1.4.1 Epidemiologic Studies

Epidemiologic evidence presented in the 2006 Pb AQCD and the 2013 Pb ISA indicated a
consistent pattern of associations between higher childhood blood Pb (i.e., blood Pb metrics including
early childhood, lifetime average and concurrent) or tooth Pb levels reflecting pre- or early postnatal Pb
exposure, and poorer performance on tests of executive function in children and young adults (see Table
4-8 (U.S. EPA. 2013)). Associations were found with indices of executive function such as strategic
planning, organized search, flexibility of thought and action to a change in situation, and control of
impulses assessed by various tests including the Intra-Extra Dimensional Set Shift, Wisconsin Card
Sorting Test, and Stroop Color-Word test (SCWT). The strongest evidence was provided by prospective
analyses. These analyses included several birth cohorts in Boston and Rochester and examined BLLs that
preceded the outcome assessment, with adjustment for several potential confounding factors (Canfield et
al., 2004; Canfield et al., 2003b; Bellinger et al., 1994a; Stiles and Bellinger, 1993). Moderate to high
follow-up participation that was not biased to those with higher BLLs and lower cognitive function was
an additional strength of the studies. A small number of cross-sectional studies also found concurrent

3-54

-------
blood Pb-associated decrements in executive function, including an analysis of the Rochester cohort
(Froehlich et al.. 2007). and some studies were limited due to their lack of consideration of potential
confounding (Nelson and Espy. 2009; Vega-Dienstmaier et al.. 2006). A cross-sectional analysis by Cho
et al. (2010) with a concurrent child mean BLL of 1.9 (ig/dL did not find an association with performance
on SCWT.

A small number of recent studies expanded the evidence base pertaining to the effect of Pb on
executive functions in children (study details can be found in Section 3.7, Table 3-4E). Most of these
recent studies assessed executive function using parent-teacher ratings on the Behavior Rating Inventory
of Executive Function (BRIEF) (Gioia et al.. 2002). This instrument comprises three scales including a
Behavioral Regulation Index, which has several components (i.e., emotional control, shift, and inhibit).
Other scales of the BRIEF are the metacognition index and the global executive composite. Higher
BRIEF scores indicate executive function-related behavioral dysfunction.

Fruh et al. (2019) studied mother-child pairs participating in Project Viva, a longitudinal birth
cohort in eastern Massachusetts. Maternal blood Pb concentration in erythrocytes was measured during
the second trimester of pregnancy and parents rated their child's behavior using the BRIEF in mid-
childhood (median 7.7 years). The associations (i.e., |3 coefficients) with the parent and teacher-rated
BRIEF Behavioral Regulation Index were imprecise, i.e., 1.15 (95% CI: -0.22, 2.52) and 0.77 (95% CI:
-0.57, 2.10) per 1 (ig/dL increase in maternal erythrocyte Pb, respectively. In another analysis of these
data, Fruh et al. (2021) aimed to determine the association of joint exposure to Pb, Mn, Se, and MeHg
with scores on the BRIEF and Strengths and Difficulties Questionnaire (SDQ) using BKMR and quantile
g-computation. Individual beta coefficients for each metal from multiple regression models generally
agreed with the original results. Specifically, maternal Pb concentration in erythrocytes (2nd trimester)
was associated with worse parental ratings on the BRIEF global executive composite (|3 = 1.11 [95% CI:
(-0.12, 2.34] per unit increase in maternal erythrocyte Pb). Notably, the mixture was also associated with
poorer parent ratings on the BRIEF in BKMR models.

Sex-specific findings were observed in a study by Merced-Nieves et al. (2022). The researchers
examined the association of prenatal BLL with behavioral tasks on the operant test battery (OTB) (i.e.,
Conditioned Position Responding [CPR], Temporal-Response Differentiation [TRD], Delayed Matching-
to-Sample [DMTS], and Incremental Repeated Acquisition), which assess executive functions, at age 6-7
years. Maternal blood Pb in late pregnancy was not associated with greater response latencies in the CPR
(|3 = 0.00 [95% CI: -0.08, 0.08]) and DMTS (|3 = 0.08 [95% CI: -0.04, 0.20]) tasks, although a small
increase average latency to initiate a response in the TRD task (|3 = 0.14 [95% CI: -0.00, 0.29]). The
association of blood Pb concentrations with two operant tasks were modified by child sex, indicating Pb-
associated changes in the CPR task were more pronounced in girls, and Pb-associated changes in the TRD
task were more pronounced in boys. The mean BLLs during the first trimester, second trimester, and
delivery, in umbilical cord blood, and postnatal were 3.7, 3.9, 4.3, 3.4, and 2.4 (ig/dL, respectively.

3-55

-------
Ruebner et al. (2019) evaluated the association between BLLs and executive function among
children with CKD. In addition to adjusting for important potential confounders including SES and
maternal education, the author adjusted for clinical variables in their models (i.e., CKD stage, duration,
glomerular versus non-glomerular diagnosis, hypertension, proteinuria, and anemia). Executive
functioning was assessed with the Delis-Kaplan Executive Function System Tower Subset (subjects >6
years) and rated by parents using BRIEF for Preschool Children (BRIEF-P; 2-5 years) and the standard
BRIEF (6-18 years) or self-reported by adults (18 years and older) using BRIEF for Adults (BRIEF-A).
Associations between BLL and behavioral symptoms on BRIEF did not persist in models that controlled
for potential confounders including race, poverty, maternal education, and clinical factors related to CKD
(quantitative results not reported). The median BLL in this study was 1.2 (ig/dL.

Summary

Strong evidence of associations between Pb exposure and indices of executive function was
described in the 2013 Pb ISA. Studies included prospective analyses of several birth cohorts with
moderate to high follow-up rates in Boston and Rochester that examined the BLLs that preceded the
outcome assessment and adjusted for several potential confounding factors (Canfield et al.. 2004;

Canfield et al.. 2003b; Bellinger et al.. 1994a; Stiles and Bellinger. 1993). Recent studies that assessed
executive functions using parent or teacher behavioral ratings on BRIEF are mixed; however, findings
from these studies do not diminish the evidence from the earlier well-conducted studies that relied on
neuropsychological testing.

3.5.1.4.2 Toxicological Studies of Executive Function

The epidemiologic evidence reviewed above indicated associations between higher childhood
BLLs and poorer performance on tests of executive functions in children and young adults. Pb was
associated with impaired strategic planning, organized search, flexibility of thought and action to a
change in situation, and control of impulses (described in Section 3.5.1.4.1). In rodents, reversal learning
is one of the main frameworks used to measure cognitive flexibility, an important component of executive
function. Reversal learning tasks assess the ability of animals to actively suppress reward-related response
and disengage from ongoing behavior when the conditions governing the response are altered. For
example, if rodents are trained to press the leftmost of two levers to receive a reward, then the
experimenters cease to reward presses of the left lever and begin rewarding right-lever presses, the
animals must learn that the conditions have changed and adjust their behavior accordingly. Perseverance
(i.e., continuing to press the left lever after the rules have changed) represents impaired cognitive
flexibility and execution dysfunction. This basic paradigm can be expanded to parse specific components
of executive function.

3-56

-------
Two recent studies from the same group assessed executive function using an attention-set
shifting task paradigm following relevant Pb exposures during development. Neuwirth et al. (2019c)
compared performance in Long-Evans rats exposed to Pb during different windows of development.
Briefly, rats were trained to dig for treats by relying on environmental cues that indicated which of two
bowls contained a buried treat. Trained rats were run through a series of discrimination trials including
interdimensional and extradimensional shifts to test cognitive flexibility. An interdimensional shift occurs
when the relevant cue changes but remains within the same dimension (e.g., the baited bowl is still
indicated by a scent, but the correct scent has changed from lavender to peppermint). More complicated
extradimensional shifts require animals to recognize that the relevant cue has changed dimensions (e.g.,
the baited bowl is no longer indicated by scent but by the texture of the media).

Male rats exposed to Pb via lactation (150 ppm in maternal chow) during the early postnatal
period (PND 0 to 22), which yielded BLLs of ~6 (ig/dL, displayed substantial learning deficits in the form
of increased Trials-to-criterion for the olfactory (the relevant dimension) discrimination component.
Indeed, males exposed during the early postnatal period were unable to complete discrimination training
and progress to the next task in the same manner as the control males, indicative of a substantial learning
impairment as the result of Pb exposure. No effect was observed in female rats following exposure in the
postnatal period, despite similar BLLs of ~5 (ig/dL. Even though male rats perinatally exposed (GD -14
to PND 22, BLLs of ~6 (ig/dL) successfully completed discrimination training, they struggled during
testing and displayed significant increases in Trials-to-criterion across simple and complex discrimination
tasks yet solved extradimensional shift tasks in a fashion comparable to control males. In contrast to
males, perinatally exposed females performed poorly in extradimensional shift tasks and showed
improved performance relative to controls in discrimination training and reversal (Neuwirth et al.. 2019c).

In a subsequent study by the same group (Neuwirth et al.. 2019b). the authors reported that
perinatal exposure (GD 0 to PND 22), which resulted in BLLs -10 (ig/dL on PND 22, also produced sex-
specific learning deficits in discrimination training and impaired reversal learning. These studies provide
evidence that exposure to Pb during different developmental windows may produce differential patterns
of executive dysfunction and that these changes may be sex-specific. While the previous ISA did not
include any toxicological evidence that explicitly addressed executive function, the findings of (Neuwirth
et al.. 2019c) and (Neuwirth et al.. 2019b) are consistent with the evidence reviewed in the previous ISA
that indicated Pb exposure contributed to cognitive dysfunction and that the effects of Pb on cognition
were often sex-specific.

Summary

The previous ISA did not incorporate any evidence of the relationship between Pb exposure and
executive function in animal models. Two recent studies from the same group provided evidence that Pb
exposure broadly impairs measures of executive function in a reversal learning paradigm. These effects
were sex-specific, with greater effects reported in males. While these reports are consistent with one

3-57

-------
another, evidence for the association between Pb exposure and impaired executive function in animal
models with BLLs <30 (ig/dL remains limited.

3.5.1.5 Academic Performance and Achievement in Children

Poorer academic performance and achievement is linked with lower FSIQ and may have
important implications for success later in life (U.S. EPA, 2013). The 2006 Pb AQCD and the 2013 Pb
ISA described associations of higher blood and tooth Pb levels, which reflected Pb exposure at various
time periods and lifestages, in children aged 5-18 years with poorer performance on tests of math,
reading, and spelling skills, lower probability of high school completion, lower class rank, and lower
teacher ratings of academic functioning. Notably, associations were reported in prospective studies
examining performance on academic achievement tests (Chandramouli et al., 2009; Min et al., 2009;
Miranda et al., 2009) and an additional analysis of adolescents participating in NHANES (Lanphcar et al..
2000). Several prospective studies (Min et al., 2009; Miranda et al., 2009) and cross-sectional studies
(Krieg et al., 2010; Chiodo et al„ 2007; Surkan et al„ 2007; Lanphear et al„ 2000) were conducted in
populations with population or group mean BLLs <5 (ig/dL. In addition, prospective studies in Boston
and New Zealand found associations of tooth Pb levels collected at an earlier age (e.g., ages 6-8 years)
and generally reflecting pre- or early postnatal Pb exposure, with school performance ascertained at age
18 from school records (Fergusson et al., 1997; Needleman et al„ 1990), suggesting the effect of early
exposure on Pb may be persistent. The strengths of the Fergusson et al. (1997) analysis included a low
probability of selection bias, coherence with results indicating associations between higher tooth Pb levels
and lower teacher ratings of math, reading, and writing abilities at ages 12-13 years (Fergusson et al.,
1993), and consideration of important covariates including SES, parental education, and HOME score.
The Needleman et al. (1990) study was relatively small with no adjustment for parental caregiving
quality.

Recent prospective studies of groups or populations with mean BLLs <5 (ig/dL add to the
evidence supporting an effect of Pb exposure on academic achievement and performance (study details
can be found in Section 3.7, Table 3-5E). Prospective studies have been conducted in Detroit, Chicago,
North Carolina, and a 57-county region in New York State (all counties outside New York City).

Zhang et al. (2013) studied children enrolled in Detroit public schools to determine the
association between childhood BLL measured before age 6 and performance on standardized tests for
math, reading, and science in grades 3, 5, and 8. Compared with students with lower BLLs (defined as
levels <1 (ig/dL), students with higher BLLs (defined as levels between 1 and 5 (ig/dL) had increased risk
of scores that were classified as less than proficient (OR = 1.42 [95% CI: 1.24, 1.63] for math, OR = 1.33
[95% CI: 1.10, 1.62] for science, and OR = 1.45 [95% CI: 1.27, 1.67] for reading. Logistic regression
models were adjusted for covariates including SES (i.e., free and reduced school lunch participation) and
maternal education.

3-58

-------
Evens et al. (2015) conducted a study in children enrolled in the Chicago public school system.
The association between childhood BLLs measured before 72 months and failure on standardized tests for
math and English in grade 3 was examined. This study found that a 1 (ig/dL increase childhood BLL was
associated with an increased risk of failure on the reading and math tests (RR = 1.06 [95% CI: 1.05, 1.07]
and RR = 1.06 [95% CI: 1.05, 1.07], respectively) after adjustment for covariates including child
characteristics, preterm birth, maternal education, and SES (i.e., participation in the free and reduced
lunch program). The associations of BLL with reading failure in white, Black and Hispanic children were
1.14 (95%CI: 1.08, 1.20), 1.05 (95% CI: 1.04, 1.06) and 1.08 (95%CI: 1.05, 1.11), respectively. The
associations of BLL with math failure were in white, Black and Hispanic children were 1.11 (95% CI:
1.05, 1.18), 1.05 (95% CI: 1.04, 1.06) and 1.09 (95% CI: 1.06, 1.12), respectively. The mean BLL was
4.81 (ig/dL in this study. Blackowicz et al. (2016) extended this analysis through their examination of the
association of BLL and failure on standardized tests for math or reading among Hispanic children
enrolled in the Chicago school system. An association between 1 (ig/dL change in BLL and failures in
reading (RR= 1.07 [95% CI: 1.05, 1.10]) and math (RR= 1.09 [95% CI: 1.06, 1.12]) were observed. The
mean BLL was 4.16 (ig/dL in Hispanic children in this study.

In a statewide study of North Carolina school children (Shadbegian et al.. 2019). children with
higher BLLs had, on average, lower scores in both math and reading (averaged over grades 3 and 8) than
children with lower BLLs. Compared with children with BLL <1 (ig/dL, the authors reported a decrease
in the test-score percentile of 0.95 (0.66, 1.24) for math and 1.41 (1.12, 1.70) for reading in children with
a BLL of 5 (ig/dL. Shadbegian et al. (2019) included interaction terms between BLL and the grade of
testing that further indicated that the deficit in the test score persisted from grade 3 to grade 8.

SkerfVing et al. (2015) studied the association of childhood BLL (age 7-12 years) with school
performance in the ninth grade at age 16 among Swedish school children. School performance was based
on a 1-5 point passing grade scale or a 4-level merit system in which 0, 10, 15, or 20 points were
assigned for each increasing level of performance. This study found a 0.11-point decrease (95% CI:
-0.18, -0.05) per 1 (ig/dL increase in BLL for school performance using the grading scale and a -10.90
(95% CI: -15.49, -6.31) point decrease using the merit scale, among school children with BLLs <5
(ig/dL. The models were adjusted for covariates including parent's income, education, and father's IQ
score on the military conscription exam. The association with IQ (|3 = -0.20 [95% CI: -0.39, -0.02])
among those evaluated for military conscription at age 18 was also examined (see Section 3.6.1).

3.5.1.5.1 Summary

Associations of higher blood and tooth Pb levels, which reflect various time periods including
earlier childhood lifestages, in children aged 5-18 years with poorer performance on tests of math,
reading, and spelling skills, lower probability of high school completion and lower-class rank, and
lower teacher ratings of academic functioning were observed in previous assessments (U.S. EPA, 2013).

3-59

-------
Recent studies in populations of children (age 6-16 years) with BLLs <5 (ig/dL support and extend these
observations of poorer academic performance in association with increasing Pb exposure.

3.5.1.6 Relevant Issues for Interpreting the Evidence Base

3.5.1.6.1 Concentration-Response Function

With each previous assessment (U.S. EPA, 2013, 2006), the epidemiologic and toxicological
study findings have shown that progressively lower BLLs or Pb exposures are associated with cognitive
deficits in children. The 2006 AQCD found that cognitive effects in children were associated with BLLs
of 10 (ig/dL and lower, while the evidence assessed in the 2013 Pb ISA found that an association between
BLLs and cognitive effects in children was substantiated to occur in populations of young children with
mean BLLs between 2 and 8 (ig/dL. The conclusions of the 2013 Pb ISA were based on studies that
examined early childhood BLLs (i.e., age <3 years), considered peak BLLs in their analysis (i.e., peak
BLL <10 (ig/dL), or examined concurrent BLLs in young children (i.e., age 4 years). The lower bound of
this mean BLL range was derived from Miranda et al. (2009), who examined the association between
early childhood BLL and academic performance among school-aged (grade 4) children. A recent study of
Canadian preschool children from generally middle- to upper-middle SES families with low Pb exposure
(Desrochers-Couture et al., 2018) did not find an association between concurrent Pb exposure and
performance on the WPPSI at age 3-4 years. Although some individual recent studies found associations
of Pb exposure with cognitive effects in children with mean BLLs <2 (ig/dL (e.g., (Martin et al„ 2021;
Dantzer et al.. 2020; Hong et al„ 2015)). the studies generally involved somewhat older children with
lengthier exposure histories, or employed modeling strategies designed to answer relatively narrow
research questions (e.g., the effect of joint exposure to Pb and other metals or the effect of concurrent Pb
exposure independent from prenatal exposure). Consequently, the studies did not provide evidence that
would change the conclusion of the 2013 Pb ISA that cognitive effects in children are best substantiated
in young children with mean BLLs between 2 and 8 (ig/dL. Studies that might extend the evidence related
to exposure-response relationships (i.e., recent studies that reflect the lower early childhood Pb exposures,
which are now more common in the U.S.]) are limited. Overall, the recently available studies were not
designed, and may not have the sensitivity (Cooper et al., 2016), to detect the effect or hazard at very low
BLLs; however, recent studies generally corroborated the epidemiologic observations of associations
between Pb exposure and IQ in children with relatively low blood Pb concentrations (<5 (.ig/dL).
Consistent with findings from the 2013 Pb ISA, studies do not provide evidence of a threshold for the
effects across the range of BLLs examined. The finding of higher mean IQ with decreasing blood Pb
concentration observed across epidemiologic studies, however, indicates that the absolute magnitude of
the effect of Pb exposure on cognitive function is smaller with decreasing BLL.

Despite limitations, several recent studies describe the cognitive effects over the range of Pb
exposure examined. Shadbcgian et al. (2019) extended the analysis conducted by Miranda et al. (2009)

3-60

-------
also using data from the statewide study of North Carolina school children while focusing on lower BLLs
(<10 (ig/dL) and characterized the persistence of Pb effects across grades. Among children with BLLs of
5 (ig/dL or lower, a decrease in the test score percentile of 0.95 was found for math and a decrease of 1.41
was found for reading when comparing children with a BLL of 5 to those with a BLL <1 (ig/dL. Figure
3-6 depicts the association of blood Pb levels with math and reading test score performance (average
percentile decrement), with 95% CIs, among children across all grades.

Dotted lines denote 95% confidence intervals.
Source: CShadbeqian et al.. 20191.

Figure 3-6 Association of blood Pb level with reading and math scores
among North Carolina school children (average across all
grades). Left panel displays impact of blood Pb level on math test
score. Right panel displays impact of blood Pb level on reading
test score.

Compelling evidence for a larger decrement in cognitive function per unit increase in blood Pb
among children with lower mean blood Pb concentrations, compared with children with higher mean
blood Pb concentrations, was presented in previous assessments (U.S. EPA, 2013, 2006). Individual
studies as well as an international pooled analysis of seven prospective cohort studies by Lanphear et al.
(2019, 2005) that examined prenatal or early childhood BLLs or considered peak BLLs in school-aged
children or concurrent BLLs in young children <3 years old showed greater decrements in cognitive
function per unit increase in BLL among children in lower strata of blood Pb levels compared with
children in higher strata of blood Pb level (Figure 4-15, and Table 4-16 of U.S. EPA (2013) corroborated
the finding of a nonlinear C-R function over the range of the BLLs evaluated (5th to 95th percentile BLL,
2.5 to 33.2 (ig/dL), i.e., a larger incremental effect of Pb exposure on IQ at lower blood Pb concentrations
as indicated by a log-linear C-R function (Crump et al„ 2013). Notably the reanalysis by Crump et al.
(2013) extended the findings of the original study by employing a different modeling strategy.
Specifically, several covariates were defined in a site-specific manner (i.e., HOME score, maternal
education, maternal IQ, ethnicity, maternal alcohol consumption, and maternal smoking), which enabled
finer scale control for potential confounding factors (e.g., amount of alcohol consumed as opposed to a

3-61

-------
binary variable for alcohol consumption). In addition, Crump et al. (2013) used more of the available
blood Pb measurements by computing weighted averages for lifetime and early childhood BLLs. Further,
the authors used formal methods (i.e., nested F-tests and splines) to decide between alternative
specifications of BLL (linear versus log-linear) and chose to add 1 to the BLL prior to log transformation
to ensure it would equal zero when BLL was zero. The key findings of Lanphear et al. (2005), Lanphear
et al. (2019), and Crump et al. (2013) contribute to the strong evidence regarding the effect of low-level
Pb exposure on cognitive function and the supralinear concentration relationship between Pb exposure
and FSIQ. The beta coefficients from the log-linear models, which indicate larger incremental effects of
Pb at lower blood Pb concentration, for (Lanphear et al., 2019, 2005) and Crump et al. (2013) were
comparable (i.e., |3 = -2.65 [95% CI: -3:69, -1:61] per unit of natural log transformed BLL and |3 = -3.32
[95% CI: -4.55, -2.08] per unit of natural log transformed BLL + 1, respectively).

Attenuation of C-R relationships at higher exposure or dose levels has been reported in the
occupational literature. Reasons proposed to explain the attenuation include greater exposure
measurement error and saturation of biological mechanisms at higher levels as well depletion of the pool
of susceptible individuals at higher exposure levels (Stayner et al„ 2003). Possible explanations specific
to nonlinear relationships observed in studies of Pb exposure in children include a lower incremental
effect of Pb due to covarying risk factors such as low SES, poor caregiving environment, and higher
exposure to other environmental factors (Schwartz, 1994a), differential activity of mechanisms at
different exposure levels, and confounding by omitted or mis-specified variables (U.S. EPA, 2013).
Review of the evidence did not reveal a consistent set of covarying risk factors to explain the differences
in blood Pb IQ C-R relationship across high and low Pb exposure groups observed in epidemiologic
studies. Recent studies in populations with mean concentrations of 2 (ig/dL or lower indicated that some
of the observed heterogeneity at lower BLLs may be explained by the underlying distribution in at-risk
factors, including other metals. In addition, some recent studies with similarly low blood Pb
concentrations reported effect modification by sex. These studies are discussed in more detail in Section
3.5.1.6.2.

A limited number of recent studies examine the shape of the C-R function for the relationship
between Pb exposure and cognitive effects in children. Lucchini et al. (2012) conducted a cross-sectional
analysis of children between the ages of 11 and 14 to examine the relationship between concurrent BLL
and FSIQ. The mean BLL was 1.71 (ig/dL in this study. The relationship between BLL and IQ using a
restricted cubic spline fit is plotted in Figure 3-7. As shown in the plot, the decrement in IQ is not
constant over the range in BLLs (0.44-10.2 (.ig/dL). The study was conducted in an area where ferroalloy
plants had operated and the extent to which the children in the study were exposed to higher Pb levels
during early childhood was not clear from this cross-sectional analysis.

3-62

-------
2	4	e	ft	10

Pi> Cuflt'dL}

IQ = intelligence quotient.

Source: Lucchini et al. (2012).

Figure 3-7 Relationship between concurrent blood Pb level and intelligence
quotient among Italian adolescents using a cubic spline fit.

Lucchini et al. (2012) also plotted relationship between the log-transformed BLL (ordinary least
squares fit) and FSIQ (Figure 3-8). A decrement in FSIQ score was observed in association with In
concurrent BLL after adjustment for covariates including SES and maternal education (-2.24 [95% CI:
-4.10, -0.37)]. Consistent with evidence reviewed in the 2013 Pb ISA, the log transformation of BLL
implies a larger incremental decrement in IQ at lower BLLs. Lucchini et al. (2012) calculated the
benchmark dose (BMD) for blood Pb, which is the dose that results in a specific IQ loss (i.e., a loss of one
IQ point), and its lower 95% confidence limit (BMDL) using this C-R function. The BMDL calculated
from these data is 0.11 (ig/dL. As noted previously the older children in this study may have had higher
past Pb exposure that was not reflected in their concurrent BLL.

3-63

-------
130 -
120
110 -

g
75

o	*

100

90 -

80 -

0,5	12	5	10

Pb (na/dL)

IQ = intelligence quotient.

Source: Lucchini et al. (20121.

Figure 3-8 Relationship between log-transformed blood Pb level and
intelligence quotient using an ordinary least squares fit.

Experimental animal studies support the findings in epidemiologic studies indicating the effect of
Pb exposure on cognition at low exposures. Clear support from animal toxicological studies that
demonstrated decrements in learning, memory, and executive function with dietary exposures resulting in
relevant BLLs was assessed in the 2013 Pb ISA. Recent experimental animal studies further support
impairments in cognitive function at BLLs <20 (ig/dL. It is well-documented (and reviewed in the
previous ISA) that Pb exposures resulting in BLLs >20 (ig/dL consistently produced deficits in cognitive
function. Recent evidence (reviewed in the current ISA; see Section 3.5.1.3.2) suggests that BLLs
resulting from lower-level exposures (5-10 (ig/dL) also lead to cognitive function deficits in animal
models. For example, Zhou et al. (2020a); Zhao et al. (2018); Xiao et al. (2014); Betharia and Maher
(2012); Corv-Slechta et al. (2012) all reported significant cognitive deficits following exposures yielding
BLLs <10 (ig/dL. Betharia and Maher (2012) reported the lowest BLLs for animals tested in the Morris

3-64

-------
water maze (0.98 (ig/dL at PND 21 and later 0.03 (ig/dL at PND 56) which produced mild effects on
memory in females at the later time point only.

The reader should be aware that BLLs reported in many of these studies were measured later
during experimentation and do not necessarily reflect peak Pb burden or Pb burden during the most
sensitive window of brain development for young animals and children; BLLs in toxicological studies
should be interpreted in the context of both exposure time course and blood collection. Discussion of
BLLs and cognitive function is further complicated by the exposure window. The evidence generally
supports the notion that Pb exposures during brain development led to changes in cognition that persist
after cessation of exposure and BLLs have decreased. For example, Barkur and Bairy (2015b) compared
multiple different windows of exposure and reported the greatest decreases in learning and memory
following gestational and lactational windows, consistent with the altricial nature of rodent brain
development.

While nonlinear C-R relationships including U- or inverted U-shaped curves for various
endpoints, including those related to cognitive impairment, were demonstrated in the toxicological
literature discussed in the previous ISA, these toxicological findings are distinct from epidemiologic
findings of supralinear relationships in that some U- or inverted U-shaped relationships do not indicate
Pb-induced impairments at higher exposure concentrations (U.S. EPA, 2013). Recent animal studies do
not provide evidence for an inverted dose relationship, rather increased Pb doses (and resulting BLLs)
generally resulted in greater cognitive impairment. This may be related to the refined PECOS used in the
current ISA, which did not incorporate studies reporting BLLs higher than 30 (ig/dL, thus narrowing the
range of doses integrated. Thus, recent evidence generally supports dose-dependent effects of Pb on
cognitive function at relevant BLLs.

In summary, recent studies support and extend the evidence pertaining to the effect of Pb
exposure on cognitive function in children at low BLLs. These effects are best substantiated to occur in
study populations with mean BLLs between 2 and 8 (ig/dL. Association between Pb exposure in
populations of children below 2 (ig/dL are reported, extending the evidence described in the 2013 Pb ISA;
however, heterogeneity at lower exposure levels (i.e., not all studies report positive associations) has been
observed. Recent experimental studies of rodents continue to support impairments in cognitive function at
BLLs <30 (ig/dL. Compelling evidence for a larger decrement in cognitive function per unit increase in
blood Pb among children with lower mean blood Pb concentrations, compared with children with higher
mean blood Pb concentrations, across a broad range of BLLs (e.g., 5th percentile of 2.5 (ig/dL up to 95th
percentile of 33 (ig/dL) was supported by a reanalysis of a pooled international dataset Crump, 2013,
3838553}. Recent studies with an adequate range of Pb exposure measured during relevant time periods
that would be required to evaluate exposure-response relationships are generally lacking. Considering the
collective body of studies, no evidence of a threshold for cognitive effects in children across the range of
BLLs examined in epidemiologic studies was reported.

3-65

-------
3.5.1.6.2 Confounding

The 2013 Pb ISA described multiple factors that influence cognitive function and behavior in
children including parental IQ and education, SES of the family, quality of the caregiving environment
(i.e., HOME score), and other environmental exposures (U.S. EPA, 2013; Wasserman and Factor-Litvak,
2001). These other risk factors often are correlated with blood, tooth, and bone Pb levels, and thus, are
considered as potential confounding factors in epidemiologic analyses. The collective epidemiologic
evidence consistently demonstrates associations of higher blood and tooth Pb levels with cognitive
function decrements and poorer behavior in children. These associations were observed in diverse
populations in the U.S., Mexico, Europe, Asia, and Australia. Associations have been observed across
studies that used different methods to control for confounding and adjusted for different potential
confounding factors, commonly maternal IQ and education, SES, and HOME score. Several studies have
found associations with additional adjustments for smoking exposure, birth outcomes, and nutritional
factors. Multiple recent studies adjusted for exposure to other metals or environmental chemicals e.g.,
Zhou et al. (2020b) and Liu et al. (2015); however, there remains uncertainty regarding the
appropriateness of the adjustment for other metals as confounders in some studies that did not examine
the potential for interactions (see Section 3.5.1.6.5 for evidence related to interactions between Pb and
other metals).

As noted in the 2013 Pb ISA, no single method to control for potential confounding is without
limitation, and there is potential for residual confounding by unmeasured factors. However, the
consistency of findings among different populations and study methods with consideration of several well
characterized potential confounding factors as described above increases confidence that the associations
observed between Pb biomarker levels and neurodevelopmental effects in children represented a
relationship with Pb exposure. Recent studies expanded the evidence reporting associations between Pb
exposure and nervous system effects in children after consideration of covariates including sex, maternal
stress and race or ethnicity as potential effect modifiers as opposed to confounders (see Section 3.5.1.6.5).
Biological plausibility was derived from extensive evidence provided by animal toxicological studies that
are experimental in design and thus, not vulnerable to confounding. These experimental animal studies
demonstrate the effect of Pb on cognition and behavior as well as changes in neurogenesis, synaptic
pruning, and neurotransmitter function in the hippocampus, prefrontal cortex, and nucleus accumbens of
the brain (U.S. EPA, 2013). Recent experimental animal studies support the evidence described in the
2013 Pb ISA, provide additional evidence for Pb-induced impairments in learning and memory (short and
long-term) assessed by several methods not discussed in the 2013 Pb ISA, and extend the limited
evidence related to Pb-induced impairment of executive functions. These experimental animal studies
provide strong support that the effects observed in epidemiologic studies cannot be explained by
confounding.

3-66

-------
3.5.1.6.3 Lifestages

Epidemiologic studies reviewed in the 2013 Pb ISA consistently showed that BLLs measured
during various lifestages and time periods (i.e., prenatal, early childhood, childhood average, and
concurrent with the outcome) were associated with cognitive function decrements in children (U.S. EPA,
2013). Epidemiologic studies consistently pointed to inverse associations between FSIQ in school-aged
children and BLLs measured at various lifestages and time periods (Table 4-14 U.S. EPA (2013)). In an
analysis of data from seven prospective studies Lanphear et al. (2019) found that increases in early
childhood (age 6-24 months on average), peak, concurrent, and lifetime average BLLs were associated
with decreases in FSIQ in children at ages 4-10 years. The investigators reported that the best predictor of
IQ decrement, as indicated by the model R2 value, was early childhood blood Pb concentration (R2 =
0.6433), although the R2 value for the concurrent metric (0.6414) was nearly identical (Lanphear et al„
2019; Crump et al., 2013). These results illustrated the challenge of distinguishing a critical time period
when exposures are highly temporally correlated. Epidemiologic studies that aimed to improve the
characterization of important lifestages and time periods of Pb exposure by examining children in whom
BLLs were not strongly correlated with exposure over time indicated FSIQ decrements in association
with higher concurrent BLLs but did not conclusively demonstrate stronger findings for early versus
concurrent BLLs (Table 4-15 of U.S. EPA (2013)). Considering the collective body of epidemiologic
evidence reviewed in the 2013 Pb ISA, there was no clear indication of a single critical lifestage or
duration of Pb exposure that is uniquely associated with the risk of neurodevelopmental effects in
children. These observations in the epidemiologic literature were supported by experimental animal
evidence. Consistent with findings from the 2013 Pb ISA, more recent studies continue to report
associations with prenatal BLLs (maternal and cord blood Pb) and postnatal BLLs measured at various
childhood lifestages despite some heterogeneity in the magnitude and direction of the associations at
BLLs <5 (ig/dL.

Maternal Pb exposure presents an exposure risk during gestation and early infancy, when
important neurodevelopmental processes are known to occur. Substantial fetal Pb exposure may occur
from mobilization of maternal skeletal Pb stores (Gulson et al., 2003; Hu and Hernandez-Avila, 2002)
and its transfer across the placenta (Section 3.2.2.4 of U.S. EPA (2013)). Among studies that examined
BLLs at multiple time periods, some found a larger decrement in MDI per unit increase in prenatal blood
Pb than concurrent blood Pb ((Hu et al„ 2006; Gomaa et al„ 2002), Table 4-14 of U.S. EPA (2013)).
Prenatal and early postnatal (age 6 months) BLLs were also associated with cognitive function in studies
that included school-aged children (ages 5-17 years) (Table 4-14 of U.S. EPA (2013)). Sanchez et al.
(2011) extended the analysis of Hu et al. (2006), which was designed to elucidate the time window during
pregnancy that the effect of Pb exposure on neurodevelopment is most pronounced among participants in
a birth cohort study in Mexico City. These authors compared methods to model exposure and found that
the MDI score at age 2 was sensitive to the choice of method. A decrease in MDI score of 2.74 (95% CI -
5.78 to 0.29) per natural log increase in BLL during the first trimester was observed, using a window-

3-67

-------
specific regression, while the corresponding decrease was larger [|3=—4.13 (95% CI, -7.54, -0.72) using a
multiple informant model.

As described above, however, most of these studies also found cognitive function decrements in
association with postnatal BLLs, and the results did not identify an individual critical postnatal time
period of blood Pb measurement associated with cognitive function decrements. Maternal pregnancy-cord
BLL correlations of 0.53-0.81, depending on the stage of pregnancy, were reported by Schell et al.
(2003). Depending on the magnitude of child exposure, the contribution of maternal blood Pb to child
BLLs appears to diminish rapidly over a period of a few months following birth, after which child BLLs
may be influenced mainly by postnatal Pb exposures (Section 3.4.1 of U.S. EPA (2013)).

Recent studies observed associations between Pb exposure during prenatal and childhood
lifestages (i.e., maternal (Y Ortiz et al„ 2017; Vigeh et al., 2014; Kim et al„ 2013b, c), cord (Valeri et al„
2017), and postnatal exposure (Lin et al., 2013)) and poorer performance on tests of neurodevelopment
among mothers and infants with mean BLLs <5 (ig/dL. The is some evidence indicating that there
heterogeneity in the magnitude and direction of the observed associations in recent studies may be
explained, in part, by co-exposure to other metals or maternal stress

Experimental animal studies demonstrated that prenatal or early postnatal or lifetime Pb exposure
alters brain development via changes in synaptic architecture and neuronal outgrowth, leading to
impairments in memory and learning (Sections 4.3.10.4, 4.3.10.10, and 4.3.2.3 of U.S. EPA (2013)).
Gestational or infancy Pb exposures are not necessary to induce cognitive function decrements in juvenile
animals, however. Studies of monkeys have found that Pb exposures during lifestages and time periods
extending from infancy through the juvenile or adult periods resulted in impaired cognitive function
(Rice, 1992; Rice and Gilbert, 1990a; Rice, 1990; Rice and Karpinski, 1988). These findings are
consistent with studies of individuals aged 3 to 30 years, which showed that brain development
ascertained using MRI continues throughout adolescence, indicating the potential for alterations to
neurodevelopment later in childhood (Gerber et al., 2009; Lenroot and Giedd, 2006).

Additional recent animal studies support the notion that various exposure periods (i.e.,
preconception, gestation, lactation) may represent a critical periods during which Pb exposure can cause
cognitive impairment later in life. In rodents, developmental exposure to Pb was consistently associated
with persistent cognitive effects observed both early (Tartaglionc et al., 2020; Zhao et al„ 2018; Barkur
and Bairy, 2015b; Anderson et al., 2012) and later in life (Liu et al., 2022c; Xiao et al„ 2014; Betharia
and Maher, 2012). Few studies were designed to compare exposures across multiple different
developmental windows (Barkur and Bairy, 2015b; Xiao et al., 2014). These studies reported similar
magnitudes of effects between developmental windows, suggesting that individual periods of
development may be similarly sensitive to Pb. Generally, longer exposures that spanned multiple
developmental periods (e.g., preconception through lactation) produced not only the highest BLLs but the
largest effects on cognition (Zhou et al„ 2020a; Zhu et al„ 2019b).

3-68

-------
Unlike other organ systems, the unidirectional nature of CNS development limits the capability of
the developing brain to compensate for cell loss, and environmentally induced cell death can result in a
permanent reduction in cell numbers (Bayer, 1989). Hence, when normal development is altered, the early
effects have the potential to persist into adult life even in the absence of concurrent exposure, magnifying
the potential public health effects. A limited number of studies examined the persistence of the effects of
Pb on cognitive function. A recent study by Shadbegian et al. (2019) indicated that poorer performance on
tests of reading and math associated with earlier childhood Pb exposure persisted from grade 3 to grade 8.
Some epidemiologic evidence reviewed in the 2013 Pb ISA indicated associations of earlier childhood
blood or tooth Pb levels in adolescents or adults with decreased cognitive function (Mazumdar et al.,
2011; Ris et al., 2004; Stiles and Bellinger, 1993). Recent studies support and extend this evidence.
Specifically, childhood Pb exposure was observed to have long-term cognitive consequences in young-
(18-19 years) (SkerfVing et al., 2015) or mid-adulthood (38 or 45 years of age) (Reuben et al., 2020;
Reuben et al„ 2017). These epidemiologic studies did not examine adult BLLs, thus the relative influence
of adult Pb exposure was not ascertained. The persistence of effects of early exposures, however, is
supported by findings of impaired learning in adult monkeys exposed to Pb only during infancy (Rice,
1992; Rice and Gilbert, 1990a; Rice, 1990). Additional recent studies in rodents provided support for the
persistence of effects of early exposures of Pb (Xiao et al., 2020; Li et al., 2016a; Xiao et al., 2014; Zhang
et al., 2014; Rahman et al., 2012b; Zhang et al„ 2012; Kuhlmann et al., 1997).

There is some evidence that the effects of early Pb exposure on cognitive function are not fixed.
Results indicated higher cognitive function in children at ages 1-8 years who had declines in BLL over
durations of 6 months to 5 years compared with children with smaller declines, no change, or increases in
BLLs in some studies (Hornung et al„ 2009; Chen et al., 2005; Liu et al., 2002; Ruff et al„ 1993;

Bellinger et al„ 1990). This evidence pertains to populations with declines from higher BLLs at baseline
(20-55 (ig/dL) or larger declines over time (i.e., 8, 14 (ig/dL) than those expected for most of the current
population of U.S. children. No recent studies that provided additional information on this topic were
identified.

To conclude, the collective body of epidemiologic evidence reviewed in the 2013 Pb ISA did not
provide strong evidence to identify an individual critical lifestage or timing of Pb exposure with regard to
neurodevelopmental effects in children (U.S. EPA, 2013). Recent studies support this conclusion.
Evidence indicates that prenatal BLLs are associated with mental development in very young children
aged <2 years. Several studies indicated that increases in postnatal (earlier childhood, lifetime average,
concurrent) BLLs were associated with larger cognitive function decrements in children aged 4-10 years
than were similarly sized increases in prenatal BLLs. These results suggest that per unit increase,
postnatal Pb exposures that are reflected in concurrent or cumulative BLLs or tooth Pb levels may have a
larger magnitude of effect on cognitive function decrements as children age (U.S. EPA, 2013). The
identification of critical lifestages and time periods of Pb exposure is complicated by the fact that BLLs in
older children, although affected by recent exposure, are also influenced by Pb stored in their bone and
maternal Pb stores. Thus, associations of neurodevelopmental effects with concurrent BLL in children

3-69

-------
may reflect the effects of past and recent Pb exposures. Nonetheless, the epidemiologic evidence for
associations of neurodevelopmental effects with multiple lifestages or time periods of Pb exposure,
including more recent exposures, is supported by evidence in monkeys that Pb exposures in infancy,
lifetime exposure starting from birth, or lifetime exposure starting during the juvenile period induce
impairments in cognitive function when assessed between the ages of 6 and 10 years.

3.5.1.6.4 Public Health Significance

The 2006 Pb AQCD and the 2013 Pb ISA (U.S. EPA. 2013. 2006) concluded that
neurodevelopmental effects in children were among the effects best substantiated as occurring at the
lowest BLLs. Evidence from several cohorts of children indicated that there is a supralinear C-R
relationship between blood Pb and FSIQ (i.e., larger incremental effect of blood Pb on FSIQ at lower
levels) and no threshold was identified for Pb-associated neurodevelopmental effects in the range of BLLs
examined. The evidence reviewed in the current assessment supports these conclusions and continues to
clearly indicate that neurodevelopmental effects in children are among the greatest public health concern
associated with Pb exposure.

Cognitive function in children has been assessed using a variety of tests, including FSIQ, BSID,
academic performance, and academic achievement. As noted in the 2013 Pb ISA (U.S. EPA, 2013), FSIQ
has strong psychometric properties (i.e., reliability, consistency, validity), is among the most rigorously
standardized cognitive function measures, is relatively stable in school-age children, and has been
predictive of educational achievement and life success. Variation in IQ score across different populations,
however, may be influenced by differential access to resources in those populations
(Shuttleworth-Edwards, 2016; Marks, 2010). In children aged 6 months to 3 years, the BSID is commonly
used to assess mental development; however, the BSID MDI is not an intelligence test and MDI scores
are not necessarily strongly correlated with later measurements of FSIQ in children with normal
development. Lower FSIQ is also linked to poorer academic performance and achievement, both of which
have important implications for success later in life including reduced earning potential and productivity
(Lin et al., 2016; U.S. EPA, 2013; Salkever, 1995; Schwartz, 1994b). Analyses of end-of-grade tests from
North Carolina indicated that early childhood BLL is associated with reduced performance on the tests,
the cumulative effect of Pb and low SES is more pronounced at the lower end of the test score distribution
(Miranda et al„ 2009), and the effects of Pb exposure persisted from grade 3 to grade 8
(Shadbegian et al„ 2019). Tests of academic achievement generally measure a child's understanding of a
given curriculum that is developed and implemented through the school system; thus, because exams are
typically specific to each state, data cannot be directly compared across states.

The World Health Organization (WHO) definition of "health" is "the state of complete physical,
mental, and social well-being and not merely the absence of disease or infirmity" (WHO, 1948). By this
definition, decrements in health status that are not severe enough to result in the assignment of a clinical
diagnosis might reflect a decrement in the well-being of an individual. Further, deficits in subtle indices

3-70

-------
of health or well-being may not be observable except in aggregate, at the population level; therefore, a
critical distinction between population and individual risk is essential for interpreting the public health
significance of study findings. This concept of population risk is relevant to the interpretation of findings
regarding IQ in the assessment of their public health significance. Specifically, Weiss (1988) discussed
the hypothetical effects of a small shift in the population distribution of IQ score. As shown in Figure 3-9,
these authors anticipate that even a small shift in the population mean IQ may be significant from a public
health perspective because such a shift, given a normal distribution, could yield a larger proportion of
individuals functioning in the low range of the IQ distribution, which is associated with increased risk of
educational, vocational, and social failure (Section 4.3.13), as well as reduce the proportion of individuals
with high IQ scores. Although the change in population mean IQ score may be small relative to the
standard error for the IQ measurement, a study that is large enough will have adequate statistical power to
detect small changes at the population level. Bias may be introduced if the measurement error of the
outcome is highly correlated with the exposure, but there is no evidence to suggest that individuals with
higher BLLs test systematically lower than their true IQ.

IQ = intelligence quotient.

Note: Two distributions of intelligence test scores. (Left): Based on a mean of 100 (the standardized average, with SD of 15).
(Right): Demonstrating a 5% reduction model, based on a mean score of 95. This is a conceptual model that assumes that the
incremental C-R between Pb exposure and IQ is similar across the full range of IQ and is not based on actual data. The figure
shows that the effect of a small shift in population mean IQ score may result in a larger proportion of individuals with IQ scores
below 70 and a smaller proportion with IQ scores above 130.

Source: Reproduced with permission of Elsevier; from Weiss (19881.

Figure 3-9 Two distributions of intelligence test scores demonstrating the
consequence in a small shift in the mean score.

3.5.1.6.5 Potentially At-Risk Populations

The 2013 Pb ISA described physiologic factors that influence the internal distribution of Pb (U.S.
EPA, 2013). Blood and bone Pb measurements are influenced to varying degrees by biokinetic processes
including absorption, distribution, metabolism, and excretion. These processes are affected by age,

3-71

-------
genetics, diet, and co-exposures to other metals and chemicals, which are summarized in the Executive
Summary and Integrated Synthesis (https://asscssmcnts.cpa.gov/isa/documcnt/&dcid=359536). In
addition to these physiological factors, several population characteristics that explain differential Pb
exposure have been identified. These factors included age, sex, race and ethnicity, proximity to Pb
sources, and residential sources and are also discussed in the Executive Summary and Integrated
Synthesis (https://asscssmcnts.cpa.go\7isa/documcnt/&dcid=359536). The factors potentially related to
increased risk of Pb-induced cognitive effects (i.e., factors identified in epidemiologic studies that
conduct stratified analyses and compare the magnitude of the observed association across stratum) are
discussed below.

Age

This section on Potentially At-Risk Populations emphasizes stratified results described in some
epidemiologic studies, as opposed to the large body of longitudinal studies following mothers and infants
throughout childhood that comprises the most compelling body of evidence in support of conclusions
regarding childhood as an "at-risk" factor. As noted in previous sections of the document (i.e., Section
3.5.1.1 and 3.5.1.6.1), recent evidence supports the finding from the 2013 Pb ISA that cognitive effects in
young children is the outcome that best substantiated to occur at the lowest exposure levels. Strong
evidence indicates increased risk of Pb-induced neurocognitive effects during several childhood lifestages
throughout gestation, childhood, and into adolescence (see Section 3.5.1.6.3). Moreover, the integrated
synthesis (Section 7.4.2.2) of this document concludes that, "In consideration of the evidence base (e.g.,
stratified and longitudinal analyses) and integrating across disciplines of toxicokinetics, exposure, and
health, there is adequate evidence to conclude that children are an at-risk population."

Sex

Multiple epidemiologic studies included in the 2013 Pb ISA examined Pb-related effects on
cognition separately in males and females. Studies on cognition from the CLS cohort and a study in
Poland reported larger magnitude Pb-associated cognitive effects in males (Jedrvchowski et al.. 2009a;
Ris et al.. 2004; Dietrich et al.. 1987). whereas studies from Australia indicated that females were at
increased risk of Pb-associated cognitive effects (Tong et al.. 2000; Baghurst et al.. 1992; Mcmichael et
al.. 1992). While toxicological evidence supporting sex-specific effects of Pb on cognitive function was
summarized in the previous ISA (Virgolini et al.. 2008; Yang et al.. 2003; Mcgivern et al.. 1991). the
number of studies considering sex as a factor was limited. Recent evidence provides more support for the
sex-biased effects. One study reported a male-specific effect (Anderson et al.. 2016) and several studies
demonstrated female-specific effects (Tartaglione et al.. 2020; Verma and Schneider. 2017; Anderson et
al.. 2012; Betharia and Maher. 2012). The evidence supports a conclusion that there are sex-related
differences in the effects of Pb on cognitive function, yet it remains difficult to parse the exact nature and

3-72

-------
direction of sex-specific effects given the variation in outcomes examined, exposure timing and the
considerable number of studies that only reported data from one sex at a time.

Several recent epidemiologic studies examined sex-stratified associations of Pb exposure with
cognitive effects (Tatsuta et al.. 2020; Zhou et al.. 2020b; Desrochers-Couture et al.. 2018; Taylor et al..
2017). Tatsuta et al. (2020) found a decrement in FSIQ score in association with postnatal BLL (|3 =
-9.880 [95% CI: -2.905, 5.831]) among boys, with a smaller less precise association among girls (|3 =
-4.406 [95% CI: -15.94, 7.129]). Prenatal BLL was associated with a smaller and less precise decrease in
FSIQ score among boys (|3 = --3.683 [95% CI: -10.714, 3.349]) but not among girls (|3 = 1.463 [95% CI:
-2.905, 5.831]). A lower BNT score (with cues) was associated with both pre- and postnatal BLL among
boys in this study. Desrochers-Couture et al. (2018) studied the association between cord, maternal and
childhood (3-4 years old) BLLs with cognitive function (WPPSI-III at age 3-4 years) among preschool
aged children in Canada. An association was observed between cord BLL and performance IQ in boys (|3
= -3.28 [95% CI: -5.31, -1.18] per doubling) that was not present in girls (|3 = 0.16 [95% CI: -1.76,
2.06] per doubling). Although associations were imprecise, Taylor et al. (2017) found an association
between increased maternal BLL and IQ decrements in boys but not in girls enrolled in the ALSPAC
study (e.g., -0.29 [95% CI: -1.02, 0.44] versus 0.73 [95% CI: 0.39, 1.33], respectively on the WISC).
Using models that adjusted for co-exposure to metals (Mn and Cd), Zhou et al. (2020b) found no
association between cord BLL and FSIQ in boys or girls. Further, using models that adjusted for other
chemicals (i.e., DDE, HCB, PCBs, and Mn), Oppenheimer et al. (2022) found no statistical evidence of
an interaction between prenatal Pb exposure and sex.

Maternal Stress

Toxicological studies assessed in the 2013 Pb ISA demonstrated that early life exposure to Pb and
maternal stress can result in dysfunction of the HPA axis (U.S. EPA, 2013). Recent toxicological evidence
provides further support for this interaction between maternal stress and Pb exposure during development.
Anderson et al. (2012) demonstrated that exposure to Pb blunted the positive effects of environmental
enrichment on learning in the Morris water maze paradigm. Interestingly, Cory-Slechta et al. (2012)
reported that contrary to the effect previously reported in females, maternal stress improved the
performance of Pb-exposed male offspring in a repeated performance and learning paradigm compared
with Pb-exposed males without maternal stress. While this study supports the interaction between
maternal stress and Pb exposure, it remains unclear whether stress would positively influence other facets
of cognitive function.

Recent epidemiologic studies examined maternal stress as a modifier of the association between
Pb exposure and neurodevelopment. Y Ortiz et al. (2017) used the CRISYS-R questionnaire, which
assesses negative life events across several domains (i.e., financial, legal, career, relationships,
community and home violence, medical problems, other home issues, discrimination or prejudice, and
difficulty with authority) to examine this effect. Third trimester maternal BLL was associated with the

3-73

-------
cognitive component of the BSID in this study and a weak interaction (i.e., lower cognitive scores as BLL
and stress increase) between log-transformed maternal blood Pb and stress was observed (|3 = 1.02 [95%
CI: -0.78, 2.82]). In another study, Zhou et al. (2017) assessed mother-child pairs from the Shanghai
Stress Birth Cohort. Maternal whole blood and maternal prenatal stress levels were assessed at 28-36
weeks of gestation, and the GDS adapted for a Chinese population were administered to children 24-36
months old in the study. No association between prenatal maternal BLL and child cognitive development
was observed; however, an interaction effect was observed such that high maternal stress appeared to
exacerbate the effect of prenatal Pb exposure in several domains, including language (|3 = —33.82 [95%
CI: -60.04, -7.59] per log-10 transformed unit of BLL), while low maternal stress did not (|3 = -1.76
[95% CI: -13.03, 9.51] per log-10 transformed unit of BLL, p-interaction = 0.02).

Other Metal Exposure (Cd, Mn, Hg, As)

A limited number of studies included in the 2013 Pb ISA examined the modification of
association between Pb exposure and cognitive function by other metals (U.S. EPA, 2013). Larger Pb-
associated decrements in IQ (Kim et al., 2009) and neurodevelopment (Henn et al„ 2012) were observed
in children with higher Mn levels. Henn et al. (2012) also observed an interaction between the highest
quintile of Mn and BLL at 12 months (Figure 3-10).

Several recent epidemiologic studies of the association of Pb exposure with FSIQ examined
interactions between Pb exposure and other metals or modification of the Pb-FSIQ association by other
metals. For example, some cross-sectional analyses found evidence that coexposure to Mn may heighten
the effect of Pb in some populations (Martin et al., 2021; Menezes-Filho et al„ 2018), while another study
found no interaction between Pb exposure and Mn (or ALAD) (Lucchini et al., 2012). Several studies of
the association between Pb exposure and infant development also point to possible interactions with other
metals. Lin et al. (2013) observed an interaction with Mn such that children who were highly exposed to
both Mn and Pb had larger neurodevelopmental deficits compared with those with low exposure to just
one or both these metals. Kim et al. (2013b, 2013c) observed a larger decrement in MDI in association
with late pregnancy maternal BLL among those with Cd levels above the median (|3 = -3.20 [95% CI:
-5.35, -1.06]) compared with the decrement among those with Cd levels below the median (|3 = -0.29
[95% CI: -2.88, 2.30]). In contrast to the findings of Henn et al. (2012). Valeri et al. (2017) observed an
association between increasing cord Pb level and neurodevelopmental decrements in children with lower
cord blood Mn and As (|3 = -0.01 [95% CI: -0.02, 0.00]), but not in the group with higher concentrations
of these metals (or metalloids) (|3 = 0.01 [95% CI: -0.05, 0.07]). Nvanzaet al. (2021) also observed
modification of maternal blood Pb and global neurodevelopmental status by blood Hg concentrations.

3-74

-------
BPb = blood Pb; MDI = Mental Developmental Index; Mn = manganese.
Source: Henn et al. (20121.

Figure 3-10 Scatter plots and regression lines of blood Pb level and 18-month
Mental Developmental Index among children in manganese (A)
quintiles 1-4 and (B) quintile 5.

Only one recent animal study incorporated combined exposure to Pb and Mn at relevant levels
(<30 (ig/dL Pb). Betharia and Maher (2012) reported that both Pb and Mn individually impaired memory
in the Morris water maze, but the effects of the mixture were not significantly different from those of the
control. Interestingly, during the learning (acquisition) phase only, the mixture enhanced performance,
suggesting a possible antagonistic effect of these two metals on the development of spatial learning
processes. Given the lack of evidence available on combined metal exposures, the possible interaction
between multiple metals at relevant levels in animals remains unclear.

Many recent toxicological studies provided evidence for the interaction of Pb and other metals
(Mn, Cd, Ar, Hg, Fe) but were not PECOS-relevant (e.g., in vitro studies, high levels, non-mammalian
models) and are summarized in the biological plausibility Section 3.3

Socioeconomic Status

SES has been examined as an effect modifier in multiple studies of Pb-induced cognitive effects
(U.S. EPA, 2013, 2006). Larger blood Pb-associated decreases in cognitive function were found with
lower SES in several studies (Ris et al., 2004; Tong et al„ 2000; Bellinger et al., 1990). In contrast, a
meta-analysis of eight studies found a smaller decrement in FSIQ for studies in disadvantaged
populations than for studies in advantaged populations (Schwartz, 1994a). While the results indicate that
BLL is associated with FSIQ deficits in both higher and lower sociodemographic groups, they do not

3-75

-------
clearly indicate whether groups with different SES differ in Pb-related changes for cognitive function
(Murphy et al., 2013).

No recent epidemiologic studies examined SES as a modifier of the association between Pb
exposure and cognitive effects in children.

Race/Ethnicity

The evidence reviewed in the 2013 Pb ISA pertaining to the modification of the effect of Pb
exposure on cognitive function in children by race or ethnicity was limited to one study (U.S. EPA,
2013). Miranda et al. (2007) presented data indicating that the association between early childhood
exposure to Pb and declines in reading and mathematics scores was similar between Black and white
children. In a recent study, Braun et al. (2018) examined the effects of residential exposure interventions
on dust Pb loadings, BLL, and neurodevelopmental outcomes in children (4 to 8 years old). Although no
intervention effect on BLL was found, overall, the geometric mean childhood BLLs for children 1 to 8
years old was lower in non-Hispanic Black children (See Appendix 2: https://assessments.epa.gov/isa/
document/&deid=359536). No intervention effect on other neurodevelopmental outcomes (e.g., FSIQ,
BSID, BRIEF) were observed.

Pre-existing Disease

Studies that examined the effect of Pb exposure on cognitive function in children across strata
defined by pre-existing disease status were not reviewed in previous assessments (U.S. EPA, 2013,
2006).

A recent study examined the association of Pb exposure with IQ and executive functioning using
BRIEF among children with CKD (Ruebner et al., 2019). Concurrent BLL assessment was associated
with FSIQ decrement in adjusted models (|3 = -2.1 [95% CI: -3.9, -0.2] per 1 (ig/dL increase in BLL).
Associations between BLL and behavioral symptoms indicating executive function problems did not
persist in models that controlled for potential confounders including race, poverty, maternal education,
and clinical factors related to CKD.

Nutritional Factors

The 2006 Pb AQCD included studies that indicated individuals with Fe deficiency and
malnourishment had greater inverse associations between Pb and cognition (U.S. EPA, 2006);
nutritional factors were not examined as effect modifiers of the association between Pb exposure and
cognitive effects in children in more recent studies reviewed in the 2013 Pb ISA (U.S. EPA, 2013).

3-76

-------
No recent epidemiologic studies were available to inform this topic. Recent toxicological studies
(that were PECOS-relevant) investigated the influence of different dietary factors on the effects of Pb. Liu
et al. (2022c) exposed rats to 0.2% Pb in drinking water in combination with either standard rodent chow
or a high-fat diet and then assessed cognitive function using the Morris water maze paradigm. High-fat
diet increased the BLL of rats compared with Pb-exposed rats maintained on standard chow. High-fat diet
enhanced the effect of Pb on learning during the acquisition phase compared with the Pb-control diet
group. During the probe trial, Pb-exposed animals (both diets) had significantly fewer crossings into the
target quadrant compared with untreated animals. Interestingly, a similar memory impairment was
observed in the non-Pb + high-fat diet group, suggesting a role for high-fat diets in cognitive impairment
independent of Pb exposure. The contribution of Pb versus high-fat diet remains unclear based on this one
study.

Al-Qahtani et al. (2022) supplemented Pb exposure in mice with green tea extract and reported
that green tea ameliorated the negative effects of Pb exposure on both learning and memory assessed in
an active avoidance paradigm. Additionally, Long et al. (2022) reported that probiotic supplementation
(Limosilactobacilitis fermentum) in Pb-exposed rats partially mitigated the cognitive deficits observed in
an active avoidance paradigm. These studies support a role for dietary factors in the neurotoxicity of Pb
but the diversity of nutritional factors investigated and the small number of studies make it difficult to
determine their importance.

Genetics

Polymorphisms in certain genes have been implicated in the absorption, retention, and
toxicokinetics of Pb in humans (U.S. EPA, 2013, 2006). Studies assessed in the 2013 Pb ISA indicated
that the presence of ALAD variants was associated with an increase in Pb-related cognitive effects in
adults, but there was limited information for children. In studies of children, inverse associations with
poorer rule learning and reversal, spatial span, and planning were exacerbated among those lacking the
DRD4 gene (Froehlich et al., 2007). Two additional studies found no evidence that the
methylenetetrahydrofolate reductase 677T allele or variants of the DRD2 or dopamine transporter
(DAT1) genes modified the effect of Pb on neurodevelopment (Kordas et al„ 2011; Pilsner et al„ 2010).

Several recent studies add to the limited body of evidence in children. Bah et al. (2022) found that
the effect of low Pb exposure on children's IQ was less among those with the ALAD1 genotype. Rooney
et al. (2018) found interaction effects between variants of glutamate ionotropic receptor NMDA-type
subunits 2A and 2B (GRIN2A and GRIN2B) and Pb exposure on performance on tests of learning,
memory, and executive function, which were more pronounced in boys. Kordas et al. (2011) found that
children with the DRD2 TT genotype (variant) scored higher than children with CC genotype (wild type)
on the Bayley MDI and McCarthy memory scale. However, the variants did not modify the relationship
between BLLs and MDI or McCarthy memory scale scores. Bozack et al. (2021) found that prenatal Pb
exposure was associated with DNA methylation in regions annotated to genes involved in

3-77

-------
neurodevelopment. Overall, the evidence pertaining to interactions between genes and Pb exposure in
children remains limited.

Other Factors

No studies that examined maternal smoking as a modifier of the association between Pb exposure
and cognitive effects were included in previous assessments (U.S. EPA, 2013, 2006). Recent studies did
not examine maternal smoking as a modifier of the association between Pb exposure and cognitive
effects in children.

BMI was not examined as an effect modifier of the association between Pb exposure and
cognitive effects in children in studies reviewed in the 2013 Pb ISA (U.S. EPA, 2013). No recent studies
have examined this factor as an effect modifier.

Maternal self-esteem modified the association between BLL and infant development in Surkan et
al. (2008), which was assessed in the 2013 Pb ISA. No recent epidemiologic studies were available to
inform this topic.

Cognitive reserve was not examined as an effect modifier of the association between Pb exposure
and cognitive effects in children in studies reviewed in the 2013 Pb ISA (U.S. EPA, 2013). No recent
epidemiologic studies were available to inform this topic.

3.5.1.7 Summary and Causality Determination: Cognitive Effects in Children

The evidence from epidemiologic and experimental evidence that supports the causality
determination for cognitive effects in children is outlined in Table 3-2. Overall, recent evidence
supports the conclusion from the 2013 Pb ISA that there is a causal relationship between Pb
exposure and cognitive effects in children.

Studies evaluated in the 2013 Pb ISA found a consistent pattern of associations between higher
BLLs and lower FSIQ in children aged 4-17 years (see Figure 4-2 and Table 4-3 (U.S. EPA, 2013)). The
strongest evidence was provided by prospective studies with analyses of the association of blood Pb
measured in early childhood before FSIQ was assessed or with tooth Pb levels typically measured in
dentin reflecting prenatal or early childhood Pb exposure. These prospective studies typically considered
potential confounding by maternal IQ and education, SES, birth weight, smoking exposure, parental
caregiving quality, and in a few cases, other birth outcomes and nutritional factors. Associations were
found in diverse populations (e.g., Boston, MA; Cincinnati, OH; Rochester, NY; Cleveland, OH; Mexico
City, Mexico; Port Pirie, Australia; and Kosovo, Yugoslavia) in studies that examined children recruited
from prenatal clinics, hospital maternity departments, or schools. Studies generally reported high follow-
up participation supported by evidence that selection bias did not explain the associations observed.

3-78

-------
Multiple recent longitudinal studies of children with mean BLLs <5 (ig/dL add to the evidence
informing the relationship between BLL and IQ in children. Heterogeneity in the magnitude and direction
of the associations was present across these studies, however. In a study of Canadian preschool children
with low blood Pb levels, an association between cord blood Pb level and FSIQ was observed, while the
association of childhood concurrent BLLs with FSIQ effectively null (Desrochers-Couture et al., 2018).
In addition, associations were observed in boys but not in girls in several studies (Tatsuta et al., 2020;
Desrochers-Couture et al., 2018; Taylor et al., 2017). There was also some indication that the
heterogeneity across studies could be explained by modeling choices such as confounder adjustment for
other metals. For example, cross-sectional analyses found evidence that exposure to Mn may modify the
association between Pb exposure and IQ in some populations Martin et al. (2021); (Menezes-Filho et al.,
2018). However, studies that adjusted for multiple metals (e.g., Mn, Hg, Cd, and Pb) in regression
models, without examining the interaction between metals, found little evidence of an association
between cord or postnatal BLL and IQ (Zhou et al„ 2020b; Liu et al., 2015), imprecise associations only
in boys (Tatsuta et al., 2020; Desrochers-Couture et al., 2018), or large IQ decrements after adjustment
for Mn, Hg, and ADHD rating score Hong et al. (2015). Overall, recent studies generally corroborated the
epidemiologic observations of associations between Pb exposure and IQ in children with relatively low
blood Pb concentrations (<5 (ig/dL) among some groups of children (see Section 3.5.1.1). Consistent with
findings from the 2013 Pb ISA, studies continue to report associations with prenatal BLL (maternal and
cord blood Pb) and postnatal BLLs measured at various childhood lifestages despite the aforementioned
heterogeneity at BLLs <5 (ig/dL. Overall, the heterogeneity did not weaken the larger body of supporting
evidence.

In the review of the MDI evidence in the 2013 Pb ISA, emphasis was placed on results from
examinations at ages 2-3 years, which incorporate test items more similar to those in school-age IQ tests.
Among these studies, several included children with mean BLLs less than 5 (ig/dL (Henn et al„ 2012;
Jedrychowski et al„ 2009b; Hu et al„ 2006; Bellinger et al„ 1987). Most of the prospective studies
reviewed in previous IS As (U.S. EPA, 2013, 2006) found associations of higher prenatal (cord and
maternal BLL), earlier infancy, and concurrent BLL with lower MDI scores in children aged 2 to 3 years
(Figure 3-10). These blood Pb-associated decrements in MDI were observed in populations with mean
BLLs of 1.3 to 7.1 (ig/dL. Studies typically recruited participants before or at birth without consideration
of Pb exposure or maternal IQ and reported high to moderate follow-up participation as well as
nondifferential loss-to-follow-up. Most studies adjusted for birth outcomes, maternal IQ, and education.
Cord BLLs were associated with MDI, with additional adjustment for SES and HOME score in the
Boston cohort (Bellinger et al., 1987) and for HOME score in the Yugoslavia cohort (Wasserman et al.,
1992). Some studies found a stronger association of MDI with prenatal than child postnatal BLLs ((Hu et
al„ 2006; Gomaa et al., 2002; Bellinger et al., 1987).

Recent studies continue to support associations between Pb exposure measured during prenatal or
childhood lifestages and poorer performance on tests of neurodevelopment, among mothers and infants
with mean BLLs <5 (ig/dL (i.e., maternal (Y Ortiz et al„ 2017; Vigeh et al„ 2014; Kim et al„ 2013b, c),

3-79

-------
cord (Valeri et al., 2017), and postnatal (Lin et al.. 2013) BLLs). Although Zhou et al. (2017) found no
association overall, this study reported decrements on several domains of the GDS among infants of
mothers reporting high maternal stress. Similarly, Y Ortiz et al. (2017) found some evidence of
interactions between Pb exposure and maternal stress. Several studies found interactions between Pb and
Mn or Mn and As (Valeri et al„ 2017; Lin et al„ 2013; Henn et al„ 2012) or Cd exposure (Kim et al„
2013b); Kim et al. (2013c). Overall, recent studies support findings from the previous reviews and extend
the evidence pertaining to modification of the association between Pb exposure and infant
neurodevelopment by maternal stress and exposure to other metals. The MDI and other tests that measure
neurodevelopment in infants and toddlers are not intelligence tests. Notably, MDI scores, particularly
before ages 2-3 years, are not necessarily strongly correlated with later measurements of FSIQ in children
with normal development and thus, are not weighted heavily in the consideration of causality (U.S. EPA,
2013).

Experimental animal studies evaluated in the 2013 Pb ISA demonstrated that prenatal and early
postnatal or lifetime Pb exposure alters brain development via changes in synaptic architecture and
neuronal outgrowth, leading to impairments in memory and learning (Sections 4.3.10.4, 4.3.10.10, and
4.3.2.3 of U.S. EPA (2013)). A small number of recent experimental animal studies were designed to
compare exposures across multiple different developmental windows (Barkur and Bairy, 2015b; Xiao et
al„ 2014); these studies reported similar magnitudes of effects between developmental windows,
suggesting that individual periods of development may be similarly sensitive to Pb. Generally, longer
exposures that spanned multiple developmental periods (e.g., preconception through lactation) produced
not only the highest BLLs but the largest effects on cognition (Zhou et al., 2020a; Zhu et al„ 2019b).
Overall, these studies provide strong support for observations in epidemiologic studies that Pb exposure
during the prenatal, childhood, and adolescent lifestages is associated with cognitive effects. Recent
animal studies also provide evidence to support the observation that development (i.e., preconception,
gestation, lactation) may represent a critical window for Pb exposure to cause cognitive impairment later
in life (see Section 3.6.1). In rodents, developmental exposure to Pb was consistently associated with
persistent cognitive effects observed both early (Tartaglionc et al., 2020; Zhao et al., 2018; Barkur and
Bairy, 2015b; Anderson et al„ 2012) and later in life (Liu et al„ 2022c; Xiao et al„ 2014; Betharia and
Maher. 2012).

Learning, memory, and executive function are domains of cognitive function that are related to
intelligence, and several are evaluated in the subtests of FSIQ. Additionally, indices of memory, learning,
and executive function are comparable to endpoints examined in experimental animal studies. The studies
evaluated in the 2006 Pb AQCD and 2013 Pb ISA did not clearly indicate associations between higher
BLL and poorer performance on neuropsychological tests of memory or learning (U.S. EPA, 2013). The
ascertainment of the outcomes varied across studies, potentially explaining the heterogeneity of the
epidemiologic observations. Notably, evidence for both memory and learning decrements from
prospective analyses of several established cohorts (i.e., Rochester, Boston, and Cincinnati) was mixed
(Canfield et al„ 2004; Ris et al., 2004; Stiles and Bellinger, 1993; Bellinger et al„ 1991; Dietrich et al.,

3-80

-------
1991). Cross-sectional studies included in the previous ISA, however, generally found associations
between higher concurrent BLLs and poorer learning and memory. Several recent studies of children with
mean BLLs <5 (ig/dL add to the evidence informing the association of Pb exposure with performance on
tests of memory and learning; however, these recent studies do not enhance the consistency of the
evidence as a whole. Some of the available studies consider co-exposure to other chemicals and metals as
confounders (Tatsuta et al., 2014) despite evidence that such co-exposures may interact with or modify
the association between Pb and the outcomes (Yorifuji et al„ 2011). Several recent studies of rodents with
exposure resulting in mean BLLs <30 (ig/dL add to the evidence indicating coherence between the
epidemiologic and toxicological findings pertaining to learning and memory observed in the 2013 Pb
ISA.

Strong evidence of associations between Pb exposure and indices of executive function was
described in the 2013 Pb ISA. Studies included prospective analyses of several birth cohorts with
moderate to high follow-up rates in Boston and Rochester that examined BLLs before the outcome
assessment and adjusted for several potential confounding factors (Canfield et al., 2004; Canfield et al.,
2003b; Bellinger et al., 1994a; Stiles and Bellinger, 1993). Recent studies relying on parent or teacher
behavioral ratings on BRIEF did not generally report associations. The previous ISA did not incorporate
any evidence of the relationship between Pb exposure and executive function in animal models. Recent
studies from a single laboratory provided evidence that Pb exposure broadly impairs measures of
executive function in a reversal learning paradigm. These effects were sex-specific, with greater effects
reported in males. While these reports are consistent with one another, evidence for the association
between Pb exposure and impaired executive function in animal models with BLLs <30 (ig/dL remains
limited.

As described in Sections 3.5.1.1 and 3.5.1.2 and summarized above, heterogeneity in the
epidemiologic results for FSIQ and infant development at BLLs <5 (ig/dL may be explained in part by
sex, exposure to other metals, or maternal stress. Experimental animal studies offer some support for the
observations regarding sex and maternal stress. The limited evidence evaluated in the 2013 Pb ISA
(Virgolini et al., 2008; Yang et al„ 2003; Mcgivern et al., 1991), combined with recent evidence, provides
more consistent support for the sex-biased effects in both male (Anderson et al„ 2016) and female
(Tartaglione et al., 2020; Verma and Schneider, 2017; Anderson et al„ 2012; Betharia and Maher, 2012)
animals. The exact nature and direction of sex-specific effects given the variation in outcomes examined
remains unclear, however. Toxicological studies assessed in the 2013 Pb ISA demonstrated the potential
for Pb and maternal stress to result in dysfunction of the HPA axis (U.S. EPA, 2013). Recent
toxicological evidence provides further support for the interaction between maternal stress and Pb
exposure during the exposure period (Anderson et al., 2012; Cory-Slechta et al., 2012) but it remains
unclear whether stress would positively influence some facets of cognitive function. Given the lack of
evidence available on combined metal exposures, the possible interaction between multiple metals at
relevant levels in animals remains unclear.

3-81

-------
Poorer academic performance and achievement is linked with lower FSIQ and may have
important implications for success later in life (U.S. EPA, 2013). In children aged 5 to 18 years higher
blood Pb measured at various lifestages, including early childhood, and tooth Pb levels, which were
generally measured in dentin and reflect Pb exposure during the prenatal or early childhood period, were
associated with poorer performance on tests of math, reading, and spelling skills, lower probability of
high school completion and lower-class rank, and lower teacher ratings of academic functioning (U.S.
EPA, 2013). Recent studies in populations of children (age 6-16 years) enrolled in school districts
including North Carolina, Detroit, and Chicago with BLLs <5 (ig/dL support and extend these
observations of poorer academic performance in association with increasing Pb exposure in populations
with mean BLLs <5 (ig/dL.

Recent studies support and extend the evidence pertaining to the effect of Pb exposure on
cognitive function in children at low BLLs. Compelling evidence for a larger decrement in cognitive
function per unit increase in blood Pb among children with lower mean blood Pb concentrations,
compared with children with higher mean blood Pb concentrations, was supported by a reanalysis of a
pooled international dataset (Crump et al., 2013). The larger incremental effect of Pb on cognitive effects
at the lower (relative to higher) end of the study population Pb exposure distribution has been observed
across biomarkers [i.e., bone (Wasserman et al„ 2003; Wasserman, 2003), plasma (Hu et al„ 2006) and
blood (Section 4.3.12, Figure 4-15, and Table 4-16 of the 2013 Pb ISA U.S. EPA (2013))! and across
different cognitive function endpoints [i.e., infant development (Hu et al„ 2006), IQ (Wasserman et al„
2003; Wasserman, 2003) and academic achievement (Evens et al.. 2015).! Crump et al. (2013) also
supported the finding of Lanphear et al. (2019) that the C-R function in the pooled analysis of blood Pb
level and IQ is adequately modeled as linear at BLLs <10 (ig/dL. Recent studies with an adequate range
of Pb exposure measured during relevant time periods that would be required to further evaluate
exposure-response relationships were limited. Considering the collective body of studies, no evidence of a
threshold for cognitive effects in children across the range of BLLs examined in epidemiologic studies
was reported.

The total body of evidence evaluated is sufficient to conclude that there is a causal
relationship between Pb exposure and decrements in cognitive function in children. This causality
determination is the same as the conclusion in the 2013 Pb ISA, reflecting the consistency of the results
from epidemiologic studies of FSIQ, Bayley MDI, and academic performance and achievement, as well
as the coherence of evidence across epidemiologic and toxicological studies of learning and memory. The
pattern of associations consistently observed for tooth Pb levels and blood Pb levels measured at various
lifestages or time periods reduces uncertainty regarding the temporal association observed in cross-
sectional analyses of concurrent blood Pb levels with cognitive decrements in children. Notably
concurrently measured tooth Pb levels in dentin generally reflect the prenatal or childhood Pb exposure
that precedes the assessment of the outcome. Biological plausibility is provided by studies that describe
pathways involving the interaction of Pb with cellular proteins, in some cases competing with and
displacing other biologically relevant cations. This interaction leads to increased oxidative stress and the

3-82

-------
presence of inflammation, which can have widespread effects on brain structure and function, as well as
disruptions of Ca2+ signaling. These disruptions can result in altered brain signaling and contribute to the
development of neurological health effects. Recent studies support the conclusion of the 2013 Pb ISA that
Pb-associated cognitive effects in children occur in populations with mean BLLs between 2 and 8 (ig/dL.
As noted in the 2013 Pb ISA, this conclusion was based on studies that examined early childhood BLLs
(i.e., age <3 years), considered peak BLLs in their analysis (i.e., peak <10 (.ig/dL). or examined concurrent
BLLs in young children (i.e., age 4 years). One recent study of Canadian preschool aged children from
mainly middle- to upper-middle SES families with low Pb exposure (mean concurrent blood Pb level
0.70) did not find an association between concurrent Pb exposure and performance on WPPSI at age 3 to
4 years (Desrochers-Couture et al.. 2018). Other recent studies found associations of Pb exposure with
cognitive effects in children with mean BLLs <2 (ig/dL; however, the studies with mean BLLs <2 (ig/dL
lack the aforementioned attributes (i.e., early childhood BLLs, consideration of peak BLLs, or
examination of concurrent BLLs in young children) and exhibit heterogeneity in both the magnitude and
precision of the associations at the lowest blood Pb concentrations. The observed heterogeneity may be
explained in part by the underlying distribution and complex relationship between covariates in the
populations studied, including sex, maternal stress, and co-exposures to other metals and neurotoxic
chemicals, at relatively low BLLs (<5 (.ig/dL). Overall, epidemiologic and toxicological studies continue
to strongly support the finding that exposure during multiple lifestages (prenatal through adolescence and
early adulthood) is associated with cognitive effects in children. No evidence of a threshold for cognitive
effects in children across the range of BLLs examined in epidemiologic studies was reported.

3-83

-------
Table 3-2 Summary of evidence Indicating a causal relationship between Pb exposure and cognitive effects
in children

Rationale for C^sality Key Evidence13 References'3 Levels Associated

Determination3 wjth EffectsC

Consistent associations from multiple, Evidence from prospective studies for decrements U.S. EPA (2013) Blood Pb (various

prospective epidemiologic studies with in FSIQ in association with prenatal, earlier Section 4 3 2 1 Table 4-3 t'me Peri°ds and

relevant BLLs childhood, peak, concurrent, lifetime average BLLs ' lifestages): Means 3-

and tooth Pb levels in children ages 4-17 yr in 16 |jg/dL

multiple U.S. locations, Mexico, Europe, Australia.

Recent prospective studies observe associations of Section 3.5.1.1 Blood Pb (various

Pb with FSIQ; however, heterogeneity in the time periods and

magnitude and direction of the associations is lifestages) <5 |jg/dL

present. (<2 |jg/dL in some

studies)

Some recent epidemiologic studies indicate Section 3.5.1.6.5

potential effect modification or interactions of Pb

with sex, other metals, and maternal stress

potentially explaining heterogeneity in the observed

associations at BLLs <5 |jg/dL.

Evidence from prospective studies for lower scores
on tests of executive function and academic
performance in association with earlier childhood or
lifetime average BLLs or tooth Pb levels in children
ages 5-20 yr in multiple U.S. locations, U.K., New
Zealand.

U.S. EPA (2013)

Blood Pb (various
time periods and
lifestages) <5 |jg/dL

Recent evidence generally relies on outcome Section 3.5.1.4

ascertainment based on the BRIEF is inconsistent.

The direction and magnitude of associations were U.S. EPA (2013) and Section 3.5.1.3
not consistent for learning and memory. Recent
evidence does not enhance the consistency.

3-84

-------
Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker
Levels Associated
with Effects0

Supporting evidence from cross-sectional studies of
children ages 3-16 yr, but most did not consider
potential confounding by parental caregiving
quality. Includes large NHANES III analysis.

Several studies indicate supralinear C-R
relationship, with larger decrements in cognitive
function per unit increase in blood Pb at lower BLLs
in children ages 5-10 yr. Reanalysis of international
pooled analysis substantiates this finding.

Epidemiologic evidence helps rule out
chance, bias, and confounding with
reasonable confidence

Several epidemiologic studies found associations
with adjustment for SES, maternal IQ and
education, HOME score. Several adjust for birth
weight, smoking. A few, nutritional factors.

U.S. EPA (2013)
Section 3.5.1.6.2



Experimental animal studies with
relevant exposures provide coherence
and help rule out chance, bias, and
confounding with reasonable
confidence

Impaired learning and associative ability in juvenile
and adult animals as indicated by performance in
tasks of visual discrimination, water maze, y maze,
and operant conditioning with schedules of
reinforcement with relevant dietary Pb exposure.

U.S. EPA (2013)
Section 4.3.2.3

Blood Pb (after
prenatal/ lactation,
lactation only,
prenatal/lifetime Pb
exposure): 10-25
pg/dL



Recent studies of executive function in rodents add
to the evidence; however, evidence for impaired
executive function in animal models with BLLs <30
|jg/dL remains limited.

Section 3.5.1.4.2



Experimental animal studies with
relevant exposures provide coherence
for epidemiologic observations of effect
modification by sex or interactions of Pb
with other metals or maternal stress

Recent studies in rodents suggest that factors such
as sex and maternal stress may influence the
effects of Pb on cognitive function.

Section 3.5.1.6.5



Biological plausibility demonstrated

Pathways involving oxidative stress, inflammation
and Ca2+ signaling result in impaired neuron
development, synaptic changes, LTP, and
neurotransmitter changes.

U.S. EPA (2013)
Section 3.6



U.S. EPA (2013)	Blood Pb (various

time periods and
lifestages) <5 |jg/dL

U.S. EPA (2013)
Section 3.5.1.6.1
Crump et al. (2013)

3-85

-------
Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker
Levels Associated
with Effects0

Recent studies support and extend findings related Section 3.3
to overt nervous system effects.

BLL = blood lead level; BRIEF = Behavior Rating Inventory of Executive Functions; Ca2+ = calcium ion; C-R = concentration-response; FSIQ = full-scale intelligence quotient; HOME
= Health Outcomes and Measures of the Environment; IQ = intelligence quotient; LTP = long-term potentiation; NHANES = National Health and Nutrition Examination Survey; Pb =
lead; SES = socioeconomic status; yr = year(s).

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 2015).
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or
inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-86

-------
3.5.2 Externalizing Behaviors: Attention, Impulsivity, and Hyperactivity in
Children

The evidence evaluated in the 2013 Pb ISA was sufficient to conclude that there is a "causal
relationship" between Pb exposure and effects on attention, impulsivity, and hyperactivity in children.
Several prospective studies demonstrated associations of blood Pb measured years before outcomes or
tooth Pb levels, which were generally measured in dentin and reflect prenatal and early childhood Pb
exposure, with attention decrements and hyperactivity in children (7-20 years) as assessed using objective
neuropsychological tests and rated by parents and teachers. Most of the prospective studies examined
representative populations with no indication of participation that was conditional on BLLs and behavior.
The results from the prospective studies were generally adjusted for potential confounding by SES as well
as parental education and caregiving quality, with some studies also considering parental cognitive
function, birth outcomes, substance abuse, and nutritional factors. With respect to the timing of exposure
discerned from prospective studies, blood Pb-associated attention decrements and hyperactivity were
found in populations with prenatal (maternal or cord) or postnatal (i.e., 3-60-month average, age 6 years,
or lifetime average through age 11-13 years) mean BLLs of 7 to 14 (ig/dL and in groups with BLLs >10
(ig/dL at 30 months of age. Biological plausibility for these observations in children was provided by
experimental animal studies that demonstrated increases in impulsivity or impaired response inhibition
with relevant postweaning and lifetime Pb exposures that resulted in BLLs of 11 to 30 (ig/dL.
Demonstrated Pb-induced impairments in neurogenesis, synaptic pruning, and dopamine transmission in
the prefrontal cerebral cortex, cerebellum, and hippocampus also supported the biological plausibility of
the associations observed in the epidemiologic studies. Although coherence across and within lines of
evidence was demonstrated, the small number of studies of diagnosed ADHD were limited by their cross-
sectional or case-control design, inconsistent adjustment for SES and parental education, and lack of
consideration for potential confounding by parental caregiving quality.

There are three major domains of externalizing behavior disorders: (1) ADHD, (2)
undersocialized aggressive conduct disorder, and (3) socialized aggressive conduct disorder (as reviewed
in (Whitcomb and Merrell. 2012)). Although these domains are interrelated, to the extent possible, this
Section (3.5.2) will maintain a similar structure as the 2013 Pb ISA by focusing on the ADHD domain,
which encompasses characteristics including but not limited to short attention span, distractibility,
impulsivity, and hyperactivity. Within the ADHD domain of externalizing behaviors, most epidemiologic
studies of Pb exposure focus on attention, impulsivity, and hyperactivity. Some epidemiologic studies
examined composite indices of multiple behaviors, and a few studies have examined physician-diagnosed
ADHD. Domain-specific neuropsychological assessments of attention, impulsivity, and hyperactivity
with strong psychometric properties and rigorous validation were emphasized in the 2013 Pb ISA and
provide the strongest evidence for the causality determination. Studies that evaluated the association of Pb
exposure with externalizing behaviors assessed using teacher and parent ratings, which are generally
reliable and valid instruments that predict functionally important outcomes, contributed to the overall

3-87

-------
evidence (Desrochers-Couture et al.. 2019; Fruh et al., 2019; Nigg et al.. 2016; Hong et al., 2015;

Gittleman and Eskenazi, 1983).

Control for confounding is considered an attribute of a well-conducted, high-quality study.
Greater weight is given to studies that consider important potential confounders in their design or
statistical analyses. As noted in the 2013 Pb ISA (U.S. EPA, 2013), associations between Pb biomarker
levels and externalizing behaviors may be confounded by parental SES, education, and IQ; nutritional
status; and the quality and stability of the caregiving environment (often evaluated using the HOME score
(Totsika and Svlva. 2004)). The research available for evaluation in the 2013 Pb ISA did not establish a
direct relationship between parental psychopathology and child Pb exposure or one between parental
psychopathology and poorer parental caregiving quality. Thus, parental psychopathology itself was not
considered to be a potential confounder of associations between child Pb and externalizing behaviors.
Although parental psychopathology was hypothesized to modify the association between Pb exposure and
externalizing behavior, no studies available for evaluation in the 2013 Pb ISA evaluated parental
psychopathology as an effect modifier. To the extent that parental psychopathology could affect child Pb
exposure indirectly through parental caregiving quality, however, it was noted that confounder control
would be achieved in studies that included an adjustment for the HOME score or similar metrics.

Greater emphasis is also placed on prospective studies with repeated assessments of BLLs and
studies of children with BLLs that are less influenced by higher past Pb exposures (i.e., younger children).
Studies assessing effects in populations with BLLs that are most relevant to current U.S. children (e.g., <5
(ig/dL) are also emphasized, e.g., (Cho et al„ 2010; Nicolescu et al., 2010; Chandramouli et al„ 2009;
Nigg et al„ 2008; Chen et al„ 2007; Chiodo et al., 2007). In the current ISA, when considering the causal
relationship of Pb exposure with attention, impulsivity, and hyperactivity, PECOS statements (see Section
3.2) were refined to focus on the most informative studies. Longitudinal epidemiologic studies with mean
(or central tendency) BLLs <5 (ig/dL are highlighted in the text as are the most reliable biomarkers of Pb
exposure (i.e., blood, bone, teeth, or nails). Consideration of potential confounding and modification of
the observed associations by the aforementioned factors was evaluated when considering the overall
quality of the study. Measures of central tendency for Pb biomarker levels used in each study, along with
other study-specific details, including study population characteristics and select effect estimates, are
highlighted in evidence inventory Table 3-7E (Epidemiologic Studies) and Table 3-7T (Toxicological
Studies). In addition, studies with central tendency blood Pb concentrations that exceed 5 (ig/dL are
extracted into Table 3-8E of Section 3.7 (Evidence Inventories). An overview of the recent evidence is
provided below. Overall, recent studies generally support findings from the 2013 Pb ISA.

3-88

-------
3.5.2.1

Attention in Children

3.5.2.1.1 Epidemiologic Studies of Attention in Children

Attention is the ability to maintain a consistent focus on an activity or relevant stimuli and can be
assessed by examining sustained attention, concentration, or distractibility. The preponderance of
evidence pertaining to the externalizing behaviors included in the 2013 Pb ISA evaluated the association
of Pb exposure with measures of attention (U.S. EPA, 2013). Most prospective studies found associations
of blood or tooth Pb levels with decrements in neuropsychological tests of attention as well as parent and
teacher ratings of attention. One strength of the prospective studies is that they characterized the sequence
of Pb exposure (i.e., prenatal blood Pb, postnatal blood Pb before the outcome, concurrent, lifetime
average blood Pb, and tooth Pb [i.e., generally measured in dentin, reflecting prenatal, early childhood or
cumulative Pb exposure depending on the assessment method]), establishing the temporal relationship
between exposure and outcome. In addition, the studies reported moderate to high follow-up participation
that was not conditional on blood or tooth Pb levels and controlled for important confounders (i.e.,
parental education, IQ, and caregiving quality; SES). Overall, these studies showed a pattern of lower
attention with higher blood or tooth Pb level (see Figure 4-9 and Table 4-11 of the 2013 Pb ISA (U.S.
EPA, 2013)). Mean BLLs were generally within the range of 7-14 (ig/dL for most of the prospective
studies, and cross-sectional studies generally supported findings from the longitudinal analyses (see
Section 4.3.3.1 of the 2013 Pb ISA (U.S. EPA, 2013}!	

A small number of recent longitudinal epidemiologic studies of children with relatively low BLLs
(i.e., <5 (ig/dL) add to the evidence for associations between Pb exposure and decrements in
neuropsychological tests of attention or parent and teacher ratings of behaviors that indicate attention
problems (see Section 3.5.2.4).

Ncugcbaucr et al. (2015) conducted an analysis of the Duisburg birth cohort data to examine the
association of maternal BLLs at 32 weeks gestation with performance on neuropsychological tests of
attention (Test of Attentional Performance for Children [KiTAP]) and parent-rated ADHD behaviors on
the German Symptom Checklist for ADHD (Fremdbeurteilungsbogen fur

Aufmerksamkeitsdefizit/Hyperaktivitatstorungen [FBB-ADHS]) in childhood. Maternal blood Pb was
most strongly associated with specific KiTAP subtests, i.e., number of omissions (geometric mean ratio
[GMR] = 1.15 [95% CI: 1.00, 1.33]) and reduced performance speed (GMR= 1.14 [95% CI: 0.98, 1.33]).
Maternal blood Pb concentration was positively associated with the inattention component of the FBB-
ADHS indicating that inattention increases with increasing BLL (GMR = 1.05 [95% CI: 0.99, 1.12]).

In a study of a subset (n = 27) of Inuit children (Boucher et al., 2012b) (see Section 3.5.2.5) that
used a modified Posner paradigm to assess the association between prenatal and concurrent childhood Pb
exposure with visuospatial attention, vigilance, and impulsivity, Ethier et al. (2015) found that concurrent
ln-transformed blood Pb was associated with some tests of attention including longer reaction times (|3 =

3-89

-------
0.52 [95% CI: -0.10, 1.14] per SD increase in ln-transformed Pb) in a model adjusted for age, sex, and
current PCB exposure. In another study, Tatsuta et al. (2014) adjusted for PCBs and MeHg (in addition to
maternal IQ and family income) and found no associations with sequential processing score (-2.14 [95%
CI: -12.80, 8.53] per unit of log transformed BLL [base not specified]) or mental processing score (-3.32
[95% CI: -12.4, 5.77] per unit of log transformed BLL [base not specified])). Yorifuii et al. (2011) found
that cord blood Pb was associated with some neuropsychological tests of attention and working memory
on the WISC-R (i.e., digit span) and that the interaction between cord blood Pb and cord Hg level may be
less than additive. For example, the association of cord Pb with performance on the digit span forward at
age 7 was |3 = —0.11 [95% CI: -0.29, -0.07] per log-transformed unit of BLL without accounting for the
interaction between cord Pb and cord Hg concentration. This association was more pronounced when Hg
exposure was lower. Specifically, the unstandardized associations of log-transformed cord BLL with the
neuropsychological test outcomes were most discernable among children with hair Hg concentrations
below 2.61 (ig/g, which was the lowest cord Hg concentration. For example, a lower digit span forward
score on the WISC-R (|3 = -1.70 [95% CI: -3.12, -0.28]) at age 7 and a lower digit span backward score
on the WISC-R (|3 = -2.73 [95% CI: -4.32, -1.14]) at age 14 were observed among children with the
lowest Hg exposure.

Ruebner et al. (2019) evaluated the association between BLLs and attention and hyperactivity
among children with CKD. Attention was assessed using either Couriers' Kiddie Continuous Performance
(K-CPT; 4-5 years) or Couriers' CPT II (>6 years), which produce scores for omission and commission
errors, correct detection rate, response variability, reaction time, and summary measures for sustained
attention and inhibitory control. A 1.8 T-score point increase (i.e., worse performance) in CPT variability
(95% CI: 0.2, 3.5), which indicates problems with sustained attention and attention regulation, was
associated with childhood blood Pb (on average, ~2 years before outcome ascertainment) in this study.
Covariates considered as potential confounders included race, poverty, maternal education, and factors
related to CKD. The median BLL in this study was 1.2 (ig/dL.

A small number of studies examined the gene-environment interaction between Pb exposure and
genotypes associated with attention decrements. Roonev et al. (2018) studied children in Lisbon, Portugal
to determine the association of baseline BLL (8-12 years old) and variants of GRIN2A and GRIN2B,
which regulate neurodevelopmental processes, with performance on neuropsychological tests, including
tests of attention, during the 7-year follow-up period. A pattern of association indicating poorer
performance on tests of attention with increasing baseline Pb exposure was not observed. Choi et al.
(2020) enrolled children (5-18 years old) with ADHD and healthy controls without ADHD to evaluate
interactions between Pb exposure and noradrenergic pathway-related genotypes (i.e., [DAT1], dopamine
receptor D4 [DRD4], and alpha-2A-adrenergic receptor [ADRA2A]). ADHD was assessed using the
ADHD rating scale (ADHD-RS) and neuropsychological tests of attention (i.e., CPT and SCWT) were
also administered. BLLs were associated with omission errors (|3 = 3.75 [95% CI; 0.09, 7.40]) in models
adjusted for IQ, age, and sex, which were found to partly mediate the effect of Pb on ADHD symptoms in
a path analysis model. An interaction effect was detected between the ADRA2A Dral genotype and Pb

3-90

-------
levels on omission errors (|3 = 5.07 [95% CI: 0.20, 9.93]). Multiple comparisons were made in this
analysis (e.g., associations with additional CPT components including commission errors, response time,
and response time variability, ADHD-RS components and SCWT components were not observed),
increasing the likelihood of chance findings.

Summary

Prospective studies in the 2013 Pb ISA showed strong support for an association between pre-
and postnatal Pb exposure (range: 7-14 (ig/dL) and decreased scores on neuropsychological tests and
parent/teacher ratings of attention. Cross-sectional studies from the 2013 Pb ISA corroborated these
observations. A small number of recent studies reported associations of maternal and cord BLLs <5 (ig/dL
with some measures of inattention; however, the results for multiple subtests were reported, potentially
increasing the likelihood of chance findings. Recent studies add to the limited evidence regarding co-
exposure to Hg and gene-environment interactions (see Section 3.5.2.6.3).

3.5.2.1.2 Toxicological Studies of Attention

In support of the associations described in the preceding sections for BLL with attention
decrements in children, studies have found Pb-induced decreases in attention in animals, although results
have not been consistent across studies. Although tests in animals often measure aspects of both attention
and impulsivity, behaviors measured with signal detection tests with distraction can be inferred as
predominately assessing sustained attention. In this test, animals earn food rewards by responding to a
target stimulus and not responding to a distracting light. Poorer sustained attention and greater
distractibility are indicated by lack of response to the target and increased response to the distracter light,
respectively. The 2006 Pb AQCD (U.S. EPA, 2006) reported inconsistent effects of Pb exposure in
animals on performance in this test. For example, postweaning Pb exposure that produced BLLs of 16
and 28 (ig/dL induced small decreases in attention in adult rats, as indicated by small increases in
omission and commission errors but only during sessions with long intervals between stimuli (Brockel
and Cory-Slechta, 1999). Lifetime Pb exposure from birth (mean peak BLLs of 15 and 25 (ig/dL for the
50 and 100 (ig/kg/day groups, respectively) was found to induce distractibility in monkeys at age 9-10
years, as indicated by increased responses to irrelevant cues, i.e., distracting stimuli, in a spatial
discrimination reversal task. Repeated reversal testing revealed that these deficits likely were not due to
sensory or motor impairment (Gilbert and Rice, 1987).

In animals, Pb-induced decrements in attention have been inferred from tests designed to assess
impulsivity but that have elicited behaviors that suggest deficits in attention. For example, a study
reported that impaired performance on auditory threshold tasks in Pb-exposed monkeys was likely due to
lack of attention (Laughlin et al., 2009). Rhesus monkeys were exposed to Pb acetate from gestation
(drinking water of mothers, 3 months prior to mating) to birth or postnatally from birth to age 5.5 months

3-91

-------
at weaning, resulting in bone Pb levels of 7 and 13 |ig/g for prenatal and postnatal groups at 11 years of
age, respectively, and average BLLs of 35 and 46 (ig/dL, respectively, during Pb exposure. Animals were
tested at age 13 years when BLLs had returned to baseline levels. The inability of some of the monkeys to
engage or focus attention on the task at hand yielded fewer available measurements in Pb-exposed
animals versus controls. These observations were made in monkeys with higher peak BLLs than those
relevant to this ISA. No recent studies have evaluated attention in animals following exposure that
resulted in BLLs relevant to the current ISA.

3.5.2.2 Impulsivity in Children

3.5.2.2.1 Epidemiologic Studies of Impulsivity in Children

Measures specific to impulsivity were examined in relatively few epidemiologic studies of
children compared with measures of attention, and most studies including evaluations of impulsivity that
were included in the 2013 Pb ISA were cross-sectional in design (U.S. EPA, 2013). The available
evidence indicated Pb-associated poorer performance on tests of response inhibition. Response inhibition
is a measure of impulsivity and has been assessed in children via stop signal tasks, which measure the
execution of action in response to stimuli and the inhibition of that action when given a stop signal.
Associations of blood and tooth Pb with parent and teacher ratings of impulsivity were also reported.
These studies generally adjusted for potential confounding by SES, sex, parental education, and smoking;
however, parental IQ or caregiving quality was not examined in most studies. The relatively small body
of epidemiologic evidence (see Figure 4-9 and Table 4-11 (U.S. EPA, 2013)) was coherent with results
from experimental animal studies (see Section 4.3.3.1 (U.S. EPA. 2013)).

Analyses of Inuit children have been conducted since the 2013 Pb ISA, examining Pb exposures
and impulsivity. Boucher et al. (2012a) examined response inhibition deficits assessed with the Go/No-
Go task and event-related potentials (ERPs) derived from electroencephalogram (EEG) recordings during
task performance among Inuit school children residing in Arctic Quebec. Cord and concurrent blood Pb
concentrations were associated with increased impulsivity after adjustment for covariates including child
age, sex, SES, maternal nonverbal reasoning abilities, and Hg (Figure 3-11). In addition, Ethier et al.
(2015) studied a subset of this population (n = 27) using a modified Posner paradigm to assess the
association between pre- and concurrent childhood Pb exposure with visuospatial attention, vigilance, and
impulsivity. The study found that cord Pb was associated with greater impulsivity (|3 = 0.42 [95% CI:
0.08, 0.76] per SD increase in ln-transformed Pb).

3-92

-------
100

•= 90

T M

5 70

fBl

[

§5 70

v 60

o
o

0.8-2.5 2.5-3.7 3.7-5.6 5.6-20.9
Cord blood Pb (|jg/dL)

0.8-2.5 2.5-3.7 3.7-5.6 5.6-20.9
Cord blood Pb (|ig/dL)

Source: Boucher et al. (2012a)

Figure 3-11

0.4-1.3 1.3-2.0 2.0-2.9 2.9-12.8
11-year blood Pb (|ig/dL)

Mean ± standard deviation behavior performance in the Go/No-Go
task according to quartiles of exposure for (A and B) cord blood
Pb and (C) childhood blood Pb level at age 11 years.

Summary

Studies of impulsivity in children in the 2013 Pb ISA were limited by their quantity and lack of
temporality but generally indicated associations of Pb exposure with worse scores on tests of response
inhibition and on parent and teacher ratings of impulsivity. These studies also often lacked confounder
control for parental IQ or caregiving quality, which are key potential confounders. Recent analyses of
Inuit children add support for the relationship between Pb exposure and impulsivity with additional
consideration of potential confounders including maternal nonverbal reasoning abilities (Boucher et al..
2012a).

3.5.2.2.2 Toxicological Studies of Impulsivity

The associations described between higher BLL and greater impulsivity in children are supported
by findings in animals for Pb-induced increases in perseveration and impaired ability to inhibit
inappropriate responses. In animals, these effects are supported by studies reviewed in the 1986 and 2006
Pb AQCDs (U.S. EPA, 2006, 1986) and studies incorporated into the 2013 Pb ISA. Animal studies
provide more consistent evidence for the effects of Pb exposure on impulsivity than on sustained
attention. As mentioned earlier, behaviors displayed by animals in a variety of tests can be identified as
reflecting impulsivity. These include tests of differential reinforcement of low rates of responding, fixed
interval (FI) schedule performance, FI with extinction, or fixed ratio (FR)/waiting-for-reward. Greater
impulsivity is indicated by premature responses, decreased pause time between two scheduled events, and
increased perseveration.

3-93

-------
Behaviors observed in tests of operant conditioning with FI reinforcement schedules have also
been used to indicate impaired learning in animals (Section 3.5.1.3.2), and the interactions observed
between Pb exposure and maternal or offspring stress may also apply to effects on impulsivity. Maternal
exposure to 150 ppm Pb with and without stress co-exposure was found to increase overall FI rate and
decrease Post-reinforcement Pause (PRP) in rats. Lifetime (from gestation) Pb exposure resulting in BLLs
of 11-16 (ig/dL increased the overall FI rate without stress co-exposure and decreased PRP with stress co-
exposure (Rossi-George et al.. 2011). suggesting that stress may interact with Pb exposure to affect
attention. Discrimination reversal learning has been shown to be affected by Pb exposure. In these tasks,
an animal is trained to choose between two alternative responses and is then required to reverse the
association. Perseveration or lack of inhibition of the original response can be interpreted to involve
impulsivity. Spatial and non-spatial discrimination reversal was significantly affected in monkeys after Pb
exposure during infancy, after infancy, or continuously from birth, and was exacerbated with distracting
stimuli (Rice. 1990; Rice and Gilbert. 1990b; Gilbert and Rice. 1987). These monkeys had BLLs in the
range of 15-36 (ig/dL, which includes values relevant to this ISA. Hilson and Strupp (1997) found Pb
exposure (Pb acetate in drinking water at GD 1-PND 28, yielding a BLL of 26 (ig/dL in the lower dose
group) in rats slowed reversal learning in an olfactory discrimination task. However, analysis of the
response patterns showed that Pb exposure shortened the perseverative responding phase of reversal
learning and lengthened the post-perseverative phase of chance responding, indicating impairments in
associative ability, not response inhibition. Thus, it is more likely that Pb negatively affected associative
learning rather than impulsivity in this study (Hilson and Strupp. 1997). which is inconsistent with the FI
data in monkeys. Due to the small number of studies following relevant Pb exposures in rodents, it
remains unclear whether Pb affects impulsivity in FI.

The effects of Pb exposure on impulsivity also have been demonstrated in a study reporting that
Pb-exposed animals wait a shorter period of time for reward in FR/waiting for reward testing. In this test,
animals can obtain food by pressing a lever a fixed number of times (FR component). Free food is then
delivered at increasingly longer time intervals, so long as the animal inhibits additional lever presses.
Animals can reset the schedule to return to the FR component at any time. Brockel and Corv-Slechta
(1998) exposed male Long-Evans rats to 0, 50, or 150 ppm Pb acetate in drinking water from weaning,
which produced respective BLLs of <5, 11, and 29 (ig/dL after 3 months of exposure. After 40 days of
exposure, the 150 ppm Pb-exposed rats responded more quickly in the FR component and reset the
schedule (thus shortening the waiting period) more often than did the 50 ppm Pb-exposed rats and
controls. In the waiting component, average wait time was significantly lower in both Pb exposure groups
compared with controls. The rats exposed to 150 ppm Pb also had higher response rates and earned more
reinforcers per session but had a higher response to reinforcement-ratio than did the 50 ppm Pb group and
controls, which indicated less efficient responses.

Weston et al. (2014) used the delayed discounting paradigm following developmental exposure
with or without prenatal restraint stress. The delayed discounting protocol offered animals the choice
between a large reward after a long delay or a small reward after a short delay. Pb increased long-delay

3-94

-------
responding, slowed acquisition of delayed discounting performance, and increased failures almost
exclusively in males. Consistent with (Hilson and Strupp. 1997). these results more likely represent
impaired learning or cognitive flexibility rather than simply increased impulsivity.

In summary, several studies in animals indicate that Pb exposure of rodents and nonhuman
primates from birth or after weaning changes behavior in ways consistent with increased impulsivity,
primarily as indicated by impaired response inhibition. It is also important to note that many of the
measures of impulsivity discussed in this section are sensitive to disruption by impairments in learning
and executive function, which is consistent with several of the studies summarized in Sections 3.5.1.3.2
and 3.5.1.4.2. Some observations of Pb-induced impulsivity in animals were made with BLLs considered
relevant for this ISA. The observations for Pb-induced increases in impulsivity in animals provide support
for associations found in children of higher blood and tooth Pb levels with lower response inhibition and
higher ratings of impulsivity.

Transgenerational Effects of Pb on Impulsivity

The paradigm of combined Pb and stress exposure experienced by a laboratory animal has been
examined with a focus on the common pathway of altered HPA axis and brain neurotransmitter levels.
Studies investigating the interactions between Pb and prenatal stress on learning and memory are
reviewed in Section 3.5.1.3.2. These findings were expanded on in a recent study that investigated the
transgenerational effects of combined Pb and prenatal stress in mice (Sobolewski et al.. 2020). The
authors reported that Pb exposure in the gestating female (F0 generation) resulted in sex-specific effects
in the third filial (F3) generation (no direct exposure to Pb), with F3 females displaying significantly
elevated response rates in an FI schedule of reward compared with control lineages, suggesting an
impulsive behavioral phenotype (Sobolewski et al.. 2020). This and other transgenerational effects were
accompanied by Pb-induced alterations in neurotransmitters, BDNF expression, and DNA methylation.
The authors postulated that lineage effects may be mediated through some combination of maternal
responses to pregnancy, maternal behavior, or epigenetic modifications (Sobolewski et al.. 2020). While
these findings were limited to a single study, they support the possibility that exposure to Pb may
influence the behavior of subsequent generations.

3.5.2.3 Hyperactivity in Children

3.5.2.3.1 Epidemiologic Studies of Hyperactivity in Children

Studies reviewed in the 2006 Pb AQCD (U.S. EPA, 2006) indicated associations between higher
concurrent BLLs or tooth Pb levels and higher parent or teacher ratings of hyperactivity in children aged
6-11 years in the U.S., Asia, and New Zealand (Rabinowitz et al., 1992; Silvaet al., 1988; Gittleman and
Eskenazi, 1983; Needleman et al.. 1979; David et al.. 1976). The case-control or cross-sectional design of

3-95

-------
studies limited understanding of the temporal sequence between Pb exposure and hyperactivity. A
prospective study (Chandramouli et al.. 2009) included in the 2013 Pb ISA also found associations
between BLL and hyperactivity as rated by teachers and parents. Overall, studies indicated associations in
children 3-12 years old with mean concurrent BLLs of 3.7-12 (ig/dL. Studies of recent hyperactivity
symptoms as rated by parents and teachers are discussed in Section 3.5.2.4.

3.5.2.3.2 Toxicological Studies of Hyperactivity

The 2006 Pb AQCD (U.S. EPA, 2006) reviewed the evidence that developmental exposure to Pb
could affect locomotor activity in laboratory animals. Findings summarized in this document included
four studies showing increased activity with developmental Pb exposure and three studies showing no
change in activity. The 2013 Pb ISA (U.S. EPA, 2013) only described one new study in this category,
which showed a decrease in activity in mice after maternal Pb exposure (Leasure et al., 2008). Effects of
developmental Pb exposure on rodent locomotor activity are commonly assessed using an open-field test.
The activities examined vary across studies (e.g., distance traveled, counts of square crossings). Because
there are myriad potential explanations for changes in rodent activity, it can be difficult to draw
conclusions from "simple" tests like open-field. The results from such tests are best interpreted alongside
additional behavioral assays, which, together, may better model the complexity of human behavior.
Conclusions for an effect of developmental Pb exposure on locomotor activity were not reached in earlier
United States Environmental Protection Agency (U.S. EPA) Pb reviews due to mixed results in these
tests. Studies described in the 2013 Pb ISA (U.S. EPA, 2013) and 2006 Pb AQCD (U.S. EPA, 2006) as
observing the effects of maternal Pb exposure, which resulted in mean BLLs no higher than 30 (ig/dL, are
Munoz et al. (1989), Rodrigucs et al. (1996), Moreira et al. (2001), Trombini et al. (2001), De Marco et
al. (2005), and Leasure et al. (2008).

Recent studies (see Table 3-1 IT) observed the activity of early postnatal rodents after
developmental exposures to Pb with varying durations. Tartaglionc et al. (2020) exposed rats to Pb from
pregestation to offspring weaning and tested offspring. They observed a decrease in neonatal spontaneous
activity on PND 10 and no change in spontaneous activity on PND 4, 7, and 12. Another group reported
that two groups of CD 1 mice exposed to either a high or low dose of Pb through lactation exhibited
hyperactivity (PND 7, 11, 15, 19) compared with controls in open-field testing Duan et al. (2017).
Interestingly, when locomotor data from early postnatal studies are pooled, nine out of nine sets of
lactationally exposed animals (BLLs: 9.6-28.9 (ig/dL) tested between PND 14 and 23 were hyperactive
(Duan et al., 2017; De Marco et al., 2005; Moreira et al., 2001; Rodrigucs et al., 1996). These sets
consisted of both male and female rodents, except for one set of only males in Moreira et al. (2001).

Evidence inventory (Section 3.7) also includes recent studies that monitored the effects of
developmental exposure to Pb on immature rodents postweaning. Basha and Reddy (2015) found
decreased locomotor activity in male rats with gestational exposure to Pb when tested on both PND 21
and PND 28. Betharia and Maher (2012) studied open-field behavior in Sprague Dawley rats with

3-96

-------
gestational and lactational exposure to Pb. There were no differences in total square crossings for both
males and females tested at both PND 24 and PND 59. Flores-Montoya and Sobin (2015) saw no effects
on open-field tasks (PND 28) in two groups each of male and female C57BL/6 mice after postnatal
exposure (PND 0-28) to low levels of Pb acetate. Developmentally Pb-exposed rats from the Tartaglione
et al. (2020) study described in the previous paragraph exhibited no change compared with control in
open-field activity when tested on PND 30. Neuwirth et al. (2019a) observed no effect on locomotor
activity in open-field tests (PND 36-45) for two groups of Long-Evans rats with different levels of in
utero and lactational exposure to Pb. Zou et al. (2015) reported that exposure to Pb acetate in drinking
water for 3 weeks (PND 37-58) increased spontaneous locomotor activity in juvenile ICR mice tested on
PND 58. These recent studies do not indicate effects on the activity of rodents when tested in adolescence
after Pb exposure from mothers, and they do not support the findings of hyperactivity in the similarly
exposed and tested mice described by Trombini et al. (2001).

Faulk et al. (2014), Basha et al. (2014), and Wang et al. (2016) evaluated activity in adult rodents
after developmental Pb exposure. Faulk et al. (2014) measured activity and horizontal movements along
with ambulatory activity by adult offspring of dams (Agouti mouse) exposed to Pb for weeks from
pregestation until weaning. Mean BLLs for offspring were not reported; however, maternal BLLs tested
at weaning were below the limit of detection in the control group and 4.1, 25.1, and 32.1 (ig/dL in the
three respective exposure groups of 2.1, 16, and 32 ppm. Overall horizontal activity was different across
Pb exposures in females but not in males. At 9 months, female offspring exposed to 2.1 ppm Pb had
higher average horizontal activity compared with controls. There was a sex-specific difference in
ambulatory measurement (subset of total horizontal activity), with only exposed females showing
significant differences from controls. Ambulatory activity was lower in females at the 32-ppm exposure
level at 3 months versus control offspring. Males did not exhibit significant differences at any time point
or exposure level. Although there was suggestive evidence of differences in the life-course patterns of
vertical activity by exposure among females, neither sex showed statistically significant differences
between Pb-exposed and control offspring. Testing adult rats at 4, 12, and 18 months of age, Basha et al.
(2014) found consistent decreases in locomotor activity associated with lactational Pb exposure. While
mean BLLs at these testing periods were shown to be below the 30 (ig/dL limit for PECOS relevance, it is
notable that the mean BLL for this group of animals was determined to be 49.5 (ig/dL at PND 45. Wang
et al. (2016) measured distance traveled in the open field by Sprague Dawley rats aged 116-122 days
after adolescent Pb exposure in drinking water from PND 24 to 56. They observed no effect of this
exposure.

Evidence of Pb exposure-induced intergenerational effects on rodent behavior was also reviewed
in the 2006 Pb AQCD (U.S. EPA, 2006). Trombini et al. (2001) observed increased open-field
ambulation in F2 generation rats derived from female offspring of Pb-treated pregnant mothers. Recently,
Sobolewski et al. (2020) exposed F0 mice to Pb during pregnancy and lactation, bred offspring with
unexposed mates for two generations (F1 and F2), and then evaluated behavior in the F3 generation. F3
females demonstrated a small increase in locomotor activity, regardless of lineage.

3-97

-------
Overall, there are still mixed indications on locomotor activity from Pb exposure studies with
BLLs <30 |ig/dL. which may be due to differential dosing, timing of exposures, and activity
measurements. However, in a rare set of four individual studies wherein these critical factors were
analogous, Pb exposure during lactation induced hyperactivity in rodents when tested within a PND 14 to
23 window (Duan et al.. 2017; De Marco et al.. 2005; Moreiraet al.. 2001; Rodrigues et al.. 1996). Pb-
induced hyperactivity in rodents provides some support for hyperactivity observed in children but may be
more appropriately interpreted in the context of additional behavioral assays.

3.5.2.4 Parent and Teacher Ratings of ADHD-related Behavior

In addition to finding associations with attention, impulsivity, and hyperactivity, epidemiologic
studies also found associations between higher concurrent BLLs and higher parent and teacher ratings of
ADHD-related behaviors, calculated as a composite of the various behaviors evaluated in the diagnosis of
ADHD (see Section 4.3.3.1 of the 2013 Pb ISA (U.S. EPA. 2013)). Most of these studies were limited
due to their cross-sectional design and lack of validation of ADHD ratings with clinical diagnosis.
Although diagnostic guidelines for ADHD exist, the exact criteria or specific behaviors required can vary.
Thus, within studies, there were variations among subjects in the types of behaviors they displayed that
led to a diagnosis of ADHD. Further the available studies considered age, sex, and SES or parental
education but generally not both as potential confounders, and none of the studies considered parental
caregiving quality.

Recent longitudinal studies add to the body of evidence examining the association between
prenatal and childhood Pb exposure and parent/teacher-rated ADHD symptoms in populations with
relatively low blood Pb concentrations (<5 (ig/dL). This group of studies includes some that found
associations with hyperactivity using the SDQ, which is a screening questionnaire that includes five
domains. Sioen et al. (2013) analyzed data from the Flemish Environment and Health Study (FLEHS I,
2002-2006), a birth cohort comprising mother-infant pairs to examine the association between cord blood
Pb and ADHD-related behaviors for 281 infants whose parents returned the SDQ (26.4%). A positive
association of cord blood Pb concentration with hyperactivity score >7 was observed (OR: 2.94 [95% CI:
1.17, 7.38 per log ug/dL increase in BLL]). In another study using this assessment instrument, Fruh et al.
(2019) analyzed data from mother-child pairs participating in Project Viva, a longitudinal birth cohort in
eastern Massachusetts. Maternal blood Pb concentration in erythrocytes was measured during the second
trimester of pregnancy and parents rated their child's behavior using the SDQ in mid-childhood (median
7.7 years). The associations (i.e., |3 coefficients) with the parent and teacher-rated hyperactivity
component of the SDQ were 0.10 (95% CI: -0.21, 0.41) and 0.20 (95% CI: -0.24, 0.64), respectively.
While behavior assessments and maternal blood Pb measurements were available for fewer than half of
Project Viva participants, important confounders including HOME score, maternal IQ, and parental
education were considered in this study.

3-98

-------
Several of these longitudinal epidemiologic studies used the Behavior Assessment System for
Children (BASC) to assess both the behaviors and emotions of children (Reynolds and Kamphaus. 2015).
BASC-2 includes individual subscales for attention and hyperactivity as well as an overall behavioral
skills index (BSI) composite score. Specific rating scales and forms related to attention, hyperactivity, and
impulsivity are emphasized in this section (e.g., clinical scales such as "attention problems" or
"hyperactivity" on the teacher or parent rating scale forms).

Horton et al. (2018) analyzed data from the Early Life Exposure in Mexico to Environmental
Toxicants (ELEMENT) Project birth cohort in Mexico City to determine the association of weekly tooth
Pb concentration (prenatal through 1 year postnatal) with BASC-2 scores assessed between 8 and 11
years old. Distributed lag models were used to identify specific time windows of increased risk due to Pb
exposure. Tooth Pb concentration estimated to correspond with the 8 to 11 months postnatal period was
associated with parent-rated behavioral symptoms overall (|3 = 0.22 units [95% CI: 0.06, 0.38] per natural
log unit increase in dentine Pb concentration), and hyperactivity (|3 = 0.19 units [95% CI = 0.02, 0.37] per
natural log unit increase in dentine Pb concentration) after adjustment for gestational age and maternal
education. Approximately 12% of the original cohort was enrolled in this study; participants differed with
respect to several characteristics including child birth weight and maternal IQ. Rasnick et al. (2021)
conducted a study that estimated monthly air Pb exposure. The authors also aimed to identify sensitive
time windows of exposure; however, they attempted to distinguish exposure to Pb in air by controlling for
concurrent BLL (age 12 years) in their analysis of the Cincinnati Study of Allergy and Air Pollution study
data. Air Pb exposure was estimated using validated land use regression models and behavioral outcomes,
including attention and hyperactivity, were assessed using BASC-2 administered at age 12. Distributed
lag models to predict outcome responses based on current and past (i.e., lagged) predicted air Pb
exposures did not identify associations during any of the lifestages examined. Models were adjusted for
community deprivation, residential greenspace, and elemental carbon attributable to traffic (ECAT), in
addition to concurrent BLL.

In addition to examining attention using Couriers' (Section 3.5.2.1.1), Ruebner et al. (2019)
evaluated the association of BLLs with parent-rated attention and hyperactivity symptoms on BASC-2.
This study was unique in that it enrolled children with CKD. Associations with parent ratings did not
persist in models that controlled for potential confounders including race, poverty, maternal education,
and clinical factors related to CKD. The median BLL in this study was 1.2 (ig/dL.

Several other instruments, including FBB-ADHS, the Child Behavior Checklist (CBCL), the
Disruptive Behavior Disorder (DBD) rating scale, the Barkley Adult ADHD-IV Rating Scale (BAARS),
Couriers' Rating Scale (CRS), the Strengths and Weaknesses of ADHD Symptoms and Normal Behavior
Scale (SWAN), and DuPaul's ADHD rating scale were used to assess total ADHD in recent prospective
or case-control studies. These rating scales are generally reliable and valid instruments that predict
functionally important outcomes (Desrochers-Couture et al.. 2019; Fruh et al.. 2019; Nigg et al.. 2016;
Hong et al.. 2015; Gittleman and Eskenazi. 1983).

3-99

-------
Neugebauer et al. (2015) conducted an analysis of the Duisburg birth cohort data to determine the
association of maternal BLL at 32 weeks gestation with parent-rated ADHD behaviors in childhood
(average age 9.5 years old) assessed using FBB-ADHS. Maternal blood Pb was associated with overall
ADHD symptoms (|3 = 1.06 [95% CI: 1.01, 1.12]), with the strongest association observed forthe
impulsivity component (|3 = 1.13 [95% CI: 1.06, 1.22]). These associations were observed after
adjustment for confounders including parental education, but not SES.

Liu et al. (2014b) examined the association of early childhood blood Pb concentration at 3, 4, or 5
years old (mean: 6.8 (ig/dL) with parent and teacher ratings of ADHD behaviors among Chinese school
children at age 6 using CBCL and the Caregiver-Teacher Report Form (C-TRF). The outcome was
modeled as a continuous and also as a dichotomous variable (i.e., clinically significant behavior problems
when T-score >60). The associations (i.e., |3) between increased blood Pb concentrations (per (ig/dL) and
ADHD behavior problems were 0.001 (95% CI: -0.002, 0.002) for problems reported on CBCL and 0.07
(-0.18 to 0.32) for behavior problems reported on C-TRF. The associations (i.e., OR) with clinically
significant ADHD behavior reported by parents on CBCL were 1.08 (95% CI: 0.99, 1.18) among children
overall, 1.04 (95% CI: 0.94, 1.16) among boys, and 1.15 (95% CI: 0.98, 1.35) among girls. The
participation rate was 81% in this study. Models were adjusted for confounders including parental
caregiving quality but not SES. Another prospective study evaluated the association of Pb exposure with
caregiver ratings on CBCL. In this study of adolescents, Winter and Sampson (2017) examined the
relationship between average BLLs in childhood (6 years old or younger) with impulsivity between 16 to
18 years old. These authors found a 0.06 SD (95% CI: 0.01, 0.12) increase in impulsivity score, after
adjustment for caregiver education and SES. Participants were originally enrolled in the mid-1990s and a
random sample of those that continued to participate in 1999 and 2002 was randomly selected for this
study, with 67% of those selected agreeing to participate.

Choi et al. (2016) investigated the association of childhood BLLs (geometric mean BLL =1.56
(ig/dL) with parent-rated ADHD symptoms later in childhood assessed using DuPaul's ADHD rating
scale. Approximately 72% (n = 2,159) of 2,967 eligible participants provided blood Pb measurements and
ADHD assessments, and 2052 were free of ADHD symptoms at baseline. A positive association between
childhood blood Pb and the development of ADHD symptoms at the 2-year follow-up visit was observed
in this study (RR: 1.55 [95% CI: 1.00, 2.40] >2.17 versus <2.17 (ig/dL) after adjustment for residential
area, household income, parental marital status, family history of psychiatric disorders, preterm birth, and
birth weight. A stronger association was observed among children with higher BLLs and who resided in a
single parent home (RR: 3.57 [95% CI 1.60, 7.98]).

Another longitudinal analysis examined the association of both cord BLL (mean: 4.7 (ig/dL) and
childhood (mean 2.7 (ig/dL) BLL with ADHD symptoms among Inuit children in Quebec (Boucher et al..
2012b). In this study, teachers completed the DBD rating scale to indicate Diagnostic and Statistical
Manual of Mental Disorders (DSM)-IV symptoms of ADHD inattentive type and ADHD hyperactive-
impulsive type. Associations between child concurrent log-transformed BLL and hyperactive/impulsive-

3-100

-------
type ADHD symptoms assessed using DBD were observed (OR = 4.01 [95% CI: 1.06, 5.23] tertile 2
versus tertile 1; OR = 5.52 [95% CI: 1.38, 22.12] tertile 3 versus tertile 1). Inattentive-type ADHD
symptoms on the DBD were not associated with Pb exposure. (Desrochers-Couture et al., 2019) extended
this study by conducting a mediation analysis to estimate the direct and indirect associations of childhood
BLLs with adolescent externalizing behaviors, including ADHD symptoms assessed using BAARS. The
study found an association between childhood BLL and child hyperactivity and impulsivity, assessed by
teachers on CBCL (|3 = 0.45 [95% CI: 0.13, 0.78]). Neither a direct (|3 = 0.09 [95% CI: -0.11, 0.28]) nor
an indirect (|3 = -0.02 [95% CI: -0.06, 0.03]) association with adolescent ADHD symptomology assessed
using BAARS was observed. A wide array of covariates was considered as potential confounders
including both maternal education and SES.

Hong et al. (2015) found an association between blood Pb concentration and higher parent and
teacher-rated ADHD-RS symptoms (|3 = 1.04 [95% CI: 0.18, 1.90] and |3 = 1.90 [95% CI: 0.74, 3.05],
respectively) in a cross-sectional analysis of Korean school children from 8 to 11 years old after
adjustment for demographic factors (age, sex, residential region, paternal education level, and SES). This
association remained positive but was attenuated in models additionally adjusted for FSIQ, Mn, and Hg (|3
= 0.68 [95% CI: -0.20, 1.56] and |3 = 1.49 [95% CI: 0.32, 2.67], parent- and teacher-rated symptoms,
respectively). The mean BLL in this study was 1.80 (ig/dL. Associations indicating an increase in
commission errors on CPT were also observed.

Nigg et al. (2016) conducted a case-control study of children from Michigan (mean BLL = 0.74
(ig/dL (cases) and 0.94 (ig/dL controls). In this study, ADHD composite indices were derived for (1)
inattention/disorganization and (2) composite hyperactivity-impulsivity using relevant scales of the
DuPaul, Couriers', and SWAN scales. This study found an interaction between the hemochromatosis gene
(HFE) C282Y genotype, which is involved in iron metabolism, and BLL in predicting parent and teacher
reports of hyperactivity-impulsivity but not inattention. For example, the association between z scores of
BLL and hyperactivity was significantly stronger among those with the HFE C282Y mutation (|3 = 0.74,
[95% CI: 0.52, 0.96]) compared with those with the wild type genotype (|3 = 0.28 [95% CI: 0.15, 0.41]).
This study also found an interaction between z scores of logio-transformed BLL and sex (association
larger in boys) in predicting parent and teacher-rated hyperactivity and impulsivity but not attention.

3.5.2.4.1 Summary

Cross-sectional studies in the 2013 Pb ISA found associations between higher concurrent BLL
and higher parent and teacher ratings of ADHD-related behaviors, calculated as a composite of the
various behaviors that are evaluated in the diagnosis of ADHD (U.S. EPA, 2013). The evidence from
prospective studies was limited to Chandramouli et al. (2009), which found associations between BLL
and hyperactivity as rated by teachers and parents. Parent and teacher ratings generally considered SES or
parental education, but typically not both, as potential confounders. None of the studies considered
parental caregiving quality. Recent longitudinal studies that established the temporality between the

3-101

-------
exposure and the outcome add to the body of evidence examining the association between prenatal and
childhood Pb exposure and parent/teacher-rated ADHD symptoms in populations with relatively low
blood Pb concentrations (<6 (.ig/dL). Across studies, associations were observed with tooth Pb
concentration that were measured in dentin and generally reflect early childhood Pb exposure, childhood
BLLs, and maternal or cord (2-5 (ig/dL) BLLs. Studies of caregiver-reported ADHD symptoms generally
report associations with composite indices (Choi et al., 2016; Hong et al., 2015; Neugebauer et al., 2015;
Liu et al., 2014b; U.S. EPA, 2013), and there is some evidence indicating that the associations with
impulsivity and hyperactivity symptoms (Desrochers-Couture et al., 2019; Fruh et al., 2019; Horton et al.,
2018; Winter and Sampson, 2017; Nigg et al„ 2016; Neugebauer et al., 2015; Sioen et al„ 2013; Boucher
et al., 2012b) are stronger than the associations with inattention symptoms. The majority of recent studies
were prospective and generally reported moderate or high participation rates. Some studies addressed the
validity of caregiver assessed outcomes by evaluating internal consistency (Rasnick et al., 2021;
Desrochers-Couture et al., 2019), and Nigg et al. (2016) addressed reliability and validity concerns by
using structural equation modeling to create latent factors for inattention and hyperactivity-impulsivity for
each informant. Rating scales used in these studies are generally reliable and valid instruments that
predict functionally important outcomes (Desrochers-Couture et al., 2019; Fruh et al., 2019; Nigg et al.,
2016; Hong et al„ 2015; Gittleman and Eskenazi, 1983). Confounder adjustment remains somewhat
inconsistent across studies, although Liu et al. (2014b) and Fruh et al. (2019) adjusted for the quality of
parental caregiving, Choi et al. (2016) adjusted for family history of psychiatric disorders, and several
considered both SES and parental education (Desrochers-Couture et al., 2019; Ruebner et al., 2019;
Horton et al„ 2018; Winter and Sampson, 2017; Boucher et al., 2012b). There is uncertainty regarding the
patterns of exposure that are associated with BLLs in older children because they may be influenced by
higher past exposure.

3.5.2.5 Clinically Diagnosed ADHD

In the 2013 Pb ISA, results from a small body of cross-sectional studies indicated associations
between concurrent BLL and the prevalence of ADHD symptom ratings (Section 3.5.2.4) and clinically
diagnosed ADHD in children aged 4-17 years. The temporal relationship between Pb exposure and
ADHD was not established in these studies, and concurrent blood Pb concentrations in older children may
reflect higher past exposures. Additionally, some ADHD symptom rating studies lacked outcome
validation, and confounding was inconsistently addressed across studies. Therefore, the evidence
specifically for these total ADHD index ratings and clinically diagnosed ADHD were emphasized less in
the 2013 Pb ISA (U.S. EPA, 2013) than evidence for individual behaviors when drawing conclusions
about the effects of Pb exposure on attention, impulsivity, and hyperactivity.

Recent studies add to the evidence and address some of the uncertainties pertaining to the studies
included in the 2013 Pb ISA. Notably, Ji et al. (2018) analyzed data from the Boston Birth Cohort (1479
mother-infant pairs) to examine the association of early childhood Pb exposure (i.e., earliest (< age 4)

3-102

-------
blood Pb concentration recorded during routine screening) with the development of ADHD later in
childhood. ADHD was assessed using electronic medical records (International Classification of Diseases
[ICD]-9 codes: 314.0, 314.00, 314.01, 314.1, 314.2,314.8, and 314.9, or ICD-10 codes: F90.0, F90.1,
F90.2, F90.8, and F90.9). Several important potential confounders (i.e., parental education, SES but not
quality of parental caregiving) were controlled for in the analysis, and child sex, maternal high-density
lipoprotein (HDL), and maternal stress were considered as potential effect modifiers. Ji et al. (2018)
analyzed the association modeling blood Pb concentration as continuous and as categorical variables.
When blood Pb was analyzed using three categories, the OR comparing children with BLLs between 2
and 4 (ig/dL to children with BLLs <2 (ig/dL was 1.08 (95% CI: 0.81-1.44). The OR comparing children
with BLLs between 5 and 10 (ig/dL to children with BLLs <2 (ig/dL was 1.73 (95% CI: 1.09-2.73).

When blood Pb was modeled as a continuous variable, the OR was 1.12 [95% CI: 1.00, 1.25) per ug/dL
increase in BLL. Sex-stratified analyses comparing children with BLLs between 5 and 10 (ig/dL to
children with BLLs <5 (ig/dL indicated no association among girls (OR = 0.68 [95% CI: 0.27, 1.69]) and
a strong association among boys (OR = 2.49 [95% CI: 1.46-4.26]). Joint analyses indicated a 10-fold
increase in the magnitude of the association between childhood blood Pb concentration and ADHD
diagnosis among those with multiple risk factors (i.e., male sex, inadequate maternal HDL, higher
maternal stress).

Several additional recent studies also extend the evidence. Park et al. (2016) conducted a hospital
based case-control study in Busan, South Korea comparing the odds of higher blood Pb concentration
among diagnosed ADHD cases, which were confirmed using the Korean version of the Kiddie Schedule
for Affective Disorders and Schizophrenia Present and Lifetime (K-SADS-PL-K), to the odds of higher
blood concentration among controls that were frequency matched by age and sex and adjusted for other
potential confounders. Blood Pb was measured when the cases and controls were recruited into the study.
Higher blood Pb concentration was associated with increased risk of ADHD (OR: 1.60 [95 % CI: 1.04-
2.45] per unit increase in log BLL); however, blood Pb concentrations were not associated with ADHD-
RS score or CPT profiles among the ADHD cases. In a smaller case-control study that examined the
association of childhood Pb exposure with diagnosed ADHD among children living near a former smelter
in Omaha, Nebraska, Kim et al. (2013a) found a positive association (OR: 2.52 [95% CI: 1.07, 5.92] per
unit increase in natural log BLL).

In a recent cross-sectional analysis of NHANES (2003-2004) data, Geier et al. (2018) examined
the association of concurrent blood Pb concentration with self-reported doctor diagnosed attention deficit
disorder (ADD) among children and adolescents 10-19 years old. This study observed a positive
association between concurrent blood Pb concentration and ADD after adjusting for age, race, sex, and
SES (OR: 1.29 [95% CI: 1.03, 1.55]). In a previous analysis Braun et al. (2006) found an association in
children aged 4-15 years participating in NHANES (1999-2002). ADHD ascertained by the parent report
of ADHD diagnosis is subject to reporting bias; however, the examination of multiple risk factors and
outcomes in NHANES reduces the likelihood of biased participation and reporting of ADHD by parents
of children specifically with higher Pb exposure.

3-103

-------
3.5.2.5.1 Summary

The 2013 Pb ISA assessed a small body of cross-sectional studies that examined the associations
between concurrent BLLs and the prevalence of clinically diagnosed ADHD. The temporal relationship
between Pb exposure and ADHD was not established in these studies and it was noted that concurrent
blood Pb concentration in older children potentially reflects higher past exposures. As noted in the 2013
ISA, clinically diagnosed ADHD was emphasized less than evidence for individual behaviors in drawing
conclusions about the effects of Pb exposure on attention, impulsivity, and hyperactivity in the 2013 Pb
ISA (U.S. EPA, 2013). Further, the available studies did not consistently adjust for SES, parental
education, and quality of parental caregiving. A small number of recent studies add to the evidence
showing consistent associations between Pb exposure and diagnosed ADHD. One recent epidemiologic
study (Ji et al., 2018) addressed several of the uncertainties identified in the literature included in the 2013
Pb ISA. Specifically, this study employed a prospective design, and adjusted for parental education and
SES (although not quality of parental caregiving). Notably, ADHD was ascertained using ICD codes
recorded on electronic records and ADHD type was not distinguished in this study.

3.5.2.6 Relevant Issues for Interpreting the Evidence Base
3.5.2.6.1 Lifestages

Environmental exposures during critical lifestages spanning from childhood into adolescence can
affect key physiological systems that orchestrate brain development and plasticity (see Section 3.4.1.6.4
of U.S. EPA (2013)). Epidemiologic studies examined in the 2013 Pb ISA consistently showed that BLLs
measured during various lifestages and time periods, including the prenatal period, early childhood, later
childhood, and averaged over multiple years, are associated with attention decrements, impulsivity, and
hyperactivity in children. These observations of Pb-associated elevated risk are well supported by
findings in animals that prenatal and early postnatal or lifetime Pb exposures alter brain development via
changes in synaptic architecture (Section 4.3.10.4 of U.S. EPA (2013)) and neuronal outgrowth (Section
4.3.10.10 of U.S. EPA (2013)), potentially leading to increases in impulsivity (Section 4.3.3.1 of U.S.
EPA (2013)). Potential mechanisms of lifestage-specific sensitivities are further reviewed in Section 3.3.
Recent studies support this conclusion from the 2013 Pb ISA.

A limited number of epidemiologic studies employed methods designed to further elucidate
critical lifestages for Pb exposure but did not change the overall conclusion in the 2013 Pb ISA. Horton et
al. (2018) used distributed lag models to identify specific time windows of increased Pb-associated
externalizing behaviors. This study found that tooth Pb concentration corresponding with the 8 to 11
months postnatal period was associated with parent-rated behavioral symptoms overall (0.22 units [95%
CI: 0.06, 0.38] per natural log unit increase in dentine Pb concentration), and hyperactivity (|3 = 0.19 units
[95% CI = 0.02, 0.37] per natural log unit increase in dentine Pb concentration) after adjustment for

3-104

-------
gestational age and maternal education. In another study, Rasnick et al. (2021) also aimed to distinguish
critical windows of exposure to Pb, focusing on Pb concentration in air by controlling for concurrent BLL
(age 12 years). Air Pb exposure was estimated using validated land use regression models and behavioral
outcomes, including attention and hyperactivity, were assessed using BASC-2 administered at age 12.
Distributed lag models to predict outcome responses based on current and past (i.e., lagged) predicted air
Pb exposures did not identify associations during any of the lifestages examined.

3.5.2.6.2 Public Health Significance

The strongest evidence indicating a causal relationship between Pb exposure and attention,
impulsivity, and hyperactivity assessed in the 2013 Pb ISA was derived from studies that relied on
neuropsychological testing (U.S. EPA, 2013). Domain-specific neuropsychological assessments of
attention, impulsivity, and hyperactivity have strong psychometric properties and rigorous validation;
however, deficits on these neuropsychological tests do not directly correspond to a diagnosis of ADHD
nor do they necessarily predict long-term consequences that might be associated with some types of
ADHD. Studies that evaluated the association of Pb exposure with behavioral symptoms of ADHD
assessed using teacher and parent ratings contributed to the overall evidence in the 2013 Pb ISA, but the
limitations of these studies were noted. The bulk of the recent evidence comprises prospective studies of
parent or teacher ratings of ADHD behavioral symptoms. The recent studies addressed some uncertainties
in the previous ISA related to the temporal association of the exposure with the outcome and controlled
for potential confounding. Studies of diagnosed ADHD are also subject to limitations. Although
diagnostic guidelines for ADHD exist, the exact criteria or specific behaviors required for diagnosis may
vary across studies. The recent study by Ji et al. (2018) addressed several of the uncertainties regarding
the association of Pb exposure with clinical ADHD. This study was prospective in design, assessed early
childhood BLL (<4 years old), and adjusted for parental education and SES (although not quality of
parental caregiving); however, ADHD was ascertained using ICD codes recorded on electronic records
and ADHD type was not distinguished.

3.5.2.6.3 Potentially At-Risk Populations

Sex

Studies examining sex as an at-risk factor for attention, hyperactivity and impulsivity outcomes
were not assessed in the 2013 Pb ISA. A recent study by Nigg et al. (2016) found an interaction between
BLL and sex in predicting parent and teacher-rated hyperactivity and impulsivity but not attention. The
association was larger in boys in this study.

3-105

-------
Maternal Smoking

Maternal smoking during pregnancy was examined in a study of children's concurrent BLLs and
the prevalence of ADHD among children aged 8-15 years. An interaction was observed between
children's current BLLs and prenatal tobacco smoke exposure; those children with high Pb levels and
prenatal tobacco smoke exposure had the highest odds of ADHD (Frochlich et al.. 2009). Recent studies
have not examined maternal smoking as an at-risk factor.

Co-exposure to Other Metals or Chemicals

Studies examining other metals as an at-risk factor for attention, hyperactivity and impulsivity
outcomes were not assessed in the 2013 Pb ISA. Some recent studies adjusted for other metals or
chemicals (e.g., PCBs) (Ethicr et al.. 2015; Tatsuta et al.. 2014). and effect modifications were observed
in other studies (Yorifuji et al.. 2011). For example, Yorifuji et al. (2011) found a less-than-additive
interaction between cord Pb and Hg concentrations. Specifically, a lower digit span forward score on the
WISC-R (|3 = -1.70 [95% CI: -3.12, -0.28] per log-transformed BLL) at age 7 and a lower digit span
backward score on the WISC-R (|3 = -2.73 [95% CI: -4.32, -1.14] per log-transformed BLL) at age 14
were observed among children with the lowest Hg exposure.

Gene-Environment Interactions

Studies examining gene-environment interactions in the context of attention, hyperactivity and
impulsivity outcomes were not assessed in the 2013 Pb ISA. Interactions between child BLL and genes
that regulate neurodevelopmental processes were observed in studies of attention (Choi et al.. 2020;
Roonev et al.. 2018). Genes that were implicated included variants of GRIN2A and GRIN2B and
genotypes involved in the regulation of noradrenergic pathways. In addition, Nigg et al. (2016) found an
interaction between the HFE C282Y genotype and BLL in predicting parent and teacher reports of
hyperactivity-impulsivity but not inattention. Specifically, the association between z scores of BLL and
hyperactivity was significantly stronger among those with the HFE C282Y mutation (|3 = 0.74 [95% CI:
0.52, 0.96]) compared with those with the wild type genotype (|3 = 0.28 [95% CI: 0.15, 0.41]).

3.5.2.7 Summary and Causality Determination: Attention, Impulsivity, and
Hyperactivity

Attention, hyperactivity, and impulsivity are included within the ADHD domain of externalizing
behaviors. Although not studied as extensively as cognitive function, several epidemiologic studies have
examined the relationship between Pb exposure in children and attention, impulsivity, and hyperactivity
in children and young adults.

3-106

-------
The majority of these studies examined attention, and some also examined impulsivity or
hyperactivity. Thus, the focus of the evaluation is on the evidence related to attention, but the evaluation
also draws on coherence with evidence for impulsivity and hyperactivity, including evidence in animals
and that suggesting potential modes of action. The collective epidemiologic evidence base for attention in
children comprises many prospective and cross-sectional studies, which were also reviewed in the 2006
Pb AQCD and the 2013 Pb ISA and some recently published studies. Most of these studies reported
associations between childhood blood Pb or tooth Pb levels, that reflected early postnatal Pb exposure,
and attention decrements, impulsivity, and hyperactivity (Table 3-7E). A small number of recent
longitudinal studies contributed to this evidence. Not all results were uniform with regard to precision and
the magnitude of the association, but results mostly showed a pattern of attention decrements, impulsivity,
and hyperactivity with higher blood or tooth Pb levels.

Whether prospective, cross-sectional, or longitudinal, most studies relied on population-based
recruitment from prenatal clinics, hospitals at birth, or schools and reported moderate to high
participation. Several of the studies reviewed in the previous ISA demonstrated increased loss-to-follow-
up in certain groups (e.g., lower SES or HOME scores), which has the potential to introduce selection
bias and reduce the generalizability of findings. A strong indication that participation in the study was
biased to those with higher BLLs and greater deficits in attention, hyperactivity, or impulsivity was not
observed. Recent studies incorporated adjustments for these and other covariates. Repeated testing in
children was common but the consistent pattern of association observed across the ages, BLL, and
behavioral outcomes examined increases confidence that the evidence is not unduly biased by the
increased probability of finding associations by chance alone. Coherence with animal studies, which are
less vulnerable to confounding, further supports the pattern of associations described in the preceding
sections.

The strongest epidemiologic evidence indicating an association of Pb exposure with inattention
and hyperactivity is described in the 2013 Pb ISA U.S. EPA (2013). Prospective studies showed strong
support for an association between Pb exposure (range: 7-14 (ig/dL) and decreased scores on
neuropsychological tests and parent/teacher ratings of attention and hyperactivity. Cross-sectional studies
from the 2013 Pb ISA generally corroborated these observations. Studies of impulsivity in children in the
2013 Pb ISA were limited by their quantity and lack of temporality but generally indicated associations of
Pb exposure with worse scores on tests of response inhibition and on parent/teacher ratings of impulsivity
in cross-sectional analyses. A small number of recent prospective studies with mean maternal and cord
BLLs <5 (ig/dL report associations with some measures of inattention (Ethier et al„ 2015; Neugebaueret
al., 2015). In addition, recent analyses of Inuit children add support for the relationship between child and
cord BLL and impulsivity (Boucher et al., 2012a).

Most of the aforementioned studies of parent and teacher ratings of ADHD-related behaviors in
the 2013 Pb ISA were largely cross-sectional in design. The evidence from prospective studies was
limited to Chandramouli et al. (2009), which found associations between BLLs and hyperactivity as rated

3-107

-------
by teachers and parents. The available studies considered SES or parental education but generally not
both as potential confounders, and none of the studies considered parental caregiving quality. The bulk of
the recent evidence comprises prospective studies that establish the temporality of the association
between Pb exposure and parent or teacher ratings of ADHD symptoms and clinical ADHD. Across
studies, associations were observed with tooth Pb concentration, childhood (<6 |ig/dL). and maternal or
cord (2-5 (ig/dL) BLLs. Studies of caregiver-reported ADHD symptoms generally reported associations
with composite indices, and there is some evidence that the associations with impulsivity and
hyperactivity symptoms are stronger than the associations with inattention symptoms. The majority of the
recent studies were prospective and generally reported moderate or high participation rates. Some studies
addressed the validity of caregiver assessed outcomes by evaluating internal consistency (Rasnick et al.,
2021; Desrochers-Couture et al„ 2019), and Nigg et al. (2016) addressed reliability/validity concerns by
using structural equation modeling to create latent factors for inattention and hyperactivity-impulsivity for
each informant. Confounder adjustment has become more consistent across recent studies. In addition to
the studies relying on parent and teacher behavior ratings, a small number of recent studies add to the
evidence showing consistent associations between Pb exposure and diagnosed ADHD. One recent
epidemiologic study (Ji et al., 2018) addressed several of the uncertainties identified in the literature
included in the 2013 Pb ISA. Specifically, this study employed a prospective design, assessed early
childhood BLL, and adjusted for parental education and SES (although not quality of parental
caregiving). Notably, in this study, ADHD was ascertained using ICD codes recorded on electronic
records and ADHD type was not distinguished. Uncertainty remains regarding the patterns of exposure
associated with BLLs in older children because they may be influenced by higher past exposures.

The findings from epidemiologic studies are generally coherent with findings of studies in
experimental animals. Available evidence in animals supports the effect of developmental Pb exposure in
rodents and nonhuman primates on behavioral measures consistent with increased impulsivity, primarily
indicated by impaired response inhibition. Measures of impulsivity are additionally sensitive to disruption
by impairments in learning and executive function, which is consistent with several of the studies
summarized in Sections 3.5.1.3.2 and 3.5.1.4.2. While no recent animal toxicological studies of attention
are available, evidence from the 2013 Pb ISA demonstrates Pb-induced decreases in attention in rodents
and monkeys, although results are not entirely consistent across studies. Pb has been observed to have
mixed effects on locomotor activity in rodents, but several studies have demonstrated Pb-induced
hyperactivity in rodents during the postnatal phase, which provides some support for the epidemiologic
findings.

In summary, the total body of evidence evaluated in this and previous assessments is
sufficient to conclude that there is a causal relationship between Pb exposure and attention,
impulsivity, and hyperactivity. This conclusion reflects the consistency of the results from
epidemiologic studies of externalizing behaviors in children and young adults, incorporating various
objective neuropsychological tests and reporting from teachers and parents, which are generally reliable
and valid instruments that predict functionally important outcomes. The conclusion also incorporates the

3-108

-------
coherence of evidence across epidemiologic and toxicological studies of externalizing behaviors and
biological plausibility provided by studies that outline pathways by which Pb may interfere with the
proper development, connectivity, and function of systems underlying externalizing behaviors.

3-109

-------
Table 3-3 Summary of evidence indicating a causal relationship of Pb exposure with attention, impulsivity,
and hyperactivity

Rationale for

Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with Effects0

Consistent
associations from
multiple prospective
epidemiologic
studies with relevant
BLLs

Evidence from prospective
studies for attention decrements
and hyperactivity in association
with prenatal (maternal or cord),
early childhood, and lifetime
blood Pb and tooth Pb levels in
children ages 7-17 yr and young
adults 19-20 yr in the United
States, United Kingdom
Australia, New Zealand.

Evidence from prospective
studies of parent or teacher-
rated ADHD composite
symptom indices derived from
widely used, structured
instruments.

Burns et al. (1999)

Ris et al. (2004)

Ferqusson et al. (1993)

Bellinger et al. (1994a)
Chandramouli et al. (2009)
Leviton et al. (1993)

Section 4.3.3.1, U.S. EPA (2013)
Neuqebauer et al. (2015)

Ethier et al. (2015)

Choi et al. (2016)

Neuqebauer et al. (2015)

Liu et al. (2014b)

Blood Pb:

Means 2 to 8.3 |jg/dL (prenatal maternal or cord),
8.3 |jg/dL (age 6 yr), 13.4 |jg/dL (age 3-60 mo), 14
|jg/dL (lifetime avg to age 11-13 yr)

Group with age 30 mo >10 |jg/dL

Tooth Pb (ages 6-8 yr): Means: 3.3, 6.2 |jg/g

Childhood BLL <6 |jg/dL; maternal and cord BLL 2-5
pg/dL.

Ratings for impulsivity and
hyperactivity more strongly
associated with Pb exposure

Sioen et al. (2013)

Fruh et al. (2019)

Horton et al. (2018)

Neuqebauer et al. (2015)

Winter and Sampson (2017)
Desrochers-Couture et al. (2019)
Boucher et al. (2012b)

Niqq et al. (2016)

Childhood BLL <6 |jg/dL; maternal and cord BLL 2-5
pg/dL.

Prospective analysis found
associations with impulsivity in
Inuit children

Boucher et al. (2012a)

Mean 4.7 (cord), 2.7 (concurrent, average age 11.3
yr)

3-110

-------
Rationale for







Causality

Key Evidence13

References'3

Pb Biomarker Levels Associated with Effects0

Determination3







Prospective analysis find

Ruebner et al. (2019)

Child blood Pb ~2 yr before outcome assessment at



association with attention



age 4-18 yr



decrements in children with CKD





Limited evidence

Co-exposure to Hg modified the

Yorifuii et al. (2011)



evaluates the

risk of Pb-associated effects on





potential modification

attention (less than additive





of Pb associations to

effect observed)





other metals or







genes









Interactions between BLL and

Niqa et al. (2016)





genes that regulate

Roonev et al. (2018)





neurodevelopmental processes
observed.

Choi et al. (2020)





Association observed in

Ji et al. (2018)

Mean: 2.2 |jg/dL (<4 yr of age)



prospective study of clinical







ADHD diagnosed before age 6,







with adjustment for parental







education and SES.







No association found with

Wasserman et al. (2001)

Blood Pb: Mean 7.2 |jg/dL for lifetime (to age 4-5 yr)



ratings of attention problems in



avg



children ages 4-5 yr in whom







ratings may be measured less







reliably.





Supporting evidence

Associations of concurrent BLL

Section 4.3.3.1, U.S. EPA (2013)

Concurrent (ages 5-7.5 yr) blood Pb: Means 5.0-5.4

from cross-sectional

with attention decrements,

pg/dL

studies

impulsivity, and hyperactivity in







children ages 5-7.5 yr. Some







populations had high prenatal







drug or alcohol exposure.





3-111

-------
Rationale for

Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with Effects0

Epidemiologic
studies help rule out
chance, bias, and
confounding with
reasonable
confidence

Most prospective and some
cross-sectional studies found
associations with adjustment for
SES, maternal education, and
parental caregiving quality
(HOME score). Some also
considered parental IQ,
smoking, birth outcomes. A few
considered substance abuse,
nutritional factors, and family
history of psychiatric disorders.
Studies had population-based
recruitment with moderate to
high follow-up participation not
conditional on blood or tooth Pb
level.

Section 4.3.3.1, U.S. EPA (2013)

HOME score:

Liu etal. (2014b)

Fruh etal. (2019)

Family history of psychiatric disorders:

Choi etal. (2016)

SES and parental education:

Horton et al. (2018)

Ruebner et al. (2019)

Winter and Sampson (2017)

Boucher et al. (2012b)
Desrochers-Couture et al. (2019)

Consistent evidence
in animals with
relevant exposures

Several studies report increased
open-field activity in rodents
following developmental Pb
exposure, consistent with
hyperactivity.

Rodriques et al. (1996)
Moreira et al. (2001)
De Marco et al. (2005)
Duan etal. (2017)

Blood Pb: 19-28 |jg/dL in mice with lactational
exposure (tested PND 15-19); 10-29 |jg/dL in rats
with lactational exposure (tested PND 14-23)

Evidence from
lifetime Pb exposure
in nonhuman
primates suggests
that Pb produces
attention

decrements, which
supports the findings
in humans

Lifetime Pb exposure in
nonhuman primates was
reported to increase
distractibility in a spatial
discrimination task.

Gilbert and Rice (1987)

Blood Pb: 15-25 pg/dL

Evidence from lifetime Pb
exposure in nonhuman primates
suggests that Pb increased
perseveration and errors of
commission in a spatial
discrimination reversal task

Rice (1990)

Rice and Gilbert (1990b)
Gilbert and Rice (1987)

Blood Pb; 15-36 pg/dL

3-112

-------
Rationale for

Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with Effects0

Evidence describes

Both in vitro and in vivo Section 3.3





biologically plausible

evidence suggests that Pb





pathways

exposure may influence brain







development,







neurotransmission, connectivity,







neuronal integrity, all of which







may underlie the observed







alterations in externalizing







behaviors.





ADHD = attention deficit/hyperactivity disorder; avg = average; BLL = blood lead level; CKD = chronic kidney disease; Hg = mercury; HOME = Health Outcomes and Measures of the
Environment; mo = month(s); Pb = lead; PND = postnatal day; SES = socioeconomic status; yr = year(s).

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs CU.S. EPA. 20151.
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or inconsistencies.
References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-113

-------
3.5.3 Externalizing Behaviors: Conduct Disorders, Aggression, and Criminal
Behavior in Children, Adolescents, and Young Adults

There are two domains of conduct disorders that are considered in the ISA: undersocialized
aggressive conduct disorder, and socialized aggressive conduct disorder (Whitcomb and Merrell, 2012).
As discussed in the 2013 Pb ISA (U.S. EPA, 2013), these domains are combined in this assessment
because they cannot be disentangled based on the available epidemiologic literature. This section also
considers evidence for criminal offenses, which are associated with conduct disorders (U.S. EPA, 2013).
Although not described explicitly as a part of either domain of conduct disorders, evidence for criminal
offenses is reviewed with conduct disorders because conduct disorders can be predictors of subsequent
delinquency and criminality (Soderstrom et al., 2004; Babinski et al., 1999; Pajer, 1998).

The evidence reviewed in the 2013 Pb ISA is sufficient to conclude that a "causal relationship is
likely to exist" between Pb exposure and conduct disorders in children and young adults (U.S. EPA,
2013). Prospective studies consistently indicated that earlier childhood (e.g., age 30 months 6 years) or
lifetime average (to age 11-13 years) BLLs or tooth Pb levels (shed between ages 6-8 years and typically
measured in dentin, which reflects prenatal and/or child Pb exposure depending on the tooth layer
analyzed, see Section 2.3.4.1.) were associated with criminal offenses in young adults aged 19-24 years,
and with higher parent and teacher ratings of behaviors related to conduct disorders in children ages 7-17
years (see Table 4-12 of (U.S. EPA, 2013) and (U.S. EPA, 2006)). Pb-associated increases in conduct
disorders were found in populations with mean BLLs of 7-14 (ig/dL. These associations were found
without indication of strong selection bias and with adjustment for SES, parental education and IQ,
parental caregiving quality, family functioning, smoking, and substance abuse. Supporting evidence was
provided by cross-sectional studies of children participating in NHANES, e.g., (Braun et al„ 2008), and a
meta-analysis of prospective and cross-sectional studies (Marcus et al„ 2010). In addition, there was
coherence across related measures of conduct problems in epidemiologic studies. Evidence for Pb-
induced aggression in animals was mixed, however, with increases in aggression found in some studies of
adult animals with gestational plus lifetime Pb exposure but not juvenile animals. The strongest evidence
for the 2013 causality conclusion was provided by prospective epidemiologic studies, with support from
cross-sectional studies of criminal offenses and ratings of behaviors related to conduct disorders.
Associations with lower BLLs that were not influenced by higher earlier Pb exposures as in older children
and adults were not well characterized, however.

Studies published since 2013 from both cohort and cross-sectional studies add to this evidence
base, which continues to support a "likely to be causal" relationship, as described in Table 3-4 and below.
The central tendency Pb levels, study-specific details, and selected effect estimates are highlighted in
Table 3-9E.

3-114

-------
3.5.3.1 Epidemiologic Studies of Conduct Disorders, Aggression, and Criminal
Behavior in Children and Adolescents

The 2013 Pb ISA describes cohort and cross-sectional studies demonstrating associations of Pb
exposure with behaviors related to conduct disorders, including criminal offenses (U.S. EPA, 2013).
Collectively, the evidence from prospective cohort studies indicated associations of aggressive, antisocial,
delinquent, and criminal behavior with biomarkers of Pb exposure. Cross-sectional studies also provided
evidence on these associations, though there is more uncertainty in data from this study design due to
limitations in assessing temporality.

Recent studies have evaluated associations between Pb and conduct disorders, aggressive
behavior, and other related measures of externalizing behavior. Most of these studies were prospective
cohort studies (Tlotlcng et al„ 2022; Desrochers-Couture et al., 2019; Reuben et al., 2019; Beckwith et
al„ 2018; Nkomo et al„ 2018; Nkomo et al., 2017; Liu et al., 2014b; Sioen et al„ 2013; Boucher et al.,
2012b; Tatsuta et al., 2012). Several utilized self-report tools (e.g., Youth Self-Report [YSR], Buss-Perry
Aggression Questionnaire [BPAQ], Psychopathic Personality Inventory [PPI], Antisocial Behavior
Interview) to assess aggression, violence, or other socio-behavioral problems among adolescents and
young adults aged 14-24 years. These studies reported central tendency BLLs at ages 6.5-13 years
ranging from 2.3 to 8 (ig/dL or mean bone Pb of 8.7 (ig/dL (Tlotlcng et al„ 2022; Desrochers-Couture et
al„ 2019; Beckwith et al„ 2018; Nkomo et al„ 2018; Nkomo et al., 2017). In analyses adjusted for most
key confounders, associations were observed for: physical violence (|3: 0.05; 95% CI: 0.04, 0.05) (Nkomo
et al., 2017); direct aggression (|3 [95% CI] comparing those with BLLs >10 (ig/dL to those with BLLs <5
Hg/dL: 0.43 [0.08, 0.78]) (Nkomo et al.. 2018); anger aggression (|3 = 0.25 [95% CI: 0.04, 0.37])

(Tlotleng et al.. 2022); and PPI (overall |3 = 0.22 [95% CI: 0.06, 0.38]; female |3 = 0.16 [95% CI: -0.05,
0.37]; male |3 = 0.22 [95% CI: -0.02, 0.47]) (Beckwith et al.. 2018) (Table 3-9E). Although the PPI
serves as a measure of psychopathic personality traits, psychopathy more generally includes behavioral
factors such as aggression and criminal conduct, in addition to personality traits. As discussed in the 2013
Pb ISA (U.S. EPA, 2013). an analysis of this same cohort reported associations between BLLs and
criminal and violent criminal arrests at ages 19-24 (Wright et al.. 2008). (Beckwith et al.. 2018) also
noted that BLLs were associated with volumetric reductions in gray matter in the frontal lobe and white
matter in several brain regions. Considered together, these studies provide support for an association
between childhood Pb exposure and psychopathy in adolescents and young adults that may stem from
changes in brain morphology. In addition to these studies examining total effects, one prospective study
of Pb and self-reported behavioral outcomes conducted mediation analyses and reported an association
between Pb and adolescent externalizing behavior mediated through child externalizing behavior (|3: 0.18,
95% CI: 0, 0.36) (Desrochers-Couture et al., 2019). There was also evidence of a small but imprecise
direct effect, though there was likely limited power to detect a direct effect given the small sample size
and correlation between child and adolescent externalizing behavior. The observed association between
BLLs and adolescent externalizing behavior is at least partially mediated through child externalizing
behavior in this study population.

3-115

-------
Other prospective cohort studies used observer assessments (e.g., parent or teacher ratings) to
assess conduct disorders, aggression, and related behaviors among children with central tendency blood
or cord blood Pb ranging from 0.4 to 14.3 (ig/dL (Fruh et al.. 2019; Ruebner et al.. 2019; Liu et al..
2014b; Sioen et al.. 2013; Boucher et al.. 2012b; Tatsuta et al.. 2012). Some of these studies focused on
the prenatal period as a potentially sensitive period of exposure: three of these studies evaluated Pb levels
in cord blood (Sioen et al.. 2013; Boucher et al.. 2012b; Tatsuta et al.. 2012) and one evaluated second
trimester maternal BLLs (Fruh et al.. 2019). All but two (Ruebner et al.. 2019; Boucher et al.. 2012b) of
these studies evaluated the outcome in children with mean age <8 years. These analyses reported
generally null associations (Table 3-9E). It is possible that behavioral ratings are less reliable at younger
ages or that these outcomes manifest at later ages (Blair. 2001).

There were also some cross-sectional evaluations of blood Pb and behavioral problems (e.g.,
aggression, oppositional, externalizing, antisocial) covering children with central tendency BLLs ranging
from 0.7 to 11.08 (ig/dL (Liu et al.. 2022b; Desrochers-Couture et al.. 2019; Reuben et al.. 2019; Barg et
al.. 2018; Rodrigucs et al.. 2018; Boucher et al.. 2012b; Naicker et al.. 2012; Nigg et al.. 2010). These
analyses utilized a mix of self-report and observer assessment tools to evaluate the outcomes of interest in
children aged 6-13 years. Positive associations were reported in most studies, including for child
externalizing behavior (|3: 0.23; 95% CI: 0.08, 0.38) and child oppositional defiant and conduct disorder
(OD/CD) (|3: 0.37; 95% CI: 0.06, 0.69) (Desrochers-Couture et al.. 2019); "attacking people" (boys only,
see table for unstandardized estimate; (Naicker et al.. 2012)); teacher-reported aggressive and rule-
breaking behavior (referred to in the paper as "externalizing behavior") (log-transformed concurrent Pb |3:
0.14; 95% CI: 0.01, 0.26) (Boucher et al.. 2012b); parent-reported externalizing composite (|3 for SD
increase in symptoms scores per SD increase in loglO transformed BLL = 0.21 [95% CI: 0.05, 0.37]) and
oppositional behavior (|3 for SD increase in symptoms scores per SD increase in log 10 transformed BLL
= 0.09 [95% CI: -0.09, 0.27]) (Nigg et al.. 2010); antisocial behavior (|3 = 0.02 [95% CI: 0.00, 0.04])
(Reuben et al.. 2019); and antisocial/aggressive behavior factor (Parent-reported |3 = 0.20 [95% CI: 0.05,
0.34]; Child-reported |3 = 0.20 [95% CI: 0.04, 0.35]) (Liu et al.. 2022b). Both of the null studies evaluated
the outcome in groups of children that included individuals aged <8 years (Barg et al.. 2018; Rodrigues et
al.. 2018); it is possible that behavioral ratings are less reliable at younger ages or that these outcomes
manifest at later ages (Blair. 2001). More research is needed to disentangle these issues. Overall, cross-
sectional studies were less of a consideration in drawing conclusions on the effects of Pb, given their
inherent limitations with regard to temporality (given exposure assessment using BLLs).

Among studies published since the 2013 Pb ISA, there were also evaluations of the association
between Pb and suspensions, arrests, juvenile delinquency, and crime (including violent crime) (Wright et
al.. 2021; Emer et al.. 2020; Becklev et al.. 2018; Boutwell et al.. 2017; Amato et al.. 2013). The strongest
evidence comes from three prospective studies. Using data from the CLS on multiple measures of BLLs
(from prenatal to age 6 years; mean = 14.4 (ig/dL) and arrests from ages 18-33, (Wright et al.. 2021)
observed numerous positive associations, including for adult arrests (RR =1.01 [95% CI: 1.00, 1.03]),
lifetime arrests (RR= 1.02 [95% CI: 1.00, 1.03]), arrests for violent crime (RR= 1.02 [95% CI: 0.99,

3-116

-------
1.04]), and arrests for drug crime (RR= 1.03 [95% CI: 1.01, 1.061) (Wright et al.. 2021). In a cohort
based in Milwaukee, Wisconsin, (Emer et al.. 2020) reported that elevated mean and peak BLL prior to
age 6 was associated with increased risk of firearm violence perpetration (RR for mean BLL = 1.03 [95%
CI: 1.02, 1.04]; RR for peak BLL = 1.02 [95% CI: 1.01, 1.02] (Emer et al.. 2020). Additionally, (Amato
et al.. 2013) reported that Pb exposure during the first 3 years of life (based on a BLL >10 (ig/dL and <20
(ig/dL) increased the odds of school suspensions in fourth grade, compared with those without Pb
exposure during the first three years of life (based on a BLL <5 (ig/dL) (OR: 2.66; 95% CI: 2.12, 3.32)
(Amato et al.. 2013).

Supporting evidence comes from a prospective study with limitations that affect interpretation
and confidence as well as one ecologic study. Criminal offending, comprising both criminal conviction
and self-report offending, was evaluated in a prospective cohort study based in Dunedin, New Zealand in
which the mean 11-year-old BLL was 11.01 (ig/dL (Becklev et al.. 2018). In sex-adjusted analyses of
convictions, the authors reported that increased childhood BLL was associated with increased odds of at
least one nonviolent criminal conviction forages 15-38 years (OR: 1.05; 95% CI: 1.00, 1.10). However,
sex-adjusted analyses of other criminal conviction endpoints (e.g., any criminal conviction, recidivistic
conviction, one-time conviction, violent offense) were inconclusive (see Table 3-9E). While this study
had extensive follow-up (27 years) and both subjective and objective measures of the outcome, the
limited adjustment for potential confounders is a concern. Analyses were adjusted for sex, age was
controlled in the study design, and SES was evaluated as a potential confounder but determined not to be
associated with BLL; however, important covariates (i.e., parental IQ or education and HOME score)
were not considered, leaving open the possibility of residual confounding. In an ecologic study of 106
census tracts in St. Louis, Missouri, United States, (Boutwell et al.. 2017) reported that a 1% increase in
the proportion of elevated blood Pb tests (>5 (ig/dL) among children within a census tract was associated
with increased RRs for firearm crimes (RR: 1.03; 95% CI: 1.03, 1.04), assault crimes (RR: 1.03; 95% CI:
1.02, 1.03), robbery crimes (RR: 1.03;95%CI: 1.02, 1.04), and homicides (RR: 1.03;95%CI: 1.01,
1.04). The association with rape was inconclusive (RR: 1.01; 95% CI: 0.99, 1.03). While ecologic studies
can be useful for hypothesis generation and understanding patterns among groups, the lack of control for
individual-level confounding factors in such studies leaves concern for risk of bias.

3.5.3.1.1 Summary

Overall, recently published epidemiologic studies support the findings from the previous ISA.
The strongest evidence published since 2013 comes from prospective cohort studies of 1) self-reported
conduct and aggression-related outcomes (Tlotlcng et al.. 2022; Desrochers-Couture et al.. 2019;
Beckwith et al.. 2018; Nkomo et al.. 2018; Nkomo et al.. 2017). and 2) external measures of delinquency
(e.g., criminal arrests, school suspensions) (Wright et al.. 2021; Amato et al.. 2013). These studies
evaluated outcomes among individuals aged 7-33 years in relation to earlier (or cumulative) Pb levels.
BLLs were <10 (ig/dL in the studies of self-reported conduct and aggression-related outcomes and higher

3-117

-------
in studies of external measures of delinquency (e.g., (Wright et al.. 2021); mean 14.4 (ig/dL). These
studies controlled for most relevant confounders, and the prospective study design inherently ensured
appropriate temporality between the exposure and outcome. Additional supporting evidence comes from
cross-sectional studies using either self-report or observer-reported outcome measures among individuals
aged 6-13 years with concurrent BLLs ranging from 0.7-11.08 (ig/dL (Liu et al.. 2022b; Desrochers-
Couture et al.. 2019; Reuben et al.. 2019; Boucher et al.. 2012b; Naicker et al.. 2012; Nigg et al.. 2010).

3.5.3.2 Toxicological Studies of Aggression

There are no recent PECOS-relevant studies examining the relationship between Pb exposure and
aggression. Available toxicological studies of aggression were described in the 2006 Pb AQCD (U.S.
EPA, 2006) and 2013 Pb ISA (U.S. EPA, 2013). The evidence supported effects of Pb exposure on
changes in social behavior of rodents and nonhuman primates. In animals, the social behavior most
comparable to conduct disorders in children is aggression; however, the effects of Pb on aggression in
animals were inconsistent. In animals, aggression was assessed as threats, attacks, bites, chases, and
offensive posture in encounters with other animals. Pb exposure was found to have no effect on
aggression in some studies as well as to decrease and increase aggression in others. Pb exposure generally
was not found to affect aggression in juvenile animals; however, increased aggression was found in adult
animals with high concentrations of gestational plus postnatal dietary Pb exposure. Recent PECOS-
relevant studies have not further examined the effects of Pb on aggression. Additional reported effects on
social behaviors described in the 2006 Pb AQCD (U.S. EPA, 2006) and 2013 Pb ISA (U.S. EPA, 2013)
included Pb-induced increases in social and sexual investigation, as indicated by sniffing, grooming,
following, mounting, and lordosis behavior. Despite the limited new evidence, observations for Pb-
induced changes in aggression in animals provide support for associations of altered aggression outcomes
in children. Furthermore, many of the more general overt nervous system toxicology studies discussed in
Sections 3.4.2 and 3.3 assessed a variety of endpoints, including brain structural changes and
neurotransmitter analysis, that can contribute to understandings of the mechanistic underpinning of
observed behavioral changes providing additional biological plausibility.

3.5.3.3 Relevant Issues for Interpreting the Evidence Base

3.5.3.3.1 Concentration-Response Function

The evidence base for this outcome is more limited compared with that for cognitive deficits, and
the shape of the C-R function cannot be determined from available studies. However, it is important to
highlight that in studies reviewed for the 2013 Pb ISA, effects were observed at central tendency BLLs of
5-10 (ig/dL (Nigg et al.. 2008; Wright et al.. 2008; Chiodo et al.. 2007; Wasserman et al.. 2001) and <5

3-118

-------
(ig/dL (Braun et al.. 2008). Among studies published since 2013, effects on conduct disorder, aggression,
and crime were observed at central tendency BLLs of 5-10 (ig/dL (Tlotlcng et al.. 2022; Beckwith et al..
2018; Nkomo et al.. 2018; Nkomo et al.. 2017; Naicker et al.. 2012) as well as at central tendency BLLs
<5 (ig/dL (Liu et al.. 2022b; Desrochers-Couture et al.. 2019; Boucher et al.. 2012b; Nigg et al.. 2010).
However, it should be noted that there is less confidence in these studies of BLLs <5 (ig/dL as they were
all cross-sectional analyses (Liu et al.. 2022b; Desrochers-Couture et al.. 2019; Boucher et al.. 2012b;
Nigg et al.. 2010). Further work is needed to better understand whether the potential effects of Pb on this
outcome persist at BLLs <10 (ig/dL.

3.5.3.3.2 Potentially At-Risk Populations

Sex

The 2013 Pb ISA identified one study that evaluated the role of sex as an at-risk factor. Wright et
al. (2008) examined early life BLLs and criminal arrests in adulthood and reported that risks attributable
to Pb exposure were greater among males than females (Wright et al.. 2008).

Several new studies evaluated the role of sex as an at-risk factor through sex-stratified analyses of
Pb exposure and conduct disorders. The results were generally inconclusive regarding sex as an at-risk
factor. In a prospective study of blood Pb concentrations at 3-5 years and teacher-rated behavioral
problems at age 6 years, sex-stratified results were similar to non-stratified results, with null associations
for conduct disorder and aggression-related outcomes (Liu et al.. 2014b). In a prospective study of Pb
exposure and self-reported aggressive behavioral characteristics, associations for some outcomes (e.g.,
"attacks people") were observed in boys (but not observed or reported for girls); the authors suggested
this may be due to lower BLLs in girls compared with boys (Naicker et al.. 2012).

In a cross-sectional study of first grade children (mean 6.7 years) and teacher-rated behavioral
problems, sex-stratified results were generally null and similar to the non-stratified results. However,
some analyses indicated stronger associations among females (e.g., Behavioral Regulation Index (PR
[95% CI]: girls = 1.03 [1.00, 1.05]; boys = 0.99 [0.97, 1.01]), though the sample size was limited (n = 83
for girls) (Barg et al.. 2018). The authors suggested these results could be explained by teacher
expectations and perceptions of girls compared with boys, with effects on girls being more noticeable due
to gender norms and expectations rather than greater susceptibility to Pb exposure (Barg et al.. 2018).

Finally, in a study of BLLs measured at age 6.5 years and PPI between ages 19 and 24, sex-
stratified models indicated stronger associations in males, though associations were also present in
females (Beckwith et al.. 2018). Sex-stratified analyses indicated that Pb-associated gray matter volume
loss was only present in females, while Pb-related white matter loss was more widespread in males,
including an overlap in frontal white matter loss associated with both PPI scores and BLLs (Beckwith et
al.. 2018).

3-119

-------
Pre-existing Conditions

One study evaluated the association between Pb (median BLL 1.2 (ig/dL) and aggression/conduct
problems among children with CKD, a population at elevated risk of neurocognitive dysfunction (Gerson
et al., 2006; Gipson et al., 2004). No associations were observed (Ruebner et al.. 2019).

3.5.3.3.3 Confounding

The 2013 Pb ISA described multiple factors that influence conduct disorder and related outcomes
including sex, race, SES, parental education, parental IQ, and quality of the caregiving environment (i.e.,
HOME score) (U.S. EPA, 2013). These risk factors are often correlated with blood, tooth, and bone Pb
levels, and thus, are considered as potential confounding factors in epidemiologic analyses. As noted in
the 2013 Pb ISA, no single method to control for potential confounding is without limitation, and there is
potential for residual confounding by unmeasured factors. However, consistency of results across studies
utilizing different approaches to control for confounding can increase confidence across the body of
evidence.

Recent studies demonstrate associations between Pb exposure and conduct disorder after
controlling for different combinations of the aforementioned key covariates as well as additional relevant
covariates. However, it should be noted that in the current evidence base, the vast majority of studies that
identified associations did not specifically adjust for HOME score (Liu et al„ 2022b; Tlotlcng et al., 2022;
Desrochers-Couture et al., 2019; Reuben et al., 2019; Barg et al., 2018; Beckley et al„ 2018; Nkomo et
al., 2018; Rodrigues et al., 2018; AbuShady et al., 2017; Boutwell et al., 2017; Nkomo et al., 2017; Liu et
al„ 2014b; Amato et al„ 2013; Sioen et al., 2013; Boucher et al„ 2012b; Naicker et al., 2012; Tatsuta et
al., 2012; Nigg et al., 2010). Yet, most of these studies did adjust for other potentially related covariates
such as social adversity, house crowding, family violence, and SES, which mitigates some of the concern
about residual confounding due to exclusion of HOME score. Additionally, as highlighted in the previous
ISA, a meta-analysis by Marcus et al. indicated that the lack of adjustment for variables such as SES or
HOME score does not warrant limiting inferences from a particular study (U.S. EPA, 2013; Marcus et al.,
2010).

When there is uncertainty in epidemiologic evidence due to potential confounding, it is often
helpful to consider associated toxicological data. Aggressive behavior in rodents is mediated by several
brain regions, including the hypothalamus, prefrontal cortex, dorsal raphe nucleus, nucleus accumbens,
and olfactory system (Takahashi and Miczek, 2014) along with other neurochemical systems including
neurotransmitters, neuropeptides, and neuromodulators (i.e., serotonin, dopamine, vasopressin, oxytocin,
testosterone, estrogen, corticotrophin releasing factor, opioids, neuronal nitric oxidate synthase, and
monoamine oxidase A) (Takahashi and Miczek, 2014). Pb-induced changes on many of these
neurochemical endpoints has been reported and are described in Section 3.3, which lends some limited
yet relevant biological plausibility from the animal evidence without influence of potential confounding

3-120

-------
factors. While no new studies on Pb-induced aggressive behavior in mammals were identified with BLLs
of relevance to this ISA, the previous experimental animal studies support the evidence described in the
2013 Pb ISA.

3.5.3.3.4 Lifestages

Environmental exposures during critical lifestages spanning from childhood into adolescence can
affect key physiological systems that orchestrate brain development and plasticity (see Section 3.5.1.6.4).
Epidemiologic evidence assessed in the 2013 Pb ISA indicated associations of earlier childhood blood or
tooth Pb levels with behaviors related to conduct disorders in adolescents or adults (Fergusson et al..
2008; Wright et al.. 2008); however, these epidemiologic studies did not examine adult BLLs, thus the
relative influence of adult Pb exposure cannot be ascertained.

Recent studies observed associations of Pb exposure assessed via blood, cord blood, or bone
between delivery and age 13 years with outcomes evaluated among children, adolescents, and young
adults aged 7-33 years (Liu et al.. 2022b; Tlotleng et al.. 2022; Wright et al.. 2021; Desrochers-Couture et
al.. 2019; Reuben et al.. 2019; Beckwith et al.. 2018; Nkomo et al.. 2018; Nkomo et al.. 2017; Amato et
al.. 2013; Naicker et al.. 2012). Evidence published since 2013 is weaker for exposures that occur during
the prenatal period (Fruh et al.. 2019; Sioen et al.. 2013; Tatsuta et al.. 2012) and for most studies
assessing outcomes prior to the age of 8 years (Fruh et al.. 2019; Liu et al.. 2014b; Sioen et al.. 2013;
Tatsuta et al.. 2012). It is possible that outcome assessment tools that measure conduct disorder and
related aggressive traits are less reliable in this age group, aggressive patterns have not yet stabilized, or
the particular type of aggression associated with Pb exposure does not manifest until later years (Blair.
2001). Overall, Pb exposure during lifestages spanning childhood and into adolescence may confer risk
for conduct disorders and related outcomes.

3.5.3.3.5 Public Health Significance

The global prevalence of conduct disorders in 2019 was estimated to be 40.1 million (95% CI: 29
million, 52 million), with the highest burden experienced by individuals 0-14 years of age (GBP 2019
Mental Disorders Collaborators. 2022). Early life conduct disorders and other "antisocial behaviors" are
an important public health issue due to their persistence within an individual (Lvnam et al.. 2009). their
costs (both social and economic) to society (Sumner et al.. 2015; Mccollister et al.. 2010). and their
association with risk-taking behaviors, comorbid mental health conditions, and premature mortality
(Reves. 2015; Maughan et al.. 2014; Glenn et al.. 2013). For example, in one recent study based in New
Zealand, children with conduct problems accounted for 9.0% of the population but 53.3% of convictions,
15.7% of emergency department visits, 20.5% of prescription fills, 13.1% of injury claims, and 24.7 % of
welfare benefit months (Rivenbark et al.. 2018).

3-121

-------
3.5.3.4 Summary and Causality Determination: Conduct Disorders, Aggression, and
Criminal Behavior

The 2013 Pb ISA concluded that the relationship between Pb exposure and conduct disorders was
"likely to be causal" (U.S. EPA. 2013). This causality determination was primarily based on
epidemiologic evidence. In particular, prospective cohort studies provided key evidence of the association
between blood or tooth Pb levels and 1) parent or teacher ratings of delinquent, aggressive, and antisocial
behavior (Chandramouli et al„ 2009; Dietrich et al„ 2001; Burns et al„ 1999), and 2) criminal offenses
(Fergusson et al., 2008; Wright et al., 2008) in children and adolescents across diverse locations.
Supporting evidence was provided by cross-sectional studies of these outcomes (Braun et al„ 2008;
Chiodo et al., 2007).

Recent epidemiologic studies support the findings from the previous ISA. The strongest evidence
published since the 2013 Pb ISA comes from prospective cohort studies of 1) self-reported conduct and
aggression-related outcomes (Tlotlcng et al„ 2022; Desrochers-Couture et al„ 2019; Beckwith et al„
2018; Nkomo et al., 2018; Nkomo et al„ 2017), and 2) external measures of delinquency (e.g., criminal
arrests, school suspensions) (Wright et al„ 2021; Amato et al„ 2013). BLLs were <10 (ig/dL in studies of
self-reported conduct and aggression-related outcomes and higher in studies of external measures of
delinquency (e.g., (Wright et al„ 2021); mean 14.4 (ig/dL). These studies controlled for most relevant
confounders, and the study design inherently ensured appropriate temporality between the exposure and
outcome. Additional supporting evidence comes from cross-sectional studies using either self-report or
observer-reported outcome measures among individuals aged 6-13 years with concurrent BLLs ranging
from 0.7-11.08 (ig/dL (Liu et al., 2022b; Desrochers-Couture et al., 2019; Reuben et al„ 2019; Boucher et
al„ 2012b; Naicker et al„ 2012; Nigg et al., 2010). Although the evidence generally suggests positive
associations, null results may be explained by age at outcome or exposure. For example, many studies
with null associations evaluated the outcome in groups of children that included individuals <8 years of
age. It is possible that behavioral ratings are less reliable among this younger age group and/or abnormal
behaviors do not manifest until later in childhood. It should also be noted that both studies focusing
exclusively on newborn exposure (i.e., measurement of Pb in cord blood) were null, which potentially
indicates that the prenatal period may not be a relevant sensitive period of exposure for this outcome.
Studies that provide information on sensitive periods of exposure are limited.

Despite the growing epidemiologic evidence, the central uncertainty present in the 2013 Pb ISA
database remains: there is limited and inconsistent evidence from animal toxicological studies. Available
toxicological studies of aggression were described in the 2006 Pb AQCD (U.S. EPA, 2006) and 2013 Pb
ISA (U.S. EPA, 2013). No new PECOS-relevant studies examining the relationship between Pb exposure
and aggression have been reported. Despite the lack of new PECOS-relevant studies, Pb-induced changes
on many neurochemical endpoints that contribute to aggressive behaviors have been reported and are
described in Section 3.3, which lends biological plausibility from the animal evidence.

3-122

-------
In summary, there is sufficient evidence to conclude that there is likely to be a causal
relationship between Pb exposure and conduct disorders, aggression, and criminal behavior. This
causality determination is based on positive associations observed across various populations and based
on multiple outcome assessment approaches at relevant Pb exposure levels across recently published
prospective and cross-sectional epidemiologic studies. However, limitations remain in the animal
toxicology database, given the inconsistent evidence described in the 2013 Pb ISA and the lack of
relevant studies published since then. Yet, biological plausibility for these associations is supported by
human evidence linking early life Pb exposure to later life volumetric reductions in gray matter in the
frontal lobe and white matter in several brain regions (Bcckw ith et al.. 2018) and experimental animal
studies demonstrating Pb-induced changes on neurochemical endpoints relevant to this set of outcomes.

3-123

-------
Table 3-4 Summary of evidence for a likely to be causal association between Pb exposure and conduct
disorders, aggression, and criminal behavior in children and adolescents

Rationale for Causality	Key Evidence*	References*	Pb Biomarker Levels Associated

Determination3	y	with Effects0

Consistent results from
epidemiologic studies
with relevant blood or
bone Pb levels,
adequate control of
relevant confounders

Evidence from prospective studies
(demonstration of a temporal sequence)
using self-report measures of aggressive or
related externalizing behavior among
individuals ages 14-24 yr in relation to
earlier average blood Pb or bone Pb

Nkomo etal. (2017)
Nkomo etal. (2018)
Beckwith etal. (2018)

Tlotlenq et al. (2022)

Blood Pb: age 6.5-13 mean = 5.6-
pg/dL

Bone Pb: age 9 mean = 8.7 |jg/g

Evidence from prospective studies	Wright et al. (2021)	Blood Pb: prenatal to age 6 mean =

(demonstration of a temporal sequence) of Amato et al (2013)	>10 [jg/dL

arrests (ages 18-33 yr) and suspensions

(ages 9-10 yr) in relation to earlier average

blood Pb

Supporting evidence from cross-sectional
studies using both self-report and observer-
reported measures of aggressive or
externalizing behavior among individuals
ages 6-13 yr

Desrochers-Couture et al. (2019)

Naickeret al. (2012)

Liu et al. (2022b)

Boucher et al. (2012b)

Nigg et al. (2010)

(Reuben et al.. 2019)

Blood Pb (concurrent): age 6-13
mean = 0.7-11.08 pg/dL

Blood Pb: age 11 geometric mean =
2.3 |jg/dL

indicating indirect association of BLL on
adolescent externalizing behavior via child
externalizing behavior

Supporting evidence from a mediation	Desrochers-Couture et al. (2019)

analysis from a prospective cohort study

Supporting evidence from a prospective Beckwith et al. (2018)	Blood Pb: age 6.5 mean = 8.0 |jg/dL

study demonstrating association between

early life BLL and volumetric reductions in

gray matter in the frontal lobe and white

matter in several brain regions (mean age

26.8 yr)

3-124

-------
Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated
with Effects0

Most studies had sufficient adjustment for
relevant confounders. While most did not
adjust for HOME score specifically, they did
consider other related variables, such as
income, parental IQ, parental education,
SES, and/or neighborhood safety

Evidence strongest for outcomes assessed
among children >9 yr

Experimental animal Supporting evidence from animals exposed U.S. EPA (2013)

studies with relevant prenatally and postnatally

exposures provide

coherence and help rule

out chance, bias, and

confounding with

reasonable confidence

Biological plausibility Changes in key brain regions and	U.S. EPA (2013)

demonstrated	neurochemical systems implicated in

behavioral changes.

BLL = blood lead level; HOME = Health Outcomes and Measures of the Environment; IQ = intelligence quotient; SES = socioeconomic status; yr = year(s).

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 2015).

bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or

inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-125

-------
3.5.4

Internalizing Behaviors: Anxiety and Depression in Children

The evidence evaluated in the 2013 Pb ISA was sufficient to conclude that a "causal relationship
was likely to exist" between Pb exposure and internalizing behaviors in children (U.S. EPA, 2013).
Prospective studies in a few populations found associations of higher lifetime average blood Pb (mean:
-14 (ig/dL) or childhood tooth Pb (shed between ages 6-8 years and generally reflecting prenatal or early
child Pb exposure depending on the tooth layer analyzed, see Section 2.3.4.1) levels with higher parent
and teacher ratings of internalizing behaviors such as withdrawn behavior and symptoms of depression
and anxiety in children aged 8-13 years. There was no strong indication of biased reporting of behaviors
for children with higher BLLs. The few cross-sectional associations in populations with mean concurrent
BLLs of ~5 (ig/dL were inconsistent. Pb-associated increases in internalizing behaviors were found with
adjustment for maternal education and SES-related variables. Consideration for potential confounding by
parental caregiving quality was inconsistent. Despite some uncertainty in the epidemiologic evidence, the
biological plausibility for the effects of Pb on internalizing behaviors was provided by a small number of
experimental animal study findings with dietary lactational Pb exposure, with some evidence at BLLs
relevant to humans. Additional toxicological evidence demonstrating Pb-induced changes in the HPA axis
and dopaminergic and gamma-aminobutyric acid (GABA) systems provided additional support. Overall,
the strongest evidence was from prospective studies in a few populations of children and the coherence
with evidence from a small number of experimental animal studies with relevant Pb exposures. Some
uncertainty related to potential confounding by parental caregiving quality remained.

Measures of central tendency for Pb biomarker levels used in each study, along with other study-
specific details, including study population characteristics and select effect estimates, are highlighted in
Table 3-10E (Epidemiologic Studies) and Table 3-7T (Toxicological Studies). An overview of the recent
evidence is provided below. Overall, recent studies generally support findings from the 2013 Pb ISA.

3.5.4.1 Epidemiologic Studies of Internalizing Behaviors in Children

Several epidemiologic studies evaluated in the 2013 Pb ISA linked biomarkers of Pb exposure in
children with internalizing behaviors characterized by directing feelings and emotions inward, i.e.,
withdrawn behavior, symptoms of depression, fearfulness, and anxiety. These studies did not clearly
indicate that Pb exposure affected a particular domain of internalizing behaviors, i.e., withdrawn
behavior, somatic symptoms, anxiety, and depression. However, a consistent pattern of associations with
BLLs was observed across ages and across multiple internalizing behaviors. The strongest evidence was
provided by prospective studies conducted across multiple locations, i.e., Boston, Port Pirie, Australia,
and Yugoslavia (Wasserman et al.. 2001; Burns et al.. 1999; Wasserman et al.. 1998; Bellinger et al..
1994b). Collectively, these studies found associations between internalizing behaviors in children (ages
3-13 years) and Pb levels based on cord blood, concurrent blood (age 3 years), lifetime average blood,

3-126

-------
and teeth. Moderate to high follow-up rates in most studies increased confidence that selection bias did
not explain the pattern of associations observed in the studies. Factors, which were well documented to be
correlated with both Pb exposure and internalizing behaviors, including SES, parental caregiving quality
(speculated to mediate the potential correlation between parental psychopathology and Pb exposure), and
parental education, were considered as potential confounders across most studies. Although internalizing
behaviors are likely to have a strong familial component, the available evidence did not support parental
psychopathology as a direct confounder of the child Pb-internalizing behavior association. Studies that
included both teacher and parent ratings were emphasized. The most common instrument used to assess
internalizing behaviors was CBCL. Summary scores for internalizing behaviors, associated syndromes,
and DSM-IV scales (e.g., anxiety and depression) can be derived using CBCL.

Recent studies also analyze the association of Pb exposure with internalizing behaviors assessed
using CBCL. Using community survey data from the Project on Human Development in Chicago
Neighborhoods (PHDCN), Winter and Sampson (2017) examined the relationship between average BLL
in childhood (6 years old or younger) and anxiety or depression in adolescence (mean age 17 years old).
These authors found a 0.09 SD (0.03, 0.16) increase in anxiety or depression score, after adjustment for
covariates including caregiver education and SES. Participants were originally enrolled in the mid-1990s
and a random sample of those continuing to participate in 1999 and 2002 was randomly selected for this
study, with 67% of those selected agreeing to participate. Liu et al. (2014b) examined the association of
early childhood blood Pb concentration (3, 4, or 5 years old) with both parent and teacher ratings of
internalizing behavior at age 6 using CBCL and C-TRF, respectively. The outcomes were modeled as
both continuous and dichotomous variables (i.e., clinically significant behavior problems with T-score
>60) and adjusted for potential confounders including parent's educational level, father's occupation, and
child IQ. The emotional reactivity syndrome component of the teacher-rated internalizing problem scale
and the DSM-IV oriented anxiety were associated with child BLL when scores were modeled as
continuous terms (|3 = 0.32 [95% CI: 0.06, 0.59] and |3 = 0.25 [95% CI: 0.02, 0.50], respectively). The
ORs were 1.10 (95% CI: 1.03, 1.18) for the association of child BLL with clinically significant teacher-
reported internalizing behavior and 1.10 (95% CI: 1.01, 1.19) for clinically significant anxiety problems.
The participation rate was 81% in this study. The mean BLL of the children in this study was 6.4 (ig/dL
and the study had a high participation rate and included both teacher and parent ratings of internalizing
behavior.

Joo et al. (2018) analyzed data from the MOCEH study, a Korean prospective birth cohort of
mother-child pairs that were followed for 5 years. Maternal (early and late pregnancy), cord, and multiple
postnatal blood Pb concentrations were measured, and internalizing behaviors were assessed by the parent
using the Korean-CBCL at age five. The interaction between Pb exposure and child sex was evaluated
with further model adjustment for covariates including maternal educational level, and SES. Late
pregnancy and cord BLLs were associated with increasing internalizing behavior ratings in boys (|3 = 2.55
[95% CI: 0.22, 4.88] and |3 = 2.44 [95% CI: -0.74, 5.63], respectively), while postnatal (ages 2 and 5)
BLL was associated with increasing internalizing behavior ratings in girls (|3 = 2.94 [95% CI: 0.36, 5.52]

3-127

-------
and |3 = 5.65 [95% CI: 0.5, 10.8]). A total of 579 women of the 1751 originally enrolled in the cohort
provided data for this study.

Recent studies also examined the association of Pb exposure with internalizing behaviors using
SDQ. SDQ includes five scales (i.e., peer relationship problems, hyperactivity, emotional problems,
conduct problems and prosocial behavior) with the results for emotional problems discussed in this
section. Fruh et al. (2019) studied mother-child pairs participating in Project Viva, a longitudinal birth
cohort in eastern Massachusetts. Maternal blood Pb concentration in erythrocytes was measured during
the second trimester of pregnancy and parents rated their child's behavior using the SDQ (see also
Sections 3.5.1, 3.5.2, 3.5.3) in mid-childhood (median 7.7 years). The associations (i.e., P coefficients)
with the parent- and teacher-rated emotional components of the SDQ were 0.30 (95% CI: 0.05, 0.55) and
0.07 (95% CI; -0.22, 0.35), respectively. A stronger association with the emotional component of the
SDQ for girls compared with boys was reported by parents (P = 0.52 [0.18, 0.86] for girls versus P = 0.17
[95% CI: -0.17, 0.50] for boys). Note that the higher scores on the emotional problem scale indicate
worse performance. Behavior assessments and maternal blood Pb measurements were available for fewer
than half of study participants; however, important confounders including HOME score, maternal IQ, and
parental education were considered in this study. Sioen et al. (2013) analyzed data from a birth cohort
(FLEHS I, 2002-2006) comprising mother-infant pairs born in the Netherlands. This study examined the
association of cord blood for 281 infants whose parents returned the SDQ (26.4% response rate). No
association of cord blood Pb concentration with emotional symptom score >5 was observed (OR: 0.90
[95% CI: 0.52, 1.55] per doubling of BLL on log-scale).

Rokoff et al. (2022) used Couriers' Parent and Teacher Ratings Scales (CPRS and CTRS) at age 8
years and the BASC-2 self-report of personality (SRP) at age 15 years to assess internalizing behaviors
among children enrolled in a birth cohort study in New Bedford, MA. This study examined the
association of cord blood Pb with internalizing behaviors and also considered exposure to
organochlorines (hexachlorobenzene, p.p'-dichlorodiphenyl dichloroethylene, polychlorinated biphenyls)
and Mn, which were also measured in cord blood. BKMR analysis indicated linear associations and no
interactions between cord Pb, Mn, and organochlorines. Cord blood Pb was positively associated with
BASC anxiety score at age 15 (P = 1.78 [95% CI: 0.58, 2.99] BASC-2 SRP anxiety score increase per
doubling Pb) but not with Couriers' anxious-shy score at age 8 years. Additionally, a positive association
of cord blood Pb with depression score at age 15 was observed (P = 0.79 [95% CI: -0.39, 1.97]). The
Connor's psychosomatic score was positively associated with cord Pb, and this association was stronger
in boys (P = 2.08 [95% CI: 0.07, 4.10]) than in girls (P = 0.48 [95% CI: -1.00, 1.97]). A total of 528 of
the original 788 (67%) mother-infant pairs participated in the 15-year follow-up. The models were
adjusted for SES, maternal age, smoking, seafood, alcohol intake during pregnancy, maternal IQ, quality
of parental caregiving, and child characteristics (sex, race/ethnicity, age at assessment).

Several additional studies used the BASC-2 to assess associations with Pb exposure. Rasnick et
al. (2021) designed a study to identify sensitive time windows of exposure to Pb in air. These authors

3-128

-------
controlled for concurrent BLL (age 12 years) in their analysis of the Cincinnati Study of Allergy and Air
Pollution study data. Air Pb exposure was estimated using validated land use regression models, and
behavioral outcomes, including depression and anxiety, were assessed using the BASC-2 administered at
age 12. Models were adjusted for community deprivation, residential greenspace, and ECAT, in addition
to concurrent BLL. Distributed lag models that predicted outcome responses based on current and past
(i.e., lagged) predicted air Pb exposures identified a sensitive window in late childhood for anxiety but not
depression (Figure 3-12). The sensitive time window is indicated by months when the estimated 95% CI
did not include the null value.

BASC-2 = Behavior Assessment System for Children; edf =effective degrees of freedom.

The solid lines show the predicted change in score and the gray shading indicates the 95% CIs.

Source: Rasnick et al. ('20211.

Figure 3-12 Associations of monthly airborne Pb exposure levels from birth to
age 12 with scores for anxiety and depression behaviors on the
Behavior Assessment System for Children.

Ruebner et al. (2019) evaluated the association between BLLs and attention among children with
CKD. Internalizing behavior symptoms were assessed using the parental rating scales of the BASC-2,
which includes a composite score for internalizing problems. Associations between BLL and behavioral
symptoms on BASC-2 did not persist in models that were controlled for potential confounders including
race, poverty, maternal education, and clinical factors related to CKD. The median BLL in this study was
1.2 (.ig/dL.

Two additional prospective studies examined the association Pb concentration in teeth and
toenails with internalizing behavior on the BASC; these studies relied on a low proportion of the original
cohort, however. Horton et al. (2018) analyzed data from the ELEMENT Project birth cohort in Mexico
City to determine the association of weekly tooth Pb concentration (prenatal through 1 year postnatal)
with BASC-2 scores assessed between 8 and 11 years old. Approximately 12% of the original cohort was
enrolled in this study. Participants differed with respect to child birth weight and maternal IQ. A 0.4-unit

3-129

-------
increase in anxiety score was associated with a log-transformed unit increase in tooth Pb concentration at
12 months, while no consistent pattern of association was observed with increased internalizing behavior
symptoms overall. Dohertv et al. (2020) followed children enrolled in the New Hampshire Birth Cohort
Study (NHBCS) to examine the association of toenail Pb concentration with parent-rated internalizing
behaviors on the BASC-2. Data were available for approximately 300 of the 2000 women enrolled in the
study. No consistent pattern of association between pre- or postnatal toenail Pb concentration was
observed with internalizing behaviors after adjustment for confounders, including parent education and
parent perception of the parent-child relationship.

3.5.4.1.1 Summary

The 2013 Pb ISA included several prospective studies with moderate to high participation rates
that controlled for potential confounders including SES, parental education, and quality of parental
caregiving. These studies found associations of higher lifetime average blood (mean: -14 (ig/dL) or
childhood tooth Pb levels with higher parent and teacher ratings of internalizing behavior on the CBCL in
children aged 8-13 years. Several recent longitudinal epidemiologic studies with high to moderate
participation rates, which relied on an expanded array of instruments to assess internalizing behaviors
(i.e., CBCL, SDQ, CPRS, CTRS, and BASC-2), reported associations with blood Pb concentration
(childhood average, prenatal, and postnatal BLLs <7 (.ig/dL). Several studies in children evaluated sex
(Rokoff et al.. 2022; Fruh et al.. 2019; Joo et al.. 2018) as an effect modifier. The majority of analyses
controlled for important potential confounders including the quality of parental caregiving (Rokoff et al..
2022; Fruh et al.. 2019) maternal education and SES (Rokoff et al.. 2022; Fruh et al.. 2019; Winter and
Sampson. 2017; Liu et al.. 2014b). No association with internalizing behaviors was observed for the
blood Pb of children with CKD or in prospective studies of Pb concentration in blood (Sioen et al.. 2013).
teeth (Horton et al.. 2018). or toenails (Dohertv et al.. 2020). which reported relatively low participation
rates. The limited number of studies that aimed to distinguish types of internalizing behaviors indicated
associations with the anxiety component (Rokoff et al.. 2022; Rasnick et al.. 2021).

3.5.4.2 Toxicological Studies of Anxiety and Depression

Evidence in the 2013 Pb ISA consistently supported increases in emotionality in Pb-treated
animals. Postnatal exposure to Pb in female Long-Evans rats, resulting in mean BLLs between 13 and 31
(ig/dL, increased disruption and frustration in response to errors and reward omission in discrimination
task trials (Beaudin et al.. 2007; Stanglc et al.. 2007). Pb-exposed female Rhesus macaques displayed
increased negative responses to repeated tactile stimuli (i.e., tactile defensiveness) during adolescence
(mean BLLs of 31 (ig/dL) (Moore et al.. 2008). Furthermore, decreased exploratory behaviors in the
open-field test were also reported in male Wistar rats following Pb exposure from gestation through
weaning (Souza Lisboa et al.. 2005). Additional evidence for increased anxiety-like behavior, evaluated

3-130

-------
via the elevated plus maze, was found in one study of postnatally exposed rats (mean BLLs 35 (ig/dL at
weaning) (Fox et al.. 2010); however, no significant effects were found in another study of postnatal Pb
exposure (Molina et al.. 2011). Inconsistent evidence for depression-like behaviors measured in the forced
swim test was also reported (Souza Lisboa et al.. 2005; Stewart et al.. 1996).

Tests of anxiety-like behavior (i.e., emotionality) in rodents are often designed to exploit the
approach-avoidance conflict. Rodents must balance their motivation to explore novel environments (to
gather food and resources) with the need to evade predators and other threats. The open-field test (OFT)
allows for observation of rodent behavior within a bare, brightly lit, open area. Decreases in measures of
exploration (e.g., rearing, sniffing) indicate a shift towards an anxiety-like phenotype, although some
metrics may also be affected by other factors such as decreased motor function, to varying degrees. Basha
et al. (2014) found that postnatal Pb exposure in male rats decreased rearing and sniffing in the OFT
between PND 45 and 18 months, well after exposure was terminated. Grooming was also decreased at
PND 45, 4 months, and 12 months, which may indicate an altered response to stress in comparison to
controls. The same study also utilized the hole board test as another method to evaluate rodents" interest
in exploration of a novel environment. Animals displayed anxiety-like behavior (i.e., decreases in head
dip count and head dip duration) between PND 45 and 18 months. A follow-up study evaluated male
Wistar rats using a prenatal Pb exposure paradigm that resulted in BLLs of 11 (ig/dL at PND 21 and
found decreased exploratory behaviors in both the OFT and hole board test between PND 21 and 4
months (Basha and Reddv. 2015). Decreases in head dipping behavior were also reported by Flores-
Montovaand Sobin (2015). who evaluated male and female C57BL/6 mice following exposure to Pb
from PND 0 to PND 28 that resulted in low BLLs (mean between 3 and 12 (ig/dL). Further analysis of
individual BLLs and head dipping behavior suggested a negative association (i.e., head dipping behaviors
decreased as BLLs increased).

Enhanced thigmotaxis (i.e., tendency to remain close to the walls of the arena) within the OFT is
also associated with an anxiety-like phenotype. Betharia and Maher (2012) reported that low dose Pb
treatment had no significant effects on the latency of rodents to enter the center of the arena at PND 24 or
PND 59. The Sprague Dawley rats used in this study were exposed to Pb through their mothers from
gestation until PND 20 and had a mean BLL of 9 (ig/dL at PND 2, which decreased to <1 (ig/dL when
behavior was assessed. Another recent study found that adolescent Pb exposure (between PND 24 and
PND 56) in male Sprague Dawley rats, resulting in mean BLLs of 13 (ig/dL, significantly decreased the
time spent exploring the center of the arena compared with controls shortly after exposure was terminated
(Wang et al.. 2016). However, Shvachiv et al. (2018) found no significant effect of developmental Pb
exposure on adult Wistar rats using the same measure, despite employing a longer exposure paradigm that
resulted in higher BLLs than Wang et al. (2016). Interestingly, Abazvan et al. (2014) reported OFT
findings suggestive of an anxiolytic effect of Pb exposure (i.e., increased central activity and increased
rearing) in male transgenic mice that were heterozygous for mDISCl (associated with increased risk for
psychiatric disorders including schizophrenia) but phenotypically normal.

3-131

-------
Six recent studies have evaluated the potential anxiogenic effects of Pb using the elevated plus
maze (EPM). The EPM is comprised of four arms—two closed and two open (i.e., with or without
walls)—and anxious behavior is indicated by an increase in the preference for the closed arms (or
inversely, decreased preference for the open arms). Despite not finding conclusive results indicative of
increased anxiety in the OFT, Shvachiv et al. (2018) found that Pb-treated Wistar rats (male and female
adults exposed consistently or intermittently since gestation) spent significantly less time in the open arms
of the EPM. This finding was corroborated in a subsequent study by the same research group, which
investigated lifetime Pb exposure in Wistar rats and found that the percent of time animals spent in the
open arms significantly decreased at 12, 20, and 28 weeks of age (Shvachiv et al.. 2020). Interestingly,
the greatest decrease in open arm presence was observed at 20 weeks. Abazvan et al. (2014) also
demonstrated an anxiety-like phenotype using the EPM in 6-month-old male and female mice following
lifetime exposure to Pb. Tartaglionc et al. (2020) found that exposure to Pb from gestation to weaning
significantly decreased entries into the open arms, decreased head dipping behavior and stretch-attend
postures in female Wistar rats at PND 60 (mean BLLs of 25 (.ig/dL): however, only the decreases in
stretch-attend postures were observed in males. One study, Neuwirth et al. (2019a). found no significant
behavioral differences in the EPM in adolescent Long-Evans rats following gestational and
developmental exposure in either dosing group (peak BLLs 3-11 (ig/dL for lower dose group and 9-18
(ig/dL for higher dose group).

Sobolewski et al. (2020) investigated the potential for transgenerational effects of Pb on this
endpoint by exposing female C57BL/6J mice (F0) prior to mating and during gestation, resulting in
offspring (Fl) with BLLs of 10-15 (ig/dL at PND 6-7. The developmentally exposed F1 generation was
paired with unexposed mice at PND 60 to produce the F2 generation, and the process was repeated to
produce the F3 generation which had no direct Pb exposure. F3 females spent significantly more time in
the open arms of the EPM. This effect could be further traced to descendants of the F1 sire line instead of
the Fl dam line. No significant effects were detected in F3 males.

The influence of Pb exposure on rodent behavior in the forced swim test (FST) and tail
suspension test (TST) has also been evaluated in recent studies. These tests are classically considered
models of emotional despair, with animals exhibiting both escape-directed behaviors and periods of
immobility (e.g., floating or hanging). Originally used to screen for antidepressant drugs, decreases in
immobility in the FST or TST following chemical exposure are interpreted as an antidepressant effect;
however, it was recently suggested that immobility is instead an adaptive response to the acute stress of
the FST or TST, and decreased immobility may be reflective of a maladaptive coping strategy or,
potentially, an anxiety-like phenotype (Anvan and Amir. 2018; Molendijk and de Kloet. 2015). Corv-
Slechtaet al. (2013) reported that C57BL/6 mice which had been exposed to Pb from gestation to
adulthood had significantly decreased immobile bouts in the FST compared with control animals. In
another recent study, postnatal exposure to Pb in male and female CD1 mice significantly increased their
time spent resisting in the TST (Duan et al.. 2017). These recent results indicate that, at least under some
experimental testing paradigms (producing mean BLLs as low as roughly 6 (.ig/dL). Pb exposure results in

3-132

-------
what has classically been considered an antidepressant effect but may be more aptly attributed to an
altered response to stress.

3.5.4.2.1 Summary

Studies in the previous ISA consistently supported increases in emotionality in rodents and
nonhuman primates following developmental Pb exposure that produced mean BLLs as low as 13 (ig/dL.
Recent studies largely support and expand on this conclusion. Consistent decreases in rodent exploratory
behaviors in the OFT and hole board test (e.g., rearing, sniffing, head dipping) were found in Pb-exposed
rodents with peak BLLs from 3 to greater than 30 (ig/dL, lower than previously demonstrated. An
anxiety-like phenotype was also demonstrated in the EPM by multiple studies, with only one study
reporting null effects. Sobolewski et al. (2020) also demonstrated potential sex-specific transgenerational
effects of Pb exposure on this endpoint. Inconsistent effects of Pb on thigmotactic behavior were reported
by a few studies, which was not an endpoint discussed in the previous ISA. Two studies demonstrated
decreased immobility in classical tests of depression-like behavior, suggestive of an antidepressant effect,
but the relevance of these tests to human depression is unclear. While limited studies reported null results,
they were not stronger with respect to design or methodology and did not significantly weaken the larger
body of evidence.

3.5.4.3 Relevant Issues for Interpreting the Evidence Base

3.5.4.3.1 Concentration-Response Function

Bayesian kernel machine regression (BKMR) and five-chemical linear regression models were
used to examine covariate adjusted associations between Pb exposure and CPRS Anxious-Shy T-score at
age 8 and BASC-second revision Anxiety T-score at age 15 Rokoff et al. (2022). BKMR analysis
indicated linear associations between Pb exposure and these outcomes, and no interactions between cord
Pb, Mn, and organochlorines.

3.5.4.3.2 Potentially At-Risk Populations

The 2013 Pb ISA did not describe populations of children potentially at higher risk of Pb-
associated internalizing behaviors. Recent epidemiologic studies presented sex-stratified results or
examined interactions between Pb exposure and other chemicals.

3-133

-------
Sex

Fruh et al. (2019) studied mother-child pairs participating in Project Viva, a longitudinal birth
cohort in eastern Massachusetts. This study found a stronger association of maternal BLL with the
emotional component of the SDQ measured in mid-childhood for girls compared with boys (|3 = 0.52
[0.18, 0.86] for girls v. (3 = 0.17 [95% CI: -0.17, 0.50] for boys). Note that higher scores on the emotional
problem scale indicate worse performance. In another study, Joo et al. (2018) found that late pregnancy
and cord BLL was associated with increasing internalizing behavior ratings on the CBCL in boys (|3 =
2.55 [95% CI: 0.22, 4.88] and |3 = 2.44 [95% CI: -0.74, 5.63], respectively), while postnatal (age 2 and 5)
BLL was associated with increasing internalizing behavior ratings on the CBCL in girls (|3 = 2.94 [95%
CI: 0.36, 5.52] and |3 = 5.65 [95% CI: 0.5, 10.8]). In a study that used Conners" rating scale to ascertain
internalizing behaviors, Rokoff et al. (2022) found the psychosomatic score was positively associated
with cord Pb and this association was stronger in boys than in girls (|3 = 2.08 [95% CI: 0.07, 4.10] versus
|3 = 0.48 [95% CI: -1.00, 1.97]). Of the experimental animal studies that evaluated both sexes, a small
number identified behavioral changes in Pb-exposed females while detecting minimal or no changes in
their male counterparts on the EPM, which could indicate that females are more sensitive to changes in
anxiety-like behavior after exposure to Pb (Sobolewski et al.. 2020; Tartaglionc et al.. 2020). Overall, no
consistent pattern was observed across the limited number of epidemiologic and toxicologic studies that
presented sex-stratified results. Each study used a different instrument to ascertain the outcomes.

Other Metals

A recent study examined the interaction effect between prenatal Pb exposure and other metals on
internalizing behavior scores on the BASC and the CPRS. BKMR analysis indicated no interactions
between cord blood Pb, Mn, and organochlorines that would indicate a deviation from additivity in a
study by Rokoff et al. (2022).

3.5.4.3.3 Lifestages

Epidemiologic studies consistently show that BLLs measured during various lifestages and time
periods, including the prenatal period, early childhood, and later childhood, and averaged over multiple
years, are associated with increases in internalizing behaviors. The identification of critical lifestages and
time periods of Pb exposure is complicated further by the fact that BLLs in older children, although
affected by recent exposure, are also influenced by Pb stored in bone due to rapid growth-related bone
turnover in children relative to adults. Thus, associations of neurodevelopmental effects with concurrent
BLL in children may reflect the effects of past and recent Pb exposures. Recent prospective studies add to
the evidence from the strongest studies in the 2013 Pb ISA that found associations with childhood
average blood and tooth Pb levels in children. These recent studies found associations between
internalizing behaviors and early childhood, maternal, and cord BLLs. Toxicological studies also provide

3-134

-------
support that the sensitive exposure window is not limited to a single phase of development. Rather,
effects of Pb exposure on anxiety or depression-like behavior in animals have been found following
gestational and postnatal exposure, exposure starting in adolescence, and lifetime exposure.

3.5.4.4 Summary and Causality Determination of Internalizing Behaviors in Children

The 2013 Pb ISA concluded that a causal relationship was likely to exist between Pb exposure in
children and internalizing behaviors based on the available evidence (U.S. EPA, 2013). Prospective
studies demonstrated associations between higher average blood (roughly 14 (ig/dL) or tooth Pb (i.e.,
reflective of prenatal or early postnatal Pb exposure depending on the tooth layer analyzed) levels and
higher parent and teacher ratings of internalizing behaviors, including withdrawn behavior and symptoms
of depression, fearfulness, and anxiety in children (aged 8-13). These associations were present after
adjustment for SES, birth outcomes, and parental education, but some uncertainty regarding potential
confounding by parental caregiving quality remained. Results from cross-sectional studies evaluating
lower concurrent BLLs (5 (ig/dL) were inconsistent. Increased emotionality in rodents and monkeys was
demonstrated at BLLs as low as 13 (ig/dL after exposure to Pb during development, and biological
plausibility was supported by findings of alterations in the HPA axis and dopaminergic and GABAergic
systems.

Several recent longitudinal epidemiologic studies with high to moderate participation rates relied
on an expanded array of instruments to assess internalizing behaviors (i.e., CBCL, SDQ, PRS, CTRS, and
BASC-2) compared with the studies in the 2013 Pb ISA. These studies observed associations with blood
Pb exposure (early childhood and prenatal BLLs <7 (.ig/dL). A limited number of studies evaluated child
sex (Rokoff et al., 2022; Fruh et al., 2019; Joo et al., 2018) as an effect modifier but were not consistent
with regard to sex-specific effects. The majority of analyses controlled for important potential
confounders including the quality of parental caregiving (Rokoff et al., 2022; Fruh et al., 2019), maternal
education, and SES (Rokoff et al., 2022; Fruh et al., 2019; Winter and Sampson, 2017; Liu et al., 2014b);
however, each potential confounder was not uniformly considered across studies. No association between
blood Pb and internalizing behaviors was observed among children with CKD or in prospective studies of
Pb concentration in blood (Sioen et al., 2013), teeth (Horton et al., 2018) or toenails (Doherty et al.,
2020), which reported relatively low participation rates. The limited number of studies that aimed to
distinguish types of internalizing behaviors indicated associations with the anxiety component (Rokoff et
al„ 2022; Rasnick et al„ 2021). Recent studies that found associations with prenatal or cord BLLs add to
the evidence. Uncertainty remains, however, regarding the exposure patterns associated BLLs in older
children and adults.

Recent experimental animal studies provide coherence with the previous findings that moderate
to high peak BLLs (12 to >30 (ig/dL) increase anxiety-like behaviors on the EPM, hole board test, and
OFT following Pb exposure during a single developmental window (including prenatal (Basha and

3-135

-------
Reddv. 2015). postnatal (Basha et al.. 2014). or adolescent periods (Wang et al.. 2016)) or throughout
development and beyond (Shvachiv et al.. 2020; Tartaglionc et al.. 2020; Shvachiv et al.. 2018; Abazvan
et al.. 2014). Overall, experimental animal studies provide more extensive support for anxiety-like
behaviors than for depression-like behaviors. However, two recent studies reported that Pb exposure
decreased immobility in classical tests of emotional despair following postnatal or lifetime Pb exposure
(Duan et al.. 2017; Corv-Slechta et al.. 2013). In addition to the well demonstrated effects at moderate to
high BLLs, two recent studies found altered behaviors in a nose poke task and FST following Pb
exposures resulting in low BLLs (3.2-10 (.ig/dL); moreover, one study was able to demonstrate exposure-
response relationships (i.e., higher BLLs were associated with greater behavioral changes) (Flores-
Montovaand Sobin. 2015).

Overall, the evidence is sufficient to conclude that there is likely to be a causal relationship
between Pb exposure and internalizing behaviors in children. This determination is based on
consistent evidence from both recent and past prospective epidemiologic studies, which demonstrate
positive associations between average blood Pb (prenatal, early childhood, lifetime) or childhood tooth Pb
levels (generally reflecting prenatal or early postnatal exposure) and multiple measures of internalizing
behaviors in children (aged 4-17) after adjustment for multiple confounding factors (e.g., SES, birth
outcomes, parental education). Recent toxicological studies provide further support for anxiety-like
behaviors following developmental and cumulative exposures that result in BLLs that are relevant to
humans. Despite these findings, some uncertainties have not been addressed in the epidemiologic
literature, including full consideration of certain confounding factors (e.g., parental caregiving quality)
and uncertainty regarding the exposure patterns associated with observed BLLs. Furthermore,
inconsistencies remain in the limited number of cross-sectional studies available in populations with
BLLs below 5 (ig/dL.

3-136

-------
Table 3-5 Summary of evidence for a likely to be causal relationship between Pb exposure and internalizing
behaviors in children

Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0

Consistent results from
prospective epidemiologic
studies with relevant
exposures

Evidence from prospective studies for
higher ratings of internalizing behaviors in
children ages 8-13 yr in Boston and Port
Pirie cohorts in association with tooth or
lifetime average BLLs.

Section 4.3.4.1, (U.S. EPA, 2013)
Burns et al. (1999)

Bellinger et al. (1994b)

Blood Pb lifetime (to age 11-13 yr)
average mean: -14 |jg/dL

Tooth Pb (age 6 yr) mean: 3.4 |jg/g

Evidence from prospective studies for
higher rating of internalizing behaviors in
children 6-17 yr (cohorts in eastern MA,
Chicago, Cincinnati, and China) in
association with early childhood and
prenatal BLLs.

Winter and Sampson (2017)
Liu et al. (2014b)

Fruh et al. (2019)

Rokoff et al. (2022)

Early childhood <7 |jg/dL (median/mean)
Maternal and cord blood Pb, <2 |jg/dL
(median)

Associations also found in children aged
4-5 yr in former Yugoslavia in association
with lifetime average BLL

Wasserman et al. (2001)

Blood Pb lifetime (to age 4-5 yr) average
mean: 7.2 |jg/dL

Prospective studies had population-based
recruitment with moderate follow-up
participation. Participation not conditional
on tooth/BLLs and behavior

Inconsistent results in cross-sectional
studies with mean BLLs < 5

Section 4.3.4.1, (U.S. EPA, 2013)

Uncertainty regarding	Epidemiologic associations found with Section 3.7, Table 3-10E

potential confounding	adjustment for SES, birth outcomes,

parental education. Studies did not
uniformly adjust for parental caregiving
quality.

Uncertainty regarding the Uncertainty in regarding past exposure in
exposure patterns associated older children,
with observed BLLs.

3-137

-------
Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0

And, supporting animal
evidence with relevant
exposures from multiple
studies

Gestational, lactational, and adolescent
exposures increasing anxiety-like
behaviors and altered stress coping
response.

Corv-Slechta et al. (2013)
Flores-Montova and Sobin (2015)
Shvachiv et al. (2020)

Peak BLLs: 3-27 pg/dL

BLL = blood lead level; Pb = lead; yr = year(s); SES = socioeconomic status.

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 2015).
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or
inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-138

-------
3.5.5

Motor Function in Children

The evidence assessed in the 2013 Pb ISA is sufficient to conclude that a "causal relationship is
likely to exist" between Pb exposure and decrements in motor function in children. Evidence from
prospective studies of Cincinnati and Yugoslavia birth cohorts indicated associations of decrements in
fine and gross motor function with higher neonatal, concurrent, and lifetime average BLLs in children
aged 4.5-6 years and with higher earlier childhood (ages 0-5 years on average, age 78 months) BLLs in
children aged 15-17 years (Bhattacharya et al„ 2006; Ris et al., 2004; Bhattacharya et al., 1995; Dietrich
et al., 1993). The means for these blood Pb metrics ranged from 4.8 to 12 (ig/dL. These associations were
found with adjustment for several potential confounding factors, including SES, parental caregiving
quality, and child health with no indication of substantial selection bias. Evidence from cross-sectional
studies was less consistent, however (see Section 4.3.8 of (U.S. EPA. 2013)). The biological plausibility
for associations observed in children was supported by a study that found poorer balance in male mice
with relevant gestational to early postnatal (PND 10) Pb exposures. Overall, the strongest evidence was
from a small number of prospective cohort studies of children with limited support from studies in mice
with relevant exposures.

Measures of central tendency for Pb biomarker levels used in each study, along with other study-
specific details, including study population characteristics and select effect estimates, are highlighted in
Table 3-1 IE (Epidemiologic Studies) and Table 3-1 IT (Toxicological Studies). An overview of the recent
evidence is provided below. Overall, recent epidemiologic studies support findings from the 2013 Pb ISA
and a limited number of recent experimental animal studies provide coherence for their observations
demonstrating effects at relevant exposure concentrations.

3.5.5.1 Epidemiologic Studies of Motor Function

Evidence from prospective studies of Pb exposure and decrements in motor function in the 2013
Pb ISA indicated associations between higher neonatal, concurrent and lifetime average BLLs and motor
function decrements. Several recent epidemiologic studies examined the association between Pb exposure
and decrements in motor function in children. The findings generally support an association between Pb
exposure and decrements in motor function; however, they varied by the specific measure of motor
function as well as the timing of exposure measurement. Most studies were cohort studies and assessed
motor function using a comprehensive motor score, such as the Psychomotor Developmental Index (PDI)
score, from a version of the BSID (Jiang et al.. 2022; Kao et al.. 2021; Rygiel et al.. 2021; Shekhawat et
al.. 2021; Kim et al.. 2018b; Y Ortiz et al.. 2017; Paraiuli et al.. 2015b; Paraiuli et al.. 2015a; Liu et al..
2014c; Kim et al.. 2013c; Henn et al.. 2012). A few studies used a motor score from the Chinese version
of the GDS (Liu et al.. 2022a; Zhou et al.. 2017). The remaining studies assessed specific tasks, such as

3-139

-------
balance, manual dexterity, coordination, and fine motor speed (Taylor et al.. 2018; Boucher et al.. 2016;
Taylor et al.. 2015).

Studies using the Bayley scales to measure motor function in infants and toddlers (i.e., through
age 3) generally found associations between some Pb exposure metrics and decreased motor score. Kim et
al. (2013c). Kim et al. (2018b). Y Ortiz et al. (2017). Liu et al. (2014c). Rvgiel et al. (2021). and
Shekhawat et al. (2021) observed a decrease in motor score using maternal or cord BLLs, as well as other
blood Pb metrics, in several birth cohorts in multiple countries. Associations between BLLs and PDI are
presented in Figure 3-13.

3-140

-------
Study

Prospective Studies

TKim etal.2013

tKim etal.2018

tLiu et al. 2014
TRygiel etal.2021

Location

3 Cities, S Korea

4 Cities, S Korea

Guangdong, China
Mexico City, Mexico

Blood Pb	Mean	Age at Outcome

(Mg/dl_)	(months)

Prenatal (early pregnancy) 1.4 (GM)	6

Prenatal (late pregnancy) 1.3 (GM)	6

Prenatal (early pregnancy) 1.4 (GM)	6

Prenatal (late pregnancy) 1.3 (GM)	6

Prenatal (early pregnancy) 1.4 (GM)	6

Prenatal (late pregnancy) 1.3 (GM)	6
Prenatal (late pregnancy) 2.7 (median) 13-24

Prenatal (late pregnancy) NR	13-24

Prenatal (late pregnancy) NR	13-24
Prenatal (cord) 5.63 (ref: 1.35) 36

Prenatal (T1)	5.27 (GM)	12

Prenatal (T2)	4.74 (GM)	12

Prenatal (T3)	4.98 (GM)	12

Prenatal (T1)	5.27 (GM)	24

Prenatal (T2)	4.74 (GM)	24

Prenatal (T3)	4.98 (GM)	24

Strata

Cd <1.47 jjg/L
Cd >1.47 pg/L
Cd <1.51 pg/L
Cd >1.51 pg/L

Boys	*

Girls	"

Pb £3.92 vs. *1.89 pg/dL

Claus Henn et al. 2012 Mexico City, Mexico

Child (12 months)
Child (24 months)

5.1
5

12-36
12-36

	1	1	1	1	

-15.00	-10.00	-5.00	0.00

Beta values (95% CI) per 1 ug/dL increase in blood Pb

Figure 3-13 Associations between biomarkers of Pb exposure and Bayley Score of Infant Development
Psychomotor Developmental Index.

Note: Effect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb. If the Pb biomarker is log-transformed, effect estimates are standardized to
the specified unit increase for the 10th -90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated interval. Categorical effect
estimates are not standardized. Associations that could not be standardized are not included on the plot.
tStudies published since the 2013 Integrated Science Assessment for Lead.

3-141

-------
Kim et al. (2013c) found that PDI score at 6 months of age decreased with increasing BLLs
measured in the third trimester (median = 39th week) (|3 = -1.38 [95% CI: -3.31, 0.55] per 1 (ig/dL
increase in BLL) in the Korean MOCEH study. Another Korean study using the CHECK cohort (Kim et
al.. 2018b) also observed decrements in PDI among 13-24 month old infants in association with perinatal
maternal BLLs (|3 per l-(ig/dL blood Pb = -15.45 [95% CI: -30.12, -0.79]). Sex-stratified results were
slightly negative but not significant. In China, Liu et al. (2014c) observed an association between
increasing prenatal (umbilical cord blood) Pb levels and worse PDI score at 36 months of age. Compared
with low prenatal Pb (<1.89 (.ig/dL). children exposed to high prenatal Pb (>3.92 (ig/dL) were more likely
to have a lower PDI score (|3 = -1.30 [95% CI: -1.57, -1.03]), after adjusting for potential confounders.

Several studies in Mexico also examined the association of BLL with PDI assessed in infants.
Rvgiel et al. (2021) found a small negative association between prenatal (trimester-specific) BLLs and
PDI scores at 12 months in the ELEMENT Project study (|3 per 1 (ig/dL increase in 1st trimester Pb =
-0.24 [95% CI: -0.95, 0.48]; |3 per 1 (ig/dL increase in 2nd trimester Pb = -0.38 [95% CI: -1.10, 0.35]; |3
per 1 (ig/dL increase in 3rd trimester Pb = -0.33 [95% CI: -1.06, 0.40]). At 24 months, the negative
association persisted but with a smaller magnitude of effect. Rvgiel et al. (2021) also examined whether
DNA methylation mediated the association and found that DNA methylation of cgl8515027 located
within glucosaminyl (N-acetyl) transferase 1 (GCNT1) had a suppressive (positive indirect) effect on the
inverse relationship between second trimester BLLs (ln-transformed) and PDI scores at 12 months (|3indirect
= 1.25 (95% CI: -0.11, 3.32]). In the Programming Research in Obesity, Growth, Environment and
Social Stressors (PROGRESS) birth cohort in Mexico, Y Ortiz et al. (2017) found a negative association
between motor score at 24 months of age and log-transformed BLLs measured during the third trimester
(|3 = -11.01 [95% CI: -17.55, -4.48]), but not for BLLs measured during the second trimester (|3 = 1.97
[95% CI: -2.46, 6.40]). In a study using childhood BLLs, Henn et al. (2012) found a small negative
association with repeated measures of PDI scores in another cohort of children in Mexico. From adjusted
mixed-effects models with repeated measures of PDI scores at 12, 18, 24, 30, and 36 months, there was a
negative association between 12-month blood Pb and PDI scores (|3 per 1 -(.ig/dL blood Pb = -0.27 [95%
CI: -0.56, 0.02]). Similarly, from adjusted mixed-effects models with repeated measures of PDI scores at
24, 30, and 36 months, there was a negative association between 24-month blood Pb and PDI scores (|3
per 1-ng/dL blood Pb = -0.18 [95% CI: -0.53, 0.17]).

Several studies are not pictured in Figure 3-13. Shekhawat et al. (2021) found that children with
cord blood Pb concentrations of 5-10 (ig/dL had reduced gross motor skills on the BSID at an average
age of 6.5 months (|3 = -0.29 [95% CI: -5.00, 0.11]) for each 1 (ig/dL increase in cord BLL. Additionally,
in a birth cohort of mother-child pairs recruited from Bharatpur General Hospital in Nepal, Paraiuli et al.
(2015a) and Paraiuli et al. (2015b) assessed the association of cord BLLs with PDI at 24 months old and
36 months of age, respectively. Negative but non-significant associations were observed between log-
transformed cord BLLs and 24-month PDI (|3 = -4.83 [95% CI: -16.53, 6.86]) nor 36-month PDI (|3 =
-2.56 [95% CI: -9.71, 4.59]). Notably, two additional studies that used biomarkers other than blood did
not find associations, i.e., Jiang et al. (2022) measured Pb in meconium (at birth) and in hair and

3-142

-------
fingernails (at 3 years of age) in Taiwan and did not find an association with any motor score (total, fine
motor, or gross motor) at 3 years of age. Another study in Taiwan (Kao et al.. 2021) that used hair and
fingernail biomarkers of Pb concentrations similarly did not report significant associations with motor
development among infants less than 3 years old.

Several other instruments were used to assess motor function in infants and toddlers. Paraiuli et
al. (2013) measured Pb, As, and Zn levels in cord blood and used the third edition of the Brazelton
Neonatal Behavioral Assessment Scales (NBAS III) to assess neurodevelopment in one-day-old newborns
in Chitwan, Nepal. The NBAS III contains 27 behavioral and 18 reflex items and is used for infants up to
2 months old. The multivariate model was adjusted for parity, family income, mother's age, education,
BMI, birth weight, gestational age, and age in hours at NBAS assessment. The NBAS motor cluster score
was inversely associated with the log-transformed cord BLLs (|3 = -2.15 [95% CI: -4.27, -0.03]). Liu et
al. (2014d) used the Neonatal Behavioral Neurological Assessment (NBNA), which is based on the
NBAS and has five clusters of behavior: passive tone, active tone, primary reflexes, and general
assessment. Newborns in this study were assessed at 3 days old, and the NBNA has been validated among
Chinese newborns between 2 and 28 days old. Associations between maternal BLL in the first trimester
and the NBNA scores were observed (|3 = -4.86 [95% CI: -8.83, -0.89] per unit of log-transformed Pb).
Less precise associations of second trimester, third trimester, and cord BLLs with decreased motor
function were also observed. Among toddlers (2-3 years old) Zhou et al. (2017) and Liu et al. (2022a)
both used the Chinese version of the GDS to calculate a motor score. For every loglO (|ig/dL) increase in
maternal blood Pb (measured at 28-36 weeks of gestation), Zhou et al. (2017) observed a positive
association for gross motor development ([3 = 3.31 [95% CI: -6.11, 12.73] per log-10 transformed unit of
BLL) as well as fine motor development (|3 = 0.49 [95% CI: -11.27, 12.24] per log-10 transformed unit
of BLL); however, the effect estimates were extremely imprecise. On the other hand, for each In (|ig/L)
increase in maternal Pb, Liu et al. (2022a) observed a negative association for gross motor development
(|3 = -2.32 [95% CI: -3.61, -1.03] per ln-transformed unit of BLL). Furthermore, Nvanza et al. (2021)
did not find associations between high Pb exposure and fine or gross motor impairment assessed by the
MDAT.

Several additional studies were conducted using assessment instruments that measure children's
(7 years or older) ability to perform certain tasks. In the ALSPAC, Taylor et al. (2015) conducted a heel-
to-toe test in children at age 7 years, beam walking test (to measure dynamic balance) at age 10 years, and
balancing test with eyes closed (to measure static balance) also at age 10 years. Pb levels measured in
maternal blood (<18 weeks of gestation) and Pb levels measured in child blood (30 months old) were not
associated with any measure of motor function in this study (Taylor et al.. 2015). In another analysis of
ALSPAC data, Taylor et al. (2018) examined the association between first trimester BLLs and different
measures of coordination. Compared with prenatal blood Pb <5 (ig/dL, children exposed to higher levels
(>5 (ig/dL) of prenatal Pb were more likely to fail the tests of manual dexterity (threading lace, peg board
using preferred hand, and peg board using non-preferred hand). When comparing the highest blood Pb
quartile to the lowest blood Pb quartile, the only association remaining was for failing the peg board using

3-143

-------
the preferred hand (OR for quartile 4 versus quartile 1 = 1.23 [95% CI: 0.92, 1.66]). Prenatal Pb exposure
was not associated with tests of balance and the results were inconsistent for ball skills (inverse
association for >5 (ig/dL versus <5 (ig/dL; positive association for quartile 4 versus quartile 1). In the
Nunavik Child Development Study in Canada, Boucher et al. (2016) measured manual dexterity, fine
motor speed, and visuomotor integration in children (ages 8.5-13.3 years). BLLs (log-transformed)
measured at birth (cord blood) and at age 11 years were negatively associated with manual dexterity (|3
for cord blood Pb = -0.08 [p > 0.10]; (3 for child blood Pb = -0.17 [95% CI: -0.34, 0.00]) and fine motor
speed (|3 for cord blood Pb = -0.19 [95% CI: -0.33, -0.05]; |3 for child blood Pb = -0.21 [95% CI: -0.37,
-0.05]). For visuomotor integration, there was no association with cord blood Pb (|3 for cord blood Pb =
-0.01 [p > 0.10]) and apositive association with child blood Pb (|3 for child blood Pb = 0.10 [p > 0.10]).
The magnitude of effect was greater for child BLLs. Nozadi et al. (2021) collected blood samples from
pregnant mothers at the 36-week visit or at the time of delivery and administered the ASQ:I at 10-13
months of age to evaluate communication, gross motor, fine motor, problem-solving, and personal-social
development. A 1 -(.ig/dL increase in prenatal blood Pb was associated with a decrease in fine motor (|3 =
-0.63 [95% CI: -1.19, -0.08]) scores. Palaniappan et al. (2011) observed decrements ofWRAVMA
scores in association with l-(ig/dL increase in concurrent BLLs (Drawing: |3 = -0.29 [95% CI: -0.51,
-0.07]; Matching: |3 = -0.14 [95% CI: -0.31, 0.02]; Pegboard: |3 = -0.19 [95% CI: -0.38, 0.01];
Composite: |3 = -0.26 [95% CI: -0.45, -0.07]).

3.5.5.1.1 Summary

Evidence from prospective studies of Cincinnati and Yugoslavia birth cohorts indicated
associations of decrements in fine and gross motor function with higher neonatal, concurrent, and lifetime
average BLLs in young children with higher earlier childhood BLLs. Several recent birth cohort studies
observed lower scores on the Bayley PDI in association with maternal Pb exposure (no clear pattern by
trimester of pregnancy), cord BLL, and postnatal concurrent blood Pb (Rygiel et al.. 2021; Y Ortiz et al..
2017; Liu et al.. 2014c; Kim et al.. 2013c; Henn et al.. 2012). Pb-associated decrements in motor function
were observed in neonates (Liu et al.. 2014d; Paraiuli et al.. 2013) and in some but not all studies of
toddlers that assessed motor function using GDS (Liu et al.. 2022a; Zhou et al.. 2017) or children's
(greater than 7 years old) abilities to perform certain tasks indicative of gross motor function (i.e.,
balance) (Taylor et al.. 2015). although associations with fine motor function were observed (Taylor et al..
2018; Boucher et al.. 2016).

3.5.5.2 Toxicological Studies of Motor Function

As described above and in previous reviews (U.S. EPA. 2013. 2006). epidemiologic studies
provide evidence of associations between Pb exposures and fine and gross motor decrements, mainly in
children. Evaluating performance in neurobehavioral toxicological studies with Pb exposure in rodents

3-144

-------
can substantiate observed Pb exposure effects on motor function seen in humans. In past assessments,
evidence in animal toxicological studies has been limited due to a lack of investigations with relevant Pb
exposures. The purpose of this section is to update the collection of evidence available concerning Pb
exposure-induced effects on motor function in animal models. Studies that examined various indices of
locomotor activity are evaluated above in the Toxicological Studies of Hyperactivity section (Section
3.5.2.3.2).

Previous IS As (U.S. EPA, 2013, 2006) highlighted rotarod and air righting reflex experiments
with rodents to discuss the effects of Pb on development of motor coordination and balance. Typical
rotarod tests compare the latency to fall for subjects placed on a rotating rod. Falling off more quickly
indicates decreased coordination and/or balance. There are two rotarod studies discussed in previous U.S.
EPA reviews that describe effects of developmental Pb exposure on rotarod performance that resulted in
relevant BLLs less than 30 (ig/dL. Interestingly, Moreira et al. (2001) saw no effect of Pb exposure, from
the beginning of gestation through lactation, on Wistar rat rotarod performance at PND 70 with PND 23
mean BLLs of 21 (ig/dL. In contrast, Leasure et al. (2008) observed substandard performance in
pregestational through lactation Pb-exposed male, but not female, mice with peak BLLs of less than 10
(ig/dL. Since Leasure et al. (2008), no other PECOS-relevant studies have assessed rotarod performance
in rodents exposed to Pb throughout the entire developmental period. Two recent rotarod studies with
mice, by Flores-Montoya and Sobin (2015) and Zou et al. (2015), showed no decrements in performance
in rotarod tests after postnatal-only exposure to Pb in drinking water for PND 0-28 and 37-58 for
respective studies.

While the outcomes of these two latest rotarod studies were mostly negative, additional
investigations evaluating the effects of developmental Pb exposure on coordination and balance in
neonatal rats using surface righting reflex, negative geotaxis reflex, and ascending wire mesh tests yielded
mixed results. Surface righting reflex tests are run by placing pups in a supine position and then recording
the time it takes to flip onto their feet. Slower times to flip indicate postural imbalances. For negative
geotaxis reflex, or slant-board tests, pups are placed on a slanted board and the time it takes for the pup to
face upward is recorded. Slower times to turn upward indicate that the vestibular response to gravity cues
or motor coordination required for turning are underdeveloped. Success in ascending wire mesh tests also
requires coordination, as the animals are required to climb to the top of a mesh out of a water bath in a
predetermined period. In a study comparing the developmental effects on male Wistar rats with
pregestational, gestational, or lactational Pb exposure, pups exposed during gestation achieved negative
geotaxis significantly faster than unexposed counterparts when tested on PND 8, 10, and 12 (Rao Barkur
and Bairy, 2016). In contrast, in the same study, Rao Barkur and Bairy (2016) observed no difference in
negative geotaxis times between control, pregestation alone, and lactation alone Pb-exposed pups. No
effects on surface righting reflex on PND 3 through 5 were observed for pups belonging to the previously
mentioned exposure groups. The day of achievement in ascending wire mesh tests (PND 14-18) was
delayed for animals in both gestation and lactation Pb-exposed groups but not for those in the
pregestational group (Rao Barkur and Bairy, 2016). Betharia and Maher (2012) exposed pregnant

3-145

-------
Sprague Dawley rats to Pb (II) acetate trihydrate via drinking water from the beginning of gestation
through lactation and until weaning. Development of the surface righting reflex of control and exposed
offspring was tested from PND 1 to 10. Slower righting times were observed for Pb-exposed offspring on
PND 1; however, from PND 2 through 10, there were no differences between Pb-exposed and control
groups. Basha and Reddv (2015) observed a significant increase in righting time in righting reflex tests
done on PND 6 and 7 and an increase in latency to turn in negative geotaxis tests for male Wistar rats
tested on PND 8, 9, and 10 after in utero exposure to Pb. Tartaglione et al. (2020) saw no decrements in
righting reflex time or negative geotaxis achievement on PND 4, 7, 10, and 12 from Pb exposure in the
offspring of dams exposed to Pb from 1 month pre-mating to offspring weaning.

Additional motor function experiments with early postnatal weaning in Wistar rats were carried
out in studies with developmental Pb exposures. Tartaglione et al. (2020) recorded on PND 4, 7, 10, and
12 the duration of neonatal motor patterns of rat pups from dams exposed to Pb before mating until
offspring weaning. On PND 10, Pb-exposed pups spent less time in locomotion compared with controls,
in favor of head rising and wall climbing movements, demonstrating a stereotyped/preservative profile.
Basha and Reddv (2015) observed a prenatal Pb-induced strength deficit when rats were subject to
fore limb hang tests on PND 13, 14, 15, and 16 but not on day 12. This indicated Pb-induced
underdevelopment of fine motor ability. Rao Barkur and Bairv (2016) tested rats on PND 6, 8, 10, and 12
for Pb-induced effects on swimming development. They observed no difference in swimming body angle
or limb movements for ISA-relevant pregestation, gestation, or lactation-exposed groups compared with
control. These novel studies warrant further investigation into the effects of Pb exposure at different
concentrations and stages of development on neonatal movement patterns and forelimb hang tests.

3.5.5.2.1 Summary

The evidence supporting the link between developmental Pb exposure and deficits in motor
function in animal models has expanded on account of recent studies utilizing Pb-exposed rodents with
mean BLLs <30 (ig/dL. These new studies illustrate the effects of Pb exposure on both gross and fine
motor development in novel paradigms. In addition to the effect on rotarod performance (Leasure et al..
2008) described in the previous ISA, developmental Pb-induced decrements in righting reflex, negative
geotaxis reflex (Basha and Reddv. 2015). ascending wire mesh (Rao Barkur and Bairv. 2016). and
forelimb hang tests (Basha and Reddv. 2015) were observed. Interestingly, gestational Pb exposure was
present among each type of study that yielded decrements in these measurements of motor function;
therefore, it may be a more sensitive window compared with lactation or postnatal exposures. In terms of
design or methodology, studies that found weak or null relationships were not stronger and did not
weaken the overall body of corroborating data. Key aspects such as exposure levels and timing, ages of
animals at testing, and slant-board angles were variable between the few relevant studies. Altogether, the
results from these recent studies support the conclusions from the previous ISA. However, due to the
limited number of reproduced experiments, these recent studies do not enhance the consistency of the

3-146

-------
evidence. In addition to the fine motor, motor reflex, and coordination, and balance studies described in
this section, effects of developmental Pb exposure on locomotor activity are evaluated separately in the
Toxicological Studies of Hyperactivity section (Section 3.5.2.3.2) above. Briefly, due to heterogeneity in
study design, the evidence for effects of developmental Pb on locomotor activity is mixed; however, a set
of four independent studies with analogous conditions showed hyperactivity in rodents when tested within
a PND 14 to 23 window after lactational Pb exposure (Duan et al.. 2017; De Marco et al.. 2005; Moreira
et al.. 2001; Rodrigues et al.. 1996).

3.5.5.3 Relevant Issues for Interpreting the Evidence Base

3.5.5.3.1 Potentially At-Risk Populations

Sex

A limited number of toxicological studies have reported sex differences in Pb-related effects on
motor function. Among studies in the 2013 Pb ISA, sex-specific differences in mice were observed for
gross motor skills, with balance and coordination most affected among males at the lowest Pb exposures
(Leasure et al.. 2008).

Recent epidemiologic studies that evaluated sex as a potential modifier of the association between
Pb exposure and motor function add to the evidence (Liu et al.. 2022a; Y Ortiz et al.. 2017). Y Ortiz et al.
(2017) found that the observed association between maternal blood Pb during the third trimester and
lower PDI scores was not different between boys and girls. Liu et al. (2022a) found that the association of
maternal blood Pb exposure with gross motor development quotient on the GDS was modified by sex
(-3.43 [95% CI: -6.16,-0.69] in boys and-1.18 [95% CI: -2.81, 0.44] in girls per ln-transformed unit).

Maternal Self-esteem

Maternal self-esteem has been shown to modify associations between BLLs and health effects in
children. In one study, high maternal self-esteem appeared to attenuate the negative effects of the child's
increased BLLs on PDI scores (Surkan et al.. 2008). In this study, larger decreases in PDI scores were
associated with increased BLLs among children whose mothers were in the lower quartiles of self-esteem
(Surkan et al.. 2008). Maternal self-esteem was not evaluated as an effect modifier in recent studies of Pb
exposure and motor function among children.

3-147

-------
Maternal Stress

In a recent epidemiologic study, Y Ortiz et al. (2017) found that the observed association of
maternal blood Pb during the third trimester with lower PDI scores differed depending on maternal stress.
Contrary to expectations, higher PDI scores were observed with higher maternal stress.

3.5.5.3.2 Lifestages

Multiple lifestages during childhood are implicated in the effects of Pb exposure on motor
function in children. Analyses of children enrolled in the Cincinnati cohort at age 6 years indicated
associations of concurrent, lifetime average, and neonatal Pb exposure with poorer upper limb dexterity
and fine motor composite score. Studies conducted in the Cincinnati cohort found that prenatal or
neonatal BLLs were not consistently associated with motor function decrements at ages 4-10 years
(Bhattacharva et al.. 1995; Dietrich et al.. 1993). Several recent birth cohort studies support findings from
the 2013 Pb ISA with observations of lower scores on the Bayley PDI in association with maternal Pb
exposure (no clear pattern by trimester of pregnancy), cord BLL, and postnatal concurrent blood Pb
(Rvgiel et al.. 2021; Y Ortiz et al.. 2017; Liu et al.. 2014c; Kim et al.. 2013c; Henn et al.. 2012). Animal
toxicological studies mentioned above and in previous ISAs indicate the potential for delays in gross
motor development with gestational and/or early postnatal Pb exposure (Rao Barkur and Bairv. 2016;
Basha and Reddv. 2015; Leasure et al.. 2008) and for fine motor decrements with gestational Pb exposure
(Basha and Reddv. 2015). Apart from the study by Leasure et al. (2008). which tested balance in adults,
these studies measured and found diminished motor development in early postnatal rodents (Rao Barkur
and Bairv. 2016; Basha and Reddv. 2015).

3.5.5.4 Summary and Causality Determination: Motor Function in Children

The evidence assessed in the 2013 Pb ISA is sufficient to conclude that a "causal relationship is
likely to exist" between Pb exposure and decrements in motor function in children. Key evidence came
from prospective analyses of the CLS and Yugoslavia cohorts demonstrating associations of BLLs with
poorer motor function with consideration of potential confounders including SES, parental caregiving
quality and education, smoking birth outcomes, sex, and child health. Among children that participated in
the Cincinnati cohort, higher earlier childhood BLLs (age 0-5 year average [median: 11.7 |ig/dL| or age
78 month) were associated with poorer fine (i.e., grooved pegboard and finger tapping) (Ris et al.. 2004)
and gross motor function (i.e., postural balance) (Bhattacharva et al.. 2006) assessed in adolescence (ages
12, 15-17 years). In addition, assessments of children enrolled in the Cincinnati cohort at age 6 years
indicated associations of concurrent (mean: 10.1 (.ig/dL). lifetime average (mean: 12.3 (.ig/dL). and
neonatal (mean: 4.8 (ig/dL) but not prenatal maternal (mean: 8.4 (ig/dL) BLLs with poorer upper limb
dexterity, fine motor composite score (Dietrich et al.. 1993). and poorer postural balance (Bhattacharva et

3-148

-------
al.. 1995). Wasserman et al. (2000) also examined the association of Pb exposure with motor function. In
this prospective analysis of the Yugoslavian cohort, an association of lifetime average BLL (exact levels
not reported) with decrements in fine but not gross motor function at age 4.5 years was observed
(Wasserman et al.. 2000). Evidence from cross-sectional studies for associations between motor function
and concurrent BLL was mixed in populations with mean BLLs of 2-5 (ig/dL (Min et al.. 2007; Surkan et
al.. 2007; Despres et al.. 2005). Recent epidemiologic and toxicologic studies generally support findings
from the 2013 Pb ISA. The key evidence, as it relates to the causal framework, is summarized in Table
3-6.

Several recent birth cohort studies report lower scores on the Bayley PDI in association with
maternal Pb exposure (no clear pattern by trimester of pregnancy), cord BLL, and postnatal concurrent
blood Pb (Rygiel et al.. 2021; Y Ortiz et al.. 2017; Liu et al.. 2014c; Kim et al.. 2013c; Henn et al.. 2012).
Pb-associated decrements in motor function were also observed in neonates (Liu et al.. 2014d; Paraiuli et
al.. 2013). A limited number of studies of children greater than 7 years old were conducted. Taylor et al.
(2015) did not report associations with certain tasks indicative of gross motor function (i.e., balance),
although associations with decreased fine motor function were observed (Taylor et al.. 2018; Boucher et
al.. 2016).

Recent toxicological studies provide limited biological plausibility by showing effects on motor
function in rodent models from developmental Pb exposure resulting in BLLs <30 (ig/dL within one order
of magnitude of recent concentrations observed in humans. Epidemiologic evidence of developmental Pb-
induced impairment of balance and coordination is supported by observations of poorer rotarod
performance in male mice exposed to Pb during gestation (Leasure et al.. 2008). In addition, evidence
from epidemiologic studies indicating Pb-induced delayed gross motor development in children is
reinforced by toxicological studies that display slower times to achievement by postnatal rats
gestationally exposed to Pb in surface righting reflex (gestational Pb), negative geotaxis reflex
(gestational Pb) (Basha and Reddv. 2015). and ascending wire mesh tests (gestational Pb; lactational Pb)
(Rao Barkur and Bairv. 2016). Epidemiologic studies revealing Pb-induced decrements in children's fine
motor skills are supported by the observed grip strength deficits for gestational Pb-exposed early postnatal
rats in forelimb hang tests (Basha and Reddv. 2015). Additional studies on Pb-induced changes on several
neurochemical endpoints that factor into impaired motor function have been reported and are described in
Section 3.3

Overall, the evidence is sufficient to conclude that there is likely to be a causal relationship
between Pb exposure and motor function in children. This determination is based on consistent
evidence from prospective epidemiologic studies, which demonstrate an association between higher
childhood BLLs (neonatal, earlier childhood, concurrent and lifetime average) and poorer fine and gross
motor function in children (aged 4.5-17) with adjustment for maternal IQ, parental education, SES, and
HOME score. Additional prospective studies have also demonstrated consistent evidence in infants and
toddlers using the Bayley PDI, but evidence supporting neonatal effects is more limited. Epidemiologic

3-149

-------
evidence is supported by limited experimental animal studies that demonstrate impairments in balance,
coordination, and grip strength, as well as delayed reflex development. There is some remaining
uncertainty arising from cross-sectional studies using concurrent BLLs that have reported mixed results.

3-150

-------
Table 3-6 Summary of evidence indicating a likely to be causal relationship between Pb exposure and motor
function in children

Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0

Consistent findings from a few
prospective epidemiologic studies
with relevant BLLs

Evidence from prospective studies for fine and gross
motor function decrements in children ages 4.5-17
yr in Cincinnati, Yugoslavia in association with
neonatal, earlier childhood, concurrent, lifetime avg
BLLs.

High follow-up participation, no selective attrition in
Cincinnati cohort, higher loss-to-follow-up in
Yugoslavia cohort with lower maternal IQ, HOME.

Both studies adjusted for maternal IQ, parental
education, SES, HOME score

Studies used various, widely used tests to assess
outcomes.

Ris et al. (2004)
Dietrich et al. (1993)
Bhattacharya et al.
(1995)

Wasserman et al.
(2000)

Section 4.3.7, (U.S.
EPA, 2013)

Blood Pb means

Cincinnati: neonatal (10 day) 4.8 |jg/dL,
concurrent (age 6 yr) 11.6 |jg/dL, lifetime
(to age 15-17 yr) avg 12.3 |jg/dL, age 0-5
yravg 11.7 |jg/dL

Former Yugoslavia: NR

Mixed evidence for lower (concurrent) BLLs from Section 4.3.7,
cross-sectional studies that considered several EPA, 2013)
potential confounding factors.

(U.S.

Consistent findings from prospective
studies of infants and toddlers

Lower scores on the Bayley PDI in association with
maternal Pb exposure (no clear pattern by trimester
of pregnancy), cord BLL and postnatal concurrent
blood Pb

Kim et al. (2013c)
Y Ortiz etal. (2017)
Liu et al. (2014c)
Rvaiel et al. (2021)
Henn etal. (2012)

Limited evidence in neonates

Pb-associated decrements in motor function
reflexes) observed

e.g., Paraiuli et al. (2013)
Liu etal. (2014d)

Limited experimental animal
evidence at relevant exposures

Deficient gross motor coordination and balance in
rodents with developmental Pb exposure (less time
on rotarod, slower righting and negative geotaxis
reflexes, delayed day of achievement for ascending
wire mesh test)

Leasure et al. (2008)
Basha and Reddv

(2015)

Rao Barkur and Bairv

(2016)

Blood Pb: -10 |jg/dL in mice after
pregestational through lactation exposure,
5-11 |jg/dL in rats after gestational
exposure, 27 |jg/dL in rats after lactational
exposure

3-151

-------
Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0



Fine motor (grip strength) deficits in early postnatal
rats with gestational Pb exposure

Basha and Reddv
(2015)

Blood Pb: 11.2 |jg/dL after gestational
exposure

Limited experimental animal
evidence at relevant exposures
provide coherence for epidemiologic
observations of effect modification
by sex

Poorer balance (fell off rotarod more quickly) in adult
male but not female mice with pregestational
through lactation dietary Pb exposure

Leasure et al. (2008)

Blood Pb: -10 |jg/dL in mice after
pregestational through lactation exposure

Biological plausibility demonstrated

Pathways involving oxidative stress, inflammation
and Ca2+ signaling result in impaired neuron
development, synaptic changes, and
neurotransmitter changes.

Recent studies support and extend findings related
to overt nervous system effects

U.S. EPA (2013)
Section 3.3

Section 3.4.2



avg = average; BLL = blood lead level; Ca2+ = calcium ion; HOME = Health Outcomes and Measures of the Environment; IQ = intelligence quotient; NR = not reported; Pb = lead;
PDI = Psychomotor Developmental Index; SES = socioeconomic status; yr = year(s).

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 20151.
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or
inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-152

-------
3.5.6

Sensory Organ Function in Children

The 2013 Pb ISA included separate causality conclusions for auditory and visual function. This
ISA combines these categories and makes one causality determination for Sensory Organ Function
because there are relatively few studies within this outcome grouping.

3.5.6.1 Auditory Function in Children

The evidence assessed in the 2013 Pb ISA was sufficient to conclude that "a causal relationship is
likely to exist" between Pb exposure and decrements in auditory function in children (U.S. EPA, 2013).
Evidence from a prospective study (Dietrich et al.. 1992) and small number of cross-sectional studies of
U.S. children, including NHANES and Hispanic Health and Nutrition Examination Survey (HHANES)
analyses (Schwartz and Otto, 1991, 1987) indicated associations of higher BLLs with increases in hearing
thresholds as well as decreases in auditory processing or auditory evoked potentials, with adjustment for
potential confounding by SES in most studies and by child health and nutritional factors in some studies.
The high participation rates in a prospective birth cohort study (Dietrich et al., 1992) reduced the
likelihood of biased participation by children with higher BLLs. Across studies, associations were found
with BLLs measured at various time periods, including prenatal maternal, neonatal (10 days, mean 4.8
(ig/dL), lifetime average (to age 5 years), and concurrent (ages 4-19 years) BLLs (median 8 (.ig/dL).
Evidence for Pb-associated increases in hearing thresholds or latencies of auditory evoked potentials was
also found in adult monkeys with lifetime dietary Pb exposure. However, these effects in adult animals
were demonstrated at higher peak or concurrent BLLs (i.e., 33-150 (ig/dL) than those relevant to this
ISA; thus, the biological plausibility for epidemiologic observations was unclear.

In the current ISA, several recent cross-sectional studies support the conclusion in the 2013 Pb
ISA regarding the association of Pb exposure with hearing loss; however, results were inconsistent for
other audiometric parameters. Recent toxicological studies provide additional evidence for hearing loss
and auditory processing deficits in rodents at relevant BLLs. Measures of central tendency for Pb
biomarker levels used in each study, along with other study-specific details, including study population
characteristics and select effect estimates, are highlighted in Table 3-12E (Epidemiologic Studies) and
Table 3-16T (Toxicological Studies). An overview of the recent evidence is provided below.

3.5.6.1.1 Epidemiologic Studies of Auditory Function

Several recent epidemiologic studies examined the association between Pb exposure and
decrements in auditory function in children. The findings generally support a positive association between
Pb exposure and hearing loss. For other audiometric parameters, however, the results were inconsistent.

3-153

-------
Most studies of auditory function were cross-sectional. In a meta-analysis of studies from Iran, Korea,
China, and the United States, Yin et al. (2021) observed a positive association between Pb exposure and
hearing loss indicated by pure-tone average (PTA) >25 dB in children and adolescents (3-19 years)
(combined OR per unit increase in Pb = 1.53 [95% CI: 1.24, 1.87]). The pooled OR was based on only
two studies (Xu et al.. 2020; Choi and Park. 2017). Xu et al. (2020) conducted a case-control analysis of
preschool-aged children (3 to 7 years of age) who resided in an area contaminated with Pb and Cd and in
an uncontaminated reference area. This study found associations of exposures with hearing loss,
potentially affected by epigenetic changes. The OR for Pb-associated hearing loss in both ears was 1.40
(95% CI: 1.06, 1.84 per unit change log-transformed BLL) after adjustment for characteristics of the
child, parental education, SES, and noise exposure. Choi and Park (2017) measured speech- and high-
frequency hearing loss in adolescents (12-19 years) and adults (20-87 years) in the Korea National
Health and Nutrition Examination Survey (KNHANES). Hearing loss was defined as PTA >15 dB in
adolescents. For each doubling of blood Pb, there was a positive association with speech-frequency
hearing loss (OR =1.2 [95% CI: 0.48, 3.05]) and high-frequency hearing loss (>25 dB) (OR = 1.26 [95%
CI: 0.73, 2.16]) among adolescents.

In addition to the aforementioned studies, among adolescent NHANES participants (ages 12 to
19 years), Shargorodskv et al. (2011) found a positive association between blood Pb and hearing loss.
Hearing loss was defined as low or high-frequency PTA >15 dB in either ear. Compared with study
participants with low BLLs (<1 (ig/dL), those with the highest level (>2 (ig/dL) were more likely to have
any hearing loss (OR = 1.95 [95% CI: 1.24-3.07]), particularly high-frequency hearing loss (OR = 2.22
[95% CI: 1.39-3.56]). The direction of effect for low-frequency hearing loss was the same but at a
smaller magnitude (OR= 1.13 [95% CI: 0.61-2.07]). Among younger children (3-7 years with median
BLL <5 (ig/dL) in China, a positive association between blood Pb and hearing loss was also observed
(Liu et al.. 2018c). For each (ig/dL increase in blood Pb, the odds of any hearing loss increased by 1.24
times (OR= 1.24 [95% CI: 1.03, 1.49]). This association was not as evident for high-frequency hearing
loss (OR= 1.08 [95% CI: 0.84, 1.38]) and low-frequency hearing loss (OR= 1.02 [95% CI: 0.87, 1.19]).

Auditory function in children was also measured according to the auditory brainstem response
(ABR) (Silver et al.. 2016; Alvarenga et al.. 2015; Pawlas et al.. 2015). In an unadjusted descriptive
analysis of children (4-13 years) in Poland, BLLs were positively correlated with brainstem auditory
evoked potentials (BAEP) and pure-tone audiometry and negatively correlated with acoustic otoemission
(Pawlas et al.. 2015). In multivariable analyses stratified by polymorphisms in the ALAD and vitamin D
receptor (VDR) genes, the associations for BAEP per (ig/dL increase in blood Pb were generally null
(Pawlas et al.. 2015). Silver et al. (2016) measured ABR in newborns (average 2 days old) in China.
Compared with a low (<2 (ig/dL) BLL measured during late pregnancy, infants exposed to medium (2-
3.8 (ig/dL) and high (>3.8 (ig/dL) Pb levels were more likely to have a higher ABR central-to-peripheral
(C-P ratio) (Silver et al.. 2016). When using Pb levels measured in cord blood and during mid-pregnancy,
however, the association for ABR C-P ratio moved toward the null (Silver et al.. 2016). Although

3-154

-------
quantitative results were not provided for a study of children (18 months-14 years) in Brazil, Alvarenga
et al. (2015) observed no association between cumulative BLLs and BAEP.

Summary

A prospective study in the 2013 Pb ISA (Dietrich et al., 1992) found an association of Pb
exposure with decreased auditory processing. In addition, cross-sectional studies found increased hearing
thresholds in children aged 4-19 years that participated in NHANES and HHANES in association with
higher concurrent BLLs. Recent cross-sectional and case-control studies of young children and
adolescents generally support a positive association between Pb exposure and hearing loss (Xu et al„
2020; Choi and Park, 2017; Shargorodskv et al., 2011), whereas the results were inconsistent for ABR
(Silver et al„ 2016; Alvarenga et al„ 2015; Pawlas et al„ 2015).

3.5.6.1.2 Toxicological Studies of Auditory Function

Toxicological evidence for effects on auditory function in the 2013 Pb ISA was limited to one
study (U.S. EPA, 2013). This study evaluated auditory thresholds using a behavioral task in 13-year-old
monkeys (Macaca mulatto) who had previously been exposed to Pb either gestationally or postnatally
(Laughlin et al., 2009). Potentially due to limitations noted within the study, small but nonsignificant
increases in the auditory threshold were reported in Pb-exposed animals compared with controls. Stronger
associations between Pb exposure, auditory threshold shifts, and latency in BAEP were reviewed in the
2006 Pb AQCD (U.S. EPA, 2006). Importantly, the associations demonstrated in the 2013 Pb ISA and
2006 Pb AQCD occurred at higher BLLs (>30 (ig/dL) that would not be considered PECOS-relevant for
this ISA.

Changes in auditory thresholds using BAEP have been further assessed in three recent rodent
studies (Table 3-16T). Jamesdaniel et al. (2018) exposed male C57B1/6 mice from PND 33 to PND 61 to
Pb and subsequently detected 8-12-dB upward shifts in hearing thresholds (indicative of hearing loss)
between 4 and 32 kHz. In contrast, another recent study using similarly aged male CBA/CaJ mice and a
longer exposure paradigm (11 weeks) found no significant effect of Pb on hearing thresholds at 8, 16, and
32 kHz (Carlson et al., 2018). The final study, which exposed male and female Sprague Dawley rats
postnatally to Pb did not detect significant differences in hearing thresholds between 4 and 28 kHz at
PND 60 (Zhu et al., 2016). Animals in the two studies that did not detect an effect had lower BLLs than
those in the study that did (3-8 (ig/dL versus 29 (ig/dL). However, due to the small number of studies, the
existence of an exposure threshold for this effect remains uncertain.

Recent studies have also investigated the effect of Pb exposures on auditory processing, which
was not discussed in previous ISAs. Zhu et al. (2016) exposed rat pups to Pb through their dams" drinking
water until weaning, when they began drinking Pb-free water. BLLs of the pups were roughly 8 (ig/dL

3-155

-------
during exposure and had returned to baseline levels by PND 40. At PND 60, the Pb-exposed rats were
found to have a decreased ability to discriminate between target and nontarget sound bursts. Additionally,
these rats were found to have a reduced spike rate-following ability and decreased cortical response
synchronization, indicative of a deficit in auditory cortical temporal processing. The same research group
published a follow-up study using a similar exposure paradigm to investigate another aspect of auditory
processing (i.e., sound localization) (Liu et al.. 2019). In a sound-azimuth discrimination task, Pb-exposed
animals took significantly longer to reach target accuracy and had significantly greater deviations (i.e.,
difference between the location of the desired response versus the location of incorrect response)
compared with control animals. These behavioral impairments were accompanied by a degraded sound-
azimuth selectivity in the primary auditory cortex neurons.

Summary

Earlier experimental animal studies have found decreased auditory function in adult monkeys and
rodents after lifetime exposure to Pb in animals with peak BLLs greater than 30 (ig/dL, but the persistence
of these effects at lower BLLs and in juvenile animals was uncertain. Three recent studies evaluated
auditory thresholds using BAEP in rodents exposed to Pb starting in the postnatal or juvenile period.
Jamesdaniel et al. (2018) found 8-12-dB upward shifts in hearing thresholds between 4 and 32 kHz in
young adult mice (peak BLLs of 29 (ig/dL). Studies evaluating lower mean BLLs from 3 to 8 (ig/dL did
not report differences in BAEP thresholds. However, mice with mean peak BLLs of 8 (ig/dL had
significant deficits in auditory processing, including decreased sound discrimination and sound
localization ability paired with dysfunction in the auditory cortical neurons (Liu et al.. 2019; Zhu et al..
2016).

3.5.6.2 Visual Function

The evidence reviewed in the 2013 Pb ISA was inadequate to determine whether a causal
relationship exists between Pb exposure and visual function in children (U.S. EPA. 2013). A study in
children and a few studies in animals showed Pb-associated increases in supernormal electroretinograms;
however, the biological plausibility of the observations was unclear. Overall, the available epidemiologic
and toxicological evidence was of insufficient quantity, quality, and consistency to support a causality
conclusion.

3.5.6.2.1 Epidemiologic Studies of Visual Function

Only a few epidemiologic studies examined the association between Pb exposure and decrements
in visual function in children (Silver et al., 2016; Fillion et al.. 2013). Since the measures of visual
function differed between studies, it is difficult to draw any conclusions about Pb exposure and visual

3-156

-------
function in children. Silver et al. (2016) measured grating visual acuity (VA) in 6-week-old infants in
China. Compared with low (<2 (ig/dL) BLLs measured during late pregnancy, infants exposed to medium
(2-3.8 (ig/dL) and high (>3.8 (ig/dL) Pb levels were more likely to have lower grating VA (Silver et al..
2016). When using Pb levels measured in cord blood and during mid-pregnancy, however, the association
for grating VA was attenuated and moved closer toward the null (Silver et al., 2016). In Brazil, Fillion et
al. (2013) measured contrast sensitivity (cycles per degree [cpd]) and acquired color vision loss (color
confusion index, CCI) in study volunteers that included adolescents (age range: 15-66 years). Based on
the entire study population, blood Pb exposure was negatively associated with the intermediate spatial
frequency of contrast sensitivity (12 cycles/degree); however, results varied by spatial frequency (Fillion
et al., 2013). For CCI, there was a small positive association with blood Pb (Fillion et al., 2013).

Summary

Overall, the available epidemiologic and toxicological evidence assessed in the 2013 Pb ISA was
of insufficient quantity, quality, and consistency to support a causality conclusion. A limited number of
recent epidemiologic studies are available for consideration; however, measures of visual function
differed between studies limiting observations regarding the consistency of the evidence overall.

3.5.6.2.2 Toxicological Studies of Visual Function

The evidence base pertaining to effects on visual function in the 2013 Pb ISA was largely
supported by seminal literature reviewed previously in the 1986 and 2006 Pb AQCDs showing reduced
VA, retinal alterations, and changes in CNS visual processing areas and subcortical neurons involved in
vision (U.S. EPA, 2013, 2006, 1986). Electroretinography (ERG), which measures the bioelectrical
response of the retina to a light stimulus, is used to detect abnormalities in retinal functioning. Fox et al.
(2008) found that Pb exposure in female Long-Evans rats (gestation through PND 10, measured at PND
90) induced supernormal ERGs (i.e., increases in the response amplitude) at low and moderate exposure
levels (BLLs of 12 and 24 (ig/dL) and subnormal ERGs (i.e., decreases in the response amplitude) in the
high exposure group (BLL of 46 (ig/dL). Earlier studies have also found Pb-related aberrations in ERGs,
but the direction of this effect is inconsistent (i.e., both subnormal and supernormal responses have been
detected) (Fox et al.. 1997; Lilienthal et al.. 1988). As discussed in Giddabasappa et al. (2011). the effect
direction may be related to both the lifestage during exposure (gestational versus postnatal) and the Pb
dose. This study also demonstrated that low to moderate gestational Pb exposure (BLLs: 10 and 27
(ig/dL) increased and prolonged retinal progenitor cell proliferation, resulting in selectively increased rod
photoreceptor and bipolar cell neurogenesis in C57BL/6 mice at PND 60 (Giddabasappa et al.. 2011).
Adult monkeys {Macaca fascicidaris) with lifetime Pb exposure, producing BLLs from 50-115 (ig/dL,
had temporal vision dysfunction but no change in spatial function (Rice. 1998). In contrast to these
effects, Laughlin et al. (2008) found that Pb exposure in Rhesus monkeys (exposed from PND 8-26
weeks; BLLs of 35-40 (ig/dL) did not significantly affect the development of photopic spatial acuity

3-157

-------
assessed using a modified Teller preferential looking paradigm. Recent PECOS-relevant studies have not
further examined the effects of Pb on visual function.

3.5.6.3	Relevant Issues for Interpreting the Evidence Base

3.5.6.3.1 Potentially At-Risk Populations

Genes

Pawlas et al. (2015) conducted multivariable analyses stratified by polymorphisms in the ALAD
and VDR genes and found that the associations for BAEP and pure-tone audiometry per |ig/L increase in
blood Pb were generally null (Pawlas et al.. 2015).

3.5.6.4	Summary and Causality Determination: Sensory Organ Function

The 2013 Pb ISA presented two causality determinations related to sensory function in children:
auditory function and visual function (U.S. EPA, 2013). At the time, the evidence was sufficient to
conclude that a causal relationship was likely to exist between Pb exposure and auditory function
decrements in children. For visual function, the evidence was inadequate to determine if a causal
relationship exists. In 2015, the Preamble to the ISA introduced minor changes to the language used in the
causality framework descriptors (U.S. EPA, 2015). This change has affected the causality determination
for this section. Importantly, the new determination is not intended to be interpreted as a weakening of the
evidence base, as recent evidence has remained consistent with previously reviewed studies.

Auditory processing decrements were previously demonstrated in a prospective study by Dietrich
et al. (1992). In 5-year-old children, elevated BLLs during infancy (mean BLLs of 4.8 (ig/dL at 10 days
old) were associated with poorer performance on a test for auditory processing disorders after adjusting
for confounding factors including SES, HOME score, a variety of birth outcomes, maternal alcohol
consumption, maternal smoking, and overall child health. Recently, experimental animal studies
demonstrated that postnatal Pb exposure resulting in mean peak BLLs of 8 (ig/dL also caused significant
deficits in auditory processing, including decreased sound discrimination and sound localization ability
paired with dysfunction in the auditory cortical neurons (Liu et al„ 2019; Zhu et al„ 2016).

Multiple large cross-sectional NHANES and HHANES studies have shown that higher BLLs
(children aged 4-19; BLLs 8 (ig/dL) are associated with increased hearing thresholds (Schwartz and Otto,
1991, 1987). These associations remained after adjustment for age, sex, race, family income, parental
education, and nutritional factors. Recent cross-sectional and case-control studies continued to
demonstrate associations with BLLs and hearing loss in young children (aged 3-7, BLLs ~3 to 6 (ig/dL)

3-158

-------
and adolescents (aged 12-19, BLLs ~1 to 8 (.ig/dL). particularly at higher frequencies (Xu et al.. 2020; Liu
et al., 2018c; Choi and Park, 2017; Shargorodskv et al., 2011). Furthermore, hearing threshold increases
were previously demonstrated in adult nonhuman primates after developmental or lifetime Pb exposure,
although BLLs in these studies were greater than 30 (ig/dL (Laugh 1 in et al., 2009; Rice, 1997). Recent
experimental animal studies have not further evaluated hearing thresholds in nonhuman primates and
instead have focused on BAEPs in rodents. Jamesdaniel et al. (2018) found 8-12-dB upward shifts in
auditory thresholds between 4 and 32 kHz in young adult mice exposed during adolescence (peak BLLs
29 (ig/dL). Similar studies did not detect differences in BAEPs in rodents with lower peak BLLs (3 to 8
(ig/dL). Likewise, a few recent epidemiologic studies also evaluated BAEP with inconsistent results.

Although Pb-induced alterations in subcortical visual neurons, visual processing areas, and retinal
development have been demonstrated, supporting the biological plausibility of Pb-associated effects on
vision (U.S. EPA, 2013), evidence relating to visual function in epidemiological and toxicological studies
remains limited and inconsistent. Silver et al. (2016) found that decreased visual acuity in infants was
associated with maternal BLLs higher than 2 (ig/dL in late pregnancy, but this association was weaker
with BLLs in both mid-pregnancy and cord blood. Studies in nonhuman primates failed to detect changes
in visual acuity at BLLs above 35 (ig/dL, although one reported decrements in temporal acuity as a result
of Pb exposure (Laugh 1 in et al., 2008; Rice, 1998). Another recent study found associations with blood Pb
and decrements in contrast sensitivity and color vision, an endpoint that has not been previously studied,
in a study population that included adolescents (15-66 years old) (Fillion et al., 2013). Studies in both
humans and animals have found significant but inconsistent changes in ERGs (Fox et al., 2008;
Rothcnbcrg et al., 2002; Fox et al., 1997), though it is unclear if these findings translate to functional
visual changes.

In conclusion, the evidence is suggestive of, but not sufficient to infer, a causal relationship
between Pb exposure and sensory function in children. This determination is based primarily on the
strongest line of evidence within the sensory function grouping {i.e., auditory function). No recent
epidemiologic studies have further investigated the auditory processing decrements shown in Dietrich et
al. (1992), but recent experimental animal studies have demonstrated Pb-induced effects on auditory
processing. Cross-sectional and case-control studies focusing on the impact of Pb exposure on hearing
loss generally support an association but are not entirely consistent. Experimental animal studies
evaluating hearing loss at human relevant BLLs in young animals are not available. Limited
epidemiologic studies have evaluated Pb exposure and visual function in children with inconsistent
findings, but evidence for biological plausibility has been demonstrated.

3-159

-------
Table 3-7 Evidence that is suggestive of, but not sufficient to infer, a causal relationship between Pb
exposure and sensory organ function in children

Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0

Auditory Function

Consistent findings from a few Prospective study found associations of Dietrich et al. (1992)
epidemiologic studies with prenatal (maternal), neonatal, yearly age
relevant BLLs	1 to 5 yr, lifetime avg BLLs with poorer

auditory processing in children at age 5 yr
in Cincinnati.

Blood Pb means: neonatal (10 d) 4.8
|jg/dL, yearly age 1 to 5 yr 10.6-17.2
|jg/dL, lifetime (to age 5 yr) avg NR

Cross-sectional and case-control studies
for increased hearing thresholds in
children ages 3-19 yr, including analyses
of NHANES, HHANES and KNHANES in
association with higher concurrent BLLs.

Section 4.3.6.1, (U.S. EPA, 2013)

Xu et al. (2020)
Liu et al. (2018c)

Blood Pb median:

HHANES: 8 pg/dL; NHANES: NR

Means 3.63-5.69 pg/dL (3-7 yr)

Sharqorodsky et al. (2011)

NHANES (2005-2008): med ~1 pg/dL
(12-19 yr)

Choi and Park (2017)

KNHANES: GM: 1.26 pg/dL (15.6 yr)

Epidemiologic evidence helps
to rule out chance, bias and
confounding with reasonable
confidence

Prospective study adjusted for SES,
HOME score, birth outcomes, obstetrical
complications, maternal smoking. Several
other factors considered.

Dietrich et al. (1992)

Cross-sectional and case-control studies
considered potential confounding by age,
sex, race, income, parental education,
nutritional factors.

Xu et al. (2020)

Liu et al. (2018c)
Sharqorodsky et al. (2011)
Choi and Park (2017)

3-160

-------
Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0

Uncertainty due to lack of
animal evidence in juveniles
and limited evidence at
relevant exposure levels

Hearing loss in adult monkeys and
decreased BAEP in young adult rodents
at higher exposure levels.

Rice (1997)

Lauahlin et al. (2009)
Jamesdaniel et al. (2018)

Peak BLLs >29 |jg/dL



Decrements in sound discrimination and
localization in young adult rodents.

Zhu et al. (2016)
Liu et al. (2019)

Peak BLLs 8.2 pg/dL

Visual Function

Limited evidence from
epidemiologic studies

Associations with some tests of grating
VA and contrast sensitivity observed.

Silver et al. (2016)
Fillion et al. (2013)



Uncertainty due to limited
animal evidence in juveniles
and at relevant exposures

Higher than relevant postnatal Pb
exposure did not cause changes in VA in
infant nonhuman primates in infants but
did decrease temporal acuity in adults.

Lauahlin et al. (2008)
Rice (1998)

BLLs >35 pg/dL

Biological plausibility
demonstrated

Pb-induced alterations in ERGs,
subcortical visual neurons, visual
processing areas, and retinal
development demonstrated.

(U.S. EPA, 2013)



avg = average; BAEP = brainstem auditory evoked potentials; BLL = blood lead level; d = day; ERG = electroretinography; GM = geometric mean; HHANES = Hispanic Health and
Nutrition Examination Survey; HOME = Health Outcomes and Measures of the Environment; KNHANES = Korea National Health and Nutrition Examination Survey; NHANES =
National Health and Nutrition Examination Survey; NR = not reported; Pb = lead; SES = socioeconomic status; VA = visual acuity; yr = year(s).

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 20151. Note that

the change from "likely to be causal" for auditory effects in children in the 2013 Lead ISA, to "suggestive of, but not sufficient to infer, a causal relationship" for sensory organ function

in children reflects minor changes to the causal framework, rather than a weakening of the evidence base pertaining to auditory effects in children.

bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or

inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-161

-------
3.5.7

Social Cognition and Behavior in Children

In addition to neurodevelopmental disorders covered in previous sections—including ADHD,
intellectual and developmental disabilities, and motor disorders—there is an emerging body of research
on autism spectrum disorder (ASD) and other conditions related to social cognition and behavior. The
2013 Pb ISA (U.S. EPA, 2013) did not evaluate any epidemiologic studies examining associations
between Pb exposure and autism. ASD is generally characterized by restricted interests and behaviors,
including stereotyped patterns of behavior and sensory sensitivities. To meet the DSM criteria for ASD, a
child must have persistent deficits in social communication and demonstrate repetitive behaviors (APA,
2013). Social cognition, which is often impaired among individuals with ASD, involves the ability to
interpret and respond to social cues, communication, and interaction. These traits (or behaviors) can be
measured on a continuum in the general population with scores exhibiting a fairly normal distribution,
with scores at the extreme impaired end indicating a higher risk for ASD (Constantino, 2011). Deficits in
social cognition have been associated with lifelong educational, vocational, adaptive functioning, and
mental health challenges among individuals with and without a clinically diagnosed disorder. Autism
diagnosis (e.g., via the ICD code), the CBCL, Social Responsiveness Scale (SRS), BASC-2, BSID-II and
III, CDIIT, ASQ:I, GDS, Social Maturity Scale (SMS), MDAT, and ECDI have been used in studies
examining the association of Pb exposure with social cognition and behavior.

3.5.7.1 Epidemiologic Studies of Social Cognition and Behavior

There have been a number of recent studies of ASD and deficits in social cognition and related
behaviors. Many of these recent studies did not control for potential confounders and/or did not include
robust statistical methods to estimate C-R relationships between Pb exposure and outcome, and are not
considered further in this section (Filon et al.. 2020; Qin et al.. 2018; Skalnv et al.. 2017; Macedoni-
Luksic et al.. 2015; Alabdali et al.. 2014; Yassa. 2014; De Palma et al.. 2012; Blaurock-Busch et al..
2011; Tian et al.. 2011). Instead, the ensuing discussion focuses on a number of autism and social
cognition studies that include more comprehensive control for potential confounders. The relevant studies
provide some evidence of a positive association between Pb exposure and ASD, along with generally
consistent supporting evidence of an association with decrements in social cognition. Measures of central
tendency for BLLs used in each study, along with other study-specific details, including study population
characteristics and select effect estimates, are highlighted in Table 3-13E of Section 3.7. An overview of
the recent evidence is provided below.

Two recent studies used robust modeling approaches to assess the C-R relationship between
exposure to Pb and ASD (Arora et al.. 2017; Kim et al.. 2016). While each study examined different
biomarkers of exposure and exposure windows, both indicated associations between Pb exposure and
ASD. Arora et al. (2017) conducted a difference-in-differences analysis of a small case-control study of

3-162

-------
8- to 12-year-old twins with discordant or concordant ASD status. Pb was measured in shed deciduous
teeth using a method that provided temporal estimates of tooth Pb levels ranging from 20 weeks before
birth to 30 weeks after birth. To estimate the relationship between tooth Pb and ASD across this exposure
window, the authors used distributed lag models to estimate the smoothed mean differences in tooth Pb
levels in discordant pairs minus the mean differences in concordant twins at each time point. In this case,
concordant twins served as the control group to account for natural variations in Pb exposure within a
dyad. One analysis used concordant twins without ASD and the other used concordant twins with ASD as
the control groups. In both cases, the difference in tooth Pb levels between discordant twins was greater
than the difference in concordant twins across the entire exposure window, though there appeared to be
bimodal peaks in tooth Pb differences from about 10 to 15 weeks before birth and 10 to 20 weeks after
birth (see Figure 3-14).

Time since birth (weeks) Time since birth (weeks)

ASD = autism spectrum disorder.

Black line represents the difference in mean differences in tooth Pb levels between discordant ASD twins and: A) control twins; or B)
concordant ASD twins. Gray bands are unadjusted 95% CIs, while blue bands are adjusted for intra-twin correlations. Values above
zero represent increased levels in ASD cases compared with the non-ASD sibling after taking into account average difference in
control twins.

Source: Arora et al. (20171.

Figure 3-14 Differences in mean difference tooth Pb levels for autism

spectrum disorder in discordant twin pairs versus (A) non-autism
spectrum disorder twin pairs or (B) autism spectrum disorder
concordant twin pairs.

In a large cohort study of children in South Korea, Kim et al. (2016) analyzed blood Pb in relation
to autistic behaviors measured by parental response to the Autism Spectrum Screening Questionnaire
(ASSQ) and the SRS at ages 11-12 years old. BLLs at study enrollment (7-8 years old) were associated
with higher scores on the ASSQ (number of autistic behaviors) and SRS (severity across domains of
social awareness, cognition, communication, motivation, and mannerisms). There were null associations

3-163

-------
with blood Pb measured at 9-10 years and attenuated, but still positive, associations with concurrent
BLLs (11-12 years old). Nonparametric generalized additive models indicated an approximately linear
relationship between BLLs at enrollment and scores on the SRS. In addition to continuous models, the
authors also dichotomized ASSQ scores and reported 45% higher odds (95% CI: 10%, 93%) of a positive
screen for autism (ASSQ score >17) per 1 (ig/dL higher BLL at enrollment. Notably, symptoms of ASD
manifest as early as infancy and although BLLs in this study were measured prior to assessment of
autistic behaviors, the relevant exposure window likely preceded exposure measurement.

In contrast to the results from Kim et al. (2016) and Arora et al. (2017). recent case-control
studies did not observe an association between adjusted mean childhood BLLs and ASD cases (Rahbar et
al.. 2021; Rahbar et al.. 2015). In addition to matching cases and controls on age and sex, the authors
estimated mean differences using a linear model controlling for a variety of demographic and SES factors,
including maternal age. Similarly, a recent case-control study reported a null association between tertiles
of maternal BLLs and ASD in children, though there was some evidence of a nonlinear association in a
cubic spline model (Skogheim et al.. 2021). The study populations for these analyses included children
and mothers with lower (<2 (ig/dL; (Skogheim et al.. 2021; Rahbar et al.. 2015)) and higher (>7 (ig/dL;
(Rahbar et al.. 2021)) mean or median BLLs.

One additional large retrospective study in Northeast China (Dong et al.. 2022) compared current
BLLs in children with moderate/severe versus mild autism, as determined by CARS scores. Mean BLLs
for the mild and moderate/severe groups were 2.58 (SD: 1.08) (ig/dL and 3.25 (SD: 1.89) (ig/dL,
respectively. After adjusting for age, residence, parental caregiving, parental education, and
gastrointestinal conditions, autism severity was positively associated with BLL (|3 = 0.03 [95% CI: 0.01,
0.05]).

Additional supporting evidence was provided by several cohort studies that investigated
associations between Pb exposure and social cognition in children without autism. Most of these studies
reported inverse relationships between prenatal Pb exposure and social cognition measures. Several
studies additionally investigated effect modification by various other factors.

Rvgiel et al. (2021) assessed the relationship between maternal blood Pb and infant behavioral
development at 12 to 24 months of age in a small analysis of three birth cohorts from the ELEMENT
study in Mexico City. The authors used the behavioral rating scale (BRS) of the BSID-II to examine
attention, social engagement, orientation, motivation, and emotional response, giving rise to two social
cognition outcomes: orientation/engagement (ORIEN) and emotional regulation (EMOCI). The authors
reported that children had lower 24-month EMOCI and ORIEN percentile ranks with higher maternal
BLLs. Associations were observed in relation to maternal BLLs measured during each trimester, but
greatest for the second trimester, with 1.13% (95% CI: -2.63%, 0.37%) and 0.98% (95% CI: -2.83%,
0.88%) lower 24-month EMOCI and ORIEN percentiles, respectively, for 1 (ig/dL higher second
trimester BLLs. In an examination of the mediation of trimester-specific Pb exposure by DNA
methylation at several previously identified CpG sites, the authors observed both enhancing and

3-164

-------
suppressive effects of DNA methylation on the association between blood Pb and neurocognitive
outcomes, depending on the gene locus, with methylation at the majority of loci playing a suppressive
role. Shekhawat et al. (2021) similarly reported null but slightly inverse associations between cord blood
Pb and social-emotional scores from the BSID-III in a study of mother-child pairs in western Rajasthan,
India. Neurocognitive assessments were conducted at the average age of 6.5 months. Additionally, in an
analysis of the Navajo Birth Cohort Study, Nozadi et al. (2021) reported imprecise inverse associations
between maternal BLLs and communication (|3 = -0.15 [95% CI: -0.58, 0.28]) and personal-social (|3 =
-0.11 [95% CI: -0.72, 0.50]) domain scores on the ASQ:I at 10 months.

Some cohort studies examined interactions between Pb levels and the levels of other trace
elements (Nvanza et al.. 2021; Dohertv et al.. 2020; Lin et al.. 2013). Lin et al. (2013) measured maternal
blood Pb and assessed child development (including social and self-care skills) in the TBPS with the
CDIIT, as described in Section 3.5.1.2. The authors observed that children with high Pb exposure (>75th
percentile: 1.65 (ig/dL) had lower social DQs (|3 = -5.89 [95% CI: -10.81, -0.97]) compared with those
with low prenatal Pb exposure. In addition, the authors reported lower social or self-help DQs among
those with higher Pb and Mn concentrations in an interaction analysis. Nvanza et al. (2021) measured Pb,
Hg, Cd, and As concentrations using dried blood spots from pregnant mothers at 16-27 weeks of
gestation in Northern Tanzania. Adjusting for maternal age, maternal education, maternal and parental
occupation, number of under-five siblings at home, family socioeconomic wealth quintile, infant sex,
infant age, birth weight, and height and weight at neurocognitive testing, the authors did not observe an
association between high Pb exposure (>3.5 (ig/dL) and social impairment on the MDAT, which is
described in Section 3.5.1.2. However, an interaction analysis with maternal blood Hg levels showed that
children highly exposed to both Hg (>0.08 (ig/dL) and Pb were more likely to have global
neurodevelopmental impairment (PR= 1.40 [95% CI: 0.90, 2.10]). Dohertv et al. (2020) measured
concentrations of Pb and other metals (As, Cu, Mn, Se, and Zn) in maternal prenatal and postnatal
toenails and infant toenails at 6 weeks of life from mother-infant pairs in the New Hampshire Birth
Cohort. The three exposure assessments estimated exposures that occurred during periconception and
early pregnancy, mid-pregnancy, and late pregnancy and early neonatal life, respectively. The authors
observed mostly negative but imprecise associations between prenatal and child toenail Pb levels and total
SRS-2 scores. They also observed mostly positive but imprecise associations between postnatal maternal
and child toenail Pb levels and the adaptive skills composite on the BASC-2 (see Section 3.7, Table
3- 13E). Pb concentrations did not appear to interact with other metals on the total SRS-2 score or the
BASC-2 adaptive skills composite, and sex-stratified analyses revealed inconsistent associations among
girls.

An additional study assessed effect modification by maternal psychosocial measures. Zhou et al.
(2017) investigated the interactions of maternal BLL in whole blood and maternal prenatal stress levels
with child development (including adaptive behavior and social domains) using the GDS. Among those
with high maternal stress levels (GSI: P75-P100), adaptive behavior DQs were 17.93 points lower (95%
CI: -35.83, -0.03) per loglO-transformed (ig/dL higher maternal BLL. Social behavior DQs were also

3-165

-------
inversely associated with maternal BLL in children of mothers with high stress levels (|3 = -41.00 [95%
CI: -63.11, -18.89] per log-10 transformed unit of BLL).

One cross-sectional study (Ruebner et al.. 2019) evaluated the association between concurrent
BLL and neurocognitive outcomes including adaptive skills among children with CKD, using parent
ratings on the BASC-2. This study is discussed in more detail in Section 3.5.2.1.1. Higher BLL was
associated with worse adaptive skills composite scores (|3 = —3.1) in univariable analyses; however, this
association did not remain after adjusting for key sociodemographic and clinical confounders.

One additional prospective study (Vigeh et al.. 2014) measured domains of social and self-help
skills but presented only associations with composite neurodevelopment test scores (described in Section
3.5.1.2), which impedes parsing of specific social cognition effects of Pb exposure. In addition, Kim et al.
(2018b) evaluated concentrations of Pb in maternal serum, cord blood, urine, and breast milk in
association with neurodevelopmental and behavioral outcomes, including social quotient (SQ) measures
from the SMS among 13-24-month-old children. However, they reported only statistically significant
results in the paper, precluding quantitative results for blood Pb and SQ.

Recent epidemiologic studies utilized a wide range of outcome measures, including diagnostic
tests of autism (e.g., ICD code, DSM classification, ASSQ, and the Autism Diagnostic Observation
Schedule [ADOS]), behavior rating systems (e.g., CBCL, SRS-2, BASC-2, and SMS), and
neurodevelopmental assessments with social behavioral subtests (e.g., BSID-II and III, CDIIT, ASQ:I,
GDS, MDAT, and ECDI). Psychometric tests of social cognition often add valuable dimensional
information regarding the severity and type of social deficit among children with autistic traits, and the
wide variety of tests used in the evaluated studies provided insight into diverse aspects of problems with
social cognition, including communication, adaptive and self-help skills, social engagement, and
emotional behavior. One limitation, however, is that this variety complicates a straightforward
interpretation of results due to the lack of consistency of measures. Many behavioral tests provide
outcomes that overlap with domains discussed in other sections such as externalizing behavior (Section
3.5.3) and internalizing behavior (Section 3.5.4), which can limit parsing of effects. Vigeh et al. (2014)
examined social and adaptive skills but reported quantitative results using only the global
neurodevelopmental composite score from the ECDI. Other studies (Nvanza et al.. 2021; Rvgicl et al..
2021) used rating subscale measures (i.e., EMOCI and ORIEN from the BRS; social development score
from the MDAT) that are not widely used in the literature, making it difficult to compare results across
studies.

3.5.7.1.1 Summary

Two recent high-quality studies of Pb exposure and ASD reported positive associations between
increased Pb exposure and higher risk of ASD diagnosis or symptomatology (Aroraet al.. 2017; Kim et
al.. 2016). One retrospective study also observed a positive association between greater autism severity

3-166

-------
and current BLL (Dong et al.. 2022). However, some case-control studies did not find evidence of a
positive association (Rahbar et al.. 2021; Skogheim et al.. 2021; Rahbar et al.. 2015). Although most
autism studies except one (Kim et al.. 2016) were case-control studies, two of the case-control studies
accounted for temporality of exposure and outcome by analyzing prenatal maternal blood (Skogheim et
al.. 2021) or using tooth Pb measurement methods that allow ascertainment of perinatal Pb exposure
levels (Arora et al.. 2017). Several cohort studies observed null or slight impairments of social
dimensions scores. Recent studies of social cognition in children without ASD used a wide variety of
psychosocial and neurodevelopmental instruments, such as BASC-2, BSID-II and III, CDIIT, ASQ:I,
GDS, SRS, SMS, MDAT, and ECDI, to obtain scores of social, emotional, and adaptive abilities. These
studies were mostly prospective in design and accounted for some key potential confounders, including
maternal age, parental education, SES, and caregiving.

3.5.7.2 Toxicological Studies of Social Cognition and Behavior

The previous ISA incorporated evidence of the effects of Pb exposure on social cognition and
behavior. Donald et al. (1986) reported sex-specific effects of Pb exposure on social investigatory
behavior in mice, wherein males and females exposed to Pb displayed enhanced social interaction but at
different times after exposure. In a subsequent publication, Donald et al. (1987) reported that Pb exposure
increased non-social behavior in males while females displayed decreased non-social behavior. The
previous evidence suggests that Pb may influence social behavior in rodents in a sex-specific manner, but
the direction of the effect was not clear.

There is limited recent toxicological evidence available on the effects of Pb exposure on social
cognition and behavior. A single study by Tartaglione et al. (2020) examined homing test and ultrasonic
vocalizations (USV). USV are calls emitted by pups when separated from their mother and siblings and
are markers of early emotional and communication development. Pups prenatally and lactationally
exposed to Pb exhibited reduced numbers of calls at PND 4 and 12, with no significant differences at
PND 7 and 10 from control animals. The same study also performed a homing test, which assesses
discriminative performance and maternal preference behavior by separating the pup from the dam and
recording the time taken to return to the nest from a maze. The time spent is a measure of both olfactory
discrimination and social preference. The authors reported no difference in homing test performance
between control and Pb-exposed pups at PND 12 (Tartaglione et al.. 2020). In summary, there is limited
evidence from the toxicological literature examining potential relationships between developmental Pb
exposure and social behavior, which represents an area of uncertainty.

3-167

-------
3.5.7.3 Relevant Issues for Interpreting the Evidence Base

3.5.7.3.1 Concentration-Response Function

Evaluation of the shape of the C-R function in recent studies of social cognition is limited,
making it challenging to draw conclusions. Across studies, associations between Pb exposure and social
cognition and behavior were observed at median or geometric mean maternal and cord BLLs ranging
from 3.3 to 5.5 (ig/dL, and BLLs measured in children ranging from 1.6 to 3.9 (ig/dL (Table 3-8). Kim et
al. (2016) used penalized regression splines to examine the C-R relationship between BLLs at 7-8 years
old and SRS scores at 11-12 years old. The C-R relationship was approximately linear across the range of
the BLL distribution, though there is more confidence in the shape of the C-R relationship (i.e., more
narrow confidence limits) closer to the mean, where there is a higher density of observations. Spline
models for most of the SRS subscales are also approximately linear, except for social cognition, which
has a sublinear relationship with BLLs (i.e., a smaller slope below the mean).

3.5.7.3.2 Potentially At-Risk Populations

Maternal Stress

There is limited evidence that maternal stress modifies the association between Pb exposure and
social cognition. Stratifying by maternal stress, Zhou et al. (2017) found that social behavior (|3 = -41.00,
95% CI: -63.11, -18.89 per log-10 transformed unit of BLL) and adaptive behavior (|3 = -17.93, 95%
CI: -35.83, -0.03 per log-10 transformed unit of BLL) in toddlers were inversely associated with BLLs
among children of mothers with high prenatal stress. In contrast, adaptive behavior appeared to have a
positive but imprecise relationship (|3 = 7.57, 95% CI: -0.12, 15.27 per log-10 transformed unit of BLL)
with BLLs among children of mothers with low prenatal stress, while the association with social behavior
was null in the same population.

Co-exposure to Other Metals or Chemicals

A limited number of studies examined co-exposures to other metals as potential modifiers of the
relationship between Pb and social cognition. Lin et al. (2013) observed slight impairments to social and
self-help DQs among those with high concentrations of both Pb (>1.65 (ig/dL) and Mn (>5.93 (.ig/dL).
Nvanza et al. (2021) conducted interaction analyses of Pb and various neurodevelopmental outcomes with
Hg, Cd, and As, but did not report results for social skills.

3-168

-------
Gene-Environment Interactions

A single study evaluated the role of DNA methylation as a mediator of the relationship between
Pb exposure and social cognition and behavior. Rygiel et al. (2021) found both enhancing and
suppressing effects of DNA methylation at several CpG sites in mediation analyses. Methylation of
cg23280166 within CCSER1, a gene which has been associated with ADHD, suppressed the association
between second trimester Pb levels and ORIEN and EMOCI scores at 24 months old, while methylation
at cgl8515027 (GCNT1), positively mediated the association between first and second trimester BLLs
and 24-month EMOCI scores. Likewise, DNA methylation of cg23280166 (VPS11) also positively
mediated the relationship between third trimester BLLs and 24-month EMOCI scores.

Pre-existing Conditions

Although no recent studies evaluated pre-existing conditions as potential effect modifiers, one
study evaluated the relationship between Pb exposure and adaptive behavior among children with CKD.
After adjusting for sociodemographic and CKD-related variables, they did not report quantitative results
because they did not observe a statistically significant association (Rucbncr et al.. 2019).

Sex

There is limited evidence on sex as a modifier of the association between Pb exposure and social
cognition and behavior. Dohertv et al. (2020) observed inconsistent associations between Pb and SRS-2
total and BASC-2 adaptive skills composite scores in sex-stratified analyses. Female infant toenail Pb
concentration was positively associated with adaptive skills (|3 = 0.26 [95% CI: 0.07, 0.45] per log-2
transformed unit of BLL) and maternal prenatal toenail Pb was negatively associated with adaptive skills
in female infants (|3 = -0.19 [95% CI: -0.34, -0.04] per log-2 transformed unit of BLL), but associations
in male infants were null. Sample size limited statistical precision in sex-stratified analyses, which may
help explain these inconsistencies.

3.5.7.3.3 Confounding

Several sociodemographic characteristics were considered as potential confounders in recent
epidemiologic studies. Child age at outcome measurement was included in all but three studies
(Shekhawat et al.. 2021; Kim et al.. 2018b: Vigeh et al.. 2014) and child sex was included in all studies
but two (Nozadi et al.. 2021: Vigeh et al.. 2014). Parental education, which was consistently associated
with BLLs and measures of social cognition and/or autism status, was adjusted for or considered in all
studies except Rygiel et al. (2021). However, Rygiel et al. (2021) was the only study to include maternal
IQ as a potential confounder. Many studies also included SES among their modeled covariates (Nvanza et

3-169

-------
al., 2021; Rygiel et al.. 2021; Ruebner et al., 2019; Zhou et al., 2017; Vigeh et al.. 2014). Quality of

parental caregiving (e.g., HOME score) was included in only one study (Lin et al., 2013).

Various pregnancy and birth factors are also relevant for consideration as potential confounders.
Maternal age is strongly associated with autism risk (Sandin et al„ 2012) and is correlated with Pb
exposure (Ettinger et al„ 2020); hence, lack of inclusion in models may introduce bias. Additionally,
autism, like many developmental disorders, is more prevalent as delivery diverges in both directions from
40 weeks of gestation. As such, gestational age, birth weight, and maternal age were consistently included
as a potential confounder in most analyses. Breastfeeding, parity, maternal smoking and alcohol intake,
and food consumption during pregnancy were also included in multiple studies.

Genetics may also play a large role in the association between Pb exposure and social cognition
abilities. Aroraet al. (2017) used a case-control design with twin pairs, which allowed for matching on
genetic factors to some extent. Rahbar et al. (2021) evaluated interaction effects of glutathione S-
transferase (GST) genes (GSTP1, GSTM1, and GSTT1), which have been linked to detoxification of
environmental pollutants and to autism status. Rygiel et al. (2021) examined mediation by DNA
methylation at various CpG sites linked to prenatal Pb levels.

Co-exposures and mixtures with other trace metals were considered in several studies. Nozadi et
al. (2021) found positive correlations of BLLs with Mn and Cd. The authors used an algorithm to identify
the control variables for each metal they analyzed, including all co-occurring metals and demographics;
however, none met the inclusion criteria of being significantly associated with both the exposure and
outcome, and the final model did not include any covariates. Additionally, Lin et al. (2013) and Nyanza et
al. (2021) reported that Pb was positively correlated with Mn and As, and Cd and Pb, respectively.
However, neither study adjusted for metals in their analyses. Kim et al. (2016) adjusted for Hg and was
the only study to adjust for a co-occurring metal in its final model.

3.5.7.3.4 Lifestages

No epidemiologic studies examining the relationship between Pb exposure and social cognition
and behavior in children were included in the 2013 Pb ISA (U.S. EPA, 2013). Recent studies
demonstrated that BLLs measured during various lifestages and time periods (i.e., prenatal, early
childhood, later childhood, and concurrent with outcome assessment) are associated with ASD and
decrements in social cognition. Due to differences in study designs and the variety of psychometric tests
used to assess aspects of social cognition, it is difficult to compare the magnitude of associations across
studies to characterize important lifestages and time periods of Pb exposure. There is some examination
of different exposure measurement windows within studies. In the case-control study of twins described
previously, Arora et al. (2017) used laser ablation-inductively coupled plasma-mass spectrometry (ICP-
MS) to estimate pre- and postnatal Pb exposure from shed deciduous teeth. Differences in tooth Pb levels
were consistently higher in discordant ASD twins across the exposure period (20 weeks prenatal to 30

3-170

-------
weeks postnatal) compared with concordant and control twins, with bimodal peaks around 10 to 15 weeks
before birth and 10 to 20 weeks after birth (see Figure 3-14). This is consistent with results from a birth
cohort study that reported negative associations between maternal BLLs and social cognition in infants
(Rvgiel et al.. 2021). The observed associations were strongest in magnitude with maternal BLLs
measured in the second trimester compared with BLLs in the first and third trimesters. Although the
limited number of studies that evaluate different exposure windows makes it difficult to draw firm
conclusions on critical lifestages, the nature of ASD as a developmental disorder suggests that prenatal
and early infant exposures may be of particular importance.

It should be noted that children with ASD have a high prevalence of pica, a compulsive eating
behavior of non-food items (Fields et al.. 2021). Thus, children with ASD may have elevated BLLs due to
their higher likelihood of ingesting soil or other materials contaminated with Pb, rather than Pb exposure
causing ASD. As there is potential for reverse causation, accurately ascertaining the time of exposure
measurement is crucial in order to determine whether a causal effect of Pb on ASD exists, and studies
with exposure metrics that precede pica behavior would mitigate this concern. Such metrics include bone
Pb, tooth Pb (Arora et al.. 2017). and cord or maternal blood Pb (Nozadi et al.. 2021; Nvanza et al.. 2021;
Rvgiel et al.. 2021; Shekhawat et al.. 2021; Skogheim et al.. 2021; Kim et al.. 2018b; Zhou et al.. 2017;
Vigeh et al.. 2014; Lin et al.. 2013).

3.5.7.4 Summary and Causality Determination: Social Cognition and Behavior

The 2013 Pb ISA (U.S. EPA. 2013) did not include a causality determination for social cognition
and behavior in children. There were no epidemiologic studies on social cognition and behavior in
children in the previous ISA, and only a few toxicological studies that examined social behavior in mice.
The number of studies examining autism and social cognition in relation to Pb exposure has increased
substantially since the 2013 Pb ISA (U.S. EPA, 2013), highlighted by recent epidemiologic studies that
provide some evidence that Pb exposure is associated with increased ASD incidence and symptomology,
as well as decrements in social, emotional, and adaptive abilities. Recent toxicological evidence, along
with studies reviewed in the 2013 Pb ISA, provide some evidence of Pb-induced changes in social
behavior in mice, but the direction of the observed changes was inconsistent.

A recent novel epidemiologic analysis of twins provides strong evidence of an association
between Pb exposure and ASD. Arora et al. (2017) examined tooth Pb levels with respect to ASD status
among discordant and concordant twin pairs and observed higher Pb levels in the affected twin among
discordant monozygotic and dizygotic pairs. In contrast, concordant twins demonstrated similar levels of
exposure. This study also provided some insight into potentially sensitive time windows of exposure in
which the association between tooth Pb levels and autistic status was highest between 10-15 weeks
before birth and 10-20 weeks after birth. Additional support was provided by a prospective cohort study,
which reported that the number and severity of autistic behaviors in young children was positively

3-171

-------
associated with low BLLs (geometric mean: 1.58-1.64 (ig/dL) at several points prior to outcome
assessment (Kim et al.. 2016). There is some uncertainty about the relevance of the exposure window in
this study given that the earliest Pb measurements occurred at 7-8 years old, which is close to the
outcome assessment age (11-12 years old) and later than autistic behaviors typically manifest.
Additionally, covariates examined in this study did not include maternal age, which is an important
potential confounder for developmental disorders like autism; therefore, lack of adjustment for this
variable weakens the conclusions that can be drawn from the analysis. Dong et al. (2022) provides
support for the positive association between autism severity and BLLs among children 2 to 13 years old at
low levels of current blood Pb (mild group mean: 2.58 (ig/dL; moderate/severe group mean: 3.25 (.ig/dL);
however, the study's retrospective design and the wide range of the ages of assessed children introduce
uncertainty regarding potential reverse causality.

Several prospective studies among children without autism provide some additional support for
associations between Pb exposure and measures of social impairment in children (Nozadi et al.. 2021;
Nvanza et al.. 2021: Rygiel et al.. 2021: Shekhawat et al.. 2021: Zhou et al.. 2017: Lin et al.. 2013).
Median or geometric mean maternal and cord BLLs in these studies ranged from 3.3 to 5.5 (ig/dL, and
BLLs measured in children ranged from 2.7 to 3.9 (ig/dL. These studies had moderate to good follow-up
participation rates, and follow-up durations ranged from 6.5 months to 3 years. Furthermore, they
demonstrated good confounder control, adjusting for maternal age and some measure of SES or parental
education. Notably, the use of non-specific composite test scores (Vigeh et al.. 2014) and lesser-used
subscales (Nvanza et al.. 2021: Rygiel et al.. 2021) limits the specificity and generalizability of some
studies. Additionally, results from recent studies were not entirely consistent, as some analyses did not
observe associations (Rahbar et al.. 2021: Skogheim et al.. 2021; Dohertv et al.. 2020; Ruebner et al..
2019; Rahbar et al.. 2015). These included mostly case-control studies, one prospective cohort study, and
one cross-sectional study. Although all three case-control studies adjusted for maternal age and various
relevant covariates among matched pairs, Ruebner et al. (2019) did not.

Two toxicological studies in the 2013 Pb ISA reported a potential sex-based effect modification
of the effect of Pb exposure on social behavior (Donald et al.. 1987. 1986). Female and male mice
exhibited social interaction and non-social behavior at different timings and in different directions. One
recent study observed that rats exposed to Pb made fewer ultrasonic vocalizations than did control rats at
PND 4 and 12 but not at PND 7 and 10 (Tartaglione et al.. 2020). The authors additionally did not
observe differences between exposed and control rats on the homing test, which evaluates olfactory
discrimination and social preference.

In summary, the body of evidence is suggestive of, but not sufficient to infer, a causal
relationship between Pb exposure and social cognition and behavior in children. The strongest
evidence supporting this causality determination comes from a novel case-control study in twins that
provides strong support for a positive association between dentine Pb levels and autism risk. There are a
number of recent prospective epidemiologic studies that provide supporting evidence of a positive

3-172

-------
association of increases in BLLs with reduced social cognition and increased autistic behaviors in
children, but the evidence is not entirely consistent and is limited by the potential for unmeasured
confounding by maternal age or the potential for reverse causality due to the timing of exposure in studies
examining blood Pb levels. Furthermore, the wide range of social cognition measures used in the
evaluated studies simultaneously adds dimensionality and complicates interpretation of the results. Only
one recent experimental animal study on Pb exposure and social cognition was available. This study,
combined with the toxicological evidence reviewed in the previous ISA suggests that Pb exposure may
influence social cognition and communication, though the direction of these effects is inconsistent. Thus,
while the limited experimental animal evidence provides some coherence with the epidemiologic
evidence, a number of uncertainties remain. The key evidence, as it relates to the causal framework, is
summarized in Table 3-8.

Table 3-8 Evidence that is suggestive of, but not sufficient to infer, a causal
relationship between Pb exposure and social cognition and
behavior in children

Rationale for

Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels
Associated with Effects0

Consistent evidence
from a few high-
quality epidemiologic
studies with relevant
blood, bone, and
tooth Pb levels

Greater difference in tooth Pb
levels among twins discordant
for ASD status than among
concordant twins.

Lower scores on test of social
cognition in a prospective study
in South Korea in association
with earlier childhood and
concurrent mean BLLs.

Arora et al. (2017)

Kimetal. (2016)

Deciduous tooth Pb NR
(early and postnatal Pb
levels)

Child blood Pb GM:
7-8 yr: 1.64 pg/dL
9-10 yr: 1.58 pg/dL
11—12 yr: 1.58 pg/dL

Evidence from multiple
prospective cohort studies for
small decrements in scores on
tests of social cognition among
children without autism ages
6.5 mo-3 yr at low levels of
exposure.

Shekhawat et al. (2021)

Rvaiel et al. (2021)

Cord blood Pb GM: 4.14
pg/dL

Mat. blood Pb GM (SD):
1st tri.: 5.27 (1.93) pg/dL
2nd tri.: 4.74 (1.96) pg/dL
3rd tri.: 4.98 (1.93) pg/dL

Infant blood GM (SD):
12 mo: 3.92 (1.80) pg/dL
24 mo: 3.49 (1.93) pg/dL

Zhou et al. (2017)

Mat. blood Pb GM (95%
CI): 3.30 (3.05, 3.57) pg/dL

Section 3.5.7.1

3-173

-------
RcL°u"alTtJ0r	Key Evidence"	References'

Determination3	Associated witn tracts

Epidemiologic
studies help rule out
chance, bias, and
confounding with
reasonable
confidence

Prospective studies had
population-based recruitment
with moderate to good follow-up
participation not conditional on
bone Pb/BLLs and social
cognition scores.

Section 3.5.7.1

All studies controlled for	Table 3-13E

maternal age, education and/or
SES. Some controlled for
HOME score, maternal IQ, and
exposures to other pollutants.

Limited supporting
evidence from case-
control and cross-
sectional studies

Null findings from case-control
studies conducting adjusted
mean comparisons of ASD
cases and typically developing
controls with lower and higher
mean or median BLLs,
adjusting for maternal age,
various demographic and
lifestyle factors and dietary
consumption.

Skoaheim et al. (2021)

Rahbaret al. (2015)

Rahbaret al. (2021)

Mat. blood Pb GM
cases: 0.83 |jg/dL
controls: 0.88 |jg/dL

Child blood Pb GM (SD)
cases: 2.25 (2.23) |jg/dL
controls: 2.73 (1.85) pg/dL

Child Blood Pb GM
cases: 7.11 pg/dL
controls: 8.48 pg/dL

Greater BLLs among children
2-13 years old with
moderate/severe vs. mild
autism in a retrospective study.

Dong et al. (2022)

Child blood Pb mean (SD)
Mild: 2.58 (1.08) pg/dL
Moderate/severe: 3.25
(1.89) pg/dL

Null finding from cross-sectional
study. Lacked control for
maternal age at delivery.

Ruebner et al. (2019)

Child blood Pb med: 1.2
pg/dL

3-174

-------
RcL°u"alTtJ0r	Key Evidence"	References'

Determination3	Associated witn tracts

Limited experimental Mixed evidence of enhanced or
animal evidence at reduced social interaction
relevant exposures behavior among Pb-exposed
mice. Some suggestion of sex-
specific effect modification.

Reduced ultrasonic
vocalizations among Pb-
exposed rats. No evidence of a
difference on homing tests of
olfactory discrimination and
social preference.

Donald et al. (1986)
Donald et al. (1987)

Tartaqlione et al. (2020)	Med blood Pb after

exposure during
pregnancy and lactation:
0.26 |jg/mL PND 23

ASD = autism spectrum disorder; BLL = blood lead level; CI = confidence interval; GM = geometric mean; HOME = Health
Outcomes and Measures of the Environment; IQ = intelligence quotient; Mat = maternal; med = median; mo = month(s); NR = not
reported; Pb = lead; PND = postnatal day; SD = standard deviation; SES = socioeconomic status; tri = trimester; yr = year(s).
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble to the ISAs fU.S. EPA. 2015).

bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and,
where applicable, to uncertainties or inconsistencies. References to earlier sections indicate where the full body of evidence is
described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3.6 Nervous System Effects Ascertained during Adult Lifestages

The strongest evidence of Pb-associated nervous system effects in adults without occupational
exposure pertained to cumulative exposure and cognitive effects (U.S. EPA, 2013). Prospective studies
indicated associations of higher baseline tibia (means 19, 20 (ig/g) or patella (mean 25 |ig/g) Pb levels
with declines in cognitive function in adults (age >50 years) over 2- to 4-year periods. Pb-associated
cognitive function decrements were found with adjustment for potential confounding factors such as age,
education, SES, current alcohol use, and current smoking. Supporting evidence was provided by cross-
sectional studies, which found stronger associations with bone Pb level than concurrent BLL. Cross-
sectional studies also considered more potential confounding factors, including dietary factors, physical
activity, medication use, and comorbid conditions. The multiple exposures and health outcomes examined
in many studies reduced the likelihood of biased participation specifically by adults with higher Pb
exposure and lower cognitive function. Uncertainties remained due to residual confounding by age and
lack of information on the patterns of exposure associated with the BLLs observed in the epidemiologic
studies.

3.6.1 Cognitive Function in Adults

The evidence reviewed in the 2013 Pb ISA was sufficient to conclude that "a causal relationship
is likely to exist" between long-term cumulative Pb exposure and cognitive function decrements in adults

3-175

-------
(U.S. EPA, 2013). Prospective studies of the Normative Aging Study (NAS) and Baltimore Memory
Study (BMS) cohorts indicated associations of higher baseline tibia (means 19, 20 (ig/g) or patella (mean
25 (ig/g) Pb levels with declines in cognitive function in adults (age >50 years) over 2- to 4-year periods
among adults without occupational exposure (see Table 4-10 (U.S. EPA. 2013)). While the specific
covariates differed between studies, these bone Pb-associated cognitive function decrements were found
with adjustment for potential confounding factors such as age, education, SES, current alcohol use, and
current smoking. Supporting evidence was provided by cross-sectional analyses of the NAS, BMS, and
the Nurses" Health Study (NHS), which found stronger associations with bone Pb level than concurrent
BLL indicating the relative importance of long-term Pb exposure. Cross-sectional analyses considered
more potential confounding factors, including dietary factors, physical activity, medication use, and
comorbid conditions. The multiple exposures and health outcomes examined in many studies reduced the
likelihood of biased participation specifically by adults with higher Pb exposure and lower cognitive
function. The effects of recent Pb exposures on cognitive function decrements in adults were indicated in
Pb-exposed workers by associations found with BLLs, although these studies did not consider potential
confounding by other workplace exposures. The biological plausibility for the effects of Pb exposure on
cognitive function decrements in adults was provided by findings that relevant lifetime Pb exposures from
gestation, birth, or after weaning induce learning impairments in adult animals and by evidence for the
effects of Pb altering neurotransmitter function in the hippocampus, prefrontal cortex, and nucleus
accumbens (U.S. EPA, 2013).

Recent epidemiologic studies provide consistent evidence that higher cumulative exposure
indicated by bone Pb levels or childhood BLLs are associated with decrements in cognitive function
during young-, mid- or older-adulthood periods (Table 3-14E). Across populations, higher Pb levels were
associated with decrements in FSIQ, global cognitive function, executive function, visuospatial and
visuomotor skills, language, and memory. Much of this evidence was provided by extended analyses
(about 15 years of follow-up data) of the NAS and NHS cohorts considered in the 2013 Pb ISA, and
prospective cohort studies from Sweden and New Zealand that explored the effects of early childhood Pb
exposure (7-12 years) on IQ and various cognitive domains during young adulthood (18-19 years).
Findings from these recent prospective cohort studies, emphasize the important role of early childhood Pb
exposure and persistent effects on adult cognition after adjustments of various sociodemographic factors
and maternal and childhood IQ. Overall, the longitudinal design with longer follow-up periods, multiple
and repeatedly measured cognitive outcomes, and multiple risk factors and confounders accounted for in
the studies reduce the bias and strengthen the study findings related to the effects of Pb exposure on adult
IQ and cognitive function. Recent evidence from animal studies provide support that postnatal exposure
to Pb (either during adolescence or continuing into adulthood) negatively affects learning and memory in
rodents. Additionally, adult rodents exposed during early developmental periods displayed impairments in
tests of learning and memory conducted in adulthood (reviewed in Section 3.4). This suggests that early
life Pb exposure contributes to cognitive dysfunction that persists into adulthood, which is new evidence
in this review. Additionally, animals exposed to Pb during adulthood display similar cognitive
impairment, though there is still uncertainty regarding the influence of age on Pb exposures during

3-176

-------
adulthood. These studies add to the current evidence base suggesting a potential role of both early and
later life Pb exposures and biological plausibility for the effects of Pb exposure on cognitive function
decrements in adults.

A summary of the recent evidence, which is interpreted in the context of the entire body of
evidence, is provided in the subsequent sections. Measures of central tendency for Pb biomarker levels
used in each study, along with other study-specific details, including study population characteristics and
select effect estimates, are highlighted in Section 3.7, Table 3-14E (Epidemiology) and Table 3-4T
(Toxicology).

3.6.1.1 Epidemiologic Studies of Cognitive Function in Adults

Studies in the 2013 Pb ISA (U.S. EPA. 2013) found that higher bone Pb levels, indicating long-
term exposure to Pb, were associated with decrements in cognitive function in adults without
occupational Pb exposure. There was variability in associations across the various domains of cognitive
function tested within studies; however, higher bone Pb levels were associated with poorer performance
in most of the tests conducted. Further, discordant Pb associations across domains of cognitive function
are likely to reflect biologic variability or differences in the outcome pathophysiology. Across
populations, higher bone Pb levels were associated with decrements in executive function, visuospatial
skills, learning, and memory. Much of this evidence was provided by analyses of the BMS and NAS, with
additional findings reported in the NHS and smaller populations. The strongest evidence for bone Pb-
associated cognitive decrements demonstrated that higher tibia (means: 19, 20 (ig/g) and patella (mean:
25 (ig/g) bone Pb levels measured at baseline were associated with subsequent declines in cognitive
function over 2- to 4-year periods (Bandeen-Roche et al., 2009; Weisskopf et al., 2007). These findings
indicated that long-term Pb exposure may contribute to ongoing declines in cognitive function in adults.
These associations were found with adjustment for potential confounding by age, education, smoking, and
alcohol use in the NAS and age, sex, race, household wealth, and education in the BMS. While the NAS
and Nurses" Health Study included primarily white men and white women, respectively, the BMS
examined a more diverse population of men and women of various races and ethnicities.

In a recent analysis of the NAS cohort, Farooqui et al. (2017) examined the associations between
long-term Pb exposure quantified using bone biomarkers (mean patella Pb: 30.6 (ig/g and tibia Pb: 21.6
(ig/g) and longitudinal changes in cognition (repeatedly measured up to five visits over the 15 years
follow-up period) adjusted for age at the first cognitive test, education level, baseline smoking status, and
alcohol intake. The study found that higher patella bone Pb concentration (IQR: 21 |ig/g) was associated
with a 0.062 point lower baseline Mini Mental State Examination (MMSE) score (95% CI: -0.012,
0.003), 0.008 units/year MMSE decline (95% CI: -0.015, 0) over 15 years, and an increased risk of
having an MMSE score below 25 (threshold considered to represent cognitively not normal or at risk for
dementia) (hazard ratio [HR] = 1.10 (95% CI: 0.99, 1.21)). Similar but weaker and less precise

3-177

-------
associations were observed when tibia Pb and MMSE outcomes were assessed. The study also used
"global cognition" as a separate proxy for worsening cognitive impairment, and combined seven test
scores assessed in NES2, CERAD, and WAIS-R. Weaker associations were observed between both
patella and tibia Pb and global cognition (both baseline and longitudinal change). When separate
cognitive domains were assessed, patella Pb was associated with faster longitudinal decline in language
and memory domains, whereas similar but weaker associations were observed with tibia Pb.

A longitudinal study of women aged 45-74 enrolled in the NHS cohort (Power et al.. 2014) added
to the evidence provided by previous analyses. The authors examined the associations between Pb
exposures using bone and blood biomarkers (mean patella Pb: 12.6 (ig/g and tibia Pb: 10.5 |ig/g: mean
blood Pb: 2.9 (ig/dL) and cognitive decline (repeatedly measured using a telephone battery of cognitive
tests assessing learning, memory, executive function, and attention during 2-4 waves over the 13-year
follow-up period). Results were adjusted for alcohol consumption, smoking status, education, husband's
education, menopausal status and hormone therapy use, physical activity, ibuprofen use, aspirin use,
vitamin E supplementation, percentage of residential census tract of white race or ethnicity, and median
income of residential census track. A weak and imprecise association was observed for an excess annual
decline in the overall cognitive test Z-score per SD increase in tibia bone Pb concentration (-0.002
standard units; 95% CI: -0.005, 0.000). When individual cognitive tests were considered, a decline on the
East Boston Memory Test as well as immediate (a measure of episodic memory) and category fluency (a
measure of executive function and memory) was observed in relation to increased tibia Pb concentration
(Power et al.. 2014). There was little evidence for associations between patella Pb or blood Pb and the
decline in overall cognition, verbal memory, or individual cognitive tests.

Associations between either tibia or patella Pb concentration and cognitive function decline
observed in these longitudinal studies are supported by (Weuve et al.. 2013). a cross-sectional study that
assessed bone Pb and cognitive function (assessed using telephone cognitive assessment battery) among
participants from an existing case-control study of Parkinson's disease. The Pb concentration observed in
this study is similar to that reported in the NHS cohort. Separate analyses were performed for the group of
participants with PD and for all participants including both PD and control groups. Analysis of the PD
group showed that higher tibia Pb was significantly associated with worse overall performance (as shown
by the global cognitive score) and worse performance on the majority of telephone cognitive tests (in the
model adjusted for age at cognitive assessment, sex, race, education, smoking history). Patella Pb
concentration, however, was not consistently associated with cognitive performance. In the model with
both PD and control groups, interactions were observed for the association between tibia Pb and global
cognition by case-control status. Among the participants, a 10 (ig/g increase in tibia Pb corresponded to a
decrease in the global cognitive score by 0.12 standard units (95% CI: -0.22, -0.01), but the association
among controls was weak (0.06 standard units, 95% CI: -0.09, 0.20).

Evidence of the cognitive effects of cumulative Pb exposure observed in the NAS and NHS
studies is strengthened and extended further by the findings from cohort studies that examined the

3-178

-------
associations of childhood Pb exposure and continued long-term exposures with cognitive impairments in
young adults (18-19 years) (Skerfving et al.. 2015) or in mid-adulthood (38 or 45 years of age) (Reuben
et al.. 2020; Reuben et al.. 2017). SkerfVing et al. (2015) included samples of 7-12 year-old school
children in southern Sweden and followed them over time to examine the association between childhood
BLL (age 7-12 years old, mean blood Pb: 3.4 (ig/dL) and cognitive performance (IQ) assessed for
military conscription at 18-19 years of age using generalized linear models. The study found an IQ loss of
0.127 (95% CI: -0.209, -0.045) points per (ig/dL increase in childhood BLL for all participants and an IQ
loss of 0.204 (95% CI: -0.392, -0.016) points per (ig/dL increase in childhood BLL among those with
childhood BLLs <50 |ig/L. even in multivariable models adjusted for parent's income, education, and
father's IQ. Reuben et al. (2017) and Reuben et al. (2020) examined a New Zealand birth cohort with
participants born in 1972-1973 (a time when Pb exposure in New Zealand cities were higher than
international standards) who were part of the Dunedin Multidisciplinary Health and Development Study.
Infants were followed from birth through adulthood. Blood collection at 11 years of age provided blood
biomarker data for Pb (mean blood Pb: 10.99 (ig/dL). Cognitive performance was assessed using
objective tests of cognitive performance such as the WISC-R during childhood at ages 7 and 9 years.
Cognitive performance was also assessed using the WAIS-IV when participants were 38 years old
(Reuben et al.. 2017) and again when they were 45 years old (Reuben et al.. 2020). Reuben et al. (2020).
in addition to the objective tests, also included subjective reports of everyday cognitive functioning
(memory or attention problems) at age 45 years as provided by study participants and their nominated
informants. The studies examined the association between childhood blood Pb and adult cognitive
outcomes or cognitive decline (change in IQ score between childhood and mid-adulthood), using OLS
multiple regression models. Reuben et al. (2017) found that each 1 (ig/dL higher level of blood Pb in
childhood was associated with a 0.39-point lower score in adult FSIQ (95% CI: -0.67, -0.12), and a 0.32-
point decline (95% CI: -0.50, -0.15) after adjusting for sex, childhood IQ, maternal IQ, and childhood
SES. Similarly, with additional years of data (Reuben et al.. 2020) continued to show significant
associations between childhood BLL and IQ at 45 years of age. Each 1 (ig/dL higher childhood BLL was
associated with a -0.41 (95% CI: -0.68, -0.15) point decline in full-scale IQ from baseline. When using a
residualized change model to adjust for autocorrelation between baseline and follow-up IQ, the decline in
IQ was similar (-0.39 [95% CI: -0.58, -0.21]). The study also found that the relationship between
childhood blood Pb and adult IQ persisted and remained significant even after adjustment for brain
structure measures. Results pertaining to brain structure are discussed in Section 3.4.1.

Several recent cross-sectional studies also examined the associations of Pb exposure and
decrements in cognitive function. The majority used concurrent blood biomarkers (Sasaki and Carpenter.
2022; Xiao et al.. 2021; Przvbvla et al.. 2017; Souza-Talarico et al.. 2017; Khalil et al.. 2014; van
Wiingaarden et al.. 2011). A few studies used other biomarkers such as toenails (Meramat et al.. 2017).
bone (Weuve et al.. 2013). or urine (Sasaki and Carpenter. 2022). As described further below, results
were not entirely consistent across studies.

3-179

-------
Three of the cross-sectional studies examining the blood Pb-cognitive function association used
data from various NHANES cycles including participants aged 60-84 years (Sasaki and Carpenter. 2022;
Przvbvla et al.. 2017; van Wiingaarden et al.. 2011). van Wiingaarden et al. (2011) examined the
associations of blood Pb (mean blood Pb: 2.46 (ig/dL) with self-reported confusion and memory problems
using data from NHANES (1999-2008). Data from the 1999-2002 NHANES cycle were used to estimate
the association between BLL and performance on the Digit Symbol Substitution Test (DSST) scores.

After adjustment for age, sex, education level, ethnicity, poverty-income ratio (PIR), self-reported health
status, and comorbid conditions, no association of BLLs with self-reported confusion or memory
problems or DSST performance was observed (van Wiingaarden et al.. 2011). Two other studies,

Przvbvla et al. (2017) and Sasaki and Carpenter (2022). included the NHANES participants from 1999-
2002 and 2011-2014 cycles, respectively. These studies explored the cross-sectional associations of blood
and urine biomarkers of multiple metals and metalloids (separately and jointly) on cognitive function.
(Przvbvla et al.. 2017) used a path analysis approach to model multiple exposures of 14 chemicals
simultaneously while adjusting for multiple comparisons. The study found that the association of BLL
(Geometric mean: 2.17 (ig/dL) with lower cognition scores was attenuated when the model controlled for
smoking status. Specifically, a 1-SD increase in BLL was weakly associated with a slightly lower Digit
Symbol Coding (DSC) test score from WAIS-III (|3 = -0.10, 95% CI: -0.20, -0.00) after controlling for
co-exposure and sociodemographic covariates. The study also performed stratified analysis by sex and
age (above and below median age) and found a greater magnitude of associations for female and higher
age categories (>10%), despite a lack of statistical evidence of an interaction. (Sasaki and Carpenter.
2022) used two stage linear regression models. First, they performed single metal analyses separately for
each of seven metals or metalloids followed by a second analysis including multiple metals or metalloids
from the stage 1 analysis to examine the associations with immediate, delayed, and working memory
quantified using CERAD and DSST. When single metals were assessed, increased blood Pb concentration
was associated with decrements in performance on all three cognitive tests after adjusting for
sociodemographic, behavior, and clinical characteristics (immediate recall: |3 = -0.58, 95% CI: -0.91,
-0.24; delayed recall: |3 = -0.19, 95% CI: -0.35, -0.02; Digital Symbol Substitution: |3 = -1.08, 95% CI:
-2.12, -0.05). Multi-metal analysis stratified by age group (60-70 and >70 years old) suggested greater
declines in immediate recall among participants over the age of 70. Khalil et al. (2014) examined the
association between concurrent blood Pb concentration (mean blood Pb: 2.25 (ig/dL) and cognitive
function among a subset of non-institutionalized community dwelling non-Hispanic Caucasian men 65
years and older who participated in the Osteoporotic Fractures in Men Study (MrOS) cohort study.
Cognitive function was assessed using the Modified MMSE (3MS) and the Trail Making Test Part B.
Higher scores on the 3MS and faster time on the Trail Making Test Part B both represent better
performance. Multivariable analysis found no association between blood Pb concentration and cognitive
function (Khalil et al.. 2014).

Souza-Talarico et al. (2017) examined the association between blood Pb (and interactions
between blood Pb and Cd) and working memory capacity (WMC) in a population of 125 older adults
aged 50-82 years, in the metropolitan area of Sao Paulo, Brazil. The study also explored the mediating

3-180

-------
role of antioxidant capacity (using various oxidative stress biomarkers) in the heavy metals-memory
associations. Using regression models accounting for age, sex, income, and hemoglobin, the study did not
find an association between blood Pb (mean 2.1 (ig/dL) and WMC (|3 = 0.106, 95% CI:-0.208, 0.417);
however, an interaction between blood Pb and blood Cd level was observed as well as a significant
inverse association between the blood Cd x blood Pb interaction term and WMC was observed (Souza-
Talarico et al.. 2017). The Monte Carlo Method for Assessing Mediation test for mediation revealed that
the association between the blood Cd x blood Pb interaction term and WMC was significantly mediated
by total antioxidant capacity.

(Xiao et al.. 2021) examined the association between multiple metals (22 metals including Pb;
mean blood Pb: 5.15 (ig/dL) and cognitive function measured using MMSE in participants aged >60 years
from Guangxi, Southern China. The study used least absolute shrinkage and selection operator (LASSO)
penalized regression to identify main metals associated with cognitive function. Twelve metals (including
Pb) selected from LASSO were then explored in a multi-metal generalized linear regression model
adjusted for age, gender, education attainment, annual income, BMI, smoking, alcohol, insomnia, and
physical activity. No association was observed for blood Pb and cognitive function after adjustment for
other metals.

Notably, a limitation of cross-sectional studies of concurrent BLLs is that the relative contribution
of the recent versus past Pb exposure is not well characterized. A recent prospective study was designed
to address the uncertainties related to the exposure patterns associated with BLLs observed in studies of
adults (Yu et al.. 2021). Yu et al. (2021) examined the association of BLLs and neurocognitive
performance among newly hired employees at battery manufacturing and Pb recycling plants with no
previous occupational Pb exposure, a subset of participants in the Study for Promotion of Health in
Recycling Lead (SPHERL) cohort study. Baseline blood Pb concentration was measured, and the
participants were followed annually over a 2-year period to measure blood Pb biomarkers and assess if
higher recent occupational exposure to Pb was associated with neurocognitive dysfunction. The
participants completed the DSST and SCWT at baseline and annual follow-up visits. The geometric mean
blood Pb at baseline and first and second follow-up visits were 3.97 (ig/dL, 13.4 (ig/dL, and 12.8 (ig/dL,
respectively, showing an almost three-fold increase in blood Pb over the 2 years of occupational
exposure. The study used a linear mixed model to examine the changes in DSST and SCWT
corresponding to changes in blood Pb separately for the 1- and 2-year visits. Despite the three-fold
increase in blood Pb concentration, the study found no association between blood Pb and cognitive
function. The change in latency time and error rate based on the DSST test showed an increase from
baseline to follow-up, with an increase in the follow-up-to-baseline blood Pb concentration ratio, but the
association was weak and imprecise in the fully adjusted models (change in latency: 0.55%, 95% CI:
-0.33, 1.42; error rate: OR: 1.01, 95% CI: 1.00, 1.03)

3-181

-------
3.6.1.1.1 Summary

Longitudinal cohort studies evaluated in the 2013 Pb ISA found consistent evidence of an
association between increased long-term exposure to lead indicated by bone Pb levels and decreased
cognition in adults. Recent prospective cohort studies add to the body of evidence informing the
relationship between Pb exposure and cognitive performance in adults without occupational Pb exposure.
More specifically, recent cohort studies indicated that higher adult bone Pb levels, which indicate
cumulative Pb exposure (tibia mean range: 10.5, 21.6 (ig/g, patella mean range: 12.6, 30.6 |ig/g) or
childhood BLLs (mean range: 3.4 (ig/dL, 10.99 (ig/dL at 7-12 years of age), were associated with
decrements in cognitive function or IQ during young-, mid-, or older-adulthood periods (Table 3-14E).
There was some variability in the associations with various domains of cognitive function tested within
studies; however, variability in the associations observed across domains of cognitive function generally
reflects biologic variability or differences in the outcome pathophysiology rather than inconsistent study
results. Across studies, higher Pb levels were associated with decrements in FSIQ, global cognitive
function, executive function, visuospatial and visuomotor skills, language, and memory. Extended
analyses of the NAS and NHS cohorts with 13 to 15 years of follow-up add to the evidence base
(Farooaui et al.. 2017; Power et al.. 2014). These studies found associations of cumulative Pb exposure
with decrements in cognitive function in adults after adjustment for potential confounding by
combinations of factors including demographic, socioeconomic, behavioral, clinical, and neighborhood-
level factors. In addition, findings from recent prospective cohort studies in Sweden and New Zealand
that explored the effects of Pb exposure during childhood lifestages (7-12 years) on IQ and cognitive
effects during young adulthood (18-19 years) (Skcrfving et al.. 2015) and mid-adulthood (38-45 years)
(Reuben et al.. 2020; Reuben et al.. 2017). These studies found that higher childhood BLLs were
associated with declines in IQ ascertained in adulthood after adjustment for demographic and
socioeconomic factors, maternal IQ, and childhood IQ scores. These findings provide new insight into the
persistence of Pb-associated cognitive function decrements. Overall, the longitudinal design with longer
follow-up periods, multiple and repeatedly measured cognitive outcomes, and multiple risk factors and
confounders accounted for in epidemiologic studies investigating long-term cumulative exposure and
early childhood exposure reduce uncertainties and strengthen the evidence related to the association of Pb
exposure with cognitive function in adulthood. Sex (male versus female, premenopause versus
postmenopause) and age (young versus mid-aged versus old-aged adults) differences in bone kinetics and
turnover, as well as disease comorbidity, particularly at middle- and older-adulthood lifestages may
potentially lead to differences in bone Pb and blood Pb levels and add complexity when modeling the
associations since inclusion of only age or sex in the model may not fully account for these differences.

3-182

-------
3.6.1.2

Toxicological Studies of Cognitive Function in Adults

3.6.1.2.1 Learning and Memory - Morris Water Maze

This section specifically reviews studies that exposed animals to Pb during either adulthood or
late adolescence. Studies that exposed animals during development (i.e., pregestation, gestation, lactation)
are reviewed in Section 3.5.1.3.2. Animals exposed to Pb via drinking water in adulthood displayed
impaired learning and memory. Using the Morris water maze, Mansouri et al. (2012) found that short-
term Pb exposure (50 mg/L Pb in drinking water, PND 70 to 100), which produced mean BLLs of 8
(ig/dL, significantly impaired both learning and memory, though the magnitude of the effect on memory
was smaller compared with the effect observed in other studies of developmental exposure (Section
3.5.1.3.2). Also using the Morris water maze, Mansouri et al. (2013) reported that long-term Pb exposure
(50 ppm in drinking water, PND 60 to 240), which produced peak BLLs of 11-19 (ig/dL, significantly
impaired learning and memory performance in both sexes. Cognitive impairment was also observed in
studies that utilized daily administration of Pb via gavage. Singh et al. (2019) reported significantly
increased escape latencies and path lengths (i.e., distance traveled to reach the platform, another measure
of learning) in exposed rats following long-term Pb exposure via gavage (2.5 mg/kg, PND 90 to 180),
which produced peak BLLs of 28 (ig/dL. No probe phase was conducted in this study. Additionally, Su et
al. (2016) found that male rats gavaged with Pb solutions daily (200 ppm, PND 20 to 76) displayed
significant impairments in both the learning and memory components of the Morris water maze.

In two recent studies, Zou et al. (2015) and Han et al. (2014) reported significant learning and
memory deficits following short-term exposure of juvenile animals. Zou et al. (2015) exposed mice from
PND 35 to 56 with mean BLLs of 22 (ig/dL, while Han et al. (2014) exposed rats from PND 21 to 42 and
reported mean BLLs of 15 (ig/dL. Other studies examined the effects of long-term Pb exposure on
juvenile rodents and found similar effects. For example, (An et al.. 2014) exposed groups of juvenile rats
to multiple doses of Pb for 56 days (PND 28 to 84). All examined doses (100, 200, and 300 ppm in
drinking water) produced BLLs relevant to this ISA. At the time of Morris water maze assessment (PND
84), the mean BLL ranged from 11 to 23 (ig/dL. All exposed animals displayed impaired memory during
the probe trial relative to controls. Only animals in the two highest dose groups were reported to show
learning deficits during training, suggesting that, with juvenile exposures, Pb may have a greater effect on
memory than learning processes. Another study that exposed juvenile mice to Pb in drinking water (0.2%)
for 90 days (PND 28 to 112) assessed Morris water maze performance in the same animals at multiple
time points during and immediately following exposure (Wu et al.. 2020b). This long-term exposure
produced relatively high mean BLLs of 28 (ig/dL, and all exposed animals showed signs of impaired
learning and memory in the maze. Interestingly, both measures of cognition improved in the exposed
animals over time, which may reflect either increasing familiarity with the task or clearance of Pb over
time. In contrast to all other studies in young and juvenile animals, Li et al. (2013) reported that rats given
Pb in drinking water for 84 days (from PND 28 to 112) with peak BLLs of 16 (ig/dL showed no

3-183

-------
indication of learning or memory impairment in the Morris water maze. Despite this one discrepant study,
recent evidence supports the notion that postnatal exposure to Pb (either during adolescence or continuing
into adulthood) negatively affects learning and memory in rodents, which contrasts with several of the
key studies reviewed in the previous ISA.

3.6.1.2.2 Summary

Four recent studies of rodents with exposure resulting in mean BLLs <30 (ig/dL add to the
evidence informing the association of both short- and long-term Pb exposure during adulthood with
measures of learning and memory in rodents. While these studies are consistent with one another,
toxicological evidence for the effects of Pb on cognitive function in adults remains limited. Additionally,
a few recent studies in juvenile rodents also provide some support for the association between postnatal
Pb exposure either during adolescence or continuing into adulthood and cognitive impairment,
specifically learning and memory.

3.6.1.3 Relevant Issues for Interpreting the Evidence Base

3.6.1.3.1 Concentration-Response Function

The 2013 Pb ISA reviewed a small number of studies that examined the shape of the C-R
relationship between blood or bone Pb levels and cognitive function. Studies using BMS and NAS
cohorts assessed nonlinearity using quadratic terms, penalized splines, or visual inspection of bivariate
plots. Prospective analyses of the NAS cohorts provided some evidence of nonlinearity (Wang ct al..
2007; Weisskopf et al.. 2007) Figures 4-7 and 4-8 from 2013 Pb ISA). Weisskopf et al. (2007) found that
a 20 (ig/g difference in patella Pb level was associated with a 0.07-ms increase in response latency (95%
CI: 0.04, 0.12; larger values mean slower reaction times in the pattern comparison test) among all men
and a 0.15-ms increase among men with patella Pb level <60 |ig/g. These results suggest that Pb-
associated latency worsens with increasing Pb up to 60 (ig/g and levels off at higher values. Wang et al.
(2007) found that among NAS men with an HFE gene variant, there was a larger decline in MMSE score
(a global examination of cognitive function with low scores indicating poor cognitive performance) per
unit increase in tibia Pb level at higher tibia Pb levels.

In the current review, the shape of the C-R function was not assessed in studies that examined the
associations of Pb biomarkers with cognitive function in adults. The majority of studies selected
analytical models that assumed linear associations in the Pb-cognitive function associations. A few
studies in the recent review examining the influence of childhood Pb exposure on cognitive impairments
at the young- (18-19 years) (SkerfVing et al.. 2015) or mid-adulthood periods (38 or 45 years of age).
Reuben et al. (2017) and Reuben et al. (2020) performed separate analyses for the subsets of the

3-184

-------
population exposed to higher and lower Pb levels to examine possible nonlinear relationships or threshold
effects. Reuben et al. (2017) found a 1.97-IQ-point reduction in adulthood (95% CI: -3.34, -0.59) for the
overall sample, a 4.25-IQ-point reduction for individuals above the level of concern, and a 2.73-IQ-point
reduction for individuals below the level of concern for each 5-(.ig/dL increase in the childhood blood Pb.
Similarly, IQ decline from childhood to adulthood suggested a mean decline of a 1.61 IQ points (95% CI:
-2.48, -0.74) in adulthood for the overall sample, a mean decline of 1.68 IQ points for participants above
the level of concern, and a mean increase of 1.22 IQ points for participants below the level of concern for
each 5-(ig/dL increase in the childhood blood Pb. Similar results were observed by (Reuben et al.. 2020)
in the same population studied in (Reuben et al.. 2017) and followed till 45 years of age. SkerfVing et al.
(2015) examined the influence of early childhood Pb exposure on long-term cognitive impairments at
young adulthood (18-19 years) using generalized linear models. The study found an IQ loss of 0.127
(95% CI: -0.209, -0.045) points per (ig/dL increase in childhood BLLs for all participants, and a slightly
larger IQ loss (i.e., 0.204 [95% CI: -0.392, -0.016] point per (ig/dL increase in childhood BLL) for the
populations with childhood BLLs <50 |ig/L. Reuben et al. (2017) and Reuben et al. (2020) examined the
association of childhood blood Pb (11 years) with cognitive performance during mid-adulthood and
cognitive decline (change in IQ score between childhood and mid-adulthood) for the overall sample as
well as separately for participants above or below the historic level of concern (i.e., >10 (.ig/dL). Overall,
these childhood exposure studies suggested persistence and continued cognitive effects of childhood Pb
exposure through mid-adulthood, and the strength of associations were higher in magnitude for the
participants with childhood exposure above the historic level of concern (>10 (.ig/dL).

The limited recent toxicological evidence generally supports the dose-dependent effects of Pb on
cognitive function at relevant BLLs in adult animals. Only one study examined juvenile animals exposed
to multiple concentrations of Pb and reported greater decrements in learning and memory at higher doses
(An et al.. 2014).

3.6.1.3.2 Potentially At-Risk Populations

Age and Sex:

In the 2013 Pb ISA, an analysis using the NAS cohort reported an interaction between Pb and age
(Wright et al.. 2003). The study reported that the inverse association between age and cognitive function
was greater among those with high blood or patella Pb levels. Specifically, in the highest quartile of
patella Pb, each year increase in age led to a four-fold steeper decline in the MMSE score relative to the
effect of age in the lowest quartile of patella Pb. Effect estimates were in the same direction for tibia Pb,
but the interaction was not statistically significant.

Two recent epidemiologic studies using NHANES data from the 1999-2002 and 2011-2014
cycles explored the cross-sectional associations of blood and urine biomarkers of multiple metals and

3-185

-------
metalloids (separately and jointly) with cognitive function and provided some insights into potential
effect modifications of Pb-associated decrements in cognitive function (Sasaki and Carpenter. 2022;
Przvbvla et al.. 2017). Stratified analysis by sex and/or age groups (above and below median age groups)
performed in these studies suggested a greater magnitude of Pb-cognitive function associations (beta
estimates >10%) for females and for older age categories, however, the statistical test for the interactions
suggested no difference between the sex and age categories.

Toxicological studies investigating potential sex differences in Pb-induced cognitive impairment
are limited. A study by Mansouri et al. (2013) reported that Pb produced similar decrements in learning
and memory in both male and female animals. Given the lack of toxicological evidence available, the
possible influence of sex on Pb-induced cognitive impairment in adult animals remains unclear.

Pre-existing conditions:

One study evaluated the association of bone Pb (mean ranges for various age groups: tibia Pb:
4.4-9.2 (ig/g; patella Pb: 5.9-15.2 |ig/g) with cognitive function among individuals with PD and controls
participating in a case-control study (Weuve et al.. 2013). The patella Pb and tibia Pb concentrations
reported in this study for all study participants increased with increasing age. The highest Pb
concentrations were found in study participants in the 75-81 years old category, and the lowest
concentrations were found in participants in the 54-65 years old category. When the data were analyzed
separately for participants with PD, higher tibia Pb concentration was significantly associated with lower
scores on all of the telephone cognitive tests (adjusted difference in scores per 10 (ig/g increase in bone
Pb: Telephone Interview for Cognitive status (TICS) test: -0.20 [-0.4, -0.00]; digit span forward: -0.23
[-0.43, -0.03]; digit span backward: -0.19 [-0.37, -0.00]) and global cognitive score (adjusted
difference in scores per 10 (ig/g increase in bone Pb: -0.13 [-0.25, -0.01]). When the overall (cases and
control) data were analyzed, significant interactions were observed for the association between tibia Pb
and global cognition by case-control status. Participants with PD showed worse scores compared with
controls (1 SD increase in tibia Pb led to worsening of the global cognitive score by 0.12 units [95% CI:
-0.22, -0.01] among cases; controls: 0.06 [95% CI: -0.09, 0.20]).

Genetics:

Studies investigating the association between Pb levels and cognitive function in 2013 Pb ISA
extensively evaluated the effect modification by ALAD and HFE gene variants. The evidence was
provided by an NHANES analysis (Krieg et al.. 2009) as well as multiple analyses from the NAS cohort
examining different tests of cognitive function (Raian et al.. 2008; Weuve et al.. 2006). In the study using
a cohort from NHANES III, associations with concurrent BLLs were more pronounced in groups with CC
and CG ALAD genotypes (i.e., ALAD2 carriers) for several indices of cognitive function (Krieg et al..
2009). In the NAS cohort of men, Weuve et al. (2006) found that higher concurrent BLL but not bone Pb
level was associated with a larger decrease in a test of general cognitive function among ALAD2 carriers.

3-186

-------
Another NAS study examined the function of specific cognitive domains (e.g., vocabulary, memory,
visuospatial skills) and found variable evidence for effect modification by ALAD genotype across tests
(Raian et al.. 2008). For example, among ALAD2 carriers, concurrent BLL was associated with a more
pronounced decrease in vocabulary score but less pronounced decrease in a memory index and no
difference in the associations with other cognitive tests. For tibia and patella Pb levels, ALAD genotype
was found to modify associations with different tests, for example, executive function and perceptual
speed. It is not clear why the direction of effect modification would vary among different cognitive
domains. The limited number of populations examined, and the different cognitive tests performed in
each study, make it difficult to conclusively summarize findings for effect modification by ALAD
variants. However, in the limited available body of evidence, blood and bone Pb levels were generally
associated with lower cognitive function in ALAD2 carriers.

Longitudinal analysis of the NAS cohort also indicated that HFE gene variants modified the
blood Pb-cognition association (Wang et al.. 2007). Wang et al. (2007) found an IQR higher tibia Pb level
(15 (ig/g) was associated with a 0.22 point steeper annual decline (95% CI: -0.39, -0.05) in the MMSE,
which assesses cognitive impairment in a number of domains, among the men with at least one HFE
variant allele (H63D or C282Y variant). The association was found to be nonlinear, with larger Pb-
associated declines observed at higher tibia Pb levels. Tibia Pb level was not associated with a decline in
MMSE score in men with the HFE wildtype genotype. Moreover, the deleterious association between
tibia Pb and cognitive decline appeared progressively worse in participants with increasingly more copies
of HFE variant alleles (p-trend = 0.008). These findings suggest that HFE polymorphisms greatly enhance
susceptibility to Pb-related cognitive impairment in a pattern consistent with allelic dose.

None of the studies in the current review examined the effect modification by genetic variants.
Other Metal Exposure:

Various studies that examined other metals either evaluated the relationship of each metal
separately with the outcomes of interest or included multiple metals jointly in the model (Sasaki and
Carpenter. 2022; Xiao et al.. 2021; Przvbvla et al.. 2017). In a study of adults 50 to 82 years old in Sao
Paulo, Brazil, Souza-Talarico et al. (2017) examined the associations of heavy metals (Cd and Pb) in
blood and WMC separately as well as together in a model for metal interactions. The study found no
significant association between blood Pb and WMC in the model including Pb only, but significant
interactions were observed between blood Cd and blood Pb and the inverse association with WMC (|3 =
-0.38, p< 0.001).

3.6.1.3.3 Lifestages

The identification of critical lifestages and time periods of Pb exposure is complicated by the fact
that the majority of adult cognitive studies used concurrent adult BLLs. Although possibly affected by

3-187

-------
recent exposure, BLLs are also influenced by Pb stored in bone. Thus, associations in adult studies using
concurrent BLL may reflect the effects of past and recent Pb exposures on cognitive outcomes. Some
cohort studies in the 2013 Pb ISA and the current review using bone Pb suggested the effects of
cumulative long-term Pb exposure on cognitive impairment during adulthood. However, it is still difficult
to specifically identify exposures at particular lifestages (prenatal, infancy, early and late childhood, early
adulthood, etc.) that could have led to the long-term cognitive impairment observed in these studies. Few
recent prospective studies evaluating early childhood exposures at 7-12 years of age and long-term
cognitive impairment and decline at young- (18-19 years) (Skerfving et al.. 2015) and mid-adulthood (38
or 45 years of age) (Reuben et al.. 2020; Reuben et al.. 2017) provided an insight into critical lifestages
(i.e., early childhood and persistence of the cognitive effects through adulthood). Skerfving et al. (2015)
examined the association of childhood BLL in children from southern Sweden (age 7-12 years old, mean
blood Pb: 3.4 (ig/dL) with cognitive performance (IQ) at the age of 18-19 years. They found an IQ loss of
0.127 (-0.209, -0.045) points per (ig/dL increase in childhood BLL for all participants and an IQ loss of
0.204 (-0.392, -0.016) points per (ig/dL increase in childhood BLL among those with childhood BLLs
<50 |ig/L even in multivariable models adjusted for parent's income, education, and father's IQ. Reuben
et al. (2017) and Reuben et al. (2020) followed a New Zealand birth cohort to examine the association of
childhood Pb level (age 11 years, mean blood Pb: 10.99 (ig/dL) with cognitive performance and decline at
38 years (Reuben et al.. 2017) and 45 years (Reuben et al.. 2020). Reuben et al. (2017) found that each 5-
(ig/dL higher level of blood Pb in childhood was associated with a 1.97-point decrease in IQ score (95%
CI: -3.34, -0.59) and a 1.61-point decline (95% CI: -2.48, -0.74) in adult FSIQ after adjusting for sex,
childhood IQ, maternal IQ, and childhood SES. With additional years of data, Reuben et al. (2020) also
showed a significant association between childhood BLL and IQ at 45 years of age. Each 5-(ig/dL higher
level of blood Pb in childhood was associated with a 2.07-point decrease in the full-IQ score (95% CI:
-3.39, -0.74), and a 1.97-point decline (95% CI: -2.92, -1.03) after adjusting for covariates. These study
findings suggest that Pb exposure during childhood lifestages can influence cognition in adulthood.

Overall, recent rodent studies of learning and memory evaluating Pb exposure at various
lifestages suggest cognitive impairment. However, the magnitude of the effect at different lifestages has
been shown to differ. Adult animals may be less sensitive than juvenile animals, and juvenile animals
may be less sensitive than animals exposed during development (reviewed in Section 3.5.1). This general
pattern is consistent with evidence describing critical windows for brain development (Section 3.3).
Additionally, critical evidence for the association of Pb with cognitive impairment across lifestages comes
from a series of studies describing the effects of lifetime Pb exposure on nonhuman primates (Rice. 1992;
Rice and Gilbert. 1990a; Rice. 1990; Rice and Karpinski. 1988). Cynomolgus monkeys (Macaco
fascicularis) were dosed continuously from birth and tested repeatedly throughout their lifetime. While
these exposures yielded BLLs beyond values considered relevant for the current assessment (>30 (.ig/dL).
they provide key evidence of Pb-induced cognitive impairments that persisted into adulthood in a
translationally relevant species. However, given the limited number of studies conducted in juvenile and
adult animals, and the lack of studies examining the same endpoint across multiple age groups, the precise
role of exposures at various stages on the cognitive effects in adult animals remains unclear.

3-188

-------
3.6.1.4

Summary and Causality Determination: Cognitive Function in Adults

The 2013 Pb ISA (U.S. EPA, 2013) concluded that the available evidence was sufficient to
conclude "a causal relationship is likely to exist" between long-term cumulative Pb exposure and
cognitive function decrements in adults. This causality determination was based on a small body of
prospective studies that indicated strong associations of higher baseline tibia (means 19, 20 (ig/g) or
patella (mean 25 (ig/g) Pb levels with declines in cognitive function in adults (age >50 years) over 2- to 4-
year periods among adults without occupational exposure (i.e., NAS and BMS cohorts). Supporting
evidence was provided by analyses of the NAS, BMS, and NHS cohorts, that found stronger associations
with cognitive impairment for cumulative exposure (i.e., bone Pb level) than for concurrent BLL, which
reflects recent Pb exposure and Pb that has been mobilized from the bone. The timing, frequency,
duration, and magnitude of Pb exposures that contributed to the associations observed with blood Pb
levels were not discernable from cross-sectional associations reported in these studies. The biological
plausibility for the effects of Pb exposure on cognitive function decrements in adults was provided by
findings that relevant lifetime Pb exposures from gestation, birth, or after weaning induce learning
impairments in adult animals and by evidence for the effects of Pb altering neurotransmitter function in
the hippocampus, prefrontal cortex, and nucleus accumbens.

Results from recent epidemiologic and animal studies add to the evidence base reviewed in the
2013 Pb ISA. Recent epidemiologic studies consistently report that higher cumulative Pb exposure (i.e.,
bone Pb levels) or childhood BLLs, were associated with poor cognitive performance or decrements in
cognitive function during young-, mid-, or older-adulthood periods (Table 3-14E). Across populations,
higher Pb levels were associated with decrements in FSIQ, global cognitive function, executive function,
visuospatial and visuomotor skills, language, and memory. Discordant Pb associations across domains of
cognitive function are likely to reflect inherent biologic variability or differences in the outcome
pathophysiology as opposed to inconsistency in the evidence. Much of this evidence on adult cognitive
outcomes was obtained from analyses of the NAS and NHS cohorts, including recent analyses that
extended follow-up periods beyond the analyses evaluated in the 2013 Pb ISA. Recent evidence also
comes from early childhood exposure cohort studies conducted in Sweden and New Zealand. These
studies strengthen findings that childhood Pb exposures are associated with decrements in IQ and
cognitive function during young- and mid-adulthood. Longitudinal study designs with longer follow-up
periods, multiple and repeatedly measured cognitive outcomes, and multiple risk factors and confounders
accounted for in the studies reduce the bias and strengthen the study findings related to Pb exposure and
adult cognitive function. Further, significant findings from new studies specifically investigating the
influence of early childhood Pb exposure on adult IQ and cognitive outcomes, even after adjustments for
various confounders including childhood IQ, provide evidence for the role of early childhood Pb
exposures on decrements in cognitive function in adulthood.

Strong evidence for cognitive function declines associated with cumulative Pb exposures was
provided by prospective cohort studies that demonstrated increased bone Pb levels (tibia mean: 10.5, 21.6

3-189

-------
(ig/g, patella mean: 12.6, 30.6 |ig/g) measured at baseline were associated with cognitive decline over the
follow-up period of 13-15 years (Farooqui et al.. 2017; Power et al.. 2014). Findings from these studies
suggest that long-term Pb exposure may contribute to ongoing declines in cognitive function in adults.
These associations remained significant even after adjustment for potential confounding by combinations
of factors including demographic, socioeconomic, behavioral, clinical, and neighborhood level factors.
Increased bone Pb level (tibia mean range: 4.4-9.2 (ig/g) was associated with cognitive function outcomes
among cases of PD (Wcuvc et al.. 2013). Additional support for the effects of cumulative or past Pb
exposure is provided by an analysis of past blood Pb exposures during childhood (either low or high Pb
exposure scenarios; blood Pb mean: 3.4 (ig/dL at 7-12 years, 10.99 (ig/dL at 11 years) and studies that
followed study participants through young or mid-adulthood (Reuben et al.. 2020; Reuben et al.. 2017;
SkerfVing et al.. 2015). These studies indicated that higher childhood BLL was associated with declines in
IQ at 18 to 19 years old and at 38 years or 45 years old.

Findings from cross-sectional studies that assessed the relationships of concurrent blood (2.1
(ig/dL to 5.1 (ig/dL) and cognitive function outcomes were more mixed. Concurrent blood Pb level does
not clearly indicate recent Pb exposure in adults because Pb is mobilized from the bone in various adult
lifestages complicating the interpretation of these studies. Two studies including NHANES data found
inverse association between concurrent BLLs and cognitive outcomes (Sasaki and Carpenter. 2022;
Przvbvla et al.. 2017). while others suggested null associations (Xiao et al.. 2021; Souza-Talarico et al..
2017; Khalil et al.. 2014; van Wijngaardcn et al.. 2011). The NHANES studies demonstrating significant
associations considered multiple metals in their analytical models including Pb, used advanced model
approaches to handle multiple exposures and issues around multiple comparison and multi-collinearity,
and adjusted for sociodemographic, behavioral, and clinical characteristics. These approaches reduced the
bias and uncertainty in the study findings. A recent study using a prospective design addressed some of
the concern around the health effects of recent Pb exposure (Yu et al.. 2021). The study included a group
of individuals with no prior occupational exposure and recently hired young workers at battery
manufacturing and Pb recycling plants. The association between neurocognitive performance and blood
Pb was examined prior to and up to 2 years after the first occupational exposure (geometric mean
baseline: 3.97 (ig/dL; 13.4 (ig/dL, and 12.8 (ig/dL at the first and second follow-up visits). The study did
not observe significant associations of changes in neurocognitive function in the workers with an over
three-fold increase in blood Pb concentration over the 2-year follow-up period, though the follow-up time
in this study may not have been adequate to detect the long-term effects of Pb on cognitive function.

Sex and age differences in bone kinetics and turnover could have contributed to differences in the
magnitude of associations observed for different bone Pb biomarkers and cognitive function in the NHS
and NAS cohorts. Specifically, the role of specific bone biomarkers (i.e., tibia or patella Pb) as indicators
of Pb exposure for specific age and sex groups in relation to individual cognitive domains is yet to be
fully understood. For instance, the findings from the NHS cohort with shorter follow-up, Weuve et al.
(2009). suggested that higher tibia Pb in women was inversely associated with the overall cognitive score.
The associations with the majority of the domain-specific cognitive scores were also negative (except for

3-190

-------
letter fluency, which was positive) but the estimates were imprecise. The recent extended analysis of the
NHS cohort Power et al. (2014), on the other hand, suggested that higher tibia Pb in women was inversely
associated with individual cognitive scores representing executive function and memory domains, and
was unexpectedly positively associated with immediate verbal memory domain. Associations with other
cognitive domains or the overall cognitive score were imprecise. Similarly, analyses from the NAS men
cohort with shorter versus extended follow-up periods indicated heterogeneous associations with global
and domain scores as well. Results from the analysis with shorter follow-up (Weisskopf et al., 2007)
suggested stronger inverse association between higher patella Pb and declines in the visuospatial and
visuomotor domains over time, but weaker and imprecise associations were observed for other domains.
In contrast, results from the extended analysis Farooqui et al. (2017) observed that higher patella Pb was
associated with faster longitudinal decline in MMSE (a measure of global cognition) and declines in the
language and memory domains, whereas associations with other cognitive scores or domains were
imprecise.

Recent studies of rodents with exposure resulting in mean BLLs <30 (ig/dL add to the evidence
informing the association of both short- and long-term Pb exposure during adulthood with measures of
learning and memory in rodents. While these studies are consistent with one another, toxicological
evidence for effects of Pb on cognitive function in adults remains limited., A few recent studies in
juvenile rodents also provide some support for the association between postnatal Pb exposure either
during adolescence or continuing into adulthood and cognitive impairment, specifically learning and
memory. Previous studies in nonhuman primates demonstrated that early life exposure to Pb may produce
cognitive impairment in adulthood. Hence, these findings add to the current evidence base suggesting
potential roles of both early and later life Pb exposures to produce cognitive function decrements in
adults. Additionally, animal, and in vitro studies lend biological plausibility to the association between
adult Pb exposure and adult cognitive impairment, showing that Pb has negative effects on neuronal
function and integrity, neurotransmission, and synaptic plasticity in regions of the brain associated with
learning and memory (Section 3.6).

Overall, the collective evidence is sufficient to conclude that there is a causal relationship
between Pb exposure and cognitive effects in adults. Recent prospective epidemiologic studies expand
and strengthen the previous body of evidence. These recent prospective studies include extended analyses
with longer follow-up periods, repeated measurements of cognitive outcomes, and adjustment for an array
of important potential confounders. Together, they provide compelling evidence for an association
between Pb exposure during various lifestages including childhood and decreased cognitive function in
adulthood. Discordant Pb associations across domains of cognitive function likely reflect differences in
the outcome pathophysiology rather than inconsistent results, and they do not detract from the strength of
the evidence overall. Recent evidence from animal studies supports the biological plausibility for the
effects of Pb exposure on cognitive function in adults and demonstrates that postnatal exposure to Pb
(either during adolescence or continuing into adulthood) may also negatively affect learning and memory.

3-191

-------
Table 3-9 Summary of evidence for a causal relationship between Pb exposure and cognitive effects in
adults

Rationale for Causality Determination3

Key Evidence13

References'3

Pb Biomarker
Levels Associated
with Effects0

Consistent findings from prospective epidemiologic
studies with relevant adult bone Pb levels or early
childhood BLLs suggesting significant association
of long-term and early childhood Pb exposure with
cognitive impairment and decline

Prospective analyses in NAS cohort of white men and NHS
cohort of white women found cognitive function decrements
over the 13 to 15 yr follow-up in association with patella or
tibia Pb levels.

Power et al. (2014)

Farooaui et al. (2017)

Mean patella Pb:
12.6 |jg/g

Mean tibia Pb: 10.5

pg/g

Mean patella Pb:
30.6 |jg/g

Mean tibia Pb:

21.6 |jg/g

Prospective analyses of childhood Pb exposure and long-term
cognitive impairments suggested persistent effects on
cognition during adulthood by showing lower IQ at young- and
mid-adulthood periods, and significant decline in IQ between
childhood and adulthood periods due to exposure to higher
childhood Pb levels.

Skerfvinq et al.
(2015)

Reuben et al. (2017)
Reuben et al. (2020)

Mean childhood (7-
12 yr) blood Pb: 3.4
pg/dL

Mean childhood (11
yr) blood Pb: 10.99
pg/dL

Models adjusted for various confounding factors including
baseline individual-level, socioeconomic, demographics,
behavioral, and clinical factors, as well as various
neighborhood level variables. The early childhood Pb
exposure studies also adjusted for parental education, HOME
scores, parent IQ and childhood IQ.

Supporting epidemiologic studies

Analysis of bone Pb and cognitive function among cases and
controls of PD found lower cognitive performance score with
increased tibia Pb among cases.

Weuve et al. (2013)

Mean tibia Pb: 4.4-
9.2 |jg/g (for age
groups)

Mean patella Pb:
5.9-15.2 |jg/g (for
age groups)

3-192

-------
Rationale for Causality Determination3

Key Evidence13

References'3

Pb Biomarker
Levels Associated
with Effects0



Cross-sectional analysis of concurrent blood Pb (along with
multi-metals) and cognitive function using NHANES data and
advanced modeling approach suggested significant inverse
association between concurrent BLL and cognitive outcomes.

Przvbvla et al. (2017)

Sasaki and
Carpenter (2022)

Geometric mean
blood Pb: 2.17 pg/dL

Mean blood Pb: 1.9
pg/dL



Models adjusted for socioeconomic and demographic factors,
education, health status, comorbidities, and co-exposure to
other metals.





Consistent evidence in animals with relevant
exposures

Recent evidence from animal studies supports the notion that
postnatal exposure to Pb (either during adolescence or
continuing into adulthood) negatively affects learning and
memory in rodents.

Mansouri et al.

(2012)

Mansouri et al.

(2013)

Sinqh et al. (2019)
Su et al. (2016)

Mean BLL: 8 pg/dL

Peak BLL: 11-19
pg/dL

Peak BLL: 28 pg/dL
Mean BLL: 8.4 pg/dL

Some uncertainty remains

Sex and age differences in bone kinetics and turnover may
contribute differences in biomarker Pb levels.





BLL = blood lead level; HOME = Health Outcomes and Measures of the Environment; IQ = intelligence quotient; NAS = Normative Aging Study; NHANES = National Health and
Nutrition Examination Survey; NHS = Nurses' Health Study; Pb = lead; PD = Parkinson's disease, yr = year.

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 20151.
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or
inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-193

-------
3.6.2

Psychopathological Effects in Adults

The evidence assessed in the 2013 Pb ISA was sufficient to conclude that "a causal relationship is
likely to exist" between Pb exposure and psychopathological effects in adults. Cross-sectional studies in a
small number of distinct U.S. populations demonstrated associations of higher concurrent blood or tibia
Pb levels with self-reported symptoms of depression and anxiety in adults (Bouchard et al.. 2009; Raj an
et al., 2007; U.S. EPA, 2006). The examination of multiple exposures and outcomes in the available
studies does not provide a strong indication of biased reporting of psychopathological effects specifically
by adults with higher Pb exposures. In adults, Pb-associated increases in depression and anxiety were
found with adjustments for age, SES, and in the NAS, daily alcohol intake. The biological plausibility for
epidemiologic evidence was provided by observations of depression-like behavior in animals with dietary
lactational Pb exposure, with some evidence at relevant BLLs. In addition, Pb-induced changes in the
HPA axis and dopaminergic and GABAergic systems were demonstrated in animals. Overall, the
strongest evidence was from epidemiologic studies of adults without occupational Pb exposure, with
additional support from a few experimental animal studies; however, uncertainties related to residual
confounding of bone Pb associations by age in epidemiologic studies remained. Overall, recent studies
add to the evidence and generally support the findings from the 2013 Pb ISA.

3.6.2.1 Epidemiologic Studies of Psychopathological Effects in Adults

A limited number of epidemiologic studies evaluated in the 2006 Pb AQCD (U.S. EPA, 2006)
and 2013 Pb ISA (U.S. EPA, 2013) examined the relationship between blood or bone Pb levels and
psychopathological effects in adults. All of these studies were cross-sectional and most examined
occupationally exposed populations. In addition to the occupational studies, which provided consistent
evidence of positive associations between BLLs (mean levels >15 (ig/dL) and the prevalence of self-
reported symptoms of depression, anxiety, and tension, a prospective analysis of the NAS cohort (Raj an
et al., 2007) and a cross-sectional analysis of NHANES participants (Bouchard et al., 2009) reported
positive associations between much lower concentrations of concurrent blood (NHANES; mean ~ 6
(ig/dL) or bone (NAS) Pb levels and symptoms of depression and anxiety or prevalent major depressive
disorder, respectively. The findings from the NAS cohort are notable because tibia and patella Pb reflect
cumulative exposures that occur over several years to decades and thus serve as a retrospective
assessment of Pb exposure despite the cross-sectional study design.

Recent evidence includes several prospective cohort studies and cross-sectional analyses of Pb
exposure and general psychopathological effects or internalizing symptoms, as well as a case-control
study of schizophrenia. In general, recent prospective studies provide evidence of an association between
blood or bone Pb levels and psychopathological effects in adults. Results from cross-sectional studies are
inconsistent. Additionally, with blood or bone Pb levels, it is difficult to characterize the specific timing,

3-194

-------
duration, frequency, and level of Pb exposure that contributed to associations observed with cognitive
function. This uncertainty may apply particularly to assessments of BLLs, which in nonoccupationally
exposed adults reflect both current exposures and cumulative Pb stores in bone that are mobilized during
bone remodeling. Measures of central tendency for blood and bone Pb levels used in each study, along
with other study-specific details, including study population characteristics and select effect estimates, are
highlighted in Table 3-15E. An overview of the recent evidence is provided below.

A few recent prospective cohort studies report generally consistent evidence of associations
between blood or bone Pb levels and psychopathological effects in adults, although the results for specific
endpoints are not entirely consistent. In a recent prospective cohort study examining a subset of the NAS
cohort, Peters et al. (2011) used structural equation models to examine the interrelations of childhood and
adult SES, bone Pb levels, pessimism, and depression in older adults. After controlling for childhood and
adult SES through a combination of latent variables—including parental and participant education,
occupation, and home ownership—10 (ig/g higher tibia Pb levels measured prior to psychological
measurement were independently associated with a 0.3-unit (95% CI: 0.0, 0.6) higher score for pessimism
level on the Life Orientation Test. An independent association between bone Pb levels and depression
was not observed when controlling for pessimism (quantitative results not reported), but pessimism was
strongly associated with increased odds of depression (OR = 1.04 [95% CI: 1.02, 1.05] per 1 unit higher
pessimism level), indicating a potential mediating effect of pessimism on the relationship between bone
Pb levels and depression. Other recent prospective cohort studies examined the relationship between
childhood BLLs and a wider range of psychopathological effects (Reuben et al.. 2019; McFarlane et al..
2013). In an analysis of the Dunedin cohort in New Zealand, Reuben et al. (2019) reported that each 1-
(ig/dL higher BLL at age 11 was associated with 0.27-point (95% CI: 0.02, 0.51) higher standardized
general psychopathology scores (mean [SD]: 100 [15]) in early adulthood. In more specific analyses of
psychopathology components, the association with general psychopathology appeared to be driven by
positive associations with symptoms of internalizing behavior and thought disorder. These results are
somewhat consistent with a follow-up analysis of the Port Pirie cohort, a birth cohort from a South
Australian Pb-smelting town (McFarlane et al.. 2013). In this study, BLLs averaged over the first 7 years
of life were not associated with depressive symptoms measured during follow-up at ages 25 through 29.
However, the authors did report positive associations between BLLs and some internalizing behaviors in
women (e.g., social phobia, specific phobia, and anxiety problems). There were generally null
associations for the same outcomes in men. A notable limitation of this study is that attrition in the
original cohort led to a small sample size that was even further reduced in sex-stratified models, resulting
in limited power to detect an association. Additionally, due to high community exposure to Pb, the
participants in the Reuben et al. (2019) and McFarlane et al. (2013) studies had a high mean childhood
BLL (11.08 and 17.2 (ig/dL, respectively). The mean BLLs in these studies are not directly comparable to
other studies in this section, which used concurrent BLLs in adult populations that likely had higher past
exposures.

3-195

-------
Associations between BLLs and mood states were inconsistently observed in cross-sectional
studies of pregnant women in China (Li et al.. 2017) and Japan (Ishitsuka et al.. 2020). In a small study of
pregnant women in Shanghai, Li et al. (2017) observed nonlinear associations between BLLs and
depression, anxiety, and psychological distress in nonparametric models. Based on visual inspection of
the spline curves, the authors ran separate piecewise linear regression models for each outcome with a
knot at 2.57 (ig/dL. In the piecewise models, BLLs were associated with higher prevalence of depression,
anxiety, and psychological stress at levels below 2.57 (ig/dL, whereas smaller inverse associations were
observed above the knot. In contrast, a much larger cross-sectional study of pregnant women across Japan
noted that BLLs measured during middle or late pregnancy were not associated with increased odds of
Kessler Psychological Distress Scale (K6) scores greater than or equal to 5 or 13 (Ishitsuka et al.. 2020).
The authors used different cut points to account for potential differences in the optimal sensitivity-
specificity tradeoff for assessing depression in their study population. The contrasting results in these
studies are not readily explained by variations in BLLs, as Li et al. (2017) observed positive associations
at low BLLs in linear spline models, and the population analyzed by Ishitsuka et al. (2020) had a
geometric mean BLL <1 (ig/dL.

Other recent cross-sectional studies assessed the relationship between blood or bone Pb levels and
mood states measured by validated questionnaires or self-reported physician's diagnosis in a range of
study populations, including a population-based analyses of NHANES (Berk et al.. 2014) and KNHANES
participants (Nguyen et al.. 2022). older women participating in subcohorts of the NHS (Eum et al..
2012). and older adults from two communities selected using cluster-based sampling of communities in
Luan, China (Fan et al.. 2020). Similar to previously discussed studies, results from these analyses were
inconsistent. Fan et al. (2020) reported monotonic increases in odds of depression associated with
increasing blood Pb exposure quartiles. In this study, older adults in the highest quartile of exposure
(BLLs <3.06 (ig/dL) had just over twice the odds of depression as those with BLLs <2.03 (ig/dL (OR =
2.03 [95% CI: 1.23, 3.35]). In contrast, Eum et al. (2012). Berk et al. (2014). and Nguyen et al. (2022)
reported null associations between bone or blood Pb levels and anxiety or depression. However, in a
subgroup analysis restricting the study population to pre- and postmenopausal women taking hormone
replacement therapy (HRT), Eum et al. (2012) reported that increasing tibia Pb tertiles were associated
with monotonically increasing odds of phobic anxiety and lower scores on the Mental Health Index 5-
item (MHI-5; indicating worse depressive symptoms). The authors restricted the analysis by HRT status
to account for potential exposure measurement error in the non-HRT population, resulting from higher
variability in bone turnover. Notably, patella Pb was not associated with depressive symptoms in this
population and was inversely associated with phobic anxiety. Since tibia Pb has a longer half-life than
patella Pb, the results could be indicative of a long-term exposure window contributing to changes in
anxiety and depression. However, the subgroup analyses also had small sample sizes and thus a lack of
precision and higher probability of chance findings.

While most recent studies of psychopathological effects in adults examined internalizing
behaviors or general psychopathological effects, a small case-control study in China evaluated serum

3-196

-------
heavy metal levels in association with the risk of schizophrenia (Ma et al., 2019). The authors reported
that higher serum Pb levels were associated with higher odds of schizophrenia, but the association was
imprecise (OR = 3.15 [95% CI: 1.24, 7.99] per ng/mL higher BLL). Although the cases and controls were
matched on age and sex, it is unclear what, if any, other potential confounders were included in the
adjusted models. This finding was ostensibly consistent with a pooled analysis of two small cohorts
evaluated in the 2013 Pb ISA that observed an association between higher S-ALAD levels and increased
odds of schizophrenia spectrum disorder in adolescents and adults (Opler et al., 2008). However, this
pooled analysis had a number of limitations that precluded any conclusions regarding a relationship
between Pb exposure and schizophrenia, including the lack of direct measurements of Pb biomarker
levels and limited consideration for potential confounding. As noted previously, Reuben et al. (2019)
reported a positive association between childhood BLLs and symptoms of thought disorder in early
adulthood. However, the metric for thought disorders included factor loadings for obsessive compulsive
disorder and mania in addition to schizophrenia, making it difficult to distinguish an independent
relationship between BLLs and schizophrenia.

3.6.2.1.1 Summary

A limited number of cross-sectional studies evaluated in the 2013 Pb ISA (U.S. EPA, 2013)
provided consistent evidence of positive associations between blood and bone Pb levels and the
prevalence of self-reported symptoms of depression, anxiety, and tension. Recent prospective analyses
provide additional support for a positive association between bone and BLLs and psychopathological
effects in older adults, although results from cross-sectional studies are inconsistent.

3.6.2.2 Toxicological Studies of Psychopathological Effects in Adults

No studies in the 2013 Pb ISA evaluated the effect of adult-only exposure on anxiety and
depression-like behaviors in animals. Nevertheless, developmental studies consistently supported an
effect of Pb exposure on these endpoints in adult animals (U.S. EPA, 2013). Of particular importance, the
increased reactivity to errors and reward omission reported by Beaudin et al. (2007) and Stangle et al.
(2007) extended to adulthood, well after postnatal exposure was terminated. Furthermore, Pb exposure
during adolescence reduced immobility on the FST in adult rats (Stewart et al., 1996).

A few recent studies have investigated adult-only exposure and anxiety-like outcomes using the
OFT and EPM. Similar to developmental exposures, adult male mice displayed anxiety-like behavior (i.e.,
increased time spent in closed arms) in the EPM following 6 weeks of Pb exposure given via oral gavage
(mean BLLs 7.1 (ig/dL) (Al-Qahtani et al., 2022). Singh et al. (2019) also found that oral gavage of Pb for
90 days decreased the amount of time rodents spent in the open arms of the EPM compared with control
animals. Another study of long-term adult exposure (126 days) to Pb in Wistar rats found increased
rearing and grooming, but not sniffing, in males, and no significant effects in females in the OFT

3-197

-------
(Mansouri et al.. 2013). The same research group found that Pb did not affect these endpoints (i.e.,
rearing, sniffing) following shorter-term exposure (30 days) in adult animals (Mansouri et al.. 2012).

Studies that employed developmental exposure paradigms are discussed in more detail in Section
3.5.4.2. However, they provide consistent evidence to support the notion that the effects of developmental
exposures may persist into adulthood. In one study of gestational exposure, there were significant
decreases in exploratory behaviors in the OFT and hole board test in Wistar rats at 4 months old (BLLs
peaked at 12 (ig/dL and decreased to 6 (ig/dL by 4 months) (Basha and Reddv. 2015). A similar study
utilizing postnatally exposed rats found that some measures of decreased exploratory behavior in the OFT
and hole board test persisted until 18 months, when the study was terminated (Basha et al.. 2014). The
mean BLL at PND 45 in this study was high (50 (ig/dL) but had decreased to 11 (ig/dL by 18 months.
Several additional studies investigated lifetime Pb exposure (gestation through 6-12 months), and all
found significant treatment effects on EPM or FST behavior during adulthood (Shvachiv et al.. 2020.
2018; Abazvan et al.. 2014; Corv-Slechta et al.. 2013).

Toxicological studies also provide biological plausibility to support a connection between
exposure to Pb and schizophrenia. As discussed in the 2013 Pb ISA, antagonists of NMDAR's glycine
site have been shown to exacerbate schizophrenia symptoms in affected individuals and induce a
schizophrenic phenotype in unaffected subjects (Covle and Tsai. 2004). Previous studies have shown that
Pb is a potent allosteric inhibitor at NMDARs (Hashcmzadch-Gargari and Guilarte. 1999; Guilarte. 1997).
A recent study found that developmental Pb exposure (BLLs of 22 (ig/dL at PND 50) reduced the number
of parvalbumin-positive GABAergic interneurons in the median prefrontal cortex and hippocampus and
induced subcortical dopaminergic hyperactivity, consistent with studies of schizophrenic patients
(Stansfield et al.. 2015; Volman et al.. 2011). Pb exposure may also affect DISC 1, a gene-protein pair
associated with increased susceptibility to schizophrenia and other mental disorders. You et al. (2012)
found that expression of the DISCI protein was increased in the hippocampus of rats exposed to Pb
during gestation and lactation. Abazvan et al. (2014) utilized a transgenic mouse model expressing mutant
DISCI (mDISCl) to evaluate a potential gene-environment interaction using lifetime Pb exposure. Pb-
exposed mDISC 1 mice exhibited behavioral and structural abnormalities consistent with schizophrenia,
which were not found in unexposed mDISC 1 mice or Pb exposure regular mice (i.e., heterozygous for
mDISCl but phenotypically normal).

3.6.2.2.1 Summary

Toxicologic studies providing support for adult psychopathological effects in the previous ISA
used developmental exposure paradigms. Recent developmental exposure studies were consistent with the
previous evidence and were predominantly focused on anxiety-like behaviors. Multiple studies
demonstrated the persistence of these effects into adulthood (up to 1.5 years), in some cases long after
termination of Pb exposure. A few recent studies focused on adult-only exposures found some
associations with anxiety-like behavior after 42-126 days of exposure but not following a 30-day

3-198

-------
exposure; additional studies are needed to strengthen this line of evidence. Two recent studies also
provided further biological plausibility support for an NMDAR-mediated association between Pb
exposure and schizophrenia.

3.6.2.3 Relevant Issues for Interpreting the Evidence Base

3.6.2.3.1 Concentration-Response Function

One recent study used a nonparametric model to assess the C-R relationship between BLLs and
depressive symptoms in pregnant women (Li et al.. 2017). The authors observed nonlinear associations
between BLLs and depression, anxiety, and psychological distress scores and used the nonparametric
models to determine knots for a piecewise linear regression model. The subsequent models indicated
positive associations between BLLs and depression, anxiety, and psychological stress at levels below 2.57
(ig/dL, whereas smaller inverse associations were observed for BLLs above 2.57 (ig/dL. Other cross-
sectional studies reported inconsistent evidence of associations despite evaluating populations with low
mean BLLs. However, in studies analyzing BLLs in adult populations with higher past exposures, it is a
challenge to ascertain the level, timing, frequency, and duration of Pb exposure that contributed to
observed associations.

3.6.2.3.2 Potentially At-Risk Populations

A few of the recent epidemiologic studies detailed in this section evaluated populations that are
potentially at-risk for Pb-related health effects. The conclusions that can be drawn from these analyses are
limited. A prospective analysis of young adults reported sex-specific associations between childhood
BLLs and internalizing symptoms in early adulthood (McFarlanc et al.. 2013). The observed associations,
which were only present in stratified models including women, were extremely imprecise due to a small
sample size that was even further reduced by stratification. The small sample size in this study reduced
the statistical power to detect an association and the likelihood that an observed result reflects a true
effect, making it difficult to draw firm conclusions on these sex-specific comparisons. Additionally, two
cross-sectional studies examining depressive symptoms in pregnant women observed inconsistent
evidence of an association with BLLs (Ishitsuka et al.. 2020; Li et al.. 2017).

3.6.2.3.2.1 Lifestages

Toxicological studies provide consistent evidence that developmental and lifetime exposures to
Pb can lead to increases in anxiety-like behaviors. Comparatively fewer studies have investigated the
effects of adult-only exposure, but some effects have been demonstrated. Epidemiologic studies provide

3-199

-------
some supporting evidence for the importance of developmental and cumulative exposures. In a recent
prospective cohort study, Peters et al. (2011) reported that increased depressive symptoms in older adults
were associated with tibia Pb levels, a measure of cumulative exposure. Other recent prospective analyses
observed associations between childhood BLLs and increased internalizing symptoms in young adults
(Reuben et al.. 2019; McFarlane et al.. 2013). Additionally, (Reuben et al.. 2019) also observed positive
associations between childhood BLLs and internalizing symptoms at the time of BLL testing, which is
coherent with toxicological evidence that suggests the persistence of developmental effects into
adulthood. Given the uncertainties regarding potentially higher historical exposures in adults, the cross-
sectional epidemiologic studies evaluated in this section are less suited to address the importance of
concurrent exposures.

3.6.2.3.3 Confounding

The studies evaluated in this section controlled for a range of potential confounding variables that
may be associated with both Pb exposure and psychopathological effects, including age, sex, SES factors,
and marital status (see Table 3-15E).

3.6.2.4 Summary and Causality Determination: Psychopathological Effects in Adults

The 2013 Pb ISA (U.S. EPA, 2013) concluded that the available evidence was sufficient to
conclude that "a causal relationship is likely to exist" between Pb exposure and psychopathological
effects in adults. This causality determination was based on a small body of epidemiologic evidence that
demonstrated consistent positive associations between concurrent blood or bone Pb levels and self-
reported symptoms of depression, anxiety, and panic disorder in large studies of adults (i.e., NHANES,
NAS). The epidemiologic evidence was supported by coherence in animal toxicological studies that
demonstrated depression-like behavior and emotionality in rodents exposed to dietary lactational Pb with
or without additional post-lactational exposure. Epidemiologic associations were observed in study
populations of young (20-39 years old) and older (44-98 years old) adults. Because of the cross-sectional
design of the epidemiologic studies, there was uncertainty regarding the temporal sequence between Pb
exposure and psychopathological symptoms in adults. This uncertainty is somewhat reduced with results
for tibia Pb since it is an indicator of cumulative Pb exposure. Nonetheless, because these studies included
adults with likely higher past Pb exposures, uncertainties exist regarding the Pb exposure level, timing,
frequency, and duration contributing to the associations observed with blood or bone Pb levels. An
uncertainty in the toxicological evidence base was the limited number of studies that administered
exposures resulting in BLLs that are not relevant to humans. Recent epidemiologic and toxicological
evidence continues to link Pb exposure to psychopathological effects in adults, though some uncertainties
still remain. The key evidence, as it relates to the causality determination, is presented in Table 3-10 and
Table 3-11 and summarized below.

3-200

-------
Recent evidence from prospective epidemiologic studies provides further support for positive
associations between Pb exposures and pathological effects, including increased internalizing symptoms.
Specifically, a study of older men in the NAS cohort provides evidence of an (indirect) association
between Pb exposure and depression in adults (Peters et al., 2011), and an analysis of a cohort in New
Zealand similarly reported that increased childhood BLLs were associated with increased internalizing
symptoms in young adults (Reuben et al., 2019). Together, these studies address an uncertainty from the
previous ISA regarding the temporality of the exposure and outcome. Another recent prospective study of
young adults from the Port Pirie cohort reported null associations between childhood BLLs and
depression in adults but noted several sex-specific associations between BLLs and other internalizing
symptoms in young women (McFarlane et al., 2013). The small analytic sample of the stratified models
used in this analysis reduces the likelihood of detecting a true effect. Notably, supporting evidence from
recent cross-sectional epidemiologic studies conducted in diverse populations is largely inconsistent.
However, these studies are less informative given the limitations of the study design.

The epidemiologic evidence is supported by coherence with results from an expanded number of
toxicological studies conducted at BLLs relevant to humans. Recent toxicological studies examine
multiple exposure windows and provide strong support for Pb-induced anxiety-like behaviors following
developmental and cumulative exposures. Multiple studies demonstrate the persistence of these effects
into adulthood (up to 1.5 years), in some cases long after termination of Pb exposure. The evidence for
effects resulting from adult-only exposures is more limited, though there is some evidence for an increase
in anxiety-like behavior following 42-126 days of exposure but not following a 30-day exposure. The
2013 Pb ISA also highlighted Pb-induced changes in the dopaminergic and GABAergic systems and the
HPA axis, which underlie biological plausibility for the changes in mood and emotional state that have
been observed in epidemiologic and toxicological studies (U.S. EPA, 2013). Recent studies continue to
demonstrate changes in corticosterone and glucocorticoid receptors (i.e., HPA axis changes (Cory-Slechta
et al., 2012)) and the dopaminergic system (Section 3.4.2.2).

In addition to studies of depression, anxiety, and mood-related disorders, some recent studies
examined the relationship between Pb exposure and schizophrenia. A recent case-control study in China
reported a positive, but imprecise association between serum Pb levels and schizophrenia prevalence in
adults (Ma et al„ 2019). Because serum Pb was measured after schizophrenia was diagnosed, the results
do not establish temporality between exposure and outcome. Additionally, the reported analytic
methodology does clarify the confounding variables considered outside of the age- and sex-based
matching of cases and controls. Recent toxicological studies provide biological plausibility to support a
connection between exposure to Pb and schizophrenia. Consistent with evidence from the 2013 Pb ISA,
two recent studies support Pb-induced pathophysiological features in rodents consistent with
schizophrenia, likely through inhibition of NMDAR activity. Although the toxicological evidence
presents a biologically plausible pathway through which exposure to Pb could lead to schizophrenia, the
limited quantity and quality of the epidemiologic evidence precludes meaningful consideration of this
endpoint in the causality determination for Pb exposure and psychopathological effects.

3-201

-------
Overall, the collective evidence is sufficient to conclude that there is likely to be a causal
relationship between Pb exposure and psychopathological effects in adults. The strongest evidence
comes from a limited number of recent prospective epidemiologic studies that add to previous evidence of
a positive association between bone or BLLs and psychopathological effects in older adults and addresses
prior uncertainties regarding the temporality of exposure and outcome. Recent toxicological studies
strengthen the overall evidence base, providing further support for anxiety-like behaviors following
developmental and cumulative exposures that result in BLLs that are relevant to humans. Despite
generally consistent evidence from prospective epidemiologic studies that Pb is associated with general
internalizing behavior scores or some components of internalizing behavior, the evidence from these
limited number of studies is not consistent for any single component. Although these inconsistencies may
reflect differences in outcome pathophysiology rather than inconsistent results, there is remaining
uncertainty given the limited body of evidence. The key evidence, as it relates to the causal framework, is
summarized in Table 3-10.

3-202

-------
Table 3-10 Summary of evidence for a likely to be causal relationship between Pb exposure and
psychopathological effects in adults

Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels
Associated with
Effects0

Limited epidemiologic evidence from Prospective analyses reported positive associations between tibia Pb levels
high-quality prospective cohort studies and depressive symptoms in older men, mediated by pessimism; and
with relevant bone Pb levels childhood BLLs and psychopathology scores in early adulthood.

Some supporting evidence from a prospective analysis reported imprecise
positive associations between childhood BLLs and prevalence of social
phobia, specific phobia, PTSD, anxiety problems, somatic problems, and
antisocial personality problems in young adult women.

Peters et al. Mean: 20.6 |jg/g
(2011)

Reuben et al.
(2019)

Mean: 11.08 pg/dL

McFarlane et Mean: 17.2 |jg/dL
al. (2013)

Consistent evidence in animals with
relevant exposures

Increased anxiety-like behaviors in adulthood (up to 18 mo) following
developmental or lifetime Pb exposure provides coherence with epidemiologic
evidence.

Section 3.5.4.2

Peak blood Pb after
lifetime exposure: 7-24
pg/dL

Peak blood Pb after
developmental exposure:
12-50 |jg/dL

Inconsistent supporting evidence from
cross-sectional epidemiologic studies
with relevant blood and bone Pb levels

A limited body of cross-sectional studies provides inconsistent evidence of
associations between generally lower blood and bone Pb levels and
depression and anxiety.

Section 3.6.2.1 Mean/median range
across studies:

Bone: 10.3-12.5 |jg/dL
Blood: 0.58-3.97 pg/dL

Uncertainty regarding potential
confounding

Most studies included adjustment for age, sex, SES factors, and marital status,
Table 3-15E but did not consider use of antidepressants

BLL = blood lead level; mo = months; Pb = lead; PTSD = post-traumatic stress disorder; SES = socioeconomic status.

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 2015).
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or
inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-203

-------
3.6.3

Sensory Organ Function in Adults

The 2013 Pb ISA included separate causality conclusions for auditory and visual function. This
ISA combines these categories and makes one causality determination for Sensory Organ Function (see
Section 3.6.3.5). Recent studies are summarized in Table 3-16E and Table 3-16T. An overview of the
recent evidence is provided below.

3.6.3.1 Auditory Function

The evidence assessed in the 2013 Pb ISA was "suggestive of a causal relationship" between Pb
exposure and auditory function decrements in adults (U.S. EPA, 2013). The strongest evidence was
provided by the analysis of NAS men, which revealed associations between higher tibia Pb levels and a
higher rate of elevations in hearing threshold over 20 years (Park et al., 2010). Findings demonstrating
decreased auditory evoked potentials in animals provided biological plausibility for the observations in
this epidemiologic study, but uncertainties related to effects on auditory function in adult animals with
relevant Pb exposures remained.

3.6.3.1.1 Epidemiologic Studies of Auditory Function

Several recent epidemiologic studies examined the association between Pb exposure and
decrements in auditory function in adults (Tu et al., 2021; Yin et al., 2021; Wang et al„ 2020; Kang et al.,
2018; Choi and Park, 2017; Shiue, 2013; Choi et al., 2012). The findings generally supported an
association between Pb exposure and hearing loss in adults. These studies are described below.

Most studies were cross-sectional, using data from NHANES (Tu et al„ 2021; Shiue, 2013; Choi
et al., 2012) and KNHANES (Kang et al„ 2018; Choi and Park, 2017). Choi et al. (2012) and Tu et al.
(2021) measured blood Pb in adult NHANES participants (20-69 years) in 1999-2004 and 2011-2012,
respectively. Choi et al. (2012) observed an increased likelihood of hearing loss per doubling of blood Pb
(OR = 1.09 [95% CI: 0.95, 1.26]) as well as an increased percent change in hearing threshold (% change
= 5.41 [95% CI: 2.12, 8.81]). Similar findings were noted in quintile analyses, with higher blood Pb
quintiles having a greater magnitude of effect when compared with the lowest quintile (0.20-0.80 (ig/dL)
(Choi et al., 2012). Tu et al. (2021) measured hearing loss at speech frequency and at high frequency. In
quartile analyses, the magnitude of effect increased with each blood Pb quartile for each type of hearing
loss (Tu et al., 2021). Compared with the lowest quartile (<0.07 |ig/dL). the highest blood Pb quartile
(>0.16 (ig/dL) was positively associated with speech-frequency hearing loss (OR = 1.46 [95% CI: 0.81,
2.64]) and high-frequency hearing loss (OR = 1.98 [95% CI: 1.27, 3.10]). In sex-stratified analyses, the
direction of effect remained but the magnitude of effect was greater among males. In age-stratified

3-204

-------
analyses, a positive association remained for ages <35 years and 35-52 years for both types of hearing
loss; however, ORs were generally imprecise for the younger age group (<35 years). On the contrary,
there appeared to be an inverse association between BLLs and hearing loss in the oldest age group (>52
years) (Tu et al.. 2021). Also using NHANES data, Shiue (2013) measured Pb exposure in urine and did
not find an adverse association with self-reported hearing among older adults (>50 years). For self-
reported ear ringing, urinary Pb had a slightly positive association (Shiue. 2013). In KNHANES, Choi
and Park (2017) measured speech- and high-frequency hearing loss in adolescents (12-19 years) and
adults (20-87 years). Hearing loss was defined as pure-tone average >25 dB in adults. For each doubling
of blood Pb, there was an increased likelihood of speech-frequency hearing loss (OR = 1.15 [95% CI:
0.94, 1.41]) and high-frequency hearing loss (OR = 1.30 [95% CI: 1.08, 1.57]). In quartile analyses, the
magnitude of effect increased with each blood Pb quartile when compared with the lowest quartile (Choi
and Park. 2017). In another analysis conducted in the KNHANES population, Kang et al. (2018) observed
an association between BLLs and hearing impairment in adults (20-87 years). In quartile analyses, the
magnitude of effect for high-level frequency hearing impairment was greatest in the highest blood Pb
quartile compared with the lowest blood Pb quartile in males (OR = 1.63 [95% CI: 1.16, 2.29]) as well as
in females (OR = 1.50 [1.03-2.20]). For low-frequency hearing impairment, the associations were less
consistent by quartile and more attenuated (Kang et al.. 2018).

In a meta-analysis of studies from Iran, Korea, China, and the United States, Yin et al. (2021)
observed consistent positive associations between Pb exposure and any hearing loss (combined OR per
unit increase in Pb = 1.42 [95% CI: 1.22, 1.67]), low-frequency hearing loss (combined OR = 1.31 [95%
CI: 1.17, 1.47]), and high-frequency hearing loss (combined OR= 1.96 [95% CI: 1.48, 2.60]). When
stratified by age group, the association persisted in adults (>20 years; combined OR per unit increase in
Pb = 1.34 [95% CI: 1.18, 1.52]) (Yin et al.. 2021). Despite these results, there are still some
inconsistencies in the recent literature. In a case-control study of adults in China who participated in a
survey of hearing loss, Wang et al. (2020) did not observe an association with blood Pb before and after
adjusting for workplace noise exposure.

Summary

The strongest evidence described in the 2013 Pb ISA was provided by the analysis of NAS men
for associations of higher tibia Pb level with a higher rate of elevations in hearing threshold over 20 years
(Park et al.. 2010). Several recent cross-sectional analyses of NHANES and KNHANES generally support
an association of Pb exposure (i.e., concurrent BLLs) with hearing loss; however, recent studies are not
entirely consistent.

3-205

-------
3.6.3.1.2 Toxicological Studies of Auditory Function

Recent animal studies on the auditory effects of Pb exposures have investigated exposures
beginning during development (postnatal or adolescent), discussed in Section 3.5.6.1.2. In particular,
Carlson et al. (2018) did not detect any significant changes in BAEP in 4-month-old adult mice with very
low BLLs (3 (ig/dL). The strongest evidence for adult effects of Pb exposure was presented in the
previous ISA (U.S. EPA, 2013). Lifetime Pb exposure was found to increase hearing thresholds and
latencies in BAEP in adult monkeys (aged 8-13 years) (Rice, 1997; Lilienthal and Winneke, 1996).
Moreover, Laugh 1 in et al. (2009) detected small nonsignificant shifts in auditory threshold in 13-year-old
Rhesus monkeys following gestational or postnatal Pb exposure. However, these effects were
demonstrated at higher BLLs than are relevant to this ISA (33-150 (.ig/dL).

3.6.3.2 Visual Function

The evidence pertaining to visual function assessed in the 2013 Pb ISA was limited. A case-
control study found higher Pb in retinal tissue from macular degeneration cases but lacked rigorous
statistical analysis and examination of potential confounding. Studies in adult animals showed differential
effects on ERGs, depending on the timing and concentration of exposure. Because the available
epidemiologic and toxicological evidence was of insufficient quantity, quality, and consistency, the 2013
Pb ISA concluded the "evidence is inadequate to determine that a causal relationship exists between Pb
exposure and visual function decrements in adults."

3.6.3.2.1 Epidemiologic Studies of Visual Function

Only a few epidemiologic studies examined the association between Pb exposure and decrements
in visual function in adults (Paulsen et al.. 2018; Fillion et al.. 2013; Shiue. 2013). Fillion et al. (2013)
measured contrast sensitivity (cpd) and acquired color vision loss (CCI) in adolescents and adults (15-66
years) in Brazil. Blood Pb exposure was negatively associated with the intermediate spatial frequency of
contrast sensitivity (12 cycles/degree); however, results varied by spatial frequency. For CCI, there was a
slightly positive association with blood Pb, but the effect estimate was imprecise (Fillion et al.. 2013). In
another study of contrast sensitivity, Paulsen et al. (2018) used the Pelli-Robson letter sensitivity chart
and did not observe an association with blood Pb in a U.S.-based cohort (HR for blood Pb >2.06 |ig/L
versus. <2.06 |ig/L = 0.91 [95% CI: 0.69, 1.18]). Visual function has also been measured using self-
reported eyesight. Using NHANES data, Shiue (2013) measured Pb exposure in urine and did not find an
association with self-reported visual impairment among older adults (>50 years).

3-206

-------
Summary

The epidemiologic evidence pertaining to the association of Pb exposure with visual function in
adults remains limited. A small number of recent studies examining contrast sensitivity or acquired color
vision loss found inconsistent results for associations with blood or urine Pb level (Paulsen et al., 2018;
Fillion et al., 2013; Shiue. 2013).

3.6.3.2.2 Toxicological Studies of Visual Function

No recent PECOS-relevant studies have evaluated the effects of Pb exposure on visual function.
Section 3.5.6.2.2 summarizes the literature discussed in the 2013 Pb ISA (U.S. EPA, 2013).

3.6.3.3	Olfactory Function

The 2013 Pb ISA did not assess any evidence on the relationship between Pb exposure and
olfactory function in adults (U.S. EPA, 2013).

3.6.3.3.1 Epidemiologic Studies of Olfactory Function

In the Heinz Nixdorf Recall Study (HNRS) in Germany, male participants were recruited in
2000-2003 and followed up in 2011-2014 (Casjens et al., 2018). Casjens et al. (2018) examined the
effect of Pb exposure on odor identification using the Sniffin' sticks odor identification test of 12 odors.
Participants were classified as normosmic (identified >9 odors), hyposmic (identified 7-9 odors), and
functionally anosmic (identified <7 odors). Compared with the lowest BLL at baseline (<5 |ig/dL). the
highest BLL (>9 (ig/dL) was associated with impaired odor identification (proportional OR = 1.96 [95%
CI: 0.94, 4.11]). Similar results were observed using BLLs measured at follow-up (proportional OR =
1.57 [95% CI: 0.47, 5.19]).

3.6.3.4	Relevant Issues for Interpreting the Evidence Base

3.6.3.4.1 Potentially At-Risk Populations

Sex

Tu et al. (2021) measured hearing loss at speech frequency and at high frequency. In quartile
analyses, the magnitude of effect increased with each blood Pb quartile for each type of hearing loss (Tu
et al.. 2021). In sex-stratified analyses, the direction of effect remained but the magnitude of effect was

3-207

-------
greater among males. Kang et al. (2018) also observed an association between BLLs and hearing
impairment in adults. In quartile analyses, the magnitude of effect for high-level frequency hearing
impairment was greatest in the highest blood Pb quartile compared with the lowest blood Pb quartile, with
similar associations observed in males (OR= 1.63 [95% CI: 1.16, 2.29]) and females (OR= 1.50 [1.03-
2.20]).

3.6.3.5 Summary and Causality Determination: Sensory Organ Function in Adults

In the 2013 Pb ISA, causality determinations were separately determined for auditory and visual
function in adults, while no studies of olfactory function were evaluated (U.S. EPA, 2013). For auditory
function in adults, the evidence was suggestive of, but not sufficient to infer, a causal relationship, based
primarily on findings from a few epidemiologic studies. Similar to the conclusion for children, the
evidence relating to visual function in adults was inadequate to determine if a causal relationship exists.
In the current ISA, the evidence for auditory, visual, and olfactory function are evaluated together,
forming a single causality determination for sensory organ function.

The strongest evidence described in the 2013 Pb ISA was provided by the analysis of NAS men
for associations between higher tibia Pb levels and a higher rate of elevations in hearing threshold over 20
years (Park et al., 2010). Several recent cross-sectional analyses of NHANES and KNHANES generally
support an association of Pb exposure (i.e., concurrent BLLs) with hearing loss (Tu et al., 2021; Kang et
al„ 2018); Choi and Park (2017); (Choi et al„ 2012); however, recent studies are not entirely consistent
(Wang et al., 2020). Hearing loss and altered responses on BAEPs in adult nonhuman primates and
rodents following lifetime or developmental Pb exposure have been demonstrated at BLLs as low as 29
(ig/dL (Jamesdaniel et al., 2018; Laughlin et al., 2009; Rice, 1997). The few studies that investigated
BLLs from 3-8 (ig/dL did not report altered BAEP in rodents, though effects on auditory processing may
occur at these lower exposure levels (Liu et al., 2019; Carlson et al„ 2018; Zhu et al„ 2016).

The epidemiologic and experimental animal evidence pertaining to the association of Pb exposure
with visual function in adults remains limited. Toxicological studies have demonstrated biological
plausibility for Pb-induced effects on vision, including dysfunction of subcortical visual neurons, visual
processing areas, and retinal development (U.S. EPA, 2013). A small number of recent studies examining
contrast sensitivity or acquired color vision loss in humans found inconsistent results for associations with
blood or urine Pb level (Paulsen et al„ 2018; Fillion et al„ 2013; Shiue, 2013). Deficits in visual temporal
acuity have been demonstrated in adult nonhuman primates, although peak exposure levels are higher
than considered relevant for this assessment (Rice, 1998). Altered responses to ERGs have been detected
at a wide range of BLLs, but the direction of this effect (i.e., supernormal or subnormal responses) is
overall inconsistent (Fox et al„ 2008; Rothenberg et al„ 2002; Fox et al„ 1997).

Olfactory function was not discussed in the 2013 Pb ISA. Recently, baseline BLL was associated
with reduced odor identification in a prospective analysis of the German HNRS (Casjens et al., 2018).

3-208

-------
Overall, the evidence is suggestive of, but not sufficient to infer, a causal relationship between
Pb exposure and sensory function in adults. This determination is supported by generally consistent
prospective and cross-sectional analyses demonstrating Pb-associated hearing loss in adults. The few
experimental animal studies available provide coherence for this endpoint when BLLs are greater than 29
(.ig/dL. but not at BLLs more relevant to current human exposures. Human and experimental animal
studies investigating the effect of Pb on visual and olfactory function are limited and inconsistent.

3-209

-------
Table 3-11 Summary of the evidence that is suggestive of, but not sufficient to infer, a causal relationship
between sensory function in adults

Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0

Auditory Function

Generally consistent
associations observed in
multiple epidemiologic
studies.

Prospective study found association
between tibia Pb level and higher rate of
increase in hearing threshold over 23 yr in
males enrolled in the NAS.

Park etal. (2010)

Tibia Pb mean: 22.5 |jg/g, measured near
end of follow-up

Cross-sectional analyses of NHANES and
KNHANES find Pb-associated effects on
hearing loss

Shiue (2013)

Choi et al. (2012) Tu et al. (2021)
Kana etal. (2018)

Choi and Park (2017)

Uncertainty at relevant
exposure levels in
experimental animal studies

Hearing loss and altered responses on
BAEPs in adult nonhuman primates and
rodents demonstrated

Rice (1997)

Lauqhlin et al. (2009)
Jamesdaniel et al. (2018).

Lifetime or developmental Pb exposure
>29 |jg/dL



No altered BAEP in rodents, though
effects on auditory processing may occur

Carlson et al. (2018)
Zhu etal. (2016)
Liu et al. (2019)

3-8 |jg/dL

Visual Function

Inconsistent results across
limited epidemiologic studies

Associations of blood or urine Pb level
with contrast sensitivity or acquired color
vision loss were inconsistent

Shiue (2013)

Paulsen et al. (2018)
Fillion et al. (2013)



Limited evidence from
experimental animal studies

Deficits in visual temporal acuity
demonstrated in adult nonhuman
primates

Rice (1998)

>30 |jg/dL

3-210

-------
Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker Levels Associated with
Effects0

Biological plausibility
demonstrated

Dysfunction of subcortical visual neurons,
visual processing areas, and retinal
development

(U.S. EPA, 2013)



Olfactory Function

Single study indicates
association

Impaired odor identification associated
with BLL in a single study

Casiens et al. (2018)

>9 vs. <5 |jg/dL

BAEP = brainstem auditory evoked potentials; BLL = blood lead level; KNHANES = Korea National Health and Nutrition Examination Survey; NAS = Normative Aging Study;
NHANES = National Health and Nutrition Examination Survey; Pb = lead; yr = year(s).

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs (U.S. EPA. 20151.
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or
inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-211

-------
3.6.4

Neurodegenerative Diseases

The 2013 Pb ISA concluded that the evidence was "inadequate to determine that a causal
relationship exists" between Pb exposure and neurodegenerative diseases (U.S. EPA, 2013). Evidence
was inconclusive for amyotrophic lateral sclerosis (ALS) (see Section 4.3.9.2 of (U.S. EPA, 2013)) and
Alzheimer's disease (AD; see Section 4.3.9.1 of (U.S. EPA. 2013)); however, a few case-control studies
each found higher BLLs in adults with essential tremor and higher bone Pb levels in adults with PD
(Weisskopf et al„ 2010; U.S. EPA, 2006).The evidence was considered inconclusive overall due to a
limited number of studies, the potential for reverse causation (specifically for case-control studies in
which the reduced physical activity among cases could result in greater bone turnover and greater release
of Pb from bones into blood as compared with controls), and limited consideration for potential
confounding factors.

Recent epidemiologic studies indicated potential relationships between Pb exposure and some
neurodegenerative disease endpoints among non-occupational cohorts. The strongest evidence in the
current review includes well-designed studies of ALS and PD outcomes. Findings for AD, tremor, and
motor function were inconclusive. Measures of central tendency for Pb biomarker levels used in each
study, along with other study-specific details, including study population characteristics and select effect
estimates, are highlighted in Section 3.7, Table 3-17E. A large number of toxicological studies adds to the
evidence suggesting that developmental exposure to Pb increases the expression of pathophysiological
markers of AD, including amyloid beta (A|3) peptides, tau, and phosphorylated tau (p-tau), at lower BLLs
than were investigated in the previous ISA (<10 (.ig/dL). Toxicological studies investigating potential
associations between Pb exposure and PD, ALS, and essential tremor remain limited in this review.
However, toxicological evidence for PD and ALS are supported by some studies in the 2013 review that
provided pathophysiological evidence for Pb-induced decreases in dopaminergic cell activity in the
substantia nigra, which can contribute to PD development, and Pb exposure affecting neurophysiologic
changes associated with ALS.

3.6.4.1 Epidemiologic Studies of Neurodegenerative Diseases

3.6.4.1.1 Alzheimer's Disease

MMSE is a widely used screening tool for AD and other types of dementia. Lower scores on
MMSE were consistently associated with higher bone Pb levels, which indicated long-term or cumulative
exposure to Pb, in the NAS studies assessed in the 2013 Pb ISA (Wang et al.. 2007; Weisskopf et al..
2004; Wright et al.. 2003). There was heterogeneity in the results of studies that examined associations
between MMSE scores and BLLs in adults (Weuve et al.. 2006; Nordberg et al.. 2000). Blood Pb is

3-212

-------
generally considered a marker of recent exposure; however, in studies of adults, blood Pb level also
reflects Pb that is mobilized from the bone introducing uncertainty and complicating the interpretation of
cross-sectional studies that assess exposure using concurrent blood Pb level. Evidence regarding the
association of Pb exposure with clinical diagnosis of AD was limited to studies which did not find
associations with higher occupational exposure to Pb (Graves et al.. 1991) or higher Pb concentration in
the brains (Haraguchi et al.. 2001) in AD cases compared with unaffected controls. The latter studies were
limited because of their case-control designs, which may be subject to reverse causation where AD leads
to higher Pb levels, limited consideration for potential confounding, and because of the potential
misalignment with relevant exposure window.

Recent studies add to the evidence base, including an analysis of the NAS cohort that examined
the association of bone Pb biomarkers with cognitive impairment, including MMSE score, Farooqui et al.
(2017). and two studies that examined the association of blood Pb biomarkers with the clinical endpoints
of AD risk or AD mortality (Horton et al.. 2019; Yang et al.. 2018) in non-occupational cohorts (Table
3-16T). Among the older male participants in the NAS, higher patella Pb concentration (IQR: 21 |ig/g)
was associated with increased risk (HR: 1.10, 95% CI: 0.99, 1.21) ofhaving an MMSE score below 25
(threshold that represent cognitively not normal or at risk for dementia), while less support was observed
for an association with tibia Pb concentration (HR: 1.03, 95% CI: 0.88, 1.22) (Farooqui et al.. 2017).
Studies that specifically assessed clinically diagnosed AD or mortality did not provide strong evidence of
an association. A case-control study by Yang et al. (2018) included participants from clinical settings in
Taiwan and used standard case-control as well as propensity score-matched approaches to assess the
relationship between the heavy metals (Pb, Cd, Se, Hg) and AD risk. Findings from the multivariable
analysis showed the association between BLL and AD risk in tertiles, either in the full population tertile 2
(OR 1.00, 95% CI 0.56-1.79) and tertile 3 (OR 0.87, 95% CI 0.49-1.55) or propensity score-matched
population (tertile 2: OR 1.16, 95% CI 0.55-2.47; and tertile 3: OR 1.12, 95% CI 0.53-2.39), was
imprecise. A cohort study by Horton et al. (2019) used national data from five NHANES cycles (1999-
2008) and followed a large cohort of 8,080 participants from 1999 till December 2014 for AD-related
mortality to examine the longitudinal association between blood Pb and AD mortality. Results from Cox
proportional hazard models adjusted for various confounders and competing risks for AD mortality (death
due to cancer, cardiovascular disease (CVD), cerebrovascular accident [CVA], nephritis, and respiratory
disease) indicated that BLLs of 1.5 and 5 (ig/dL had 1.2 (95% CI = 0.70, 2.1) and 1.4 (95% CI = 0.54,
3.8) times the rate of AD mortality compared with those with a BLL of 0.3 (ig/dL, respectively. The
associations observed for various BLL categories with respect to the reference category of BLL 0.3 (ig/dL
were in a positive direction with increased AD risk for increasing BLL categories; however, the
associations were imprecise. The imprecise effect estimates are likely due to the small number of AD
mortality cases (n = 81), which resulted from AD mortality being determined from the listing of the
immediate cause of death rather than the underlying cause of death. This means the study may be
underpowered, potentially resulting in an unstable effect estimate.

3-213

-------
3.6.4.1.2 Amyotrophic Lateral Sclerosis

Case-control and cohort studies examining the association of BLL and ALS risk or ALS survival
that were included in the 2006 AQCD for Pb or the 2013 Pb ISA produced inconsistent results (Fang et
al.. 2010; Kamel et al.. 2008; Kamel et al.. 2002; Vinceti et al.. 1997). Recent case-control and cohort
studies that assessed biomarkers of Pb before disease development addressed uncertainties related to
temporality and reverse causality identified in previous reviews, thus expanding the support for an
association of BLL with ALS risk and survival.

Strong evidence for the association of Pb and ALS is provided by a recent prospective cohort
study that used data from the National Registry of Veterans in the United States with ALS cases
ascertained between April 2003 and September 2007 (with blood samples collected from January to
September 2007) and followed through the date of death or July 2013 (April 2003-Sep 2007) Fang et al.
(2017). The study was novel in that it assessed ALS mortality and survival and its association with blood
Pb level, and also bone turnover (formation and resorption) biomarkers. The association of ALS survival
time with blood Pb indicated that increased blood Pb was significantly associated with the increased
mortality and thus shorter survival after ALS diagnosis (HR: 1.23 [95% CI: 1.02, 1.49]) in the model
mutually adjusted for bone resorption and formation and other confounding variables. The observation of
the association between BLL and ALS after adjustment for biomarkers of bone turnover reduced
uncertainties related to assessing exposure using Pb concentration in the blood. In another study, Peters et
al. (2020) conducted a nested case-control study within the prospective European Prospective
Investigation into Cancer and Nutrition (EPIC) cohort study. ALS cases were defined to include subjects
with motor neuron disease (ICD10 G12.2) as the underlying cause of death. Pb concentration was
measured in erythrocytes as a marker of ongoing exposure. The associations of Pb in erythrocytes
comparing categories of >56.8 to < 89.0 ng/g and >89.0 ng/g to the reference category (<56.8 ng/g) with
ALS mortality were 1.83 (95% CI: 0.99, 3.35) and 1.89 (95% CI: 0.97, 3.67), respectively.

Additional studies indicating associations of Pb concentration in cerebrospinal fluid (CSF) and Pb
in air also provide some support for an association between Pb exposure and ALS. A case-control study
in Italy used CSF biomarkers for heavy metals including Pb Vinceti et al. (2017). The odds of ALS were
greater in the highest tertile of CSF Pb concentration than in the lowest tertile; however, the effect
estimate was imprecise (OR: 1.39, 95% CI 0.48-4.25). Another case-control study of a large nationally
representative U.S. sample (cases: 26,199 and controls: 78,597) used the U.S. healthcare claims database
from the Symphony Health's Integrated Dataverse (Andrew et al.. 2022). Participants with the first ALS
diagnosis after 6 months of enrollment in the database were included (diagnosis years 2013-2019).
Controls were matched based on age and sex, selected from the Symphony Health network, and also
required to have a minimum of 6 months enrollment in the database. The study did not use the Pb
biomarkers but rather used the airborne contaminants level data for 268 contaminants (including Pb)
obtained from the U.S. EPA National Emissions Inventory (NEI) for 2008 to estimate the past exposure
prior to the ALS onset (2013-2019) at the participants" location of residence. The study used athree-

3-214

-------
phase approach to assess ALS risk nationwide: discovery, validation, and confirmation. First, in the
discovery phase, the study identified major contaminants (out of 268 contaminants) that were
significantly associated with the ALS risk. Second, in the validation phase, the study evaluated various
combinations of contaminants associated with the ALS risk. In the final confirmatory phase, the study
used cohorts only from NH, VT, and OH, incorporating their detailed residential history information to
capture changes in exposure due to residential move prior to ALS diagnosis. The discovery phase
identified 49 airborne contaminants (including Pb) associated with ALS risk. The relationship between
these 49 contaminants and the risk of ALS was further analyzed in the validation cohort, and airborne Pb
and five PCBs were identified associated with an increased risk for ALS (Pb: OR: 1.02, 95% CI: 1.01—
1.03). The confirmatory analysis using NH/VT and OH based cohorts with detailed residential history to
calculate 5-, 10-, and 15-year past exposure prior to diagnosis suggested significant increased risk for
ALS associated with 10-year Pb exposure history when the >75th percentile group was compared with the
<50th percentile group (NH/VT: OR: 2.03, 95% CI: 1.46-2.80, and OH: OR: 1.60, 95% CI: 1.28-1.98) in
a multivariable model. Despite the strong design and larger sample size of the study, the inference
regarding the Pb-ALS risk for this study should be interpreted with caution given the uncertainty
regarding the relationship between the estimated concentration of Pb in the air and Pb concentration in
biomarkers as well as the influence of potential unmeasured confounders.

3.6.4.1.3	Parkinson's Disease

A limited number of case-control studies assessed in the 2013 Pb ISA found positive associations
between bone Pb concentration and PD. A recent study by Paul et al. (2021) examined participants from
two large and independent population-based case-control studies (total n > 2,600)—the System Genomics
of Parkinson's Disease (SGPD), a consortium of three studies from across Australia and New Zealand;
and the Parkinson's Environment and Genes (PEG) study, a population-based study from three
agricultural counties of Central California—to explore the association of cumulative Pb exposure on the
PD risk. The study used novel epigenetic biomarkers of cumulative Pb exposure (i.e., DNA methylation
[DNAm] Pb data in the patella and tibia developed in the NAS cohort). The study analyzed the
relationship between DNAm Pb and PD separately for two cohorts and meta-analyzed the results. The
findings from the multivariable adjusted model suggested that PD risk was strongly associated with the
DNAm biomarker for tibia Pb levels in both cohorts (SGPD cohort: OR: 2.06, 95% CI: 1.66-2.56; PEG
cohort: OR: 1.60, 95% CI: 1.20, 2.15; meta-analyzed results [meta-OR: 1.89, 95% CI: 1.59-2.24]).

3.6.4.1.4	Tremor

A limited number of studies examined the association between BLLs and tremor. The studies
were potentially influenced by reverse causation because inactivity due to disease condition and
subsequent bone resorption can lead to increased BLLs. A recent study used a cohort of men in the NAS

3-215

-------
and assessed the longitudinal relationship of tremor score with bone Pb in the tibia (n = 670, mean: 21.23
(ig/g) and patella (n = 672, mean: 27.98 |ig/g). in addition to BLL (n = 807, mean: 5.01 (ig/dL). Ji et al.
(2015) found that over 8 years of follow-up, neither blood Pb nor bone Pb was associated with the tremor
score. However, among men younger than the median age (68.9 years), the tremor score increased as the
quintile of blood Pb increased (p = 0.03), with men in the highest quintile scoring 0.35 (95% CI: 0.03,
0.67) points higher on the tremor scale than those in the lowest quintile. This pattern was not evident for
bone Pb. The tremor score in the study was based on assessments of drawing capability and not the
clinical diagnosis.

3.6.4.1.5 Motor Function

Several epidemiologic studies examined the association between Pb exposure and decrements in
motor function in adults (Casiens et al.. 2018: Khalil et al.. 2014: Grashow et al.. 2013: Ji et al.. 2013:
Shiue. 2013: Min et al.. 2012). Motor function was assessed using measures of balance, walking speed,
coordination, and strength. Inconsistencies in the results made it difficult to draw conclusions about the
association between Pb exposure and motor function in adults.

Most studies were cross-sectional in design. Results from studies that measured balance were
inconsistent. Among older adults (>50 years) in NHANES 2003-2004, Shiue (2013) observed an inverse
association between urinary Pb and balance disorders defined as self-reported dizziness, difficulty with
balance, or difficulty with falling in the past 12 months. For every log unit increase in urinary Pb (unit not
specified), the likelihood of having a balance disorder decreased (OR = 0.56 (95% CI: 0.38, 0.84]) (Shiue.
2013). Among adults (>40 years) who participated in the NHANES Balance Component, Min et al.
(2012) found that higher levels of blood Pb were generally associated with an increased likelihood of
failing a balance test. Balance dysfunction was evaluated using the Romberg Test of Standing Balance on
Firm and Compliant Support Surfaces, which measured a participant's ability to maintain balance under
various test conditions. Compared with the lowest quintile of blood Pb (<1.2 (.ig/dL). the likelihood of any
balance dysfunction (failing any balance test) increased in the fourth quintile (2.3-3.2 (ig/dL; OR = 5.23
[95% CI: 0.59, 46.43]) and fifth quintile (3.3-48 ^ig/dL; OR = 33.33 [95% CI: 1.94, 573.16]) (Min et al..
2012).

Among older adults (50-85 years) in NHANES, Ji et al. (2013) assessed the relationship between
blood Pb and walking speed. Walking speed was measured by timing a participant's walk for 20 feet at
their usual walking pace. Compared with the lowest quintile of blood Pb (0.2 to <1.2 (.ig/dL). walking
speed decreased with increasing quintiles of blood Pb in women (p-trend = 0.005). In the highest quintile
of blood Pb (3.0 to <53.0 (.ig/dL). the estimated mean walking speed in women was 0.11 feet/second
slower (|3 = -0.11 [95% CI: -0.19, -0.04]). On the contrary, BLL did not appear to be associated with
walking speed in men (Ji et al.. 2013). In MrOS, Khalil et al. (2014) examined the association between
blood Pb and walking speed and strength among older non-Hispanic Caucasian men (>65 years). In this
cross-sectional analysis, blood Pb did not appear to be associated with grip strength, walking speed, or

3-216

-------
narrow-walk pace. Leg extension power (watts) measured with the Nottingham power rig had a negative
association for every log-(ig/dL increase in blood Pb (|3 = -0.03 [95% CI: -1.97, 2.03]). In addition, blood
Pb had a negative association with the participants" ability to stand from a chair without using their arms
(OR per log-(ig/dL increase in blood Pb = 0.97 [95% CI: 0.88, 1.07]) (Khalil et al.. 2014). The reported
associations were very imprecise.

A few cohort studies measured fine motor abilities and hand-eye coordination in adults. In the
HNRS in Germany, Casiens et al. (2018) examined the effect of Pb exposure on fine motor abilities
among male participants who were recruited in 2000-2003 and followed up in 2011-2014. Fine motor
abilities were measured at follow-up (at ages 55-86 years) and included four tasks (tapping, aiming, line
tracing, and steadiness) carried out separately with each hand. Compared with the lowest BLL at baseline
(<5 (ig/dL), the highest BLL (>9 (ig/dL) was positively associated with tapping hits (OR = 1.35 [95% CI:
0.49, 3.70]) and steadiness errors (OR = 1.36 [95% CI: 0.50, 3.66]) but negatively associated with aiming
errors (OR = 0.56 [95% CI: 0.22, 1.42]) and line tracing errors (OR = 0.93 [95% CI: 0.32, 2.74]). The
magnitude of each association increased when using BLLs measured at follow-up, except for tapping hits,
which changed to a negative association (OR = 0.98 [95% CI: 0.82, 1.16]). In general, the ORs were
imprecise (Casiens et al.. 2018). In the Department of Veterans Affairs NAS, Grashow et al. (2013)
examined the association between bone Pb and a coordination task which involved inserting metal pegs
into a grooved pegboard. Bone Pb was measured at the patella and the midtibial shaft and was positively
associated with the grooved pegboard completion time. In other words, the pegboard test took longer to
complete for every 10 (ig/g increase in patella Pb (|3 = 1.97 [95% CI: 0.55, 3.38]) and 10 |ig/g increase in
tibia bone Pb (|3 = 3.11 [95% CI: 1.16. 5.061) (Grashow et al.. 2013).

3.6.4.1.6 Summary

In summary, recent epidemiologic studies found relationships between Pb exposure and some
neurodegenerative disease endpoints among non-occupational cohorts. Similar to the conclusion of the
2013 Pb ISA, the direction and strength of the association is stronger for some endpoints than for others.
In the 2013 Pb ISA, evidence was inconclusive for ALS and AD while a limited number of case-control
studies indicated relationships between higher BLLs in adults and essential tremor, and between higher
bone Pb levels in adults and PD. In the current review, the epidemiologic evidence pertaining to ALS and
PD has strengthened due to the availability of better designed case-control and cohort studies. Recent
studies of Pb exposure and clinically diagnosed AD were also conducted. These studies added to the
previous evidence which generally relied on assessing AD using screening instruments; however, studies
that specifically assessed clinically diagnosed AD or mortality did not provide strong evidence of an
association.

Studies for ALS in this review included one cohort study that examined the relationship between
BLLs and ALS survival among U.S. veterans (Fang et al.. 2017) and a large case-control study of
participants from a healthcare claims dataset examining associations between past airborne Pb exposures

3-217

-------
and ALS risk (Andrew et al.. 2022). The findings from the cohort study by (Fang et al.. 2017) indicate
that increased levels of baseline blood Pb are associated with shorter ALS survival even after mutually
accounting for bone resorption and formation. This finding is in agreement with the results of a study
included in the previous review that suggested positive associations between Pb in blood or bone with
ALS risk (Fang et al.. 2010; Kamel et al.. 2002). An important uncertainty in these studies is the potential
for reverse causality because increased bone turnover in ALS patients could cause higher blood Pb levels,
potentially explaining the observed associations. Fang et al. (2017) addressed this uncertainty by
demonstrating that the association of blood Pb level with ALS survival persisted after adjustment for a
biomarker of bone turnover. Another study investigating ALS risk using a case-control design with a
large sample from a healthcare database and well characterized exposure and outcomes examined whether
previous airborne Pb exposures were related to ALS development. The study found significant positive
associations between Pb exposure and increased ALS risk (Andrew et al.. 2022). specifically for
residential Pb exposure over the past 10 years. The use of estimated airborne exposure without
corresponding measurement of Pb exposure biomarkers is a potential limitation of this study.

In a study of PD, Paul et al. (2021) conducted a case-control analysis using a novel, epigenetic
biomarker to estimate cumulative Pb exposure measured in tibia and patella bone. The study found an
association between DNA methylation (DNAm), as a biomarker of tibia Pb levels, and PD risk. This
empirically-derived DNA methylation signature is potentially a more sensitive predictor of the effect of
Pb exposure on PD than bone Pb concentration (Paul et al.. 2021). With regard to tremor outcomes, no
association with blood and bone Pb biomarkers was observed (Ji et al.. 2015). However, the results
indicated an association of blood Pb and tremor score, particularly in younger men, when analysis was
stratified by age categories. Studies reviewed for Pb exposure and motor function in adults yielded
inconsistent findings.

Studies of the association of Pb exposure with AD in the previous ISA examined cognitive
impairment (e.g., as indicated by MMSE scores, which are used to screen for dementia and AD) rather
than the clinical diagnosis of AD. Recent studies add to the evidence through their examination of clinical
diagnosis of AD and AD mortality in non-occupational cohorts. In a case-control study, Yang et al.
(2018) reported an imprecise positive association between BLL and AD risk. The inability to establish
temporal relationships with this case-control study, given the inclusion of prevalent cases of AD, leads to
uncertainty about potential reverse causality. A recent prospective study that investigated the relationships
between BLLs and AD mortality addressed the temporality of the association (Horton et al.. 2019). The
study calculated HRRs for selected BLLs (i.e., 0.5, 1, 1.5, 2, 3, 5 (ig/dL) compared a reference category of
0.3 (ig/dL. In addition to considering confounders in their model, the authors specified model to consider
the design effect (i.e., accounting for the NHANES survey design utilized by incorporating survey
weights) and competing risks. The study observed a positive association between blood Pb and AD
mortality risk, but the association was imprecise (Horton et al.. 2019). For example, HRR for participants
with BLL of 1.5 1.5 (95% CI = 0.81, 2.9) compared to those with BLL of 0.3 (ig/dl, respectively, after
accounting for design effect.

3-218

-------
3.6.4.2 Toxicological Studies of Neurodegenerative Diseases

Although the evidence was inconclusive overall, a few toxicological studies in the 2013 Pb ISA
suggested that Pb exposure in early life could influence AD-like pathologies (U.S. EPA, 2013). AD
exhibits several neuropathologic hallmarks, such as senile plaques and neurofibrillary tangles (comprised
of A|3 and hyperphosphorylated tau aggregates, respectively), as well as synaptic loss and neuronal death.
Developmental exposure to Pb in rodents, resulting in BLLs >40 (ig/dL, increased A|3 peptides,
hyperphosphorylated tau, and other related endpoints (Li et al„ 2010; Basha et al., 2005). Importantly,
these effects were not found following adult-only exposure. Wu et al. (2008) also demonstrated that 23-
year-old monkeys (Macaco fasciculctris) given Pb from birth to PND 400 had elevated A|3 and amyloid
plaques in their frontal cortex compared with unexposed age-matched controls. BLLs in these animals
ranged from 19 to 26 (ig/dL at PND 400 but had returned to baseline by adulthood. One previous study
performed in transgenic superoxide dismutase 1 (SOD1) mice (a model of ALS) found that adolescent
exposure to Pb reduced astrocyte reactivity and extended the survival time but had no significant effects
on the onset of disease in this model (Barbeito et al., 2010). Tavakoli-Nezhad et al. (2001), reviewed in
the 2006 AQCD, demonstrated Pb-induced decreases in dopaminergic cell activity in the substantia nigra,
which is associated with PD. No recent PECOS-relevant studies have investigated the effects of Pb
exposure on ALS-relevant endpoint or endpoints related to essential tremor.

The potential relationship between Pb exposure and AD has been further explored in recent
literature (Table 3- 17T). Of the two major A|3 isoforms (A|340 and A|342), the less predominant isoform,
A|342, is typically considered more prone to aggregation (Xiao et al„ 2015). A low A|342/A|340 ratio in
plasma and CSF has also been shown to indicate an increased risk for AD (Graff-Radford et al., 2007).
(Zhou et al., 2018) found increased expression of A|342 in the cerebral cortex and hippocampus of
Sprague Dawley rats following Pb exposure during adolescence. Gestational and lactational Pb exposure
resulting in lower BLLs, between 4-10 (ig/dL, also significantly increased A|340 in the cerebral cortex of
Kunming mice (Li et al., 2016c). Utilizing a transgenic mouse model (Tg-SwDI), Gu et al. (2012)
reported that adolescent Pb exposure significantly increased both A|340 and A|342 in the cerebral cortex,
hippocampus, and CSF; however, the ratio of A|342/A|340 was not significantly different. Pb-treated
animals in this study also had increased amyloid plaque formation, which was co-localized with brain Pb
deposits. Increases in the expression of amyloid precursor protein (APP) and beta-secretase 1 (BACE1;
the enzyme that cleaves APP into A|3) have been demonstrated in some recent studies (Wu et al., 2020b;
Zhou et al„ 2018; Sun et al., 2014), but not all (Gu et al„ 2012).

Changes in the expression of both total tau (t-tau) and p-tau are considered biomarkers of AD in
humans. One recent study demonstrated that developmental Pb exposure (GD 0-PND 21; resulting in
BLLs of 7 (ig/dL) increased t-tau and p-tau in the cerebral cortex and cerebellum but not the hippocampus
of juvenile Wistar rats (Gassk et al., 2016b). These changes coincided with some evidence of
enhanced activity of two tau kinases (glycogen synthase kinase-3|3 and cyclin-dependent kinase 5
[CDK5]) in relevant brain regions. Another study found that postnatal Pb exposure in Wistar rats caused

3-219

-------
significant changes in t-tau, p-tau, and related phosphatases in the hippocampus, but these effects were
transient and inconsistent at the time points tested (PND 21 and PND 30) (Rahman et al.. 2012b). Wu et
al. (2020b) also found no significant increases in hippocampal p-tau in Pb-exposed C57B1/6 mice at 4
months old; however, increases in hippocampal p-tau became apparent at 13 months old and persisted
until 16 months old. In the prefrontal cortex, p-tau was significantly elevated above age-matched control
values at 4 month and 13 months, but not 16 months. Importantly, BLLs in these rodents at 4 months
(when exposure was terminated) were nearly 60 (ig/dL and did not decrease to PECOS-relevant values
until 16 months (28 (.ig/dL). One additional study Zhang et al. (2012) reported that 8 weeks of Pb
exposure increased p-tau in the hippocampus at BLLs as low as 10 (ig/dL. Notably, this study also
reported increased alpha-synuclein in the hippocampus. Alpha-synuclein is a major constituent of Lewy
bodies, which are a neuropathologic hallmark of PD (Baba et al.. 1998).

In the 1980s, Rice (1990) established a cohort of monkeys {Macaca fascicidaris) exposed to Pb in
the first 400 days of life (resulting in BLLs between 19-26 (ig/dL) and terminated at 23 years old. At the
time of termination, BLLs had returned to control levels. Using tissues from these animals, Wu et al.
(2008) found increases in the protein expression of A|3 and APP, as well as increases in the gene
expression of APP and specificity protein 1 (Spl, a transcriptional regulator of APP and tau), reported in
the previous ISA. Recently, Bihaqi and Zawia (2013) extended these findings by analyzing the cerebral
cortex tissue for changes in tau-related endpoints. Compared with age-matched controls, Pb-exposed
animals had significantly increased expression of t-tau, p-tau, and CDK5 (a tau kinase). These findings
were further supported by neuropathological changes (i.e., increases in p-tau immunoreactivity and
deposits). This study also found that mRNA levels of tau, CDK5, Spl, and Sp3 were significantly
increased.

Recent studies have also measured endpoints outside of those related to A|3 and tau.

Dysregulation of lipid pathways has been implicated in AD and neurodegenerative disorders (Di Paolo
and Kim. 2011). Zhou et al. (2018) found that Pb exposure decreased total and free cholesterol levels via
dysregulation of cholesterol metabolism in the cerebral cortex and hippocampus. Feng et al. (2019) found
that lifetime exposure to Pb significantly decreased neuronal density in the cerebral cortex at PNW 70.
This change was accompanied by a decrease in overall brain volume. In addition, several studies that
reported on AD-related neuromolecular changes also reported significant impairment of learning and
memory assessed via the Morris water maze (Wu et al.. 2020b; Li et al.. 2016c; Gu et al.. 2012; Rahman
et al.. 2012b). However, these results cannot be definitively attributed to AD-related neurobehavioral
changes due to the well-known effects of Pb on cognitive function, which are unrelated to AD.

Deficiencies in balance, walking speed, coordination, and strength can also arise from insults to
the motor system in adulthood. Recent toxicological studies exposed mature rodents to Pb and
investigated the effects on motor function. Typical rotarod tests compare the latency to fall for subjects
placed on a rotating rod. Falling off more quickly indicates decreased coordination or balance. Locomotor
activity tests (e.g., measurements of distance traveled, counts of square crossings) can detect gross motor

3-220

-------
problems as well; however other influences on behavior may factor into differences in the amount of
movement. Fine motor forelimb grip strength can be determined in rodents by pulling subjects holding
onto a measurement-taking tension bar. Mansouri et al. (2012) subjected Wistar rats to Pb acetate for 30
days and observed hyperactivity in open-field tests in males but not females on the final day (PND 100).
They also observed no effect on rotarod performance in both males and females on PND 100. However,
in a subsequent study, Mansouri et al. (2013) found that long-term exposure to Pb acetate in drinking
water (155-159 days starting at PND 55-60) resulted in substandard rotarod performance (7.5 months
old) for male Wistar rats, whereas the exposure had no effect on performance of female rats. Similarly, in
a long-term exposure study by Singh et al. (2019). male Wistar rats exposed daily to Pb acetate by oral
gavage from 3 months to 6 months of age performed worse in rotarod and grip strength tests compared
with their saline-treated counterparts. Singh et al. (2019) also found decreased activity in 6-month-old Pb-
treated rats. Al-Qahtani et al. (2022) observed a decrease in locomotor activity in 15-week-old male mice
after a 6-week Pb treatment period.

Summary

In summary, recent studies have significantly expanded the toxicological literature base
established in the last Pb ISA for AD. Significant increases in A|340 and A|342 following Pb exposure
were consistently detected in the cerebral cortex, hippocampus, and CSF in multiple studies at BLLs of 4-
30 (ig/dL. Pb-induced amyloid plaque formation was also reported in a transgenic mouse model of AD
(Tg-SwDI) (Gu et al.. 2012). Aged cynomolgus monkeys (23 years old), exposed to Pb during infancy,
had both amyloid plaques and tau deposits in their cerebral cortex; however, these findings are limited
somewhat by the small sample size (Bihaqi and Zawia. 2013; Wu et al.. 2008). In rodents, mean BLLs
<10 (ig/dL were shown to increase the expression and phosphorylation of tau in multiple brain regions,
but this effect was not entirely consistent between studies. Overall, recent studies have primarily focused
on exposure paradigms beginning during development, but one study demonstrated effects from an
exposure beginning in early adulthood (8 weeks) (Gu et al.. 2012). One study reported that the PD-related
protein, alpha-synuclein, was increased in the hippocampus in Pb-treated male rats (Zhang et al.. 2012).
Recent studies have not expanded on previous findings on ALS-related endpoints or contributed evidence
related to essential tremor.

3.6.4.3 Relevant Issues for Interpreting the Evidence Base

3.6.4.3.1 Concentration-Response Function

The shape of the C-R function was not examined in the studies of the association of Pb
biomarkers with neurodegenerative diseases in adults in the past review. The majority of studies in the
current review also did not explore the shape of the C-R function in the Pb-neurodegenerative disease

3-221

-------
associations. Horton et al. (2019) explored the relationship between BLL and AD mortality using Cox
regression models that incorporated design effect or competing risks. The study found an increase in the
hazard rate ratio (HRR) by 14%-30% with each unit increase in BLL (see Figure 3-15 below). Given the
small number of AD mortality events in the study population, the effect estimates were imprecise and had
larger CIs. The authors also performed categorical analysis to explore blood Pb-AD mortality association.
Results from Cox proportional hazard models adjusted for various confounders and competing risks for
AD mortality (death due to cancer, CVD, CVA, nephritis, and respiratory disease) indicated that
participants in the 1.5 and 5 (ig/dL BLL categories had 1.2 (95% CI = 0.70, 2.1) and 1.4 (95% CI = 0.54,
3.8) times the rate of AD mortality compared with those with a blood Pb reference of 0.3 (ig/dL,
respectively. The associations observed for various BLL categories with respect to the reference category
of BLL 0.3 (ig/dL were imprecise.

BLL = blood lead level; HRR = hazard rate ratio; LCI = lower confidence interval.
Source: Reproduced with permission from Horton et al. (20191.

Figure 3-15 Hazard rate ratios for Alzheimer's disease mortality by blood Pb
level including the lower 95% confidence interval.

1.5 2 2.5 3 3.5 4 4.5 5
BLL (ng/dL)

Design effect

—	— — LCI-Design effect

	Null line

	Competing risk

—	— — LCI-Competing risk

3-222

-------
3.6.4.3.2 Potentially At-Risk Populations

Genetics

Several studies in the 2013 Pb review evaluating the association between Pb and MMSE (a
marker for AD) and effect modification by genetic variants provided support for effect modification of
the association by the ALAD genotype (details on Potentially At-Risk Populations in the Cognitive
Function Section: 3.6.1.3.2). A study by (Fang et al.. 2010) examining the association of BLL and ALS
risk and effect modification by the ALAD genotype suggested a significant Pb-ALS association among
ALAD1-1 carriers but a weaker and imprecise association among ALAD2 carriers. Tests to identify an
interaction between Pb and the ALAD genotype in the Pb-ALS association suggested no significant
difference in association between ALAD1-1 versus ALAD2 carriers, however (p = 0.32).

In the current Pb review, (Ji et al.. 2015) considered both bone and blood Pb biomarkers among
the NAS cohort and performed stratified analysis by ALAD gene (ALAD-2 carriers or non-carriers).

They found no effect modification by the ALAD genotype for the association between Pb biomarkers and
elevated tremor.

Age and Sex

A few studies in the current review explored the effect modification of the Pb-neurodegenerative
disease associations by age or sex. (Ji et al.. 2015) examined the associations of bone and blood Pb
biomarkers with tremor among the NAS cohort. The results suggested that among younger cohorts (i.e.,
below the median age of 68.9 years), the tremor score increased significantly with increasing quintile of
blood Pb (p = 0.03), and those in the highest quintile scored 0.35 (95% CI: 0.03, 0.67) points higher than
those in the lowest quintile. This pattern was not apparent when bone Pb biomarkers were used. Similarly,
(Paul et al.. 2021) performed stratified analysis of the DNAm estimated tibia and patella Pb
concentrations and PD risk by sex and found a significant association when tibia Pb concentration was
used. The magnitude of risk was higher for men in the SGPD cohort (OR and 95% CI: men: 2.48 [1.86,
3.34]; women: 1.67 [1.21, 2.33]), and the risk was higher for women in the PEG cohort (OR and 95% CI:
men: 1.49 [1.02, 2.20]; women: 1.81 [1.17, 2.85]).

3.6.4.4 Summary and Causality Determination: Neurodegenerative Diseases

The 2013 Pb ISA (U.S. EPA, 2013) concluded that the available evidence was "inadequate to
determine that a causal relationship exists between Pb exposure and neurodegenerative diseases in
adults." This conclusion was based on a limited number of studies that examined the association of blood
Pb or bone Pb levels with essential tremor, PD, ALS, and AD. These studies were not sufficient to reach a
conclusion regarding the presence or absence of an effect due largely to the potential for reverse causation

3-223

-------
(i.e., reduced physical activity among cases resulting in greater bone turnover and higher BLLs), and
limited consideration for potential confounding factors. Limited studies in monkeys and rodents found
that developmental Pb exposure induced pathologies that underlie AD, and rodent studies suggested
neurophysiologic characteristics and changes related to ALS and PD. Recent epidemiologic studies
expanded the evidence base indicating associations between Pb exposure and some neurodegenerative
diseases among non-occupational cohorts. The strongest evidence in the current review includes well-
designed case-control and cohort studies of ALS and PD, whereas findings from recent epidemiologic
studies of clinically diagnosed AD or AD mortality did not provide strong evidence to address
uncertainties in the body of evidence. Findings from recent toxicological studies, however, add to the
evidence suggesting that developmental exposure to Pb increases the expression of proteins related to
AD, including A|3, tau and p-tau at lower BLLs than the values investigated in the previous ISA (<10
(ig/dL). Alterations in neuropathologic hallmarks of AD in older monkeys were also demonstrated
following developmental Pb exposure.

Studies for ALS in this review included a well-designed cohort study that examined the
relationship of the blood Pb biomarker and ALS survival after ALS diagnosis among U.S. veterans (Fang
et al.. 2017) and a large case-control study that examined associations between past airborne Pb exposures
and ALS risk in participants from a healthcare claims dataset (Andrew et al.. 2022). The findings from the
cohort study by (Fang et al.. 2017) suggested that increased levels of past blood Pb prior to mortality
follow-up were associated with shorter ALS survival even after mutually accounting for bone resorption
and bone formation, thus reducing the uncertainty due to reverse causality. (Andrew et al.. 2022) found
significant positive associations between airborne Pb exposure and increased ALS risk, specifically for
residential Pb exposure over the past 10 years. For PD outcomes, only one case-control study investigated
the relationship between the risk of PD and epigenetic biomarkers by quantifying DNAm tibia and patella
Pb concentrations as cumulative Pb exposure. The authors found an association between DNAm tibia Pb
levels and PD risk (Paul et al.. 2021). This empirically-derived DNA methylation signature is potentially
a more sensitive predictor of the effect of Pb exposure on PD than bone Pb concentration.

Toxicological studies investigating potential associations between Pb exposure and ALS or PD
remain limited. In one study reviewed in the previous ISA, Pb exposure was found to induce
neurophysiologic changes in a rodent model of ALS. Neurophysiologic characteristics of PD, such as
decreased activity of dopaminergic neurons in the substantia nigra and increased expression of
hippocampal alpha-synuclein, have also been demonstrated following Pb exposure.

The bulk of the epidemiologic evidence in the 2013 Pb ISA drawn upon to evaluate the
association of Pb exposure AD focused on cognitive impairment identified using dementia screening
instruments such as the MMSE. Recent studies that examined the association of blood Pb biomarkers
with clinical endpoints of AD risk (Yang et al.. 2018) or AD mortality (Horton et al.. 2019) in non-
occupational cohorts add to the evidence. The association between blood Pb exposure and AD risk
observed in Yang et al. (2018) was imprecise. Uncertainties related to potential reverse causality and

3-224

-------
timing of exposure were not addressed in this study. (Horton et al.. 2019) addressed concerns raised for
the case-control study design used in Yang et al. (2018) but also observed imprecise relationships
between blood Pb and AD mortality risk, (Horton et al.. 2019). For tremor outcomes, no significant
association was observed between blood Pb and the tremor score when all study participants were
analyzed; however, an increase in the tremor score was observed for increasing BLL among younger
participants (Ji et al.. 2015). Studies reviewed for Pb exposure and motor function in adults also provided
inconsistent findings. Overall, the relationships of Pb biomarkers with AD risk, tremor, or motor function
are inconclusive. In contrast to the inconclusive epidemiologic study findings on AD, a large number of
recent toxicological studies add to the evidence which suggests that developmental exposure to Pb
increases the expression of proteins related to AD, including A|3, tau, and p-tau, at lower BLLs than the
values investigated in the previous ISA (<10 (.ig/dL). In older monkeys (23 years), the neuropathologic
hallmarks of AD (i.e., amyloid plaques and tau deposits) were also demonstrated following
developmental Pb exposure.

In summary, the evidence from epidemiologic and experimental animal studies is suggestive
of, but not sufficient to infer, a causal relationship between Pb exposure and neurodegenerative
diseases. This determination reflects a strengthening of the evidence since the 2013 Pb ISA, which found
that the evidence was "inadequate." Recent epidemiologic studies of varying quality strengthen the
evidence for the association of Pb exposure with ALS and PD, and reduce uncertainty related to the
potential for reverse causality by better establishing the temporal association between Pb exposure and
these clinical endpoints. Although recent epidemiologic studies of clinically diagnosed AD and AD
mortality add to the evidence, findings from these studies do not substantially strengthen the evidence
overall. In contrast to the epidemiologic evidence, multiple recent toxicological studies add to the
evidence indicating that developmental exposure to Pb increases the expression of proteins related to AD,
including A|3, tau, and p-tau at lower BLLs than the values investigated in the previous ISA (<10 (.ig/dL).
Alterations in neuropathologic hallmarks of AD in older monkeys were also demonstrated following
developmental Pb exposure. However, toxicological studies investigating potential associations between
Pb exposure and ALS or PD remain limited.

3-225

-------
Table 3-12 Summary of evidence that is suggestive of, but not sufficient to infer, a causal relationship
between Pb exposure and neurodegenerative diseases in adults

Rationale for Causality
Determination3

Key Evidence13

References'3

Pb Biomarker
Levels Associated
with Effects0

ALS and Parkinson's Disease

At least one high-quality
prospective cohort or case-control
study finds associations with ALS
and PD.

A prospective analysis of U.S. veterans found that higher baseline BLL was Fang et al.

associated with increased mortality / shorter survival after ALS diagnosis. The (2017)

association persisted after controlling for confounders including a biomarker of bone
turnover (formation and reabsorption), thus addressing the issue of reverse causality.

Support from recent case-control studies using novel exposure metrics that found Paul et al.
associations between higher estimated air Pb exposure and ALS risk and between a (2021)
DNAm biomarker of tibia Pb and PD.

Limited number of studies address Well-designed case-control studies and prospective studies of ALS and PD assessed
uncertainty due to temporality and exposure prior to disease development and accounted for increased bone turnover
reverse causation. resulting from the disease state.

Alzheimer's Disease

Coherence for AD provided by
consistent evidence in animals
with relevant exposures.

Amyloid plaques and/or tau deposits in transgenic rodents and aged monkeys
following Pb exposure.

Bihaai and Peak BLLs: 19-30
Zawia (2013) pg/dL

Increased expression of A(3, tau, and other AD-related proteins across multiple brain Wu et al.
regions in rodents. (2008)

Gu et al.

(2012)

Section 3.6.4.2

Peak BLLs: 4-58
pg/dL

3-226

-------
Rationale for Causality
Determination3

Key Evidence13

Pb Biomarker
References'3 Levels Associated
with Effects0

Evidence describes biologically
plausible pathways.

Evidence suggests that exposure to Pb results in neuronal cell death associated with Section 3.3
oxidative stress, neuroinflammation and altered energy metabolism, all of which may
underlie general neurodegenerative processes.

AD = Alzheimer's disease; ALS = amyotrophic lateral sclerosis; BLL = blood lead level; DNAm = DNA methylation; Pb = lead; PD =Parkinson's disease.

aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the Preamble to the ISAs CU.S. EPA. 20151.

bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and, where applicable, to uncertainties or

inconsistencies. References to earlier sections indicate where the full body of evidence is described.

°Describes the Pb biomarker levels at which the evidence is substantiated.

3-227

-------
3.7

Evidence Inventories - Data Tables to Summarize Study Details

Table 3-1E Epidemiologic studies of Pb exposure and overt nervous system toxicity

Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Yuan et al. (2006)
Cincinnati, OH

CLS
n: 24

Blood

Age at measurement:
BLL from 3 to 78 mo
averaged

Mean: 14.18

Range: 4.77-31.06
[jg/dL

MRI (subject asked to generate birth weight and
verbs to activate language with marijuana use;
bilateral finger tapping)	consideration of IQ,

sex, SES, gestational
age

Increasing BLL
associated with
decreased brain
activation in the left
frontal gyrus and left
middle temporal gyrus,
regions (semantic
language function)

3-228

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tReuben et al. (2020)

Dunedin
New Zealand
1972-2019
Cohort

Dunedin Study
n: 512

Whole blood Pb (pg/dL)
was measured via
GFAAS

Age at measurement:
11 yr

Mean (SD): 10.99 (4.63)
|jg/dL

Max: 31 |jg/dL

Cortical thickness, cortical
surface area, hippocampal
volume, WMH volumFe,
BrainAGE index

High resolution images
showing cortical thickness,
cortical surface area, bilateral
hippocampal volume, WMH,
and FA were produced using
T1-weighted, fluid-attenuated
inversion recovery and
diffusion-weighted sequences
with a Siemens Skyra 3T
scanner with 64-channel head
and neck coil. BrainAGE index
was calculated as a composite
measure of all measured
indices. Outcomes were
assessed at 45 yr of age.

Sex, maternal IQ,
childhood :

Betas

BrainAGE Index: 0.03
(0.00, 0.06)

Hippocampal Volume:
0.00 (-0.01, 0.00))

Cortical Surface Area:
-0.05 (-0.09, 0.00)

Cortical Thickness

(mm):

0.00 (0.00, 0.00)
WMH:

0.00 (0.00, 0.01)

Age at outcome:

3-229

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Cecil (2011)

Cincinnati, Ohio, Cincinnati
Children's Hospital Medical
Center
United States

Enrollment mothers): 1979-
1984. Follow-up: Birth to 24 y
Cohort

CLS
n: 159

Whole blood Pb
measured using anodic
stripping voltammetry

Age at measurement:
3-78 mo old

Pb prenatally and at
intervals to age 17 yr;
imaging 19-24 yr

Mean (SD) mcg/dL:
prenatal 8.3 (3.7), 3-12
mo 10.6 (5.1).
(Reported in Dietrich et
al. 1993)

MRI brain assessments of 4
types: volumetric (morphology),
spectroscopy (chemical
concentrations), diffusivity
(organization), and functionality
(activation related to tasks).

Brain MRI/fMRI measures of
four types. (1) Volume of gray
matter. (2) Spectroscopy -
metabolites linked to neuronal
function and myelin
architecture: N-acetyl
aspartate, creatine and
phosphocreatine,
phosphocholines and
glycerolphosphocholine, myo-
inositol, glutamate and
glutamine. (3) Diffusivity in
white matter regions reflecting
axonal and myelin effects: FA;
mean, axial and radial
diffusivity. (4) Functionality:
activation related to task
performance.

Age at outcome:

19-24 yr

Varied by outcome;
included age at imaging
and birth weight.

Reported a negative
association between
childhood BLL and gray
matter volume in
several regions: medial
and superior frontal
gyri, inferior parietal
lobule, cerebellar
hemispheres

Reported an
association between
higher childhood BLL
and lower metabolite
concentrations in
several regions: white
matter, left basal
ganglia, left cerebellar
hemisphere, vermis

3-230

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tBeckwith et al. (2021)

Cincinnati, OH
United States

in utero to up to 33 yr of age
Cohort

CLS
n: 123

Pb was measured in
whole blood at 10 d, on
a quarterly basis up to
60 mo, and monthly at
66, 72, and 78 mo.
Samples were collected
mainly by venipuncture,
but occasionally by heel
or finger stick. Pb
concentrations were
quantified using anodic
stripping voltammetry

Age at measurement:
0-78 mo

10 days, on a quarterly
basis up to 60 mo, and
monthly at 66, 72, and
78 mo

MRI brain volumetrics of white
and gray matter, focusing on
regions involved in cognitive
and emotional function.

MRI scans (Voxel based
morphometry) were used to
examine spatial differences in
regional gray and white matter
volumes in adulthood (mean
age 26.8 yr) associated with
childhood blood Pb
concentrations at 78 mo.

Age at outcome:

18-33 yr

Age at time of imaging,
birth weight, total
intracranial volume.

BLLs were associated
with MRI-derived
decreases in white and
gray matter volumes in
the frontal, parietal, and
temporal lobes.
Decreased gray matter
volume in brain regions
responsible for
cognition and emotional
regulation associated
with criminal arrests

Mean (SD) blood Pb at
78 mo: 7.82 pg/dL (4.2)
Max: 24.75 pg/dL

3-231

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tLamoureux-Tremblav et al.
(2021)

Nunavik, Northern Quebec

Canada

Cohort

NCDS
n: 71

Blood

Cord and concurrent
blood; GFAAS with
Zeeman background
correction

Age at measurement:
16-22

Cord blood: median
3.73 |jg/dL, mean 4.56
pg/dL. Adolescent
blood: median 1.52
|jg/dL, mean 1.78 |jg/dL
Max: 17.81 pg/dL

Activation of the human neural
fear circuitry

fMRI analysis of brain
activation in response to a
validated fear conditioning and
extinction stimulus test

Age at outcome:

16-22 yr

Sex, age, SES, and

alcohol/drug

consumption.

Higher differential
activation in the right
dorsolateral prefrontal
cortex in association
with higher postnatal
BLL.

tEthieret al. (2012)

Nunavik, Quebec
Canada
11 yr
Cohort

Prospective 11 yr Blood, Maternal Blood Neurological
study of Inuit
children from
Nunavik
n: 149

Concurrent venous
blood; GFAAS with
Zeeman background
correction (Perkin Elmer
model ZL 4100).

Age at measurement:
Pre-natal and 11 yr

At birth mean: 4.6
pg/dL, SD: 3.1; At 11 yr
mean: 2.6 pg/dL; SD:

2.3

Achromatic pattern-reversal
VEPs with different visual
contrast levels were
administered, using generated
vertical sinusoidal gratings with
a spatial frequency of 2.5
cycles per degree. Children
viewed stimuli binocularly and
were instructed to fixate on a
small red dot. Pattern-reversal
VEPs were recorded from the
scalp over the visual cortex at
Oz derivation according to the
International 10-20 system

Age at outcome:

10-13 yr

Analysis of variance
models controlled for
current Se, cord Se,
and gender.

Betas

N150 Latency
95% Contrast Level:
0.056 (0.099, 0.014)

*Note- 95% CIs were
converted from author
reported p-values

3-232

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tKimetal. (2018a)

n: 150

Seoul
Korea

Enrollment 2010-
Case-Control

¦2015

Blood

Cortical Thickness

Blood Pb was measured Cortical thickness of brain
using atomic absorption regions was ascertained via

Age, intracranial
volume, gender.

spectrometer graphite
furnace

Age at measurement:
6-17 years

Mean (SD) - Cases: 1.3
(0.6) |jg/dL; Controls:
1.5 (0.7) |jg/dL

whole-brain structural MRI.

Age at outcome:

6-17 yr

An interaction between
DRD2 and BLL on the
cortical thickness of the
frontal lobe in the
ADHD group, and a
brain-behavior
correlation between
cortical thickness and
the ADHD-RS
inattention score was
observed.

BLL = blood lead level; BrainAGE = Brain Age Gap Estimation; CI = confidence interval; CLS = Cincinnati lead study; d = day(s); FA = fractional anisotropy; fMRI = functional

magnetic resonance imaging; GFAAS = graphite furnace atomic absorption spectrometry; IQ = intelligence quotient; mo = month(s); MRI = magnetic resonance imaging; Pb = lead;

SD = standard deviation; Se = selenium; SES = socioeconomic status; VEP = visual evoked potential; WMH = white matter hyperintensities; yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a

change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

bResults are not standardized (e.g., BLL distribution data needed to calculate the standardized estimate was not reported or categorical data was analyzed).

tStudies published since the 2013 Integrated Science Assessment for Lead.

Table 3-1T Animal toxicological studies of Pb exposure and brain function

Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Graham et al. (2011)

Rat (Sprague Dawley) PND 4 to PND 28, Oral, gavage PND 29:

Control (vehicle), M/F, n every other day

= 4-8	0.289 pg/dL for Control

1 mg/kg, M/F, n = 4-8
10 mg/kg, M/F, n = 4-E

3.27 pg/dL for 1 mg/kg
12.6 pg/dL for 10 mg/kg

PND 11, 19, 29: Neurotransmitter

3-233

-------
Study	Species (Stock/Strain), TimingoJ	Exposurc BLL as Reported (Mg/dL)	Endpoints Examined

Liu et al. (2012)	Rat (Sprague Dawley) PND24toPND80 Oral, drinking PND21:	PND 56: Electrophysiology

Control (tap water), M, n	water

= 20	15 |jg/L (1.5 |jg/dL) for Control

100 ppm, M, n = 20	45 |jg/L (4.5 |jg/dL) for 100

ppm

PND 28:

14 |jg/L (1.4 |jg/dL) for Control

94 |jg/L (9.4 pg/dL) for 100
ppm

PND 35:

16 |jg/L (1.6 pg/dL) for Control

103 |jg/L (10.3 pg/dL) for 100
ppm

PND 42:

13	|jg/L (1.3 pg/dL) for Control

94 |jg/L (9.4 pg/dL) for 100
ppm

PND 49:

14	|jg/L (1.4 pg/dL) for Control

98 |jg/L (9.8 pg/dL) for 100
ppm

3-234

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Corv-Slechta et al. (2012) Rat (Long-Evans) GD-60 to 10 mo Oral, drinking PND 5-6
Control (tap water), M, n	water

= 12	Oral,

lactation

50 ppm Pb, M, n = 12	In utero

10-11 mo: Neurotransmitter

<5 pg/dL for Control
12.5 pg/dL for 50 ppm
2.5 mo:

<5 pg/dL for Control
6.43 pg/dL for 50 ppm
10 mo:

<5 pg/dL for Control
8.98 pg/dL for 50 ppm

Weston et al. (2014)

Rat (Long-Evans) GD -60 to PND 21

Oral,

PND 5-6 - Males:

PND 60: Neurotransmitter



Control (tap water), M/F,

lactation







n = 18-22 (9-11/9-11)

In utero

0.76 pg/dL for Control





50 ppm, M/F, n = 18-22



15.7 pg/dL for 50 ppm





(9-11/9-11)













PND 5-6 Females:









0.82 pg/dL for Control









14.7 pg/dL for 50 ppm



3-235

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Han et al. (2014)

Rat (Wistar)	PW group: PND 21 Oral, drinking PND21:

Control (tap water), M, n to PND 42	water

= 8	Oral,

ME group: GD -21	lactation

2 mM - postweaning to PND 20	In utero
(PW), M, n = 8

PND 68: Histopathology

7.36 |jg/L (0.74 pg/dL) for
Control

2 mM - ME, M, n = 8

NR for 2 mM - PW

146.6 pg/L (14.7 pg/dL) for 2
mM-ME

PND 63:

9.22 pg/L (0.92 pg/dL) for
Control

147.9 pg/L (14.8 pg/dL) for 2
mM-PW

46.13 pg/L (4.6 pg/dL) for 2
mM-ME

Barkur and Bairv (2015a) Rat (Wistar)

Pregestation

Control (untreated), M/F, exposure - GD -30

n = 9

to GD 0
Lactation only

0.2% solution -
Pregestation, M/F, n = 9 exposure - PND 0
to PND 22

0.2% solution -
Lactation, M/F, n = 9

0.2% solution -
Gestation, M/F, n = 9

0.2% solution -
Gestation and Lactation,
M/F, n = 9

Gestational
exposure - GD 0 to
GD 20

Gestation and
Lactation exposure
-GD Oto PND 22

In utero PND 22:

0.19 pg/dL for Control

3.04 pg/dL for 0.2%
Pregestation

5.26 pg/dL for 0.2% Gestation

26.8	pg/dL for 0.2% Lactation

31.9	pg/dL for 0.2% Gestation
and Lactation

PND 30: Brain Weight

3-236

-------
Study	Species (Stock/Strain), TimingoJ	Exposurc BLL as Reported (Mg/dL)	Endpoints Examined

Bashaetal. (2014)

Rat (Not Specified)

PND 1 to PND21

Oral,

PND 45:

PND 45, 4 mo, 12 mo, 18 mo:



Control (deionized



lactation



Neurotransmitter Analysis



water), M, n = 6





0.42 |jg/dL for Control





0.2% solution, M, n = 6





49.5 |jg/dL for 0.2% solution











4 mo:











0.56 |jg/dL for Control











14.4 |jg/dL for 0.2% solution











12 mo:











0.46 |jg/dL for Control











6.96 |jg/dL for 0.2% solution











18 mo:











0.12 |jg/dL for Control











11.2 |jg/dL for 0.2% solution



Rahman et al. (2012b)

Rat (Wistar)

PND 1 to PND 30

Oral, drinking

PND 21:

PND 21, 30: Brain Weight,



Control (tap water), M/F,



water



Histopathology



n = 4-10



Oral,

1.4 |jg/dL for Control









lactation







0.2% solution, M/F, n =





12.1 |jg/dL for 0.2% solution





4-10

















PND 30:



1.2 |jg/dL for Control
12.8 |jg/dL for 0.2% solution

3-237

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Nam et al. (2019b)

Rat (Sprague Dawley) GD 0 to 22

Control (not specified),

M/F, n = 12

0.2 % Solution, M/F, n =

12

In utero PND21:

0.64 pg/dL for Control
17.30 pg/dL for 0.2% solution

PND21: Histopathology

Saleh et al. (2018)

Rat (Sprague Dawley) GD 1 to GD 20

Control (deionized
water), M/F, n = 8 litters

160 ppm, M/F, n = 8
litters

In utero Maternal Blood Pb GD 20:
5.1 pg/dL for Control
27.7 pg/dL for 160 ppm

GD 20: Brain Weight, Histopathology

Menq et al. (2016)

Rat (Sprague Dawley) PND 0 to PND 21

Control (deionized
water), M/F, n = 7

300 ppm, M/F, n = 7

Oral,
lactation

PND 35:

7.61 |jg/L (0.76 pg/dL) for
Control

84.3 pg/L (8.43 pg/dL) for 300
ppm

NR: Histopathology

Amos-Kroohs et al.
(2016)

Rat (Sprague Dawley) PND 4 to PND 28

Control (sodium
acetate), M/F, n = 16
(8/8) per time point

1 mg/kg Pb, M/F, n = 16
(8/8) per time point

10 mg/kg Pb, M/F, n =

16 (8/8) per time point

Oral, gavage PND 29:

1.27 pg/dL for Control
2.76 pg/dL for 1 mg/kg
9.07 pg/dL for 10 mg/kg

PND 29: Neurotransmitter

3-238

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Rahman et al. (2018)

Rat (Wistar)

Control (tap water), M/F,
n = 20

0.2% solution, M/F, n =
20

PND 1 to PND 21 Oral, drinking PND 21:

PND 21, 30: Brain Weight

water

Oral,

lactation

2.2	pg/dL for Control

12.4 pg/dL for 0.2% solution
PND 30:

3.3	pg/dL for Control

22.7 pg/dL for 0.2% solution

Baranowska-Bosiacka et Rat (Wistar)	GD 0 to PND 21 Oral,	PND 28:

al. (2017)	Control (distilled water),	lactation

M/F, n = 8	In utero 0.05 pg/dL for Control

0.1% solution, M/F, n =

6.90 pg/dL for 0.1 % solution

PND 28: Histopathology

Baranowska-Bosiacka et Rat (Wistar)
al. (2013)	Control (distilled water),

M/F, n = 36 (17/19)

0.1% solution, M/F, n =
36 (18/18)

GD Oto PND 21

Oral,
lactation
In utero

PND 28:

0.93 pg/dL for Control
6.86 pg/dL for 0.1 % solution

PND 28: Histopathology

Wang et al. (2013)

Rat (Sprague Dawley) GD 0 to PND 1,

Control (untreated), M/F, PND 1 to PND 21,
n = 6	PND 21 to 42

0.2% Pb (w/v), M/F, n =

6 - Gestational
Exposure

0.2% Pb (w/v), M/F, n =

6 - Lactational Exposure

0.2% Pb (w/v), M/F, n =

6 - Ablactational
Exposure

Oral, drinking PND 72:

water
Oral,
lactation
In utero

PND 72: Brain Weight

34.99 |jg/L (3.5 pg/dL) for
Control

35.78 pg/L (3.58 pg/dL) for 0.2
% solution Gestational

65.97 pg/L (6.60 pg/dL) for
0.2% solution Lactational

110.67 pg/L (11.07 pg/dL) for
0.2% solution Ablactational

3-239

-------
Study	Species (Stock/Strain), TimingoJ	Exposurc BLL as Reported (Mg/dL)	Endpoints Examined

Wang et al. (2016)	Rat (Sprague Dawley) PND24toPND56 Oral, drinking PND 56:	PND 60-66: LTP, Neuronal Morphology

Control (tap water), M, n	water

= 7	11 |jg/L (1.1 |jg/dL) for Control

100 ppm, M, n = 9	133 |jg/L (13.3 |jg/dL) for 100

ppm

Li et al. (2016b)

Mouse (Kunming) GD0toPND21
Control (deionized
water), M/F, n = 10

0.1% (1000 ppm), M/F, n
= 10

0.2% (2000 ppm), M/F, n
= 10

0.5% (5000 ppm), M/F, n
= 10

Oral,
lactation
In utero

PND 21:

8.27 |jg/L (0.827 pg/dL) for
Control

41.05 ug/L (4.11 ug/dL) for
0.1% solution

82.93 ug/L (8.29 ug/dL) for
0.2% solution

105.33 |jg/L (10.53 pg/dL) for
0.5% solution

PND 21: Histopathology

Sobolewski et al. (2018)

Mouse (C57BL/6) GD -60 to PND 21

Oral,

PND 6-7:

PND 60: Epigenetics



Control (deionized

lactation







water), M/F, n = 6-12

In utero

0.37 pg/dL for Control





100 ppm, M/F, n = 6-12



10.2 pg/dL for 100 ppm



3-240

-------
Study	Species (Stock/Strain), TimingoJ	Exposurc BLL as Reported (Mg/dL)	Endpoints Examined

Barkur and Bairv (2016)

Rat (Wistar)

GD -30 to PND21

Oral,

PND 22:

PND 30: Histopathology



Control (tap water with



lactation







acetic acid), M/F, n = 8



In utero

0.5 |jg/dL for Control





0.2% solution,





9.4 |jg/dL for 0.2% solution,





pregestation only (PG),





pregestation only (PG)





M/F, n = 8

















16.6 |jg/dL for 0.2% solution,





0.2% solution, gestation





gestation only





only, M/F, n = 8

















30.1 |jg/dL for 0.2% solution,





0.2% solution, lactation,





lactation





M/F, n = 8

















33.4 |jg/dL for 0.2% solution,





0.2% solution, gestation





gestation and lactation





and lactation, M/F, n = 8









Shvachiv et al. (2018)

Rat (Wistar)

Intermittent

Oral, drinking

PND 196:

PND 189: Brain Histopathology



Control (tap water), M/F,

Exposure: GD 7 to

water







n = 8

PND 84, PND 140

Oral,

<0.1 |jg/dL for Control







to PND 196

lactation







0.2% (p/v) solution



In utero

18.8 |jg/dL for 0.2%





(distilled water), M/F, n =

Continuous



(Intermittent)





9 - Intermittent exposure

Exposure: GD 7 to











PND 196



24.4 |jg/dL for 0.2%





0.2% (p/v) solution, M/F,





(Continuous)





n = 9 - Continuous











exposure









Stansfield et al. (2015)

Rat (Long-Evans)

GD Oto PND 50

Oral, diet

PND 50:

PND 50: Neurotransmitter Analysis,



Control (chow), M/F, n =



Oral,



Brain Histopathology



4-7



lactation

0.6 |jg/dL for Control









In utero







1500 ppm, M/F, n = 4-7





22.2 |jg/dL for 1500 ppm



Listos et al. (2013)

Rat (Wistar)

GD Oto PND 28

Oral, drinking

PND 60:

PND 60: Neurotransmitter



Control (tap water), M/F,



water







n = 6-11



Oral,

0.93 |jg/dL for Control









lactation







0.1% solution, M/F, n =



In utero

20.45 |jg/dL for 0.1% solution





6-11









3-241

-------
Zhao etal. (2018)

Rat (Sprague Dawley) GD -14 to PND 10 Oral,
Control (tap water), M, n	lactation

= 8	In utero

0.005% solution, M, n =
8

0.01% solution, M, n = 8
0.02% solution, M, n = 8

3-242

PND 0:

1.9 |jg/dL for Control
17.9 |jg/dL for 0.005% solution

23.2	|jg/dL for 0.01% solution
48.8 |jg/dL for 0.02% solution
PND 3:

I.9	|jg/dL for Control

6.7 |jg/dL for 0.005% solution

II.5	|jg/dL for 0.01% solution
23.1 |jg/dL for 0.02% solution
PND 7:

1.3 |jg/dL for Control

8.1	|jg/dL for 0.005% solution

12.3	|jg/dL for 0.01 % solution
18.7 |jg/dL for 0.02% solution
PND 10:

1.2	|jg/dL for Control

5.6 |jg/dL for 0.005% solution
7.0 |jg/dL for 0.01% solution
12.3 |jg/dL for 0.02% solution
PND 14:

0.7 |jg/dL for Control

PND 30: Electrophysiology,
Histopathology

-------
Study	Species (Stock/Strain), TimingoJ	Exposurc BLL as Reported (Mg/dL)	Endpoints Examined

4.0	|jg/dL for 0.005% solution

5.5 |jg/dL for 0.01% solution

8.9 |jg/dL for 0.02% solution
PND 21:

1.1	|jg/dL for Control

2.5 |jg/dL for 0.005% solution
2.5 |jg/dL for 0.01% solution
2.98 |jg/dL for 0.02% solution
PND 30:

1.5 |jg/dL for Control
1.0 |jg/dL for 0.005% solution
1.5 |jg/dL for 0.01% solution
1.5 |jg/dL for 0.02% solution

PND 28 - Females:	PND 28: Histopathology

0.02 |jg/dL for Control
3.03 |jg/dL for 30 ppm
12.79 |jg/dL for 330 ppm
PND 28 - Males:

0.03 |jg/dL for Control
3.68 |jg/dL for 30 ppm
15.42 |jg/dL for 330 ppm

Dominquez et al. (2019) Mouse (C57BL/6)	PND 0 to PND 28 Oral,

Control (tap water), M/F,	lactation

n = 10 (7/3)

30 ppm, M/F, n = 10
(6/4)

330 ppm, M/F, n = 10
(4/6)

3-243

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Du etal. (2015)

Rat (Sprague Dawley) PND 0 to PND 90

Control (distilled water),

M/F, n = 8

250 ppm, M/F, n = 8

Oral, drinking PND 30:
water

Oral,	13.9 pg/L (1.4 pg/dL) for

lactation	Control

205.6 pg/L (20.6 pg/dL) for 250
ppm

PND 60:

15.0 pg/L (1.5 pg/dL) for
Control

321.9 pg/L (32.2 pg/dL) for 250
ppm

PND 90:

11.8 pg/L (1.2 pg/dL) for
Control

379.2 pg/L (37.9 pg/dL) for 250
ppm

PND 30, 60, 90: Histopathology

Mansouri et al. (2013)

Rat (Wistar)	PND 55 to PND

Control (tap water or 181
water + NaAc), M/F, n =

16 (8/8)

50 ppm, M/F, n = 16
(8/8)

Oral, drinking PND 178-181 - Females:
water

NR for Control
10.6 pg/dL for 50 ppm
PND 178-181 - Males:
NR for Control
18.9 pg/dL for 50 ppm

PND 161-179: Neurotransmitter

3-244

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Zhou etal. (2018)

Rat (Sprague Dawley) PND 24 to PND 52 Oral, drinking PND 52:
Control (distilled water),	water

M, n = 10

0.5% solution, M, n = 10
1.0% solution, M, n = 10
2.0% solution, M, n = 10

13.3 |jg/L (1.3 pg/dL) for
Control

148.9 |jg/L (14.9 pg/dL) for
0.5% solution

231.3 pg/L (23.1 pg/dL) for
1.0% solution

PND 24, 31, 38, 45, 52: Brain Weight,
Brain Histopathology

293.4 ug/L (29.3 pg/dL) for
2.0% solution

Dumkova et al. (2017) Mouse (ICR)	NR(24g)-6wk Inhalation After 6 wk treatment:	After 6 wk treatment: Histopathology

Control, F, n = 10	continuous

exposure	11 ng/g (1.16 pg/dL) for

106/cm3Pb0	Control

nanoparticles, F, n = 10

132 ng/g (13.99 pg/dL) for
106/cm3

3-245

-------
Study

Species (Stock/Strain), Timing of	Exposure

n, Sex	Exposure	Details

BLL as Reported (pg/dL)

Endpoints Examined

Xiao etal. (2014)

Rat (Wistar)	Pre-weaning: GD

Control (tap water), M/F, -21 to PND 21
n = 10 (5/5)

Postweaning: PND
Pre-weaning: 2 mM 21 to PND 84
solution, M/F, n = 10
(5/5)

Postweaning: 2 mM
solution, M/F, n = 10
(5/5)

Oral, drinking PND 21 - Pre-weaning:
water

10.09 |jg/L(1 pg/dL) for
Control

Oral,
lactation
In utero

PND 84 and PND 91: Histopathology

103.8 |jg/L (10.4 pg/dL) for 2
mM solution

PND 21 - Postweaning:

Not Reported

PND 91 - Pre-weaning:

10.32 pg/L (1 pg/dL) for
Control

39.27 pg/L (3.9 pg/dL) for 2
mM solution

PND 91 - Postweaning:

10.32 pg/L (1 pg/dL) for
Control

105.45 pg/L (10.5 pg/dL) for 2
mM solution

Sobin etal. (2013)

Mouse (C57BL/6)	PND 1 to PND 28

Control (tap water), M/F,
n = 30

30 ppm, M/F, n = 30
230 ppm, M/F, n = 30
330 ppm, M/F, n = 30

Oral,
lactation

PND 28:

0.22 pg/dL for Control
4.12 pg/dL for 30 ppm
10.31 pg/dL for 230 ppm
13.84 pg/dL for 330 ppm

PND 28: Histopathology

3-246

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Sobolewski et al. (2020) Mouse (C57BL/6)

F0:

Control (deionized
water), F, n = 10

100 ppm, F, n = 10

F1:

see Figure 1, n = 12

F1: GD -60 to PND Oral,
23-27	lactation

In utero

F1 PND 6-7:

0 pg/dL for Control

12.5 pg/dL for 100 ppm (F0
dosing)

F3 PND 6-7:

0 ng/dL for Control

0 pg/dL for 100 ppm (F0
dosing)

PND 60-120 (variable by endpoint):
Neurotransmitter, Epigenetics

F2:

see Figure 1, n = 12
F3:

see Figure 1, n = 8-10

Ouvanq et al. (2019)

Rat (Sprague Dawley) GD0toPND679 Oral, drinking wk97:
Control (tap water), M/F,
n = 6-10

PND 679: Histopathology

0.05/0.01% solution,
M/F, n = 6-10

water
Oral,
lactation
In utero

0 mg/L (0 pg/dL) for Control

0.216 mg/L (21.6 pg/dL) for
0.05/0.01% solution

Saleh et al. (2019)

Rat (Sprague Dawley) NR (190-220g) -
Control (deionized	20 d of treatment

water), F, n = 8

Oral, drinking After 20 d treatment:
water

5.4 pg/dL for Control

After 20 d treatment: Brain Weight
Histopathology

160 ppm, F, n = 8

23.8 pg/dL for 160 ppm

3-247

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Singh etal. (2019)

Rat (Wistar)

Control (distilled water),
M, n = 5

2.5 mg/kg, M, n = 5

3 mo to 6 mo

Oral, gavage

6 mo:

5.76 pg/dL for Control
28.4 pg/dL for 2.5 mg/kg

6 mo: Brain Weight Brain
Histopathology

Xiao et al. (2020)

Rat (Sprague Dawley)

Control (tap water), F, n
= 10

125 ppm, F, n = 10

GD -7 to PND 68 Oral, drinking PND 68:
water
Oral,
lactation
In utero

PND 22, 68: Histopathology

24.23 ng/mL (2.4 pg/dL) for
Control

205 ng/mL (20.5 pg/dL) for 125
ppm

Sun et al. (2014)

Rat (Sprague Dawley)

Control (tap water), NR,
n = 20

580 ppm, NR, n = 20

NR (230-260 g) — 3 Oral, drinking
mo of treatment water

After 3 mo treatment:	After 3 mo treatment: Histopathology

3.0 pg/L (0.3 pg/dL) for Control

56.8 pg/L (5.7 pg/dL) for 580
ppm

Su etal. (2016)

Rat (Sprague Dawley)

Control (deionized water
with 0.9% saline), M, n =
4

200 ppm, M, n = 4

PND 20 to PND 76 Oral, gavage PND 76:

PND 76: Histopathology

7.99 pg/L (0.8 pg/dL) for
Control

84.17 pg/L (8.4 pg/dL) for 200
ppm

Song etal. (2014)

Rat (Sprague Dawley)

Control (tap water), M, n
= 9

100 pg/mL, M, n = 9
200 pg/mL, M, n = 9
300 pg/mL, M, n = 9

PND 20-22 to PND Oral, drinking
76-78	water

PND 76-78:

0.73 pg/dL for Control
4.7 pg/dL for 100 pg/mL
10.1 pg/dL for 200 pg/mL
12.3 pg/dL for 300 pg/mL

PND 76-78: Histopathology

3-248

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Zhou et al. (2020a)

Rat (Sprague Dawley) GD 1 to PND 364

Control (distilled water),

M, n = 5-11

0.5 g/L solution, M, n =

5-11

2.0 g/L solution, M, n =

5-

Oral, drinking PND 21:
water

0 mg/L (0 pg/dL) for Control

0.1 mg/L (10 pg/dL) for 0.5 g/L
solution

0.36 mg/L (36 pg/dL) for 2.0
g/L solution

PND 364:

0 mg/L (0 pg/dL) for Control

0.15 mg/L (15 pg/dL) for 0.5
g/L solution

0.51 mg/L (51 pg/dL) for 2.0
g/L solution

PND 21, 364: Histopathology,
Electrophysiology

Liu et al. (2019)

Rat (Sprague Dawley) PND 1 to PND 21

Control (tap water), F, n
= 12

58 mg/L, F, n = 11

Oral,
lactation

PND 9:

0 pg/dL for Control
7.9 pg/dL for 58 mg/L
PND 21:

0 pg/dL for Control,
8.2 pg/dL for 58 mg/L
PND 40:

0 pg/dL for Control
0 pg/dL for 58 mg/L

PND 93: Histopathology

3-249

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Nan et al. (2016)

Mouse (C57BL/6)

Control (sterile water),
NR, n = 30

9.6 mmol/L, NR, n = 30

PND 21 to PND 56

Oral, drinking
water

PND 7:

0 |jg/L (0 pg/dL) for Control

106.3 |jg/L (10.6 pg/dL) for 9.6
mmol/L

PND 14:

0 |jg/L (0 pg/dL) for Control

293.2 |jg/L (29.3 pg/dL) for 9.6
mmol/L

PND 35:

0 |jg/L (0 pg/dL) for Control

959.6 |jg/L (96 pg/dL) for 9.6
mmol/L

PND 56: Histopathology

Singh et al. (2017)

Rat (Wistar)	NR (160-200 g) -

Control (distilled water),	14 days of

M, n = 3-6	treatment

7.5 mg/kg, M, n = 3-6

Oral, gavage 12 hr after last treatment:
5.54 pg/dL for Control
30.28 pg/dL for 7.5 mg/kg

12 hr after last treatment: Brain Weight,
Histopathology

Biioor et al. (2012)

Rat (Wistar)

Control (deionized
water), M/F, n = 10

50 ppm, M/F, n = 10

GD Oto PND 45

Oral, drinking
water
Oral,
lactation
In utero

PND 45:

4.06 pg/dL for Control
10.65 pg/dL for 50 ppm

PND 45: Neurotransmitter

3-250

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Wanq et al. (2021a)

Rat (Sprague Dawley)

GD Oto PND21

Oral,

PND 21:

PND 21: Histopathology



Control (deionized



lactation







water), M, n = 3



In utero

23.1 |jg/L (2.31 pg/dL) for











Control





0.05% solution, M, n = 3

















248 pg/L (24.8 pg/dL) for





0.1% solution, M, n = 3





0.05% solution











302 pg/L (30.2 pg/dL) for 0.1%











solution











361 pg/L (36.1 pg/dL) for 0.2%











solution



Liu et al. (2022c)

Rat (Sprague Dawley)

PND 35 to PND

Oral, drinking

PND 119:

PND 119: Histopathology



Control (tap water), M, n

119

water







= 10





10.9 pg/L (1.09 pg/dL) for











Control





0.2% solution, M, n = 10

















176 pg/L (17.6 pg/dL) for 0.2%











solution



Hsu et al. (2021)

Rat (Sprague Dawley)

PND 42 to PND 77

Oral, drinking

PND 84:

PND 78 to PND 84: Electrophysiology



Control (deionized



water







water), M, n = 6





0.9 pg/L (0.09 pg/dL) for











Control





250 ppm, M, n = 6

















15.3 pg/L (1.53 pg/dL) for 250











ppm



Sadeqhi et al. (2021)

Rat (Wistar)

GD Oto PND 50

Oral, drinking

PND 50:

PND 50: Histopathology



Control (untreated), M, n



water







= 5



Oral,

0.58 pg/dL for Control









lactation







1500 ppm, M, n = 5



In utero

3.4 pg/dL for 1500 ppm



Viaueras-Villasenor et al.

Rat (Wistar)

GD 0 to PND 21

Oral,

PND 110:

PND 90 to PND 110: Histopathology

(2021)

Control (tap water), M, n



lactation







= 20



In utero

2.04 pg/dL for Control





320 ppm, M, n = 20





26.3 pg/dL for 320 ppm



3-251

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

^etaMs'0 BLL as RePorted (MQ/dL)

Endpoints Examined

Long et al. (2022)

Rat (Sprague Dawley) 6 wk to 18 wk

Control (untreated), M, n
= 12

200 mg/L solution, M, n
= 12

Oral, drinking 18wk:
water

2.14 [jg/L (0.214 pg/dL) for
Control

32.48 |jg/L (3.25 pg/dL) for 200
mg/L solution

NR: Histopathology, Neurotransmitter

Abazvan et al. (2014)

Mouse (CAMKII-tTA;
heterozygous or
homozygous for
mDISCI)

Control (het), M/F, n =
5-10

Control (mutant), M/F, n
= 5-10

1500 ppm (het), M/F, n =
5-10

1500 ppm (mutant), M/F,
n = 5-10

GD Oto PND 180

Oral, diet 6 mo - Females:

Oral,

lactation, in 0.6 pg/dL for Control (het)
utero

0.8 pg/dL for Control (mutant)

34.9 pg/dL for 1500 ppm (het)

33.3 pg/dL for ppm (mutant)

6 mo - Males:

1.1 pg/dL for Control (het)

1.1 pg/dL for Control (mutant)

26.1 pg/dL for 1500 ppm (het)

25.0 pg/dL for 1500 ppm
(mutant)

PND 180: Brain Volume, Brain MRI,
Morphometric Measurements in several
regions

3-252

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

^etaMs'0 BLL as RePorted (MQ/dL)

Endpoints Examined

Zhu et al. (2013)

Rat (Sprague Dawley) GD 0 to PND 490

Control (untreated), M, n
= 11-13

510 mg/L, M, n = 11-13

Oral, drinking PND 21

0 |jg/L for control

water
Oral,

lactation, in
utero

0.27 mg/L (27 pg/dL) for 510
mg/L

PND 287
0 |jg/L for control

0.24 mg/L (24 pg/dL) for 510
mg/L

PND 490
0 |jg/L for control

PND 21 to PND 490: Histopathology

0.25 mg/L (25 pg/dL) for 510
mg/L

Nam et al. (2018a)

Rat (Sprague Dawley) GD 0 to PND 21

Control (distilled water),

M/F, n = 12

0.2% solution M/F, n =

12

Oral, drinking PND 21

1.28 pg/dL for control

PND 21: Histopathology

water
Oral

lactation, in
utero

12.67 pg/dL for 0.2% solution

Gassowska et al. (2016a) Rat (Wistar)

Control (drinking water),
M/F, n = 4-8

0.1% solution M/F, n =
4-8

GD 0 to PND 28 Oral, drinking PND 28

0.93 pg/dL for control

PND 28: Histopathology

water
Oral

lactation, in
utero

6.86 pg/dL for 0.1%

3-253

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Gassowska et al. (2016b) Rat (Wistar)	GD0toPND28

Control (drinking water),

M/F, n = 4-8

0.1% solution, M/F, n =

4-8

Oral, drinking PND 28

0.93 pg/dL for control

PND 28: Histopathology

water
Oral

lactation, in
utero

6.86 pg/dL for 0.1%solution

Sepehri and Ganii (2016) Rat (Wistar)

Control, M, n = 8

0.05% solution, M, n = i

GD 5 to PND 25 Oral, drinking PND 25

water	0.78 pg/dL for control

Oral

lactation, in 28.3 pg/dL for 0.05%
utero

PND 25: Histopathology

Zhu et al. (2019a)

Rat (Sprague Dawley) GD -10 to 12 mo

Control (distilled water),

M, n = 10

0.5 g/L, M, n = 10

Oral, drinking	12 mo

water	0 pg/dL for control
Oral

lactation, in	0.27 mg/L (27 pg/dL) for 0.5

utero	g/L

12 mo: Electrophysiology

Zhanq et al. (2015b)

Rat (Long-Evans)

GD-10 to PND 50 Oral, diet

PND 50

PND 50: Histopathology,



Control (0 ppm), M/F, n

Oral

0.8 pg/dL for control

Electrophysiology



= 10

lactation, in
utero

21.1 pg/dL for 1500 ppm





1500 ppm, M/F, n = 10







Wang, 2021,

Rat (Sprague Dawley)

GD-28 to PND 21 Oral,

PND 21:



10296633@@author-

Control (deionized

lactation





year

water), M/F, n = 12

In utero

23.9 pg/L (2.39 pg/dL) for







Control





0.05% solution, M/F, n =









10



206 pg/L (20.6 pg/dL) for
0.05% solution



3-254

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL) Endpoints Examined

Wu et al. (2020a)

Mouse (C57BL/6J)

GD -7 to 7 mo

Oral, diet

PND 207-210 7 mo: Brain Weight, Histopathology



Control (ultra-pure
water), F, n = 8



Oral

lactation, in
utero

Mouse (C57BL/6J)

19.71 |jg/L (1.97 pg/dL) for
control



200 mg/L, F, n = 8





84.53 pg/L (8.45 pg/dL) for 200



Mouse (APP/PS1)





mg/L



Control (ultra-pure
water),F, n = 8





Mouse (APP/PS1)

19.96 pg/L(1.99 pg/dL) for
control



200 mg/L, F, n = 8





205.49 pg/L(20.54 pg/dL) for
200 mg/L

Mani et al. (2020)	Rat (Wistar)	8 mo to 9 mo	Oral, gavage NR	NR: Histopathology, Brain Weight

Control (distiller water),	2.3 pg/dL for Control

M, n = NR

8.5 |jg/dL for 10 mg/kg

10 mg/kg, M, n = NR

16.4 |jg/dL for 50 mg/kg

50 mg/kg, M, n = NR

16.3 |jg/d for 100 mg/kg

100 mg/kg, M, n = NR

3-255

-------
Study

Species (Stock/Strain), Timing of
n, Sex	Exposure

^etaMs'0 BLL as RePorted (MQ/dL)

Endpoints Examined

Shvachiv et al. (2020)

Rat (Wistar)	GD 7 to 3 mo

Control (tap water), M/F,

GD 7 to 5 mo

n = 12

0.2% solution (3 mo)
M/F, n = 12

0.2% solution (5 mo)
M/F, n = 12

0.2% solution (7 mo)
M/F, n = 12

GD 7 to 7 mo

Oral, drinking 3 mo:
water

Oral,
lactation

In utero

<1 for Control

24.0 |jg/dL for 0.2% solution
5 mo:

<1 for Control

24.8 |jg/dL for 0.2% solution

PND 189: Histopathology

7 mo:

<1 for Control

26.9 |jg/dL for 0.2% solution

3-256

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Zhao etal. (2021)

Rat (Sprague Dawley) GD-14toPND10 Oral, drinking PND 10

water

Oral,
lactation

In utero

Control, M, n = 6
109 ppm, M, n = 6

0.6 pg/dL for control
11.4 pg/dL for 109 ppm

PND 21

0.85 pg/dL for control
3.5 pg/dL for 109 ppm

PND 30

0.98 pg/dL for control

PND 30: Histopathology,
Electrophysiology

1.8 pg/dL for 109 ppm

APP = amyloid precursor protein; BLL = blood lead level; Dl = deionized; F = female; F0 = gestating female; GD = gestational day; LTP = long-term potentiation; M = male; ME =
maternal exposure; MRI = magnetic resonance imaging; mo = month(s); NaAc = sodium acetate; NR = not reported; Pb = lead; PG = pregestation; PND = postnatal day; PW =
postweaning; wk = week(s); yr = year(s).

3-257

-------
Table 3-2E Epidemiologic studies of Pb exposure and full-scale intelligence quotient

Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Lanphear et al. (2005)
Lanphear et al. (2019)

International pooled
analysis: Prospective
cohorts from Boston,
Cincinnati, Cleveland,
Mexico City, Port Pirie,
Rochester, and
Yugoslavia.

Followed from birth
(1979-1995) up to age 10
yr

n = 1,333 children

Blood

Median (5th-95th)

Early childhood (6-24
mo):

12.7 (3.5-34.5)

Peak: 18 (6.0-47.0)

Lifetime avg (through
outcome measurement at
4.8-10 yr):

11.9 (3.6-34.5)
Concurrent: 9.7 (2.5-
33.2)

FSIQ: WISC-III, WISC-R, HOME score, birth weight, Early Childhood:

WPPSI, WISC-S

maternal IQ and

(depending on the cohort) education. Also

Ages 4.8-10 yr

considered potential
confounding by child sex,
birth order, marital status,
maternal age, prenatal
smoking status and
alcohol use.

-0.137 (-0.209, -0.064)

Lifetime avg:

-0.206 (-0.285, -0.126)

Concurrent:

-0.187 (-0.26, -0.114)

Peak:

-0.126 (-0.182, -0.071)

Lanphear et al. (2005)
Lanphear et al. (2019).
subset of with peak BLLs
<7.5 |jg/dL

n = 103 children

Same

Concurrent
Mean: 3.2

Same

Same

-2.53 (-4.48, -0.58)

3-258

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

al

Crump et al. (2013)	N = 1355 children (note Blood

that the reanalysis

„ ,	, x included prenatal BLLs in

Reanalysis ofLanphearet |jfetj	d , .

12019, 2005)	a different strategy P '

including covariates that

allowed a larger number

of observations to be

included in the analysis)

FSIQ: WISC-III, WISC-R,
WPPSI, WISC-S

Same but defined to be
cohort specific

BLL distribution reported
by cohort not for the
pooled dataset as a
whole.

Concurrent: (3= -3.315
(-4.546, -2.084)

Peak

(3= -2.484 (-3.825,
-1.142)

Early childhood
(3= -2.459 (-3.817,
-1.102)

Lifetime avg
(3= -3.246 (-4.659,
-1.833)

24-month

(3= -1.955 (-3.193,
-0.717)

Note: the estimates are
not standardized)

Van Landinaham et al. NR
(2020)

Reanalysis of Lanphear et
al. (2005) and Lanphear
et al. (2019)

HOME x In (BLL+1):
0.0437

NR	FSIQ: WISC-III, WISC-R, Defined highly likely B=-4.945

WPPSI, WISC-S	confounders: HOME Interaction terms'

score, maternal education ,, x
and maternal IQ and Maternal IQ x ""(BLL+1)
included interaction terms -0.0003
between BLL and each of Mother's education x
these covariates	inmi i+-iv .n rrai

Note: the estimates are
not standardized)

Canfield et al. (2003a)

n = 101

Blood

FSIQ

Child sex, Fe status, birth -1.8 (-3.0, -0.60)

Rochester, NY

Children recruited from



Stanford-Binet

weight, maternal race,

Prospective cohort

dust control study

Concurrent, children with
peak <10

Age 5 yr

education, IQ, income,
and prenatal smoking

Born 1994-1995 followed





status, HOME score.

from age 6 mo to 5 yr



Mean: 3.3





3-259

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Bellinger and Needleman n = 48 children
(2003)	Recruited at birth

Boston area, MA.

Prospective

Followed from birth
(1979-1981) to age 10 yr.

Blood

Early childhood (age 2 yr)
Mean (SD)

Peak <10: 3.8 (range: 1-
9.3)

Detection limit NR

FSIQ: WISC-R
Age 10 yr

HOME score (age 10 and
5), child stress events,
race, maternal IQ, age,
marital status, SES, sex,
birth order, # residence
changes before age 5 yr.
Also considered potential
confounding by family
stress, maternal age,
psychiatric factors, child
serum ferritin levels

-1.6 (-2.9, -0.2)

Surkan et al. (2007)
Boston, MA and
Farmington, ME

Cross-sectional

Sep 1997-Mar 2005

n = 389

Children recruited from
trial of amalgam dental
fillings.

Blood

Concurrent
Groupl: 1-2
Group 2: 3-4
Group 3: 5-10

Mean (SD):
2.2(1.6)

WISC-III
Age 6-10 yr

Caregiver IQ, child age,
SES, race, birth weight.
Also considered site, sex,
birth order, caregiver
education and marital
status, parenting stress,
and maternal utilization of
prenatal and annual
health care (not parental
caregiving quality.)

1.0 (reference)
-0.12 (-3.3, 3.1)
-6.0 (-11, -1.4)

Chiodo et al. (2007)
Detroit, Ml area

Cross-sectional

495 children (born 1989-
1991) age 7 yr,

Blood

Concurrent
Mean (SD): 5.0 (3.0)

WPPSI
Age 7 yr

Maternal

psychopathology, IQ,
prenatal smoking,
prenatal marijuana, SES,
HOME score, caretaker
education and marital
status, # children in
home, child sex. Also
considered child age,
maternal age, custody,
cocaine use, prenatal
alcohol use.

-0.19 (-0.30, -0.08)
Note: standardized
regression coefficient.
95% CI estimated using
the reported p-value of
0.01

3-260

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Kim et al. (2009)	261 children

Seoul, Seongnam, Ulsan,

and Yeoncheon, South Schoo, reCruitment

Korea

Cross-sectional
Children born 1996-1999

Blood

Concurrent

Mean (SD): 1.7 (0.80)

Age: 8-11 yr

KEDI-WISC
Ages 8-11 yr

Blood Mn <1.4 |jg/dL
-2.4 (-6.0, 1.1)

Maternal age, education
and prenatal smoking
status, paternal
education, yearly income, Blood Mn >1-4 M9/c|L
smoking exposure status ~3.2 (-6.1, -0.24)
after birth, child age, sex,
and birth weight (not
parental caregiving quality
or IQ)

tBraun et al. (2018)

Cincinnati, OH
United States

Mar 2003-Jan 2006

Followed for 8 yr
Cohort

HOME study
n: 355 (Intervention
group: 174, Control
group: 181)

Clinical trial of pregnant
women, mean gestation
of 16 wk and residence in
a house built in or before
1978

Intervention to reduce Pb
exposure

Blood

Maternal and child blood;
ICP-MS

Dust Pb loadings floor,
interior windowsill and
window at 20 wk
gestation, child age 1 and
2 yr; GFAAS.

Age at Measurement:
16, 26 wk of gestation,
delivery (maternal);
1,2,3,4,5, 8 yr (child)

Baseline GM (Intervention
and control groups):
maternal: 0.7 and 0.7
|jg/dL, floor dust Pb: 1.5
and 1.9 pg/sq ft,
windowsill dust Pb: 28
and 33 pg/sq ft, window
trough dust Pb: 574 and
510 |jg/sq ft.

WPPSI

Age at outcome:
5-8 yr

NA

Mean FSIQ score
difference15: 0.5 (-3.3,
24.2), comparing the
treatment to the injury
prevention control group.

3-261

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

t Taylor et al. (2017)

UK

Cohort

April 1, 1991-Dec 31,
1992 (followed until age
4-8 yr)

ALSPAC
n: 4285

Mother-infant pairs.

Blood	FSIQ, VIQ, PIQ (WPPSI

or WISC-111).

Maternal and child venous
blood; ICP-MS	Age at outcome:

4-8 yr

Age at Measurement:

Prenatal (mean
gestational age 11 wk)
and postnatal (30 mo)

Prenatal: 3.67 |jg/dL;

Child BLL: 4.22 pg/dL.

Family adversity index,
housing tenure,
household crowding,
smoking in the first
trimester, alcohol
consumption in the first
trimester, maternal age at
index birth, parity,
maternal education,
length of time the mother
lived in Avon, child sex,
child age at testing,
weighted life events
score, and hemoglobin
level.

WISC-Boysb: -0.29
(-1.02, 0.44)

WISC-Girlsb: 0.73 (0.13,
1.33)

WPPSI-Girlsb: -0.65
(-2.065, 0.765)
WPSSI-Boysb: -0.54
(-2.015, 0.935)

tTatsuta et al. (2020)

Tohoku district, coastal
area Japan

2002-2006 (enrollment)
through 2015-2018 (12 yr
followup)

Cohort

TSCD coastal cohorts
n: 289 mother-child pairs
(singleton births); 148
boys and 141 girls

Blood

Cord and child venous
blood; ICP-MS

Age at Measurement:
Delivery (cord), 12 yr
(child)

Median: Cord = 0.8 pg/dL,
12-yr = 0.7 pg/dL
95th: Cord: 1.4 pg/dL, 12-
yr: 1.1 pg/dL

FSIQ (Japanese version
WISC-IV), age equivalent
ranking and scores for
verbal comprehension,
perceptual reasoning,
working memory, and
processing speed
composites; BNT (cues
and no cues)

Age at outcome:

12 yr

Birth weight, drinking or
smoking during
pregnancy, the Raven's
score (parent assessment
for child at 18 mo of age),
passive smoking status at
12 yr old, family income,
WISC/BNT tester, and
cord blood total Hg

Cord-Boys: p=—3.683 (-
Cord-Girls: (3 = 10.714,
3.349)

Child-Boys: (3 = 1.463
(-2.905, 5.831)

Child-Boys: (3 = -9.88
(-18.977, -0.782)
Child-Girls: -4.406
(-15.94, 7.129)

3-262

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tDesrochers-Couture et
al. (2018)

10 study sites
Canada

2008-2011

Followed 3-4 yr from birth
Cohort

MIREC Study	Blood

n: 609

Maternal, cord, and
Birth cohort: Mother-infant postnatal child (venous)
pairs recruited during 1st ICP-MS
trimester	Age at Measurement:

Maternal (6-13 wk, 32-34
wk), birth (cord); and 3-4
yr (postnatal child)

GM: 1st trimester: 0.62
|jg/dL; 3rd trimester: 0.59
|jg/dL; cord blood: 0.76
|jg/dL; child blood: 0.70
pg/dL

Max: 1st trimester: 4.14
|jg/dL; 3rd trimester: 3.93
|jg/dL; cord blood: 3.52
|jg/dL; child blood: 5.49
Mg/dL.

FSIQ, VIQ, PIQ, General
Language composite

WPPSI-3rd Edition, short
version. The age-
standardized WPPSI-III
Canadian norms
were used to calculate the
scores.

Age at outcome:

Between 2 yr 6 mo and 3
yr 11 mo

Cord blood model: child
age, child sex, maternal
education, evaluation site
and cord blood Hg (log—2
scale).

Child blood model: child
age, child sex, evaluation
site, marital status,
familial income, HOME
total score, Parenting
Stress Index, and cord
blood Pb (log—2 scale).

Cord: (3=-0.123 (-0.251,
0.005)

Child: (3 = 0.027 (-0.135,
0.188)

Cord-Boys: p=—5.686
(-9.968, -1.405)
Cord-Girls: (3 = 0.287
(-3.787, 4.361)

tZhou et al. (2020b)

Jiangsu Province
China

June 2009-Jan 2010 to
June 2016-July 2017

Cohort

Sheyang Mini Birth
Cohort Study
n: 296

Blood, Urine

FSIQ, VIQ, PIQ (Chinese Sex, maternal age,

Cord blood tested for Mn,
Cd and Pb using GFAAS.
Birth cohort- mother-infant Postnatal urine samples
pairs from an agricultural urine also tested for the
region.	elements.

Age at Measurement:
Cord; postnatal Urine NR

GM: cord blood: 15.88
|jg/L, urine: 1.43 |jg/L,
75th: cord blood: 21.83
|jg/L, urine: 2.27 |jg/L,
Max: cord blood: 1168.20
|jg/L, urine: 62.47 |jg/L,

version WISC-R).

Age at outcome:
6-7 yr (school-aged
children)

maternal education,
family annual income,
family inhabitation area,
and passive smoking;
multiple effects were also
assessed by entering all
other metals in the model.
Sex-stratified analysis
conducted.

Cord-Girls: 0.615 (-0.909,
2.138)

Cord-Boys: 0.835
(-1.164, 2.833)

Cord-All: 0.67 (-0.514,
1.854)

3-263

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tLiu etal. (2015)

Chongqing, China

March 4, 2003-June 19,
2003 (enrollment)
Followed 5 yr

Cohort

Birth cohort with mother-
infant pairs
n: 149

Mothers gave birth in 4
hospitals in Tongliang
county.

Blood

VIQ, PIQ, and FSIQ
(Shanghai version
WPPSI)

Maternal and cord serum
(Hg, Cd, Pb); Pb
determined with AAS.

Ratio of maternal to cord
serum levels (i.e.,
placental transport ratio of equivalent percentile
metal) estimated.	ranks were estimated

Raw scores were
converted to composite
scores and age

Age at Measurement:
Delivery

Age at outcome:
5 yr

Maternal age, educational
level, vitamins, placental
transport ratios, maternal
exposure to ETS. Pb, Hg
and Cd were considered
in multivariable models
(covariates retained
based on evaluation of
VIFs).

Final predictive model did
not include associations
with Pb

Mean: 3.45 |jg/dL.

3-264

-------
tWana et al. (2022) n: 148

Blood

Cognitive Effects in

General linear models

"Table 3, Beta (95% CI):





children - FSIQ

adjusted for child sex,



Wujiang

Blood Pb measured via



maternal age at delivery,

VIQ, total population, cord

China

AAS.

Child health personnel

age of children, maternal

blood:

Birth cohort established

Age at Measurement:

conducted the WISC-CR

education level, paternal



from 2009-2010; follow-

Mean, SD (mo): 89.90,

for FSIQ, PIQ, and VIQ.

education level, monthly

Q2 vs. Q1: -0.296

up from 2016-2017.

3.77

Scores were age

household income, parity,

(-7.005, 6.413)

Cohort



converted and

inhabitation area, passive





Cord blood: GM = 28.26

standardized.

smoking.

Q3 vs. Q1: 4.468 (-2.840,



|jg/L, Median = 27.56





11.776)



|jg/L; Venous blood: GM =

Age at outcome:







22.99 |jg/L, Median =





Q4 vs. Q1: -1.275



23.80 |jg/L





(-8.231, 5.682)



75th: Cord blood: 38.42









|jg/L; Venous blood:





VIQ, boys, cord blood:



33.00 |jg/L









Max: Cord blood: 249.00





Q2 vs. Q1: -2.752



|jg/L; Venous blood:





(-12.594, 7.091)



71.40 |jg/L.















Q3 vs. Q1: 5.681 (-6.288,









17.649)









Q4 vs. Q1: 2.119 (-8.913,









13.150)









VIQ, girls, cord blood:









Q2 vs. Q1: 5.679 (-5.242,









16.600)









Q3 vs. Q1: 6.510 (-4.501,









17.521)









Q4 vs. Q1: -1.331









(-11.933, 9.271)









VIQ, total population,









venous blood:









Q2 vs. Q1: 4.942 (-3.957,









13.841)









Q3 vs. Q1: -0.536









(-9.560, 8.487)

3-265

-------
3-266

Q4 vs. Q1: -3.304
(-12.117, 5.509)

VIQ, boys, venous blood:

Q2 vs. Q1: 5.592 (-7.193,
18.378)

Q3 vs. Q1: 8.858 (-4.271,
21.988)

Q4 vs. Q1: 7.143 (-5.649,
19.935)

VIQ, girls, venous blood:

Q2 vs. Q1: 3.179
(-10.810, 17.169)

Q3 vs. Q1: -13.548
(-27.506, 0.411)

Q4 vs. Q1: -14.964
(-28.412, -1.517), p-
value = 0.036

PIQ, total population, cord
blood:

Q2 vs. Q1: -5.584
(-14.011, 2.842)

Q3 vs. Q1: -0.441
(-9.620, 8.738)

Q4 vs. Q1: -9.365
(-18.103, -0.628), p-
value = 0.038

PIQ, boys, cord blood:

Q2 vs. Q1: -13.080
(-24.907, -1.254), p-

-------
value = 0.035

Q3 vs. Q1: -9.686
(-24.067, 4.695)

Q4 vs. Q1: -7.592
(-20.848, 5.663)

PIQ, girls, cord blood:

Q2 vs. Q1: -0.002
(-14.192, 14.188)

Q3 vs. Q1: 4.023
(-10.284, 18.331)

Q4 vs. Q1: -13.293
(-27.069, 0.483)

PIQ, total population,
venous blood:

Q2 vs. Q1: -7.293
(-17.605, 3.020)

Q3 vs. Q1: -8.176
(-18.633, 2.281)

Q4 vs. Q1: -4.507
(-14.720, 5.706)

PIQ, boys, venous blood:

Q2 vs. Q1: -8.218
(-22.276, 5.841)

Q3 vs. Q1: -6.701
(-21.138, 7.737)

Q4 vs. Q1: -4.294
(-18.360, 9.772)

PIQ, girls, venous blood:

3-267

-------
3-268

Q2 vs. Q1: -10.417
(-29.104, 8.270)

Q3 vs. Q1: -16.397
(-35.043, 2.248)

Q4 vs. Q1: -6.994
(-24.957, 10.968)

FSIQ, total population,
cord blood:

Q2 vs. Q1: -3.369
(-10.711, 3,974)

Q3 vs. Q1: 2.396 (-5.603
10.394)

Q4 vs. Q1: -6.087
(-13.700, 1.527)

FSIQ, boys, cord blood:

Q2 vs. Q1: -8.599
(-19.015, 1.818)

Q3 vs. Q1: -1.434
(-14.100, 11.232)

Q4 vs. Q1: -2.552
(-14.227, 9.123)

FSIQ, girls, cord blood:

Q2 vs. Q1: 2.986 (-9.126
15.098)

Q3 vs. Q1: 5.743 (-6.469
17.955)

Q4 vs. Q1: -8.635
(-20.394, 3.123)

FSIQ, total population,

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

venous blood:

Q2 vs. Q1: -0.950
(-10.606, 8.706)

Q3 vs. Q1: -4.930
(-14.722, 4.861)

Q4 vs. Q1: -4.773
(-14.336, 4.790)

FSIQ, boys, venous
blood:

Q2 vs. Q1: -1.367
(-14.675, 11.941)

Q3 vs. Q1: 1.366
(-12.300, 15.033)

Q4 vs. Q1: 1.929
(-11.386, 15.243)

FSIQ, girls, venous blood:

Q2 vs. Q1: -3.771
(-20.071, 12.529)

Q3 vs. Q1: -17.326
(-33.590, -1.062), p-
value = 0.044

Q4 vs. Q1: -13.625
(-29.293, 2.0"3)"

3-269

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tlalesias et al. (2011)

Antofagasta,

Northern Chile
1998-2005

Cohort

Sepulveda study
n: 192

Children lived or attended
school in area
contaminated by Pb
mineral concentrate
stored in open sites at
railroad terminal (closed
in 1998).

Blood

Child's venous blood;
AAS.

Age at Measurement:
In 1998 at 0-7 yr old; in
2005 at 7-16 yr old

Mean and Median: blood
Pb 1998: 10.8 and 10
|jg/dL; blood Pb 2005: 3.5
and 3.2 |jg/dL
75th: blood Pb 1998: 14
|jg/dL; blood Pb 2005: 4.3
|jg/dL

Max: Blood Pb 1998: 33
|jg/dL; blood Pb 2005: 14
(jg/dL.

FSIQ, VIQ, PIQ (Chilean
version of WISC-R).

Age at outcome:

7-16 yr

Sex, birth weight, birth
order, # of siblings, milk
type during the first six
months of life, history of
anemia, SES (household
income, home ownership,
and school type: public or
private), parental
education, maternal
smoking during
pregnancy, maternal IQ,
children's stimulation at
home, HOME score.

Concurrent: -0.94 (-1.77,
-0.11)

Early childhood: -0.14
(-0.445, 0.165)

tRuebner et al. (2019)

46 centers
U.S.

Cohort

3 enrollment periods,
2005-2009, 2011-2014,
2016-2020

Followed up to 9 yr

CKiD Cohort study
n: 412

Children with mild to
moderate CKD

Blood

ICP-

Child venous blood
MS. The BLL
measurement closest to
the time of neurocognitive
testing was used for
analysis (concurrent).

Age at measurement:
NR; 2, 4, or 6 yr after
study entry

Median: 1.2 |jg/dL
75th: 1.8 |jg/dL
Max: 5.1 |jg/dL

FSIQ

Mullen Scales of Early
Learning (age 12-29 mo),
WPPSI (30 mo-5 yr), and
WAS I (6-18 yr).

The last available test
results were to evaluate
long-term effects. Mean
time between BLL and
neurocognitive testing
was 2.3 yr.

Age at outcome:

1-16 yr

Age, sex, race, poverty,
and maternal education

Concurrent: (3=-2.1
(-3.95, -0.25)

3-270

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tLee etal. (2021)

Seoul, Gyeonggi, and
Incheon provinces
South Korea

2008-2017 (Recruitment
2008-2010; follow-up
from 2012-2017)

Cohort

Environment and
Development of Children
n: 502

Blood

Cognitive Effects (FSIQ)

Intelligence quotients of
children assessed using
'KEDI-WISC.

Whole blood Pb from
mothers during their
second trimester of
pregnancy and children at
ages 4 yr and 6 yr were Age at outcome:
analyzed by atomic
absorption
spectrophotometry.

Age at Measurement:

Maternal mean age (SD)

= 31.3 (3.5) yr. Children at
4 yr and 6 yr.

Prenatal GM (SD) = 1.32
(1.32) |jg/dL; children at 4
yr = 1.43 (1.38) pg/dL;
children at 6 yr = 1.43
(1.35) pg/dL
75th: Prenatal = 1.56
pg/dL; children at 4 yr =

1.72 pg/dL; children at 6
yr = 1.70 pg/dL
95th: Prenatal = 2.11
pg/dL; children at 4 yr =

2.47 pg/dL; children at 6
yr = 2.28 pg/dL.

Multivariate models
adjusted by maternal
education level, exposure
to ETS during the
pregnancy, maternal age,
and maternal IQ.

Table-3 - Estimated
coefficients and 95% CI of
associations between
single metals and
children's IQ at 6-years
old (standardized)

Prenatal period: -1.202
(-4.87, 2.467)

At age 4: -1.829 (-4.664,
1.006)

At age 6: -2.614 (-5.623,
0.396)

tDantzer et al. (2020)

Greater Cincinnati, Ohio

Metro area

U.S.

Cross-sectional analysis
of data collected at age
12

CCAAPS
n: 344

Cohort recruited at birth
Oct 2001 —Jul 2003

Blood, nails

Postnatal child venous
blood, toenail; ICP-MS.
Mean blood Pb: 0.57
pg/dL; toenail Pb: 0.66
pg/g; information also
available by gender and
race

Age: 12 yr.

FSIQ (WISC-IV)
Age at outcome:
12 yr

Caregiver IQ, community
deprivation index, and
BMI. Sex considered as a
potential confound.

Concurrent (blood):
B=-10.871 (-16.893,
-4.848)

B=-1.70 (-4.27, -0.862)

3-271

-------
tMartin et al. (2021)

East Liverpool, Ohio
United States
2013-2014
Cross-Sectional

CARES
n: 66

BLLs from the children
were analyzed by ICP-
MS.

Age at Measurement:
Mean (SD) = 8.4 (0.9) yr

GM (SD) = 1.13 (1.96)
|jg/dL

Max: 6.64 |jg/dL.

Cognitive Effects

Cognitive performance
was assessed using the
WISC-IV.

Age at outcome:

Regression models were
adjusted for sex, income,
and In (serum cotinine).

"Table 3 - Interaction
effects between Ln Blood
Pb-Ln Hair Mn, 13 (95%
CI)

Blood Pb (per 1 In |jg/dL
difference)

FSIQ

At In hair Mn = 5 ng/g:
1.686 (-3.039, 6.412)

At In hair Mn = 6.25 ng/g:
-4.447 (-8.333, -0.561)

At In hair Mn = 7 ng/g:
-8.133 (-13.4, -2.867)

At In hair Mn = 7.5 ng/g:
-10.596 (-17.172, -4.02)

Perceptual Reasoning

At In hair Mn = 5 ng/g:
2.392 (-4.055, 8.839)

At In hair Mn = 6.25 ng/g:
-4.612 (-9.918, 0.694)

At In hair Mn = 7 ng/g:
-8.808 (-15.996, -1.62)

At In hair Mn = 7.5 ng/g:
-11.616 (-20.592,
-2.639)

Processing Speed

At In hair Mn = 5 ng/g:
-0.62 (-4.263, 3.024)

At In hair Mn = 6.25 ng/g:
-2.565 (-5.565, 0.435)

3-272

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

At In hair Mn = 7 ng/g:
-3.725 (-7.788, 0.337)

At In hair Mn = 7.5 ng/g:
-4.502 (-9.576, 0.573)

Verbal Comprehension

At In hair Mn = 5 ng/g:
1.208 (-3.451, 5.867)

At In hair Mn = 6.25 ng/g:
-4.141 (-7.976, -0.306)

At In hair Mn = 7 ng/g:
-7.349 (-12.549, -2.149)

At In hair Mn = 7.5 ng/g:
-9.49 (-15.976, -3.004)

Working Memory

At In hair Mn = 5 ng/g:
2.376 (-2.404, 7.157)

At In hair Mn = 6.25 ng/g:
-2.133 (-6.067, 1.8)

At In hair Mn = 7 ng/g:
-4.831 (-10.161, 0.498)

At In hair Mn = 7.5 ng/g:
-6.635 (-13.286, 0.0"6)"

3-273

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tHavnes et al. (2015)

Marietta or Cambridge,
Ohio, and surrounding
communities
U.S.

Oct 2008-M arch 2013
Cross-sectional

CARES
n: 404

Blood

Child venous blood: ICP-
MS

Age at Measurement:

Participants resided in
study area throughout
their life; not moving for at -jIq yr
least 1 yr

GM: 0.82 pg/dL
Max: NR.

FSIQ (WISC-IV), 4
domains of intellectual
functioning (reasoning,
processing speed,
working memory and
verbal comprehension)

Age at outcome:
7-9 yr

Sex, parent's IQ, parent Processing speed: (3 = -
education, parent	3.53 [ -6.95, -0.12)

confidence T-score, Mn or

Pb, community residence Association with FSIQ NR
(FSIQ models only); other
sets of variables added
depending on domain.

(Note: main effect is Mn)

tHonq et al. (2015)

5 administrative regions
South Korea
NR

Cross-sectional

n: 1001

General population of
children

Blood

Venous blood; GFAAS
Age at Measurement:
8-11 yr old

Median: 1.81 pg/dL
75th: 2.25 pg/dL,
95th: 3.01 pg/dL
Max: 6.16 pg/dL.

KEDI-WISC

Age at outcome:
8-11 yr old

Age, sex, residential
region, paternal education
level, and yearly income
Iog10-transformed blood
Hg, Mn, urine
concentrations of cotinine,
phthalate metabolites

B=-1.948 (-3.608,
-0.288), adjusted for
other metals
B=-2.113 (-3.73,
-0.496), adjusted for
ADHD and CPT
B=-2.118 (-3.792,
-0.445), adjusted for
socio-demographic
factors

tMenezes-Filho et al.
(2018)

Salvador, Bahia
Brazil

Cross-sectional
Study years: NR

School-based cohort
n: 225

Children from 4
elementary schools in
industrial town.

Blood

Child venous blood Pb,
hair and toenails tested
for Mn; GFAAS

Age at Measurement:
7-12 yr

Mean: 1.64 pg/dL,
Median: 1.15 pg/dL, only
about 2% of children
above the Centers for
Disease Control and
Prevention ref value of 5
pg/dL

75th: 2.1 pg/dL
Max: 15.6 pg/dL.

IQ estimated using
vocabulary and matrix
reasoning (WASI).

Age at outcome:
7-12 yr

Age, Maternal IQ

Effect modification by Mn
assessed. Both Pb and
Mn were log-transformed
to include in the model.
Interaction between Pb
and Mn assessed.

Put results for Model B
which is adjusted for Mn,
age and maternal IQ

3-274

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tLucchini et al. (2012)

Junior high school-age

Blood

IQ tested using WISC-III

Sex, age at testing,

B=-2.248 (-4.111,



children from 20 local



(verbal IQ and

parental education, SES,

-0.385)

Valcamonica and Garda

public schools

Children's venous blood;

performance IQ

family size, parity order,



Lake areas in Province of

n: 299

GFAAS

assessed).

BMI



Brescia











Italy



Age at Measurement:

Age at outcome:





Cross-sectional



11-14 yr

11-14 yr









1.71 pg/dL, Median: 1.50











75th: 2.10 pg/dL











Max: 10.2 pg/dL.







AAS = atomic absorption spectrometry; ALSPAC = Avon Longitudinal Study of Parents and Children; avg = average; BLL = blood lead level; BNT = Boston Naming Test; CARES =
Communities Actively Researching Exposure Study; CCAAPS = Cincinnati Childhood Allergy and Air Pollution Study; Cd = cadmium; CI = confidence interval; CKD = chronic kidney
disease; CKiD = Chronic Kidney Disease in Children Study; ETS = environmental tobacco smoke; FSIQ = full-scale intelligence quotient; GFAAS = graphite furnace atomic absorption
spectrometry; GM = geometric mean; Hg = mercury; HOME = Health Outcomes and Measures of the Environment; ICP-MS = inductively coupled plasma mass spectrometry; IQ =
intelligence quotient; KEDI = Korean Educational Development Institute; MIREC = Maternal-Infant Research on Environmental Chemical; Mn = manganese; mo = month(s); NA = not
available; NR = not reported; Pb = lead; PC = primary caregiver; PIQ = performance IQ; Q = quartile; SD = standard deviation; SES = socioeconomic status; VIF = variance inflation
factor; VIQ = verbal IQ; WASI = Wechsler Abbreviated Scale of Intelligence; WISC = Weschler Intelligence Scale for Children; wk = week(s); WPPSI = Wechsler Preschool and
Primary Scale of Intelligence; yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bResults are not standardized (e.g., BLL distribution data needed to calculate the standardized estimate was not reported or categorical data was analyzed).

tStudies published since the 2013 Integrated Science Assessment for Lead.

3-275

-------
Table 3-3E Epidemiologic studies of Pb exposure and infant development

Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Bellinger et al. (1987)

Boston, MA
U.S.

Apr. 1979 - Apr. 1981
(enrollment)

Followed through 2 yr
Cohort

Birth cohort, n = 182
infants

Recruitment from births
at Brigham and Women's
Hospital

Blood

Cord blood; anodic stripping
voltammetry (ASV)

Age at measurement:
Delivery

Mean (SD): 6.6 (3.2) pg/dL
Low: <3 pg/dL
Medium: 6-7 pg/dL
High: >10 pg/dL

MDI assessed using
BSID-II

Age-standardized
scores (mean: 100,
SD: 16)

Age at outcome: 2 yr

Maternal age, race, IQ,
education, years of
smoking, and alcohol
drinks/wk in 3rd
trimester, SES, HOME
score, child sex, birth
weight, gestational age,
birth order.

Beta:

Cord blood:

Low vs. high: -3.8 (-6.3,
-1.3)

Medium vs. high: -4.8
(-7.3, -2.3)

Concurrent blood
reported not to be
associated with MDI,
quantitative data not
reported.

Jedrvchowski et al.
(2009b)

Krakow, Poland

2001-2004 (enrollment)
Followed through 3 yr

Cohort

Birth cohort, n = 381-415 Blood
children

Recruited pregnant
mothers from prenatal
clinics in Krakow inner
city in 1st and 2nd
trimesters.

Cord blood; ICP-MS

Age at measurement:
Delivery

GM (95% CI): 1.29 (1.24,
1.34) pg/dL

Median: 1.23 pg/dL

MDI assessed using
BSID-II (Polish
version)

Standardized scores

Age at outcome:
12, 24, 36 mo

Maternal education and
prenatal smoking, child
sex and birth order

Beta

Age 2 yr: -1.8 (-3.4,
-0.14)

Age 3 yr:

-0.21)

-1.6 (-2.9,

Henn et al. (2012)

Mexico City
Mexico

1997-2000 (enrollment)
Followed through 24 mo

N: 455

Blood

MDI assessed using Sex, gestational age,

Women recruited during Child venous blood; ICP-MS
pregnancy or at delivery

Age at measurement:
12, 24 mo

Mean (SD):

BSID-II (Spanish
version)

Age at Outcome:
12, 18, 24, 30, 36
mo

hemoglobin, maternal IQ,
maternal education, and
visit

Beta

12-months: -0.07 (-0.39,
0.25)

24-months: -0.08 (-0.46,
0.30)

12 mo Pb * Mn <2 pg/dL:
-0.31 (-1.25, 0.62)

3-276

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Cohort



12 mo: 5.1 (2.6) pg/dL
24 mo: 5.0 (2.9) pg/dL





24 mo Pb * Mn <2 pg/dL:
-1.27 (-2.18, -0.37)

Hu et al. (2006)

N: 83 (cord) - 146 (24-

Blood

MDI assessed using

Maternal age and IQ,

Maternal T1: -0.76



months child blood)



BSID-II (Spanish

child sex, current weight,

(-1.50, -0.03)

Mexico City
Mexico

Women recruited during

Maternal blood, cord blood,
and child venous blood; ICP-

version)

height-for-age Z score,
and concurrent blood Pb
(in models examining
prenatal blood Pb)

Maternal T3: -0.43
(-1.10, 0.27)

pregnancy or at delivery

MS

Age at Outcome:

Cord: -0.06 (-0.87, 0.74)

1997-2000 (enrollment)
Followed through 24 mo



Age at measurement: T1, T2,
T3 (maternal), delivery (cord),
12, 24 mo (child)

24 mo

Child 24 mo: -0.23
(-0.92, 0.45)

Cohort











Mean (SD):

Maternal T1: 7.07 (5.10)
|jg/dL; T3: 6.86 (4.23) pg/dL

Cord: 6.20 (3.88) pg/dL
Child 24 mo: 4.79 (3.71)
pg/dL

tY Ortiz etal. (2017)

Mexico City
Mexico

Jul 2007-Feb 2011
Followed through 24 mo

Cohort

PROGRESS birth cohort
n: 536

Women <20 wk of
gestation and planning to
reside in Mexico City for
the next 3 yr.

Blood

Maternal blood; ICP-MS.

Age at measurement:
T2, T3

Mean:

T2: 3.7 pg/dL
T3: 3.9 pg/dL.

Cognitive and
language
development
assessed using
BSID-III.

Standardized scores
(mean: 100, SD:
15). Cognitive,
language and motor
scores were jointly
considered for
standardizing.

Age at outcome:
24 mo

Infant sex, birth weight,
gestational age, maternal
age, maternal IQ (WAIS
Spanish version), HOME
score.

Beta

Cognitive
Development:

T2: 0.76 (-3.35, 4.87)bc
T3: -6.60 (-13.49,
0.29)bc

Stress2: -0.23 (-0.45,
-0.01 )bc

T3*Stress: 1.02 (-0.78,
2.82)bc

Language
Development:

T2: 0.97 (-3.18, 5.12)bc
T3: -6.00 (-12.94,
0.94)bc

3-277

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tKimetal. (2013c) and MOCEH study
Kimet al. (2013b)

Seoul,
Ulsan
Korea

Cheonan and

2006-2010

Followed through 6 mo
Cohort

n: 884

Mothers recruited before
20th wk of pregnancy
between and were in
locations (Seoul,
Cheonan and Ulsan).

Blood

Maternal venous blood;
GFAAS with Zeeman
background correction,
measured for Pb and Cd

Age at measurement:

Early (<20 wk) and late
pregnancy (med = 39 wk)

Early pregnancy: 1.4 (GM),
2.1 (90th), 9.8 (max) pg/dL
Late pregnancy: 1.3 (GM); 2.1
(90th), 4.3 (max) pg/dL

MDI assessed using
BSID-II (Korean
version)

Age-standardized
scores (mean: 100,
SD: 15).

Age at outcome:
6 mo

Birth weight, infant sex,
maternal age and
education, family income,
breastfeeding status,
residential area.

Beta
Early:

Overall: 0.02 (-1.20,
1.24)

Cd <1.47 pg/L: 2.44
(0.04, 4.83)
Cd >1.47 pg/L: -0.87
(-2.52, 0.78)

Late:

Overall: -1.74 (-3.37,
-0.12)

Cd <1.51 pg/L: -0.29
(-2.88, 2.30)

Cd >1.51 pg/L: -3.20
(-5.35, -1.06)

tKimetal. (2018b)

4 cities: Seoul, Anyang,
Ansan and Jeju
Korea

2011-2012 (enrollment)
Followed through 24 mo

Cohort

CHECK cohort
n: 140

birth cohort- pregnant
women recruited from 4
cities in Korea before
delivery.

Blood

Maternal and cord blood;
method NR

Age at measurement:
Delivery

Median (IQR):

Maternal: 2.7 (3.5, 5.7) pg/dL
Cord: 1.2 (0.8, 1.7) pg/dL

MDI assessed using
BSID-II (Korean
version)

Age at outcome:
13-24 mo

BPA, and phthalates,
maternal age
(continuous), birth
delivery mode
(categorical), monthly
household income
(categorical), child's sex,
and BDI (continuous) of
the mother, gestational
age (continuous),
primiparous (categorical),
and

pre-pregnancy BMI
(categorical).

Associations of blood Pb
concentrations and MDI
were assessed but not
reported because they
lacked statistical
significance.

tValeri et al. (2017)

Birth cohort

Blood

Cognitive and

Child sex, age attesting,

Beta



n: 825 (Pabna: 409,



language

maternal age and

Pabna

Pabna and Sirajdikhan

Sirajdikhan: 416)

Cord blood; ICP-MS,

development using

education, maternal IQ,

Cognitive: 0.012 (-0.05,
0.074)

districts



measured for Pb, As, and Mn

BSID-III (Bengali

HOME score, ETS,

Bangladesh

Mother-infant pairs,



version, adapted for

protein intake.

2010-2013 (enrollment)



Age at measurement:

rural Bangaldesh)





3-278

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Followed through 20-40
mo

Cohort

pregnant women enrolled Delivery
in the first trimester.

GM: Pabna: 1.8 |jg/dL,
Sirajdikhan: 6.0 |jg/dL.
75th: Pabna: 2.4, Sirajdikhan:
9.7

Max: Pabna: 79.1 |jg/dL,
Sirajdikhan: 36.0 |jg/dL.

Two primary
outcomes derived by
summing across raw
scores of cognitive
and language
development. Z-
scores were
calculated.

Age at outcome:
20-40 mo

Language: -0.014
(-0.076, 0.048)

Sirajdikhan

Cognitive: -0.011
(-0.024, 0.001)
Language: -0.004
(-0.026, 0.017)

tKoshv et al. (2020)

Old Town, Salavanpet
and neighboring areas in
Vellore, South India

Mar 2010-Feb 2012
(enrollment)

Followed through 5 yr

Cohort

Etiology, Risk Factors
and Interactions of
Enteric Infections and
Malnutrition and the
Consequences for Child
Health and Development
(MAL-ED) Network
n: 228 (followed 2 yr) and
212 children (followed 5
yr)

Birth cohort of mother-
infant pairs in eight
adjacent urban slum
dwelling areas

Blood

Child venous blood; GFAAS.
Mean BLL derived by
averaging BLLs at 15 and 24
mo for analysis at 2 yr, and
15, 24 and 36 mo for analysis
at 5 yr.

Age at measurement:
7, 15, 24 and 36 mo

Mean: 15 mo: 0.5 pmol/L, 24
mo: 0.6 pmol/L, 36 mo: 0.6
pmol/L

Cognitive and
language
development
assessed using
BSID-III (culturally
adapted and
translated)

Raw scores of
cognition and
expressive and
receptive language
domains.

Age at outcome:
24 mo

Child sex, maternal
intelligence raw scores,
SES, mean body Fe
levels.

Beta

0.2,

Cognitive: -0.2
-0.03)

Expressive language:
-0.2 (-0.3, -0.1)

Receptive language:
-0.04 (-0.1, 0.02)

tShekhawat et al. (2021) n:117

Western Rajasthan
India

2018-2019 (enrollment)
Followed through 6.5 mo
(average)

Mother-child pairs in third
trimester or at delivery

Blood

Cord blood; ICP-OES

Age at measurement:
Delivery

GM =4.14 |jg/dL; mean =
4.77 ± 3.3 |jg/dL; median =

Cognitive and
language
development
assessed using
BSID-III

Age at outcome: 6.5
mo (average)

Maternal age, gravida,
gestational age, maternal
education, child sex and
weight, preterm birth,
maternal food intake
during pregnancy,
smoking, alcohol
consumption, maternal
residential and

13 (95 % CI)
(A) Umbilical cord Pb
level <5 |jg/dL (n = 70)
Composite cognitive:
0.19 (-0.03, 0.34)
Composite language:
0.21 (-0.23, 0.42)
Subscale receptive
language: 0.11 (-0.6,
1.66)

Subscale expressive

3-279

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Cohort

4.23 |jg/dL
75th: 5.1 |jg/dL.

occupational history,
delivery type.

language: 0.22 (-0.03,
1.87)

(B) Umbilical cord Pb
level = 5.0-10.5 |jg/dL (n
= 47)

Composite cognitive:
-0.13 (-0.77, 0.28)
Composite language:
-0.05 (-0.7, 0.47)
Subscale receptive
language: -0.04 (-3.5,
2.5)

Subscale expressive
language: 0.04 (-3.8,
2.9)

tParaiuli et al. (2015a) Birth cohort from
Bharatpur General
Hospital
n: 100

Chitwan, Bharatpur

District

Nepal

Sep-Oct 2008
(enrollment)

Followed through 24 mo

Resided in area for at
least 2 yr delivered at
term (i.e., >37 wk).

Blood

Cord blood; ICP-MS,
measured for Pb, As and Zn

Age at measurement:
Delivery

Median: 2.06 |jg/dL
Max: 22.08 pg/dL.

MDI assessed using
BSID-II

Age at outcome:
24 mo

Maternal age and
education, BMI,
gestational age, family
income, parity, birth
weight, weight at 24 mo,
child age assessment,
As, Zn, HOME score
(smoking and alcohol
consumption not included
given low prevalence).

Beta
-4.21

-13.62, 5.20)c

Cohort

tParaiuli et al. (2015b)

Chitwan, Bharatpur

district

Nepal

Sep-Oct 2008
(enrollment)

Followed through 36 mo
Cohort

Birth cohort from
Bharatpur General
Hospital
n: 100

Resided in area for at
least 2 yr delivered at
term (i.e., >37 wk).

Blood

Cord blood; ICP-MS,
measured for Pb, As and Zn

Age at measurement:
Delivery

Median: 2.06 pg/dL
Max: 22.08 pg/dL.

MDI assessed using Maternal age and

BSID-II

Age at outcome:
36 mo

education, BMI,
gestational age, family
income, parity, birth
weight, weight at 24 mo,
child age at assessment,
As, Zn, HOME score
(smoking and alcohol
consumption not included
given low given low
prevalence).

Beta

4.05 (-3.21, 11.311

3-280

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

t Zhou et al. (2017)

Shanghai
China

2010-2012 (enrollment)
Followed through 24-36
mo

Cohort

Shanghai Stress Birth
Cohort Study
n: 139

Women enrolled in
prenatal clinics of
maternity hospitals
during mid-to-late
pregnancy.

Blood

Maternal blood; AAS

Age at measurement:
28-36 wk of gestation

GM (95% CI): 3.30 (3.05,
3.57) |jg/dL.

Language
development
assessed using
GDS (Chinese
version)

Age at outcome:
24-36 mo

Maternal age at
enrollment, economic
status, maternal
education, gestational
week, child sex, birth
weight and age.

Beta per log—10
transformed BLL

Language development
Overall: -6.76 (-17.29,
3.77)d

Low stress: -1.76
(-13.03, 9.51 )d
High stress: -33.82
(-60.04, —7.59)d

tViqeh et al. (2014)

Tehran
Iran

October 2006 - March
2011

Followed through 36 mo
Cohort

Birth cohort
n: 174

Mother-infant pairs
recruited in first trimester
(8-12 wk).

Blood

Maternal blood,
ICP-MS

cord blood;

Age at measurement:
3 trimesters during pregnancy
and delivery

Mean:

Maternal T1: 4.15 pg/dL, T2:
3.44 pg/dL, T3: 3.78 pg/dL
Cord: 2.86 pg/dL
Max:

Maternal T1: 20.5 pg/dL, T2:
7.5 pg/dL, T3: 8.0 pg/dL

Cord: 6.9 pg/dL

Mental development
composite assessed
using the ECDI by
Harold Ireton
(language
comprehension,
expressive
language, gross
motor, self-help,
social interaction).

Cutoff point scores
for development
delay were score
<20% of that
expected for
children's age.

Age at outcome:
36 mo

Maternal educational,
BMI, family income,
gestational age, birth
weight, birth order (first
born).

OR

Total ECDI: 1.74 (1.1?
2.5).

tLin et al. (2013)
Taipei, Taiwan

April 2004-Jan 2005
(enrollment)

Birth cohort
n: 230

Mother-infant pairs from
medical center, local
hospital, and obstetric
clinics.

Blood

Maternal blood, cord blood;
ICP-MS, measured for Pb,
Mn, As, and Hg.

Cognitive and
language
development
assessed using
CDIIT

Maternal age, education,
infant gender, ETS
during pregnancy and
after delivery, fish intake,
and HOME score.

Beta

Cognitive

High vs. Low Pb: -5.35
(-9.642, —1.058)b
High Mn*low Pb: -4.15
(-9.618, 1.318)b

3-281

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Followed through 2 yr

Panel Study

Pb categories:
Low: <16.45 |jg/L
High: >16.45 pg/L
Mn categories:
Low: <59.59 |jg/L
High: >59.59 pg/L

Age at measurement:
delivery

Mean: 13 pg/L, GM: 10.61
pg/L

75th: 16.45 pg/L
Max: 43.22 pg/L

Age at outcome:
2 yr

Low Mn*high Pb: -4.79
(-10.298, 0.718)b

High Mn*high Pb: -8.19
(-14.403, —1.977)b

Language

High vs. Low Pb: -2.53
(-6.234, 1.174)b

High Mn*low Pb: -1.56
(-6.264, 3.144)b
Low Mn*high Pb: 0.22
(-4.523, 4.963)b
High Mn*high Pb: -6.81
(-12.161, -1.459)b

tNozadi etal. (2021)

Navajo Nation
United States

February 2013-June
2018 (enrollment)
Followed through 10-13
mo

Cohort

Navajo Birth Cohort

Study

n: 327

Blood

Maternal blood, child blood;
ICP-DRC-MS.

Age at measurement:
Delivery or 36-wk visit
(maternal); 10, 13 mo (child)

GM = 0.410 pg/dL; median =
0.37 pg/dL
75th: 0.51 pg/dL
95th: 1.20 pg/dL

Problem-solving
scores assessed
using ASQ:I.

Age-adjusted
scores.

Age at outcome: 10-
13 mo

Urine strontium and
arsenic.

Beta

Problem-Solving: -0.67
(-1.54, 0.20)

tNvanza et al. (2021)

Northern Tanzania
Tanzania

2015-2017 (enrollment)
Followed through 6-12
mo

Mining and Health
Prospective Longitudinal
Study in Northern
Tanzania
n: 439

Birth cohort of mother-
child pairs recruited in
2nd trimester

Maternal dried blood spots;
ICP-MS, measured for Pb,
Hg, and Cd

Age at measurement:

T2

Median: 2.72 pg/dL

Language and
global

neurodevelopment
assessed using
MDAT. Scores in
each domain
classified as normal
(>90th percentile on
all items in that

Maternal age and
education, maternal and
paternal occupation,
number siblings under 5
yr at home, and family
SES, infant sex, age,
birth weight, height, and
weight as a proxy for
nutritional status.

Prevalence ratio

Language Development:
1.0 (1.0, 1.0)

Global

neurodevelopmental
status: 1.0 (0.9, 1.0)

3-282

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Cohort



75th: 4.25 pg/dL
Max: 14.5 pg/dL

domain or <90th
percentile on one or
two items in the
domain) or impaired
(<90th percentile on
more than two items
in a domain).

Age at outcome:
6-12 mo

(Covariates with p < 0.20
retained in the final
models.)

Hg >0.08 pg/dL * Pb> 3.5
pg/dL: 1.4 (0.9, 2.1)

AAS = atomic absorption spectrometry; As = arsenic; ASQ:i = Ages and Stages Questionnaires: Inventory; BDI = Beck Depression Inventory; BLL = blood lead level; BMI = body
mass index; BPA = bisphenol A; BSID = Bayley Scales of Infant and Toddler Development; CARES = Communities Actively Researching Exposure Study; CCAAPS = Cincinnati
Childhood Allergy and Air Pollution Study; Cd = cadmium; CDIIT = Comprehensive Developmental Inventory for Infants and Toddlers; CHECK = Children's Health and Environmental
Chemicals in Korea; CI = confidence interval; CKD = chronic kidney disease; CKiD = Chronic Kidney Disease in Children Study; ECDI = Early Child Development Inventory;
ELEMENT = Early Life Exposure in Mexico to Environmental Toxicants; ETS = environmental tobacco smoke; FSIQ = full-scale IQ; GDS = Gesell Developmental Schedules; GFAAS
= graphite furnace atomic absorption spectrometry; GM = geometric mean; Hg = mercury; HOME = Health Outcomes and Measures of the Environment; ICP-DRC-MS = dynamic
reaction cell for inductively coupled plasma mass spectrometry; ICP-MS = inductively coupled plasma mass spectrometry; ICP-OES = inductively coupled plasma optical emission
spectrometry; IQ = intelligence quotient; MDAT = Malawi Development Assessment Tool; MDI = Mental Developmental Index; Mn = manganese; mo = month(s); MOCEH = Mothers'
and Children's Environmental Health; NBAS = Neonatal Behavioral Assessment Scale; NR = not reported; OR = odds ratio; Pb = lead; PDI = Psychomotor Developmental Index;
PROGRESS = Programming Research in Obesity, Growth, Environment and Social Stressors; SD = standard deviation; SES = socioeconomic status; T1 = first trimester of
pregnancy; T2 = second trimester of pregnancy; T3 = third trimester of pregnancy; WISC = Weschler Intelligence Scale for Children; wk = week(s); yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

The CI was calculated from a p-value and the true CI may be wider or narrower than calculated.

°Results are unstandardized because the log base used for exposure transformation was unspecified in the study.
dResults are unstandardized because the Pb level distribution data was not available.
tStudies published since the 2013 Integrated Science Assessment for Lead.

Table 3-4E Epidemiologic studies of Pb exposure and performance on neuropsychological tests of cognitive
function, i.e., learning, memory, and executive function

Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Lanphear et al. (2000)

United States

U.S. NHANES	Blood

n = 4,853 children ages

6-16 yr (born 1972-1988) Concurrent

GM (SD): 1.9 (7.0)

Digit span
WISC-R

Age at outcome: 6-16 yr

Child sex, race/ethnicity, -0.05 (-0.09, -0.01)

poverty index ratio,

reference adult education,

serum ferritin and cotinine

levels. Did not consider

potential confounding by	

3-283

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Effect Estimates and
Confounders 950/o C|sa

1988-1994
Cross-sectional

Large U.S. representative 63.5% <2.5

study of multiple risk Detection limit = 0.5
factors and outcomes , x , , , „

Interval analyzed: 1-5

Linear regression

parental cognitive function
or caregiving quality.

Kriea et al. (2010)

United States

1991-1994
Cross-sectional

U.S. NHANES
n = 773 children ages 12-
16 yr (born 1972-1982)

Concurrent
GM (SD): 1.9 (7.0)
63.5% <2.5
Detection limit = 0.5

Large U.S. representative , x , , , „
study of multiple risk lnterval analyzed: 1-5

factors and outcomes

Digit span
WISC-R

Age at outcome: 6-16 yr
Log-linear regression

Child sex, caregiver
education, family income,
race/ethnicity, test
language. Did not
consider potential
confounding by parental
cognitive function or
caregiving quality.

-0.34 (-0.59, -0.08)

Surkan et al. (2007)

Boston, Massachusetts
and Farmington, Maine

United States
Cross-sectional

n = 389 children
6-10 yr

Recruitment from trial of
amalgam fillings

Blood

Concurrent
Group 1: 1-2
Group 2: 3-4
Group 3: 5-10
Mean (SD): 2.2 (1.6)

General memory index,
WRAML

Age at outcome: 6-10 yr

Caregiver IQ, child age,
SES, race, birth weight.
Also considered potential
confounding by site, sex,
birth order, caregiver
education and marital
status, parenting stress,
and maternal utilization of
prenatal and annual
health care but not
parental caregiving
quality.

-0.69 (-4.4, 3.0)
-6.7 (-12, -1.2)

tYorifuii et al. (2011)

Faroese island
Denmark

1986-1987 (enrollment)
Followed through 7-14 yr

Cohort

Birth cohort

n: At age 7: 896, At age
14: 808

Birth cohort of mother-
infant pairs

Blood, hair

Cord blood;
electrothermal AAS.
Age at measurement:
At birth

GM of cord blood Pb:
1.57 |jg/dL
75th: 2.2 pg/dL

Verbal and visuospatial
reasoning, language,
learning, and memory
assessed using WISC-R
similarities, WISC-R block
designs, BNT, and CVLT-
C.

Age at outcome:

7, 14 yr

Age, sex, maternal
Raven's score, paternal
employment and
education, maternal
education, daycare at age
7, medical risk, and
maternal alcohol use and
smoking during
pregnancy

Beta

WISC-R at 7 yr old with
cord mercury
Block Design: -0.011
(-0.083, 0.062)

Similarities: -0.122
(-0.38, 0.135)

Digit Span Forward: -0.1
(-0.183, -0.016)

WISC-R at 7 yr old

3-284

-------
Reference^and Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Block Design: -0.004
(-0.013, 0.006)

Similarities: 0.019
(-0.051, 0.088)

Digit Span Forward:
-0.028 (-0.049, -0.006)

CVLT-C at 14 yr old plus
interaction with cord
mercury

Recognition: -0.053
(-0.136, 0.031)

Long-term Recall: -0.1
(-0.256, 0.057)

Short-term Recall: -0.009
(-0.178, 0.16)

Learning: -0.438 (-0.965,
0.089)

CVLT-C at 14 yr old
Recognition: -0.001
(-0.022, 0.021)

Long-term Recall: 0.041
(0, 0.082)

Short-term Recall: 0.013
(-0.031, 0.058)

3-285

-------
Reference^and Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Learning: 0.037 (-0.103,
0.177)

Boston Naming Test at 14
yr old plus interaction with
cord mercury
With Cues: 0.002 (-0.337,
0.342)

No Cues: -0.095 (-0.473,
0.283)

Boston Naming Test at 14
yrold

With Cues: 0.033 (-0.056,
0.122)

No Cues: 0.003 (-0.096,
0.102)

WISC-R at 14 yr old plus
interaction with cord
mercury

Block Design: 0.241
(-0.575, 1.057)

Similarities: -0.147
(-0.383, 0.089)

Digit Span Backward:
-0.16 (-0.253, -0.067)

Digit Span Forward:
-0.107 (-0.199, -0.016)

3-286

-------
Reference^and Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Digit Span: -0.267
(-0.423, -0.111)

WISC-R at 14 yr old
Block Design: -0.023
(-0.238, 0.191)

Similarities: -0.007
(-0.069, 0.055)

Digit Span Backward:
-0.035 (-0.06, -0.009)

Digit Span Forward:
-0.024 (-0.048, 0)

Digit Span: -0.059 (-0.1,
-0.017)

CVLT-C at 7 yr old plus
interaction with cord
mercury

Recognition: -0.094
(-0.2, 0.011)

Long-term Recall: 0.037
(-0.141, 0.215)

Short-term Recall: -0.068
(-0.22, 0.084)

Learning: -0.501 (-0.981,
-0.02)

CVLT-C at 7 yr old

3-287

-------
Reference^and Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Recognition: -0.003
(-0.032, 0.026)

Long-term Recall: 0.033
(-0.014, 0.081)

Short-term Recall: 0.043
(0.003, 0.084)

Learning: 0.073 (-0.057,
0.202)

Boston Naming Test at 7
yr old plus interaction with
cord mercury
With Cues: -0.138
(-0.469, 0.193)

No Cues: -0.046 (-0.376,
0.284)

Boston Naming Test at 7
yrold

With Cues: 0.042 (-0.045,
0.129)

No Cues: 0.039 (-0.049,
0.127)

tTatsuta et al. (2014)

Sendai, Tohoku region
Japan

Study years NR
Followed through 42 mo

TSCD birth cohort
n: 387

Mother-infant pairs urban
areas of the Tohoku
district

Blood

Cord blood; ICP-MS.

Age at measurement:
Delivery

Intelligence and
achievement (K-ABC)

Age at outcome:
42 mo

Child sex, birth order,
alcohol and smoking
habits, duration of
breastfeeding, annual
family income at 42 mo,
and maternal IQ (Raven
SPM)

Beta

Adjusted Model

Mental Processing Score:
-3.319 (-12.41, 5.774)

3-288

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Cohort



Median: 1.0 pg/dL
Max: 1.8 pg/dL





Sequential Processing
Score: -2.136 (-12.80,
8.531)

tOppenheimer et al.
(2022)

New Bedford Harbor,

Massachusetts

United States

1993-1998 (enrollment)

Followed through 2008-

2014

Cohort

New Bedford Cohort
n: 373

Blood

Cord blood; isotope
dilution ICP-MS

Age at measurement:
Delivery

Mean (SD): 1.4 (0.9)
pg/dL

Max: 9.4 |jg/dL

Cognitive Effects

Cognitive effects were
assessed using four
subtests of the Delis-
Kaplan Executive
Function System. These
included Trail Making:
Number-Letter Switching
condition, Verbal Fluency:
Category Switching
condition, Design
Fluency: Filled Dots and
Empty Dots Switching
condition, and Color-Word
Interference:
Inhibition/Switching
condition.

Multiple linear regression
models adjusted for child
race, sex, age at exam,
year of birth, HOME
score, maternal marital
status at child's birth,
maternal IQ, maternal
seafood consumption
during pregnancy,
maternal smoking during
pregnancy, maternal and
paternal education, and
annual household income
at child's birth, and study
examiner.

Beta

WRAML

Verbal Working Memory:
0.12 (-0.20, 0.45)

Symbolic Working
Memory: 0.09 (-0.246,
0.42)

Working Memory Index
Differences: 0.59 (-0.97,
2.15)

Age at Outcome:
14-18 yr

Choet al. (2010)

Seoul (metropolitan),
Seongnam (suburban),
Ulsan and Incheon
(industrial), and
Yeoncheon (rural)
South Korea

2009

Cross-sectional

n = 639 children (8-11 yr) Blood

Color-Word score

School-based recruitment Child blood; GFAAS with SCWT

Zeeman background
correction

Age at measurement:
8-11 yr

Mean (SD): 1.9 (0.67)
pg/dL

10th—90th: 1.2-2.8 pg/dL

Age at outcome:
8-11 yr

Age, sex, paternal
education, maternal IQ,
child IQ, birth weight,
urinary cotinine,
residential area. Did not
consider potential
confounding by parental
caregiving quality.

Beta

0 (-0.02, 0.02)

3-289

-------
Reference^and Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tFruh etal. (2019)

Eastern Massachusetts
U.S.

1999-2002 (enrollment)
Followed through 7 yr

Cohort

Project Viva
n: 1006

Birth cohort of mother-
child pairs

Blood

Maternal venous
erythrocyte blood
specimens; ICP-MS

Age at measurement:

2nd to 3rd trimester of
pregnancy (median: 27.9
wk)

Med (IQR): 1.1 (0.06)
pg/dL

Executive Function (see
also Section 3.5.1)

Parent teacher ratings on
BRIEF

Age at outcome:

7 yr

Scores standardized for
child age and sex;
Additional adjustment for
maternal 2nd trimester Hg
and Mn levels, nulliparity,
smoking during
pregnancy, IQ, and
education; Paternal
education; HOME
composite score and
household income; and
child race/ethnicity.

Beta

BRIEF Parent-Reported

Behavioral Regulation
Index

All: 1.15 (-0.217, 2.517)

Girls: 1.717 (0.025, 3.408)

Boys: 0.85 (-1.058,

2.758)

Metacognition Index
All: 0.95 (-0.25, 2.15)

Girls: 1.483 (-0.108,
3.075)

Boys: 0.6 (-1.05, 2.25)

General Executive
Composite

All: 1.217 (-0.1, 2.533)

Girls: 1.95 (0.1, 3.8)

Boys: 0.783 (-0.958,
2.525)

BRIEF Teacher-
Reported

Behavioral Regulation
Index

All: 0.767 (-0.567, 2.1)

3-290

-------
Reference^and Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Girls: 0.933 (-1.142,
3.008)

Boys: 0.75 (-0.9, 2.4)

Metacognition Index
All: 0.683 (-0.717, 2.083)

Girls: 0.9 (-1.308, 3.108)

Boys: 0.683 (-1.142,
2.508)

General Executive

Composite

All: 0.7 (-0.65, 2.05)

Girls: 0.883 (-1.258,
3.025)

Boys: 0.683 (-1.008,
2.375)

tFruh etal. (2021)

Eastern Massachusetts,
U.S.

1999-2002 (enrollment),
Followed through 6-11 yr

Cohort

Project Viva
n: 2128

Birth cohort of mother-
child pairs

Blood

Maternal venous
erythrocyte blood
specimens; ICP-MS

Age at measurement:
2nd to 3rd trimester of
pregnancy (median: 27.9
wk)

Global Executive
Composite (GEC);
Difficulties score

BRIEF; SDQ

Age at Outcome:
6-11 yr

Scores on BRIEF
Total standardized forage, sex
and model adjusted for
child race/ethnicity,
maternal parity, maternal
smoking status, maternal
IQ, maternal education,
paternal; education,
hemoglobin, HOME
score, household income,
and fish consumption.
SDQ model adjusted for
age, sex, and above
covariates. Pb, Mn, Se

Beta

SDQ Total Difficulties:
0.617 (-0.058, 1.292)

BRIEF GEC:
1.11 (-0.12, 2.34)

3-291

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa



Med (IQR): 1.1 (0.06)
pg/dL



and MeHg included
together in the models.



tRuebner et al. (2019)

46 centers
U.S.

Study Years: NR
Followed through: 1-16 yr

Cohort

CKiD Cohort study
n: 412

Children with mild to
moderate CKD

Blood

Child venous blood; ICP-
MS. The BLL
measurement closest to
the time of neurocognitive
testing was used for
analysis (concurrent).

Age at measurement:
NR; 2, 4, or 6 yr after
study entry

Median: 1.2 |jg/dL
75th: 1.8 |jg/dL
Max: 5.1 |jg/dL

Executive function (see
also Section 3.5.1 [FSIQ],
Section 3.5.2 [attention
and hyperactivity])

Age-specific assessments
administered at visit 3, 5,
7, or 9. Last available
results used (mean time
between BLL and
outcome assessment =
2.3 yr). Delis-Kaplan
Executive Function
System Tower Subset (>6
yr), BRIEF-P (2-5 yr),
BRIEF (6-18), BRIEF-A
(>18 yr)

Age at outcome:
1 to >18 yr

Age, sex, race, poverty,
maternal education.

Adjusted BRIEF results
were not reported
because they were not
statistically significant.

tMerced-Nieves et al.
(2022)

Mexico City
Mexico

2007-2011 (enrollment)
Followed through 6-7 yr
Cohort

PROGRESS Cohort
n = 549

Birth cohort

Blood

Maternal, cord, and child
blood; Agilent 8800 ICP
Triple Quad

Age at measurement:
Maternal: T2, T3, delivery
Cord: delivery
Postnatal: 4-6 yr

Mean (SD)

Maternal T2: 2.7 (2.7)
pg/dL

Various measures from
Condition Position
Responding (CPR),
Temporal Response
Differentiation (TRD),
Delayed Matching-to-
Sample (DMTS), and
Incremental Repeated
Acquisition (IRA) from the
OTB

Age at Outcome:
6-7 yr

Child's age at testing, Betas for BLL at T3
maternal education (high school), and SES
Modification by sex
examined.

Observing response
latency:

0.001s (-0.08, 0.08s)

TRD

Average latency:
0.14s (-0.001, 0.29s)

DMTS

Average observing
response latency:
0.08s (-0.04, 0.20s)

3-292

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa



Maternal T3: 3.9 (2.8)
pg/dL

Maternal at delivery: 4.3
(3.2) pg/dL

Cord: 3.4 (2.6) pg/dL

Child: 2.4 (2.6) pg/dL





IRA

Effective response rate:
-0.01s (-0.03, -0.002s)

AAS = atomic absorption spectrometry; BLL = blood lead level; BNT = Boston Naming Test; BRIEF = Behavior Rating Inventory of Executive Functions; CANTAB = Cambridge
Neuropsychological Test Automated Battery; CI = confidence interval; CKD = chronic kidney disease; CKiD = Chronic Kidney Disease in Children Study; CVLT-C = California Verbal
Learning Test-Children's version; ETS = environmental tobacco smoke; FSIQ = full-scale IQ; GFAAS = graphite furnace atomic absorption spectrometry; GM = geometric mean;
HOME = Health Outcomes and Measures of the Environment; ICP-MS = inductively coupled plasma mass spectrometry; K-ABC = Kaufman Assessment Battery for Children; MANAs
= Metals, Arsenic and Nutrition in Adolescents Study; MeHg = methyl mercury; NHANES = National Health and Nutrition Examination Survey; NR = not reported; OTB = Operant Test
Battery; Pb = lead; SCWT = Stroop Color-Word test; SD = standard deviation; SDQ = Strengths and Difficulties Questionnaire; SES = socioeconomic status; SPM = Standard
Progressive Matrices; T1 = first trimester of pregnancy; T2 = second trimester of pregnancy; T3 = third trimester of pregnancy; TSCD = Tohoku Study of Child Development; WISC =
Weschler Intelligence Scale for Children; WRAML = Wide Range Assessment of Memory and Learning; WRAT = Wide Range Achievement Test; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
tStudies published since the 2013 Integrated Science Assessment for Lead.

Table 3-4T Animal toxicological studies of Pb exposure and cognitive function

Studv Species (Stock/Strain), n, Timing of
" Sex Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Corv-Slechta et al. (2012) Rat (Lonq-Evans) GD-60 to 10 mo

Control (tap water), M, n =

12

Oral, drinking water
Oral, lactation
In utero

PND 5-6:

<5 pg/dL for Control
12.5 pg/dL for 50 ppm

2-3 mo to 10 mo:
Operant Behavior

50 ppm, M, n = 12



2.5 mo:

<5 pg/dL for Control
6.43 pg/dL for 50 ppm
10 mo:



<5 |jg/dL for Control
8.98 |jg/dL for 50 ppm

3-293

-------
Study Species (Stock/Strain), n,	Timingof Exposure Details BLL as Reported (Mg/dL)	eSKKS

Zou et al. (2015) Mouse (ICR)	~5wkto8wk Oral, drinking water 8wk:	8 wk: Morris

Control (distilled water), M,	water maze

n = 10	1.8 |jg/dL for Control

250 mg/L solution, M, n =	21.7 |jg/dL for 250 mg/L
10

3-294

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

Corv-Slechta et al. (2013) Mouse (C57BL/6) GD-60 to 12 mo

Oral, drinking water

PND 75 - Females:

7-12 mo:

Control (distilled deionized

Oral, lactation



Operant Behavior

water) - NS, M/F, n = 10-

In utero

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

Weston etal. (2014)

Rat (Long-Evans) GD -60 to PND 21

Control (tap water), M/F, n
= 22(11/11)

50 ppm, M/F, n = 22 (11/11)

Oral, lactation
In utero

PND 5-6 - Males:
0.76 pg/dL for Control
15.7 pg/dL for 50 ppm
PND 5-6 - Females:
0.82 pg/dL for Control
14.7 pg/dL for 50 ppm

>PND 60:

Operant Behavior

Betharia and Maher

Rat (Sprague Dawley) GD 0 to PND 20

Oral, lactation

PND 2:

PND 21-25, 56-

(2012)

PND 21-25:

In utero



60: Morris water







1.77 ng/g (0.188 pg/dL) for

maze



Control (RO Dl water), M/F,



Control





n = 11-13













85.17 ng/g (9.02 pg/dL) for 10





10 pg/mL, M/F, n = 11-13



pg/mL





PND 56-60:



PND 25:





Control (RO Dl water), M/F,



0.83 ng/g (0.088 pg/dL) for





n = 9-11



Control





10 pg/mL, M/F, n = 9-11



9.21 ng/g (0.98 pg/dL) for 10









pg/mL









PND 60:









0.23 ng/g (0.024 pg/dL) for









Control









0.30 ng/g (0.032 pg/dL) for 10









pg/mL



3-296

-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Han et al. (2014)

Rat (Wistar)

Control (tap water), M, n =
2 mM - PW, M, n = 8
2 mM - ME, M, n = 8

PW group: PND 21
8 to PND 42

ME group: GD -21
to PND 20

Oral, drinking water
Oral, lactation
In utero

PND 21:

7.36 |jg/L (0.74 pg/dL) for Control

NR for 2 mM - PW

146.6 pg/L (14.7 pg/dL) for 2 mM
-ME

PND 63:

9.22 pg/L (0.92 pg/dL) for Control

147.9 pg/L (14.8 pg/dL) for 2 mM
-PW

46.13 pg/L (4.6 pg/dL) for 2 mM -
ME

PND 63 to PND
68: Morris water
maze

Flores-Montova et al.
(2015)

Mouse (C57BL/6)	GD 0 to PND 28

Control (Sodium treated
water), M/F, n = 10 (8/2)

30 ppm, M/F, n = 10 (5/5)

330 ppm, M/F, n = 13 (7/6)

Oral, drinking water

PND 28 - Females:
0.02 pg/dL for Control
2.63 pg/dL for 30 ppm
12.92 pg/dL for 330 ppm
PND 28-Males:
0.31 pg/dL for Control
3.10 pg/dL for 30 ppm
15.21 pg/dL for 330 ppm

PND 28: Novel
Odor Recognition

Rahman et al. (2012a)

Rat (Wistar)

Control, M/F, n = 6

0.2% solution (0.002 g/mL),
M/F, n = 10

PND 1 to PND 21 Oral, drinking water

PND 21:

1.35 pg/dL for Control
12.40 pg/dL for 0.2% solution

PND 21: Morris
water maze

3-297

-------
Study

Species (Stock/Strain), n,
Sex

Exposure Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Rahman et al. (2012b)

Rat (Wistar)

PND 1 to PND 30 Oral, drinking water

PND 21:

PND 21, 30:



Control (tap water), M/F, n

Oral, lactation



Morris water



= 6



1.4 pg/dL for Control

maze



0.2% solution, M/F, n = 10



12.1 pg/dL for 0.2% solution









PND 30:









1.2 pg/dL for Control









12.8 pg/dL for 0.2% solution



Mansouri et al. (2012)

Rat (Wistar)

Control (distilled water),
M/F, n = 16 (8/8)

50 mg/L, M/F, n = 16 (8/8)

PND 70 to PND 100 Oral, drinking water

PND 100-Males:
2.05 pg/dL for Control
8.8 pg/dL for 50 mg/L
PND 100 - Females:
2.17 pg/dL for Control
6.8 pg/dL for 50 mg/L

PND 100: Morris
water maze,
Novel Object
Recognition

3-298

-------
Anderson et al. (2016)

Rat (Long-Evans)

Control (untreated), M/F, n
= 16 (8/8)

150 ppm, M/F, n = 16 (8/8)
per duration

375 ppm, M/F, n = 16 (8/8)
per duration

750 ppm, M/F, n = 16 (8/8)
per duration

Perinatal exposure
group: GD -10 to
PND 21

Early postnatal
exposure group:
PND 0 to PND 21

Long-term postnatal
exposure group:
PND 0 to PND 55

Oral, diet
Oral, lactation
In utero

3-299

PND 65 - Perinatal exposure
females:

0 |jg/dL for Control
1.36 |jg/dL for 150 ppm
2.13 |jg/dL for 375 ppm

2.08	|jg/dL for 750 ppm

PND 65 - Early postnatal
exposure females:

0 |jg/dL for Control

2.11 |jg/dL for 150 ppm

2.0 |jg/dL for 375 ppm

3.09	|jg/dL for 750 ppm

PND 65 - Long-term exposure
females:

0 |jg/dL for Control

4.5 |jg/dL for 150 ppm

5.75 |jg/dL for 375 ppm

9.58 |jg/dL for 750 ppm

PND 65 - Perinatal exposure
males:

0 |jg/dL for Control
1.25 |jg/dL for 150 ppm
2.42 |jg/dL for 375 ppm
2.47 |jg/dL for 750 ppm

PND 55, 56, 57,
and 65: Trace
Fear Conditioning

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

PND 65 - Early postnatal
exposure males:

0 |jg/dL for Control

1.64 |jg/dL for 150 ppm

1.95 |jg/dL for 375 ppm

2.83 |jg/dL for 750 ppm

PND 65 - Long-term exposure
males:

0 |jg/dL for Control
2.01 |jg/dL for 150 ppm
8.0 |jg/dL for 375 ppm
7.46 |jg/dL for 750 ppm

7.61 |jg/L (0.76 |jg/dL) for Control PND 35^0:

Morris water
84.3 |jg/L (8.43 |jg/dL) for 300 maze
ppm

Mena et al. (2016)	Rat (Sprague Dawley) PND 0 to PND 21 Oral, lactation

Control (deionized water),

M/F, n = not specified

300 ppm Pb, M/F, n = not
specified

3-300

-------
Study

Species (Stock/Strain), n, Timing of
Sex	Exposure

Exposure Details	BLL as Reported (pg/dL)

Li etal. (2016a)

Mouse (Kunming)

GD 1 to PND 21 Oral, drinking water

PND 21:

PND 21, 22, 23,



Control (untreated), M/F, n





24, 25, 26: Morris



= 10



9.8 |jg/L (0.98 pg/dL) for Control

water maze



0.1% solution (1000 ppm),



42.5 |jg/L (4.25 pg/dL) for 1,000





M/F, n = 10



ppm





0.2% solution (2000 ppm),



85.3 pg/L (8.53 pg/dL) for 2000





M/F, n = 10



ppm





0.5% solution (5000 ppm),



106.4 pg/L (10.64 pg/dL) for 5000





M/F, n = 10



ppm



Li etal. (2016c)

Mouse (Kunming)

GD 0 to PND 21 Oral, lactation

PND 21:

PND 21: Morris



Control (distilled water),

In utero



water maze



M/F, n = 10



10.62 pg/L (1.1 pg/dL) for Control





0.1% solution (mass



40.71 pg/L (4.1 pg/dL) for 0.1%





fraction), M/F. n = 10



solution





0.2% solution (mass



81.77 pg/L (8.2 pg/dL) for 0.2%





fraction), M/F, n = 10



solution





0.5% solution (mass



103.36 pg/L (10.3 pg/dL) for





fraction), M/F, n = 10



0.5% solution



Mena et al. (2016)

Rat (Sprague Dawley)

PND 1 to PND 21 Oral, lactation

PND 35:

NR: Morris water



Control (deionized water),





maze



M/F, n = 7



5.6 pg/L (0.56 pg/dL) for Control





300 ppm, M/F, n = 7



84.84 pg/L (8.48 pg/dL) for 300



ppm

3-301

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

Wana et al.

(2013)

Rat (Sprague Dawley)

GD Oto PND 1,

Oral, drinking water

PND 72:

PND 65 to PND





Control (untreated), M/F, n

PND 1 to PND 21,

Oral, lactation



69: Morris water





= 6

PND 21 to 42

In utero

34.99 |jg/L (3.5 pg/dL) for Control

maze





0.2% solution (w/v), M/F, n





35.78 pg/L (3.58 pg/dL) for 0.2 %







= 6 - Gestational Exposure





solution Gestational







0.2% solution (w/v), M/F, n





65.97 pg/L (6.60 pg/dL) for 0.2%







= 6 - Lactational Exposure





solution Lactational







0.2% solution (w/v), M/F, n





110.67 pg/L (11.07 pg/dL) for







= 6 - Ablactational





0.2% solution Ablactational







Exposure









Wana et al.

(2016)

Rat (Sprague Dawley)

PND 24 to PND 56

Oral, drinking water

PND 56:

PND 60-66:





Control (tap water), M, n = 7







Trace Fear











11 pg/L (1.1 pg/dL) for Control

Conditioning





100 ppm, M, n = 9



















133 pg/L (13.3 pg/dL) for 100













ppm



Zhanq et al.

(2014)

Mouse (Kunming)

GD Oto PND 21

Oral, lactation

PND 36:

PND 29, 30:





Control (distilled water),



In utero



Passive





M/F, n = 12





18.5 pg/L (1.9 pg/dL) for Control

Avoidance Test,













PND 31-35:





0.4% solution, M/F, n = 13





136.7 pg/L (13.7 pg/dL) for 0.4%

Morris water











Solution

maze

3-302

-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Barkur and Bairv (2015b) Rat (Wistar)	GD -30 to PND 21 Oral, lactation

Control (untreated), M, n =	In utero

6

0.2% solution - Gestational,

M, n = 6

0.2% solution - Lactational,

M, n = 6

0.2% solution - Gestation +

Lactation, M, n = 6

0.2% solution -
Pregestational, M, n = 6

PND 22:

0.18 pg/dL for Control

3.02 pg/dL for 0.2% solution -
Pregestation

5.30 pg/dL for 0.2% solution -
Gestational

26.65 pg/dL for 0.2% solution -
Lactational

32.0 pg/dL for 0.2% solution -
Gestation + Lactation

PND 30 to PND
36: Morris water
maze, PND 26,
27, 28: Passive
Avoidance Test

Barkur et al. (2011)

Rat (Wistar)	GD 0 to PND 21

Control (tap water), M, n = 9

0.2% solution (w/v), M, n =

9

Oral, lactation
In utero

PND 120:

0.24 pg/dL for control
0.47 pg/dL for 0.2% solution

PND 120:
Passive

Avoidance Test

3-303

-------
Verma and Schneider Rat (Long-Evans)
(2017)	Control, M/F, n = 32 (16/16)

PERI: GD -14 to Oral, lactation
PND 21	In utero

150 ppm chow (PERI), M/F,
n = 32 (16/16)

150 ppm chow (EPN), M/F,
n = 32 (16/16)

EPN: PND 0 to PND
21

3-304

PND 14-PERI Males:
-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined









-------
Verma and Schneider Rat (Sprague Dawley)

(2017)	Control, M/F, n = 36 (18/18)

PERI: GD -14 to Oral, lactation
PND 21	In utero

150 ppm Chow (PERI),
M/F, n = 36 (18/18)

150 ppm Chow(EPN), M/F,
n = 36 (18/18)

EPN: PND 0 to PND
21

3-306

PND 14-PERI Males:
-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined









-------
Anderson et al. (2012) Rat (Long-Evans)	GD-10toPND21 Oral, lactation

Control (untreated chow),

M/F, n = 28 (11/17)

250 ppm Chow, M/F, n = 15
(10/5)

750 ppm Chow, M/F, n = 25
(13/12)

1500 ppm Chow, M/F, n =

23 (12/11)

PND 1 - Males:	PND 55: Morris

water maze

0 |jg/dL for Control
18.9 |jg/dLfor250 ppm
52.5 |jg/dL for 750 ppm

52.5	|jg/dL for 1500 ppm
PND 1 - Females:

0 |jg/dL for Control
21.9 |jg/dLfor250 ppm
47.2 |jg/dL for 750 ppm
56.7 |jg/dL for 1500 ppm
PND 7-Males:

0 |jg/dL for Control
8.5 |jg/dL for 250 ppm
29.1 |jg/dL for 750 ppm
35.7 |jg/dL for 1500 ppm
PND 7 - Females:

0 |jg/dL for Control
14.7 |jg/dLfor250 ppm
26.9 |jg/dL for 750 ppm

37.6	|jg/dL for 1500 ppm
PND 14- Males:

0 |jg/dL for Control

3-308

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

10.5	[jg/dL for 250 ppm

18.6	[jg/dL for 750 ppm
24.8 [jg/dL for 1500 ppm
PND 14 - Females:

0 [jg/dL for Control
11.8 [jg/dL for 250 ppm
20.2 [jg/dL for 750 ppm
26.4 [jg/dL for 1500 ppm

PND 21 - Males:

0 [jg/dL for Control

18.6	[jg/dL for 250 ppm

28.8	[jg/dL for 750 ppm

28.7	[jg/dL for 1500 ppm
PND 21 - Females:

0 [jg/dL for Control

17.9	[jg/dL for 250 ppm
27.4 [jg/dL for 750 ppm

29.8	[jg/dL for 1500 ppm

3-309

-------
Zhao et al. (2018)	Rat (Sprague Dawley) GD-14toPND10 Oral, lactation

Control (tap water), M, n = 8	In utero

0.005% solution, M, n = 8

0.01% solution, M, n = 8

0.02% solution, M, n = 8

3-310

PND 0:

1.9 |jg/dL for Control
17.9 |jg/dL for 0.005% solution

23.2	|jg/dL for 0.01% solution
48.8 |jg/dL for 0.02% solution
PND 3:

I.9	|jg/dL for Control

6.7 |jg/dL for 0.005% solution

II.5	|jg/dL for 0.01% solution
23.1 |jg/dL for 0.02% solution
PND 7:

1.3 |jg/dL for Control

8.1	|jg/dL for 0.005% solution
12.3 |jg/dL for 0.01 % solution
18.7 |jg/dL for 0.02% solution
PND 10:

1.2	|jg/dL for Control

5.6 |jg/dL for 0.005% solution
7.0 |jg/dL for 0.01% solution

12.3	|jg/dL for 0.02% solution
PND 14:

0.7 |jg/dL for Control

PND 30: Morris
water maze

-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

4.0	pg/dL for 0.005% solution
5.5 pg/dL for 0.01% solution
8.9 pg/dL for 0.02% solution
PND 21:

1.1	pg/dL for Control

2.5 pg/dL for 0.005% solution
2.5 pg/dL for 0.01% solution
2.98 pg/dL for 0.02% solution
PND 30:

1.5 pg/dL for Control
1.0 pg/dL for 0.005% solution
1.5 pg/dL for 0.01% solution
1.5 pg/dL for 0.02% solution

Neuwirth et al. (2019b)

Rat (Long-Evans)

Control (tap water), M/F, n
= 12 (6/6)

363.83 |jM solution, M/F, n
= 12 (6/6)

GD Oto PND 22

Oral, lactation
In utero

PND 22:

NR for Control

5.3-15 pg/dL for 364 pM solution
PND 56-90:

ND for Control,

ND for 364 pM

PND 56-90:
Attention Set
Shifting Test

3-311

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

Neuwirth et al. (2019c)

Rat (Long-Evans)

PERI: GD -14 to

Oral, lactation

PND 14 - Females:

NR: Attention Set



Control, M/F, n = 12 (6/6)

PND 22

In utero



Shifting Test









-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Tartaqlione et al. (2020) Rat (Wistar)

GD -28 to PND 23

Control (tap water), M/F n =
16 (9/7)

50 mg/L, M/F, n = 16 (9/7)

Oral, lactation
In utero

PND 23:

0.007 |jg/mL (0.7 pg/dL) for
Control

0.255 pg/mL (25.5 pg/dL) for 50
mg/L

PND 35: Y Maze
- Spontaneous
Alternation, PND
63-65: Novel
Object

Recognition, PND
68-72: Morris
water maze

Xiao et al. (2014)

Rat (Wistar)

Control (tap water), M/F, n
= 10 (5/5)

Pre-weaning: 2 mM
solution, M/F, n = 10 (5/5)

Postweaning: 2 mM
solution, M/F, n = 10 (5/5)

Pre-weaning: GD
-21 to PND 21

Postweaning: PND
21 to PND 84

Oral, drinking water
Oral, lactation
In utero

PND 21 - Pre-weaning:

10.09 pg/L (1 pg/dL) for Control

103.8 pg/L (10.4 pg/dL) for 2 mM
solution

PND 21 - Postweaning:

Not Reported

PND 91 - Pre-weaning:

10.32 pg/L (1 pg/dL) for Control

39.27 pg/L (3.9 pg/dL) for 2 mM
solution

PND 91 - Postweaning:

10.32 pg/L (1 pg/dL) for Control

105.45 pg/L (10.5 pg/dL) for 2
mM solution

PND 85 to 90:
Morris water
maze

3-313

-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Sobolewski et al. (2020) Mouse (C57BL/6)

F0:

Control (distilled Dl water),
F, n = 10

100 ppm, F, n = 10
F1:

see Figure 1, n = 12
F2:

see Figure 1, n = 12
F3:

see Figure 1, n = 8-10

F1: GD -60 to PND Oral, lactation

23-27

In utero

F1 PND 6-7:

0 pg/dL for Control

12.5 pg/dL for 100 ppm (F0
dosing)

F3 PND 6-7:

0 ng/dL for Control

0 pg/dL for 100 ppm (F0 dosing)

PND 60-120
(variable by
endpoint): Fl
Training

Ouvana et al. (2019)

Rat (Sprague Dawley) GD 0 to PND 679

Control (tap water), M/F, n
= 6-10

0.05/0.01% solution, M/F, n
= 6-10

Oral, drinking water
Oral, lactation
In utero

wk 97:

0 mg/L (0 pg/dL) for Control

0.216 mg/L (21.6 pg/dL) for
0.05/0.01% solution

PND 674 to PND
679: Morris water
maze

Singh et al. (2019)

Rat (Wistar)	3 mo to 6 mo

Control (distilled water), M,
n = 5

2.5 mg/kg, M, n = 5

Oral, gavage

6 mo:

5.76 pg/dL for Control
28.4 pg/dL for 2.5 mg/kg

6 mo: Morris
water maze

Xiao et al. (2020)

Rat (Sprague Dawley)

Control (tap water), F, n
10

125 ppm, F, n = 10

GD -7 to PND 68

Oral, drinking water
Oral, lactation
In utero

PND 68:

24.23 ng/mL (2.4 pg/dL) for
Control

205 ng/mL (20.5 pg/dL) for 125
ppm

PND 56 -61:
Morris water
maze, PND 55: Y
Maze -
Spontaneous
Alternation

3-314

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

Su etal. (2016)

Rat (Sprague Dawley)

Control (deionized water
with 0.9% saline), M, n = 15

200 ppm, M, n = 16

PND20toPND76 Oral, gavage

PND 76:

7.99 |jg/L (0.8 |jg/dL) for Control

84.17 |jg/L (8.4 pg/dL) for 200
ppm

PND 76: Morris
water maze

3-315

-------
An et al. (2014)	Rat (Sprague Dawley) 4 wk to 12 wk

Control (deionized water
with NaAc), M, n = 12

100 ppm, M, n = 12

200 ppm, M, n = 12

300 ppm, M, n = 12

Oral, drinking water

5wk:	12-wk: Morris

water maze

0.96 |jg/dL for Control
7.07 |jg/dL for 100 ppm
11.54 |jg/dLfor200 ppm
14.76 |jg/dL for 300 ppm

6	wk:

0.96 |jg/dL for Control
8.13 |jg/dL for 100 ppm
12.92 |jg/dL for 200 ppm
16.65 |jg/dL for 300 ppm

7	wk:

0.96 |jg/dL for Control
9.68 |jg/dL for 100 ppm
13.37 |jg/dL for 200 ppm
19.48 |jg/dL for 300 ppm

8	wk:

0.96 |jg/dL for Control
9.64 |jg/dL for 100 ppm
17.07 |jg/dL for 200 ppm
22.02 |jg/dL for 300 ppm

9	wk:

0.96 |jg/dL for Control

3-316

-------
Study	Species (Stock/Strain), n, Timingof	Exposure Details	BLL as Reported (Mg/dL)	eSKKS

12.12 [jg/dL for 100 ppm
20.7 [jg/dL for 200 ppm
22.28 [jg/dL for 300 ppm

10	wk:

0.96 [jg/dL for Control
11.48 [jg/dL for 100 ppm
17.75 [jg/dL for 200 ppm
24.69 [jg/dL for 300 ppm

11	wk:

0.96 [jg/dL for Control

11.51	[jg/dL for 100 ppm

17.52	[jg/dL for 200 ppm
22.18 [jg/dL for 300 ppm

12	wk:

0.96 [jg/dL for Control
11.41 [jg/dL for 100 ppm
17.23 [jg/dL for 200 ppm
22.57 [jg/dL for 300 ppm

3-317

-------
Study

Species (Stock/Strain), n, Timing of
Sex	Exposure

Exposure Details	BLL as Reported (pg/dL)

Li etal. (2013)

Rat (Wistar)

Control (tap water), M/F, n
= 16 (8/8)

500 ppm, M/F, n = 16 (8/8)

4 wk to 16 wk

Oral, drinking water

4 mo:

29.99 |jg/L (3 pg/dL) for Control

159.54 |jg/L (16 |jg/dL) for 500
ppm

4 mo: Morris
water maze

Zhu et al. (2019b)

Rat (Sprague Dawley)

GD 0 to 12 mo

Oral, drinking water

12 mo:

NR: Morris water



Control (deionized water),



Oral, lactation



maze



M/F, n = 32



In utero

-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Zhang et al. (2012)

Rat (Sprague Dawley)

Control (deionized water),

NR (40-60 g)

Oral, drinking water

+8 wk from start of exposure: +8 wk from start

of exposure:

M, n = 10



49.9 ng/mL (5 pg/dL) for Control

100 ppm, M,

n = 10

100.9 ng/mL (10.1 pg/dL) for 100





ppm

200 ppm, M,

n = 10







128.6 ng/mL (12.9 pg/dL) for 200

300 ppm, M,

n = 10

ppm





147.7 ng/mL (14.8 pg/dL) for 300





ppm

maze

Hong et al. (2021)

Rat (Sprague Dawley) GD 0 to PND 21

Control (tap water), M/F, n
= 50

1 g/L Pb solution, M/F, n =

50

Oral, lactation
In utero

0.009 mg/L for Control, 0.291
mg/L for 1 g/L Pb - PND 21

PND 21-27:
Morris water
maze

Biioor et al. (2012)

Rat (Wistar)	GD 0 to PND 45

Control (deionized water),

M/F, n = 10

50 ppm, M/F, n = 10

Oral, drinking water
Oral, lactation
In utero

PND 45:

4.06 pg/dL for Control
10.65 pg/dL for 50 ppm

PND 45: Passive
Avoidance Test

Wang et al. (2021a)

Rat (Sprague Dawley)

Control (deionized water),
M, n = 8

0.05% solution, M, n = 8
0.1% solution, M, n = 8

GD Oto PND 21

Oral, lactation
In utero

PND 21:

23.1 |jg/L (2.31 pg/dL) for Control

248 pg/L (24.8 pg/dL) for 0.05%
solution

302 pg/L (30.2 pg/dL) for 0.1%
solution

PND 21: Morris
water maze

361 pg/L (36.1 pg/dL) for 0.2%
solution

3-319

-------
Study

Species (Stock/Strain), n,
Sex

Timing of
Exposure

Exposure Details

BLL as Reported (pg/dL)

Endpoints
Examined

Liu et al. (2022c)

Rat (Sprague Dawley)

Control (tap water), M, n =
10

0.2% solution, M, n = 10

PND 35 to PND 119 Oral, drinking water

PND 119:

10.9 |jg/L (1.09 pg/dL) for Control

176 pg/L (17.6 pg/dL) for 0.2%
solution

PND 119: Morris
water maze

Wang et al. (2021b)

Rat (Sprague Dawley)

Control (deionized water),
M/F, n = 12

GD -28 to PND 21

Oral, lactation
In utero

PND 21:

PND 21: Morris
water maze

23.9 pg/L (2.39 pg/dL) for Control

0.05% solution, M/F, n = 10

206 pg/L (20.6 pg/dL) for 0.05%
solution

Al-Qahtani et al. (2022)

Mouse (Albino)

Control (distilled water), M,
n = 10

0.2 mg/kg, M, n = 10

8-9 wk to 14-15 wk Oral, gavage

14-15 wk:

1.2 pg/100 mL (1.2 pg/dL) for
Control

7.1 pg/100 mL (7.1 pg/dL) for 0.2
mg/kg

NR: Active
Avoidance Test

Long et al. (2022)

Rat (Sprague Dawley)

Control (untreated), M, n
12

200 mg/L solution, M, n =
12

6 wk to 18 wk

Oral, drinking water

18 wk:

2.14 pg/L (0.214 pg/dL) for
Control

32.48 pg/L (3.25 pg/dL) for 200
mg/L solution

NR: Morris water
maze, NR: Active
Avoidance Test

BLL = blood lead level; CI = confidence interval; EPN = early postnatal; F = female; F1 = first filial generation; Fl
male; ME = maternal exposure; mo = month(s); NaAc = sodium acetate; NR = not reported; NS = no stress; Pb =
PW = postweaning; RO Dl = reverse osmosis deionized; SD = standard deviation; wk = week(s).

: fixed interval; GD = gestational day; LOD = limit of detection; M =
lead; PERI = perinatal; PND = postnatal day; PS = prenatal stress;

3-320

-------
Table 3-5E

Epidemiologic studies of Pb exposure, academic performance, and achievement

Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Chandramouli et al.
(2009)

Avon
U.K.

Jul.-Dec. 1992 (birth)
Followed 8 yr

10% random subsample
of Avon Longitudinal
Study of Parents and
Children (ALSPAC)
n = 488

School children

Blood

Earlier childhood venous
blood; AAS using micro
sampling flame
atomization

Age at measurement: 30
mo

Academic achievement

Standardized
Achievement Test

Age at outcome:

7 yr

Maternal education and
smoking, home
ownership, home
facilities score, family
adversity index, paternal
SES, parenting attitudes
at 6 mo, child sex. Also
considered child IQ

Per doubling BLLb
-0.3 (-0.5, -0.1)

Cohort

Mean (SD): NR
Group 1: 0-<2 |jg/dL
Group 2: 2-<5 |jg/dL
Group 3: 5—<10 |jg/dL
Group 4: >10 pg/dL

Miranda et al. (2009)

School children, n=

Blood

Academic achievement

Sex, age of blood Pb

Score vs. blood Pb



57,568

Surveillance database



measurement, race,

category 1 |jg/dL

North Carolina





4th grade EOG test score

enrollment in
free/reduced lunch

2 |jg/dL:

U.S.

Screened for Pb at age

Age at measurement: 9-

for reading (2001-2005)

program, parental

-0.30 (-0.58, -0.01)



9-36 mo in 100 NC

36 mo



education, charter

3 |jg/dL:

1995 through 1999

counties





school.

-0.46 (-0.73, -0.19)

(screening)









4 |jg/dL:











-0.52 (-0.79, -0.24)

Cohort









5 |jg/dL:











-0.80 (-1.08, -0.51)

3-321

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Min et al. (2009)

Cleveland, OH
1994-1996 (birth)
Followed to age 11 yr
Cohort

Birth cohort, n = 267

86% African-American
with high prevalence of
prenatal drug and alcohol
exposure

Blood

Earlier childhood (age 4
yr)

Mean (SD): 7.0 (4.1)

Interval analyzed: 3.0
(10th percentile)-10

WJTA (math and reading Sex, caregiver education, Math score: -2.5 (-4.6,

scores

Age at outcome:
11 yr

family income,
race/ethnicity, test
language.

-0.38)

Reading score: -2.9
(-4.4, -1.4)

Lanphear et al. (2000)

United States

1988-1994

Cross-sectional

U.S. NHANES
n = 4,853 children ages
6-16 yr (born 1972-
1988)

Large U.S.

representative study of
multiple risk factors and
outcomes

Blood

Concurrent
GM (SD): 1.9 (7.0)
63.5% <2.5
Detection limit = 0.5
Interval analyzed: 1-5

WRAT (arithmetic and
reading scores)

Age at outcome: 6-16 yr

Linear regression

Child sex, race/ethnicity,
poverty index ratio,
reference adult
education, serum ferritin
and cotinine levels. Did
not consider potential
confounding by parental
cognitive function or
caregiving quality.

-0.05 (-0.09, -0.01)

Krieq et al. (2010)

United States

1991-1994

Cross-sectional

U.S. NHANES
n = 773 children ages
12-16 yr (born 1972-
1982)

Large U.S.

representative study of
multiple risk factors and
outcomes

Blood

Concurrent
GM (SD): 1.9 (7.0)
63.5% <2.5
Detection limit = 0.5
Interval analyzed: 1-5

WRAT (arithmetic and
reading scores)

Age at outcome: 6-16 yr

Log-linear regression

Child sex, caregiver
education, family income,
race/ethnicity, test
language. Did not
consider potential
confounding by parental
cognitive function or
caregiving quality.

-0.34 (-0.59, -0.08)

3-322

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Surkan et al. (2007)

Boston, Massachusetts
and Farmington, Maine

United States
Cross-sectional

n = 389 children
6-10 yr

Recruitment from trial of
amalgam fillings

Blood

Concurrent
Group 1:1-2
Group 2: 3-4
Group 3: 5-10
Mean (SD): 2.2 (1.6)

WIAT (reading and math
composites)

Age at outcome: 6-10 yr

Caregiver IQ, child age,
SES, race, birth weight.
Also considered potential
confounding by site, sex,
birth order, caregiver
education and marital
status, parenting stress,
and maternal utilization
of prenatal and annual
health care but not
parental caregiving
quality.

-0.69
-6.7 (-

-4.4, 3.0)
12, -1.2)

Chiodo et al. (2007)
Detroit, Michigan
United States

Cross-sectional

n = 495 children (born
1989-1991) age 7 yr

Blood

Concurrent
Mean (SD): 5.0 (3.0)

Test of Early Reading
Ability—2

MAT (math and reading
scores)

Age 7 yr

Maternal

psychopathology, IQ,
prenatal smoking,
prenatal marijuana, SES,
HOME score, caretaker
education and marital
status, # children in
home, child sex. Also
considered child age,
maternal age, custody,
cocaine use, prenatal
alcohol use.

-0.19 (-0.30, —0.08)c

Ferausson et al. (1997) n = 881 children

Christchurch
New Zealand

1977 (birth)
Followed to age 18

Cohort

Tooth Pb (age 6-8 yr)
Mean (SD): 6.2 (3.7) pg/g

Percent leaving school
without school certificate

Ages 16-18 yr

Christchurch Health and
Development Study birth
cohort

Maternal age,
punitiveness, standard of
living, breastfeeding
duration, parental
conflict, grade, residence
on busy roads. Also
considered potential
confounding by sex,
ethnicity, maternal
education, family size,
HOME, SES, ethnicity,
parental change, birth
order, single parent.

0-2 pg/g: 15.6
3-5 pg/g: 16.7
6-8 pg/g: 18.1
9-11 pg/g: 19.7
12+ pg/g: 24.1
p < 0.05

3-323

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Needleman et al. (1990) n = 132 children (1st/2nd Tooth Pb (1st/2nd grade) Failure to graduate high Maternal age at birth,

Failure to graduate: 7.4

Chelsea and Somerville,
MA

United States

1975-1978 (enrollment)
followed to age 18 yr

Cohort

grade) in Massachusetts
schools

distribution
<10 ppm: 50%
10-19.9 ppm: 22.7%
>20 ppm: 27.3%

school

Highest grade achieved

Logistic regression

education, and IQ, family (1.4, 41) d

size, SES, sex, age at
testing, birth order,
alcohol use, mother and
child left hospital
together. Did not
examine potential
confounding by parental
caregiving quality.

OR >20 ppm vs. <10
ppm

Highest grade achieved:
-0.03 (-0.05, -0.01)
per natural log increase
in tooth Pb

3-324

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tZhana et al. (2013)

Detroit, Michigan
United States

1990-2008 (born)
Followed through grade
3-8 (2008-2010)

Cohort

Students in public
schools in Detroit
n: 21281 (8831-3rd
grade, 7708- 5th grade,
4742- 8th grade)

At least 1 of the 3 tests
(math, science and
reading) taken and BLL
before 6 yr of age

Blood

Venous blood collected
for surveillance by Detroit
Department of Health
and Wellness Promotion
Age at measurement:
Before 6 yr of age (mean
age: 3.1)

Max: Highest BLL before
age 6 yr: 7.12 |jg/dL

Academic achievement
(math, science and
reading) in grade 3, 5,
and 8

Educational attainment in
math, science and
reading on MEAP.

Age at outcome:
3, 5, and 8 grades

Grade level, gender,
race, language, maternal
education, SES (i.e.,
school lunch status).

ORs of Scoring "Less
Than Proficient" on
MEAP Tests (Ref= <1
fjg/dL)

1-5 fjg/dL

Mathematics: 1.42 (1.24,
1.63)

Science: 1.33 (1.10,
1.62)

Reading: 1.45 (1.27,
1.67)

6-10 fjg/dL

Mathematics: 2.00 (1.74,
2.30)

Science: 2.22 (1.82,
2.72)

Reading: 2.21 (1.92,
2.55)

>10 fjg/dL

Mathematics: 2.40 (2.07,

2.77)

Science: 2.26 (1.84,

2.78)

Reading: 2.69 (2.31,
3.12)

3-325

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tEvensetal. (2015)

Chicago metropolitan
area, 6 counties
U.S.

1994-1998 (born)
Followed 9-10 yr, 2003-
2006
Cohort

Chicago public school

children

n: 46796

Blood

BLLs obtained from
Chicago Blood Pb
Surveillance program;
ICP-MS or AAS.
Age at measurement:
<72 mo (mean age: 45
mo)

Mean: 4.81 pg/dL

Academic achievement

3rd grade ISAT scores in
Reading and Math; 4
score categories, i.e.,
failure, below standard,
meets standard and
exceeds standard.

Age at outcome:

9-10 yr

Sex, mother's education,
low-income, very low
birth weight/preterm,
child's age at time of
BLL, ISAT vs. Iowa, race
(Interaction with race
ethnicity explored).

RR

Reading Failure
1 iJg/dL increase

All Children:
1.06 (1.05, 1.07)
NH White:
1.14 (1.08, 1.20)
NH Black:
1.05 (1.04, 1.06)
Hispanic:
1.08 (1.05, 1.11)

Math Failure
11Jg/dL increase

All Children:
1.06 (1.05, 1.07)
NH White:
1.11 (1.05, 1.18)
NH Black:
1.05 (1.04, 1.06)
Hispanic:
1.09 (1.06, 1.12)

3-326

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tBIackowicz et al. (2016)

Chicago
U.S.

1994-1998 (birth)

Followed through 2003-
2006 (3rd grade)

Cohort

School children
n: 12319

Chicago Public Schools.

Blood

Chicago Blood Pb
Registry provided data
on BLL measured
between birth and 2006
Age at measurement:
between birth and 2006
(most recent was used in
analysis)

4.16 |jg/dL

School performance

3rd grade performance
based on ISAT scores

Age at outcome:
3rd grade

Child sex, maternal
education, low-income,
preterm birth, small for
gestational age, child's
age at time of BLL, ISAT
vs. Iowa, and Hispanic
subgroup (Mexican-
American vs. other
Hispanic and Puerto
Rican vs. Other
Hispanic);

Beta

Reading scores: -0.11
(-0.134, -0.086)

Math scores: -0.096
(-0.12, -0.072)

RR

Reading failure: 1.07
(1.05, 1.10)

Math failure: 1.09 (1.06,
1.12)

tShadbeaian et al.

(2019)

North Carolina Statewide
U.S.

1990-2004 (birth)
Followed 6 yr (3-8
grade)

Cohort

NC Pb Poisoning
Prevention Program
Cohort

n: 560,624 (54% of the
Pb surveillance registry)

Living in NC between
2000-2012 with BLL <10
|jg/dL at 0-5 yr

Blood

Child blood (BLL <10,
BLL <5, and a matched
group via CEM with BLL
<5 |jg/dL)

Age at measurement:
0-5 yr

Full sample (BLL <10
|jg/dL) mean: 3.66, Full
sample (BLL <5 |jg/dL)
mean: 2.89, CEM
Matched sample (BLL <5
|jg/dL) mean: 2.40

Academic achievement

Percentile scores on
standardized EOG tests
for math and reading

Age at outcome:

Grade 3 and Grade 8

Child's sex,
race/ethnicity, SES,
Medicaid enrollment,
birth month, and age
upon entry to grade 3,
mother's age, marital
status, parental alcohol
and tobacco use, highest
educational achievement
at the time of the child's
birth, vector representing
school, grade and year
combination

CEM to balance
distributions between
groups

Beta

Decrease in Test-Score
Percentile in Children
with 5-6 |jg/dL vs. BLL <
1 |jg/dL

Math: 0.95 (0.66, 1.24)

Reading: 1.41 (1.12,
1.70)

2- and 3-way interactions
for BLL*grade,
covariates*grade,
BLL*grade*covariates.

3-327

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tSkerfvina et al. (2015)

Landskrona and

Trelleborg

Sweden

1978-2007 (enrollment
during primary school)

Followed through age 16

Cohort

Primary school children
n: 3176

Blood

Child venous blood;
flame or electrothermal
atomization AAS

Age at measurement:
7-12 yr

34 |jg/L; Median: 30
75th: 44
90th: 60
Max: 162

School performance (see
also Section 3.6.1 [adult
cognitive function])

School performance after
nine-year compulsory
schooling. 4-5 categories
from not passing to
passing with merit (based
on ranking)

Age at outcome:

16 yr

Child and parent country
of birth, parental
education, total family
income, father's IQ.

Beta

Merits

Children (BLL < 5 |jg/dL):
-10.9 (-15.486, -6.314)

All Children: -6.36
(-9.986, -2.734)

Grades

Children (BLL < 5 pg/dL):
-0.112 (-0.177, -0.047)

All Children: -0.155
(-0.21, -0.1)

Note: CIs estimated from
p-values.

AAS = atomic absorption spectrometry; BLL = blood lead level; CEM = coarsened exact matching; CI = confidence interval; EOG = end of grade; GM = geometric mean; HOME =
Health Outcomes and Measures of the Environment; ICP-MS = inductively coupled plasma mass spectrometry; ISAT = Illinois Standard Achievement Test; IQ = intelligence quotient;
MAT = Metropolitan Achievement Test; MEAP = Michigan Educational Assessment Program; Ml = Michigan; NHANES = National Health and Nutrition Examination Survey; NR = not
reported; Pb = lead; SD = standard deviation; SES = socioeconomic status; WIAT = Wechsler Individual Achievement Test; WJTA = Woodcock-Johnson Test of Achievement; WRAT
= Wide Range Achievement Test; yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bResults are not standardized (e.g., BLL distribution data needed to calculate the standardized estimate was not reported or categorical data was analyzed).

The CI was calculated from a p-value and the true CI may be wider or narrower than calculated.
tStudies published since the 2013 Integrated Science Assessment for Lead.

3-328

-------
Table 3-6E Epidemiologic studies of Pb exposure and cognitive effects: population or group mean blood Pb
levels >5 ugldL

RefereDCesfgnnd StUdy Study Population Exposure Assessment

Al-Saleh et al. (2020)

Saudi Arabia
2011-2013 (enrollment)
Followed through 2011 —
2013 and 2017-2018
Cohort

Lactating mother-infant
pairs

n: 82 (36 males and 46
female children).

Blood, Hair, Urine, Breast
Milk

Maternal blood, spot
urine, breast milk, and
hair, child spot urine and
hair; AAS with
electrothermal atomizer

Age at measurement:
Maternal measurements
made during lactation
Infants at 3-12 mo
(lactation) and children at
5-8 yr old

Lactation:

GM: maternal urine
:5.881 |jg/L, hair :1.717
|jg/g, blood GM: 2.346
|jg/dL, breastmilk: 46.483
|jg/L; Infant urine: 4.946
|jg/L, hair: 2.894 |jg/g;

Outcome

Neurodevelopmental
performance and visual-
motor integration (Test of
Nonverbal Intelligence
2nd edition and Beery
VMI 3rd edition,
respectively.)

Age at outcome:
5-8 yr old

Confounders

Child's age and sex,
maternal age, BMI,
parity, lifestyle,
educational level, SES,
residential

characteristics, urinary
cotinine levels (an index
of exposure to
secondhand smoke).

Effect Estimates and
95% CIs

Beta (95% Cl)a

BVMI: 0.012 (-1.989,
2.014)

TONI: -0.044 (-1.809,
1.721)

Early childhood:
GM: urine: 2.563 |jg/L,
hair: 0.850 |jg/g max:
urine: 20.826 |jg/L, hair:
4.470 |jg/g

3-329

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Bara et al. (2018)

Montevideo
Uruguay
Study years NR
Cross-sectional

n: 206

Children living in areas
considered high risk for
metal exposure.

Blood

Child venous blood;
flame AAS or GFAAS
Age at measurement:
5-8 yr

4.2 |jg/dL

Executive function

BRIEF (teacher rating)

Age at outcome:

5-8 yr

Child IQ, iron status, and
BMI, blood Pb testing
method, household
possessions, maternal
education, current parent
smoking.

PR (95% Cl)b

BRIEF- Global Executive
Composite

Children with BLL >5 vs.
<5 |jg/dL: 1.02 (0.96,
1.09)

Boys: 1.00 (0.98, 1.01)

Girls: 1.01 (0.99, 1.04)

Cai et al. (2021)

Guangxi
China

Study years NR
Cross-sectional

School children
n: 255

Participants living near a
Pb and zinc mine (-500
m distance between
school and mine).

Blood

Child venous blood (Pb
intoxication and non-Pb
intoxication groups);
GFAAS

Age at measurement:
7-12 yr

Median Pb level: Rice
samples: 0.10 mg/kg,
Blood: 84.8 |jg/L
75th: Blood: 115.4 pg/L
Max: Rice: 0.53 mg/kg

Perception and
reasoning

Raven's SPM

Age at outcome:
7-12 yr

Age, gender, physical
condition, lifestyle habits,
educational attainment
and smoking habit of
parents, family
environment and
economy.

Beta (95% Cl)b

IQ, RSPM: -0.58 (-1.031,
-0.129)

Jeona et al. (2015)

Multi-center
South Korea
Cross-sectional

May 2006-Dec 2010

MOCEH study
n: 194

Birth cohort- mother-
infant pairs followed
through 60 mo of age.
Cross-sectional analysis
conducted.

Blood

Child venous blood;
GFAAS

Age at measurement:
60 mo

GM: 13.01 pg/L
Max: 35.05 pg/L

FSIQ, VIQ, PIQ (Korean
WPPSI-R)

Age at outcome:

60 mo

Sex, parental education,
family income,
breastfeeding status,
CRP level, mother's BLL
during pregnancy

Note: Mediation analysis
to examine the
relationship of BLL, iron
deficiency and IQ.

Beta (95% CI)b
Verbal IQ and In-BLL
(pg/L): -9.587 (-16.829,
-2.344)

3-330

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Kao et al. (2021)

Taipei
Taiwan
2011-2014
Cross-Sectional

recruited from Taipei
MacKay Memorial
Hospital

n:139 children less than
3 yr of age

Hair, fingernails

Child hair, fingernails;
ICP-MS

Age at Measurement:
Mean (SD) 2.8 (0.4)
years (children under 3
yr)

Mean (SD): hair 2.9 (4.8)

pg/g,

nails 0.8 (5.1) |jg/g

BSID-III cognitive and
language development
scores

Age at outcome: 2.8 ±
0.4 yr

General linear models
adjusted for sex,
gestational age at birth,
age of the house (years),
leafy-vegetable intake
(servings/week), and the
area of surface roads
within 100 m of the
residence

Regression results were
not reported because
they were not statistically
significant.

3-331

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Kordas et al. (2011)

Mexico City, Mexico
Jan 1994-June 1995
Followed for 48 mo
Cohort

Birth cohort

n: 24 mo = 220, 48 mo =
186

Mother-infant pairs from
3 hospitals serving low-
and middle-income
women

Blood

Maternal, cord blood, and
child blood; GFAAS

Age at measurement:
Delivery (maternal, cord),
24 and 48 mo (child)

Mean: Maternal BLL at
delivery: 8.6 |jg/dL, Cord
blood: 6.6, BLL at 24 mo:
8.1, BLL at 48 mo: 8.1

Neurodevelopment using
BSID-II (MDI), MSCA
(general cognitive index
and memory scale).

Age at outcome:

24 (BSID) and 48 mo
(MSCA)

OLS linear regression

Birth weight, gestational
age, child sex; maternal
age, years of schooling,
IQ, smoking status,
marital status crowding in
the house, type of floor in
the house.

(Stratified analysis by
child development 48 mo
and gene polymorphism
also conducted.)

Beta (95% Cl)b

McCarthy Scales of
Children's Abilities, 48
mo

GCI

Concurrent BLL: -0.6
(-0.992, -0.208)

Cord BLL: -0.2 (-0.788,
0.388)

Memory Score

Concurrent BLL: -0.3
(-0.496, -0.104)

Cord BLL: 0.1 (-0.096,
0.296)

Bayley Scales of Infant
Development II, 24 mo

MDI

Concurrent BLL: -0.1
(-0.492, 0.292)

Cord BLL: -0.7 (-1.288,
-0.112)

Kuana et al. (2020)

Nanjing

China

2012

Cross-sectional

Public primary school
children

n: 742

Excluded students with
congenital mental
retardation (third-degree
relatives included) and
diseases.

Blood

Child venous blood; ICP-
MS

Age at measurement:
7-11 yr

Mean: 30.4 |jg/L; Median:
26.1 |jg/L

School performance

Standardized scores on
Chinese, Math and
English added for total
scores.

Age at outcome:
7-11 yr

Age and sex (tests
administered on the
same day).

Beta (no p-value, Cis,
or SE reported)13

Total: -0.168

Chinese: -0.042

Math: -0.039

English: -0.087

3-332

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Lee etal. (2017)

Korea

Enrolled 2006-2015,
followed to age 5 (2015)
Cohort

Mothers' and Children's
Environmental Health
(MOCEH)
n: 251

Maternal Blood

GFAAS with Zeeman
background correction

Age at Measurement:
At birth (cord blood)

GM 0.957 |jg/dL
Max: 3.17 |jg/dL

Cognitive Development

The mental
developmental index
(MDI) of the Korean
BSID-II (K-BSID-II) was
administered to infants
who were 6, 12, 24, and
36 mo-old. The Korean
language version of the
Wechsler Preschool and
Primary Scale of
Intelligence - Revised (K-
WPPSI-R) was
administered to children
at 60 mo.

Partial correlation
analysis adjusted for
maternal education, sex
of child, and family
income.

Strongest correlations
between scores
measured at closest
time. Scores more stable
in those at extreme ends
of cognitive development.

3-333

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Liu etal. (2013a)

Jiangsu province
China

2004-2005 (enrollment)
followed 5 yr

Cohort

China Jintan Child
Cohort Study
n: 1341 children (603
girls and 738 boys)

Community based cohort
of preschool children.

Blood

Early child blood;
GFAAS.

Age at measurement:
3, 4 or 5 yr

mean: 6.43 |jg/dL

FSIQ, VIQ, PIQ (Chinese
version ofWPPSI-R).
See also Table 3-5

(School performance
was assessed by
standardized tests on 3
major subjects: Chinese,
English and Math.)

Age at outcome:

6 yr (IQ), 8-10 yr (school

performance)

Child age at blood Pb
test, child gender,
residence as defined as
school location, blood
iron level, parent
education, parent
occupation, and father's
smoking.

Beta for log-transformed
BLL (95% Cl)b (ref: <8
pg/dL)

FSIQ

8-10 |jg/dL: -1.28
(-4.01, 1.46)

>10 |jg/dL: -1.45 (-3.50,
0.67)

Chinese score

8-10 |jg/dL: -3.20

(-5.78, -0.63)

>10 |jg/dL: -4.02 (-7.11,

-0.93)

Math score

8-10 |jg/dL: -5.25
(-8.14, -2.36)

>10 |jg/dL: -5.27 (-8.73,
-1.81)

English score
8-10 |jg/dL: -4.33
(-7.32, -1.34)

>10 |jg/dL: -5.18 (-8.76,
-1.59)

Liu etal. (2018b)

PROGRESS study

Blood

Cognitive development

SES, mother's NR



n: 665



using BSID-III. BSID

hemoglobin during the

Mexico City, Mexico



Maternal blood; joint

scores were centered

second trimester of

Followed for 24 mo

Mother-infant pairs.

exposure to Mn, Pb, Co,

and scaled and

pregnancy, mother's

Cohort



Cr, Cs, Cu, As, Cd, and

presented as z-scores

educational level, child





Sb

normalized to expected
mean of 100 and SD of

gender, mother's WASI
IQ, and Fenton's birth





Age at measurement:

15.

weight z-scores.





Prenatal exposure (2nd









trimester)

Age at outcome:







NR

6, 12, 18, and 24 mo



3-334

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Marques et al. (2014)

State of Rondonia,
Western Amazon
Brazil
Cohort

2007-2012

Population-based cohort
n: 96 (TOKS = 51 and
Itapua = 45)

Low SES populations
living in rural and urban
areas including children
living in the vicinity of
TOKS (i.e., multiple
metal exposure).

Breastmilk

Breastmilk; GFAAS
Age at measurement:
6 mo of breastfeeding
(from Marques 2013c)

TOKS: 10.04 pg/L
(mean), 8.2 pg/L
(median); 29.4 pg/L
(max)

Itapua: 3.89 pg/L (mean)
2.5 pg/L (median), 16.2
pg/L (max)

Neurodevelopment
(milestones including age
of walking and talking,
Bayley MDI and PDI);
milestones assessed
based on mothers'
recollection at the time of
visit.

Age at outcome:

6 and 24 mo (Bayley
MDI, PDI)

Birth weight, income,
maternal education,
breastfeeding status.

Beta (95% Cl)b

MDI 6 M -0.293 (-0.50,
0.08)

MDI 24 M -0.234 (-0.60,
0.13)

PDI 6 M -0.062 (-0.28,
0.16)

PDI 24 M -0.129 (-0.34,
0.08)

Age of walking -0.219
(-0.43, 0.002)

Age of talking -0.066
(-0.28, -0.16)

3-335

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Maazamen et al. (2015)

Wisconsin (Milwaukee or
Racine)

United States
Enrollment: born during
1996-2000 (Follow-up of
blood lead level: before
child's third birthday;
Follow-up for WKCE
scores: 4th grade)

Cohort

Wisconsin Children's
Lead Levels and
Educational Outcomes
Project (CLLEO)
n: 1076

Blood

Blood lead records used
to categorize children as
not exposed (<5 |jg/dL)
or exposed (>10 |jg/dL
and <20 |jg/dL)

Age at Measurement:
18-36 mo

43% of sample defined
as exposed

Academic achievement:
Wisconsin Knowledge
and Concepts Exam
(WKCE) math and
reading scores

WKCE reading and math
scores obtained from the
Wisconsin Department of
Public Instruction with
parental consent.

Child gender, race,
parental < HS education,
free lunch program,
English language learner,
and child health rating by
parents (excellent vs.
other); interactions with
Pb tested for each
covariate

Beta for entire
distribution of math
scores'3

OLS = -8.94 (-14.84,
-3.05)

Beta for math scores in
quantilesb

10th percentile = -17.00
(-32.13, -3.27)
50th percentile = -8.00
(-15.24, -0.36)
90th percentile = -4.50
(-10.55, 4.50)

Beta for entire
distribution of reading
scores'3

OLS = -13.66 (-19.94, -
7.37)

Beta for reading scores
in quantilesb

10th percentile = -18.00
(-48.72, -3.32)
50th percentile = -14.50
(-20.72, -5.61)
90th percentile = -7.50
(-15.58, 2.07)

3-336

-------
Rawat et al. (2022)	n: 43	Blood

India	Blood Pb was measured

Not reported	via LeadCare II testing

Cross-Sectional	analyser

Age at Measurement:

4-12 yr

GM (SD) 19.93 (9.22)
ug/dL

Max: 37.4 |jg/dL

% change in results from
Draw-A-Person test for
Group B compared to
Group A:

Line characteristic: Thick
and sharp = 41%; Soft =
-32%

Detailing: With = 32%;
Without = -9%

Shading: With = -21%;
Without = 24%

Distortion: With = 50%;
Without = -17%

Colours: Warm = 41%;
Cool = -27%

IQ level, performance on
Draw-A-Person Test

The Draw-A-Person test
and the IQ test were
administered in the study
setting

Age at Outcome:

4-12 yr

There were no
adjustments for
confounders, as simple
statistics were employed.

IQ - Mean (SD) scoreb

Group A (<10 |jg/dL Pb,
n = 9): 122.33 (4.03)

Group B (>10 |jg/dL Pb,
n = 34): 96.03 (12.76) p-
value for difference =
0.006

Group A (<10 |jg/dL, n =
9) vs. Group B (>10
|jg/dL, n = 34) drawings
had the following
characteristics:

Thick and sharp lines =
33% vs. 47%; Soft lines =
78% vs. 53%%

3-337

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

With detailing = 22% vs.
29%%; Without detailing
= 78% vs. 71%%

With shading = 33% vs.
26%; Without shading =
67% vs. 82%

With distortion = 33% vs.
50%; Without distortion =
89% vs. 74%

Warm colors = 67% vs.
94%; Cool colors = 89%
vs. 65%

Rodriaues et al. (2016) Birth cohort in

Sirajdikhan and Pabna

districts

Bangladesh

2008-2011 (enrollment)
Cross-sectional

Bangladesh
n: 525 (Sirajdikhan:
Pabna: 286)

239;

Pregnant women
(gestational age <16 wk).

Blood

Child concurrent whole
blood tested using the Pb
Care II

Water samples from tube
well tested for As and Mn
during first trimester of
pregnancy and follow-up
visits at age of 1 mo, 12
mo and 20-40 mo.

Age at measurement:
20-40 mo

Cognitive development Maternal age and

using the culturally
adapted BSID-III. Age-
adjusted Z-scores

Age at outcome:
20-40 mo of age

education, child's sex,
ETS, HOME score,
maternal Raven score,
child hematocrit levels,
As, Mn.

Beta (95% CI)

Cognitive development
Pabna region3: 0.02 (SE:
0.12) per In-transformed
BLL*

Sirajdikhan regionb:
-0.02 (-0.04, 0.00)

*Unable to standardize
because P25 and median
BLL were 
-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Rov et al. (2011)

Chennai, India (4
representative industrial
and traffic zones)
2005-2006
Cross-sectional

School Children (3-7 yr)
n: 725

3 schools from each of
the 4 zones randomly
selected (12 schools
total); children lower and
upper kindergarten and
first grades.

Blood

Postnatal venous blood;
PbCare Analyzer

Age at measurement:
3-7 yr of age

Overall mean: 11.5
|jg/dL; mean by
genotypes: Taq A1/A1:
11.66 |jg/dL; Taq
A1/A2+A2/A2: 11.42
|jg/dL

Max: Overall: 40.5 |jg/dL

IQ using BKT (mental
age divided by
chronological age and
multiplied by 100)

Tamil-translated Binet-
Kamat Scales of
Intelligence

Age at outcome:
3-7 yr of age

Age + age 2, sex,
midarm circumference,
average monthly family
income, and family size,
parental education.
Note: stratified analysis
by genotypes (3
categories) conducted.

Beta (95% CI)b

IQ (BKT, Tamil-
translated): -4.22 (-7.10,
-1.36)

tRyqiel et al. (2021)

Mexico City
Mexico
1997-2005
Cohort

ELEMENT project
n: 85

Mother-child pairs
recruited at the Mexican
Social Security Institute

Blood

Maternal and child
venous blood; ICP-MS,
GFAAS

Age at measurement:
T1, T2, T3 (maternal);
12, 24 mo (child)

Maternal blood GM (SD):
T1: 5.27 (1.93) pg/dL
T2: 4.74 (1.96) pg/dL
T3: 4.98 (1.93) pg/dL

Infant blood GM (SD):
12 mo: 3.92 (1.80) pg/dL
24 mo: 3.49 (1.93) pg/dL

MDI assessed using
BSID-II (Spanish version)

Age at outcome: 12-24
mo

Maternal IQ (WAIS),
maternal age, infant
weight, length, SES,
infant age and sex,
current infant BLL.

Beta (95% CI) for 12-
month MDIb

T1: 0.31 (0.00, 0.62)

T2: 0.11 (-0.63, 0.86)

T3: 0.41 (-0.34, 1.17)

Beta (95% CI) for 24-
month MDIb
T1: -0.16 (-0.99, 0.66)
T2: -0.23 (-1.05, 0.59)
T3: 0.28 (-0.50, 1.06)

3-339

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tSanchez et al. (2011)

Mexico City
Mexico

1997-1999 (enrollment)
Followed through 24 mo

Cohort

ELEMENT study
n: 169

Mother-child pairs
recruited during
pregnancy or before
conception.

Blood

Maternal blood; ICP-MS

Age at measurement:
each trimester of
pregnancy

Mean (SD):

1st trimester (n = 139):
13.7 (3.4) |jg/dL
2nd trimester (n = 159):
24.5 (2.8) |jg/dL
3rd trimester (n = 147):
35.2 (1.9) |jg/dL
Max:

1st trimester: 20.4 |jg/dL
2nd trimester: 33.7 |jg/dL

3rd trimester: 39.0 |jg/dL.

MDI assessed using
BSID-II (Spanish version)

Scores standardized for
mother's age, mother's
IQ, duration of
breastfeeding, sex, and
weight and height z-
score at 24 mo

Age at outcome:

24 mo

Maternal age, IQ,
duration of breastfeeding,
sex, weight, and height
Z-score at 24 mo.

Beta (95% Cl)b

T1
T2
T3

-5.42 (-10.2, -0.64)
0.88 (-5.34, 7.09)
1.22 (-3.65, 6.08)

3-340

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Saxena et al. (2022)

Araihazar
Bangladesh
2012-2016
Cross-Sectional

Metals, Arsenic, &
Nutrition in Adolescents
study (MANAs)
n: 572

Blood

Whole blood Pb
quantified using ICP-MS.
Age at Measurement:
Mean (SD) = 14.6 (0.7)
years

Mean = 98.7 |jg/L;
Median = 91.29 |jg/L

Cognitive Effects

The Cambridge
Neuropsychological Test
Automated Battery
(CANTAB) was
administered to the
adolescents to assess
aspects of executive
function.

Linear regression models
adjusted for BMI, head
circumference, child's
years of education,
maternal intelligence
(WASI), paternal years of
education, wall type, sex,
and other blood metals -
arsenic, cadmium,
manganese, and
selenium.

Beta (95% Cl)c

Delayed Match to
sample: -3.67 (-6.59, -
0.75)

Planning: -0.05 (-0.42,
0.32)

Rapid visual processing:
-0.01 (-0.03, 0.01)
Reaction time: 0.03
(-0.01, 0.07)

Spatial recognition
memory: 1.9 (-0.88,
4.68)

Spatial span: -0.16
(-0.43, 0.11)

Spatial working memory:
1.09 (-2.54, 4.72)

3-341

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Soetrisno and Delaado-
Saborit (2020)

West Java (Depok,
Bogor and Bekasi)
Sukatani village (control)
Indonesia
Cross-sectional

School children living in
urban locations near e-
waste facility; control site
n: 44 (22 from Bogor and
22 from Sukatani)

Children selected from
schools per teachers/
principal

recommendation.

Hair, soil, water

Hair samples from
children in Bogor and
Sukatani village. BLLs
from 36 children in Bogor
area (2010).

Age at measurement:
6-9 yr

Soil Pb mean: Depok-
Bekasi: 3653 mg/kg;
Sukatani: 93.2 mg/kg;
Water Pb: all 10 samples
below LOD; Hair Pb:
Depok-Bekasi: 0.155
mg/g; Sukatani: 0.0729
mg/kg

Max: Soil Pb: Depok-
Bekasi: 7662 mg/kg;
Sukatani: 115 mg/kg;

Hair Pb: Depok-Bekasi:
0.841 mg/g; Sukatani:
0.255 mg/kg

Academic achievement
(see also Section 3.5.1.4,
executive function)

Performance on reading,
math, writing expression
and oral language, arts,
science, social sciences,
and sports collected from
the school official alumni
report. TMT B.

Age at outcome:

6-9 yr

Age, parental education,
environmental tobacco
smoke at home, and
residential traffic
exposure.

Beta (95% Cl)d

Change in TMT-B
(seconds) per mg/g unit
of hair Pb: 54 (-3.8, 114)

3-342

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Sun et al. (2015)

Jiangsu Province
China
Nov 2011
Cross-sectional

Chinese National Health Blood, Urine
Research Program
n: 446

Participants recruited
from three primary
schools located in the
three towns.

Child venous blood
samples; ICP-MS
method. Morning urine
samples collected was
tested for heavy metals.
Age at measurement:
9-13 yr

GM BLL: 33.13 pg/L;
Arithmetic mean BLL:
36.99 pg/L
75th: 43.39 pg/L
90th: 56.85 pg/L
Max: 101 pg/L

IQ (CRT). Primary score Father's education,

was converted to
standard IQ scores.

Age at outcome:
9-13 yr

mother's education, BMI,
annual family income,
gender, age.

Beta (95% CI)b
-6.61 (-13.15, -0.07)

Tassiopoulos et al.
(2017)

22 PHACS clinical
research sites in the
United States, including
Puerto Rico
USA

Enrollment began in
2007; BLL data available
from 1998-2014,
developmental data
available from
1996-2010,
developmental data
available from
1996-2010
Cohort

Surveillance Monitoring
of ART Toxicities
(SMARTT)
n: 546 children with a
Bayley-111 at one year of
age who had a BPb
between 9 mo of age and
up to 3 mo after the
Bayley-111; 634 children
with a Bayley Screen at 3
yr of age and a BPb
between 9 mo of age and
up to 3 mo after the
Bayley Screen

Blood

Blood lead obtained
between the ages of 1
and 3 yr as part of
standard of care or local
guidelines are abstracted
from the medical chart
when available
Age at Measurement: 1
yr (n = 546) and 3 yr (n =
634)

Cognitive Effects

Cognition and language
neurodevelopment using
BSID-III.

Age at outcome: 1 yr

At 3 yr of age,
developmental function
was assessed with the
Bayley Screening Test
(Bayley Screen), 19 which
includes a subset of
items from the Bayley-111
with the domains of
cognition, receptive
communication,
expressive

communication, and fine
and gross motor
development.

Sex; race; ethnicity;
maternal IQ (evaluated
with the Wechsler
Abbreviated Scale of
Intelligence); maternal
education, primary
language, living
arrangement, and living
situation; household
income; geographic
region; prenatal tobacco
exposure; postnatal
tobacco exposure within
the home; age at the
developmental
evaluation; and help from
others caring for the
child.

OR (95% Cl)b (for BLLs
>=5 vs. <5 pg/dL)
Cognitive delay 1.64
(0.95, 2.90)

Receptive

communication 0.83
(0.47, 1.43)

Expressive
communication 0.91
(0.52, 1.58)

3-343

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Tuna et al. (2022)

Providence, Rhode
Island

United States
Mother-newborn
assessed for exposure
and outcome within 2 hrs
of delivery and 24 hrs of
delivery, respectively.
Cross-Sectional

Rhode Island Health
Study (RICHS)
n: 192

Placental Blood Pb

Placental Pb
concentrations quantified
using ICP-MS.

Age at Measurement:
24 hrs

Mean = 4.49 ng/g among
those with detectable Pb

BSID

Newborns' neurologic
integrity, behavioral
function, and signs of
stress assessed by NICU
Network Neurobehavioral
Scale (NNNS). Latent
Profile Analyses used to
place children in
subgroups with discrete
profiles.

Multinomial regression
models adjusted for
infant gender, maternal
age, maternal race,
maternal BMI, education
status, and smoking
status during pregnancy.

OR (95% Cl)b for
neurobehavioral profile
membership associated
with detectable Pb
(>LOD, dichotomized) vs.
Profile 2 membership

Profile 1: 0.95 (0.38,

2.35)

Profile 3: 0.97 (0.42,

2.25)

Profile 4: 0.91 (0.38,

2.20)

Profile 5: 3.42 (0.88,
13.32), p <0.1

Profile characteristics: 5
= Highest arousal,
excitability and
hypertonicity with lowest
quality of movement and
regulation (most
extreme). Other profiles:
1 = High attention and
quality of movement; 2
[referent] = Average with
lowest lethargy; 3 =
Average, required more
handling; 4 = More signs
of lethargy, hypotonicity,
nonoptimal reflexes, low
attention and arousal.

*Unstandardized due to
BLL distribution
information

3-344

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Wan etal. (2021)	n: 333

China

Cross-Sectional

Blood

Blood samples were
collected from a previous
study and analyzed for
Pb. Authors did not
report analytical method.
Age at Measurement:
Children aged 9-11 yr;
exposure group mean
(SE) = 9.93 (0.85) years;
control group (SE) = 9.62
(0.73)

FSIQ

Intelligence was tested
using the Combined
Raven's Test in China
(CRT-C2).

Multivariable linear
regression adjusted for
children's age and
gender, father's and
mother's age, education
levels and occupations,
passive smoking of the
children, and annual
family incomes.

Beta (95% Cl)a
-1.2 (-1.7, -0.60)

Median for exposure
group = 7.163 |jg/dL;
median for control group
= 3.703 |jg/dL

3-345

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Wang et al. (2012)

Taizhou region (Luqiao
city and Lanxi city),
Zhejiang Province (for
exposure site) and
Chun'an, Zhejiang
province (reference site)
China

June 2010
Cross-sectional

School-based study
n: 329 (Luqiao: 108,
Lanxi: 151, Chun'an: 70)

Schools located near e-
waste recycling center
and tinfoil manufacturing
area (Luqiao and Lanxi
cities). Comparison
group schools in area
dominated by agriculture
(Chun'an).

Blood, urine

Child's venous blood,
urine; ICP-MS.

Age at measurement:
11-12 yr

GM: Luqiao: 6.97 |jg/dL,
Lanxi: 8.11 |jg/dL,
Chun'an: 2.78 |jg/dL
(42%-53% had BLL >10
|jg/dL in Luqiao and
Lanxi and no one had
BLL >10 in Chun'an)
Max: Luqiao: 57.24
|jg/dL, Lanxi: 59.98
|jg/dL, Chun'an: 7.59
pg/dL

IQ (CRT) calculated from
raw score.

Age at outcome:

11-12 yr

Child's sex, birth weight,
BMI, gestation at delivery
and the mother's age at
delivery, years of
education, yearly income,
tobacco exposure during
pregnancy and alcohol
exposure during
pregnancy.

Beta (95% Cl)b

IQ (CRT)

Female: -0.097 (-0.178,
-0.016)

Male: -0.096 (-0.175,
-0.016)

AAS = atomic absorption spectrometry; ADHD = attention deficit/hyperactivity disorder; As = arsenic; BASC = Behavior Assessment System for Children; BKT = Binet Kamat Test Of
Intelligence; BLL = blood lead level; BMI = body mass index; BRIEF = Behavior Rating Inventory of Executive Functions; BSID = Bayley Scales of Infant and Toddler Development;
CANTAB = Cambridge Neuropsychological Test Automated Battery; CBLI = cumulative blood lead index; CI = confidence interval; Co = cobalt; Cr = chromium; CRS = Conners'
Rating Scales; CRT = Combined Raven's Test; Cs = cesium; Cu = copper; ELEMENT = Early Life Exposure in Mexico to Environmental Toxicants; ETS = environmental tobacco
smoke; Fe = iron; FSIQ = full-scale intelligence quotient; GFAAS = graphite furnace atomic absorption spectrometry; GM = geometric mean; Hg = mercury; HNES = Home Nurture
Environment Scale; HOME = Health Outcomes and Measures of the Environment; ICP-MS = inductively coupled plasma mass spectrometry; ICP-MS-DRC = inductively coupled
plasma mass spectrometry; K-XRF = K-shell X-ray fluorescence; LOD = limit of detection; MDAT = Malawi Developmental Assessment Tool; MDI = Mental Developmental Index; Mn
= manganese; mo = month(s); MOCEH = Mothers' and Children's Environmental Health; MSCA = McCarthy Scales of Children's Abilities; NR = not reported; OLS = ordinary least
squares; Pb = lead; PDI = Psychomotor Developmental Index; PIQ = performance intelligence quotient; PROGRESS = Programming Research in Obesity, Growth, Environment and
Social Stressors; Sb = antimony; SES = socioeconomic status; SPM = Standard Progressive Matrices; TMT = Trail Making Test; TOKS = tin, ores, kiln, smelters; VIQ = verbal
intelligence quotient; VMI = visual-motor integration WAIS = Wechsler Adult Intelligence Scale-Revised; WASI = Wechsler Abbreviated Scale of Intelligence; WISC = Weschler
Intelligence Scale for Children; WPPSI = Wechsler Preschool and Primary Scale of Intelligence; WRAT = Wide Range Achievement Test.

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a

change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

bEffect estimates are not standardized because data pertaining to the BLL distribution and/or base for the log-transformation were not reported.

°Per natural log increased in centered BLLs (i.e., BLL/median).

dResults are unstandardized due to the biomarker (hair).

tStudies published since the 2013 Integrated Science Assessment for Lead.

3-346

-------
Table 3-7E Epidemiologic studies of Pb exposure and performance on neuropsychological tests of attention,
impulsivity, and hyperactivity, ADHD-related behaviors, and clinical ADHD in children

Referent^ and Study study Popu|atjon

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tNeuaebauer et al.
(2015)

Duisburg
Germany

2000-2002
(enrollment)
Followed through
2009-2011

Cohort

Duisburg birth
cohort study
n: 114

Pregnant women
and their offspring

Blood

Maternal venous blood; AAS

Age at measurement:
32 wk gestation (prenatal)

Mean (SD): 2.216 (1.083)
pg/dL

Med: 2.0 pg/dL
95th: 4.2 pg/dL
Max: 6.3 pg/dL

Attentional performance
using KiTAP with 5 subtests:
alertness, distractibility,
Go/No-go, divided attention,
flexibility; ADHD-associated
behavior using FBB-ADHS

Age at outcome:

Mean: 8.5 yr (KiTAP); 9.5 yr

(FBB-ADHS)

maternal
diseases, parental
lifestyle, childbirth
outcomes, HOME
Score

gMR:

KITAP

Inattention (omissions): 1.15
(1.00, 1.33)

Attention (performance
speed): 1.14 (0.98, 1.33)

FBB-ADHS

Overall ADHD: 1.061 (1.009,

1.115)

Impulsivity

1.133 (1.055, 1.216)

Hyperactivity

1.047 (0.992, 1.106)

Inattention

1.054 (0.989, 1.123)

3-347

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tEthieret al. (2015)

Arctic Quebec

(Puvirnituq)

Canada

1993-1998
(enrollment)
Followed through
2009-2020

Cohort

NCDS
n: 27

Subsample of
school-aged
children who
participated in the
Cord Blood
Monitoring Program

Blood

Cord blood and concurrent
venous blood; GFAAS

Age at measurement:

Delivery (cord); 8.6-12.6 yr old
(concurrent)

Mean (SD): 5.4 (4.1) pg/dL
Max: 17.8 pg/dL

Selective spatial attention

Visuo-spatial attention-shift
task (adapted from Posner
paradigm)

Age at outcome:
8.6-12.6 yr

Sex, age at testing
time, SES,
breastfeeding
duration, maternal
alcohol, marijuana,
cigarettes use

(Each model used a
different set of
confounders)

Beta per SD increase in In-
transformed Pb:

Cord Blood

Reaction time: 0.02b

Omission Error: -0.02b

False Alarm: 0.42 (0.08,
0.76)c

Accuracy: -0.27b
Validity Effect: -0.05b

Concurrent Blood

Reaction time: 0.52 (-0.10,
1.14)c

Omission Error: -0.10b
False Alarm: -0.16b
Accuracy: -0.17b
Validity Effect: -0.13b

tTatsuta et al. (2014) TSCD birth cohort
n: 387

Sendai, Tohoku region
Japan

Study years NR
Followed through 42

mo

Mother-infant pairs
urban areas of the
Tohoku district

Blood

Cord blood; ICP-MS.

Age at measurement:
Delivery

Median: 1.0 pg/dL
Max: 1.8 pg/dL

Sequential processing and Child sex, birth order, Betas: K-ABC

mental processing scores
(K-ABC)

Age at outcome:
42 mo

alcohol and smoking
habits, duration of
breastfeeding, annual
family income at 42
mo, and maternal IQ
(Raven SPM)

Sequential Processing:
-2.136 (-12.80, 8.531)d

Mental Processing:
(-12.41, 5.774)d

-3.319

Cohort

3-348

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tYorifuii et al. (2011) Faroese birth cohort Blood

Attention/working memory Child age, sex,

Faroese island
Denmark

1986-1987
(enrollment)

Followed through 7-14
yr

Cohort

n: 896 (7 yr),
(14 yr)

808

Mother-infant pairs

Cord blood; electrothermal
AAS with Zeeman background
correction.

Age at measurement:

Delivery

GM: 1.57 |jg/dL
75th: 2.2 pg/dL

assessed using WISC-R
digit span

Age at outcome:

7, 14 yr

maternal IQ (RPM),
paternal employment
and education,
maternal education,
daycare at age 7,
medical risk, and
maternal alcohol use
and smoking during
pregnancy

Beta per log-transformed
Pb:

7 yr

Digit span forward: -0.11
(-0.29, 0.07)d
<2.61 pg/g Hg: -1.70
(-3.12, -0.28)

14 yr

Digit span: -0.21 (-0.53,
0.11)d

Digit span forward: -0.04
(-0.23, 0.14)d

Digit span backward: -0.17
(-0.37, 0.04)d

<2.61 pg/g Hg: -2.73
(-4.32, -1.14)

tRuebner et al. (2019)

46 centers
United States

Study Years: NR
Followed through 1-16
yr

Cohort

CKiD Cohort study
n: 412

Children ages 1-16
yr at recruitment
with mild to
moderate CKD

Blood

Child venous blood; ICP-MS.
The BLL measurement closest
to the time of neurocognitive
testing was used for analysis
(concurrent).

Age at measurement:
NR; 2, 4, or 6 yr after study
entry

Median: 1.2 pg/dL
75th: 1.8 pg/dL
Max: 5.1 pg/dL

Attention, hyperactivity, and
response inhibition

Age-specific neurocognitive
assessments (K-CPT, CPT
III, BASC-2) administered 3,
5, 7 or 9 yr after study entry.
The last available test
results were used to
evaluate long-term effects.
Mean time between BLL
and neurocognitive testing
was 2.3 yr.

Age at outcome:

1 to >18 yr; median: 15.4 yr

Child age, sex, race,
poverty, and maternal
education

Beta: K-CPT/CPT
Attention: 1.8 (0.15, 3.45)

Adjusted BRIEF and BASC-
2 results were not reported
because they were not
statistically significant.

3-349

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders Effect Estil?fctfs and 95%

tRoonev et al. (2018)

Lisbon
Portugal

1997-2005
(enrollment age 8-12
yr)

Followed 7 yr age 15-
19 yr

Cohort

Casa Pia Clinical
Trial of Dental
Amalgams in
Children
n: 330

Children aged 8-12
yr at baseline in the
Casa Pia school
system

Blood

Child venous blood; flameless
AAS

Age at measurement:
8-12 yr old (baseline)

Mean (SD):

Boys: 5.26 (2.73) pg/dL
Girls: 4.42 (2.19) pg/dL
Max: 15.0 |jg/dL

Neuropsychological tests of
attention

Stroop word, Stroop color,
Stroop color/word, WISC-III
Digitspan, WAIS-III, WMS-
III, Trail Making A, Adult
Trail Making A

Age at outcome:

15-19 yr (annual
assessment for 7 yr)

Age at baseline, race,
and nonverbal IQ
(home environment,
parent's SES,
medical histories
similar across
subjects

Median beta: Boys

Stroop word: -0.118
(-0.257, 0.021)

Stroop color: -0.114
(-0.246, 0.018)

Stroop color/word: -0.117
(-0.232, -0.001)

WAIS-III digitspan: -0.049
(-0.112, 0.015)

WMS-III spatialspan: -0.012
(-0.077, 0.054)

Adult Trailmaking A: -0.02
(-0.148, 0.108)

Median beta: Girls

Stroop word: -0.01 (-0.182,
0.162)

Stroop color: 0.033 (-0.142,
0.208)

Stroop color/word: -0.019
(-0.165, 0.126)

WAIS-III digitspan: -0.06
(-0.139, 0.02)

WMS-III spatialspan: -0.019
(-0.103, 0.066)

Adult Trailmaking A: -0.165
(-0.35, 0.021)

3-350

-------
tChoi et al. (2020)

Seoul
Korea

Aug. 2010-Feb. 2015
Case-control

Blood

Child venous blood; GFAAS
with Zeeman background
correction

Age at measurement:
5-18 yr

Mean: 1.4 (cases) vs. 1.3
(controls) |jg/dL

ADHD status
diagnosed with K-
SADS-PL

(neuropsychological
testing by board
certified

child/adolescent
psychiatrist)

n = 355 (259
ADHD, 96 controls)

5-18 yr old patients
at a child and
adolescent
psychiatry
outpatient clinic of
Seoul National
University Hospital

Inattention and	Age, sex, IQ

hyperactivity/impulsvity
assed using ADHD-RS IV
(parent rating)

Attention and executive
function assessed using
computerized SCWT and
CPT

Age at outcome:
5-18 yr

Beta direct effects: ADHD-
RS

Total ADHD severity: 2.254
(-0.278, 4.785)

Inattention: 1.053 (-0.387,
2.493)

Hyperactivity/lmpulsivity:
1.259 (-0.042, 2.560)

Beta direct effects: Conners'
CPT

Inattention (errors of
omission): 3.748 (0.091,
7.404)

Impulsivity (errors of
commission): -0.925
(-4.412, 2.562)

Response Time: 2.515
(0.013, 5.017)

Response Time Variability:
2.647 (-0.846, 6.140)

Beta direct effects: Stroop

Stroop word: -1.143
(-3.316, 1.031)

Stroop color: -0.729
(-2.832, 1.375)

Stroop color/word: 0.491
(-1.876, 2.857)

Stroop color/word
interference: 1.618 (-0.963,
4.199)

Beta interaction: Stroop
DAT1 x Pb on Inattention
(errors omission): 10.613
(-0.237, 21.463)

3-351

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders Effect Estil?fctfs and 95%

DAT1 x Pb on Response
Time Variability: -0.198
(-10.527, 10.132)

DRD4 x Pb on Inattention
(errors omission): -0.911
(-7.380, 5.558)

DRD4 x Pb on Response
Time Variability: -4.065
(-10.166, 2.036)

ADRA2A Mspl * Pb on
Inattention (errors omission):
2.870 (-2.340, 8.079)
ADRA2A Mspl x Pb on
Response Time Variability:
-1.588 (-6.526, 3.350)

ADRA2A Dral * Pb on
Inattention (errors omission):
5.066 (0.197, 9.934)

ADRA2A Dral x Pb on
Response Time Variability:
3.392 (-1.233, 8.017)

3-352

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tBoucher et al.
(2012a)

Nunavik region,

Montreal

Canada

1993-1998
(enrollment)

Followed through Sep.
2005-Apr. 2007

Cohort

Cord Blood
Monitoring Program
(CBMP)

One child from the
Environmental
Contaminants and
Child Development
Study (1996-2000)

n: 196

School children
without known
neurodevelopmenta
I disorder or
medication for
attention problems

Blood

Cord and child blood; GFAAS
with Zeeman background
correction (cord), ICP-MS
(child).

Age at measurement:

Delivery (cord), 9-13 yr (child)

Cord: 4.8 |jg/dL (mean), 3.7
|jg/dL (med); 20.9 |jg/dL (max)

Concurrent child: 2.2 |jg/dL
(mean), 2.0 |jg/dL (med), 12.8
|jg/dL (max)

Impairment in response
inhibition (Go/No-Go, ERPs
measured by EEG)

Electro-oculogram was
recorded from bipolar
miniature electrodes placed
vertically above and below
the right eye.

Age at outcome:

9-13 yr

Child age, sex, status
as adoptee; transport
by plane from remote
to larger village for
assessment; time of
assessment;
maternal age at
delivery; SES;
maternal nonverbal
reasoning abilities;
breastfeeding
duration; maternal
smoking, marijuana
use, binge drinking
during pregnancy;
docosahexaenoic
acid concentrations in
cord and child plasma
samples; Hg, PCBs

Beta per log-transformed
Pb: Cord blood

Mean Reaction Time (RT),
correct go trials: -0.05b
Mean RT, incorrect no-go
trials: -0.10b
Percent correct go trials:
-0.21 (-0.36, —0.06)c
Percent correct no-go trials:
-0.17 (-0.29, —0.05)c

Concurrent blood

Mean RT, correct go trials:
0.03b

Mean RT, incorrect no-go
trials: 0.03b

Percent correct go trials:
-0.12b

Percent correct no-go trials:
-0.16 (-0.27, —0.05)c

Rabinowitz et al.
(1992)

Taiwan

Study period NR
Cross-sectional.

N: 493

Mix of children
residing in urban or
rural environments
or near a smelter

Children grades 1-
3 recruited from
schools

Tooth

Child deciduous tooth; method
NR

Age at measurement: grades
1-3

Mean (SD): 4.6 (3.5) pg/g

Hyperactivity Syndrome

Boston Teacher
Questionnaire (BTQ)

Age at Outcome:
3

Grades 1-

Sex, # adults at
home.

Also considered
grade, child longest
hospital stay parental
education, SES, birth
outcomes,
handedness,
language at home,
and prenatal maternal
medicine, alcohol,
and smoking.

OR vs. <2.3 pg/g as
reference

2.3-7 pg/g: 1.9 (0.53, 6.5)
>7 pg/g: 2.8 (0.68, 12)

3-353

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

Chandramouli et al.
(2009)

Avon
U.K

10% random
subsample of Avon
Longitudinal Study
of Parents and
Children (ALSPAC)

n = 488

Jul.-Dec. 1992 (birth) Schoo, chNdren
Followed through 8 yr

Cohort

Blood

Earlier childhood venous
blood; AAS using micro
sampling flame atomization

Age at measurement:
30 mo

Mean (SD): NR
Group 1: 0-<2 |jg/dL

Group 2: 2-<5 |jg/dL

Group 3: 5-<10 |jg/dL

Group 4: >10 pg/dL

Parent and teacher rated
hyperactivity and attention

SDQ (7 yr), Development
and Well-Being Assessment
(DAWBA) (8 yr), Test of
Everyday Attention for
Children (TEACh) (8 yr)

Age at outcome:

7-8 yr

Maternal education
and smoking, home
ownership, home
facilities score, family
adversity index,
paternal SES,
parenting attitudes at
6 mo, child sex. Also
considered child IQ.

OR for increased score:
TEACh

Group 1: reference
Group 2: 1.03 (0.66, 1.61)
Group 3: 0.99 (0.62, 1.57)
Group 4: 1.14 (0.54, 2.40)

SDQ hyperactivity
Group 1: reference
Group 2: 0.84 (0.47, 1.52)
Group 3: 1.25 (0.67, 2.33)
Group 4: 2.82 (1.08, 7.35)

tSioen et al. (2013)

Flanders
Belgium

Oct. 2002 - Dec. 2003
(enrollment)

Followed through June
2011

Cohort

Flemish Health and
Environment Study
(FLEHS 1)
n: 270

Birth cohort of
Flemish children
living in either rural
or urban areas

Blood

Cord blood, HR-ICP-MS

Age at measurement:
Delivery

median = 14.3 |jg/L
75th: 25.3 pg/L

Hyperactivity

SDQ with 5 domains:
emotional, conduct,
hyperactivity, peer and
social problems

Age at outcome:

7-8 yr

Maternal and paternal
BMI, maternal age,
weight increase of
mother during
pregnancy, smoking
during pregnancy,
smoking behavior of
maternal

grandmother before
birth of mother,
parental education,
current parental
smoking, child sex,
serious infections of
child since birth (also
tested interaction by
sex)

OR per doubling of log-
transformed Pb:

Hyperactivity: 2.940 (1.172,
7.380)d

Total difficulties: 2.167
(0.741, 6.334)d

3-354

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tFruh et al. (2019) Project Viva	Blood

n: 1006

Eastern	Maternal venous blood; ICP-

Massachusetts	Birth cohort of	MS

U.S.	mother-child pairs

Age at measurement:
1999-2002	T2

(enrollment)

Followed through age	Median: 1.1 pg/dL

7 yr

Parent teacher ratings of
hyperactivity using SDQ

Standardized for child age
and sex

Age at outcome:

7 yr

Maternal 2nd
trimester Hg and Mn
levels, nulliparity,
smoking during
pregnancy, IQ, and
education; paternal
education; HOME
composite score and
household income;
and child
race/ethnicity

Beta per In-transformed Pb
for hyperactivity:

SDQ-parent: 0.10 (-0.21,
0.41)

SDQ-teacher: 0.20 (-0.24,
0.64)

Cohort

tHorton et al. (2018)

Mexico City
Mexico

born 1994-2006 and
followed through age
6-16
Cohort

ELEMENT Project
n: 133

healthy, low to
moderate income
mother (18-39 yr
old)-child pairs

Teeth

tooth Pb (prenatal, postnatal
metrics derived); laser ablation
ICP-MS

Age at measurement:
tooth Pb concentration
corresponded to prenatal and
300 days after birth

Figure 1c

Externalizing behavior
(attention and hyperactivity)

BASC-2: BSI, hyperactivity
and attention symptoms

Age at outcome:

8-11 yrold

Maternal age at
delivery, maternal
education, smoking,
maternal IQ

Beta per In-transformed Pb:
Attention: 0.19 (0.02, 0.37)e
BSI (composite): 0.22 (0.06,
0.38)e

tRasnick et al. (2021) CCAAPS
n: 263

Cincinnati, OH

Born: Oct 2001—Jul
2003

Exposure: 2001-2005
Cohort

Air

LURF, air sampling at 24 sites
(C-V R2 = 0.89), predicted air
concentration at child's
residence.

Children residing >1,500 m or
<400 m from major highway
eligible.

Attention problems using
BASC-2

Age at outcome: 12 yr

Maternal education,
community-level
deprivation, blood Pb
concentrations,
greenspace, and
traffic related air
pollution.

Beta per 1 ng/m3 increase in
monthly air Pb exposure:

Attention: 0.8 (0.1, 1.5)

Median: 0.51 ng/m3 (range 0-
10.8 ng/m3)

3-355

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tLiu etal. (2014b)

Jintan, Jiangsu

province

China

Sep. 1, 2004 - Apr.
30, 2005 (age 3-5 yr)
Followed to age 6 yr

Cohort

China Jintan Child
Cohort Study
n: 1025 children

Chinese preschool
children

Blood

Venous child blood; GFAAS
Age at measurement:
3-5 yr old

Mean (SD): 6.4 (2.6) pg/dL
median = 6.0 pg/dL
75th: 7.5 pg/dL
90th: 9.4 pg/dL
Max: 32 pg/dL

Attention and ADHD
problems

CBCL (Chinese version);
Caregiver-Teacher Report
Form; normalized T scores

Age at outcome:

6 yr

Age at BLL test, sex,
preschool residence,
father's educational
level, mother's
educational level,
father's occupation,
parents' marital
status, single child
status, and child IQ

Beta:

CBCL

Attention: 0.001 (-0.002,
0.002)

ADHD: 0.136 (-0.115,

0.386)

C-TRF

Attention: 0.001 (-0.002,
0.002)

ADHD: 0.073 (-0.177,
0.322)

OR:

C-TRF

ADHD all: 1.08 (0.99, 1.18)
Boys: 1.04 (0.94, 1.16)
Girls: 1.15 (0.98, 1.35)

tWinter and Sampson PHDCN

(2017)

Chicago, Illinois
U.S.

1995-1997 (birth)

Followed through 2013
(age 17 yr)

Cohort

n: 254

Children and
caregivers living in
Chicago

Blood

Avg BLL before age 6;
methods NR
Age at measurement:
6 yr old or younger

Mean: 6.4 pg/dL

Impulsivity score

CBCL PC questionnaire

Age at outcome:

Mean: 17 yr old

Age at CBCL
assessment, sex,
race/ethnicity,
primary caregiver
immigrant
generational status,
marital status,
education level,
Temporary
Assistance for Needy
Families receipt, and
the proportion of
residential

neighborhood that is
non-Hispanic Black,
Hispanic, below the
poverty line, and
tested for Pb
exposure

Beta: CBCL
Impulsivity: 0.06 (0.005,
0.115)

3-356

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tChoi etal. (2016)

10 Cities
South Korea

2006-2010
(enrollment at 1st-2nd
grade)

Followed through age
7-9 yr

Cohort

CHEER

n: 2195

Elementary school
children

Blood

Child venous blood; AAS with
Zeeman background
correction

Age at measurement:

7-9 yr

GM: 1.56 |jg/dL

ADHD symptomology

DuPaul's ADHD rating scale
per DSM-IV

Age at outcome:

After age 7-9 yr

Age, sex, residential
area, monthly
household income,
parental marital
status, family history
of psychiatric
disorders (anxiety
disorder, ADHD,
autism and
schizophrenia),
preterm birth and
birth weight

RR (BLL>2.17 vs. <2.17
|jg/dL) for ADHD symptoms:

1.552 (1.002, 2.403)

Single parent home and BLL
>2.17 pg/dLvs. 2-parent
home and BLL <2.17 |jg/dL:
3.567 (1.595, 7.980)

3-357

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tBoucher et al.
(2012b)

Nunavik, Arctic

Quebec

Canada

1993-2000

(enrollment)

Sep. 2005-Feb. 2010

(follow-up)

Cohort

Cord Blood
Monitoring Program
and Environmental
Contaminants and
Child Development
Study
n: 279

Inuit Children

Blood

Cord and child venous blood;
AAS (cord), ICP-MS (child)

Age at measurement:
delivery (cord), 11.3 yr (child)

Mean (SD): 4.7 (3.3) pg/dL
(cord); 2.7 (2.2) pg/dL (child)
Median: 3.7 pg/dL (cord); 2.1
pg/dL (child)

Max: 20.9 pg/dL (cord); 12.8
pg/dL (child)

ADHD symptomology
assessed using the TRF
from CBCL and the DBD
rating scale

Age at outcome:
11.3 yr (average)

Child age and sex,
SES, age of the
biological mother at
birth, maternal
tobacco use during
pregnancy, and birth
weight, Hg

Cord Blood:

Attention problems Beta
(95% CI) per log-
transformed Pb: 0.05 (-0.10,
0.19)d

OR (95% CI)

ADHD inattentive type

1st fertile referent

2nd fertile 2.77 (1.00, 7.65)d

3rd fertile 2.87 (1.04, 7.94)d

ADHD hyperactive-impulsive
type

1st fertile referent

2nd fertile 0.95 (0.30, 3.00)d

3rd fertile 2.92 (1.07, 8.04)d

Child Blood:

Attention problems Beta
(95% CI) per log-
transformed Pb: 0.08 (-0.05,
0.21 )d

OR (95% CI)

ADHD inattentive type

1st fertile referent

2nd fertile 1.06 (0.42, 2.66)d

3rd fertile 1.01 (0.38, 2.64)d

ADHD hyperactive-impulsive

type

1st fertile referent

2nd fertile 4.01 (1.06,

15.23)d

3rd fertile 5.52 (1.38, 22.12)d

3-358

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tDesrochers-Couture
et al. (2019)

Nunavik, Northern

Quebec

Canada

Nov. 1993-Mar. 2002
(enrollment)
Sep. 2005-Feb. 2010
(1st follow-up)

Jan. 2013-Feb. 2016
(2nd follow-up)

Cohort

NCDS-childhood
n: 212

Inuit children from
14 coastal villages
in Nunavik,

Quebec, subsample
from the Cord Blood
Monitoring Program
and NIH-infancy
study

Blood

Cord and child venous blood;
GFAAS (cord), ICP-MS (child)

Age at measurement:

Delivery (cord), 11.4, 18.5 yr
(child)

GM (GSD): 3.80 (1.84) pg/dL
(cord); 2.34 (1.86) pg/dL
(child); 1.63 (2.00) pg/dL
(adolescent)

Median:3.73 pg/dL (cord); 2.07
pg/dL (child); 1.52 pg/dL
(adolescent)

Max: 17.80 pg/dL (cord); 12.83
pg/dL (child); 18.13 pg/dL
(adolescent)

Teacher-rated ADHD
symptomology

Teacher assessed DBD and
TRF, Achenbach's YSR,
BAARS

Age at outcome:

11.4, 18.5 yr (average)

Child age, sex, SES,
maternal age at
delivery, maternal
tobacco smoking
during pregnancy,
and birth weight

Beta (95% CI):

Child Blood:

Child externalizing behavior:

0.23 (0.08, 0.38)

Child ADHD: 0.45 (0.13,
0.78)

Direct effect: 0.09 (-0.11,
0.28)

Indirect effect: -0.02 (-0.06,
0.03)

Adolescent externalizing
behavior mediated through
child externalizing behavior
((3: 0.09, 95% CI: 0, 0.17)

tHona et al. (2015)

5 administrative
regions
South Korea

Study years NR
Case-control

n: 1001

General population
of children in 3rd to
4th grades

Blood

Child venous blood; GFAAS
with Zeeman background
correction

Age at measurement:
8-11 yr

Median: 1.81 pg/dL
75th: 2.25 pg/dL,

95th: 3.01 pg/dL
Max: 6.16 pg/dL

ADHD symptomology

Teacher/parent ratings
ADHD symptoms (ADHD-
RS); CPT

Age at outcome:

8-11 yr

Age, gender,
residential region,
paternal education
level, and yearly
income log 10-
transformed blood
Hg, Mn, urine
concentrations of
cotinine, phthalate
metabolites full-scale
IQ

Beta (95% CI) Child blood:
ADHD-RS, parent-rated
1.04 (0.18, 1.90)
ADHD-RS, teacher-rated
1.90 (0.74, 3.05)

Additionally adjusted for
FSIQ, Mn, and Hg:
ADHD-RS, parent-rated
0.68 (-0.20, 1.56)

ADHD-RS, teacher-rated
1.49 (0.32, 2.67)

3-359

-------
Referent^and Study study popu|ation Exposure Assessment

Outcome

Confounders Effect Estil?fctfs and 95%

tNiqq etal. (2016)

Michigan
United States

Study years NR

Case-control

n: 386 children, 6-
17 yr old from 267
families (148
singletons, 119
sibling pairs)

Non-ADHD: 147
ADHD: 122

Blood

Child venous blood; ICP-MS
Age at measurement: 6-17 yr

Non-ADHD: mean (SD) = 0.74
(0.35) |jg/dL

ADHD: mean (SD) = 0.94
(0.52) |jg/dL

Composite parent and
teacher ratings of ADHD
symptoms using 3 scales:

ADHD-RS: inattention and
hyperactivity-impulsivity
symptom scores

CRS-R: cognitive
(inattention) and
hyperactivity problems
subscales

Gross annual income,
HFE mutations, race,
parenting behavior,
OD/CD, Fe
hemoglobin level, sex

SWAN: inattention and
hyperactivity symptom
scores

Age at outcome: 6-17 yr

Betas of hyperactivity-
impulsivity scores perz-
score increase in Pb
modified by HFE C282Y
mutation

Parent ratings:

Mutation: 0.74 (0.52, 0.96)e
Wild-type: 0.28 (0.15, 0.41 )e
Male: 0.31 (0.14, 0.48)e
Female: 0.09 (-0.16, 0.34)e

Teacher ratings:

Mutation: 0.47 (0.22, 0.72)e
Wild-type: 0.29 (-0.04,
0.12)e

Male: 0.19 (0.07, 0.31 )e
Female: 0.11 (-0.03, 0.25)e

3-360

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tJoo etal. (2018)

Seoul, Ulsan,
Cheonan
South Korea

2006-2011
(enrollment)

Followed through 5 yr
Cohort

MOCEH

n: 575 mother-child
pairs

Pregnant women at
12-20 wk of
pregnancy in
prenatal clinics and
public health
centers

Blood

Maternal venous, cord, and
child venous blood; AAS

Age at measurement: 20 wk
(maternal); delivery (maternal
and cord); 2, 3, and 5 yr (child)

GM: Maternal 1.28 |jg/dL
(early), 1.24 (late) 0.9 (cord);
Child 1.55 (age 2), 1.43 (age
3), 1.29 (age 5)

Attention and aggressive
behavior combined

K-CBCL: Externalizing
behavior (attention and
aggressive behavior
combined);

Age at outcome:

5 yr

Maternal age at
childbirth, parity,
maternal educational
level, household
income, residential
area, and
breastfeeding

Beta (95% CI):

Externalizing behavior at 5
yr

Maternal-early pregnancy
Male: -0.72 (-3.12, 1.69)
Female: -0.45 (-2.16, 1.26)
Maternal-late pregnancy
Male: 2.99 (0.55, 5.43)
Female: 0.24 (-2.18, 2.66)

Cord blood

Male: 3.09 (-0.08, 6.26)
Female: -0.16 (-3.33, 3.01)

Child blood-2 yr
Male: 0.55 (-1.52, 2.62)
Female: 3.50 (0.97, 6.03)
Child blood-3 yr
Male: 1.13 (-1.42, 3.68)
Female: 2.05 (-1.35, 5.45)
Child blood-5 yr
(concurrent)

Male: 1.42 (-2.12, 4.95)
Female: 4.53 (-0.81, 9.86)

3-361

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

tJi etal. (2018)

Boston,

Massachusetts
U.S.

1998-2013
(enrollment)

Followed through 2016

Cohort

Boston Birth Cohort
n: 299 ADHD
cases, 1180
neurotypical
controls

Mother-infant pairs

Blood

Child blood; method NR,
obtained from electronic
medical records

BLLs

Age at measurement:
<4 yr; the earlier BLL was
selected where multiple BLLs
were recorded

Mean (SD): 2.2 (1.6) pg/dL

Diagnosed ADHD

Physician-diagnosed ADHD
from electronic medical
records

Age at outcome:

Median: 6 yr

Maternal age at
delivery, maternal
race/ethnicity,
maternal education,
smoking during
pregnancy,
intrauterine infection,
parity, child's sex,
mode of delivery,
preterm birth, and
birth weight

OR (95% CI)

Continuous BLL: 1.118
(1.003, 1.247)

Categorical BLL:

2-4 vs. <2 pg/dL: 1.08
(0.81, 1.44)e

5-10 vs. <2 pg/dL: 1.73
(1.09, 2.73)e

Sex-stratified:

Girls 5-10 vs. <5 pg/dL:
0.68 (0.27, 1.69)e

Boys 5-10 vs. <5 pg/dL:
2.49 (1.46, 4.26)e

Joint Effects of sex and
BLL category:

Girls*5-10 pg/dL: 0.69
(0.28, 1.71 )e

Boys*<5 pg/dL: 3.02 (2.24,
4.06)e

Boys*5-10 pg/dL: 7.48
(4.29, 13.02)e

tParketal. (2016)
Busan

South Korea
Apr.-Sep. 2013
Case-control

n: 114 cases
(diagnosed ADHD),
114 controls

Recruitment from
child psychiatric
and pediatric clinics
from four university
hospitals

Blood

Child venous blood; GFAAS
with Zeeman background
correction

Age at measurement:

6-12 yr

GM (GSD): 1.90 (0.86) pg/dL
(cases); 1.59 (0.68) pg/dL
(controls)

Diagnosed ADHD

Diagnosed ADHD
(confirmed by [K-SADS-PL-
K]); CPT and parent-rated
ADHD symptoms among
ADHD cases

Age at outcome:

6-12 yr

Age, sex-matched OR (95% CI) per log-

controls; gestational
age, birth weight,
SES, parental
education, and
parents' smoking
behavior

transformed Pb:
ADHD total
1.60 (1.04—2.45)d

3-362

-------
RefereDCes?gnnd	studV Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

n: 71 cases
(diagnosed ADHD);
58 controls

tKimetal. (2013a)

Omaha, Nebraska
U.S.

Children living near
Aug 2007-Dec 2009 a former refinery

Case-control

Blood

Child venous blood; ICP-MS

Age at measurement:

5-12 yr

GM: 1.29 |jg/dL (cases); 1.33
|jg/dL (controls); 1.65 |jg/dL
(inside Pb investigation area);
1.01 |jg/dL (outside Pb
investigation area)

ADHD

Physician-diagnosed
according to DSM-IV

Age at outcome:
5-12 yr

Matched on age, sex, OR (95% CI) per In
race and adjusted for transformed Pb
maternal smoking,

SES, and
environmental
tobacco exposure

ADHD Overall 2.52 (1.07-
5.92)e

tGeieretal. (2018) NHANES

Blood

ADD

Sex, age, SE

ES, race OR (95% CI):

n: 2109

Representative sample

U.S. Children

Child venous blood: ICP-MS

Self-reported doctor
diagnosed ADD



ADD 1.292 (1.025-1.545)

2003-2004

Age at measurement:
10-19 yr

Age at outcome:
10-19 yr





Cross-sectional

Mean (SD): 1.16 (1.27) pg/dL







3-363

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
Clsa

Braun et al. (2006)

Representative sample
U.S.

1999-2002
Cross-sectional

NHANES
n: 4704

Children 4-15 yr

Blood

Venous blood: GFAAS
Age at measurement:
4-15 yr old

Quintiles:
ND-0.7 |jg/dL

0.8-1.0 |jg/dL

1.1-1.3 Mg/dL

1.4-2.0 |jg/dL

679
795
857
745

Parent-reported ADHD with
prescription stimulant use

Age at outcome: 4-15 yr old

Age, sex, race,
prenatal ETS
exposure, postnatal
ETS exposure, BLLs,
preschool or
childcare attendance,
health insurance
coverage, and ferritin
levels

AOR (95% CI) Child blood:

2nd quintile (0.8-1.0): 1.1
(0.4-3.4)e

3rd quintile (1.1-1.3): 2.1
(0.7-6.8)e

4th quintile (1.4-2.0): 2.7
(0.9-8.4)e

5th quintile (>2.0): 4.1 (1.2-
14.0)e

>2.0 |jg/dL: 995

AAS = atomic absorption spectrometry; ADHD = attention deficit/hyperactivity disorder; ADHD-RS = ADHD rating scale; ADRA2A = adrenoceptor alpha 2A; AOR = adjusted odds
ratio; BAARS = Barkley Adult ADHD-IV Rating Scale; BASC = Behavior Assessment System for Children; BLL = blood lead level; BMI = body mass index; BRIEF = Behavior Rating
Inventory of Executive Functions; BSI = Behavioral Symptoms Index; CARES = Communities Actively Researching Exposure Study; CBCL = Child Behavior Check List; Cd =
cadmium; CHEER = Children's Health and Environmental Research; CI = confidence interval; CKiD = Chronic Kidney Disease in Children Study; CPT = Continuous Performance
Test; CRS-R = Conners' Rating Scale-Revised; C-TRF = Caregiver-Teacher Report Form; DAT1 = dopamine transporter; DBD = Disruptive Behavior Disorders; DRD2 = dopamine
receptor D2; DSM = Diagnostic and Statistical Manual of Mental Disorders; EEG = electroencephalogram; ELEMENT = Early Life Exposure in Mexico to Environmental Toxicants;
ERP = event-related potentials; FBB-ADHS = Fremdbeurteilungsbogen fur Aufmerksamkeitsdefizit/Hyperaktivitatstorungen; Fe = iron; FLEHS = Flemish Health and Environment
Study; GFAAS = graphite furnace atomic absorption spectrometry; GM = geometric mean; GMR = geometric mean ratio; HFE = hemochromatosis gene; Hg = mercury; HOME =
Health Outcomes and Measures of the Environment; ICP-MS = inductively coupled plasma mass spectrometry; K-ABC = Kaufman Assessment Battery For Children; K-CPT =
Conners' Kiddie Continuous Performance; KiTAP = Test of Attentional Performance for Children; K-SADS-PL-K = Kiddie Schedule for Affective Disorders and Schizophrenia Present
and Lifetime - Korean Version; LURF = Land Use Random Forest; Mn = manganese; mo = month(s); MOCEH = Mothers' and Children's Environmental Health; NCDS = Nunavik
Child Development Study; NR = not reported; OD/CD = oppositional defiant and conduct disorder; OR = odds ratio; Pb = lead; PCBs = polychlorinated biphenyls; PHDCN = Project
on Human Development in Chicago Neighborhoods; RR = relative risk; RT = reaction time; SCWT = Stroop Color-Word Test; SD = standard deviation; SDQ = Strengths and
Difficulties Questionnaire; SE = standard error; SES = socioeconomic status; SPM = Standard Progressive Matrices; SRS = Social Responsiveness Scale; SWAN = Strengths and
Weaknesses of ADHD Symptoms and Normal Behavior Scale; T1 = first trimester of pregnancy; T2 = second trimester of pregnancy; T3 = third trimester of pregnancy; TEACh =

Test of Everyday Attention for Children; TSCD = Tohoku Study of Child Development; WAIS = Wechsler Adult Intelligence Scale; wk = week(s); WMS = Weschler Memory Scale; yr
= year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 jjg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a

change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

bResults are unstandardized because they did not have an associated SE, CI, or p-value reported in the study.

The CI was calculated from a p-value and the true CI may be wider or narrower than calculated.

dResults are unstandardized because the log base used for exposure transformation was unspecified in the study.

eResults are unstandardized because the Pb level distribution data was not available.

tStudies published since the 2013 Integrated Science Assessment for Lead.

3-364

-------
Table 3-7T

Animal toxicological studies of Pb exposure and externalizing and internalizing behaviors

Study

Species (Stock/Strain), n, Sex Timing of Exposure ^Details8 BLL QjgAdL)°rted Endpoints Examined

Externalizing Behavior

Tartaalione et al. (2020)

Rat (Wistar) GD 28 to PND 23

Oral,

PND 23:

PND 4, 7, 10, 12: Ultrasonic



Control (tap water), M/F n = 16

lactation



Vocalizations



(9/7)

In utero

0.007 |jg/mL (0.7 pg/dL)









for Control





50 mg/L, M/F, n = 16 (9/7)













0.255 pg/mL (25.5









pg/dL) for 50 mg/L



Internalizing Behavior

Corv-Slechta et al.

Mouse (C57BL/6) GD -60 to 12 mo

Oral,

PND 75 - Females:

7-12 mo: FST

(2013)

Control (distilled deionized

drinking







water) - NS, M/F, n = 8-16

water

-------
Study	Species (Stock/Strain), n, Sex Timing of Exposure E^P°s.!J_re BLL f,® ^®.p°rted

Endpoints Examined

-------
Study	Species (Stock/Strain), n, Sex Timing of Exposure E^P°s.!J_re BLL f,® ^®.p°rted

Endpoints Examined

for Control

9.21 ng/g (0.98 pg/dL)
for 10 |jg/mL

PND60:

0.23 ng/g (0.024 pg/dL)
for Control

0.30 ng/g (0.032 pg/dL)
for 10 pg/mL

Faulk et al. (2014)

Mouse (Agouti) GD-14toPND21
Control (distilled water), M/F, n
= 30

2.1 ppm, M/F, n = 28
16 ppm, M/F, n = 33
32 ppm, M/F, n = 29

Oral,
lactation
In utero

PND 21 (Maternal BLL):
-------
Study	Species (Stock/Strain), n, Sex Timing of Exposure E^P°s.!J_re BLL f,® ^®.p°rted

Endpoints Examined

6.96 |jg/dL for 0.2%
solution

18 mo:

0.12 |jg/dL for Control

11.2 |jg/dL for 0.2%
solution

Mansouri et al. (2012)

Rat (Wistar) PND 70 to PND 100
Control (distilled water), M/F, n
= 16 (8/8)

50 mg/L, M/F, n = 16 (8/8)

Oral,

drinking

water

PND 100-Males: PND 100: OFT

2.05 |jg/dL for Control

8.8 |jg/dL for 50 mg/L

PND 100 - Females:

2.17 |jg/dL for Control

6.8 |jg/dL for 50 mg/L

Duan et al. (2017)

Mouse (CD1) PND 1 to PND 21

Oral,

PND 21: PND 7, 11, 15, 19: TST, OFT



Control (distilled water), M/F, n

lactation





= 5



16.2 |jg/L (1.6 |jg/dL) for







Control



27 ppm, M/F, n = 5











191.8 |jg/L (19.2 pg/dL)



109 ppm, M/F, n = 5



for 27 ppm

283.4 |jg/L (28.3 |jg/dL)
for 109 ppm

PND 35:

14.3 |jg/L (1.4 |jg/dL) for
Control

283.4 |jg/L (28.3 |jg/dL)
for 27 ppm

3-368

-------
Study	Species (Stock/Strain), n, Sex Timing of Exposure E^P°s.!J_re BLL f,® ^®.p°rted

Endpoints Examined







376.9 |jg/L (37.7 pg/dL)









for 109 ppm



Wana et al. (2016)

Rat (Sprague Dawley)

PND 24 to PND 56 Oral,

PND 56:

PND 60-66: OFT



Control (tap water), M, n = 7

drinking









water

11 pg/L (1.1 pg/dL) for





100 ppm, M, n = 9



Control









133 pg/L (13.3 pg/dL) for









100 ppm



Shvachiv et al. (2018)

Rat (Wistar)

Intermittent Exposure: GD Oral,

PND 196:

PND 189: OFT, EPM



Control (tap water), M/F, n = 8

7 to PND 84, PND 140 to drinking









PND 196 water

<0.1 pg/dL for Control





0.2% (p/v) solution (distilled

Oral,







water), M/F, n = 9 - Intermittent

Continuous Exposure: lactation

18.8 pg/dL for 0.2%





exposure

GD 7 to PND 196 In utero

(Intermittent)





0.2% (p/v) solution, M/F, n = 9



24.4 pg/dL for 0.2%





- Continuous exposure



(Continuous)



Basha and Reddv (2015)

Rat (Wistar)

GD 6 to GD 21 In utero

PND 21:

PND 21, PND 28, 4 mo: OFT,



Control (deionized water), M, n





Hole Board Test



= 8



0.21 pg/dL for Control





0.2 % solution, M, n = 8



11.2 pg/dL for 0.2%









solution









PND 28:









0.33 pg/dL for Control









12.3 pg/dL for 0.2%









solution









4 mo:









0.19 pg/dL for Control



3-369

-------
Study	Species (Stock/Strain), n, Sex Timing of Exposure E^P°s.!J_re BLL f,® ^®.p°rted

Endpoints Examined

5.9 |jg/dL for 0.2%
solution

Stansfield et al. (2015)

Rat(Long-Evans)

Control (chow), M/F, n = 11-23

1500 ppm, M/F, n = 11-23

GD Oto PND 50

Oral, diet
Oral,
lactation
In utero

PND 50:

0.6 |jg/dL for Control
22.2 |jg/dL for 1500 ppm

PND 50: Locomotor Activity

Flores-Montova and

Mouse (C57BL/6)

PND 0 to PND 28

Oral,

>PND 28 Males:

>PND 28 Hole Board Test, OFT

Sobin (2015)

Control (distilled water), M/F, n



drinking







= 19 (8/11)



water

0.2 |jg/dL for Control









Oral,







30 ppm, M/F, n = 26 (16/10)



lactation

3.93 |jg/dL for 30 ppm





230 ppm, M/F, n = 16 (12/4)





9.39 |jg/dL for 230 ppm











>PND 28 Females:











0.19 |jg/dL for Control











3.19 |jg/dL for 30 ppm











12.14 |jg/dLfor230 ppm



Neuwirth et al. (2019a)

Rat(Long-Evans)

Control (tap water), M/F, n =
NR

150 ppm, M/F, n = NR

GD Oto PND 22

Oral,
lactation
In utero

PND 22:

-------
Study

Species (Stock/Strain), n, Sex Timing of Exposure

Exposure
Details

BLL as Reported
(Mg/dL)

Endpoints Examined

-------
Study

Species (Stock/Strain), n, Sex Timing of Exposure

Exposure
Details

BLL as Reported
(Hg/dL)

Endpoints Examined



F3:









see Figure 1, n = 8-10







Sinqh et al. (2019)

Rat (Wistar) 3 mo to 6 mo

Control (distilled water), M, n =

5

2.5 mg/kg, M, n = 5

Oral, gavage

6 mo:

5.76 |jg/dL for Control
28.4 |jg/dL for 2.5 mg/kg

6 mo: EPM, Locomotor Activity

Al-Qahtani et al. (2022) Mouse (Albino)	8-9 wk to 14-15 wk Oral, gavage 14-15 wk:	NR: EPM, Locomotor Activity

Control (distilled water), M, n =

10	1.2 |jg/100 mL (1.2

|jg/dL) for Control

0.2 mg/kg, M, n = 10

7.1 |jg/100 mL (7.1
|jg/dL) for 0.2 mg/kg

BLL = blood lead level; EPM = elevated plus maze; F = female; FST = forced swim test; GD = gestational day; LOD = limit of detection; M = male; MRI = magnetic resonance
imaging; mo = month(s); NR = not reported; NS = no stress; OFT = open-field test; Pb = lead; PG = pregestation; PND = postnatal day; PS = prenatal stress; TST = tail suspension
test; wk = week(s); yr = year(s).

3-372

-------
Table 3-8E Epidemiologic studies of Pb exposure and performance on neuropsychological tests of attention,
impulsivity, and hyperactivity, attention deficit/hyperactivity disorder-related behaviors, and
clinical attention deficit/hyperactivity disorder in children; group or population mean blood Pb
level >5 (jg/dL, any study design

Referent^and Study study Population Exposure Assessment	Outcome	Confounders Effect Estimates and 95% Clsa

tArbuckle et al. (2016a)

representative population

Canada

2007-2009

Cross-sectional

CHMS
n: 1080

Representative sample of
children

Blood

Child venous blood;
analytic method NR
Age at measurement:
6-11 yr old

GM: 0.90
95th: 1.96 pg/dL

ADHD symptoms

SDQ, parent-reported
ADD/ADHD

Age at outcome:
6-11 yr old

Age, sex,	ORb

neonatal unit, Parent-Reported Outcomes
maternal

smoking child ADD/ADHD: 2.08 (1.01, 4.25)
age

(Supplemental Any Learning Disability: 1.41 (0.73,
2.70)

Psychotropic Medicine Taken: 2.91
(1.47, 5.79)

Table 1)

SDQ

Total Difficulties, Prenatal Smoking:
10.57 (2.81, 39.69)

Total Difficulties, No Prenatal
Smoking: 1.98 (1.41, 2.79)

Emotional Symptoms: 1.25 (0.60,
2.59)

Hyperactivity/lnattention: 2.75
(1.46, 5.16)

3-373

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tArbuckle et al. (2016b)

representative sample
Canada
2007-2009
Cross-sectional

CHMS	Blood

n: 2097

Child venous blood
Representative sample of Age at measurement:
children	6-19 yr old

ADHD symptomology

SDQ, reported ADD or
ADHD

Age at outcome:
6-19 yr old

Smoking, sex,
income

ORb

Parent or Self-Reported Outcomes,
Ages 6-19

ADD/ADHD: 2.39 (1.32, 4.32)

Learning Disability (Low Income):
0.81 (0.37, 1.81)

Learning Disability (High Income):
2.78 (1.40, 5.51)

Medicine Taken (Fasting Sample):
0.83 (0.34, 2.02)

Medicine Taken (Non-Fasting):
4.20 (1.92, 9.17)

SDQ, Ages 6-17

Total Difficulties: 2.16 (1.33, 3.51)

Emotional Symptoms: 1.08 (0.68,
1.71)

Hyperactivity/lnattention: 2.33
(1.59, 3.43)

3-374

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tBara et al. (2018)

Montevideo

Uruguay

Cross-sectional

n: 206

Children living in areas
considered high risk for
metal exposure

Blood

Child venous blood;
flame AAS or GFAAS
Age at measurement:
5-8-year-old

4.2 |jg/dL

teacher-rated ADHD and
hyperactive behavior

CRS-R: hyperactive,
oppositional, cognitive,
and ADHD-like behaviors
(teacher ratings)

Age at outcome:
5-8-year-old

Child IQ, iron
status, and
BMI, blood Pb
testing method,
household
possessions,
maternal
education,
current parent
smoking

PRs

Cognitive Problems/Inattention
Total population (>5 vs. 5 |jg/dL):
1.02 (0.967, 1.076)

Girls: 1.01 (0.995, 1.025)

Boys: 1.01 (0.99, 1.03)

Hyperactivity

Total population (>5 vs. 5 |jg/dL):
1.01 (0.947, 1.077)

Girls: 1.02 (1, 1.04)

Boys: 0.99 (0.97, 1.01)

ADHD Index

Total population (>5 vs. 5 |jg/dL):
1.01 (0.952, 1.072)

Girls: 1.01 (0.99, 1.03)

Boys: 1.00 (0.98, 1.02)

3-375

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tChan etal. (2015)

10 locations

U.S.

Cohort

National Institute of Child
Health and Human
Development, Study of
Early Child Care and
Youth Development
n: 266

School children

Teeth (Shed molars)

ICP-OES
Mean: 0.46 |jg/g

Disruptive behavior and
ADHD subscales

TBD completed by 3rd
grade teachers; scores
for (1) Total Disruptive
Behavior; (2) subscale
scores for ADHD,
hyperactivity/impulsivity,
inattention, and OD

Race, sex,

paternal

education,

maternal

education,

marital status,

and family SES

Change in behavior score per
|jg/g of Pb concentration in
teeth:0

DBD: -0.05

ADHD: -0.03

Impulsive: -0.06

Inattention: 0.00

Defiance: -0.09

Age at outcome:
teeth collected at 8-11 yr
old (body burden)

tFornset al. (2014)

Catalonia

Spain

Cohort

INMA
n: 385

Children of mothers
enrolled in the
population-based cohort
as part of the INMA
(Environment and
Childhood) Project

Urine

Maternal urine; ICP-MS,
values below LOD were
imputed

Age at measurement:
T1, T3

Median: 3.44, 1st; 3.63
3rd

75th: 4.64 1st, 4.84

ADHD symptoms

ADHD-DSM-IV criteria
and MSCA

Age at outcome:
4 yr old

Age, maternal Change in neuropsychological
social class, outcomes per ng/mL increase in
and maternal mother's urinary Pb
mental health concentration:

T1

GCI MSCA: 1.46 (-2.76, 5.69)

EF MSCA: 0.34 (-3.95, 4.63)

T3

GCI MSCA: -1.27 (-5.71, 3.17)

EF MSCA: -0.74 (-5.24, 3.75)

IRR:

T1

Inattention: 0.92 (0.57, 1.46)
Hyperactivity: 1.04 (0.65, 1.65)

T3

Inattention: 0.71 (0.43, 1.18)
Hyperactivity: 1.04 (0.64, 1.70)

3-376

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tGu etal. (2018)

Wuhan
China

Case-control

Hospital based case-
control, recruitment:
n: 389 cases; 392
controls

Children 6-18 yr old

Blood

Venous sample, whole
blood, AAS
Median: 5.685 |jg/dL

ADHD, subtypes
inattention, hyperactivity
and impulsivity and
combined

Cases: ADHD DSM-IV
(subtypes defined by
inattention, hyperactivity
and impulsivity [HI] and
combined type [C]);
WISC and Parent
Symptom Questionnaire

Age at outcome:
6-18 yr old

Age, sex

(cases and

controls

compared on

IQ, maternal

alcohol,

smoking,

parental

relationship,

breastfeeding)

ORb

BLL <56.85 pg/L,

BLL >56.85 pg/L,
(1.132, 3.075)

BLL <56.85 pg/L,
(0.806, 1.954)

GG: Reference

GG: 1.865

GA/AA: 1.255

BLL >56.85 pg/L, GA/AA: 1.871
(1.014, 3.451)

tGump etal. (2017)

Upstate New York
U.S.

Cross-sectional

Environmental
Exposures and Child
Health Outcomes
n: 203

children residing in low-
to middle-income
communities

Blood

venous blood;

Age at measurement:
9-11 yr old

Externalizing behavior: Sex, race, age, Beta (95%)b

attention, impulsivity,
hyperactivity

DBD for ADHD
inattentive type and
ADHD hyperactive-
impulsive type; ASQ:I
questionnaire for ASD
(parent-rated); acute
vagal response for stress
(heart rate variability)

Age at outcome:
9-11 yr old

and SES and
Hg

ADHD-Inattention (Score) 0.01
(-0.14, 0.15)

ADHD-Hyperactivity (Score) 0.16
(0.02, 0.30)

Oppositional Defiant Disorder
(Score) 0.16 (0.02, 0.31)

3-377

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

Adjusted associations between a 1-
pg/dL increase in blood lead

Cognitive Problem/Inattention -
0.03 (-0.3, 0.2)

Hyperactivity 1.2 (0.3, 2.0)

ADHD Index 0.02 (-0.2, 0.3)

CGI Restless-Impulsive 1.2 (0.3,
2.0)

CRS-R DSM-IV
Inattentive 0 (-0.3, 0.3)
Hyperactive-Impulsive 1.1 (0.2, 2.0)
Total 0.03 (-0.2, 0.3)

tJoo et al. (2017)	n: 214 cases (=19 on the

K-ARS or ADHD
Cheonan	diagnosis); 214 control

South Korea	(49 elementary schools)

2008-2010

Case-Control	Elementary school

children

tKicinski et al. (2015)

n: 606

Blood

Sustained attention,
short-term memory,

R gender, age,
smoking,

Effect estimates between BLL and
neurobehavioral outcomes not

Flanders

Third year secondary

venous blood, ICP-MS

manual motor speed

passive

reported due lack of statistical

Belgium

school students in two

Age at measurement:



smoking,

significance

2008 and 2011

industrial areas in

13.6-17 yr old

CPT, NES

household



Cross-sectional

Flanders, Belgium

Mean: 13.8 |jg/dL
95th: 28.1 pg/dL

Age at outcome:
13.6-17 yr old

income per
capita, the
highest
occupational
category of
either parent,
and the

education level
of the mother



tHuana et al. (2016)

Mexico City
Mexico
1997-2001
Cross-sectional

ELEMENT

n: 578

Mother-child pairs

Blood

Venous blood; ICP-MS
Age at measurement:
6-13 yr old

Mean: 3.4 |jg/dL

ADHD symptomology

CRS-R, CRS-R DSM-IV

Age at outcome:
6-13 yr old

Maternal
marital status,
age,

educational
years, SES,
smoking during
pregnancy,
child's age,
sex, birth
weight.

Blood

Venous blood, AA
spectrophotometry
GM: 1.65 (cases) |jg/dL;
1.49 pg/dL (controls)

ADHD symptomology
K-ARS

Age at outcome:
6 to 10 yr old

Maternal	ORb

education, AN ADHD: 1.28 (0.89, 1.83)
family history
of ADHD,

parental	Inattention: 1.63 (1.03, 2.58)

marital status,

ar|d teenage Hyperactivity/impulsivity: 1.04
mother	(Q ^ 2 Q7)

3-378

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tLin etal. (2019)

n: 164

Blood, Bone

ADHD symptoms and
comorbidities

Children's age;
sex, passive

Xinhua

Children who visited a

Child venous blood; AAS



smoking (the

China

lead specialty clinic in

Tibia bone; XRF

Vanderbilt-ADHD

frequency of

Aug. 2014 - Aug. 2015

Xinhua Hospital from

Age at measurement:

Diagnostic-Parent-Rating

smoking by

Cross-sectional

August 2014-August

3-15 yr

Scale

parents and



2015



other





GM:

Age at outcome:

household





Blood:

3-15 yr

members in







the presence





Low: 4.3 |jg/dL



of children),





High: 19.6 pg/dL



parity,





Bone:



maternal







education





Low: 0.3 pg/g



levels and





High: 12.8 pg/g



family yearly
income

ORb

Inattention

BLL <10 |jg/dL: Reference
BLL >10 |jg/dL: 3.3(0.9, 12.4)

Hyperactivity/impulsivity
BLL <10 |jg/dL: Reference
BLL >10 |jg/dL: 2.0 (0.5, 7.5)

Oppositional defiant disorder
BLL <10 |jg/dL: Reference
BLL >10 |jg/dL: 2.7 (0.8, 8.9)

tLiu etal. (2014e)

n: 240

Blood

ADHD symptomology

Age and

Measures of association for blood









gender,

lead and attention outcomes not

Guiyu

Native 3-7 yr old

Child venous blood;

ADHD (H,I,C) perDSM-

residential site,

reported. Study only reports

China

kindergarten children

GFAAS

IV; CPRS-R, CTRS-R,

time, heavy

correlation analyses.

2009 -2011

who have resided in

Age at measurement:

Rutter Child Behavior

metal exposure



Cross-sectional

Guiyu for more than 2 yr

3-7 yr

Questionnaire (antisocial







after birth



behavior, neurotic)









Median: 7.33 pg/dL











75th: 9.13 pg/dL

Age at outcome:











3-7 yr





3-379

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tLucchini et al. (2012)

Valcamonica and Garda
Lake areas in Province of
Brescia
Italy

Cross-sectional

Junior high school-age
children from 20 local
public schools
n: 299

Blood

Child venous blood;
GFAAS

Age at measurement:
11-14 yr

1.71 pg/dL, Median: 1.50
75th: 2.10 pg/dL
Max: 10.2 pg/dL

Conners'-Wells'
Adolescent Self-Report
Scale Long Form

10 subscales: family
problems, emotional
problems, conduct
problems, cognitive
problems/inattention,
anger control problems,
hyperactivity, ADHD
index, DSM-IV
(disattention), DSM-IV
(hyperactivity/impulsivity)
, and DSM-IV (Total)

Age at outcome:

11-14 yr

Sex, age at
testing,
parental
education,
SES, family
size, parity
order, BMI

Betas

Performance IQ: -1.991 (-3.918,
-0.064)

Verbal IQ: -1.863 (-3.79, 0.064)

Total IQ (Table 4): -2.237 (-4.101,
-1.372)

Total IQ (Table 5): -2.248 (-4.111,
-0.385)

tMunoz et al. (2020)

Arica
Chile

2009-2015
Cross-sectional

n: 2656

Children enrolled in a
heavy metal intervention
program

Blood

Child venous blood; AAS
Age at measurement:
3-17 yr

Median: 1.0 pg/dL
75th: 2.0 pg/dL

Parent-reported attention
deficit and hyperactivity
recorded in medical
records

Age at outcome:

3-17 yr

Age, sex,
parents' report
of children
exposure to
secondhand
tobacco
smoke,
housing
material quality

ORb

BLL >5 pg/dL: 2.33 (1.32, 4.12)

tRodriques et al. (2018) n: 225

Blood

Salvador, Bahia
Brazil

Cross-sectional

Children living near alloy GFAAS
plant	Age at measurement:

7-12 yr

1.2 pg/dL
Max: 15.6 pg/dL

Child behavior

CBCL: 8 domains
including attention

Age at outcome:
7-12 yr

Sex, age,

height-for-age

Z-score,

maternal

schooling,

socioeconomic

classification,

and community

violence index,

as well as

maternal IQ

Change in Behavior (Total raw
score) per log-yg/dL decrease in
BLL:

-1.08 (-11.5, 9.3)

Change in Behavior (Total score T)
per log-ijg/dL decrease in BLL:
-0.74 (-5.3, 3.8)

3-380

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tSkoqheim et al. (2021)

Norway
2002-2009

Case-control

Norwegian Mother,
Father and Child Cohort
Study (MoBa)

n: 397 ASD cases, 1034
controls

Children

Blood

Maternal whole blood;
ICP-SFMS atwk 17 of
gestation

Age at Measurement:
Prenatal, Week 17 og
gestation

GM (95% CI) (cases):
0.835 (0.797, 0.875)
pg/dL

GM (95% CI) (controls)
0.882 (0.860, 0.905)
pg/dL

ADHD

Diagnosis of ADHD
(NPR)

Age at outcome: 3 or less

Child sex, birth
weight, birth
year, and SGA,
maternal age
at delivery,
education,
parity, pre-
pregnancy
BMI, kg/m2),
self-reported
smoking and
alcohol intake
during
pregnancy,
FFQ-based
estimates of
seafood intake
(g/day), and
dietary iodine
intake (|jg/day)

ORb

ADHD

Q1 (Reference):

Q2
Q3
Q4

1.15 (0.87,
0.84 (0.63,
1.09 (0.82,

1.52)
1.12)
1.45)

tSobin et al. (2015)
U.S.

Cross-sectional

n: 421

Elementary school
children

Blood

Attention

2 samples 60 days apart Age at outcome:
averaged; ICP-MS or Pb 5.1-11.8 yr old
Care I

Age at measurement:

5.1—11.8 yr old

Mean: 2.7 |jg/dL (males);

2.4 |jg/dL (females)

of education

Sex, age and Beta

mother s level Motor dexterity non-dominant hand

^	1 g3 (-1 343^ 5 2Q3)

Working memory misses: 0.11
(0.051, 0.169)

Working memory false alarms
errors: 0.05 (-0.009, 0.109)

Visual attention 5-choice movement
time (ms): 26.07 (11.331, 40.809)

3-381

-------
Referencejmd Study study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tSoetrisno and Delaado-
Saborit (2020)

West Java (Depok,

Bogor and Bekasi)
Sukatani village (control)
Indonesia
Cross-sectional

School children living in
urban locations near e-
waste facility; control site
n: 44 (22 from Bogor and
22 from Sukatani)

Children selected from
schools per teachers/
principal

recommendation

Hair, soil, water

hair samples from
children in Bogor and
Sukatani village. BLLs
from 36 children in Bogor
area (2010).

Age at measurement:
6-9 yr

Soil Pb mean: Depok-
Bekasi: 3653 mg/kg;
Sukatani: 93.2 mg/kg;
Water Pb: all 10 samples
below LOD; Hair Pb:
Depok-Bekasi: 0.155
mg/g; Sukatani: 0.0729
mg/kg

Max: Soil Pb: Depok-
Bekasi: 7662 mg/kg;
Sukatani: 115 mg/kg;

Hair Pb: Depok-Bekasi:
0.841 mg/g; Sukatani:
0.255 mg/kg

Visual attention TMT A
Age at outcome:

6-9 yr

Age, parental
education,
environmental
tobacco smoke
at home, and
residential
traffic exposure

Change in TMT-A (seconds) per
mg/g unit of hair Pb

2.5 (-55, 60)

3-382

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders Effect Estimates and 95% Clsa

tZhana et al. (2015a)

Guangdong
China

Jan. 2012-May 2012
Cross-sectional

n: 243

Preschool children
residing near e-recycling
plant

Blood

Child venous blood,
GFAAS

Median: 7.9 |jg/dL
95th: 16.9 pg/dL

ADHD symptomology Age, sex,	OR

father's work in ADHD: 2.4 (1.1, 5.2)
Parent rating per DSM-IV e-waste,

ADHD criteria	Serum ferritin,

E-waste

Age at outcome:	workshops

3-7 yr	around the

house

ADD = attention deficit disorder; ADHD = attention deficit/hyperactivity disorder; ASD = autism spectrum disorder; ASQ:I = Ages and Stages Questionnaire Inventory; BLL = blood
lead level; BMI = body mass index; BRIEF = Behavior Rating Inventory of Executive Functions; CBCL = Child Behavior Check List; CHMS = Child Health Monitoring System; CI =
confidence interval; CPRS-R = Conners' Parent Rating Scale-reformed; CPT = Continuous Performance Test; CTRS = Conners' Teacher Rating Scale; DBD = Disruptive Behavior
Disorders; DSM = Diagnostic and Statistical Manual of Mental Disorders; ELEMENT = Early Life Exposure in Mexico to Environmental Toxicants; FFQ = Food Frequency
Questionnaire; GFAAS = graphite furnace atomic absorption spectrometry; GM = geometric mean; Hg = mercury; ICP-MS = inductively coupled plasma mass spectrometry; ICP-
OES = inductively coupled plasma optical emission spectrometry; ICP-SFMS = inductively coupled plasma sector field mass spectrometry; INMA = Infancia y Medio Ambiente
(Environment and Childhood); IQ = intelligence quotient; K-ARS = Korean ADHD Rating Scale; LOD = limit of detection; mo = month(s); NPR = Norwegian Patient Registry; NR = not
reported; OFT = open-field test; OD = oppositional defiant; Pb = lead; SDQ = Strengths and Difficulties Questionnaire; SES = socioeconomic status; SGA = small for gestational age;
TBD = to be determined; TMT A = Trail Making Test: attention; XRF = X-ray fluorescence; yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a

change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

bEffect estimate is unstandardized due to insufficient blood lead distribution information or insufficient information regarding log transformation.

°Result did not report confidence interval nor p-value

tStudies published since the 2013 Integrated Science Assessment for Lead.

3-383

-------
Table 3-9E Epidemiologic studies of Pb exposure and externalizing behaviors including conduct disorders,
aggression, and criminal behavior in children and adolescents

Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tTatsuta et al.
(2012)

Tohoku district
Japan

Study years NR
Followed through
30 mo

TSCD birth cohort
n: 306

Mother/child pairs in
Japan

Blood

Cord blood, HR-ICP-MS

Age at measurement:
delivery

Median = 1.0 |jg/dL
95th: 1.7 Mg/dL

Externalizing
behavior composite
(oppositional,
aggressive)

CBCL

Age at Outcome:
2.5 yr

Child age, birth weight, sex,
maternal age at pregnancy, delivery
type, birth order, drinking/smoking
habits in pregnancy, duration of
breastfeeding, maternal IQ,
Evaluation of Environmental
Stimulation score

Externalizing behavior
beta = -0.032b (not
significant)

Cohort

tSioen et al. (2013) Flemish Health and Blood

Flanders
Belgium

Oct. 2002-Dec.
2003 (enrollment)

Followed through
June 2011

Cohort

Environment Study
(FLEHS 1)
n: 270

Cord blood, HR-ICP-MS

Age at measurement:
delivery

Birth cohort of
Flemish children living
in either rural or urban Median = 14 3 Mg/L

areas

75th: 25.3 pg/L

Conduct problems

SDQ with 5
domains:

emotional, conduct,
hyperactivity, peer
and social
problems

Age at outcome:
7 - 8 yr

Maternal and paternal BMI, maternal
age, weight increase of mother
during pregnancy, smoking during
pregnancy, smoking behavior of
maternal grandmother before birth of
mother, parental education, current
parental smoking, child sex, serious
infections of child since birth (also
tested interaction by sex)

OR per doubling of log-
transformed Pb:

Conduct problems:
1.182 (0.319, 4.385)c

3-384

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tLiu etal. (2014b)

Jintan, Jiangsu

province

China

Sep. 1, 2004-Apr.
30, 2005 (age 3-5
yr)

Followed through
age 6 yr

Cohort

China Jintan Child
Cohort Study
n: 1025 children

Chinese preschool
children

Blood

Child venous blood; GFAAS
Age at measurement:

3-5 yr

Mean (SD): 6.4 (2.6) pg/dL
median = 6.0 pg/dL
75th: 7.5 pg/dL
90th: 9.4 pg/dL
Max: 32 pg/dL

Aggressive
behavior and
oppositional defiant
problems

CBCL (Chinese
version); Caregiver-
Teacher Report
Form; normalized T
scores

Age at outcome:
6 yr

Age at BLL test, sex, preschool
residence, father's educational level,
mother's educational level, father's
occupation, parents' marital status,
single child status, and child IQ

Parent:

Aggressive (3 (95% CI):
-0.018 (-0.264, 0.229)
Oppositional (3 (95%
CI): -0.03 (-0.28,

0.22)

Teacher:

Aggressive (3 (95% CI):
0.001 (-0.001, 0.003)

Oppositional (3 (95%
CI): 0.223 (-0.038,
0.484)

Aggressive OR (95%
CI): overall 1.07 (0.98,
1.17); boys 1.03 (0.93,

1.14);	girls 1.21 (0.99,
1.47)

Oppositional OR (95%
CI): overall 1.06 (0.98,

1.15);	boys 1.02 (0.92,
1.13); girls 1.12 (0.97,
1.29)

tNkomo et al.
(2017)

Soweto/Johannesb
urg

South Africa

Apr. 23-Jun. 8,
1990 (enrollment)
Followed 15-16 yr

Cohort

BT20+
n: 1322

684 females, 87.2%
Black African; 10.4%
mixed ancestry urban
residents; white and
Indian participants
excluded due to low
numbers

Blood

Child venous blood; GFAAS
with Zeeman background
correction

Age at measurement:

13 yr

Mean (SD) = 5.76 (2.42) pg/dL

median = 5.62 pg/dL
75th: 7.08 pg/dL
Max: 28 pg/dL

Violent behavior

YSR - violent
behavior

Age at outcome:
15-16 yr

Child sex, ethnicity, maternal
education, public/private hospital,
SES (unclear covariate adjustment
reporting)

physical violence (3
(95% CI): 0.05 (0.04,
0.05)

3-385

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tNkomo et al.
(2018)

Soweto/Johannesb
urg

South Africa

Apr. 23-Jun. 8,
1990 (enrollment)
Followed 14-15 yr

Cohort

BT20+
n: 1086

Black African and
mixed ancestry urban
residents; white and
Indian participants
excluded due to low
numbers

Blood

Child venous blood; GFAAS
with Zeeman background
correction

Age at measurement:

13 yr

mean (SD) = 5.6 (2.3) pg/dL
GM = 5.1 pg/dL
median = 5.4 pg/dL

Aggressive
behavior

YSR

Age at outcome:
14-15 yr

Child sex, maternal age, maternal
education at birth, marital status,
public/private hospital, SES

Direct aggression
(BLLs >10 pg/dL vs. <5
pg/dL): 0.43 (0.08,
0.78)d

tBoucher et al.
(2012b)

Nunavik, Arctic

Quebec

Canada

1992-2000
(enrollment)
2005-2010 (follow-
up)

Cohort

Cord Blood
Monitoring Program
and Environmental
Contaminants and
Child Development
Study
n: 279

Inuit Children

Blood

Cord blood; AAS

Child venous blood; ICP-MS

Age at measurement:
Avg: delivery (cord) 11.3 yr
(child)

Mean: 4.7 (cord); 2.7 (child)
Max: 20.9 (cord); 12.8 (child)

Externalizing
behavior and
OD/CD problems

CBCL andDBD
rating scale

Age at outcome:
Avg: 11.3 yr old

Child age and sex, SES, age of the
biological mother at birth, maternal
tobacco use during pregnancy, and
birth weight, Hg

Externalizing behavior

Cord: 0.09 (-0.05,
0.23 )c

Child: 0.14 (0.01,
0.26)c

OR

OD/CD

2nd vs. 1st fertile: 1.90
(0.88, 4.11 )c
3rd vs. 1st fertile: 1.53
(0.67, 3.49)c

3-386

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tBeckwith et al.
(2018)

Cincinnati, OH
U.S.

1979-1984
(enrollment)
Followed through
19-24 yr

Cohort

CLS
n: 250

Young adults from
birth cohort

Recruited pregnant
women in 1st or 2nd
trimester from inner
city neighborhoods
with historically
elevated incidence of
childhood lead
poisoning

Blood

Child venous blood; ASV
(Roda et al.. 1988))
Age at measurement:
78 mo

Mean: 7.99 |jg/dL
Max: 24.75 pg/dL

PPI score and gray
and white matter
see volume in cingulate
and ventromedial
prefrontal cortex

PPI; high resolution
anatomical MRI
with Voxel Based
Morphometry to
calculate brain
volume changes

Age at outcome:
19-24 yr old

Sex, race, age at time of imaging,
gestational age at birth, weight at
birth, maternal IQ, participant IQ,
HOME score, adult marijuana
usage, maternal prenatal alcohol
use, maternal prenatal cigarette use,
maternal narcotic use, and maternal
prenatal marijuana use

Beta
PPI

Overall: 0.22 (0.06,
0.38)e

Female: 0.16 (-0.05,
0.37)e

Male: 0.22 (-0.02,
0.47)e

tDesrochers-
Couture et al.

(2019)

Nunavik, Northern

Quebec

Canada

Nov. 1993-Mar.
2002 (enrollment)
Sep. 2005-Feb.
2010 (1st follow-up)
Jan. 2013-Feb.
2016 (2nd follow-
up)

Cohort

NCDS-childhood
n: 212

Inuit children from 14
coastal villages in
Nunavik, Quebec,
subsample from the
Cord Blood
Monitoring Program
and NIH-infancy study

Blood

Cord and venous child blood;
GFAAS (cord), ICP-MS (child)

Age at measurement:

Cord: delivery; Avg child: 11.4
and 18.5 yr

GM (GSD): 3.80 (1.84) pg/dL
(cord); 2.34 (1.86) pg/dL
(child); 1.63 (2.00) pg/dL
(adolescent)

Median:3.73 pg/dL (cord); 2.07
pg/dL (child); 1.52 pg/dL
(adolescent)

Max: 17.80 pg/dL (cord); 12.83
pg/dL (child); 18.13 pg/dL
(adolescent)

Externalizing
behavior; behavior
problems;
substance use

Behaviors -

externalizing

(CBCL),

hyperactivity-

impulsivity (DBD,

BAARS),

oppositional

defiant/conduct

disorder (DBD,

DISC); substance

use

Age at outcome:
childhood (11 yr);
adolescence (18 yr)

Child age, sex, SES, age of
biological mother at delivery,
maternal tobacco smoking during
pregnancy, birth weight, blood Hg,
house crowding, education of
primary caregiver

Beta (95% CI):

Child Blood:

Child externalizing
behavior:

0.23 (0.08, 0.38)

Direct effect on
adolescent
externalizing: 0.34
(-0.38, 1.06)

Indirect effect on
adolescent
externalizing: 0.18 (0,
0.36)

Child OD/CD: 0.37
(0.06, 0.69)

Direct effect on
adolescent CD: 0.01
(-0.10, 0.13)

Indirect effect on
adolescent CD (0.01,
-0.01, 0.03)

3-387

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tTlotlena et al.

(2022)

Johannesburg
South Africa

April-June 1990
(birth), sub-cohort
established at age
9 yr, followed
through 23-24 yr

Cohort

Young adults, n = 100 Bone

Sub-cohort (Bone
Health Cohort) of
singleton children
(born April-June
1990) from BT20
Cohort enrolling
women in 2nd and
3rd trimester residing
in Soweto-
Johannesburg

Child tibia; K-XRF

Age at measurement: NR

Mean (SD), min, med (IQR),
max: 8.7 (5.3), 0, 9 (5-12.5),

21 pg/g

Males (n = 53): 8.1 (4.4), 0, 8
(5-11), 18 pg/g

Females (n = 47): 9.4 (6.1), 0,
10 (4-14), 21 pg/g

Aggression scores
(anger, physical,
verbal, hostility)

BPAQ

Age at outcome:
23-24 yr

Age, sex, exposure to family
violence, attitude toward
neighborhood, exposure to crime
and violence in the neighborhood

Level of schooling, alcohol and drug
abuse, presence of both parents at
home, home environment, and SES
(maternal education, housing type,
participant's education/occupation)
also considered.

Beta per 1 pg/g
increase in bone Pb

Anger aggression: 0.25
(0.04, 0.37)

Physical aggression:
0.093 (-0.01, 0.27)
Verbal aggression:
0.093 (-0.05, 0.23)
Hostility: 0.03 (-0.19,
0.26)

tReuben et al.
(2019)

Dunedin
New Zealand

Apr. 1, 1972-Mar.
31, 1973
(enrollment)

Followed through
Dec. 2012

Cohort

Dunedin
Multidisciplinary
Health and
Development Study

N: 579

Birth cohort of
nationally
representative
(majority white)
children with high
rates of participation
and follow-up

Blood

Child venous blood; GFAAS

Age at measurement: 11 yr

Mean: 11.08 pg/dL
(94% above 5 pg/dL)

Antisocial behavior Sex, childhood SES, maternal IQ,
in children	and family history of mental illness.

Rutter Child Scale
(averaged
parent/teacher
ratings)

Age at outcome: 11
yr

Beta

Antisocial behavior:
0.02 (0.00, 0.04)

3-388

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

Chandramouli et al.
(2009)

Avon
U.K.

Jul. - Dec. 1992
(enrollment)

Followed through 8
yr

Cohort

10% random
subsample of Avon
Longitudinal Study of
Parents and Children
(ALSPAC)

n = 488
Birth cohort

Blood

Child venous blood; AAS with
micro sampling flame
atomization

Age at measurement: 30 mo

Mean (SD): NR
Group 1: 0-<2 |jg/dL
Group 2: 2-<5 |jg/dL
Group 3: 5-<10 |jg/dL
Group 4: >10 pg/dl_

Antisocial activities

Parent/teacher
ratings on

Antisocial Behavior
Interview

Age at outcome:
8 yr

Maternal education and smoking,
home ownership, home facilities
score, family adversity index,
paternal SES, parenting attitudes at
6 mo, child sex. Also considered
child IQ.

ORs for increased
score

Group 1 (0-<2 |jg/dL):
ref

Group 2 (2-<5 |jg/dL):
0.93 (0.47, 1.83)
Group 1 (5—< 10
|jg/dL): 1.44 (0.73,
2.84)

Group 1 (>10 |jg/dL):
2.90 (1.05, 8.03)

Wright et al. (2008) CLS

Cincinnati, OH
United States

1979-1984
(enrollment)
Followed through
19-24 yr

Cohort

n: 250

Young adults from
birth cohort

Recruited pregnant
women in 1st or 2nd
trimester from inner
city neighborhoods
with historically
elevated incidence of
childhood lead
poisoning

Blood

Child blood; ASV

Age at measurement: 6 yr

Median (5th-95th):
6 yr: 6.8(3.4-18) pg/dL
0-6 yr avg: 12 (6.0-26) pg/dL

Criminal arrests

County records

Age at outcome:
19-24 yr

Maternal IQ and education, sex,

Also considered potential
confounding by maternal prenatal
smoking, marijuana use, narcotic
use, and prior arrests, HOME score,
birth weight, # children in the home,
public assistance in childhood.

RRs (yes/no)

Age 6 blood Pb: 1.05
(1.01, 1.09)

Age 0-6 yr avg blood
Pb:

1.01 (0.98, 1.05)

3-389

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tFruh et al. (2019) Project Viva
n: 1006

Eastern
Massachusetts
U.S.

Birth cohort of
mother-child pairs

1999-2002
(enrollment)
Followed through
age 7 yr

Cohort

Blood

Maternal venous blood; ICP-
MS

Age at measurement:

T2

Median: 1.1 pg/dL

Parent teacher
ratings of conduct
problems using
SDQ

Standardized for
child age and sex

Age at outcome:
7 yr old

Maternal 2nd trimester Hg and Mn
levels, nulliparity, smoking during
pregnancy, IQ, and education;
paternal education; HOME
composite score and household
income; and child race/ethnicity

Parent ratings:

Overall (3 (95% CI):
0.10 (-0.10, 0.30)
Boys: 0.07 (-0.18,
0.32)

Girls: 0.13 (-0.13,
0.40)

Teacher ratings:

Overall (3 (95% CI):
0.18 (-0.08, 0.44)
Boys: 0.18 (-0.17,
0.53)

Girls: 0.17 (-0.13,
0.46)

tRuebner et al.
(2019)

46 centers
U.S.

Study Years: NR
Follow-up: NR

Cohort

CKiD Cohort study
n: 412

Children ages 1-16 yr
at recruitment with
mild to moderate CKD

Blood

Child venous blood; ICP-MS.
The BLL measurement closest
to the time of neurocognitive
testing was used for analysis
(concurrent).

Age at measurement:
NR; 2, 4, or 6 yr after study
entry

Median: 1.2 |jg/dL
75th: 1.8 |jg/dL
Max: 5.1 |jg/dL

Externalizing
behaviors,
composite index on
the BASC-2 (see
also 3.5.1 and
3.5.2)

The last available
test results were
used to evaluate
long-term effects.
Mean time between
BLL and
neurocognitive
testing was 2.3 yr.
Age at outcome:
1-16 yr

Age, sex, race, poverty, and
maternal education

Adjusted BASC-2
results were not
reported because they
were not statistically
significant.

3-390

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tNaicker et al.
(2012)

Johannesburg
South Africa

Apr. - Jun. 1990
(enrollment)
Followed through
20 yr

Birth to Twenty cohort Blood
(Bt20)

n: 1041 (487 boys,

554 girls)

Singleton children
representative of
South Africa
population

Child venous blood; GFAAS
Age at measurement:

13 yr

median = 5.4 |jg/dL; GM = 5.2
|jg/dL

Max: 28.1 pg/dL

Rule-breaking
behavior,
aggressive
behavior

YSR (adapted from
CBCL for use in
adolescents)

Age at outcome:
13 yr

SES, maternal education,
demographic factors

Attacking people -
boys (3 (95% CI): 0.54
(0.09, 0.98)d

Cohort

tRodriques et al.
(2018)

Simoes Filho,
Salvador, Bahia
Brazil

Study years NR
Cross-sectional

Simoes Filho, Brazil
n: 225

Children aged 7-12
yr, attending public in
town near ferro-Mn
alloy plant

Blood

Child venous blood; GFAAS
Age at measurement:
7-12 yr

median = 1.2 |jg/dL
Max: 15.6 |jg/dL

Behavioral
problems/disruptive
behavior
(externalizing
behavior,
aggressive
behavior, rule-
breaking behavior)

CBCL

Age at outcome:
7-12 yr

Gender, age
violence score

community
maternal IQ

Attacking people -
Adjusted total T-score
(3 (95% CI): -0.74
(-5.3, 3.8)d

3-391

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tBara et al. (2018) Montevideo sample
n: 206

Montevideo

Uruguay	Children in urban

area

Study years NR
Cross-sectional

Blood

Child venous blood (fasting);
AAS with flame or graphite
furnace ionization
Age at measurement:
6-8 yr (mean = 6.75 yr)

mean = 4.2 |jg/dL

Behavior problems Child IQ, iron status, BMI, household BRIEF:

(e.g., oppositional)

CRS-R; BRIEF

Age at outcome:
6-8 yr (mean =
6.75 yr)

possessions, maternal education,
current parent smoking (also looked
at sex and Pb evaluation method in
sensitivity analyses)

Behavioral Regulation
Index (PR [95% CI]):
overall = 1.01 (1.00,
1.03); girls = 1.03
(1.00, 1.05); boys =
0.99 (0.97, 1.01)

CTRS-R:

Oppositional (PR [95%
CI]): overall = 1.00
(0.98, 1.02); girls =
1.01 (0.99, 1.04); boys
= 0.99 (0.96, 1.02)

tLiu et al. (2022b) Healthy Brains and
Behavior

Philadelphia
County, PA;
Suburbs of
Philadelphia, PA
United States

Study years NR
Cross-sectional

n: 131

Blood

Child blood; HR-ICP-MS.
Age at Measurement:
11-12 yr

Mean = 2.2 |jg/dL; Median
1.10 |jg/dL
75th: 1.8 |jg/dL
Max: 35.4 |jg/dL

Parent-report and
child-report of
externalizing
behavior
(composite)

Scores derived
from factor
analyses of 14
validated measures
of antisocial/
aggressive
behavior (RPQ,
CBCL, YSR, APSD,
CODDS, AQ from
BPAQ)

OLS regression adjusted for sex and
race.

Beta for externalizing
behavior:

Parent-reported: 0.20
(0.05, 0.34)
Child-reported: 0.20
(0.04, 0.35)

Age at outcome:
11-12 yr

3-392

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tNiqq etal. (2010) n = 326

Study location and Recruitment by

year NR

Case-control

community
advertisements,
mailings, outreach to
clinics

Blood

Child venous blood; ICP-MS
Age at Measurement: 6-17 yr

Externalizing
composite score
(oppositional and
conduct symptoms)
on parent K-SADS

Mean (SE) = 0.73 (0.04) |jg/dL Oppositional

behavior on parent
and teacher CRS

Age at Outcome:
6-17 yr

Household income, maternal
smoking, child age, sex, blood
hemoglobin, child FSIQ (WISC-IV)

Beta for SD increase in
scores per SD
increase in log-10 Pb

Parent ratings:

K-SADS externalizing
composite: 0.21 (0.05,
0.37)

CRS oppositional
behavior: 0.09 (-0.09,
0.27)

Teacher ratings:
CRS oppositional
behavior: 0.11 (-0.01,
0.23)

tAmato et al.
(2013)

Milwaukee, Wl
United States

Study years NR

Followed 7-10 yr
(blood Pb before
age 3, outcome
assessment at 4th
grade)

Wisconsin Childhood
Pb Poisoning
Prevention Project /
Milwaukee Public
School

n: 1076 unexposed;
2687 exposed; 3763
total

Exposed individuals
were more likely to be
Black or Hispanic,
and be on assisted
lunch programs

Blood

Maximum child blood Pb,
methods varied by providers
Age at measurement:

<3 yr

Mean NR; reported exposed
(BLL 10-20 |jg/dL) vs.
unexposed (<5 |jg/dL)
Max: 20 |jg/dL

School
suspensions

Unduplicated
suspension count

Age at outcome:
10 yr (4th grade)

Gender, race/ethnicity, income
(free/reduced lunch)

Suspensions
OR for exposed
(10-20 |jg/dL) vs.
unexposed (<5 |jg/dL):
2.66 (2.12, 3.32)

Cohort

3-393

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tBoutwell et al.
(2017)

106 census tracts in Blood
St. Louis City, MO
n: 59,645 children; NR

St. Louis City, MO 15,734 violent crimes Age at measurement:

United States

Study years: NR
16-year period

Other - ecological
study

St Louis residents

<72 mo age

NR

Violent crime
(crimes with
firearm, assault
crimes, robbery
crimes, homicides,
rape)

Police

department/uniform
crime report -
violent crime
(crimes with
firearm, assault
crimes, robbery
crimes, homicides,
rape)

Concentrated disadvantage; mean
age of housing; proportion occupied
by renters; domestic assaults

RR for 1% increase in
proportion of elevated
blood tests in the
census tract.

Firearm crimes: 1.03
(1.025, 1.035)

Assault: 1.03 (1.025,
1.035)

Robbery: 1.03 (1.02,
1.04)

Homicides: 1.03
(1.015, 1.045)

Rape: 1.01 (0.99, 1.03)

3-394

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tBecklev et al.
(2018)

Apr. 1, 1972-Mar.
31, 1973
(enrollment)
Followed through
38 yr

Cohort

Dunedin
Multidisciplinary
Health and
Development Study
N: 553

Birth cohort of
nationally
representative
(majority white)
children with high
rates of participation
and follow-up

Blood

Child venous blood; GFAAS
Age at measurement:

11 yr

mean = 11.01 pg/dL
Max: 31 pg/dL

Criminal offending
(criminal conviction,
recidivism,
conviction for
violent offense,
self-reported
criminal offending)

Official conviction
records from
central police
computer; self-
reported offending
interview

Age at outcome:
38 yr

Sex, age

OR (ref: no conviction)

Any criminal
conviction: 1.042 (1,
1.086)

One-time: 1.046 (0.99,
1.104)

Recidivistic: 1.039
(0.986, 1.095)
Nonviolent: 1.051
(1.003, 1.101)

Violent offense: 1.025
(0.962, 1.092)

Beta (self-report
offending)

15 yr: 0.1 (0.015,
0.185)

18 yr: 0.06 (-0.02,
0.14)

21 yr: 0.01 (-0.065,
0.085)

26 yr: 0.06 (-0.015,
0.135)

32 yr: 0.04 (-0.04,
0.12)

38 yr: 0.02 (-0.06, 0.1)

3-395

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tWriaht et al.
(2021)

Cincinnati, OH
United States

1979-1984
(enrollment)
Followed through
2013

Cohort

CLS
n: 254

Young adults from
birth cohort

Recruited pregnant
women in 1st or 2nd
trimester from inner
city neighborhoods
with historically
elevated incidence of
childhood lead
poisoning

Blood

Maternal blood (n = 219) and

child blood; ASV

Age at measurement: prenatal,

<60 mo (avg child), 60-78 mo

(avg late child), 78 mo (late

child)

Mean (SD):

Average child (0-60 mo): 14.4
(6.6) pg/dL

Total number of
arrests for each
subject (2003-2013
and lifetime),
violent crimes, drug
crimes, and
property crimes

Hamilton County
public records

Age at outcome:
18-24 yr, 27-33 yr

Birth weight (grams), maternal age
at delivery, Appearance, Pulse,
Grimace, Activity, and Respiration
scores taken at 1 min, self-reported
maternal drug use during pregnancy
that includes reports of alcohol,
marijuana and tobacco use,
maternal IQ measured by the WAIS-
R, and HOME Inventory scores
across the first 3 yr

IRR

Arrests 2003-2013 for
6-year blood Pb,
controlling for prior
arrests 1998-2003:
1.008 (0.995, 1.021)

6 yr blood Pb
Lifetime Arrests: 1.016
(1.002, 1.03)

Property Arrests: 1
(0.977, 1.023)

Drug Arrests: 1.032
(1.005, 1.06)

Violent Arrests: 1.016
(0.992, 1.039)

Adult Arrests: 1.014 (1,
1.027)

EEs for other blood Pb
sources are available
but not listed for the
sake of space.

3-396

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% Clsa

tEmer et al. (2020)

Milwaukee, Wl

Born June 1 1986-
Dec 31, 2005

Outcome assessed
January 1, 2005-
December 31 2015

Cohort

Adolescents enrolled
in Milwaukee public
school (2004-2016)
with BLL before age 6

N = 82,612

Capillary or venous BLLs
measured by the Milwaukee
Health department

Male:

Mean (IQR) 5.6 (5.8)

Peak (IQR): 7.0 (7.0)

Female:

Mean (IQR): 5.3 (5.0)

Peak (IQR): 6.0 (8.0)

Firearm violence
perpetration (i.e.,
coded as arrestee,
suspect or person
of interest by
Milwaukee police
department; victim
of firearm violence
assessed using
police records

Sex, race; socioeconomic status,
and year of birth

Perpetration:

OR (mean Pb): 1.03
(1.02, 1.04)

OR (peak Pb): 1.02
(1.01, 1.02)

Victimization

OR (mean Pb): 1.04
(1.03, 1.05)
OR (peak Pb): 1.02
(1.01, 1.03)

Age: < 6 yr

BASC-2 = Behavior Assessment System for Children; BLL = blood lead level; BMI = body mass index; BRIEF = Behavior Rating Inventory of Executive Functions; BT20+ = Birth to
Twenty Plus; CBCL = Child Behavior Check List; CI = confidence interval; CKD = chronic kidney disease; CKiD = Chronic Kidney Disease in Children Study; CLS = Cincinnati Lead
Study; CTRS-R = Conners' Teacher Rating Scale-Revised; DBD = Disruptive Behavior Disorder; DISC = Disrupted-in-Schizophrenia; EES = Evaluation of Environmental Stimulation;
FLEHS = Flemish Environment and Health Study; GFAAS = graphite furnace atomic absorption spectrometry; Hg = mercury; HOME = Health Outcomes and Measures of the
Environment; ICP-MS = inductively coupled plasma mass spectrometry; IQ = intelligence quotient; IQR = interquartile range; K-SADS = Kiddie Schedule for Affective Disorders and
Schizophrenia; Mn = manganese; mo = month(s); NCDS = Nunavik Child Development Study; NR = not reported; Pb = lead; PPI = Psychopathic Personality Inventory; RR = relative
risk; SDQ = Strengths and Difficulties Questionnaire; SE = standard error; SES = socioeconomic status; T2 = second trimester of pregnancy; TSCD = Tohoku Study of Child
Development; WAIS-R = Weschler Adult Intelligence Scale-Revised; yr = year(s); YSR = Youth Self-Report.

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a

change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

bResults are unstandardized because they did not have an associated SE, CI, or p-value reported in the study.

°Results are unstandardized because the log base used for exposure transformation was unspecified in the study.

dResults are unstandardized because the Pb level distribution data was not available.

eThe CI was calculated from a p-value and the true CI may be wider or narrower than calculated.

tStudies published since the 2013 Integrated Science Assessment for Lead.

3-397

-------
Table 3-1OE Epidemiologic studies of Pb exposure and internalizing behaviors in children

Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

Wasserman et al. (2001)

Pristina
Yugoslavia

1984-1985 (enrollment)
Followed through 1999

Cohort

N: 191

Blood

Recruitment Child blood; method NR
from prenatal

clinics

Age at outcome:

Delivery to 4-5 yr

Lifetime (to age 4-5 yr) avg
blood

Mean (SD) of log—10 Pb:
0.86 (0.12) |jg/dL, Mean:
-7.2 |jg/dL

Internalizing behavior scores
and subscores (i.e.,
anxious/depressed, somatic
complaints, and withdrawn)
assessed using maternal
ratings of CBCL

Age at Outcome: 4-5 yr

Sex, ethnicity, age, maternal
education and smoking
history, HOME score, birth
weight

Betas for log—10
change in outcome:

Internalizing
composite: 0.152
(0.023, 0.281)
Anxious/depressed:
0.041 (-0.089, 0.17)

Somatic complaints:
0.107 (-0.062,
0.276)

Withdrawn: 0.066
(-0.073, 0.205)

3-398

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

Burns et al. (1999)

Port Pirie
Australia

May 1979-May 1982
(enrollment)

Followed to age 11-13 yr
Cohort

Port Pirie
Cohort Study
(PPCS)

N: 322

Recruited
90% of live
births in a
lead smelting
community

Blood

Lifetime avg (to age 11-13
yr) blood

GM (95% CI) |jg/dL

Boys: 14.3 (13.5,
Girls: 13.9 (13.2,

15.1)
14.6)

Internalizing behavior scores
and subscores (i.e.
anxious/depressed, somatic
complaints, and withdrawn)
assessed using maternal
ratings of CBCL

Age at Outcome: 11—13 yr

Maternal age, prenatal
smoking status, IQ,
concurrent psychopathology,
and education, birth weight,
type of feeding, length of
breastfeeding, paternal
education and occupation,
birth order, family functioning,
parental smoking, marital
status, HOME score, child IQ

Beta
Male:

Internalizing
composite: 0.8 (-0.9,
2.4)b

Anxious/depressed:
0.8 (-0.2, 1.8)b

Somatic complaints:
-0.1 (-0.7, 0.4)b

Withdrawn: 0.1
(-0.4, 0.7)b

Female:

Internalizing
composite: 2.1 (0.0,
4.2)b

Anxious/depressed:
1.3 (0.1, 2.5)b
Somatic complaints:
0.3 (-0.4, 0.9)b
Withdrawn: 0.6 (0.0,
1.1)b

3-399

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

Bellinger et al. (1994b)

Boston, MA
US

1979-1980 (birth) followed
to age 8 yr

Cohort

N: 1782	Blood, Tooth

Recruitment Cord blood and shed
at birth	deciduous teeth. Dentin

hospital	taken from zone representing

cumulative postnatal
deposition; ASV

Age at Measurement: 6 yr

Tooth mean (SD): 3.4 (2.4)
pg/dL

Range: 0.1-28.9
10th—90th percentiles:
1.2-6.3

95th percentile: 7.4

Internalizing behavior T
scores and subscores (i.e.
anxious/depressed, somatic
complaints, and withdrawn)
assessed using teacher
ratings of CBCL

Age at Outcome: 8 yr old

Prepregnant weight, race,
Cesarean section, maternal
marital status, prenatal care,
paternal education, colic, child
current medication use,
sibship size, sex, birth weight.
Also considered potential
confounding by public
assistance, prenatal smoking,
maternal education but not
parental caregiving quality.

Betas for In-
transformed change
in internalizing T
score:

Cord blood: -0.07
(-0.23, 0.10)

Tooth: 0.43 (0.09,
0.77)

Cord blood mean (SD): 6.8
(3.1) pg/dL

Interval analyzed: 0.1-35.1
95th percentile: 12.2

tWinter and Sampson
(2017)

Chicago, Illinois
U.S.

born 1995-1997 to 2013,
followed through 17 yrold
Cohort

PHDCN
n: 254

Children and
caregivers
living in
Chicago

Blood

Child venous and capillary
blood; methods NR
Age at measurement:
before 6 yr

Avg BLL before 6 yr
Mean: 6.4 pg/dL

Internalizing on the CBCL
(i.e., anxiety and
depression); see also
Section 3.5.2 (impulsivity)
PC questionnaire)

Age at outcome:

Mean: 17 yr old

Age, sex, race/ethnicity; PC's
immigrant generational status,
marital status, education,
Temporary Assistance for
Needy Families receipt;
proportion residential
neighborhood that is non-
Hispanic Black, Hispanic, and
below the poverty line;
proportion of the child's
residential neighborhood
tested for Pb exposure

Beta

Anxiety/depression:
0.09 (0.03, 0.16)

3-400

-------
tLiu etal. (2014b)

Jintan, Jiangsu province
China

Sep. 1, 2004-Apr. 30,
2005 (age 3-5 yr)
Followed to age 6 yr

Cohort

China Jintan
Child Cohort
Study
n: 1025
children

Chinese

preschool

children

Blood

Child venous blood; GFAAS

Age at measurement:
3-5 yr old

Mean (SD): 6.4 (2.6) pg/dL
median = 6.0 pg/dL
75th: 7.5 pg/dL
90th: 9.4 pg/dL
Max: 32 pg/dL

Internalizing problems
composite and subscores
(emotionally reactive,
anxious/depressed, somatic
complaints, withdrawn, and
sleep)

CBCL (Chinese version);
Caregiver-Teacher Report
Form; normalized T scores

Age at outcome:

6 yr

Age at BLL test, sex,
preschool residence, father's
educational level, mother's
educational level, father's
occupation, parents' marital
status, single child status, and
child IQ

Internalizing
problems

Parent beta: -0.029
(-0.280, 0.222)
Teacher beta: 0.223
(-0.037, 0.484)

Teacher OR: 1.10
(1.03, 1.18)

Emotionally Reactive

Parent beta: -0.117
(-0.365, 0.131)

Teacher beta: 0.322
(0.058, 0.587)
Teacher OR: 1.10
(1.02, 1.19)

Anxiety/Depression:
Parent beta: 0.101
(-0.151, 0.354)
Teacher beta: 0.001
(-0.001, 0.003)

Teacher OR: 1.12
(1.03, 1.23)

Somatic Complaints

Parent beta: -0.171
(-0.436, 0.094)

Teacher beta: 0.001
(-0.003, 0.001)
Teacher OR: 1.01
(0.90, 1.13)

Withdrawn
Parent beta: 0.096
(-0.158, 0.349)

3-401

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

Teacher beta: 0.001
(-0.001, 0.003)
Teacher OR: 1.02
(0.93, 1.12)

tJoo etal. (2018)

Seoul, Ulsan, Cheonan
South Korea

2006-2011 (enrollment)
Followed through 5 yr

Cohort

Clinically significant
anxiety

Parent beta: 0.044
(-0.212, 0.299)
Teacher beta: 0.253
(0.016, 0.500)

Teacher OR: 1.12
(1.03, 1.23)

MOCEH
n: 575
mother-child
pairs

pregnant
women

Blood

Maternal venous blood, cord
blood, and child blood; AAS

Age at measurement: 20 wk
gestation (maternal); delivery
(cord); 2,3 and 5 yr (child)

GM:

Maternal 1.28 |jg/dL (early),
1.24 (late) 0.9 (cord)

Child 1.55 (age 2), 1.43 (age
3), 1.29 (age 5)

Internalizing behavior

K-CBCL (emotional
reactivity, anxious/
depressed, somatic
complaints, and
withdrawn/depressed states)

See also Section 3.5.2

Age at outcome:

5 yr old

Maternal age at childbirth,
parity, maternal educational
level, household income,
residential area, and
breastfeeding

Beta (95% CI):
Internalizing at 5 yr

Maternal-early
pregnancy
Male: -0.16 (-2.54,
2.23)

Female: -0.13
(-1.86, 1.60)
Maternal-late
pregnancy

Male: 2.55 (0.22,
4.88)

Female: -0.18
(-2.66, 2.31)

Cord blood
Male: 2.44 (-0.74,
5.63)

Female: -1.00
(-4.30, 2.29)

3-402

-------
Re'erenDCifgn„d S,Udy p4SS8o„ Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

Child blood-2 yr
Male: -0.03 (-2.07,
2.00)

Female: 2.94 (0.36,
5.52)

Child blood-3 yr
Male: 0.25 (-2.33,
2.82)

Female: 2.76 (-0.73,
6.26)

Child blood-5 yr
(concurrent)

Male: 1.23 (-2.10,
4.56)

Female: 5.65 (0.50,
10.80)

tFruh etal. (2019)

Eastern Massachusetts
U.S.

1999-2002 (enrollment)
Followed through age 7 yr

Cohort

Project Viva Blood
n: 1006

Maternal venous blood; ICP-
Birth cohort of MS
mother-child Age at measurement:
pairs	T2

Median: 1.1 pg/dL

Parent teacher ratings of
emotional problems

SDQ

Standardized for child age
and sex

Age at outcome:

7 yr old

Maternal 2nd trimester Hg
and Mn levels, nulliparity,
smoking during pregnancy,
IQ, and education; paternal
education; HOME composite
score and household income;
and child race/ethnicity

Parent ratings:
Overall (3 (95% CI):
0.30 (0.05, 0.55)
Boys: 0.17 (-0.17,
0.50)

Girls: 0.52 (0.18,
0.86)

Teacher ratings:
Overall (3 (95% CI):
0.07 (-0.22, 0.35)
Boys: 0.02 (-0.33,
0.37)

Girls: 0.12 (-0.31,
0.54)

3-403

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

tSioen et al. (2013)

Flanders
Belgium

Oct. 2002 - Dec. 2003
(enrollment)

Followed through June 2011
Cohort

Flemish
Health and
Environment
Study
(FLEHS 1)
n: 270

Birth cohort of
Flemish
children living
in either rural
or urban
areas

Blood

Cord blood, HR-ICP-MS

Age at measurement:
delivery

median = 14.3 |jg/L
75th: 25.3 pg/L

Emotional problems

SDQ with 5 domains:
emotional, conduct,
hyperactivity, peer and social
problems

Age at outcome:

7 - 8 yr

Maternal BMI, age at
pregnancy, weight increase
during pregnancy, smoking,
paternal BMI, if parents
smoke, smoking behavior
maternal grandmother before
the birth of the mother,
parental education, child sex,
serious child infections

OR per doubling of
log-transformed Pb:

Emotional problems:
0.900 (0.524, 1.547)b

tRokoff et al. (2022)

New Bedford, MA
Born: 1993-1998

Cohort

Children
residing near
Superfund
site

n: 468 of 788
mother-infant
pairs.

Blood

Cord blood; ICP-MS

Child blood; medical records,
method NR

Age at measurement:
delivery

Mean (SD) cord BLL

CPRS: 1.37 pg/dL (0.94)

BASC-2: 1.37 pg/dL (0.95)

Mean (SD) peak postnatal
BLL

CPRS: 6.68 pg/dL (3.95)
BASC-2: 6.58 pg/dL(3.87)

Internalizing Behavior

Anxiety, Depression,
Somatization, and
Internalizing Problems on
BASC-2 SRP

Anxious-Shy and
Psychosomatic on CPRS

Anxious-Shy on CTRS

Age(s) at outcome: 8 yr
(CPRS and CTRS) and 15-
years (BASC-2)

Maternal age, marital status,
parity, parental education),
household income, maternal
smoking, alcohol consumption
during pregnancy, pre-
pregnancy weight, height, and
gestational weight gain, BMI,
prenatal social disadvantage
index, HOME score, maternal
IQ

No interactions between
chemicals; linear regression
models adjusted for Mn and
organochlorines,

BASC-2 SRP
Anxiety: 1.78 (0.58,
2.99)

Depression: 0.79
(-0.39, 1.97)

CPRS

Psychosomatic
Boys: 2.08 (0.07,
4.10)

Girls: 0.48 (-1.00,
1.97)

C-R functions
presented

3-404

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

tRasnick et al. (2021)

CCAAPS

Air

Internalizing and

Maternal education,

B (anxiety score) =



n: 263



Externalizing Behavior

community-level deprivation,

3.1 (95% CI: 0.4,

Cincinnati, OH

LURF, air sampling at 24



blood Pb concentrations,

5.7) per ng/m3





sites (C-V R2 = 0.89),

BASC-2; internalizing

greenspace, and traffic

Note: no association

Born: Oct2001-Jul 2003



predicted air concentration at

behaviors (anxiety,

related air pollution.

with depression,

Exposure: 2001-2005



child's residence.

depression, somatization),
externalizing behaviors



somatization,
conduct problems,

Cohort



Children residing >1,500 m or

(aggression, conduct



hyperactivity,



<400 m from major highway

problems, and hyperactivity),



withdrawal behaviors





eligible.

behavioral symptoms index









(attention problems,









Median: 0.51 ng/m3 (range 0-

atypicality, and withdrawal)









10.8 ng/m3)











Age at outcome: 12 yr





tRuebner et al. (2019)

CKiD Cohort

Blood

Internalizing behaviors,

Age, sex, race, poverty, and

Adjusted BASC-2



study



composite index on the

maternal education

results were not

46 centers

n: 412

Child venous blood; ICP-MS.

BASC-2 (see also 3.5.1 and



reported because

U.S.

Children with

The BLL measurement
closest to the time of

3.5.2)



they were not
statistically

Study Years: NR

mild to

neurocognitive testing was

The last available test results



significant.

Follow-up: 1-16 yr

moderate

used for analysis

were used to evaluate long-





CKD

(concurrent).

term effects. Mean time





Cohort





between BLL and







Age at measurement:
NR; 2, 4, or 6 yr after study
entry

Median: 1.2 |jg/dL
75th: 1.8 |jg/dL
Max: 5.1 |jg/dL

neurocognitive testing was
2.3 yr.

Age at outcome:

1-16 yr





3-405

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% Clsa

tHorton et al. (2018)

ELEMENT

Tooth

Internalizing behavior on the

Maternal age at delivery,



Project



BASC-2. See Section 3.5.2

maternal education, smoking

Mexico City

n: 133

Tooth Pb (prenatal, postnatal

(attention and hyperactivity)

SES, maternal IQ

Mexico



metrics derived); laser

and BSI



born 1994-2006 and

healthy, low

ablation ICP-MS





followed through age 6-16

to moderate

Age at measurement:

Age at outcome:



Cohort

income

tooth Pb concentration

8-11 yrold





mother (18-

corresponded to prenatal and







39 yr old)-

300 days after birth







child pairs











Figure 1c





Beta: BASC-2

Internalizing (10 mo)
NR

Anxiety (12 mo)
(95% CI NR)C

0.4

Internalizing
composite result was
not reported because
it was not statistically
significant.

tDohertv et al. (2020)

New Hampshire
U.S.

2009 to 2014-2019
Cohort

NHBCS
n: 371

(SRS-2); 318
(BASC-2)

Mother-child
pairs

Toenails

Maternal and infant toenails;
Median (maternal prenatal):
0.14 |jg/g (SRS-2), 0.13 pg/g
(BASC-2); Median (maternal
postnatal): 0.10 pg/g (SRS),
0.11 pg/g (BASC-2); Median
(infant): 0.35 pg/g (SRS-2),
0.37 pg/g

Internalizing Behaviors on
the BASC-2; see also
Section 3.5.2.2

Age at outcome:

3 yr

Maternal age, maternal BMI,
parental education, maternal
smoking, marital status,
parity, child age at last
breastfeeding, Healthy Eating
Index score, year of birth, sex,
and age of the child at testing

Exposure was log2
transformed.

Betas per 1 pg/g
increase in toenail
Pb concentration.

Total

Maternal prenatal:
-0.14 (-0.28, 0.00)d
Maternal postnatal:
0.06 (-0.05, 0.18)d
Child: 0.01 (-0.14,
0.16)d

Males

Maternal prenatal:
-0.17 (-0.36, 0.01 )d

Maternal postnatal:
0.31 (0.15, 0.47)d
Child: 0.09 (-0.13,
0.31 )d

Females

3-406

-------
Re,ereDCi?gn„d S,Udy Population Exposure Atsessmen.	Outeome

Maternal prenatal:
-0.16 (-0.33, 0.01 )d
Maternal postnatal:
-0.04 (-0.20, 0.13)d
Child: -0.15 (-0.36,
0.06)d

BASC = Behavioral Assessment System for Children; BLL = blood lead level; BMI = Body Mass Index; BRIEF = Behavior Rating Inventory of Executive Functions; CBCL = Child
Behavior Check List; CCAAPS = Cincinnati Childhood Allergy and Air Pollution Study; CI = confidence interval; CKD = chronic kidney disease; CKiD = Chronic Kidney Disease in
Children Study; CPRS = Conners' Parent Rating Scale; C-TRF = Caregiver-Teacher Report Form; CTRS = Conners' Teacher Rating Scale; C-V R2 = cross validated R-square; DSM
= Diagnostic and Statistical Manual of Mental Disorders; ELEMENT = Early Life Exposure in Mexico to Environmental Toxicants; FSIQ = full-scale intelligence quotient; GFAAS =
graphite furnace atomic absorption spectrometry; ICP-MS = inductively coupled plasma mass spectrometry; HOME = Health Outcomes and Measures of the Environment; IQ =
intelligence quotient; K-CBCL = Korean Child Behavior Check List; LURF = Land Use Random Forest; MOCEH = Mothers' and Children's Environmental Health; NHBCS = New
Hampshire Birth Cohort Study; NR = not reported; OR = odds ratio; Pb =lead; PC = primary caregiver; PHDCN = Project on Human Development in Chicago Neighborhoods; SDQ =
Strengths and Difficulties Questionnaire; SRP = Self-Report of Personality; SRS = Social Responsiveness Scale; yr = year(s); T2 = second trimester of pregnancy.
aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bResults are unstandardized because the log base used for exposure transformation was unspecified in the study.

°Results are unstandardized because they did not have an associated SE, CI, or p-value reported in the study.
dResults are unstandardized because the biomarker used for Pb exposure measurement is toenails.
tStudies published since the 2013 Integrated Science Assessment for Lead.

~	f	.	Effect Estimates

Confounders	gnd 95% C|j.a

3-407

-------
Table 3-11E Epidemiologic studies of Pb exposure and motor function in children

Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

Ris et al.
(2004)

Cincinnati, OH
US

1979-1985
(enrollment)

Cohort

Cincinnati Lead Blood
Study (CLS)

N: 195

Visuoconstructio Maternal IQ, SES,

Birth cohort
recruited
prenatally from
obstetrical
clinics

Prenatal maternal blood: NR

Average Childhood blood (mean of 20 quarterly
concentrations obtained from 3-60 mo): NR

78 mo blood lead: NR

n (Block Design
Subtest, ROCF-
Accuracy) and
Fine-Motor
(Grooved
Pegboard Test,
Finger Tapping
Test) factors

Age at outcome:
15-17 yr

total average HOME
scores,

and adolescent

marijuana

consumption

Beta

Visuoconst ruction

Prenatal: -0.157 (-0.277,
-0.037)b

Average: 0.028 (-0.052, 0.108)b
78-month: 0.014 (-0.088, 0.116)b

Fine-motor

Prenatal: -0.017 (-0.056, 0.022)b
Average: -0.016 (-0.041, 0.009)b
78-month: -0.046 (-0.077,
—0.015)b

Bhattacharva
et al. (1995)

Cincinnati, OH
US

1979-1984
(enrollment)

Cohort

Cincinnati Lead Blood
Program Project

N: 202

Pregnant
mothers living in
older houses in
poor condition
and with
chipping lead-
based paint and
lead laden dust

GM (SD) (min-max) ug/dL:

Prenatal maternal blood: 8.0 (1.58) (2-22)

Average Childhood blood (mean of 20 quarterly
concentrations obtained from birth to 5 yr): 11.9
(1.5) (4-28)

Postural
balance,
including sway
area (SA) and
sway length (SL)

Age at outcome:
5 yr

Age, height, weight,
birth length, birth
weight, Hgb, TIBC,
Minimum middle-ear
pressure, smoking
during pregnancy,
HOME score at 36
mo, foot area, sports
participation, race,
known occurrences of
bilateral otitis media

Betas
Eyes open

SA: 0.059 (0.024, 0.093)
SL: 0.145 (0.088, 0.201)
Eyes closed
SA: 0.043 (0.01, 0.076)
SL: 0.121 (0.069, 0.173)
Eyes open, foam
SA: 0.046 (-0.175, 0.266)
SL: 0.113 (0.065, 0.16)
Eyes closed, foam
SA: 0.055 (0.018, 0.091)
SL: 0.86 (0.277, 1.443)

3-408

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

Dietrich et al. N: 245
(1993)

Cincinnati, OH
US

1979-1984
(enrollment)

Cohort

Pregnant
mothers living in
older houses in
poor condition
and with
chipping lead-
based paint and
lead laden dust

Blood

Maternal and child venous blood

Age at measurement: T1 (maternal), delivery, 1
2, 3, 4, 5, 6 yr (child)

Mean (SD) (min-max) ug/dL:
T1: 8.4 (3.8) (1-27)

Neonatal: 4.8 (3.1) (1-22)

1	yr

2	yr

3	yr

4	yr

5	yr

6	yr

10.5 (4.9) (3-35)

17.1	(8.3) (6-49)

16.2	(7.6) (4-50)

14.0	(7.1) (4-45)
11.9 (6.4) (3-38)

10.1	(5.6) (2-33)

Bilateral
coordination,
visual-motor
contol, upper-
limb speed and
dexterity, and
fine motor
composite
assessed using
BOTMP

Age at outcome:
6 yr

NR

Beta

Bilateral coordination
Prenatal: -0.04 (-0.197, 0.117)b
Neonatal: -0.15 (-0.326, 0.026)b
Average (3-60 mo): -0.11
(-0.188, -0.032)b

Concurrent: -0.18 (-0.258,
—0.102)b

Visual-motor control
Prenatal: 0.06 (-0.097, 0.217)b
Neonatal: -0.1 (-0.296, 0.096)b
Average (3-60 mo): -0.05
(-0.148, 0.048)b

Concurrent: -0.12 (-0.218,
-0.022)b

Upper-limb speed and dexterity
Prenatal: -0.2 (-0.435, 0.035)b
Neonatal: -0.45 (-0.724,
—0.176)b

Average (3-60 mo): -0.19
(-0.327, -0.053)b

Concurrent: -0.31 (-0.447,
—0.173)b

Fine motor composite
Prenatal: -0.14 (-0.552, 0.272)b
Neonatal: -0.49 (-0.96, -0.02)b
Average (3-60 mo): -0.28
(-0.515, -0.045)b

Concurrent: -0.46 (-0.715,
-0.205)b

3-409

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

Wasserman et
al. (2000)

K. Mitrovica
and Pristina

Kosovo,
Yugoslavia

1985-1986
(enrollment)

Followed 54
mo

Cohort

Yugoslavia

Prospective

Study

N: 283 children

Pregnant
women
recruited from
K. Mitrovica
(lead smelter,
refinery, and
battery factory)
and Pristina
(town 40 km
south)

Blood

Prenatal maternal, delivery, and subsequent 6-
month interval venous blood: NR

Fine motor and
gross motor
composites
assessed using
BOTMP

Visual motor
integration
assessed using
Beery Test of
VMI

Age at outcome:
54 mo

Maternal age,
parental education,
number of siblings,
living arrangement,
HOME score at 3 yr,
maternal intelligence
at 2 yr (RSPM),
birthweight, BMI at 54
mo, child sex,
opportunities to
practice motor skill,
incomplete
lateralization

Beta for log—10 transformed Pb
Fine motor composite: -0.17
(-1.503, 1.163)c

Gross motor composite: 0.03
(-1.538, 1.598)°

VMI: -0.24 (-0.632, 0.152)c

tKim et al.

MOCEH study

(2013c) and

n: 884

Kim et al.



(2013b)

Mothers



recruited before '

Seoul,

20th wk of

Cheonan and

pregnancy

Ulsan

between and

Korea

were in

2006-2010

locations

Followed 6 mo

(Seoul,

Cohort

Cheonan and



Ulsan)

Blood

Maternal blood samples measured for Pb, Cd in
early (<20 wk) pregnancy and late (med = 39
wk) pregnancy

Age at measurement:

Early and late pregnancy

Early pregnancy: 1.4 (GM), 2.1 (90th), 9.8 (max)
pg/dL

Late pregnancy: 1.3 (GM); 2.1 (90th), 4.3 (max)
pg/dL

GM also available separately by 3 sites

PDI assessed
using BSID-II
(Korean version)

Age at outcome:
6 mo

Birth weight, infant
sex, maternal age
and education, family
income, breastfeeding
status, residential
area.

Beta

Early: 0.28 (
Late: -1.38 i

¦1.19, 1.75)
-3.31, 0.55)

Early:

Cd <1.47 |jg/L: 2.70 (0, 5.39)

Cd >1.47 |jg/L: -1.17 (-3.27,
0.94)

Late:

Cd <1.51 |jg/L: 0.18 (-2.70, 3.07)

Cd >1.51 |jg/L: -2.86 (-5.55,
-0.16)

3-410

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tKim et al.
(2018b)

4 cities: Seoul,
Anyang, Ansan
and Jeju
Korea

2011-2012
(enrollment)
Followed
through 24 mo

Cohort

CHECK cohort
n: 140

birth cohort-

pregnant

women

recruited from 4
cities in Korea
before delivery

Blood

Maternal blood; method NR
Age at measurement: delivery
Median (IQR):

Maternal: 2.7 (3.5, 5.7) pg/dL
Cord: 1.2 (0.8, 1.7) pg/dL

PDI assessed
using BSID-II
(Korean version)

Age at outcome:
13-24 mo

BPA, and phthalates,
maternal age
(continuous), birth
delivery mode
(categorical), monthly
household income
(categorical), child's
sex, and BDI
(continuous) of the
mother, gestational
age (continuous),
primiparous
(categorical), and
pre-pregnancy BMI
(categorical)

Beta (maternal blood)

Overall: -15.45 (-30.12, -0.79)
Boys: -18.32 (-45.35, 8.71)
Girls:-7.48 (-42.10, 27.15)

tY Ortiz et al.
(2017)

Mexico City
Mexico

Jul 2007-Feb
2011
Followed
through 24 mo

PROGRESS
birth cohort
n: 536

Women <20 wk
of gestation and
planning to
reside in Mexico
City for the next
3 yr.

Blood

Maternal blood analyzed using ICP-MS.
Age at measurement:

T2, T3

Mean:

T2: 3.7 pg/dL
T3: 3.9 pg/dL

Motor

development
assessed using
BSID-II I.

Standardized
scores (mean:
100, SD: 15).

Age at outcome:
24 mo

Infant sex, birth
weight, gestational
age, maternal age,
maternal IQ (WAIS
Spanish version),
HOME score.

Beta for log-transformed Pb
Motor Development:
T2: 1.97 (-2.46, 6.40)b c
T3: -11.01 (-17.55, -4.48)bc

Cohort

3-411

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tLiu et al.

Birth cohort

(2014c)

from 3 medical



centers

Pearl River

n: 362 mother-

Delta Region,

infant pairs (141

Guangdong

high Pb group

China

with cord BLL



>3.92 pg/dL and

Jan 2009-Jan

102 low Pb

2010

group <1.89

(enrollment)

pg/dL)

Followed for 3



yr



Cohort



tRvaiel et al.

ELEMENT

Blood

Cord blood and child blood analyzed using
GFAAS. 2 exposure groups created based on
cord BLL below 25th percentile (low) and above
the 75th percentile (high). Age at measurement:
At birth (cord), 6, 12, 24 and 36 mo (postnatal
child)

High and low Pb groups: cord BLLs: 5.63 and
1.35 |jg/dL; 6 mo BLL: 4.03 and 2.85 pg/dL; 12
mo: 4.87 and 3.79 pg/dL; 24 mo: 4.39 and 3.31
pg/dL; 36 mo: 3.94 and 3.28 pg/dL

PDI assessed
using BSID-II
(Chinese
version)

Age at outcome:
36 mo

Birth weight, sex,
maternal education,
IQ (WISC-R),
hemoglobin level,
smoking, age,
parental occupations,
household annual
income, HNES total
score

Beta comparing PDI score at 36
mo in high exposed (Cord BLL
>3.92 pg/dL) vs. low exposed
(Cord BLL <1.89 pg/dL):

-1.302 (-1.572, -1.031)

(2021)

Mexico City
Mexico

1997-2005
Cohort

project
n: 85

Mother-child
pairs recruited
at the Mexican
Social Security
Institute

Blood

Maternal and child venous blood; ICP-MS,
GFAAS

Age at measurement:

T1, T2, T3 (maternal); 12, 24 mo (child)

Maternal blood GM (SD):

T1
T2
T3

PDI assessed
using BSID-II
(Spanish
version)

Age at outcome:
12-24 mo

Maternal IQ (WAIS),
maternal age, infant
weight, length, SES,
infant age and sex,
current infant BLL

5.27 (1.93) pg/dL
4.74 (1.96) pg/dL
4.98 (1.93) pg/dL

A large number of results were
obtained from the mediation
analysis. In summary, T1, T2,
and T3 BLLs were associated
with nonsignificant decreases in
12-month PDI. This association
persisted for 24-month PDI at
less magnitude for T2 Pb.

Beta for 12-month PDI

Infant blood GM (SD):
12 mo: 3.92 (1.80) pg/dL
24 mo: 3.49 (1.93) pg/dL

T1
T2
T3

-0.24 (-0.95, 0.48)
-0.38 (-1.10, 0.35)
-0.33 (-1.06, 0.40)

Beta for mediation by GCNT1
cg18515027 methylation of In-
transformed T2 BLLs and 12-
month PDI:

Indirect: 1.25 (-0.11, 3.32)

3-412

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tShekhawat et n: 117
al. (2021)

Western

Rajasthan

India

2018-2019
(enrollment)
Follow-up at
6.5 mo
(average)

Cohort

Mother-child
pairs in third
trimester or at
delivery

Blood

Cord blood; ICP-OES

GM = 4.14 |jg/dL; mean = 4.77 ± 3.3 pg/dL;
median = 4.23 pg/dL
75th: 5.1 pg/dL

Motor

development
assessed using
BSID-III

Age at outcome:
6.5 mo

Maternal age,
gravida, gestational
age, maternal
education, child sex
and weight, preterm
birth, maternal food
intake during
pregnancy, smoking,
alcohol consumption,
maternal residential
and occupational
history, delivery type

(3 (95 % CI)

Umbilical cord Pb level <5
pg/dL (n = 70)

Composite motor:
-0.048 (-0.28, 0.19)
Subscale fine motor:

-0.10 (-1.80, 0.68)

Subscale gross motor:

0.14 (-0.84, 0.94)

Umbilical cord Pb level 5.0-
10.5 pg/dL (n = 47)

Composite motor:

0.01 (-1.19, 0.23)

Subscale fine motor:
0.03 (-3.34, 4.1)

Subscale gross motor:

-0.29 (-5.00, 0.11)

Henn et al.
(2012)

Mexico City
Mexico

1997-2000
(enrollment)
Followed for
24 mo

N: 455

Blood

Women	Child venous blood, ICP-MS

recruited during
pregnancy or at
delivery

Age at measurement: 12, 24 mo

12 mo mean (SD): 5.1 (2.6) pg/dL
24 mo mean (SD): 5.0 (2.9) pg/dL

PDI assessed
using BSID-II
(Spanish
version)

Age at Outcome:

12, 18, 24, 30,
36 mo

Sex, gestational age,
hemoglobin, maternal
IQ, maternal
education, and visit

Beta

12-month BLL:
0.02)

24-month BLL:
0.17)

-0.27 (-0.56,

-0.18 (-0.53,

Cohort

3-413

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tParaiuli et al.

Pregnant

Blood

PDI assessed

Maternal age and

Beta

(2015a)

women visiting



using BSID-II

education, BMI,

-4.83 (-16.53, 6.86)



the Bharatpur

Cord blood; ICP-MS, measured for Pb, As and



gestational age,

Chitwan,

General

Zn

Age at outcome:

family income, parity,



Bharatpur

Hospital



24 mo

birth weight, weight at



District

n: 100

Age at measurement:



24 mo, child age



Nepal

Birth cohort:

At birth



assessment, As, Zn,
HOME score



Sep-Oct 2008

women were

Median: 2.06 |jg/dL



(smoking and alcohol



Followed

selected if living

Max: 22.08 pg/dL



consumption not



through 24 mo

in the study





included given low



area for at least





prevalence)



Cohort

2 yr and were at
term pregnancy
(>37 wk of
gestation)









tParaiuli et al.

Birth cohort

Blood

PDI assessed

Maternal age and

Beta

(2015b)

from Bharatpur



using BSID-II

education, BMI,

-2.56 (-9.71, 4.59)



General

Cord blood; ICP-MS, measured for Pb, As and



gestational age,

Chitwan,

Hospital

Zn

Age at outcome:

family income, parity,



Bharatpur

n: 100



36 mo

birth weight, weight at



district



Age at measurement:



24 mo, child age at



Nepal

Resided in area
for at least 2 yr

At birth



assessment, As, Zn,
HOME score



Sep-Oct 2008

delivered at

Median: 2.06 pg/dL



(smoking and alcohol



Followed

term (i.e., >37

Max: 22.08 pg/dL



consumption not



through 36 mo

wk)





included given low



given low prevalence)

Cohort

3-414

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tJianq et al.
(2022)

Taipei
Taiwan

August 2008-

December

2009

(enrollment)
Follow-up 3 yr

Cohort

N: 53 children
Meconium (n =
36)

Hair (n = 52)
Fingernail (n =
43)

Longitudinal
birth cohort
study at a
medical center
hospital in
northern Taiwan

Meconium, Hair, Fingernail

All metals analyzed using ICP-MS.

Child meconium collected at birth

Child hair and fingernails collected at age 1 mo

Median (min, max):

Meconium: 25.6 (2.00, 8815) ng/g
Hair: 3.61 (0.31, 25.1) pg/g
Fingernail: 0.84 (0.06, 24.3) pg/g

Motor

development
assessed using
BSID-III.

Raw total motor
scores were
standardized to
expected mean
of 100 and SD of
15. Raw fine
motor and gross
motor scores
were

standardized to
expected mean
of 10 and SD of
3.

Age at Outcome:
3 yr

Maternal age and
education, newborn
birth head
circumference and
sex, and As and Cd
levels

Beta for log—10 transformed Pb
and log—10 transformed motor
development score

Meconium

Motor: -0.00001 (-0.021, 0.021)
Fine motor: 0.009 (-0.048,
0.065)e

Gross motor: -0.014 (-0.064,
0.036)e

Hair

Motor: 0.020 (-0.009, 0.049)e

Fine motor: 0.046 (-0.015,
0.107)e

Gross motor: 0.006 (-0.043,
0.054)e

Fingernails

Motor: -0.003 (-0.025, 0.019)e
Fine motor: -0.001 (-0.053,
0.052)e

Gross motor: -0.004 (-0.046,
0.037)e

3-415

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tZhou et al.

Shanghai

(2017)

Stress Birth



Cohort Study

Shanghai

n: 139

China



2010-2012

Women

(enrollment)

enrolled in

Followed for

prenatal clinics

24-36 mo after

of maternity

birth

hospitals during

Cohort

mid-to-late



pregnancy.

tLiu et al.

Guangxi Birth

(2022a)

Cohort Study



N: 703 children

Blood

Maternal blood Pb measured using AAS.
Age at measurement:

28-36 wk of gestation

GM: 3.30 pg/dL

Gross motor and
fine motor
development
assessed using
GDS (Chinese
version)

Age at outcome:
24-36 mo

Maternal age at
enrollment, economic
status, maternal
education, gestational
week, child sex, birth
weight and age

Beta per log—10 transformed BLL
Gross motor development: 3.31
(-6.11, 12.73)c

Fine motor development: 0.49
(-11.27, 12.24)c

Guangxi region
China

July-

September
2015

(enrollment)

Followed until
July-

September
2018 (3 yr)

Pregnant
women
recruited from
eight maternity
and child
healthcare
hospitals in six
cities of
Guangxi

Blood, urine

Prenatal maternal serum (first, second, and third
trimesters)

Infant urine

Age at measurement: NR

Maternal serum med (25th, 75th): 0.78 (0.54,
1.24) pg/L

Gross motor
development
using GDS
(Chinese
version)

Age at outcome:
2.57 (SD: 0.14)
yr

Maternal age, pre-
pregnancy BMI,
children's age,
children's gender,
blood sampling time,
delivery mode,
delivery gestational
week, birth head
circumference.

Beta per In-transformed pg/L
increase in Pb

Overall: -2.321 (-3.614, -1.029)c
Male: -3.426 (-6.162, -0.691 )c
Female: -1.182 (-2.805, 0.442)c

Infant urine med (25th,
pg/L

75th): 0.22 (0.14, 0.37)

Cohort

3-416

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tTavlor et al.
(2015)

Avon
UK

Apr. 1, 1991 —
Dec. 31, 1992
(expected
delivery date)

Followed 10 yr
Cohort

Subsample of
ALSPAC Study
N: 582 child
blood
N:4285
prenatal
maternal blood

Pregnant
women in
former Avon
Health Authority

Blood

Maternal blood collected in early pregnancy
(med: 11 wk of gestation); ICP-MS

Child venous blood

Age at measurement: 30 mo

Mean (SD)

Prenatal: 3.67 (1.47) pg/dL
Child: 4.22 (3.12) pg/dL

Balance (heel-
to-toe test) from
the Movement
Assessment
Battery for
Children
(Movement
ABC)

Age at outcome:
7 yr

Static and
dynamic balance
tests based on
BOTMP

Age at outcome:
10 yr

Sex, passive smoking
at 77 or 103 months
old (weekdays and
weekends), and
concurrent Ca and Fe
intakes

OR for >5 pg/dL vs. <5 pg/dL Pb
Prenatal Pb:

Heel-to-toe test: 1.01 (0.95, 1.01)
Dynamic balance: 1.02 (0.95,
1.09)

Static balance: 0.98 (0.92, 1.06)
Child Pb:

Heel-to-toe test: 0.98 (0.92, 1.05)
Dynamic balance: 1.01 (0.93,
1.09)

Static balance: 1.03 (0.94, 1.12)

3-417

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tTavlor et al.
(2018)

Avon
UK

Apr. 1, 1991 —
Dec. 31, 1992
(expected
delivery date)

Followed 10 yr

Cohort

Subsample of
ALSPAC Study

N:1558

Pregnant
women in
former Avon
Health Authority

Blood

Maternal blood; ICP-MS

Age at measurement:

Early pregnancy (med: wk 11 of gestation)

Mean (SD)

Prenatal: 3.66(1.55) pg/dL

Balance (heal-
to-toe test), ball
skills (beanbag
toss), and
manual dexterity
(threading lace
and placing
pegs)

Movement

Assessment

Battery for

Children

(Movement

ABC)

Age at outcome:
7 yr

Sex, maternal
education, smoking in
pregnancy, alcohol in
pregnancy, maternal
age and parity

OR for >5 pg/dL vs. <5 pg/dL Pb
Balance: 0.99 (0.74, 1.33)

Ball skills: 0.88 (0.58, 1.32)
Threading lace: 1.12 (0.83, 1.50)
Peg board - preferred hand: 1.19
(0.88, 1.60)

Peg board - non-preferred hand:
1.14 (0.85, 1.54)

OR for highest quartile (NR) vs.
lowest quartile (<5 pg/dL) of Pb
Balance: 0.98 (0.73, 1.31)

Ball skills: 1.07 (0.71, 1.63)
Threading lace: 1.01 (0.75, 1.35)
Peg board - preferred hand: 1.23
(0.92, 1.66)

Peg board - non-preferred hand:
0.99 (0.73, 1.32)

3-418

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tBoucher et al.
(2016)

Nunavik,
Quebec

Canada

October 2005
- February
2010 (outcome
assessment)

Cohort

Nunavik Child
Development
Study

N: 265 school
children

Phone

recruitment of
children with
umbilical cord
blood samples
obtained under
the Arctic Cord
Blood
Monitoring
Program

Blood

Fine motor
performance on
Santa Ana Form
Board (manual
dexterity), Finger
Tapping (fine

Age at measurement: birth (cord), 11.3 yr (child) motor speed),

and Stanford-
Binet Copying
(visuo-motor
integration)

Age at outcome:
11.3 yr (SD: 0.8)

Cord blood and child concurrent venous blood
ICP-MS

Cord mean, median (SD): 4.7, 3.7 (3.4) |jg/dL

Child blood mean, median, SD: 2.7, 2.0 (2.1)
pg/dL

Form Board: child age
and sex, social
environment,
maternal age, parity,
marital status,
smoking during
pregnancy; Finger
Tapping: child age
and sex, social
environment;
Stanford-Binet: child
age and sex, social
environment, marital
status

Others considered:
adoption status,
primary caregiver's
years of education,
Peabody Picture
Vocabulary Test,
RPM, parity, mother
fluency in
English/French,
assimilation to
Western culture,
alcohol and illicit drug
use during
pregnancy, other
contaminants, nutrient
biomarkers

Beta for log-transformed Pb
Cord blood

Manual dexterity: -0.08df
Fine motor speed: -0.19 (-0.33,
—0.05)bf

Visuo-motor integration: -0.01df
Child blood

Manual dexterity: -0.17 (-0.34,

0)b,f

Fine motor speed: -0.21 (-0.37,
—0.05)bf

Visuo-motor integration: 0.1df

3-419

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tParaiuli et al.

n: 79

(2013)





women living in

Chitwan Valley

the study area

Nepal

(i.e., Chitwan)

Sep-Oct 2008

for at least 2 yr,

Cross-

at term

sectional

pregnancy



when the



mothers visited



the hospital



(more than 37



wk of



gestation), age



of 18-40 yr,



singleton birth,



and no report of



diabetes,



hypertension, or



preeclampsia

tLiu et al.

Birth cohort

(2014d)

n: 415 mother-



child pair (219

Shenzhen,

high Pb group

Guangdong

>4.89 pg/dL at

China

first trimester

Jan 2009-Jan

and 196 low Pb

2010

group <1.96

Followed for

pg/dL)

26-30 wk



Cohort

Pregnant



women



recruited during



the early



pregnancy (10-



14 wk)

Blood

Cord blood Pb concentrations determined using
ICP-MS.

Age at measurement:
delivery

mean: 31.7 |jg/L; median: 20.6 |jg/L
75th: 35.1 pg/L
Max: 220.8 |jg/L

Neurodevelopm
ent assessed
using Brazelton
NBAS III

Age at outcome:
1 day old

Maternal age, parity,
mother's education
level; annual family
income, mother's
BMI, birth weight,
gestational age, age
of baby at NBAS
assessment

13 (95 % CI) change in score per
1 pg/L increase in blood Pb
Habituation: 1.44 (-1.19, 4.07)
Orientation: -0.12 (-8.34, 8.10)
Motor system: -2.15 (-4.27,
-0.03)

State organization: 2.15 (-1.58,
5.88)

State regulation: -0.75 (-3.86,
2.36)

Autonomic Stability: 0.71 (-0.48,
1.90)

Abnormal reflex: 1.07 (-1.32,
3.46)

Blood

Maternal, cord blood analyzed using GFAAS.
Maternal BLL classified as low or high

Age at measurement:

First, second and third trimester and at delivery

Low and High BLL groups: First trimester: 1.22
pg/dL and 6.49 pg/dL; second trimester: 1.01
pg/dL and 5.63 pg/dL; third trimester: 1.19 pg/dL
and 6.31 pg/dL; and delivery: 1.26 pg/dL and
6.65 pg/dL

Neurodevelopm
ent assessed
using NBNA

Age at outcome:
3 days

Infant sex, maternal
hemoglobin, IQ,
tobacco use and
parents' occupation,
education, yearly
household income.

Beta for change in NBNA score
per log-transformed Pb

T1
T2
T3

-4.86 (-8.831, —0.889)f
-3.98 (-8.180, 0.220)f
-3.65 (-6.609, 1.309)f

Cord: -3.39 (-7.531, 0.751)

3-420

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tNozadi et al.
(2021)

Navajo Nation
United States

Enrollment
February 2013
- June 2018

Follow-up at
age 10-13 mo

Navajo Birth
Cohort Study
(NBCS)

n: 327

Blood

Maternal blood Pb from the 36-week visit or at
the time of delivery was processed using ICP-
DRC-MS.

Age at Measurement:

Mean (SD) maternal age at birth = 27.4 (5.87)
years. Children assessed at 10 and 13 mo.

GM = 0.410 |jg/dL; median
75th: 0.51 pg/dL
95th: 1.20 pg/dL

0.37 pg/dL

Neurodevelopm Multivariable linear Beta

ent assessed
using Ages and
Stages

Questionnaire

Inventory

(ASQ:I)

Age at outcome:
10, 13 mo

regression for fine
motor adjusted for
blood cadmium, urine
cesium, urine arsenic,
and mother's
education; gross
motor adjusted for
urine strontium

Fine motor: -0.63 (-1.19, -0.08)
Gross motor: 0.14 (-0.47, 0.75)

Cohort

tKao et al.
(2021)

Taipei

Taiwan

2011-2014

Cross-
Sectional

recruited from
Taipei MacKay
Memorial
Hospital
n:139 children
less than 3 yr of
age

Hair, fingernails

Pb concentrations in hair and fingernails were
measured using ICP-MS

Age at Measurement:

Mean (SD) 2.8 (0.4) years (children under 3 yr)

GM (SD): hair 2.9 (4.8) i1/4g/g, nails 0.8 (5.1)
TVig/g

Motor

development
assessed using
BSID-III

Age at outcome:
2.8 ± 0.4 yr

Sex, gestational age
at birth, age of the
house (years), leafy-
vegetable intake
(servings/week), and
the area of surface
roads within 100 m of
the residence

Regression results were not
reported because they were not
statistically significant.

3-421

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tNvanza et al.
(2021)

Northern
Tanzania
Tanzania

2015-2017
(enrollment)
Followed for
12 mo
Cohort

Mining and

Health

Prospective

Longitudinal

Study in

Northern

Tanzania

n: 439

Birth cohort of
mother-child
pairs recruited
in 2nd trimester

Maternal dried blood spots; ICP-MS, measured
for Pb, Hg, and Cd

Age at measurement:

T2

Median: 2.72 |jg/dL
75th: 4.25 pg/dL
Max: 14.5 pg/dL

Gross motor and
fine motor
development
assessed using
MDAT.

Scores in each
domain
classified as
normal (>90th
percentile on all
items in that
domain or <90th
percentile on
one or two items
in the domain) or
impaired (<90th
percentile on
more than two
items in a
domain).

Maternal age and
education, maternal
and paternal
occupation, number
siblings under 5 yr at
home, and family
SES, infant sex, age,
birth weight, height
and weight as a proxy
for nutritional status.
(Covariates with p <
0.20 retained in the
final models.)

Prevalence ratio

Gross motor development: 1.0

(0.9, 1.0)

Fine motor development: 1.0
(0.9, 1.0)

Age at outcome:
between 6 and
12 mo

3-422

-------
Reference
and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% Clsa

tPalaniappan n = 755 school Blood
et al. (2011) children (age 3-
7 yr)

Chennai
India

2003-3006

Cross-
sectional

Child venous blood; LeadCare Analyzer

Children	.	x

attending public A9e at measurement: 3-7 yr

schools in

Chennai	Mean (SD): 11.5 (5.3) pg/dL

(kindergarten -
1st grade)

Visual-motor

(drawing),

visual-spatial

(matching), fine

motor

(pegboard)

subtests and

composite

assessed using

WRAVMA

Age at outcome:
3-7 yr

Gender, age,
hemoglobin level,
average monthly
income of the family
(categorical) and
parent education
(categorical)

Beta

Drawing: -0.29 (-0.51, -0.07)
Matching: -0.14 (-0.31, 0.02)
Pegboard: -0.19 (-0.38, 0.01)
Composite: -0.26 (-0.45, -0.07)

Standardized
scores (mean:
100, SD: 15)

AAS = atomic absorption spectrometry; As = arsenic; BASC = Behavior Assessment System for Children; BDI = Beck Depression Inventory; BLL = blood lead level; BOTMP =
Bruininks-Oseretsky Test of Motor Proficiency; BPA = bisphenol A; BRIEF = Behavior Rating Inventory of Executive Functions; BSID = Bayley Scales of Infant and Toddler
Development; CBCL = Child Behavior Check List; CCAAPS = Cincinnati Childhood Allergy and Air Pollution Study; CHECK = Health and Environmental Chemicals in Korea; CI =
confidence interval; CKD = chronic kidney disease; CKiD = Chronic Kidney Disease in Children; CPRS = Conners' Parent Rating Scale; C-TRF = Caregiver-Teacher Report Form;
CTRS = Conners' Teacher Rating Scale; DSM = Diagnostic and Statistical Manual of Mental Disorders; ELEMENT = Early Life Exposure in Mexico to Environmental Toxicants; FSIQ
= full-scale intelligence quotient; GDS = Gesell Development Schedules; GFAAS = graphite furnace atomic absorption spectrometry; GM = geometric mean; Hgb = hemoglobin;
HOME = Health Outcomes and Measures of the Environment; IQ = intelligence quotient; MOCEH = Mothers' and Children's Environmental Health; NBAS = Neonatal Behavioral
Assessment Scales; NBNA = Neonatal Behavioral Neurological Assessment; NHBCS = New Hampshire Birth Cohort Study; NR = not reported; OR = odds ratio; Pb = lead; PDI =
Psychomotor Developmental Index; PHDCN = Project on Human Development in Chicago Neighborhoods; SDQ = Strengths and Difficulties Questionnaire; SES = socioeconomic
status; SRP = self-report of personality; SRS = Social Responsiveness Scale; T1 = first trimester of pregnancy; T2 = second trimester of pregnancy; T3 = third trimester of
pregnancy; WAIS = Weschler Adult Intelligence Scale; WISC-R = Wechsler Intelligence Scale for Children; WRAVMA = Wide Range Assessment of Visual-Motor Abilities; Zn = zinc.
aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

The CI was calculated from a p-value and the true CI may be wider or narrower than calculated.

°Results are unstandardized because the Pb level distribution data was not available.

dResults are unstandardized because they did not have an associated SE, CI, or p-value reported in the study.
eResults are unstandardized because the biomarker used for Pb exposure measurement is not blood, tooth, or bone.

'Results are unstandardized because the log base used for exposure transformation was unspecified in the study.
tStudies published since the 2013 Integrated Science Assessment for Lead.

3-423

-------
Table 3-11T

Animal toxicological studies of Pb exposure and motor function



Study

Species ^Stock/Strain), Timing o, Exposure BLL as Reported (pg/dL)

Endpoints Examined

Flores-Montova and Sobin
(2015)

Mouse (C57BL/6)

Control (distilled water),
M/F, n = 19 (8/11)

30 ppm, M/F, n = 26
(16/10)

230 ppm, M/F, n = 16
(12/4)

PND 0 to PND 28 Oral

Oral,

PND 28 - Males:

drinking



water

0.2 |jg/dL for Control

Oral,



lactation

3.93 |jg/dL for 30 ppm



9.39 |jg/dL for 230 ppm



PND 28 - Females:



0.19 |jg/dL for Control



3.19 |jg/dL for 30 ppm



12.14 |jg/dLfor230 ppm

Oral,

8 wk:

drinking



water

1.8 |jg/dL for Control



21.7 |jg/dL for 250 mg/L

PND 28: OFT, Rotarod Test

Zou et al. (2015)

Mouse (ICR)

Control (distilled water),
M, n = 10

250 mg/L solution, M, n
= 10

-5 wk to 8 wk

8 wk: Rotarod Test,
Locomotor Activity

Rao Barkur and Bairv (2016)

Rat (Wistar)	PG: GD -30 to GD Oral,

Control (tap water), M, n 0	lactation

= 12	In utero

G: GD 1 to GD 21

0.2% solution, PG, M, n
= 12	L: PND 1 to PND

21

0.2% solution, G, M, n =

12

0.2% solution, L, M, n =

12

PND 22:

0.19 |jg/dL for Control
3.03 |jg/dL for PG
5.51 |jg/dL for G
26.86 |jg/dL for L

PND 3, 4, 5: Surface Righting
Reflex, PND 6, 8, 10, 12:
Swimming Performance, PND
8, 10, 12: Negative Geotaxis,
PND 14-18: Ascending Wire
Mesh,

3-424

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Betharia and Maher (2012)

Rat (Sprague Dawley)

PND 24:

Control (RO Dl water),
M/F, n = 11-13

GD 0 to PND 20

Oral,
lactation
In utero

PND 2:

1.77 ng/g (0.188 pg/dL) for
Control

85.17 ng/g (9.02 pg/dL) for 10

PND 1-10: Surface Righting
Reflex, PND 24, 59: OFT



10 |jg/mL, M/F, n = 11-
13





|jg/mL









PND 25:





PND 59:





0.83 ng/g (0.088 pg/dL) for





Control (RO Dl water),





Control





M/F, n = 10-11





9.21 ng/g (0.98 pg/dL) for 10





10	|jg/mL, M/F, n = 10-

11





pg/mL









PND 60:

0.23 ng/g (0.024 pg/dL) for
Control

0.30 ng/g (0.032 pg/dL) for 10
pg/mL



3-425

-------
Study

Species (Stock/Strain), Timing of
n, Sex	Exposure

EDe°aSi|Js0 BLL as ReP°rted (M9^L)

Endpoints Examined

Basha and Reddv (2015)

Rat (Wistar)

Control (deionized
water), M, n = 8

0.2 % solution, M, n = i

GD 6 to GD 21 Inutero PND21:

0.21 |jg/dL for Control

11.2	|jg/dL for 0.2% solution
PND 28:

0.33 |jg/dL for Control

12.3	|jg/dL for 0.2% solution
4 mo:

0.19 |jg/dL for Control
5.9 |jg/dL for 0.2% solution

PND 4-7: Surface Righting
Reflex, PND 8-10: Negative
Geotaxis, PND 12-16:
Forelimb Hang, PND 21, PND
28, 4 mo: Locomotor Activity

Tartaalione et al. (2020)

Rat (Wistar) GD -28 to PND 23

Oral,

PND 23:

PND 4, 7, 10, 12: Neonatal



Control (tap water), M/F

lactation



Spontaneous Movement,



n = 16 (9/7)

In utero

0.007 |jg/mL (0.7 pg/dL) for
Control

PND 4, 7, 10, 12: Surface
Righting Reflex, PND 4, 7, 10,



50 mg/L, M/F, n = 16





12: Negative Geotaxis, PND



(9/7)



0.255 pg/mL (25.5 pg/dL) for 50
mg/L

30: OFT

Faulk et al. (2014)

Mouse (Agouti)

GD-14 to PND 21 Oral,

PND 21 (Maternal BLL):

PND 90, 180, and 270:



Control (distilled water),

lactation



Locomotor Activity



M/F, n = 30

In utero

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Bashaetal. (2014)

Rat (Not Specified)

PND 1 to PND 21

Oral,

PND 45:

PND 45, 4 mo, 12 mo, 18 mo:



Control (deionized



lactation



OFT, Locomotor Activity



water), M, n = 6





0.42 pg/dL for Control





0.2% solution, M, n = 6





49.5 pg/dL for 0.2% solution











4 mo:











0.56 pg/dL for Control











14.4 pg/dL for 0.2% solution











12 mo:











0.46 pg/dL for Control











6.96 pg/dL for 0.2% solution











18 mo:











0.12 pg/dL for Control











11.2 pg/dL for 0.2% solution



Mansouri et al. (2012)

Rat (Wistar)

Control (distilled water),
M/F, n = 16 (8/8)

50 mg/L, M/F, n = 16
(8/8)

PND 70 to PND
100

Oral,

drinking

water

PND 100 - Males:
2.05 pg/dL for Control
8.8 pg/dL for 50 mg/L
PND 100 - Females:

PND 100: OFT, Rotarod Test

2.17 |jg/dL for Control
6.8 |jg/dL for 50 mg/L

3-427

-------
Study

Species (Stock/Strain), Timingof Exposurc BLL as Reported (Mg/dL)	Endpoints Examined

Duan et al. (2017)	Mouse (CD1)	PND 1 to PND 21 Oral,	PND21:	PND 7, 11, 15, 19: OFT

Control (distilled water),	lactation

M/F, n = 5	16.2 pg/L (1.6 pg/dL) for Control

27 ppm, M/F, n = 5	191.8 pg/L (19.2 pg/dL) for 27

ppm

109 ppm, M/F, n = 5

283.4 pg/L (28.3 pg/dL) for 109
ppm

PND 35:

14.3 pg/L (1.4 pg/dL) for Control

283.4 pg/L (28.3 pg/dL) for 27
ppm

376.9 pg/L (37.7 pg/dL) for 109
ppm

Wana et al. (2016)

Rat (Sprague Dawley)

PND 24 to PND 56

Oral,

PND 56: PND 60-66: OFT



Control (tap water), M, n



drinking





= 7



water

11 pg/L (1.1 pg/dL) for Control



100 ppm, M, n = 9





133 pg/L (13.3 pg/dL) for 100 ppm

Shvachiv et al. (2018)

Rat (Wistar)

Intermittent

Oral,

PND 196: PND 189: OFT



Control (tap water), M/F,

Exposure: GD 7 to

drinking





n = 8

PND 84, PND 140

water

<0.1 pg/dL for Control





to PND 196

Oral,





0.2% (p/v) solution



lactation

18.8 pg/dL for 0.2% (Intermittent)



(distilled water), M/F, n =

Continuous

In utero





9 - Intermittent

Exposure: GD 7 to



24.4 pg/dL for 0.2% (Continuous)

exposure	PND 196

0.2% (p/v) solution, M/F,
n = 9 - Continuous
exposure

3-428

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Stansfield et al. (2015)

Rat(Long-Evans)

Control (chow), M/F, n =
11-23

1500 ppm, M/F, n = 11-
23

GD Oto PND 50

Oral, diet
Oral,
lactation
In utero

PND 50:

0.6 pg/dL for Control
22.2 pg/dL for 1500 ppm

PND 50: Locomotor Activity

Neuwirth et al. (2019a)

Rat(Long-Evans)

Control (tap water), M/F,
n = 48 (30/18)

150 ppm, M/F, n = 62
(32/30)

1000 ppm, M/F, n = 49
(30/19)

GD 0 to PND 22

Oral,
lactation
In utero

PND 22:

-------
Study

Species (Stock/Strain), Timingof Exposurc BLL as Reported (Mg/dL)	Endpoints Examined

Sobolewski et al. (2020)	Mouse (C57BL/6)

F0:

Control (distilled Dl
water), F, n = 10

100 ppm, F, n = 10

F1:

see Figure 1, n = 12
F2:

see Figure 1, n = 12
F3:

see Figure 1, n = 8-10

Sinah et al. (2019)

Rat (Wistar) 3 mo to 6 mo

Control (distilled water),

M, n = 5

2.5 mg/kg, M, n = 5

Oral,
gavage

6 mo:

5.76 |jg/dL for Control
28.4 |jg/dL for 2.5 mg/kg

6 mo: Locomotor Activity,
Rotarod Test

Viaueras-Villasenor et al.

Rat (Wistar) GD 0 to PND 21

Oral,

PND 110:

PND 90 to PND 110:

(2021)

Control (tap water), M, n

lactation



Locomotor Activity



= 8

In utero

2.04 |jg/dL for Control





320 ppm, M, n = 8



26.3 |jg/dL for 320 ppm



F1: GD -60 to PND Oral,
23-27	lactation

In utero

F1 PND 6-7:

0 |jg/dL for Control

12.5 |jg/dL for 100 ppm (F0
dosing)

F3 PND 6-7:

0 ng/dL for Control

0 |jg/dL for 100 ppm (F0 dosing)

PND 60-120 (variable by
endpoint): Locomotor Activity

3-430

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Al-Qahtani et al. (2022)

Mouse (Albino)

Control (distilled water),
M, n = 10

0.2 mg/kg, M, n = 10

8-9 wk to 14-15 wk

Oral,
gavage

14-15 wk:

1.2 |jg/100 mL (1.2 pg/dL) for
Control

7.1 |jg/100 mL (7.1 pg/dL) for 0.2
mg/kg

NR: Locomotor Activity

BLL = blood lead level; F# = filial generation; F = female; GD = gestational day; LOD = limit of detection; M = male; MRI = magnetic resonance imaging; mo = month(s); NaAc =
sodium acetate; NR = not reported; OFT = open-field test Pb = lead; PG = pregestation; PND = postnatal day; wk = week(s); yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

3-431

-------
Table 3-12E Epidemiologic studies of Pb exposure and sensory organ function in children

Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

Dietrich et al. (1992)

Cincinnati, U.S.

Cross-sectional

The Cincinnati
lead study cohort

n: 259

Blood

Age at measurement: prenatal-
yr

Mean (SD) |jg/dL:

Prenatal 8.2 (3.8)

Neonatal 4.8 (3.3)

Central auditory
processing abilities and
cognitive developmental
status

Age at outcome: 5 yr

Measures of fetal distress
and growth, perinatal
complications, postnatal
indices of health and
nutritional status,
sociodemographic
characteristics, and
psychosocial features of the
home environment

Betab

Filtered Word
Score (total
number of words
correctly identified
in both ears)

Prenatal: -0.12

Neonatal: -0.26

Mean lifetime
through 5 yr:

-0.07

Schwartz and Otto (1991)

HHANES, U.S.

Cross-sectional

Hispanic Health Blood

and Nutrition
Examination
Survey
n: 3545

Age at measurement: 6-19 yr

Median (25th' 75th) pg/dL:
Mexican Americans 8 (6, 11)
Cuban American 8 (6, 10)
Puerto Ricans 8 (6, 11)

Elevated hearing
threshold

Audiometric evaluations
were performed for all
subjects Beltone model
200-C audiometers were
used in the survey;
Hearing threshold was
defined as the lowest
intensity of a pure tone
that was just audible to
the subject.

NR

An increase15 in
BLL from 7
~microg/dl to 18
pg/dl was
associated with an
approximately 2-dB
loss of hearing at
all frequencies

Age at outcome: 6-19 yr

3-432

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

Schwartz and Otto (1987)

NHANES II

U.S.

Cross-sectional

NHANES II	Blood

n: 4519	Age at measurement: 4-19 yr

Range of Pb: 6 to 47 |jg/dL

Hearing thresholds

Standard Beltone Model
200C audiometers were
used and ca liberated
weekly with B&K Model
2203 sound level meters
in accordance with 1969
ANSI specifications.

Tests were conducted at
500, 1000, 2000, and
4000Hz on each ear.

Race, lead, ear discharge,
cold in last 2-week, other
ear condition, chronic ear
discharge, income, dietary
calcium, sex, current cold,
ringing in ear(s), earache,
previous running ear,
diagnosed hearing
impairment, degree of
urbanization, head of
household education level

The risk of
elevated hearing
thresholds at 500,
1000, 2000 and
4000 Hz increased
with increasing
PbB for both ears

Age at outcome: 4-19 yr

tYin etal. (2021)

n: 234-7596 in 8 Blood
studies

Age at measurement: 3-87 yr

Hearing loss

All studies included in the
meta-analysis controlled for
age and sex. Adjustment for
other potential confounders
varies by studies, but
includes monthly income,
education levels, smoking
status, BMI, ethnicity, work
duration, ototoxic
medication, blood lead,
occupational noise, loud
noise, and firearm noise,
and hypertension and
diabetes

OR (95% Cl)b
1.53 (1.24,1.87)

tChoi and Park (2017)

Korea
2010-2012

KNHANES
n: 5187 adults
and 853
adolescents

Blood

Hearing loss (>15dB) at
speech frequency;
Hearing loss (>15dB) at
high frequency

Age, age squared, sex,
education, BMI, current
cigarette smoking

OR (95% Cl)b

Hearing Loss (>15
dB) High-frequency
PTA

Pb Quartile 2
(0.978-1.260):

0.89 (0.39, 2.03)

3-433

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

Cross-sectional

Graphite furnace atomic
absorption spectrometry

Age at Measurement:
adolescents 12-19 yr (mean±SE
15.6 ± 0.10)

Geometric mean (95% CI) (age-
adjusted): 1.26 [jg/dL (1.22, 1.30)

Pure-tone air conduction
hearing thresholds were
obtained for each ear at
frequencies of 0.5, 1, 2, 3,
4, and 6 kHz over an
intensity range of -10
to 110 dB -10 to 110 dB.

Age at outcome: 12-19 yr

Pb Quartile 3
(1.261-1.557):
1.88 (0.83, 4.25)

Pb Quartile 4
(1.562-5.904):

1.38 (0.63, 3.02)
Per doubling of Pb:
1.26 (0.73, 2.16)
Hearing Loss (>15
dB) Speech-
frequency PTA
Pb Quartile 2
(0.978-1.260):
1.17 (0.41, 3.32)

Pb Quartile 3
(1.261-1.557):

1.08 (0.38, 3.08)
Pb Quartile 4
(1.562-5.904):

1.24 (0.34, 4.49)
Per doubling of Pb:
1.2 (0.48, 3.05)

3-434

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

tXu et al. (2020)

China

October-December 2014

Cross-sectional

n: 116

Blood

Graphite furnace atomic
absorption spectrometry (GFAAS,
Jena Zeenit 650, Germany)

Age at measurement: 3-7 yr

Median ± SEM (P25, P75):

Exposed group

5.29 ± 0.29 (3.61, 7.40)

Refence group

3.63 ± 0.24 (2.98, 4.77)

DNA methylation and
hearing loss

Age at outcome: 3-7 yr

Both continuous variables
for child age, gender,
weight, height and BMI, and
categorical variables for
presence of family member
smoking, residence distance
to the road, residence
nearby noise, residence
renovation noise within a
year, often listening music
with earphones within a
year, often watching
television programs in loud
noise, and often play (i.e.,
toys or music, etc.) in loud
noise

Beta (95% Cl)b

Q1 0.139 (0.007,
2.968

Q2 0.051 (0.003,
0.977)

Q3 0.16 (0.016,
1.58)

Q4 2.765 (1.795,
15.237)

OR (95% CI)

Hearing loss in
both ears 1.40
(1.06, 1.84)

Left ear 1.46 (1.12,
1.91)

tSharaorodskv et al. (2011) NHANES

n: 2535

NHANES, U.S.

2005-2008
Cross-sectional

Blood

Inductively coupled plasma mass
spectrometry

Age at Measurement: 12-19 yr

Weighted Mean (95% CI):
Age 12-13: 1.00 pg/dL
(0.92-1.09 pg/dL)

Age 14-15: 0.93 pg/dL
(0.87-0.99 pg/dL)

Age 16-17: 0.85 pg/dL
(0.79-0.91 pg/dL)

Age 18-19: 0.93 pg/dL
(0.84-1.03 pg/dL)

Any Hearing Loss (>15
dB),

High-Frequency Hearing
Loss, Low-Frequency
Hearing Loss

Age at outcome: 12-19 yr

Age, sex, race-ethnicity,
PIR, history of 3 or more ear
infections, loud noise
exposure, and smoking

OR (95% Cl)b (<1
pg/dL reference)

Any >15 dB
1-1.99 pg/dL 0.99
(0.67-1.46)

>2 pg/dL 1.95
(1.24-3.07)

High-Frequency
1-1.99 pg/dL 1.20
(0.80-1.80)

>2 pg/dL 2.22
(1.39-3.56)

Low-Frequency
1-1.99 pg/dL 1.24
(0.82-1.86)
>2 pg/dL 1.13
(0.61-2.07)

3-435

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

tLiu etal. (2018c)

Guiyu (e-waste recycling
area) & Haojing (exposure
control, no e-waste
processing), China

2014

Cross-sectional

n: 234 (146
exposed; 88
reference)

Blood

Graphite furnace atomic
absorption spectrometry (GFAAS,
Jena Zeenit 650, Germany)

Age at Measurement:

3-7 yr

Mediant SE: 4.94 ± 0.20 |jg/dL in
exposed; 3.85 ± 1.81 pg/dL in
reference

Hearing loss, Low
frequency hearing loss,
High frequency hearing
loss

Age at outcome: 3-7 yr

Child age, gender, weight,
height, BMI, parent
education level, family
member smoking, family
monthly income, residence
distance to the road,
residence nearby noise,
residence renovation noise
within a year, often listening
to music with earphones
within a year, often watching
television programs in loud
noise, and often play (i.e.,
toys or music, etc.) in loud
noise

OR (95% Cl)b

Hearing loss total
1.24 (1.029, 1.486)
Low frequency
1.02 (0.869, 1.190)
High frequency
1.08 (0.839, 1.379)

3-436

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

tPawlas et al. (2015)

Upper Silesia, Poland

1996-2001 and 2008-2010
Cross-sectional

Two cohorts
merged
n: 483

Blood

Graphite furnace atomic
absorption spectrometry

Age at Measurement:
4-13 yr

Median: 4.50 |jg/dL

Pure-tone audiometry
(PTA),

Brainstem auditory
evoked potentials
(BAEP),

Acoustic otoemission
Age at Outcome: 4-13 yr

Cohort, mother's education
(dichotomized into
'secondary school or
higher', or 'less',
corresponding to primary
and apprenticeship) and
smoking during pregnancy,
and the child's sex, birth
weight, apgar score, history
of mumps, age, and
pressure in middle ear on
both sides

Beta (95% Cl)b

ALAD MspI
ALAD1-1 0.3
(0.15, 0.45)
ALAD*2 0.42
(-0.03, 0.87)

ALAD Rsa1
TT+TC 0.3 (0.1,
0.5)

CC 0.2 (-0.05,
0.45)

VDR Bsml
bb 0.03 (-0.22,
0.28)

Bb+BB 0.4 (0.25,

0.55)

VDR taq1

TT 0.04 (-0.21,
0.29)

Tt+tt 0.4 (0.2, 0.6)

VDR fokl

FF+Ff 0.4 (0.25,
0.55)

ff -0.1 (-0.6, 0.4)

3-437

-------
tSilveret al. (2016)

Sanhe County, Hebei
Province, China

November 2009- November
2011

Cohort

n: 391 (ARB:
auditory
brainstem
response), 1148
(VA: visual
acuity)

Maternal Blood

AAS

Age at Measurement:

Pregnant woman 18 yr or older
Mean (SD) gestational age at
mid-pregnancy visit
ABR subset 15.7 (2.2) weeks

VA subset 15.5 (1.9) weeks
Mean (SD) gestational age at
late-pregnancy

ABR subset 38.8 (1.3) weeks
VA subset 39.3 (1.3) weeks
Mean (SD) gestational age at
birth

ABR subset 39.2 (1.1) weeks
VA subset 39.7 (1.1) weeks

ABR Pb median
2.9 |jg/dL at mid-pregnancy
3.0 |jg/dL at late-pregnancy
<2.0 |jg/dL at birth (cord blood)
GM (SD)

2.4 (2.5) |jg/dL at mid-pregnancy

2.7 (2.3) |jg/dL at late-pregnancy

<2.0 |jg/dL at birth (cord blood);
VA median

2.9 |jg/dL at mid-pregnancy, 3.3
|jg/dL at late-pregnancy, 2.1
|jg/dL at birth (cord blood)

GM (SD)

2.4 (2.6) |jg/dL at mid-pregnancy
2.9 (2.2) |jg/dL at late-pregnancy
<2.0 |jg/dL at birth (cord blood)

3-438

ABR;

Grating visual acuity (VA)

Age at Outcome:
ABR mean 2 d old
VA mean 6 wk old

Sex, age attesting, cord
blood iron status,
gestational age, birth
weight, head circumference

Mid pregnancy
lead Med. (2-3.8
|jg/dL) 0.02 (-0.01
-0.05)

Late-pregnancy
lead High (>3.8
|jg/dL) 0.05 (0.02,
0.08)

Late-pregnancy
lead Med. (2-3.8
|jg/dL) 0.03 (0.01,
0.06)

Cord lead High
(>3.2 [jg/dL) 0
(-0.02, 0.03)

Cord lead Med. (2-
3.2 [jg/dL) 0
(-0.03, 0.03)

Beta (95% Cl)b

ARB C-P ratio

Mid pregnancy
lead High (>3.8
|jg/dL) 0.02 (-0.01
-0.05)

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

tAlvarenqa et al. (2015)

Brazil

Followed 35.5 mo

Contemporary cross-sectional
cohort

n: 130 children
(80 males & 50
females)

Blood

AAS with graphite furnace

Age at Measurement:

18 mo-14 yr (Mean: 6 yr 8 mo ±
2 yr 3 mo)

Mean: 12.2 pg/dL; SD = 5.7
pg/dL

Median: 10.2 pg/dL

Auditory brainstem
response

Age, gender, cumulative
blood lead levels, and

date of the audiological
Age at outcome: 18 mo- assessment
14 yr

Beta (95% Cl)b

Wave III, in relation
to wave I
Constant 4.00
(3.97, 4.04)

Wave I RE 0.58
(0.44, 0.72)

Male RE 0.09
(0.05, 0.13)

Constant 4.03
(3.99, 4.06)

Wave I LE 0.61
(0.45, 0.77)

Male LE 0.07
(0.03, 0.11)

Wave V, in relation
to wave III

Constant 5.77
(5.74, 5.80)

Wave I RE 0.81
(0.68, 0.94)

Male RE 0.073
(0.03, 0.11)

Constant
5.78 (5.75, 5.81)
Wave I LE 0.85
(0.73, 0.97)

Male LE 0.08
(0.05, 0.12)

3-439

-------
Reference and Study
Design

Study
Population

Exposure Assessment

Outcome

Confounders

Effect Estimates
and 95% CIs

tFillion et al. (2013)

n: 228

Blood

Contrast sensitivity
(cycles per degree, cpd);

Age, sex, current smoking
(yes vs. no), current drinking

Beta (95% Cl)b

Spatial frequency

Lower Tapajos River Basin,



Inductively coupled plasma mass

Acquired color vision loss
(color confusion index,
CCI)

(yes vs. no)

with %EPA



1.5 cpd -1.32
(-4.30; 1.65)

State of Para, Brazil



spectrometry (ICP-MS, Perkin
Elmer DRC II)



May to July 2006



Age at Measurement:

Age at outcome: 15-66 yr



3 cpd 2.06 (-2.87;
6.99)





15-66 yr (median = 33.0 yr)





6 cpd 0.60 (-6.04;
7.25)

12 cpd -13.33
(-23.28; -3.49)

18 cpd -2.43
(-6.64; 1.79)

CCI 0.16 (-0.03;
0.33)

Cross-sectional



Mean = 12.8 ± 8.4 |jg/dL; Median
= 10.5 |jg/dL





AAS = Atomic absorption spectrometry; ABR = Auditory brainstem response; BLL = blood lead level; CI = confidence interval; OR = odds ratio; Pb = lead; PTA = pure-tone average.
aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bEffect estimates are not standardized because data pertaining to the BLL distribution and/or base for the log-transformation were not reported.
tStudies published since the 2013 Integrated Science Assessment for Lead.

3-440

-------
Table 3-13E Epidemiologic studies of Pb exposure, social cognition, and behavior in children

Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tKimetal. (2016)

South Korea
2005-2006
(enrollment); 2009-
2010 (follow-up)
Cohort

CHEER study
n: 2,437

Children recruited
from 33 elementary
schools across 10
Korean cities

Blood

Child blood; GFAAS

Age at measurement:

7-8 yr old, 9-10 yr old, and 11-

12 yr old

GM (pg/dL):

7-8 y: 1.64; 9-10 y: 1.58; 11-12
y: 1.58

75th (pg/dL):

7-8 y: 2.36; 9-10 y: 2.08; 11-12
y: 2.05
95th (pg/dL):

7-8 y: 3.47; 9-10 y: 3.05; 11-12
y: 3.05

Autistic behaviors

Parent responses to
ASSQ and SRS

Age at outcome:
11-12 yr

Child sex, fetal and
environmental tobacco
smoke, parental
education levels,
family income, low
birth weight,
breastfeeding,
gestational age, fish
intake, and blood Hg
level

Change in SRS Scores*
Exposure at 7-8 yr
1.37 (0.75, 1.98)
Exposure at 9-10 yr
0.56 (-0.33, 1.44)
Exposure at 11-12 yr
0.39 (-0.47, 1.25)

Change in ASSQ Scores*
Exposure at 7-8 yr
0.09 (0.03, 0.14)

Exposure at 9-10 yr
-0.02 (-0.09, 0.05)
Exposure at 11-12 yr
0.03 (-0.04, 0.10)

*Higher score indicates more
autistic behaviors

OR Autism (ASSQ >17)
Exposure at 7-8 yr
1.45 (1.10, 1.93)
Exposure at 9-10 yr
0.86 (0.60, 1.23)
Exposure at 11-12 yr
0.97 (0.70, 1.35)

3-441

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tArora et al. (2017) Roots of Autism Tooth

Sweden

2011-2016
(enrollment)

Cohort

and ADHD Twin
Study in Sweden
n: 32 twin pairs
and 12 individual
twins

Monozygotic and
dizygotic twins
discordant for
ASD; discordance
defined as >2
points differences
on the Autism-Tics,
ADHD and other
Comorbidities
subscale

Shed deciduous teeth, validated
by maternal, cord, and serial
child blood Pb; laser ablation
ICP-MS

Age at measurement: estimating
various timepoints from 20 wk
prenatal to 30 wk postnatal

Mean NR

ASD diagnosis

ADOS-2, SRS-2
among discordant twins
for ASD (ICD10 [Autism
or Asperger's]; DSM-5
[ASD])

Age at outcome:

8-12 yr

Genetic factors

Child sex, zygosity,
gestational age, the
average birth weight of
the twin pairs, and the
SD of the birth weight
in the twin pairs.

OR of log-transformed Pb for
ASD case vs. non-ASD twin
control: 1.5 (0.9, 2.5)bd

More quantitative results
depicted graphically (see
Figure 3-2)

tSkoaheim et al.
(2021)

Nationwide
Norway

2002-2009
(enrollment)

Case-control

Norwegian Mother,
Father and Child
Cohort Study
(MoBa)

n: 397 ASD cases,
1034 controls

Children from a
birth cohort

Blood

ASD diagnosis

Maternal whole blood; ICP-SFMS NPR

Age at measurement:
wk 17 of gestation

Exposure Quartiles:
Q1: 0.16-0.65 pg/dL
Q2: 0.65-0.86 pg/dL
Q3: 0.86-1.12 pg/dL
Q4: 1.12-8.24 pg/dL

Age at outcome: NR

Birth year and child
sex-matched controls

Child sex, birth weight,
birth year, and SGA,
maternal age at
delivery, education,
parity, pre-pregnancy
BMI, kg/m2), self-
reported smoking and
alcohol intake during
pregnancy, FFQ-
based estimates of
seafood intake (g/day),
and dietary iodine
intake (pg/day)

OR for In-transformed Pb

Q1:

Ref.





Q2:

0.80

(0.57,

1.12)'

Q3:

0.79

(0.56,

1.12)'

Q4:

0.81

(0.57,

1.15)'

3-442

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tRahbar et al. (2015)

Jamaican Autism
Study

Blood

ASD diagnosis

Age, sex-matched
controls

GMD for In-transformed Pb
(ASD Cases vs. Controls):

Kingston

n: 100 cases; 100

Child venous blood; ICP-MS

DSM-IV-TR criteria,





Jamaica

controls



ADOS

maternal age, parental

-0.17 (-0.86, 0.52)bc



Children 2-8 yrat

Age at measurement:



education levels,

December 2009-

2-8 yr

Age at outcome:

parish at child's birth,



March 2012

enrollment

2-8 yr

SES (i.e., car



(enrollment)



GM (SD) (cases): 2.25 (2.23)
pg/dL



ownership by the
family), consumption



Case-control



GM (SD) (controls): 2.73 (1.85)



of shellfish (lobsters,







pg/dL



crabs), and Teflon use









(pots, pans, and
dishes) for cooking



tRahbar et al. (2021)

n: 30 cases; 30
controls

Blood

ASD diagnosis

Age, sex-matched
controls

GMD for In-transformed Pb
(ASD Cases vs. Controls):

Karachi



Child venous blood; ICP-MS

DSM-IV-TR criteria,





Pakistan

children at clinics



ADOS

maternal age, parental

-1.37 |jg/dL (-3.28, 0.54)bc



affiliated with Aga

Age at measurement:



education level, and

Study years NR

Khan University

2-8 yr

Age at outcome:

SES (i.e., car







2-12 yr

ownership by the



Case-control



GM (cases): 7.11 |jg/dL;
GM (controls): 8.48 |jg/dL



family) and dietary
consumptions



dummy variables that
represented the
matched pairs

3-443

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tDonq et al. (2022)

Northeast China

October 2017-
January 2020
(enrollment)

Case-Control

n: 512 children with Blood
ASD

Children diagnosed
with ASD at First
Hospital of Jilin
University

Child serum

Age at measurement: 2-13 yr
Mean (SD)

Mild Autism: 2.58 (1.08) pg/dL
Moderate/severe: 2.58 (1.08)
pg/dL

ASD severity

Severity of autism
symptoms determined
by CARS

Age at outcome 2-13 yr

Age, place of
residence, caregivers,
parental education
level, gastrointestinal
problems.

Also considered sex,
siblings, parental age
at pregnancy,
household income,
family history of mental
illness, vitamin intake
during pregnancy,
eating problems,
sleeping problems,
gastrointestinal
problems, ADHD
comorbidity

Beta

0.03 (0.01, 0.05)c

tRyqiel et al. (2021) ELEMENT project Blood

Orientation/engagement Maternal IQ (WAIS),

Mexico City
Mexico

1997-2005
(enrollment)

Cohort

n: 85

Mother-child pairs
recruited at the
Mexican Social
Security Institute

Maternal and child venous blood;
ICP-MS, GFAAS

Age at measurement:

T1, T2, T3 (maternal); 12, 24 mo
(child)

Maternal blood GM (SD):
T1: 5.27 (1.93) pg/dL
T2: 4.74 (1.96) pg/dL
T3: 4.98 (1.93) pg/dL

and emotional
regulation

ORIEN and EMOCI
scores from BRS of
BSID-IIS

Age at outcome: 12-24
mo

maternal age, infant
weight, length, SES,
infant age and sex,
current infant BLL

A large number of results were
obtained from the mediation
analysis. In summary, T2 BLLs
were consistently inversely
associated with 24-month
EMOCI and ORIEN scores

Beta

24-month EMOCI at T2:
-1.13% (-2.63, 0.37)
24-month ORIEN at T2:
-0.98% (-2.83, 0.88)

Infant blood GM (SD):
12 mo: 3.92 (1.80) pg/dL
24 mo: 3.49 (1.93) pg/dL

3-444

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tShekhawat et al.
(2021)

Western Rajasthan
India

2018-2019
(enrollment)

Followed through 6.5
mo (average)

n: 117

Mother-child pairs
in third trimester or
at delivery

Blood

Cord blood; ICP-OES

Age at measurement:

Delivery

GM = 4.14 |jg/dL; mean = 4.77 ±
3.3 |jg/dL; median = 4.23 |jg/dL
75th: 5.1 |jg/dL

Social-emotional
development score
using BSID-III

Age at outcome: 6.5 mo

Maternal age, gravida,
gestational age,
maternal education,
child sex and weight,
preterm birth, maternal
food intake during
pregnancy, smoking,
alcohol consumption,
maternal residential
and occupational
history, delivery type

13 (95 % CI) for socio-emotional
development scores

Pb < 5 |jg/dL: 0.19 (-0.46,

0.46)

Pb 5.0-10.5 |jg/dL: -0.05
(-0.60, 0.86)

Cohort

tNozadi et al. (2021)

Navajo Nation
United States

February 2013-June
2018 (enrollment)
Followed through 10-
13 mo

Cohort

Navajo Birth
Cohort Study
(NBCS)

n: 327

Children of
mothers (age 14-
45 yr) living across
Navajo Nation with
community
exposure to metal
mixtures from
abandoned
uranium mines

Blood

Maternal blood,
DRC-MS.

child blood; ICP-

Age at measurement:

Delivery or 36-wk visit (maternal);
10, 13 mo (child)

GM = 0.410 |jg/dL;
0.37 |jg/dL
75th: 0.51 pg/dL
95th: 1.20 pg/dL

median =

Communication and
personal-social domain
scores using the ASQ:I.
Age-adjusted scores.

Age at outcome: 10-13
mo

Age.

Also considered
maternal age, marital
status, maternal
occupation and
education, household
income,

concentrations of
various metals in
urine, blood, and
serum

Beta (95% CI)

Communication: -0.15 (-0.58,
0.28)

Personal-Social: -0.11 (-0.72,
0.50)

3-445

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tLin etal. (2013)
Taipei, Taiwan

April 2004-Jan 2005
(enrollment)

Followed through 2 yr

Panel Study

TBPS
n: 230

Singleton full-term
children of non-
smoking mothers
without
occupational
exposure attending
medical center,
hospital, and
clinics in Taipei

Blood

Maternal blood, cord blood; ICP-
MS, measured for Pb, Mn, As,
and Hg.

Pb categories:

Low: <16.45 |jg/L
High: >16.45 pg/L
Mn categories:

Low: <59.59 pg/L
High: >59.59 pg/L

Age at measurement:

Delivery

Mean: 13 pg/L, GM: 10.61 pg/L
75th: 16.45 pg/L
Max: 43.22 pg/L

Social and self-help
ability DQs

CDIIT

Age at outcome:
2 yr

Maternal age,
maternal education,
child sex,

environmental tobacco
smoke during
pregnancy and after
delivery, fish intake,
and HOME Inventory
score

Beta
Social

High vs. Low Pb: -5.89
(-10.81, —0.97)c
High Mn x low Pb: 2.83
(-3.442, 9.102)c

Low Mn x high Pb: -2.9
(-9.231, 3.431 )c
High Mn x high Pb: -7.01
(-14.144, 0.124)c

Self-help

High vs. Low Pb: -1.26
(-5.905, 3.385)c

High Mn x low Pb: 0.49

(-5.429, 6.409)c

Low Mn x high Pb: 0.35

(-5.608, 6.308)c

High Mn x high Pb: -2.38

(-9.103, 4.343)c

tNvanza et al. (2021)

Northern Tanzania
Tanzania

2015-2017
(enrollment)
Followed through 12
mo

Cohort

Mining and Health
Prospective
Longitudinal Study
in Northern
Tanzania

n: 439

Birth cohort of
mother-child pairs
recruited in 2nd
trimester

Maternal dried blood spots; ICP-
MS, measured for Pb, Hg, and
Cd

Age at measurement:
second trimester

Median: 2.72 pg/dL
75th: 4.25 pg/dL
Max: 14.5 pg/dL

Social development
domain using MDAT.
Scores classified as
normal (>90th
percentile on all items in
the domain or <90th
percentile on one or two
items in the domain) or
impaired (<90th
percentile on more than
two items in the
domain).

Age at outcome:
6-12 mo

Maternal age and
education, maternal
and paternal
occupation, number
siblings under 5 yr at
home, and family SES,
infant sex, age, birth
weight, height and
weight as a proxy for
nutritional status
(covariates with p <
0.20 retained in the
final models)

Prevalence ratio:

Social status development:
1.01 (1.00, 1.02)

3-446

-------
tDohertv et al.
(2020)

New Hampshire
U.S.

2009 to 2014-2019
Cohort

NHBCS

n: 371 (SRS-2);
318 (BASC-2)

Mother-child pairs

Toenails

Maternal and infant toenails;

Median (maternal prenatal): 0.14
|jg/g (SRS), 0.13 pg/g (BASC);
Median (maternal postnatal):
0.10 |jg/g (SRS), 0.11 pg/g
(BASC); Median (infant): 0.35
|jg/g (SRS), 0.37 |jg/g

Composite score
(Social Awareness,
Social Cognition, Social
Communication, Social
Motivation, and
Restricted Interests and
Repetitive Behavior) on
SRS-2.

Adaptive skills
composite on BASC-2;
see also Section 3.5.2.2

Age at outcome:

3 yr old

Maternal age,
maternal BMI, parental
education, maternal
smoking, marital
status, parity, child age
at last breastfeeding,
Healthy Eating Index
score, year of birth,
sex, and age of the
child at testing

Beta per log2-transformed |jg/g
increase in toenail Pb

Total SRS-2

Maternal prenatal: -0.08
(-0.20, 0.04)e
Maternal postnatal: 0.03
(-0.08, 0.13)e
Child: -0.06 (-0.19, 0.06)e

Males

Maternal prenatal: -0.06
(-0.23, 0.11 )e

Maternal postnatal: -0.01
(-0.14, 0.13)e

Child: -0.08 (-0.25, 0.10)e

Females

Maternal prenatal: -0.04
(-0.19, 0.11 )e
Maternal postnatal: 0.07
(-0.08, 0.21 )e

Child: -0.05 (-0.21, 0.11)e

Total Adaptive Skills
Maternal prenatal: -0.06
(-0.19, 0.07)e
Maternal postnatal: 0.08
(-0.03, 0.19)e
Child: 0.08 (-0.06, 0.22)e

Males

Maternal prenatal: -0.01
(-0.19, 0.18)e

Maternal postnatal: 0.08
(-0.07, 0.24)e

3-447

-------
StillyD^kjn* Study Population Exposure Assessment	Outcome	Confounders Effect Estimates and 95% CIs

Child: 0.10 (-0.13, 0.32)e

Females

Maternal prenatal: -0.19
(-0.34, —0.04)e
Maternal postnatal: 0.07
(-0.08, 0.23)e

Child: 0.26 (0.07, 0.45)e

tZhou etal. (2017)

Shanghai
China

2010-2012
Followed through 24-
36 mo

Cohort

Shanghai Stress
Birth Cohort study
n: 139

Mother-infant pairs
in prenatal clinics
of maternity
hospitals during
mid-to-late
pregnancy

Blood

Maternal whole blood

Age at measurement: wk 28-36
of gestation

GM (95% CI): 3.30 (3.05, 3.57)
pg/dL

Adaptive and social
behavior domain DQs
from GDS

Age at outcome: 24-
mo

¦36

Maternal age at
enrollment, SES,
maternal education,
gestational week, child
sex, birth weight and
age

Beta per log—10 transformed
BLL

Adaptive:

Overall: 3.60 (-3.64, 10.83)b
Low stress: 7.57 (-0.12,
15.27)b

High stress: -17.93 (-35.83,
—0.03)b

Social:

Overall: -6.45 (-15.55, 2.65)b
Low stress: -0.07 (-9.57,
9.44)b

High stress: -41.00 (-63.11,
—18.89)b

3-448

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tRuebner et al.
(2019)

46 centers
U.S.

Study Years: NR
Cross-sectional

CKiD Cohort study
n: 412

Children (age 1-
yr) with mild to
moderate CKD

¦16

Blood

Child venous blood; ICP-MS. The
BLL measurement closest to the
time of neurocognitive testing
was used for analysis
(concurrent).

Age at measurement:

NR; 2, 4, or 6 yr after study entry

Median: 1.2 pg/dL
75th: 1.8 Mg/dL
Max: 5.1 |jg/dL

Adaptive skills,
composite index on the
BASC-2 (see also 3.5.1
and 3.5.2)

The last available test
results were used to
evaluate long-term
effects. Mean time
between BLL and
neurocognitive testing
was 2.3 yr.

Age at outcome:
3, 5, or 7 yr after study
entry

Child age, sex, race,
poverty, and maternal
education

Adjusted BASC-2 results were
not reported because they
were not statistically significant.

tViqeh et al. (2014)

Tehran
Iran

October 2006 -
March 2011
Followed through 36
mo

Cohort

Birth cohort
n: 174

Mother-infant pairs
recruited in first
trimester (8-12
wk).

Blood

Maternal blood, cord blood;
MS

ICP-

Age at measurement:

3 trimesters during pregnancy
and delivery

Mean: 1st trimester: 4.15 |jg/dL,
2nd trimester: 3.44, 3rd trimester:
3.78, umbilical cord: 2.86

Max: 1st trimester: 20.5 |jg/dL,
2nd trimester: 7.5, 3rd trimester:
8.0, umbilical cord: 6.9

Mental development
assessed using the
ECDI by Harold Ireton
(language
comprehension,
expressive language,
gross motor, self-help,
social interaction).
Cutoff point scores for
development delay was
score <20% of that
expected for children's
age.

Age at outcome:

36 mo

Maternal educational,
BMI, family income,
gestational age, birth
weight, birth order (first
born)

OR

Total ECDI:

1.74 (1.18, 2.5)

3-449

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tKimetal. (2018b)

4 cities: Seoul,
Anyang, Ansan and
Jeju
Korea

Pregnancy (2011-
2012) through 24 mo
of age

Cohort

CHECK cohort
n: 140

birth cohort-
pregnant women
recruited from 4
cities in Korea
before delivery,

Blood

Prenatal maternal blood collected
during hospital visit: 2.7 |jg/dL

Cord blood: 1.2 pg/dL

Adaptive behaviors
assessed using SMS

Association was
examined using multiple
linear regression
analysis.

Age at outcome:
13-24 mo

BPA, and phthalates, Associations of blood Pb

maternal age
(continuous), birth
delivery mode
(categorical), monthly
household income
(categorical), child's
sex, and BDI
(continuous) of the
mother, gestational
age (continuous),
primiparous
(categorical), and
pre-pregnancy BMI
(categorical)

concentrations and SQ were
assessed but not reported
because they lacked statistical
significance.

AAS = atomic absorption spectrometry; ADHD = attention deficit/hyperactivity disorder; ADOS = Autism Diagnostic Observation Schedule; ASD = autism spectrum disorder; ASQ =

Ages and Stages Questionnaire Inventory; ASSQ = Autism Spectrum Screening Questionnaire; BASC = Behavior Assessment System for Children; BDI = Beck Depression

Inventory; BLL = blood lead level; BMI = body mass index; BPA = bisphenol A; BRS = behavioral rating scale; BSID = Bayley Scales of Infant and Toddler Development; CARS =

Childhood Autism Rating Scale; CDIIT = Comprehensive Developmental Inventory for Infants and Toddlers; CHECK = Children's Health and Environmental Chemicals in Korea;

CHEER = Children's Health and Environmental Research; CKiD = Chronic Kidney Disease in Children; DQ = development quotient; DSM = Diagnostic and Statistical Manual of

Mental Disorders; GM = geometric mean; ECDI = Early Child Development Inventory; ELEMENT = Early Life Exposure in Mexico to Environmental Toxicants; GDS = Gesell

Developmental Schedules; GFAAS = graphite furnace atomic absorption spectrometry; HOME = Home Observation Measurement of the Environment; ICP-MS = inductively coupled

plasma mass spectrometry; ISAT = Illinois Standard Achievement Test; MAT = Metropolitan Achievement Test; MEAP = Michigan Educational Assessment Program; MDAT =

Malawi Development Assessment Tool; Mn = manganese; mo = month(s); NHANES = National Health and Nutrition Examination Survey; NHBCS = New Hampshire Birth Cohort

Study; NHNPR = Norwegian Patient Registry; NR = not reported; OR = odds ratio; Pb = lead; SD = standard of deviation; SES = socioeconomic status; SGA = small for gestational

age; SMS = Social Maturity Scale; SRS = Social Responsiveness Scale; SQ = social quotient; TBPS = Taiwan Birth Panel Study; wk = week(s); yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a

change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.

bResults are unstandardized because the Pb level distribution data was not available.

The CI was calculated from a p-value and the true CI may be wider or narrower than calculated.

dResults are unstandardized because the log base used for exposure transformation was unspecified in the study.

eResults are unstandardized because the biomarker used for Pb exposure measurement is not blood, tooth, or bone.

tStudies published since the 2013 Integrated Science Assessment for Lead.

3-450

-------
Table 3-14E Epidemiologic studies of exposure to Pb and cognitive function in adults

Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tPower et al. (2014)
Boston, MA, U.S

1993-2008
Cohort

Participants selected from
cohort study (Nurse's Health
Study) and part of case-
control ancillary study

n: 584

Bone, Blood

Bone Pb: K-XRF at the
midtibial shaft and the
patella, blood Pb
concentrations; GFAAS with
Zeeman background
correction in year 1993-
2004

Age at measurement:
registered nurses aged 45-
74 yr

Tibia Pb cone: 10.5 ± 9.7
|jg/g, Patella Pb cone: 12.6 ±
11.7 pg/g.

Blood Pb cone: 2.9 ± 1.9
pg/dL

Cognitive decline

Cognitive decline
assessed using a
telephone battery
of cognitive tests
during 2-4 waves
over the period of
follow-up, 1995-
2008. All 9
cognitive scores
were Z-transformed
with high score
representing better
performance.

Alcohol
consumption,
smoking status,
education,
husband's
education,
menopausal
status/hormone
therapy use,
physical activity,
ibuprofen use,
aspirin use, vitamin
E supplementation,
the % of residential
census tract of
white race/ ethnicity,
and median income
of residential
census track.

Beta (95% Cl)a

Tibia

Verbal Memory
-0.002 (-0.006, 0.003)
Overall Cognition
-0.002 (-0.005, 0)
Patella

Verbal Memory
-0.001 (-0.005, 0.002)
Overall Cognition
-0.001 (-0.004, 0.001)
Blood

Verbal Memory
0.003 (-0.021, 0.027)
Overall Cognition
-0.007 (-0.023, 0.009)

3-451

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tFarooaui et al. (2017)

Boston, MA, U.S.

1993-2007

Cohort

Participants selected from
cohort study (Veterans Affairs
NAS

n: 741 subjects in MMSE and
715 in Global cognition

Bone

Patella (trabecular bone)
and tibia (cortical bone)
bone Pb was measured
using K-XRF spectroscopy in
1993

Age at measurement:
healthy men aged 51-98 yr

Patella Pb cone: 30.6 ±
19.44 |jg/g, and tibia Pb
cone: 21.6 ± 13.33 |jg/g

Changes in
cognition

Cognition was
assessed using the
MMSE, NES2,
CERAD and WAIS-
R during 3-5 visits
over the period of
15 yr of follow-up.

Age at first cognitive
test, past education
level, baseline
smoking status and
alcohol intake.

Beta (95% Clf Pb and MMSE
over time

Tibia

IQR change in Pb -0.051
(-0.137, 0.035)

IQR change in Pb*time -0.007
(-0.018, 0.004)

Patella

IQR change in Pb -0.061

(-0.12, -0.002)

IQR change in Pb*time -0.008

(-0.015, 0)

HR (95% Cl)b

Patella 1.095 (0.993, 1.207)

Tibia 1.033 (0.875, 1.22)

Beta (95% Cl)b Pb and Global
Cognition overtime

Patella -0.119 (-0.247, 0.009)

Tibia -0.137 (-0.318, 0.043)

3-452

-------
tWeuve etal. (2013)

Boston, MA, United
States

2003-2007

Cross-sectional

PD cases confirmed by
movement disorder specialists
using the U.K. Brain Bank
criteria

n: 151 subjects (101 cases
and 50 controls)

Bone

Bone Pb measured using K-
XRF spectrometric estimates
of Pb concentrations in Tibia
and Patella bones.

Age at measurement: cases
and controls (spouses, in-
laws, or friends of the cases)
aged 54-81 yr

Patella Pb cone by age at
cognitive interview
categories:

54-64.9 yr
65-69.9 yr
70-74.9 yr
75-80.9 yr

5.9 ± 10.3 |jg/g
9.2 ± 7.8 |jg/g
7.7 ± 10.5 |jg/g
15.2 ± 10.2 |jg/g

Tibia Pb cone by age at
cognitive interview
categories:

54-64.9 yr
65-69.9 yr
70-74.9 yr
75-80.9 yr

4.4 ± 11.1 |jg/g
8.8 ± 10.5 |jg/g
6.8 ± 8.8 |jg/g
9.2 ±11.5 |jg/g

Cognition function Age at cognitive
assessment, sex,
Cognitive function race, education,
assessed using a smoking history
telephone cognitive
assessment battery
of 9 tests based on
a validated
telephone battery
for assessing age-
related cognitive
decline. Added test
of cognitive
domains that
typically decline in
PD. All 9 cognitive
scores were z-
transformed with
high score
representing better
performance.

Adjusted difference (95% Cl)b

Patella

Telephone interview for
cognitive assessment (TICS)

-0.08 (-0.32 to 0.15)

Delayed 10-word recall
0.05 (-0.18 to 0.28)

Delayed 10-word recognition
0.01 (-0.22 to 0.24)

Animal naming
-0.11 (-0.32 to 0.10)
"F" naming
-0.07 (-0.30 to 0.17)

Digit span forward
-0.02 (-0.27 to 0.22)

Digit span backward
0.05 (-0.17 to 0.27)

Oral trails B minus A
0.03 (-0.23 to 0.28)

Global score
-0.01 (-0.14 to 0.13)

Tibia

Telephone interview for
cognitive status (TICS)

-0.20 (-0.40 to -0.00)

Delayed 10-word recall

-0.04 (-0.23 to 0.16)

Delayed 10-word recognition

-0.01 (-0.21 to 0.20)

Animal naming

-0.11 (-0.29 to 0.07)

"F" naming

-0.19 (-0.39 to 0.01)

Digit span forward

3-453

-------
Referenc^and Study	Study Population	Exposure Assessment	Outcome	Confounders Effect Estimates and 95% CIs

-0.23 (-0.43 to -0.03)

Digit span backward
-0.19 (-0.37 to -0.00)

Oral trails B minus A
-0.06 (-0.29 to 0.17)

Global score
-0.13 (-0.25 to -0.01)

tSkerfvina et al. (2015) n: 927

Landskrona and
Trelleborg, Southern
Sweden

1978-2007 followed for
4-12 yr

Cohort

Blood

Between 1978 and 1994, B-
Pb levels were determined
using flame or
electrothermal atomization
atomic absorption
spectrometry; between 1995
and 2007, B-Pb levels were
determined using inductively
coupled plasma mass
spectrometry

Age at measurement: 7-12
yr

IQ assessed for

military

conscription

IQ (measured
logical, verbal,
spatial abilities, and
technical
understanding)
assessed as a part
of military
conscription
examinations.

Age at outcome:
18-19 yr

Age at blood
sampling, sex,
parents' education,
family economy,
and country of birth
of child and parents

Beta (SE)a
IQ

All subjects

-0.127 (-0.209, -0.045)
Blood Pb <50 |jg/L
-0.204 (-0.392, -0.016)

Mean: 34 |jg/L

3-454

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tReuben et al. (2017) Dunedin Multidisciplinary
Health and Development
Dunedin, New Zealand Study

1972/73-2012

Cohort

n: 565

Blood

Graphite fumance atomic

absorption

spectrophotometry

Age at measurement: 11 yr

Mean (SD): 10.99 ±4.63
pg/dL

Full -scale IQ
(other domains
such as verbal
comprehension,
perpetual

reasoning, working
memory,

processing speed)

Cognitive function
assessed using
Wechsler Adult
Intelligence Scale -
IV (WAIS-IV) at the
age of 38 yr.

Childhood IQ scores Change in IQa (95% CI)

(age 7 and 9 yr),
their mothers' IQ
score, and their
socioeconomic
background

Adjusted by sex
-0.394 (-0.669, -0.119)
Fully adjusted
-0.322 (-0.496, -0.148)
Change in perceptual
reasoning3 (95% C)
-0.414 (-0.627, -0.201)
Change in working memory3
(95% CI)

-0.252 (-0.476, -0.028)

Change in socioeconomic
status3 (95% CI)

-0.358 (-0.635, -0.081)

tReuben et al. (2020) Dunedin Multidisciplinary
Health and Development
Dunedin, New Zealand Study

1972/73-2019

Cohort

n: 564

Blood

Furnace atomic absorption
spectrophotometry

Age at measurement: 11 yr

Mean (SD): 10.99 ±4.63
pg/dL

Full-scale IQ and

self-reported

information)

Cognitive
performance
assessed
objectively using
Wechsler Adult
Intelligence Scale -
IV (WAIS-IV) and
subjectively via
informant and self-
reports at the age
of 45 yr.

Childhood IQ scores
(age 7 and 9 yr),
their mothers' IQ
score, and their
socioeconomic
background

Change in IQ3(95% CI)

-0.414 (95% CI: -0.679,
-0.149)

Residualized Change in IQ3
(95% CI)

-0.394 (-0.583, -0.205)

3-455

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tKhalil etal. (2014)

Multicity (6 clinical
sites), U.S.

May 2007 to Nov 2008
Cross-sectional

Population-based cohort study Blood
(MrOS)

n: 445	Venous blood samples

tested for Pb levels using
AAS

Age at measurement: non-
Hispanic Caucasian men
aged >65 yr

Mean (SD)

2.25 ± 1.20 |jg/dL

Cognitive function
assessed using 3
MS

Age at outcome:
>65 yr

Age, education,
smoking, alcohol
consumption and
BMI

Beta (95% Cl)b

Cognitive Function in Adults
3MS

-0.01 (-1.10,1.07)

Cognitive Function in Adults
Trail Making B

2.72 (-7.65, 13.09)

tSouza-Talarico et al.
(2017)

Sao Paulo City,

Brazil

Cross-sectional

n: 125 (104 women and 21
men)

Blood

Venous blood samples
tested for heavy metals (Cd
and Pb) levels using ICP-MS

Age at measurement:
Healthy older adults
between 50 and 82 yr (M =
65.9)

Mean (SD)

2.1 ± 0.970 |jg/dL

MMSE and
Informant
Questionnaire on
Cognitive Decline
used to rule out
cognitive and
functional
impairments.

Age at outcome: 50
and 82 yr (M =
65.9)

Age, sex, income,
education,
hemoglobin,
hematocrit

Betab

WMC Pb 0.106 (AR: 0.057)
BCd x BPb interaction-term and
WMC-0.378 (p< 0.001)

Table 3 adjusted EE for Pb and
WMC, with and without
controlling for Oxygen Radical
Absorbance Capacity total:
standardized

3-456

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tvan Wiinaaarden et al.
(2011)

Nationwide, U.S.

1999-2008

Cross-sectional

NHANES 1999-2008 (for self-
reported confusion and
memory problems) and
NHANES 1999-2002 (for
DSST)

n: 9526 participants (7277
from NHANES 1999-2008
and 2299 participants from
NHANES 1999-2002)

Blood

Venous blood samples
tested for Pb concentration
using AAS with Zeeman
background correction.

Age at measurement: >60 yr

Blood Pb cone: 2.46 |jg/dL
(range 0.18-54.00 pg/dL)

Cognitive function

Cognitive function
assessed by self-
reported responses
on limitation in
cognitive
functioning, and
DSST (a subset of
the WAIS-III) for
subset of
participants.

Age, sex, ethnicity,
education level,
PIR, self-reported
general health
status

OR (95% Cl)b
1.01 (0.65, 1.56)

Age at outcome:
>60 yr

tPrzvbvlaet al. (2017) NHANES cycles 1999-2000 Blood

Nationwide, U.S.

1999-2002

Cross-sectional

and 2001 -2002;
n: 498

Cognitive function Race/ethnicity, age, Betas per natural log increase in

Blood samples tested for
chemicals (Pb, Cd and
PCBs) concentrations; Pb
and Cd measured using
ICP-MS.

Age at measurement: 60-84
yr

Mean 2.17 pg/dL (95% CI:
2.07, 2.27)

Cognitive function
assessed using the
DSC Module of the
WAIS-III.

Age at outcome:
60-84 yr

education level,
PIR, sex and
smoking status

BLL

Cognitive Functioning

All Participants: -0.10 (-0.20,

-0.01)

Females:

-0.12 (-0.26, 0.01)

Males:

-0.09 (-0.24, 0.06)
Age 60-69:

-0.13 (-0.28, 0.01)
Age 70-74:

-0.08 (-0.2, 0.04)

3-457

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tSasaki and Carpenter NHANES cycles 2011-12 and Blood and Urine

(2022)

Nationwide, U.S.

2011-2014

Cross-sectional

2013-14 and tested for
different sets of chemicals for
different subgroups
n: 3042

Venous blood samples and
urine samples tested for
seven metals and metalloids
(including Pb) using ICP-MS

Age at measurement: 60-80
yr

Blood mean Pb: 19.0 |jg/L
Urine mean Pb: 0.72 |jg/dL

Cognitive function	Age, sex, ethnicity,

education level,

Immediate and	depression,

delayed memory	diabetes, alcohol

assessed using the	consumption, and

CERAD, and	smoking
working memory
assessed using the
DSST.

Age at outcome:
60-80 yr

Beta (95% Cl)b

Blood

CERAD Immediate recall:
-0.58 (-0.91, -0.24)

CERAD Delayed recall:
-0.19 (-0.35, -0.02)

Digit symbol substitution:
-1.08 (-2.12, -0.05)

CERAD immediate recall as a
function of age

60s Years Old Group:

-0.37 (-0.87, 0.13)

>70 Years Old Group:

-0.85 (-1.44, -0.27)

Urine

CERAD Immediate recall:
-0.26 (-0.58, 0.06)
CERAD Delayed recall:
-0.03 (-0.19, 0.13)

Digit symbol substitution:
-1.03 (-2.01, -0.06)

tXiao etal. (2021)

Guangxi, southern
China

Aug 2016-July 2018
Cross-sectional

n: 2879

Blood

Venous blood samples
tested for 22 metals
(including Pb) using ICP-MS.

Age at measurement: >60 yr
Blood Pb: Median: 51.5 |jg/L

Cognitive function Age, gender,

Cognitive function
assessed using the
MMSE.

Age at outcome:
>60 yr

education
attainment, annual
income, BMI,
smoking, alcohol
drinking, insomnia,
and physical activity

Beta (95% Cl)a
Cognitive function
Single-pollutant model
-0.018 (-0.06, 0.023)
Multi-pollutant model
-0.019 (-0.063, 0.025)

3-458

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tMeramat et al. (2017) Neuroprotective Model for
Healthy Longevity among

Malaysia

May 2013 to January
2014

Cross-sectional

Malaysia Older adult
n: 317

Nail

Toenails (clipped from all
toes) assessed for trace
elements (Al, Ca, Cd, Co,
Fe, Pb, Zn, Se, Cu and Cr)
using ICP-MS.

Age at measurement: >60 yr

Pb cone: Cognitive impaired
group (n = 197): 0.55 ± 0.03
|jg/g; and Normal cognitive
group (n = 120): 0.35 ±
0.013 |jg/g

Cognitive
impairment
assessed using
Montreal Cognitive
Assessment - a
Malay version

Age at outcome:
>60 yr

Age, sex, years of
education and
smoking habits

OR (95% Cl)a

Cognitive impairment 2.471
(1.535-3.980)

3-459

-------
Reference and Study
Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95% CIs

tYu et al. (2021)
Nationwide, U.S.
Jan 2015-Sep 2017
Cohort

SPHERL longitudinal study Blood

n: 260 (260: DSST cohort and
168: SCWT cohort) with
baseline and annual follow-up
blood Pb measurements and
neurocognitive function
assessments.

Venous blood samples
tested for Pb concentration
using ICP-MS.

Age at measurement: mean
age 29.4 yr

Blood Pb cone: DSST
cohort: Geo mean: 3.97 (5-
95th percentage interval (PI)
0.90-14.3) |jg/dL at
baseline, 13.4 (PI 3.70-30.3)
|jg/dL and 12.8 (PI 2.80-
29.2) |jg/dL at the first and
second follow-up visits,
respectively.

Cognitive function

Cognitive function
changes assessed
using the DSST
and ST at baseline
and annual follow-
up visits.

Age at outcome:
mean age 29.4 yr

Age, sex, ethnicity,
change in age,
baseline BMI,
changes in body
weight, education,
baseline blood Pb,
baseline
neurocognitive
function test,
baseline values and
changes in smoking
status, total/HDL
ratio, cholesterol
and alcohol
consumption

OR (95% Cl)a
DSST

1.012 (0.997, 1.028)

3MS = Modified Mini Mental State Examination; BrainAGE = Brain Age Gap Estimation; CERAD = Consortium to Establish a Registry for Alzheimer's Disease; DSC = Digital Symbol
Coding; DSST = Digit Symbol Substitution Test; EE = effect estimate(s); K-XRF = K-shell X-ray fluorescence; MMSE = mini mental status exam; MrOS = Osteoporotic Fractures in Men
Study; NAS = Normative Aging Study; NES2 = Neurobehavioral Evaluation System 2; PD = Parkinson's disease; WAIS-III = Wechsler Adult Intelligence Scale, Third Edition; WAIS-R =
Wechsler Adult Intelligence Scale-Revised; WMC = working memory capacity; SPHERL = Study for Promotion of Health in Recycling Lead; SCWT = Stroop Color-Word Test.
aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a change
in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bResult not standardized because data pertaining to the BLL distribution and/or base for the log-transformation were not reported.
tStudies published since the 2013 Integrated Science Assessment for Lead.

3-460

-------
Table 3-15E Epidemiologic studies of Pb exposure and psychopathological effects in adults

Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
CIs

Raian et al. (2007)
Boston, MA, U.S.
1991-2002
Cohort

Veterans Affairs
NAS
n: 1,075

Closed cohort of
male volunteers with
no chronic medical
conditions at entry.
97% white

Bone

Bone Pb measured in the mid-
tibia shaft and patella using K-
XRF

Age at measurement: 21-80 yr
Mean: -67.5 yr old

Mean (SD):

Tibia: 22.1 (13.8) pg/g

Patella: 31.4 (19.6) pg/g

Depression and anxiety

Depressive and anxiety
symptoms were measured using
the BRIEF Symptom Inventory
(depression and anxiety were
determined to be present for
participants that scored 1 SD
above the mean for a normal
population). Participant followed
up was 3 yr.

Age, alcohol
consumption,
education, time
between

assessments, and
cumulative smoking

Anxiety OR (95% Cl)a
Tibia: 1.13 (0.99, 1.29)
Patella: 1.09 (0.99, 1.19)

Depression OR (95% Cl)a
Tibia: 1.11 (0.98, 1.38)
Patella: 1.05 (0.96, 1.16)

Bouchard et al.
(2009)

U.S.

1999-2004
Cross-sectional

NHANES
n: 1,987

Blood

Blood Pb measured in venous
whole blood samples using
ICP-MS

Age at measurement:
20-39 yr old

Geo. mean: 1.24 pg/dL
20th %ile: 0.7 pg/dL
40th %ile: 1.0 pg/dL
60th %ile: 1.4 pg/dL
80th %ile: 2.1 pg/dL

Depression

WHO CIDI was administered.
Major depressive disorder
diagnosed according to DSM-IV
criteria.

Age at outcome: 20-39 yr

Age, sex,
race/ethnicity,
education, and PIR

Major Depressive Disorder

OR (95% Cl)a

Q1
Q2
Q3
Q4
Q5

2.72)

Ref.

1.39 (0.71,
1.28 (0.69, 2.38)
1.41 (0.76, 2.6)
2.32 (1.13, 4.75)

3-461

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
CIs

tPeters et al. (2011)

Boston, MA
U.S

1991-1997(Bone Pb

measurements);

1993-2003

(Psychological

measurements)

Cohort

Veterans Affairs

NAS

n: 412

Closed cohort of
male volunteers with
no chronic medical
conditions at entry.
97% white

Bone

Bone Pb measured in the mid-
tibia shaft using K-XRF
Age at measurement:

Mean: -65.3 yr old

Mean: 20.6 |jg/g

Pessimism and Depression

A subscale of the Life
Orientation Test was used to
assess pessimistic attitudes.
Depressive symptoms were
measured using the BRIEF
Symptom Inventory (depression
was determined to be present
for participants that scored 1 SD
above the mean for a normal
population).

Age at outcome: Mean -68.3 yr

Age, health
behaviors,
childhood SES,
adult SES

Difference in Pessimism
Level on the Life Orientation
Test (95% CI)

0.21 (0.00, 0.43)

tReuben et al.
(2019)

Dunedin, New
Zealand

Enrollment: 1972-73;
Follow-up through
2012

Cohort

Dunedin
Multidisciplinary
Health and
Development Study

Cohort of children 3
yr old at enrollment
followed through 32
yr of age. Study
population was
nationally
representative
(majority white) and
had high rates of
participation and
follow-up.

Blood

Blood Pb measured in venous
blood samples using GFAAS

Age at measurement: 11 yr

Mean: 11.08 pg/dL
(94% above 5 |jg/dL)

General Psychopathology,
Externalizing Symptoms,
Internalizing Symptoms, and
Thought Disorder Symptoms in
Adults

Psychopathology symptoms
were assessed using the
Diagnostic Interview Schedule.
Factor loadings from each of 11
disorders were used to create
hierarchical measures for
psychopathology and each of its
constituent psychiatric spectra

Sex, childhood
maternal IQ, and
family history of
mental illness.

Change in symptom scores
(95%CI) (standardized to a
mean [SD] of 100 [15])a

General Psychopathology
0.27 (0.02, 0.51)

Externalizing Symptoms
0.15 (-0.10, 0.40)
Internalizing Symptoms
0.28 (0.04, 0.53)

Thought Disorder
0.26 (0.01, 0.51)

Age at outcome:
and 38 yr

18, 21, 26, 32,

3-462

-------
Reference and
Study Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
CIs

tMcFarlane et al.
(2013)

Port Pirie, Australia

1979-1982
(enrollment); 2008-
2009 (follow-up)

Cohort

Port Pirie cohort
Study
n: 210

Mother-singleton
infant pairs enrolled
in Pb-smelting town
from 1979-1982.
Assessed
periodically from
birth to 7 yr, again
from 11 to 13 yr, and
for this study, at 25
to 29 yr

Blood

Blood Pb measured in capillary
blood samples using GFAAS
Age at measurement:

6, 15, and 24 mo; 3-7 yr

Mean: 17.2 |jg/dL
(birth to 7-yr average)

Drug and alcohol abuse,

DSM-IV Disorders (Alcohol
abuse, drug abuse, social
phobia, specific phobia, PTSD,
alcohol dependence, panic
attack, major depressive
disorder) and adult self-report
DMV-IV oriented subscale
(anxiety, somatic problems,
depressive problems,
hyperactivity, inattention,
antisocial personality problems,
avoidant personality problems)

Age at outcome:

25 to 29 yr

HOME, maternal
education, paternal
occupation,
mothers' age at
birth, breastfeeding,
and single parent
family status

OR (95% Cl)a

Social Phobia

Women: 1.05 (0.93, 1.188)

Men: 0.96 (0.80, 1.15)

Specific Phobia
Women: 1.13 (0.99, 1.29)
Men: 1.02 (0.71, 1.47)

Major Depressive Disorder
Women: 0.89 (0.77, 1.03)
Men: 0.89 (0.68, 1.16)

3-463

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
CIs

tLi etal. (2017)

n: 1,701

Shanghai (inner and Stratified cluster

outer districts)
China

2010

Cross-sectional

sampling of
pregnant women
(gestational wk 28-
36)

Blood

Maternal stress

Blood Pb measured in venous Life Event Stress Scale for

blood samples using GFAAS
Geo mean: 3.97 |jg/dL
Max: 14.84 pg/dL

Pregnant Women, Symptom
Checklist-9-Revised (GSI
[measure of psychological
distress], anxiety and
depression scores)

Age at outcome:

13-42 yr old (wk 28-36)

Maternal age at
enrollment,
ethnicity, maternal
education, and
family monthly
income, years
residing in Shanghai Maternai stress

Change in maternal stress,
anxiety, and depression
scores per 10-fold increase in
BLLs (results from piecewise
linear models)3

<2.57 |jg/dL: 0.22 (0.05, 0.4)
>2.57 |jg/dL: -0.07 (-0.16,
0.01)

Depression

<2.57 |jg/dL: 0.34 (0.12,
0.56)

>2.57 |jg/dL: -0.09 (-0.19,
0.02)

Anxiety

<2.57 |jg/dL: 0.25 (0.04,
0.46)

>2.57 |jg/dL: -0.08 (-0.18,
0.02)

tlshitsuka et al.

Japan Environment

Blood

Maternal Depression

Age, parity, marital

OR (95% Cl)b

(2020)

and Children's Study





status, education,





n: 17,267

Blood Pb measured in whole

K6. Depression measured as

employment status,

K6 >13

Japan



blood samples using ICP-MS

scores >5 or 13 (two cutoff

household income,

1.00 (0.76, 1.32)



Pregnant women

Age at measurement:

points for sensitivity).

and smoking and

2011-2014

recruited out of 15

31 yr (mean)



alcohol status



regional centers



Age at outcome:



K6 >5

Cross-sectional

across Japan

Geo. mean: 0.58 |jg/dL
Max: 6.75 |jg/dL

Mean age: 31 yr



0.98 (0.88, 1.09)

3-464

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
CIs

tBerk et al. (2014)

NHANES
n: 15,140

Blood

Depression

Age, sex, poverty,
family income,

Depression OR (95% CI)

U.S.



Blood Pb measured in venous

Depression measured as >9 on

ethnicity, and

Q4 vs. Q1*:



General population,

whole blood samples using

the nine-item depression

country of birth

0.98 (0.78, 1.25)

2005-2010

>18 yr old

ICP-MS

module of the Patient Health



Age at measurement:

Questionnaire









>18 yr old
Mean: NR





*Quartile levels NR

Cross-sectional



Age at outcome:
>18 yr old





tNauven et al. (2022) KNHANES
n: 16,371

South Korea

2009-2013 and
2016-2017

Cross-Sectional

General population;
mean age: 42.6 yr
old (SD: 18.12)

Blood

Blood Pb was measured in
venous whole blood using
GFAAS

Age at measurement (mean):
42.6 yr old (SD: 18.12)

Geo. Mean:

1.84	|jg/dL (w/o depression)

1.85	|jg/dL (w/ depression)

Depression

Self-reported physician's
diagnosis or treatment for
depression

Age at outcome (mean): 42.6 yr
old (SD: 18.12)

Sex, urbanicity,
household income,
physical activity,
occupation, BMI,
alcohol
consumption,
education level, and
smoking status

OR (95% Clf
1.02 (0.90, 1.16)

3-465

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
CIs

tEumetal. (2012)

Boston, MA
United States

Subsample 1: Bone
Pb Measure 1993-
1995;

Subsample 2: Bone
Pb Measure 2001-
2004.

Psychological
Questionnaires:
1988, 19992,1996,
2000, 2004

Cohort

Nurses' Health
Study
n: 617

Women from two
subsample studies
of the NHS cohort

Bone

Midtibial shaft and patella bone
Pb measured using K-XRF
Age at measurement:

Mean: 60.9 yr

Mean:

Tibia: 10.3 |jg/g;

Patella:12.5 |jg/g

Tibia Tertiles:

T1: <7.0 |jg/g
T2: 7.0-11.5 pg/g
T3: >11.5 pg/g

Patella Tertiles:
T1
T2
T3

<8.5 pg/g
8.5-14.5 pg/g
>14.5 pg/g

Phobic anxiety and depressive
symptoms

Depression symptoms
measured using MHI-5; Anxiety
symptoms measured using
phobic anxiety scale of the
Crown-Crisp Experiential Index
(CCEI)

Age at outcome:

Mean:

MHI-5: 59.4 yr
CCEI: 59.2 yr

Substudy group,
age at bone Pb
measure, age at
MHI-5 or CCEI
measurement,
education,
husband's
education, alcohol
consumption, pack-
years of smoking,
and employment
status at MHI-5 or
CCEI assessment

OR (95% Cl)b (Tertile 3 vs.
Tertile 1)

CCEI >4
All women

Tibia: 1.10 (0.73, 1.64)
Patella: 0.75 (0.49, 1.15)

Women on HRT
Tibia: 2.79 (1.02, 7.59)
Patella: 0.23 (0.07, 0.69)

MHI-5 Point Difference
(lower scores indicate worse
symptoms)

All women (T3 vs. T1)

Tibia: -1.06 (-3.05, 0.94)

Patella: -7.78 (-11.73,
-3.83)

Women on HRT (T3 vs. T1)
Tibia: 0.61 (-1.55, 2.78)
Patella: 0.51 (-3.91, 4.94)

3-466

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and 95%
CIs

tFan et al. (2020) Cohort Study of

Luan city, Anhui
province, China

2016

Cross-sectional

Elderly Health and
Environmental
Controllable Factors
n: 994

Older adults (>60 yr
old) selected using
cluster sampling
from two
communities in
Luan, China

Blood

Blood Pb measured in venous
whole blood samples using
ICP-MS

Age at measurement:
>60 yr old

Quartiles

Q1
Q2
Q3
Q4

<2.03 |jg/dL
2.03-2.68 |jg/dL
2.68-3.06 |jg/dL
>3.06 |jg/dL

Depressive symptoms

Chinese revision of the geriatric
depression scale

Age at outcome:

>60 yr old

Age, gender, region, OR (95% Cl)b
marital status,
monthly income,
education level,
alcohol intake,
smoking, and BMI

Depression

Q1
Q2
Q3
Q4

Ref.

1.28 (0.79,
1.36 (0.84,
2.03 (1.23,

2.08)
2.22)
3.35)

tMaetal. (2019)

Hebei Province
China

2018-2019

Case-control

n: 190 (95 cases,
controls)

95 Blood

First-episode drug-
naive patients ages
18 to 60 yr old were
recruited from a
psychiatric hospital.
Age and sex-
matched controls
without known
psychiatric problems
recruited from an
affiliated hospital

Serum Pb measured in venous
blood samples using ICP-MS
Age at measurement:

18-60 yr old

Median: 0.61 ng/mL (serum)
75th: 0.79 ng/mL (serum)

Schizophrenia

Physician-diagnosed
schizophrenia using ICD-10
criteria

Age at outcome:

18-60 yr old

Marital status
(others not
specified).
Population matched
on age and sex

OR (95% Cl)b per 1 ng/mL

increase

3.15 (1.24, 7.99)

AAS = atomic absorption spectrometry; BLL = blood lead level; BMI = body mass index; CCEI = Crown-Crisp Experiential Index; CI = confidence interval; CIDI = Composite
International Diagnostic Interview; GFAAS = graphite furnace atomic absorption spectrometry; K6 = Kessler Psychological Distress Scale; K-XRF = K-shell X-ray fluorescence; MHI-5
= Mental Health Index 5-item; NAS = Normative Aging Study; NR = not reported; Pb = lead; PIR = poverty-income ratio; PTSD = post-traumatic stress disorder; Q = quartile; SD =
standard deviation; SES = socioeconomic status; WHO = World Health Organization; wk = week(s); yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bResult not standardized because data pertaining to the BLL distribution and/or base for the log-transformation were not reported.

tStudies published since the 2013 Integrated Science Assessment for Lead.

3-467

-------
Table 3-16E Epidemiologic studies of Pb exposure and sensory organ function in adults

RefereDCesfgnnd	Study Population Exposure Assessment	Outcome	Confounders Effect f|o^eS a"d

Park etal. (2010)

Eastern Massachusetts
U.S.

Enrollment and outcome
assessment: 1962-1996;
bone Pb measurements:
1991-1996

Cohort

NAS
n: 448

Bone

Bone Pb levels measured
in the midtibial shaft and
patella with a K-XRF
instrument.

Age at measurement:
Mean (SD) at bone Pb
measurement = 64.9 (7.3)
yr; mean (SD) at first
audiometric test = 42.5
(8.4) yr

Mean (SD) in tibia = 22.5
(14.2) |jg/g; mean (SD) in
patella = 32.5 (20.4) pg/g

Sensory Organ Function

Pure-tone averages
assessed by audiologists
with the modified Hughson-
Westlake procedure. Air
conduction hearing
thresholds measured for
each ear by audiologists
using either a Beltone 15C
or a Grason-Stadler 1701
audiometer.

Cross-sectional
analyses and logistic
regression analyses
adjusted for age,
race, education, BMI,
pack-years of
cigarettes, diabetes,
hypertension,
occupational noise,
and noise notch.

Hearing loss OR (95%
Cl)b

Tibia 1.19 (0.92, 1.53)

Patella 1.48 (1.14, 1.91)

EE in Hearing thresholds
(dB HL) with one
interquartile range
Increment in bone lead
measure

Tibia PTA 0.83 (-0.18,
1.83)

Patella PTA 1.58 0.62,
2.55

tShiue (2013)
U.S.

2003-2004
Cross-sectional

NHANES

n: 712 (vision); 732
(hearing); 669
(balance)

NHANES age 50 and
above

Urine

Urinary Pb was detected
by mass spectrometry
Age at measurement:
50 yr

Not Reported

Vision: excellent, good, and
fair eyesight (self-reported)
were classified as good;
poor and very poor were
classified as poor

Hearing: good and little
trouble hearing (self-
reported) were classified as
good; lots of trouble and
deaf were classified as poor

Age, sex, ethnicity,
urine creatinine,
survey weighting

OR (95% Cl)b
Vision 1.15 (0.67-1.97)
Hearing 0.97 (0.63-1.51)
Balance 0.68 (0.51-0.91)

Balance: "During the past 12
mo, have you had dizziness,
difficulty with balance, or
difficulty with failing?"

3-468

-------
Referenc^and Study study Population Exposure Assessment	Outcome	Confounders Effect Estimates and

Ear ringing: "ears ringing,
roaring, or buzzing in the last
year"

Age at outcome:

50 yr

tKana et al. (2018)
Korea

2010-2013

Cross-sectional

KNHANES
n: 6409

Representative sample
of the entire Korean
population. Study
participants were at
least 20 yr old and
underwent pure-tone
audiometry and blood
Pb test.

Blood

Blood Pb was measured
using GFAAS and
classified into quartiles by
sex

Age at measurement:
20-87 yr (mean ± SE:
47.1 ± 0.3 yr)

Weighted mean ± SE

(Men):

|jg/dL;

|jg/dL;

|jg/dL;
pg/dL;

Weighted mean ± SE
(Women): Q1 = 1.12 ±

Q1 = 1.56 ± 0.01
Q2 = 2.22 ±0.01
Q3 = 2.82 ±0.01
Q4 = 4.22 ± 0.08

0.01 |jg/dL
0.01 |jg/dL
0.01 |jg/dL
0.03 |jg/dL

Q2 = 1.61 ±
Q3 = 2.11 ±
Q4 = 3.03 ±

Low-frequency hearing
impairment;

High-frequency hearing
impairment

Pure-tone audiometry was
performed on both ears at
0.5, 1, 2, 3, 4, and 6 kHz. A
binaural pure-tone average
threshold was used and two
binaural averages were
computed, one across 0.5, 1,
and 2 kHz and the other
across 3, 4, and 6 kHz to
determine the low- and high-
frequency thresholds.
Hearing impairment was
then determined according
to whether an average
threshold exceeded 25 dB in
the respective frequency
band.

Age at outcome:

20-87 yr (mean ± SE: 47.1 ±

0.3 yr)

Age, BMI, education,
smoking, alcohol
consumption,
exercise, diabetes
mellitus,

hypertension, noise
exposure

OR (95% Cl)b

Hearing Loss - Low
Frequency

Females

Q2
Q3
Q4

1.271 (0.726, 2.224)
1.308 (0.784, 2.183)
0.932 (0.541, 1.605)

Males

Q4
Q3
Q2

1.026 (0.813, 1.295)
1.028 (0.661, 1.598)
1.17 (0.772, 1.773)

Hearing Loss - High
Frequency

Females

Q2: 0.947 (0.608, 1.475)

Q3: 1.013 (0.698, 1.471)
Q4: 1.502 (1.027, 2.196)

Males

Q2: 1.368 (1.006, 1.86)
Q3: 1.402 (1.005, 1.955)
Q4:

1.629 (1.161, 2.286)

3-469

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tChoi and Park (2017)

Korea National Health and
Nutrition Examination Survey
(KNHANES), Korea

2010-2012
Cross-sectional

KNHANES
n: 5187 adults

Blood

Measured using Graphite
furnace atomic absorption
spectrometry

Age at Measurement: 20-
87 yr

90th: Adults: Geometric
mean (age-adjusted) 2.12
|jg/dL (95% CI: 2.08, 2.15)
Adolescents: Geometric
mean (age-adjusted) 1.26
|jg/dL (95% CI: 1.22, 1.30)

Hearing loss (>25dB) at
speech frequency;

Hearing loss (>25dB) at high
frequency

Pure-tone air conduction
hearing thresholds were
obtained for each ear at
frequencies of 0.5, 1, 2, 3, 4,
and 6 kHz over an intensity
range of-10 to 110 dB
-10 to 110 dB.

Age at outcome: 20-87 yr

Adjusted for age;
age squared; sex;
education; BMI;
current cigarette
smoking; current
diagnosis of
hypertension and
diabetes; and
occupational,
recreational, and
firearm noise
exposures

OR (95% Cl)b

Hearing Loss (>25 dB)
High-frequency PTA
Pb Quartile 2 (1.594-
2.146):

1.13 (0.83, 1.53)

Pb Quartile 3 (2.48-
2.822):

1.35 (1, 1.81)

Pb Quartile 4 (2.823-

26.507):

1.7 (1.25, 2.31)
Per doubling of Pb:
1.3 (1.08, 1.57)
Speech-Frequency PTA
Pb Quartile 2:

0.94 (0.65, 1.35)
Pb Quartile 3:

1.29 (0.92, 1.78)
Pb Quartile 4:

1.25 (0.87, 1.79)
Per doubling of Pb:
1.15 (0.94, 1.41)

tWanq et al. (2020)

Zhejiang Province
(Hangzhou, Jiangshan,
Tonglu, Jiaxing, Anji,
Jinyun), China

2016 to 2018
Case-control

n: 2016

Blood

Measured by graphite
furnace atomic absorption
spectrometry

Age at Measurement:
21-89 yr

Hearing loss

The devices utilized in this
research were an
audiometer (AT235,
Interacoustics AS, Assens,
Denmark) and standard
headphones (TDH-39,
Telephonies Corporation,
Farmingdale, USA)

Income, education,
hypertension,
diabetes,

hyperlipidemia, otitis
media, migraine,
anemia, smoking,
alcohol consumption,
daily fruit and
vegetable intake,
and workplace noise
exposure

OR (95% Cl)b Q1 Ref
Q2 1.135 (0.806, 1.599)
Q3 1.038 (0.731, 1.475)
Q4 1.016 (0.7, 1.475)

3-470

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Logarithmic-transformed
levels of Pb

Case group (1.58 ± 0.17
|jg/dL) and control group
(1.57 ± 0.16 Mg/dL)

Age at outcome: 21-89 yr

3-471

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tChoi etal. (2012)

NHANES, U.S.

1999-2004
Cross-sectional

NHANES
n: 3698

Blood

Simultaneous
multielement atomic
absorption spectrometer
(SIMAA 6000;

PerkinElmer, Norwalk, CT)
with Zeeman background
correction

Age at Measurement:
20-69 yr

Age-adjusted geometric
mean (95% CI) = 1.54
|jg/dL (1.49, 1.60)

Hearing threshold;

Hearing loss

Pure-tone air conduction
hearing thresholds were
obtained for both ears at
frequencies of 0.5-8 kHz
over an intensity range of-
10 to 120 dB.

Age at outcome: 20-69 yr

Age and age2, sex,
race/ethnicity [non-
Hispanic white
(reference), Mexican
American, non-
Hispanic Black,
other], education [<
high school
(reference), high
school, > high
school], BMI
(continuous),
ototoxic medication
use (yes/no),
cigarette smoking
[never smoker
(reference), < 20
pack-years, > 20
pack-years],
hypertension
(yes/no), type 2
diabetes (yes/no),
and either blood lead
or blood cadmium
(for the

corresponding
cadmium or lead
model), occupational
noise exposure
(0*NET score,
continuous),
nonoccupational
firearm noise
(yes/no) and any
recreational noise
(yes/no)

OR (95% Cl)b
Hearing Loss:

Quintile 2 (0.90-1.30
pg/dL)

1.08	(0.55, 2.12)

Quintile 3 (1.4-1.8 pg/dL)
1.1 (0.58, 2.05)

Quintile 4 (1.90-2.70
Mg/dL)

1.21 (0.67, 2.22)

Quintile 5 (2.80-54
MQ/dL)

1.36 (0.75, 2.48)
Per doubling of Pb

1.09	(0.95, 1.26)

3-472

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tYin etal. (2021)

Iran, Korea, China, United

States

Other

n: 234-7596 in 8
studies

Blood

Age at measurement:
3-87 yr

Hearing loss

All studies included
in the meta-analysis
controlled for age
and sex. Adjustment
for other potential
confounders varies
by studies, but
includes monthly
income, education
levels, smoking
status, BMI,
ethnicity, work
duration, ototoxic
medication, blood
lead, occupational
noise, loud noise,
and firearm noise,
and hypertension
and diabetes

OR (95% Cl)b
1.34 (1.18, 1.52)

tTu et al. (2021)
NHANES, U.S.
2011-2012
Cross-sectional

NHANES
n: 1503

Blood

Measured by plasma
mass spectrometry

Age at measurement:
20-69 yr

Median = 1.07 |jg/l
95th: 1.62 pg/l

Speech-frequency hearing
loss;

High-frequency hearing loss

For each ear, 0-5, 1, 2, 3, 4
and 6 kHz frequencies

were used for assessing
pure-tone air conduction
hearing thresholds

over a-10 to 110 dB
intensity ranges. The
average of four

audiometric frequencies
(0-5, 1, 2 and 4 kHz) was
used to identify

Age, sex, education,
marital status, BMI,
smoking, noise
exposure,
hypertension
and diabetes

OR (95% Cl)b
HFHL 1 98 (1-27, 3 10)
SFHL 1.46 (0.81, 2.64)

3-473

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

speech-frequency hearing
loss (SFHL), while the
average of three audiometric
frequencies (3, 4 and 6 kHz)
was used to identify

high-frequency hearing loss
(HFHL). SFHL or HFHL >25
dB in either ear was sued to
define hearing loss, based
on the WHO definition for
this condition

Age at outcome: 20-69 yr

tPaulsen et al. (2018)

Beaver Dam Offspring Study
Beaver Dam, Wisconsin,

U.S.

Baseline data collection
June 8, 2005, through
August 4, 2008 with two
follow-up examinations
occurred at 5-year intervals:
one was conducted between
July 12, 2010, and March 21,
2013, and the other between
July 1, 2015, and November
13, 2017

BOSS
n: 1983

Blood

Measured by Inductively
coupled plasma mass
spectrometry

Age at Measurement:
21-84 yr

Central tendency BLL: NR

Contrast sensitivity
impairment

Age, alcohol
consumption,
smoking, AMD,
cataract, plaque site,
VA impairment, and
sex

HR (95% CI)b
0.91 (0.696, 1.19)

Cohort

tFillion et al. (2013)

n: 228

Blood

Measured by Inductively
coupled plasma mass
spectrometry (ICP-MS)

Contrast sensitivity (cycles
per degree, cpd);

Acquired color vision loss
(color confusion index, CCI)

Age, sex, current
smoking (yes vs. no),
current drinking (yes
vs. no)

Beta (95% CI)b

Spatial frequency with
%EPA

3-474

-------
Reference and Study
Design

Study Population Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Lower Tapajos River Basin

State of Para

Brazil

Age at Measurement:
15-66 yr (median = 33.0
yr)

Age at outcome: 15-66 yr

1.5 cpd -1.32 (-4.30;
1.65)

3 cpd 2.06 (-2.87; 6.99)
6 cpd 0.60 (-6.04; 7.25)
12 cpd -13.33 (-23.28;
-3.49)

18 cpd -2.43 (-6.64;
1.79)

CCI 0.16 (-0.03; 0.33)

May to July 2006

Mean = 12.8 ± 8.4 pg/dL;
Median = 10.5 |jg/dL

Cross-sectional

BLL = blood lead level; CI = confidence interval; CIDI = Composite International Diagnostic Interview; EPA = eicosapentaenoic acid (; K-XRF = K-shell X-ray fluorescence; NAS =
Normative Aging Study; OR = odds ratio; Pb = lead; PTA = pure tone average; Q = quartile; RR= relative risk; SD = standard deviation; SE = standard error; WHO = World Health
Organization; yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bResult not standardized because data pertaining to the BLL distribution and/or base for the log-transformation were not reported.
tStudies published since the 2013 Integrated Science Assessment for Lead

3-475

-------
Table 3-16T Animal toxicological studies of Pb exposure and sensory organ function

Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Jamesdaniel et al. (2018)

Mouse (C57BL/6)

Control (tap water), M, n =
6

2 mM, M, n = 6

PND 33 to
PND61

Oral,

drinking

water

PND 61:

10 |jg/L (1 pg/dL) for
Control

293 |jg/L (29.3 pg/dL) for
2 mM

PND 61: Auditory threshold (via BAEP)

Carlson et al. (2018)

Mouse (CBA/CaJ)

Control (deionized water),
M, n = 16

0.03 mM, M, n = 8

5 wkto 16
wk

Oral,

drinking

water

16 wk:

-------
Study

Species (Stock/Strain),
n, Sex

Timing of
Exposure

Exposure
Details

BLL as Reported (pg/dL)

Endpoints Examined

Liu etal. (2019)

Rat (Sprague Dawley)

PND 1 to

Oral,

PND 9: PND 93:

Sound-Azimuth Discrimination



Control (tap water), F, n =

PND 21

drinking

Training





12



water

0 pg/dL for Control





58 mg/L, F, n = 11





7.9 pg/dL for 58 mg/L











PND 21:











0 pg/dL for Control,











8.2 pg/dL for 58 mg/L











PND 40:











0 pg/dL for Control











0 pg/dL for 58 mg/L



BAEP = brainstem auditory evoked potentials; BLL = blood lead level; F = female; KNHANES = Korea National Health and Nutrition Examination Survey; M = male; Pb = lead; PND :
postnatal day; wk = week(s).

3-477

-------
Table 3-17E Epidemiologic studies of exposure to Pb and neurodegenerative disease in adults

Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Wang et al. NAS
n: 358

NAS
US

Bone Pb
measurement
(1991-1999),
Mini-Mental State
Examination
(MMSE) twice
(1993-1998 and
1995-2000)

Tibia and patella

Measured by K-XRF

Age at measurement 21-
81 yr

Median: 19 and 23 |jg/g
for tibia and patella

Cognitive decline	Adjusted for age, years of

cognitive assessment battery was education, nonsmoker,
the MMSE, a global examination f°rmer smoker, pack-years,
of cognitive function that
assesses orientation, immediate
and short-term recall, verbal and
written skills, and attention and
ability to follow commands

Age at outcome: 21-81 yr

nondrinker, alcohol
consumption, English as first
language, computer
experience, and diabetes

Change in MMSE score
per IQR (15 |jg/g)
increase in tibia Pb by
class of HFE genotype3
Wild-type -0.02 (-0.10
to 0.07)

One HFE variant allele
-0.14 (-0.33 to 0.04)
Two HFE variant alleles
-0.63 (-1.04 to-0.21)

Cross-sectional

3-478

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Weisskopf et al. NAS

12004}	n: 466

Normative Aging
Study, U.S.

1991 and 2002

Cross-sectional

Tibia, patella, and blood

Bone Pb measured by
ABIOMED K-XRF
instrument and blood Pb
measured by Zeeman
background-corrected
flameless atomic
absorption (graphite
furnace)

Age at measurement 21-
81 yr

Cognitive decline
cognitive assessment battery was
the MMSE, a global examination
of cognitive function that
assesses orientation, immediate
and short-term recall, verbal and
written skills, and attention and
ability to follow commands

-0.25 (-0.45, -0.05)

Age at first MMSE test, Difference in change in
alcohol intake, and days MMSE score per IQR
between the two MMSE tests increase in Pba
as continuous variables, as
well as education (<12 yr, 12
yr, 13-15 yr, >16 yr), smoking
status (never, former,
current), computer
experience (yes/no), and
English as a first language
(yes/no)

Median |jg/g (interquartile
range)

Patella 27 (19, 40) pg/g
Tibia 21 (15, 29) pg/g
Blood 5 (3, 7) pg/dL

3-479

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

(Wright et al.. NAS
2003)	n: 1033

Normative Aging
Study, U.S.

1991-1997

Cross-sectional

Tibia, patella, and blood

Bone Pb measured by
ABIOMED K-XRF
instrument and blood Pb
measured by Zeeman
background-corrected
flameless atomic
absorption (graphite
furnace)

Cognitive decline	Age, alcohol intake, and

cognitive assessment battery was education history

the MMSE, a global examination

of cognitive function that

assesses orientation, immediate

and short-term recall, verbal and

written skills, and attention and

ability to follow commands

OR (95% Cl)a MMSE
<24

Tibia 1.02 (1.00,1.04)

Patella 1.02 (1.00,
1.03)

Age at measurement 21-
81 yr

Mean (SD)

Patella 29.5 (21.2) pg/g
Tibia 22.4 (15.3) pg/g
Blood 4.5 (2.5) pg/dL

3-480

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

(Weuve et al.. NAS
2006)	n: 1171

Normative Aging
Study, U.S.

1991 and 2002

Cross-sectional

Tibia, patella, and blood

Measured by graphite
furnace atomic
absorption with Zeeman
background correction

Age at measurement 21-
81 yr

Median (and first and
third quartiles) of tibia
and patella were 19 (13,
28) and 27 (18, 39) pg/g

Blood 5.2 (<1-28) pg/dl

ALAD genotype modifications on
cognition

cognitive assessment battery was
the MMSE, a global examination
of cognitive function that
assesses orientation, immediate
and short-term recall, verbal and
written skills, and attention and
ability to follow commands

Age at cognitive assessment
and age-squared, years of
education (<8, 9-11, 12, 13-
15, 16, >17 yr), computer
experience (an additional
measure of socioeconomic
status), and length of time
between the lead and
cognitive assessments, were
smoking status (current v
past or never), alcohol
consumption (none, 0.1-4.9
g/day, 5.0-9.9 g/day, >10
g/day, or missing), calorie
adjusted calcium intake (in
tertiles), regular energy
expenditure on leisure time
physical activity (in tertiles),
and diabetes (physician
diagnosed or fasting blood
glucose >126 mg/dl)

Mean difference in
MMSE score per IQR
increase in Pb (95%
Cl)a

Tibia

Among ALAD-2
carriers -0.16 (-0.58 to
0.27)

Among ALAD wildtypes
-0.05 (-0.21 to 0.12)

Patella

Among ALAD-2
carriers -0.26 (-0.64 to
0.12)

Among ALAD wildtypes
-0.07 (-0.23 to 0.09)
Blood

Among ALAD-2
carriers -0.26 (-0.54 to
0.01)

Among ALAD wildtypes
-0.04 (-0.16 to 0.07)

(Nordbera et al.. Kungsholmen project
2000)	n: 762

Stockholm

1994-1996

Cross-sectional

Blood

Measured using Graphite
furnace atomic
absorption spectrometry

Age at measurement:
75+ (mean age of 88.4
yr)

Mental Performance
MMSE

Age and BMI

No association was
reported (quantitative
estimate NR)

Mean (SD) 3.7 (2.3) pg/dl

3-481

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tFarooqui et al. Participants selected from Bone

(2017)

Boston, MA,
United States

1993-2007
Cohort

Changes in cognition

cohort study (Veterans
Affairs NAS); healthy men
aged 51-98 yr
n: 741 subjects in MMSE
and 715 in Global cognition

Age at first cognitive test,

Patella (trabecular bone)
and tibia (cortical bone)
bone Pb was measured
using K-XRF
spectroscopy
in 1993

Patella mean (SD) 30.6 ±
19.44 |jg/g, and tibia
mean (SD)21.6± 13.33

pg/g

past education level, baseline
Cognition was assessed using smoking status and alcohol
the MMSE, NES2, CERAD and intake.

WAIS-R during 3-5 visits over the
period of 15 yr of follow-up.

HR (95% Cl)b

Cognition
MMSE < 25
Tibia

1.05 (0.82, 1.35)
Patella

1.21 (0.99, 1.49)

Beta (95% Cl)b

Global Cognition
(Summary score of
NES2, CERAD and
WAIS-R)

Tibia

-0.206 (-0.453, 0.089)
Patella

-0.25 (-0.518, 0.019)

Cognition

MMSE

Tibia

-0.077 (-0.206, 0.052)
Patella

-0.128 (-0.251,
-0.0004)

3-482

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tYanq et al.
(2018)

Multicity
(Taichung city,
Changhua, and
Nantou County),

Taiwan

Feb 2015-Oct
2016

Case-control

Participants were recruited
from the China Medical
University Hospital. Cases
recruited the Department of
Neurology, and controls
from the

Department of Family
Medicine who were
receiving general health
check-up; aged >50 yr.

n: Full sample: 434 (170 AD
and 264 controls);
Propensity-score-matched
sample: 84 AD and 84
controls.

Blood assessed for heavy AD risk

Age, gender, education,

metals

Blood samples tested for
heavy metals (Pb, Cd,
Se, Hg). Blood Pb
measured through ICP-
MS

Blood Pb cone: Full
samples: AD: 2.50 ± 1.35
|jg/dL, Controls: 2.36 ±
1.02 |jg/dL
Propensity matched
samples: AD: 2.58 ± 1.35
|jg/dL, Controls: 2.50 ±
1.18 |jg/dL

exercise habits
Physician AD diagnosis based on hypertension, diabetes
definition of the Diagnostic and
Statistical Manual Fourth Edition
Criteria; MMSE test of cognitive
function

cardiovascular diseases,
depression, anxiety

OR (95% Cl)b

Full population

Total: 1.05 (0.86-1.28)

Tertile 2 vs. 1: 1.00

(0.56-1.79)

Tertile 3 vs. 1: 0.87

(0.49-1.55)

Propensity score-
matched population

Total: 1.06 (0.83-1.35)
Tertile 2 vs. 1: 1.16
(0.55-2.47)

Tertile 3 vs. 1: 1.12
(0.53-2.39)

tHorton et al.
(2019)

Nationwide,
United States

1999-2008
followed till 2014

Cohort

Participants selected from Blood assessed for Pb AD mortality
the five NHANES cycles and
included 1999 to 2008 who
were followed till 2014 for
death; aged >60 yr.

n: 8,080 subjects

Blood samples collected
during the NHANES
mobile examination
center visit assessed for
Pb using ICP-DRC-MS.

Blood Pb cone: Geo
mean and 95% CI: 2.1
(2.02, 2.11) |jg/dL

The identification of AD mortality
is based on the immediate cause
of death in the National Death
Index record. Cause of death was
coded according to the ICD-10,
revision 10; G30 was used to
indicate AD.

Age, sex, poverty status,
race/ethnicity, and smoking
status, and competing risks
for AD mortality.

HRR (95% CI)b
0.3 |jg/dL:
ref

0.5 |jg/dL:

1.1	(0.89, 1.3)

1	|jg/dL:

1.2	(0.77, 1.8)
1.5 |jg/dL:

1.2	(0.7, 2.1)

2	|jg/dL:

1.3	(0.66. 3)

3	|jg/dL:

1.3	(0.6, 3.0)
5 |jg/dL:

1.4	(0.54, 3.8)

3-483

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

(Vinceti et al.. n: 15 cases and 36 controls Blood
1997)

ALS measured by ALS severity Patients and controls

Correlation coefficient

scale

Santa Mafia
Nuova Hospital
in Reggio Emilia,
northern Italy

December 31st,
1994

Case-control

Age at measurement:
(mean ±SD)

Patients 65.9 ± 14.0 yr
Controls 64.4 ± 12.9

Mean (SD)

Controls 108.3 ± 44.4

pg/i

Patients 127.1 ± 67.8 pg/l

matched on year of birth and ALSSS (p-value)b
gender, confounders NR Tota| -0.440 (0.101)

(Kamel et al.
2002)

New England
and U.S.

1993-1996
Case-control

n: 109 cases and 256
controls

Blood and bone

Blood lead was
measured using graphite
furnace atomic
absorption spectrometry

ALS

A board-certified neurologist (T.
L. M. or J. M. S.) evaluated
potential cases. Diagnosis of ALS
was based on criteria published
by the World Federation of

Cases and controls matched
on age, sex, and region

OR (95% Cl)b
Blood 1.9 (1.4, 2.6)
Tibia 2.3 (0.4, 14.5)
Patella 3.6 (0.6, 20.6)

Bone lead was measured Neurology,
using in vivo K-XRF

Age at measurement:

30-80 yr

Mean (SE)

Blood |jg/dl
Cases 5.2 (0.4)
Controls 3.4 (0.4)
Patella |jg/g
Cases 20.5(2.1)
Controls 16.7 (2.0)
Tibia |jg/g
Cases 14.9 (1.6)
Controls 11.1 (1.6)

3-484

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Kamel et al. n: 110
(2008)

New England

and

U.S.

Enrollment: 1993
- 1996; follow-up
through
December 31,
2003

Cohort

Bone and blood

Bone Pb measured in the
tibia and patella using K-
XRF; blood Pb measured
using atomic absorption
spectrometry

Neurodegenerative Disease ¦
ALS

Age at Measurement:
Median (range
79) years

Amyotrophic lateral sclerosis
(ALS) was diagnosed by board-
certified neurologists and based
on the World Federation of
Neurology El Escorial criteria;
related symptoms were
documented from interviews.
60 (30- Cause of death was identified by
the National Death Index (NDI).

Cox proportional hazard
analyses adjusted for age,
sex, and ever smoked,
except for sex-stratified
models, which included age
and ever smoked.

HR (95% Clf
Diagnosis to death 0.9
(0.8 to 1.0)

Symptoms to death 0.9
(0.8 to 1.0)

Blood Pb median = 4
|jg/dL; patella Pb median
= 15 |jg/g; tibia Pb mean
= 13 |jg/g

Max: Blood Pb max = 14
|jg/dL; patella Pb max =
107 pg/g; tibia Pb max =

61 pg/g

(Fang et al..
2010)

U.S.

2003-2007
Case-control

n: 184 cases and 194
controls

Blood

Pb measured by
inductively coupled
plasma mass
spectrometry

Age at measurement:
mean (SD)

Cases 63.3 (34-83)
Controls 63.4 (34-84)

Pb Mean (SD)
3.4 pg/dL (2.5)

ALS

Neurologists with expertise in
ALS reviewed medical records to
determine motor neuron disease
diagnosis in accordance with the
original El Escorial Criteria,
including ALS (International
Classification of Diseases, Ninth
Revision (ICD-9) code 335.20),
progressive muscular atrophy
(ICD-9 code 335.21), progressive
bulbar palsy (ICD-9 code 335.22),
pseudobulbar palsy (ICD-9 code
335.23), primary lateral sclerosis
(ICD-9 code 335.24), and other
motor neuron diseases (ICD-9
code 335.29).

Age

OR (95% Cl)b
Overall 1.9 (1.3, 2.7)

3-485

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tFanq et al.

(2017)

Nationwide,
United States

2007-2013
Cohort

Veterans with ALS in the
U.S. National Registry of
Veterans, and other
veterans with ALS not
treated within the Veterans
Affairs healthcare system,
with ALS from April 2003 to
September 2007 and
followed till the date of death
or July 25, 2013; non-
Hispanic Caucasian men
aged 34-83 yr

n: 145 U.S. Veterans with
ALS, who were male,
diagnosed with ALS by
neurologist.

Blood assessed for Pb ALS survival

Whole blood collected
during Jan-Sep 2007
assessed for Pb using
ICP-MS.

Biomarkers for bone
formation measured in
plasma. Bone formation
was measured using
procollagen type I N-
terminal propeptide and
bone resorption
measured using C-
terminal collagen
crosslinks.

Blood Pb cone: 2.35 ±
1.28 |jg/dL

Age at diagnosis, diagnostic
certainty, site of onset,
Trained neurologist with expertise diagnostic delay and revised
in ALS assigned diagnoses using ALS Functional Rating Scale
an algorithm based on the	Score

revised El Escorial Criteria

HR (95% Clf
1.234 (1.021, 1.49)

tPeters et al.

(2020)

EPIC

Multi-center,
Europe

1993-1999
Nested case-
control

N = 107 cases identified
after 8 yr of follow-up
3 controls per case

Pb concentration in
erythrocytes analyzed
using ICP-MS

ALS: Motor neuron disease
(ICD10 G12.2) as underlying
cause of death

Matched by age at	OR (95%CI)

recruitment, sex, study center Reference: <56.8 ng/g

>56.8-<89.0: 1.83
(0.99, 3.35)

>89.0: 1.89 (0.97, 3.67)

tVinceti et al.
(2017)

Emilia-Romagna;
Italy

May 1998—April
2011

Case-control

Cases were ALS patients
and controls were selected
from hospital-admission of
no ALS; mean age cases:
52 yr

n: 76 (38 ALS cases and 38
controls)

Cerebral spinal fluid
assessed for heavy
metals

CSF evaluated for heavy
metals (Pb, Cd, Hg) using
ICP-MS

Median Pb cone: Cases:
155 ng/L, Controls: 132
ng/L

ALS

Probable ALS diagnosis using the
revised El Escorial Criteria.

Age, sex, and total selenium

In the highest fertile of
exposure, OR (95%
Cl)b 1.39 (0.48, 4.25)

3-486

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tAndrew et al.
(2022)

Nationwide,
United States

2013-2019

Case-control

Participants are from the
healthcare claims dataset
from Symphony Health with
ALS diagnosis after 6 mo
enrollment in the database
prior to the first ALS ICD
code. Controls are
individuals similar to ALS
cases based on age, sex,
and length of database
history with min of 6 mo in
database; cases and
controls age 18-80 yr (63%
were 55-75 yr)

n: Cases: 26,199 and
controls: 78,597

268 Airborne
contaminants

Airborne exposure to Pb
and other contaminants
assessed from U.S.
EPA's NEI database for
2008 to estimate
exposure prior to ALS
onset. Data was used to
estimate residential
exposure at the zip3
locations of the ALS
patients and controls.

ALS

ALS based on the healthcare
claims

Family income, race, age,
and sex

OR (95% Cl)b
Discovery and
Validation Cohorts

1.39 (95% CI 0.48-
4.25)

New

Hampshire/Vermont

[Q1+Q2: <1.37 tons
(Ref)

Q3: 1.37-26.1
Q4: >26.1]

5-year Exposure
History

Q3: 1.79 (1.32, 2.43)
Q4: 1.11 (0.8, 1.55)

10-year Exposure
History

Q3: 2.42 (1.76, 3.33)
Q4: 2.03 (1.46, 2.8)

15-year Exposure
History

Q3: 1.83 (1.34, 2.52)
Q4: 1.73 (1.26, 2.38)

Ohio

[Q1+Q2: <14.7 tons
(Ref) Q3: 14.7-50.8

Q4: >50.8]

3-487

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

5-year Exposure
History

Q3: 0.48 (0.37, 0.61)
Q4: 0.39 (0.3, 0.51)

10-year Exposure
History

Q3: 1.05 (0.83, 1.33)
Q4: 1.6 (1.28, 1.98)

15-year Exposure
History

Q3: 0.94 (0.74, 1.18)
Q4: 1.07 (0.86, 1.34)

3-488

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tPaul et al.

(2021)

Australia and
New Zealand,
and Central
California

Case-control

Participants for this study
comes from two publicly
available PD studies: SGPD
consortium of three studies
across Australia and New
Zealand with cases and
controls. PEG a population-
based study from three
agricultural counties of
Central California with cases
and controls.

n: SGPD cohort: 959 cases
and 930 controls; PEG
cohort: 569 cases and 238
controls

Epigenetic biomarkers for PD
cumulative Pb exposure
(tibia and patella), i.e.,

DNAm Pb

Epigenetic biosensors
identified with site-by-sire
analysis and combined
via machine learning
algorithm on K-XRF in
vivo measures of bone
Pb. To determine Pb
biomarker level in two
cohorts, the regression
coefficients were
extracted from NAS and
applied to the DNAm beta
matrices.

Age (DNAm Age in SGPD),
sex, ancestry (PEG only),
smoker (PEG only), blood
cell composition, and mean
Meth By Sample

OR (95% Cl)b
Tibia SPGD
1.54 (1.22, 1.95)
Patella SPGD
0.70 (0.53, 0.93)
Tibia PEG
1.52 (1.25, 1.86)

DNAm tibia Pb: SGPD
cohort: cases: 3.41 ± 0.4,
controls: 3.48 ± 0.4; PEG
cohort: cases: 3.06 ± 0.4,
controls: 3.03 ± 0.3

tJietal. (2015)

Participants selected are

Blood, Bone assessed for Tremor

Age, age squared, alcohol

OR (95% Cl)b



subgroup of participants

Pb

consumption, smoking status,

Blood

Boston, MA,
United States

from cohort study (Veterans
Affairs NAS); healthy men

Blood samples tested for

education level

Quintile 2 vs. 1



aged 50-98 yr.

Pb concentration using



0.09 (-0.10, 0.29)

NAS



Zeeman background-



Quintile 3 vs. 1

n: 807

corrected flameless



0.06 (-0.16, 0.28)





atomic absorption



Cohort



graphite furnace. Bone



Quintile 4 vs. 1



Pb concentration
measured with K-XRF at



0.07 (-0.13, 0.27)
Quintile 5 vs. 1





both the tibia and the



0.07 (-0.16, 0.30)
Patella





patella starting in 1991.







Blood Pb concentration:



Quintile 2 vs. 1





5.01 ± 2.72 |jg/dL





3-489

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

Tibia Pb cone: 21.23 ±

13.29 |jg/g (); patella Pb Tremor score was created based
cone: 27.98 ± 18.38 |jg/g) on an approach using hand-
drawing samples that were
derived from figure copying
testing performed as part of
larger cognitive test battery
(CERAD, MMSE, and VMI)
assessed over a mean follow-up
of 8.0 ± 3.2 yr after bone Pb
measurement.

ALAD genotype was determined
by amplifications of 0.5 |jL of
whole blood using two sets of
primers specific for a portion of
the ALAD gene.

-0.07 (-0.30, 0.15)
Quintile 3 vs. 1
-0.14 (-0.37, 0.09)
Quintile 4 vs. 1
-0.22 (-0.45, 0.01)
Quintile 5 vs. 1
-0.02 (-0.27, 0.22)
Tibia

Quintile 2 vs. 1
0.03 (-0.20, 0.26)
Quintile 3 vs. 1
0.03 (-0.20, 0.26)
Quintile 4 vs. 1
0.13 (-0.11, 0.36)
Quintile 5 vs. 1
-0.07 (-0.32, 0.17)

3-490

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tKhalil et al.
(2014)

Pittsburgh, PA
United States

2007-2009

Cross-sectional

MrOS
n: 445

Non-Hispanic Caucasian
men (community dwelling
non-institutionalized) at least
65 yr of age enrolled in
MrOS at the University of
Pittsburgh clinic. Eligibility
criteria included the ability to
walk unaided and without
bilateral hip replacements.

Blood	Grip strength (kg);

Leg extension power (watts);
Blood Pb measured using Walking speed (m/s);
AAS	Narrow-walk pace (m/s);

Age at measurement: Use arms to stand up (yes/no)
Mean = 79.5 ± 5 yr

Grip strength was measured on a
Mean = 2.25 |jg/dL; SD = Jamar dynamometer;

1.20 |jg/dL; Median = 2 Leg extension power was

Age, education, smoking,
alcohol consumption, BMI

Beta (95% Clf
Leg extension power
-0.03 (-1.97, 2.03)
Ability to stand from a
chair without using their
arms 0.97 (0.88, 1.07)

|jg/dL

Max: 10 pg/dL

measured with the Nottingham
power rig;

Walking speed was assessed on
a standard 6-m walking course;
Narrow-walk pace (an indirect
measure of dynamic balance)
was assess while keeping each
foot within a 20-centimeter wide
lane on the 6-m walking course;
Stand from a chair without using
the arms was measured as
yes/no.

Age at outcome:

65 yr

3-491

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tShiue (2013)

United States

2003-2004

Cross-sectional

NHANES

n: 712 (vision); 732
(hearing); 669 (balance)

Urine

Urinary Pb was detected
by mass spectrometry
NHANES age 50 and above Age at measurement:

50 yr

Not Reported

Vision;

Hearing;

Balance;

Ear ringing

Vision: excellent, good, and fair
eyesight (self-reported) were
classified as good; poor and very
poor were classified as poor
Hearing: good and little trouble
hearing (self-reported) were
classified as good; lots of trouble
and deaf were classified as poor
Balance: "During the past 12 mo,
have you had dizziness, difficulty
with balance, or difficulty with
failing?"

Ear ringing: "ears ringing, roaring,
or buzzing in the last year"

Age at outcome:

50 yr

Age, sex, ethnicity, urine
creatinine, survey weighting

OR (95% Cl)b

Vision 1.15 (0.67-1.97)

Hearing 0.97 (0.63-
1.51)

Balance 0.68 (0.51 —
0.91)

3-492

-------
tCasiens et al.
(2018)

Ruhr area, a
German
industrial region
with a high
volume of steel
production
Germany

Baseline
recruitment
2000-2003;
Follow-up 2011-
2014

Cohort

HNRS
n: 1188

Men from the Heinz Nixdorf
Recall Study. Recruitment
details not provided.

Blood

Blood Pb was measured
in aliquots of whole blood
archived at baseline and
at follow-up using ICP-
MS

Age at measurement:
Median = 58 yr at
baseline (range 45-75 yr)
and 68 yr (range 55-86)
yr at follow-up

Median = 3.29 (IQR
2.55-4.32) |jg/dL at
baseline; 2.59 (IQR 1.99—
3.39) |jg/dL at follow-up
Max: 67.73 |jg/dL at
baseline; 39.68 |jg/dL at
follow-up

Odor identification;

Tapping hits;

Aiming errors;

Line tracing errors;

Steadiness errors

Odor identification: Sniffin sticks
odor identification test of 12
odors, participants classified as
normosmic if >9 odors identified,
hyposmic if 7-9 odors identified,
and functionally anosmic if <7
odors identified
Tapping hits: tapping a stylus
within 32 s as fast as possible;
- hits <10th percentile were
considered as substantially
impaired manual dexterity
Aiming errors: 20 small plates
with a diameter of 5 mm standing
in a line (distance 4 mm) had to
be touched with a stylus as fast
as possible; errors >90th
percentiles were considered as
substantially impaired manual
dexterity

Line tracing errors: drawing a
stylus through a curvy course of a
groove without touching side
walls or bottom; errors >90th
percentiles were considered as
substantially impaired manual
dexterity

Steadiness errors: maintain a
precise arm-hand position by
holding a stylus for 32 s in a 5.8
mm hole without touching sides
or bottom; errors >90th
percentiles were considered as
substantially impaired manual
dexterity

Age at outcome:

55-86 yr

Occupational qualification,
age, smoking status, alcohol
consumption, total test time

OR (95% Cl)b
Group 1: <5 |jg/dL (Ref)
Group 2: 5-<9 |jg/dL
Group 3: >9 |jg/dL

Odor identification

baseline

G2: 0.91 (0.65, 1.28)
G3: 1.96 (0.94, 4.11)
follow-up

G2: 1.04 (0.55, 1.94)
G3: 1.57 (0.47, 5.19)

Motor Performance
Series

Steadiness Errors

baseline

G2: 0.99 (0.62, 1.59)
G3: 1.36 (0.5, 3.66)
follow-up

G2: 1.16 (0.5, 2.69)
G3: 1.75 (0.41, 7.58)

Line tracing errors

baseline

G2: 1.09 (0.68, 1.76)
G3: 0.93 (0.32, 2.74)
follow-up

G2: 1.01 (0.41, 2.48)
G3: 0.59 (0.08, 4.11)

Aiming errors

baseline

G2: 1.07 (0.75, 1.53)
G3: 0.56 (0.22, 1.42)

3-493

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

follow-up

G2: 1.35 (0.73, 2.51)
G3: 0.42 (0.09, 2.08)

Tapping Hits

baseline

G2: 0.87 (0.53, 1.44)
G3: 1.35 (0.49, 3.7)
follow-up

G2: 2.63 (1.26, 5.49)
G3: 0.8 (0.14, 4.59)

3-494

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tJietal. (2013)

United States

1999-2002

Cross-sectional

NHANES

n: 3,593 (1,798 women;
1,795 men)

NHANES data from the
1999-2000 and 2001-2002
surveys, participants were
50 yr of age

Blood

Blood Pb was measured
using AAS

Age at measurement:
50-85 yr (Median = 61.2
yr)

Mean ± SD: Women =

2.17	± 0.04 |jg/dL; Men =

3.18	± 0.08 |jg/dL
("there's a slight
discrepancy in the SDs in
Table 1 vs. text on p.

712)

Median: Women = 1.72
|jg/dL; Men = 2.41 |jg/ dL

Walking speed (ft/sec)

Time to walk 20 ft (at usual
walking pace)

Age at outcome:

50-85 yr (median = 61.2 yr)

Model 4 (fully adjusted): age,
education, ethnicity, height,
waist circumference, alcohol,
smoking, physical activity,
arthritis, diabetes, heart
condition, hypertension,
homocysteine, C-reactive
protein

Beta (95% CI)

Walking Speed-Men
4.4 to <54.0

-0.029 (-0.155, 0.097)
Walking Speed-Women

3.0	to <53.0
-0.114 (-0.191,

-0.038)

Walking Speed-Men

3.1	to < equal to 4.3

0.082 (-0.012, 0.176)
Walking Speed-Women

2.2	to <2.9
-0.104 (-0.187,

-0.021)

Walking Speed-Men
2.4 to <3.0

-0.17 (-0.26, -0.08)

Walking Speed-Women

1.7	to <2.1
-0.024 (-0.118,

-0.063)

Walking Speed-Men

1.8	to <2.3

0.057 (-0.051, 0.165)
Walking Speed-Women

1.3	to <1.6

-0.027 (-0.055, 0)

3-495

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tMin etal. (2012) NHANES
n: 5574

United States

1999-2004

Cross-sectional

Adults who participated in
the NHANES Balance
Component and had blood
Pb and Cd measurements
and data for all covariate
variables

Blood

Blood Pb was measured
using a multielement AAS
with Zeeman background
correction.

Age at measurement:
40 yr

Weighted mean
(participant without
balance dysfunction):
2.09 |jg/dL (95% CI: 2.01,
2.18); Weighted mean
(participants with balance
dysfunction): 2.39 |jg/dL
(95% CI: 2.29, 2.49)
Max: 48 pg/dL

Balance dysfunction

Balance dysfunction was
evaluated by the Romberg Test
of Standing Balance on Firm and
Compliant Support Surfaces,
which measured the participant's
ability to maintain balance under
four test conditions: Test 1)
maintain balance while standing
(with feet together and arms
folded across the waist, holding
each elbow with the opposite
hand) for 15 sec.; Test 2)
maintain balance while standing
for 15 sec with eyes closed so
that only vestibular and
proprioceptive (i.e., leg muscle
position sense) information is
available; Test 3) maintain
balance while standing for 30 sec
on a foam-padded surface, which
reduces proprioceptive input but
does not affect visual or
vestibular input; Test 4) maintain
balance while standing for 30 sec
on a foam-padded surface with
eyes closed, so that input is
available from the vestibular
system only. Each condition was
scored on a pass or fail basis.
The time to failure (i.e., loss of
balance) was also recorded for
test condition 4, with those who
passed the test assigned the
maximum value of 30 sec.

Age, sex, race/ethnicity,
education, pack-years of
smoking, alcohol
consumption, histories of
stroke and diabetes, intakes
of Ca2+ and iron

OR (95% CI)

Balance Dysfunction
(Quintile 5 [3.3—48
Mg/dL])

33.334 (1.939,
573.157)

Balance Dysfunction
(Quintile 4 [2.3-3.2
MQ/dL])

5.234 (0.59, 46.429)
Balance Dysfunction
(Quintile 3 [1.8-2.2
MQ/dL])

0.665 (0.05, 8.783)

Balance Dysfunction
(Quintile 2 [1.3-1.7
Mg/dL])

3.707 (0.544, 25.282)

Age at outcome:
40 yr

3-496

-------
Reference and
Study Design

Study Population

Exposure Assessment

Outcome

Confounders

Effect Estimates and
95% CIs

tGrashow et al.
(2013)

Greater Boston
area, MA, United
States

Grooved
pegboard May
2005- December
2009

Neuroskill July
2004 and
November 2007

Cohort

Normative Aging Study
n: 362 for grooved pegboard
test; 328 for the Neuroskill
test

Elderly, majority Caucasian
men originally recruited from
the greater Boston,
Massachusetts area in the
1960s

Bone

Bone Pb was measured
at the patella and the
midtibial shaft using an
ABIOMED K-XRF
instrument. "Tibia and
patella bone Pb
concentrations reflect
cumulative Pb exposure
over different time
windows: patella Pb
reflects exposure over
the last decade, while
tibia Pb half-life is on the
order of decades"

Age at measurement:

NR

Mean patella Pb = 25.0
mg/g bone (SD = 20.7);
Mean tibia Pb = 19.2
mg/g bone (SD = 14.6)

Grooved pegboard (completion
time, seconds);

Neuroskill (Signature score, %);
Neuroskill (Im pattern score, %)

Grooved pegboard test: Subjects
were asked to insert the metal
pegs into each of the 25 holes in
sequence as quickly as possible
with their dominant hand without
practice trials;

Neuroskill tests (signature score,
%): Subjects were asked to
provide five samples of their
signature in succession, written in
their natural manner;

Neuroskill tests (Im pattern score,
%): Subjects were asked to
provide five samples of a series
of cursive Ims (Im pattern) using
the instrumented pen

Age at outcome:

Mean age = 69.1 yr (SD = 7.2)

Age, smoking, education, Beta (95% Cl)a
computer experience, income Neuroskill-lm pattern

score

Patella 0.45 (0.178,
0.723)

Tibia 0.847 (0.163,
1.53)

Neuroskill-Signature
Score

Patella 0.08 (-0.47,
0.63)

Tibia -0.293 (-1.063,
0.477)

Grooved pegboard-
dominant hand
completion time
Patella 1.965 (0.553,
3.378)

Tibia 3.107 (1.157,
5.057)

AAS = atomic absorption spectrometry; BLL = blood lead level; Cd = cadmium; CERAD = Consortium to Establish a Registry for Alzheimer's Disease; CI = confidence interval; CSF =
cerebrospinal fluid; EE = effect estimate(s); HNRS = Heinz Nixdorf Recall Study; HRR = hazard rate ratio; K-XRF = K-shell X-ray fluorescence; MMSE = Mini Mental State
Examination; mo = month(s); MrOS = Osteoporotic Fractures in Men Study; NAS = Normative Aging Study; NEI = National Emissions Inventory; NHANES = National Health and
Nutrition Examination Survey; OR = odds ratio; Pb = lead; SD = standard deviation; sec = second(s); VMI = visual-motor integration; yr = year(s).

aEffect estimates are standardized to a 1 |jg/dL increase in BLL or a 10 |jg/g increase in bone Pb level, unless otherwise noted. For studies that report results corresponding to a
change in log-transformed Pb biomarkers, effect estimates are assumed to be linear within the 10th to 90th percentile interval of the biomarker and standardized accordingly.
bResult not standardized because data pertaining to the BLL distribution and/or base for the log-transformation were not reported.
tStudies published since the 2013 Integrated Science Assessment for Lead.

3-497

-------
Table 3-17T Animal toxicological studies of Pb exposure and neurodegeneration

Study

Species (Stock/Strain), n,	Timing of	Exposure

Sex	Exposure	Details

BLL as Reported (pg/dL)

Endpoints Examined

Zhou etal. (2018)

Rat (Sprague Dawley)

Control (distilled water), M, n
= 10

0.5% solution, M, n = 10
1.0% solution, M, n = 10
2.0% solution, M, n = 10

PND 24 to PND 52 Oral, drinking
water

PND 52:

13.3 |jg/L (1.3 pg/dL) for Control

148.9 |jg/L (14.9 pg/dL) for 0.5%
solution

231.3	|jg/L (23.1 pg/dL) for 1.0%
solution

293.4	|jg/L (29.3 pg/dL) for 2.0%
solution

PND 24, 31, 38, 45,
52: Amyloid protein
expression, Brain
Cholesterol,
Expression of BACE1
and APP

Li etal. (2016c)

Mouse (Kunming)

Control (distilled water), M/F,
n = 10

0.1% solution (mass fraction),
M/F. n = 10

0.2% solution (mass fraction),
M/F, n = 10

0.5% solution (mass fraction),
M/F, n = 10

GD to PND 21

Oral, lactation
In utero

PND 21:

10.62 pg/L (1.1 pg/dL) for Control

40.71 pg/L (4.1 pg/dL) for 0.1%
solution

81.77 pg/L (8.2 pg/dL) for 0.2%
solution

103.36 pg/L (10.3 pg/dL) for 0.5%
solution

PND 21: Amyloid
protein expression

Gu etal. (2012)

Mouse (Tg-SwDI)

4-8 wk to 10-14 wk

Oral, gavage

10-14 wk:

10-14 wk: Beta-



Control (Na-acetate water),







amyloid and APP



NR, n = 4-7





1.83 pg/dL for Control

expression



50 mg/kg, NR, n = 4-7





29.5 pg/dL for 50 mg/kg



Wu et al. (2020b)

Mouse (C57BL/6)

Control (distilled deionized
water), M, n = 7-10

0.2% solution, M, n = 7-10

4 wk to 4 mo

Oral, drinking
water

16 mo:

66.4 pg/L (6.6 pg/dL) for Control

278.9 pg/L (27.9 pg/dL) for 0.2%
solution

16 mo: Expression of
BACE1 and APP,
Phosphorylated tau
expression

3-498

-------
Study

Species (Stock/Strain), n,	Timing of	Exposure

Sex	Exposure	Details

BLL as Reported (pg/dL)

Endpoints Examined

Sun et al. (2014)

Rat (Sprague Dawley)

Control (tap water), NR, n :
20

580 ppm, NR, n = 20

NR (230-260 g) - 3 Oral, drinking
mo of treatment water

After 3 mo treatment:

3.0 |jg/L (0.3 pg/dL) for Control

56.8 |jg/L (5.7 pg/dL) for 580 ppm

After 3 mo treatment:
Immunohistochemistry
of APP

Gassowska et al. (2016b)

Rat (Wistar)

GD Oto PND21

Oral, lactation

PND 28:

PND 28: Tau protein



Control (tap water), M/F, n = 8



In utero



expression and









0.93 pg/dL for Control

phosphorylation



0.1% solution, M/F, n = 8

















6.86 pg/dL for 0.1% solution



Rahman et al. (2012b)

Rat (Wistar)

PND 1 to PND 30

Oral, drinking

PND 21:

PND 21, 30: Tau



Control (tap water), M/F, n =



water



protein expression



6-10



Oral, lactation

1.4 pg/dL for Control

and phosphorylation



0.2% solution, M/F, n = 6-12





12.1 pg/dL for 0.2% solution











PND 30:











1.2 pg/dL for Control











12.8 pg/dL for 0.2% solution



Zhana et al. (2012)

Rat (Sprague Dawley)

NR (40-60 g)

Oral, drinking

+8 wk from start of exposure

+8 wk from start of



Control (deionized water), M,



water



exposure:



n = 10





49.9 ng/mL (5 pg/dL) for Control

Phosphorylated tau











expression, Alpha-



100 ppm, M, n = 10





100.9 ng/mL (10.1 pg/dL) for 100 ppm

Synuclein expression



200 ppm, M, n = 10





128.6 ng/mL (12.9 pg/dL) for 200 ppm





300 ppm, M, n = 10





147.7 ng/mL (14.8 pg/dL) for 300 ppm



Bihaai and Zawia (2013)

Monkey (Macaca

PND Oto PND 400

Oral, infant

PND 400:

23 yr:



fascicularis)



formula



Tau protein



Control, F, n = 4



Oral, gelatin

3-6 pg/dL for Control

expression and







capsules

phosphorylation, Tau



1.5 mg/kg/day, F, n = 5





19-26 pg/dL for 1.5 mg/kg/day

phosphorylation

3-499

-------
Study

Species (Stock/Strain), n,	Timing of	Exposure

Sex	Exposure	Details

BLL as Reported (pg/dL)

Endpoints Examined

Feng et al. (2019)

Rat (Sprague Dawley)

Control (deionized water),
M/F, n = 8

0.8 g/L (maternal) and 0.3 g/L
(pup), M/F, n = 8

1.5 g/L (maternal) and 0.9 g/L
(pup), M/F, n = 8

GD -10 to PND
490

Oral, drinking
water

Oral, lactation
In utero

PND 21:

0 mg/L (0 pg/dL) for Control
0.29 mg/L (29 pg/dL) for 0.8 g/L
0.69 mg/L (69 pg/dL) for 1.5 g/L
PND 287:

0 mg/L (0 pg/dL) for Control
0.29 mg/L (29 pg/dL) for 0.8 g/L
0.61 mg/L (61 pg/dL) for 1.5 g/L
PND 490:

0 mg/L (0 pg/dL) for Control
0.31 mg/L (31 pg/dL) for 0.8 g/L
0.58 mg/L (58 pg/dL) for 1.5 g/L

PND 21, 287, 490:
Neuronal Density,
Brain Volume

Mansouri et al. (2012)

Rat (Wistar)

Control (distilled water), M/F,
n = 16 (8/8)

50 mg/L, M/F, n = 16 (8/8)

PND 70 to PND
100

Oral, drinking
water

PND 100-Males:
2.05 pg/dL for Control
8.8 pg/dL for 50 mg/L

PND 100: Open Field
Test, Rotarod Test

PND 100 - Females:
2.17 pg/dL for Control
6.8 pg/dL for 50 mg/L

3-500

-------
Study

Species (Stock/Strain), n,	Timing of	Exposure

Sex	Exposure	Details

BLL as Reported (pg/dL)

Endpoints Examined

Mansouri et al. (2013)

Rat (Wistar)

Control (tap water or
water+NaAc), M/F, n = 16
(8/8)

50 ppm, M/F, n = 16 (8/8)

PND 55 to PND
181

Oral, drinking
water

PND 178-181 - Females:

NR for Control

10.6 pg/dL for 50 ppm

PND 155-159:
Rotarod Test

PND 178-181 - Males:
NR for Control
18.9 pg/dL for 50 ppm

Singh et al. (2019)	Rat (Wistar)	3 mo to 6 mo

Control (distilled water), M, n
= 5

2.5 mg/kg, M, n = 5

Oral, gavage 6 mo:

5.76 pg/dL for Control
28.4 pg/dL for 2.5 mg/kg

6 mo: Locomotor
Activity, Rotarod Test

Al-Qahtani et al. (2022) Mouse (Albino)	8-9 wk to 14-15 wk Oral, gavage 14-15 wk:	NR: Locomotor

Control (distilled water), M, n	Activity

= 10	1.2 |jg/100 mL (1.2 pg/dL) for Control

0.2 mg/kg, M, n = 10

7.1 pg/100 mL (7.1 pg/dL) for 0.2
mg/kg

APP = amyloid precursor protein; BACE1 = beta-secretase 1; F = female; GD = gestational day; M = male; mo = month(s); NR = not reported; Pb = lead; PND = postnatal day; wk =
week(s); yr = year(s).

3-501

-------
3.8 References

Abazvan. B: Dziedzic. J: Hua. K: Abazvan. S: Yang. C: Mori. S: Pletnikov. MY: Guilarte. TR. (2014). Clironic
exposure of mutant DISCI mice to lead produces sex-dependent abnormalities consistent with
schizophrenia and related mental disorders: A gene-enviromnent interaction study. Schizophr Bull 40: 575-
584. http://dx.doi.org/10.1093/schbul/sbt071.

Abubakar. K: Mailafiva. MM: Danmaigoro. B: Chiroma. SM: Raliim. E: Zakari. M. (2019). Curcumin attenuates
lead-induced cerebellar toxicity in rats via chelating activity and inhibition of oxidative stress.

Biomolecules 9: 453. http://dx.doi.org/10.3390/biom9090453.

AbuShadv. MM: Fathv. HA: Fathv. GA: abd el Fatah. S: Ali. A: Abbas. MA. (2017). Blood lead levels in a group of
children: The potential risk factors and health problems. J Pediatr (Rio J) 93: 619-624.
http://dx.doi.Org/10.1016/i.iped.2016.12.006.

Ahmad. F: Hague. S: Ravinavagam. V: Ahmad. A: Kamli. MR: Barreto. GE: Ghulam Md Ashraf. GE. (2020).
Developmental lead (Pb)-induced deficits in redox and bioenergetic status of cerebellar synapses are
ameliorated by ascorbate supplementation. Toxicology 440: 152492.
http://dx.doi.Org/10.1016/i.tox.2020.152492.

Ahmed. MB: Ahmed. MI: Meki. AR: Abdraboh. N. (2013). Neurotoxic effect of lead on rats: Relationship to
apoptosis. International Journal of Health Sciences 7: 192-199. http://dx.doi.org/10.12816/0006Q42.

Al-Oahtani. A: Aiarem. J: Okla. MK: Rubnawaz. S: Alamri. SA: Al-Oahtani. WH: Al-Himaidi. AR: Elgawad. HA:
Akhtar. N: Maodaa. SN: Abdel-Maksoud. MA. (2022). Protective effects of green tea supplementation
against lead-induced neurotoxicity in mice. Molecules 27: 993.
http://dx.doi.org/10.3390/molecules27030993.

Al-Saleh. I: Moncari. L: Jomaa. A: Elkhatib. R: Al-Rouai. R: Eltabache. C: Al-Raiudi. T: Alnuwavsir. H: Nester. M:
Aldhalaan. H. (2020). Effects of early and recent mercury and lead exposure on the neurodevelopment of
children with elevated mercury and/or developmental delays during lactation: A follow-up study. Int J Hyg
Environ Health 230: 113629. http://dx.doi.Org/10.1016/i.iiheh.2020.113629.

Alabdali. A: Al-Avadhi. L: El-Ansarv. A. (2014). A key role for an impaired detoxification mechanism in the
etiology and severity of autism spectrum disorders. Behavioral and Brain Functions 10: 14.
http://dx.doi.org/10T186/1744-9081-10-14.

Albers. CA: Grieve. AJ. (2007). Test review: Bayley, N. (2006). Bayley Scales of Infant and Toddler Development-
Third Edition [Review]. J Psychoeduc Assess 25: 180-198. http://dx.doi.org/10.1177/0734282906297199.

Alvarenga. KF: Morata. TC: Lopes. AC: Fenimaa MR: Corteletti. LC. (2015). Brainstem auditory evoked potentials
in children with lead exposure. Braz J Otorhinolaryngol 81: 37-43.
http ://dx. doi.org/10.1016/i ,bi orl.2013.12.001.

Amato. MS: Magzamen. S: Imm. P: Havlena. JA: Anderson. HA: Kanarek. MS: Moore. CF. (2013). Early lead
exposure (<3 years old) prospectively predicts fourth grade school suspension in Milwaukee, Wisconsin
(USA). Environ Res 126: 60-65. http://dx.doi.Org/10.1016/i.envres.2013.07.008.

Amedu. NO: Omotoso. GO. (2020). Lead acetate-induced neurodegenerative changes in the dorsolateral prefrontal
cortex of mice: The role of Vitexin. Environ Anal Health Toxicol 35: e2020001.
http://dx.doi.org/10.5620/eaht.e2020Q01.

Amos-Kroohs. RM: Graham. PL: Grace. CE: Braun. AA: Schaefer. TL: Skelton. MR: Vorhees. CV: Williams. MT.
(2016). Developmental stress and lead (Pb): Effects of maternal separation and/or Pb on corticosterone,
monoamines, and blood Pb in rats. Neurotoxicology 54: 22-33.
http://dx.doi.Org/10.1016/i.neuro.2016.02.011.

3-502

-------
An. J: Cai. T: Che. H: Yu. T: Cao. Z: Liu. X: Zhao. F: Jing. J: Shen. X: Liu. M: Du. K: Chea J: Luo. W. (2014).

The changes of iniRNA expression in rat hippocampus following chronic lead exposure. Toxicol Lett 229:
158-166. http://dx.doi.Org/10.1016/i.toxlet.2014.06.002.

Anderson. DW: Mettil. W: Schneider. JS. (2016). Effects of low level lead exposure on associative learning and
memory in the rat: Influences of sex and developmental timing of exposure. Toxicol Lett 246: 57-64.
http ://dx. doi.org/10.1016/i .toxlet. 2016.01.011.

Anderson. DW: Pothakos. K: Schneider. JS. (2012). Sex and rearing condition modify the effects of perinatal lead
exposure on learning and memory. Neurotoxicology 33: 985-995.
http://dx.doi.Org/10.1016/i.neuro.2012.04.016.

Andrade. V: Mateus. ML: Batoreu. MC: Aschner. M: Dos Santos. AM. (2017). Toxic mechanisms underlying motor
activity changes induced by a mixture of lead, arsenic and manganese. EC Pharmacol Toxicol 3: 31-42.

Andrew. A: Zhou. J: Gui. J: Harrisoa A: Shi. X: Li. M: Guetti. B: Nathan. R: Tischbein. M: Pioro. E: Stommel. E:
Bradley. W. (2022). Airborne lead and polychlorinated biphenyls (PCBs) are associated with amyotrophic
lateral sclerosis (ALS) risk in the U.S. Sci Total Environ 819: 153096.
http://dx.doi.Org/10.1016/i.scitotenv.2022.153096.

Anvan. J: Amir. S. (2018). Too depressed to swim or too afraid to stop? A reinterpretation of the forced swim test as
a measure of anxiety-like behavior [Comment]. Neuropsychophannacology 43: 931-933.
http://dx.doi.org/10.1038/npp.2017.260.

APA (American Psychiatric Association). (2013). Diagnostic and statistical manual of mental disorders (5th ed.).
Arlington, VA. http://dx.doi.org/10.1176/appi.books.9780890425596.

Arbuckle. TE: Davis. K: Bovlan. K: Fisher. M: Fu. J. (2016a). Bisphenol A, phthalates and lead and learning and
behavioral problems in Canadian children 6-11 years of age: CHMS 2007-2009. Neurotoxicology 54: 89-
98. http://dx.doi.Org/10.1016/i.neuro.2016.03.014.

Arbuckle. TE: Davis. K: Bovlan. K: Fisher. M: Fu. J. (2016b). Processed data for CHMS 2007-2009: Bisphenol A,
phthalates and lead and learning and behavioral problems in Canadian children 6-19 years of age. Data in
Brief 8: 784-802. http://dx.doi.Org/10.1016/i.dib.2016.06.017.

Arora. M: Reichenberg. A: Willfors. C: Austin. C: Gennings. C: Berggren. S: Lichtenstein. P: Anckarsater. H:

Tammimies. K: Bolte. S. (2017). Fetal and postnatal metal dysregulation in autism. Nat Commun 8: 15493.
http://dx.doi.org/10.1038/ncommsl5493.

Ashok. A: Rai. NK: Tripathi. S: Bandvopadhvav. S. (2015). Exposure to As-, Cd-, and Pb-mixture induces A(3,
amyloidogenic APP processing and cognitive impairments via oxidative stress-dependent
neuroinflammation in young rats. Toxicol Sci 143: 64-80. http://dx.doi.org/10.1093/toxsci/kfu208.

ATSDR (Agency for Toxic Substances and Disease Registry). (2020). Toxicological profile for lead [ATSDR Tox
Profile]. Atlanta, GA: U.S. Department of Health and Human Services, Agency for Toxic Substances and
Disease Registry, https://www.atsdr.cdc. gov/toxprofiles/tp 13 .pdf.

Baba. M: Nakaio. S: Tu. PH: Tomita. T: Nakava. K: Lee. VM: Troianowski. JO: Iwatsubo. T. (1998). Aggregation
of alpha-synuclein in Lewy bodies of sporadic Parkinson's disease and dementia with Lewy bodies. Am J
Pathol 152: 879-884.

Babinski. LM: Hartsough. CS: Lambert. NM. (1999). Childhood conduct problems, hyperactivity-impulsivity, and
inattention as predictors of adult criminal activity. J Child Psychol Psychiatry 40: 347-355.
http://dx.doi.org/10.1111/1469-7610.00452.

Baghurst. PA: Mcmichael. AJ: Wigg. NR: Vimpani. GV: Robertson. EF: Roberts. RJ: Tong. SL. (1992).

Enviromnental exposure to lead and children's intelligence at the age of seven years: The Port Pirie cohort
study. N Engl J Med 327: 1279-1284. http://dx.doi.org/10.1056/NEJM1992102932718Q5.

Bah. HAF: Dos Anios. ALS: Gomes-Junior. EA: Bandeira. MJ: de Carvalho. CF: Dos Santos. NR: Martinez. VO:
Adorno. EV: Menezes-Filho. JA. (2022). Delta-aminolevulinic acid dehydratase, low blood lead levels,
social factors, and intellectual function in an Afro-Brazilian children community. Biol Trace Elem Res 200:
447-457. http://dx.doi.Org/10.1007/sl2011-021-02656-8.

3-503

-------
Balasundaram. P: Avulakunta. ID. (2021). Bayley scales of infant and toddler development [StatPearls]. Treasure
Island, FL: StatPearls Publishing. Retrieved from https://www.ncbi.nlm.nih.gov/books/NBK567715/

Bandeen-Roche. K: Glass. TA: Bolla. KI: Todd. AC: Schwartz. BS. (2009). Cumulative lead dose and cognitive
function in older adults. Epidemiology 20: 831-839. http://dx.doi.org/10.1097/EDE.0b013e3181b5flQ0.

Baranowska-Bosiacka. I: Falkowska. A: Gutowska. I: Gassowska. M: Kolasa-Wolosiuk. A: Tarnowski. M:

Cliibowska. K: Goschorska. M: Lubkowska. A: Clilubek. D. (2017). Glycogen metabolism in brain and
neurons - astrocytes metabolic cooperation can be altered by pre- and neonatal lead (Pb) exposure.
Toxicology 390: 146-158. http://dx.doi.Org/10.1016/i.tox.2017.09.007.

Baranowska-Bosiacka. I: Gutowska. I: Marchetti. C: Rutkowska. M: Marchlewicz. M: Kolasa. A: Prokopowicz. A:
Wiernicki. I: Piotrowska. K: Baskiewicz. M: Safranow. K: Wiszniewska. B: Clilubek. D. (2011). Altered
energy status of primary cerebellar granule neuronal cultures from rats exposed to lead in the pre- and
neonatal period. Toxicology 280: 24-32. http://dx.doi.Org/10.1016/i.tox.2010.ll.004.

Baranowska-Bosiacka. I: Struzvnska. L: Gutowska. I: Machalinska. A: Kolasa. A: Klos. P: Czapski. GA:

Kurzawski. M: Prokopowicz. A: Marchlewicz. M: Safranow. K: Machalinski. B: Wiszniewska. B:

Clilubek. D. (2013). Perinatal exposure to lead induces morphological, ultrastructural and molecular
alterations in the hippocampus. Toxicology 303: 187-200. http://dx.doi.Org/10.1016/i.tox.2012.10.027.

Barbeito. AG: Martinez-Palma. L: Vargas. MR: Pehar. M: Mafiav. N: Beckman. JS: Barbeito. L: Cassina. P. (2010).
Lead exposure stimulates VEGF expression in the spinal cord and extends survival in a mouse model of
ALS. Neurobiol Dis 37: 574-580. http://dx.doi.Org/10.1016/i.nbd.2009.ll.007.

Barg. G: Daleiro. M: Oueirolo. EI: Ravenscroft. J: Mafiav. N: Peregalli. F: Kordas. K. (2018). Association of low
lead levels with behavioral problems and executive function deficits in schoolers from Montevideo,
Uruguay. Int J Environ Res Public Health 15: 2735. http://dx.doi.org/10.3390/iierphl5122735.

Barkur. RR: Bairv. LK. (2015a). Assessment of oxidative stress in hippocampus, cerebellum and frontal cortex in rat
pups exposed to lead (Pb) during specific periods of initial brain development. Biol Trace Elem Res 164:
212-218. http://dx.doi.org/10.1007/sl2011-014-Q221-3.

Barkur. RR: Bairv. LK. (2015b). Evaluation of passive avoidance learning and spatial memory in rats exposed to
low levels of lead during specific periods of early brain development. Int J Occup Med Environ Health 28:
533-544. http://dx.doi.org/10.13075/iiomeh.1896.00283.

Barkur. RR: Bairv. LK. (2016). Histological study on hippocampus, amygdala and cerebellum following low lead
exposure during prenatal and postnatal brain development in rats. Toxicol Ind Health 32: 1052-1063.
http://dx.doi.org/10.1177/0748233714545624.

Barkur. RR: Rao. MS: Bairv. LK. (2011). Low lead exposure during foetal and early postnatal life impairs passive
avoidance learning in adulthood in rats. Arh Hig Rada Toksikol 62: 147-153.
http://dx.doi.org/10.2478/10004-1254-62-2011-207Q.

Basha. CD: Reddv. RG. (2015). Long-term changes in brain cholinergic system and behavior in rats following
gestational exposure to lead: Protective effect of calcium supplement. Interdiscip Toxicol 8: 159-168.
http ://dx. doi.org/10.1515/intox-2015-0025.

Basha. DC: Reddv. NS: Rani. MU: Reddv. GR. (2014). Age related changes in aminergic system and behavior
following lead exposure: Protection with essential metal supplements. Neurosci Res 78: 81-89.
http://dx.doi.Org/10.1016/i.neures.2013.09.007.

Basha. MR: Murali. M: Siddiqi. HK: Ghosal. K: Siddiqi. OK: Lashuel. HA: Ge. YW: Laliiri. DK: Zawia. NH.

(2005). Lead (Pb) exposure and its effect on APP proteolysis and A beta aggregation. FASEB J 19: 2083-
2084. http://dx.doi.org/10.1096/fi.05-4375fie.

Batliina. S: Das. UN. (2015). Brain-derived neurotrophic factor and its clinical implications. Archives of Medical
Science 11: 1164-1178. http://dx.doi.org/10.5114/aoms.2015.56342.

Baver. SA. (1989). Cellular aspects of brain development [Review]. Neurotoxicology 10: 307-320.

Bavlev. N. (1969). Bayley scales of infant development: Manual. New York, NY: The Psychological Corporation.

3-504

-------
Beaudin. SA: Stangle. DE: Smith. PR: Levitskv. DA: Strupp. BJ. (2007). Succimer chelation nomializes reactivity
to reward omission and errors in lead-exposed rats. Neurotoxicol Teratol 29: 188-202.
http ://dx. doi.org/10.1016/i .ntt.2006.11.004.

Becklev. AL: Caspi. A: Broadbent. J: Harrington. H: Houts. RM: Poulton. R: Ramrakha. S: Reuben. A: Moffitt. TE.
(2018). Association of childhood blood lead levels with criminal offending. JAMA Pediatr 172: 166-173.
http://dx.doi.org/10.1001/iamapediatrics.2017.40Q5.

Beckwith. TJ: Dietrich. KN: Wright. JP: Altave. M: Cecil. KM. (2018). Reduced regional volumes associated with
total psychopathy scores in an adult population with childhood lead exposure. Neurotoxicology 67: 1-26.
http://dx.doi.Org/10.1016/i.neuro.2018.04.004.

Beckwith. TJ: Dietrich. KN: Wright. JP: Altave. M: Cecil. KM. (2021). Criminal arrests associated with reduced

regional brain volumes in an adult population with documented childhood lead exposure. Environ Res 201:
111559. http://dx.doi.Org/10.1016/i.envres.2021.111559.

Bellinger. D: Hu. H: Titlebaum. L: Needleman. HL. (1994a). Attentional correlates of dentin and bone lead levels in
adolescents. Arch Environ Health 49: 98-105. http://dx.doi.org/10.1080/00Q39896.1994.9937461.

Bellinger. D: Leviton. A: Allred. E: Rabinowitz. M. (1994b). Pre- and postnatal lead exposure and behavior
problems in school-aged children. Environ Res 66: 12-30. http://dx.doi.org/10.1006/enrs.1994.1041.

Bellinger. D: Leviton. A: Sloman. J. (1990). Antecedents and correlates of improved cognitive performance in
children exposed in utero to low levels of lead. Environ Health Perspect 89: 5-11.
http://dx.doi.org/10.2307/3430890.

Bellinger. D: Leviton. A: Waternaux. C: Needleman. H: Rabinowitz. M. (1987). Longitudinal analyses of prenatal
and postnatal lead exposure and early cognitive development. N Engl J Med 316: 1037-1043.
http://dx.doi.org/10.1056/NEJM198704233161701.

Bellinger. D: Needleman. HL. (2003). Intellectual impairment and blood lead levels [Letter], N Engl J Med 349:
500-502. http://dx.doi.org/10.1056/NEJM20030731349Q515.

Bellinger. D: Sloman. J: Leviton. A: Rabinowitz. M: Needleman. HL: Waternaux. C. (1991). Low-level lead
exposure and children's cognitive function in the preschool years. Pediatrics 87: 219-227.
http://dx.doi.Org/10.1542/peds.87.2.219.

Berk. M: Williams. LJ: Andreazza. AC: Pasco. JA: Dodd. S: Jacka. FN: Movlan. S: Reiner. EJ: Magalhaes. PVS.
(2014). Pop, heavy metal and the blues: Secondary analysis of persistent organic pollutants (POP), heavy
metals and depressive symptoms in the NHANES National Epidemiological Survey. BMJ Open 4:
e005142. http://dx.doi.org/10.1136/bmiopen-2014-005142.

Betharia. S: Maher. TJ. (2012). Neurobehavioral effects of lead and manganese individually and in combination in
developmentally exposed rats. Neurotoxicology 33: 1117-1127.
http://dx.doi.Org/10.1016/i.neuro.2012.06.002.

Bhattacharva. A: Shukla. R: Dietrich. K: Bornschein. R: Berger. O. (1995). Effect of early lead exposure on
children's postural balance. Dev Med Child Neurol 37: 861-878. http://dx.doi.org/10.1111/i. 1469-
8749.1995.tb 11939.x.

Bhattacharva. A: Shukla. R: Dietrich. KN: Bornschein. RL. (2006). Effect of early lead exposure on the maturation
of children's postural balance: A longitudinal study. Neurotoxicol Teratol 28: 376-385.
http://dx.doi.Org/10.1016/i.ntt.2006.02.003.

Bhattacharvva. MH. (1983). Bioavailability of orally administered cadmium and lead to the mother, fetus, and
neonate during pregnancy and lactation: An overview. Sci Total Environ 28: 327-342.
http://dx.doi.org/10.1016/S0048-9697(83)80030-8.

Bihaai. SW: Zawia. NH. (2013). Enhanced taupathy and AD-like pathology in aged primate brains decades after

infantile exposure to lead (Pb). Neurotoxicology 39: 95-101. http://dx.doi.Org/10.1016/i.neuro.2013.07.010.

Biioor. AR: Sudha. S: Venkatesh. T. (2012). Neurochemical and neurobehavioral effects of low lead exposure on
the developing brain. Indian J ClinBiochem 27: 147-151. http://dx.doi.org/10.1007/sl2291-012-019Q-2.

3-505

-------
Blackowicz. MJ: Hrvhorc/uk. DO: Rankin. KM: Lewis. DA: Haider. D: Lanphear. BP: Evens. A. (2016). The

impact of low-level lead toxicity on school performance among Hispanic subgroups in the Chicago Public
Schools. Int J Environ Res Public Health 13: 774. http://dx.doi.org/10.3390/iierphl3080774.

Blair. RJR. (2001). Neurocognitive models of aggression, the antisocial personality disorders, and psychopathy. J
Neurol Neurosurg Psychiatry 71: 727-731. http://dx.doi.Org/10.1136/innp.71.6.727.

Blaurock-Busch. E: Amin. OR: Rabah. T. (2011). Heavy metals and trace elements in hair and urine of a sample of
arab children with autistic spectrum disorder. Maedica 6: 247-257.

Bouchard. MF: Bellinger. DC: Weuve. J: Matthews-Bellinger. J: Gilman. SE: Wright. RO: Schwartz. J: Weisskopf.
MG. (2009). Blood lead levels and major depressive disorder, panic disorder, and generalized anxiety
disorder in US young adults. Arch Gen Psychiatry 66: 1313-1319.
http://dx.doi.org/10.1001/archgenpsvcliiatrv.20Q9.164.

Boucher. O: Burden. MJ: Muckle. G: Saint-Amour. D: Avotte. P: Dewaillv. E: Nelson. CA: Jacobson. SW:

Jacobson. JL. (2012a). Response inhibition and error monitoring during a visual go/no-go task in Inuit
children exposed to lead, polychlorinated biphenyls, and methylmercury. Environ Health Perspect 120:
608-615. http://dx.doi.org/10.1289/ehp.1103828."

Boucher. O: Jacobson. SW: Plusauellec. P: Dewaillv. E: Avotte. P: Forget-Dubois. N: Jacobson. JL: Muckle. G.

(2012b). Prenatal methylmercury, postnatal lead exposure, and evidence of attention deficit/hyperactivity
disorder among Inuit children in Arctic Quebec. Environ Health Perspect 120: 1456-1461.
http://dx.doi.org/10.1289/ehp. 1204976.

Boucher. O: Muckle. G: Avotte. P: Dewaillv. E: Jacobson. SW: Jacobson. JL. (2016). Altered fine motor function at
school age in Inuit children exposed to PCBs, methylmercury, and lead. Environ Int 95: 144-151.
http://dx.doi.Org/10.1016/i.envint.2016.08.010.

Boutwell. BB: Nelson. EJ: Oian. Z: Vaughn. MG: Wright. JP: Beaver. KM: Barnes. JC: Petkovsek. M: Lewis. R:
Schootman. M: Rosenfeld. R. (2017). Aggregate-level lead exposure, gun violence, homicide, and rape.
PLoS ONE 12: e0187953. http://dx.doi.org/10.1371/iournal.pone.0187953.

Bozack. AK: Rifas-Shiman. SL: Coull. BA: Baccarelli. AA: Wright. RO: Amarasiriwardena. C: Gold. PR: Oken. E:
Hivert. MF: Cardenas. A. (2021). Prenatal metal exposure, cord blood DNA methylation and persistence in
childhood: An epigenome-wide association study of 12 metals. Clinical Epigenetics 13: 208.
http://dx.doi.org/10.1186/sl3148-021-01198-z. '

Braun. JM: Froehlich. TE: Daniels. JL: Dietrich. KN: Hornung. R: Auinger. P: Lanphear. BP. (2008). Association of
enviromnental toxicants and conduct disorder in U.S. children: NHANES 2001-2004. Environ Health
Perspect 116: 956-962. http://dx.doi.org/10.1289/ehp. 11177.

Braun. JM: Hornung. R: Chen. A: Dietrich. KN: Jacobs. DE: Jones. R: Khourv. JC: Liddv-Hicks. S: Morgan. S:
Vanderbeek. SB: Xu. Y: Yolton. K: Lanphear. BP. (2018). Effect of residential lead-hazard interventions
on childhood blood lead concentrations and neurobehavioral outcomes: A randomized clinical trial. JAMA
Pediatr 172: 934-942. http://dx.doi.org/10.1001/iamapediatrics.2Q18.2382.

Braun. JM: Kalin. RS: Froehlich. T: Auinger. P: Lanphear. BP. (2006). Exposures to enviromnental toxicants and
attention deficit hyperactivity disorder in U.S. children. Environ Health Perspect 114: 1904-1909.
http://dx.doi.org/10.1289/ehp.9478.

Brockel. BJ: Corv-Slechta. DA. (1998). Lead, attention, and impulsive behavior: Changes in a fixed-ratio waiting-
for-reward paradigm. Pharmacol Biochem Behav 60: 545-552. http://dx.doi.org/10.1016/S0Q91-
3057(98)00023-9.

Brockel. BJ: Corv-Slechta. DA. (1999). The effects of postweaning low-level Pb exposure on sustained attention: A
study of target densities, stimulus presentation rate, and stimulus predictability. Neurotoxicology 20: 921-

933."

Burns. JM: Baghurst. PA: Sawyer. MG: Mcmichael. AJ: Tong. SL. (1999). Lifetime low-level exposure to

enviromnental lead and children's emotional and behavioral development at ages 11-13 years: The Port
Pirie cohort study. Am J Epidemiol 149: 740-749. http://dx.doi.org/10.1093/oxfordiournals.aie.a009883.

3-506

-------
Cai. OL: Peng. DJ: Zhao. L: Chen. JW: Li. Y: Luo. HL: Ou. SY: Huang. ML: Jiang. YM. (2021). Impact of lead

exposure on thyroid status and iq performance among school-age children living nearby a lead-zinc mine in
China. Neurotoxicology 82: 177-185. http://dx.doi.Org/10.1016/i.neuro.2020.10.010.

Canfield. RL: Gendle. MH: Corv-Slechta. DA. (2004). Impaired neuropsychological functioning in lead-exposed
children. Dev Neuropsychol 26: 513-540. http://dx.doi.org/10.12Q7/sl5326942dn2601 8.

Canfield. RL: Henderson. CR. Jr: Corv-Slechta. DA: Cox. C: Jusko. TA: Lanphear. BP. (2003a). Intellectual

impairment in children with blood lead concentrations below 10 |ig per deciliter. N Engl J Med 348: 1517-
1526. http://dx.doi.org/10.1056/NEJMoa022848.

Canfield. RL: Kreher. DA: Cornwell. C: Henderson. CR. Jr. (2003b). Low-level lead exposure, executive
functioning, and learning in early childhood. Child Neuropsychol 9: 35-53.
http://dx.doi.Org/10.1076/cliin.9.l.35.14496.

Cao. XJ: Huang. SH: Wang. M: Chen. JT: Ruan. DY. (2008). S-adenosyl-L-methionine improves impaired

hippocampal long-term potentiation and water maze performance induced by developmental lead exposure
in rats. Eur J Pharmacol 595: 30-34. http://dx.doi.Org/10.1016/i.eiphar.2008.07.061.

Carlson. K: Schacht. J: Neitzel. RL. (2018). Assessing ototoxicity due to chronic lead and cadmium intake with and
without noise exposure in the mature mouse. J Toxicol Environ Health A 81: 1041-1057.
http://dx.doi.org/10.1080/15287394.2018.152132Q.

Casiens. S: Pesch. B: van Thriel. C: Zschiesche. W: Behrens. T: Weiss. T: Pallapies. D: Arendt. M: Dragano. N:
Moebus. S: Jockel. KH: Briining. T. (2018). Associations between blood lead, olfaction and fine-motor
skills in elderly men: Results from the Heinz Nixdorf Recall Study. Neurotoxicology 68: 66-72.
http://dx.doi.Org/10.1016/i.neuro.2018.06.013.

Cecil. KM. (2011). Effects of early low-level lead exposure on human brain structure, organization and functions. J
Dev Orig Health Dis 2: 17-24. http://dx.doi.org/10.1017/S204017441000Q486.

Chan. TJH: Gutierrez. C: Qgunseitan. OA. (2015). Metallic burden of deciduous teeth and childhood behavioral
deficits. Int J Environ Res Public Health 12: 6771-6787. http://dx.doi.org/10.3390/iierphl20606771.

Chandramouli. K: Steer. CD: Ellis. M: Emond. AM. (2009). Effects of early childhood lead exposure on academic
performance and behaviour of school age children. Arch Dis Child 94: 844-848.
http://dx.doi.org/10.1136/adc.2008.149955.

Chen. A: Cai. B: Dietrich. KN: Radcliffe. J: Rogan. WJ. (2007). Lead exposure, IQ, and behavior in urban 5- to 7-
year-olds: Does lead affect behavior only by lowering IQ? Pediatrics 119: e650-e658.
http://dx.doi.org/10.1542/peds.2006-1973.

Chen. A: Dietrich. KN: Ware. JH: Radcliffe. J: Rogan. WJ. (2005). IQ and blood lead from 2 to 7 years of age: Are
the effects in older children the residual of high blood lead concentrations in 2-year-olds? Environ Health
Perspect 113: 597-601. http://dx.doi.org/10.1289/ehp.7625.

Chibowska. K: Korbecki. J: Gutowska. I: Metrvka. E: Tarnowski. M: Goschorska. M: Barczak. K: Chlubek. D:

Baranowska-Bosiacka. I. (2020). Pre- and neonatal exposure to lead (Pb) induces neuroinflammation in the
forebrain cortex, hippocampus and cerebellum of rat pups. International Journal of Molecular Sciences 21:
1083. http://dx.doi.org/10.3390/iims21031Q83.

Chichinadze. K: Chichinadze. N: Lazarashvili. A. (2011). Hormonal and neurochemical mechanisms of aggression
and a new classification of aggressive behavior. Aggression and Violent Behavior 16: 461-471.
http://dx.doi.Org/10.1016/i.avb.2011.03.002.

Cliiodo. LM: Covington. C: Sokol. RJ: Hannigan. JH: Jannise. J: Ager. J: Greenwald. M: Delanev-Black. V. (2007).
Blood lead levels and specific attention effects in young children. Neurotoxicol Teratol 29: 538-546.
http://dx.doi.Org/10.1016/i.ntt.2007.04.001.

Cho. SC: Kim. BN: Hong. YC: Shin. MS: Yoo. H.T: Kim. JW: Bhang. SY: Cho. TH: Kim. HW (2010). Effect of
enviromnental exposure to lead and tobacco smoke on inattentive and hyperactive symptoms and
neurocognitive performance in children. J Child Psychol Psychiatry 51: 1050-1057.
http://dx.doi.Org/10.llll/i.1469-7610.2010.02250.x.

3-507

-------
Choi. JW: Jung. AH: Nam. S: Kim. K: Kim. J: Kim. S: Kim. BN: Kim. JW. (2020). Interaction between lead and

noradrenergic genotypes affects neurocognitive functions in attention-deficit/hyperactivity disorder: A case
control study. BMC Psychiatry 20: 407. http://dx.doi.org/10.1186/sl2888-020-02799-3.

Choi. WJ: Kwoa HJ: Lim. MH: Lim. JA: Ha. M. (2016). Blood lead, parental marital status and the risk of

attention-deficit/hyperactivity disorder in elementary school children: A longitudinal study. Psychiatry Res
236: 42-46. http://dx.doi.Org/10.1016/i.psvclires.2016.01.002.

Choi. YH: Hu. H: Mukheriee. B: Miller. J: Park. SK. (2012). Enviromnental cadmium and lead exposures and

hearing loss in U.S. adults: The National Health and Nutrition Examination Survey, 1999 to 2004. Environ
Health Perspect 120: 1544-1550. http://dx.doi.org/10.1289/ehp.1104863.

Choi. YH: Park. SK. (2017). Enviromnental exposures to lead, mercury, and cadmium and hearing loss in adults and
adolescents: KNHANES 2010-2012. Environ Health Perspect 125: 067003.
http://dx.doi.org/10.1289/EHP565.

Constantino. JN. (2011). The quantitative nature of autistic social impairment [Review]. Pediatr Res 69: 55R-62R.
http://dx.doi.org/10.12Q3/PDR.0b013e318212ec6e.

Cooper. GP: Manalis. RS. (1984). Interactions of lead and cadmium on acetylcholine release at the frog
neuromuscular junction. Toxicol Appl Pharmacol 74: 411-416. http://dx.doi.org/10.1016/0Q41-
008X(84)90294-1.

Cooper. GS: Lunn. RM: Agerstrand. M: Glenn. BS: Kraft. AD: Luke. AM: Ratcliffe. JM. (2016). Study sensitivity:
Evaluating the ability to detect effects in systematic reviews of chemical exposures. Environ Int 92-93:
605-610. http://dx.doi.Org/10.1016/i.envint.2016.03.017.

Corv-Slechta. DA: Merchant-Borna. K: Allen. JL: Liu. S: Weston. D: Conrad. K. (2013). Variations in the nature of
behavioral experience can differentially alter the consequences of developmental exposures to lead,
prenatal stress, and the combination. Toxicol Sci 131: 194-205. http://dx.doi.org/10.1093/toxsci/kfs260.

Corv-Slechta. DA: Virgolini. MB: Liu. S: Weston. D. (2012). Enhanced stimulus sequence-dependent repeated

learning in male offspring after prenatal stress alone or in conjunction with lead exposure. Neurotoxicology
33: 1188-1202. http://dx.doi.Org/10.1016/i.neuro.2012.06.013.

Counter. SA: Buchanan. LH: Ortega. F. (2008). Zinc protoporphyrin levels, blood lead levels and neurocognitive
deficits in Andean children with chronic lead exposure. ClinBiochem 41: 41-47.
http://dx.doi.Org/10.1016/i.clinbiochem.2007.10.002.

Covle. JT: Tsai. G. (2004). The NMD A receptor glycine modulatory site: A therapeutic target for improving

cognition and reducing negative symptoms in schizophrenia [Review]. Psychophannacology 174: 32-38.
http://dx.doi.Org/10.1007/s00213-003-1709-2.

Crump. KS: Van Landingham. C: Bowers. TS: Cahov. D: Chandalia. JK. (2013). A statistical reevaluation of the
data used in the Lanphear et al. (2005) pooled-analysis that related low levels of blood lead to intellectual
deficits in children [Review]. Crit Rev Toxicol 43: 785-799.
http://dx.doi.org/10.3109/10408444.2Q13.832726.

da Silva. DRF: Bittencourt. LO: Aragao. WAB: Nascimento. PC: Leao. LKR: Oliveira. ACA: Crespo-Lopez. ME:
Lima. RR. (2020). Long-term exposure to lead reduces antioxidant capacity and triggers motor neurons
degeneration and demyelination in spinal cord of adult rats. Ecotoxicol Environ Saf 194: 110358.
http://dx.doi.Org/10.1016/i.ecoenv.2020.110358.

Dabrowska. A: Luis Venero. J: Iwasawa. R: Hankir. MK: Rahman. S: Boobis. A: Haiii. N. (2015). PGC-la controls
mitochondrial biogenesis and dynamics in lead-induced neurotoxicity. Aging 7: 629-647.
http://dx.doi.org/10.18632/aging.100790.

Dantzer. J: Ryan. P: Yolton. K: Parsons. PJ: Palmer. CD: Cecil. K: Unrine. JM. (2020). A comparison of blood and
toenails as biomarkers of children's exposure to lead and their correlation with cognitive function. Sci Total
Environ 700: 134519. http://dx.doi.Org/10.1016/i.scitotenv.2019.134519.

David. OJ: Hoffman. SP: Sverd. J: Clark. J: Voeller. K. (1976). Lead and hyperactivity. Behavorial response to
chelation: A pilot study. Am J Psychiatry 133: 1155-1158. http://dx.doi.org/10.1176/aip. 133.10.1155.

3-508

-------
De Marco. M: Halpern. R: Barros. HMT. (2005). Early behavioral effects of lead perinatal exposure in rat pups.
Toxicology 211: 49-58. http://dx.doi.Org/10.1016/i.tox.2005.02.007.

De Palma. G: Catalan! S: Franco. A: Brighenti. M: Apostoli. P. (2012). Lack of correlation between metallic
elements analyzed in hair by ICP-MS and autism. J Autism Dev Disord 42: 342-353.
http://dx.doi.org/10.1007/si08Q3-011-1245-6.

Despres. C: Beuter. A: Richer. F: Poitras. K: Veilleux. A: Avotte. P: Dewaillv. E: Saint-Amour. D: Muckle. G.

(2005).	Neuromotor functions in Inuit preschool children exposed to Pb, PCBs, and Hg. Neurotoxicol
Teratol 27: 245-257. http://dx.doi.Org/10.1016/i.ntt.2004.12.001.

Desrochers-Couture. M: Courtemanche. Y: Forget-Dubois. N: Belanger. RE: Boucher. O: Avotte. P: Cordier. S:
Jacobson. JL: Jacobson. SW: Muckle. G. (2019). Association between early lead exposure and
externalizing behaviors in adolescence: A developmental cascade. Environ Res 178: 108679.
http://dx.doi.Org/10.1016/i.envres.2019.108679.

Desrochers-Couture. M: Oulhote. Y: Arbuckle. TE: Fraser. WD: Seguin. JR: Quelle! E: Forget-Dubois. N: Avotte.
P: Boivin. M: Lanphear. BP: Muckle. G. (2018). Prenatal, concurrent, and sex-specific associations
between blood lead concentrations and IQ in preschool Canadian children. Environ Int 121: 1235-1242.
http://dx.doi.Org/10.1016/i.envint.2018.10.043.

Pi Paolo. G: Kim. TW. (2011). Linking lipids to Alzheimer's disease: Cholesterol and beyond [Review]. Nat Rev
Neurosci 12: 284-296. http://dx.doi.org/10.1038/nrn3012.

Dietrich. KN: Berger. OG: Succop. PA. (1993). Lead exposure and the motor developmental status of urban six-
year-old children in the Cincinnati Prospective Study. Pediatrics 91: 301-307.
http://dx.doi.Org/10.1542/peds.91.2.301.

Dietrich. KN: Douglas. RM: Succop. PA: Berger. OG: Bornschein. RL. (2001). Early exposure to lead and juvenile
delinquency. Neurotoxicol Teratol 23: 511-518. http://dx.doi.org/10.1016/S0892-0362(01)00184-2.

Dietrich. KN: Krafft. KM: Shukla. R: Bornschein. RL: Succop. PA. (1987). The neurobehavioral effects of early
lead exposure. In SR Schroeder (Ed.), Toxic substances and mental retardation: Neurobehavioral
toxicology and teratology (pp. 71-95). Washington DC: American Association on Mental Deficiency.
https://www.scopus.com/inward/record.uri?eid=2-s2.0-
0023076878&partnerID=40&md5=8ccbl8d53f25b5ab0be523a3e6fld4ec.

Dietrich. KN: Succop. PA: Berger. OG: Hammond. PB: Bornschein. RL. (1991). Lead exposure and the cognitive
development of urban preschool children: The Cincinnati lead study cohort at age 4 years. Neurotoxicol
Teratol 13: 203-211. http://dx.doi.org/10.1016/0892-0362(91)90012-L.

Dietrich. KN: Succop. PA: Berger. OG: Keith. RW. (1992). Lead exposure and the central auditory processing
abilities and cognitive development of urban children: The Cincinnati lead study cohort at age 5 years.
Neurotoxicol Teratol 14: 51-56. http://dx.doi.org/10.1016/0892-0362(92)90028-9.

Diouf. A: Garcon. G: Diop. Y: Ndiave. B: Tliiaw. C: Fall. M: Kane-Barry. O: Ba. D: Haguenoer. JM: Shirali. P.

(2006).	Enviromnental lead exposure and its relationship to traffic density among Senegalese children: A
cross-sectional study. Hum Exp Toxicol 25: 637-644. http://dx.doi.org/10.1177/0960327106Q74591.

Dohertv. BT: Romano. ME: Gui. J: Punshon. T: Jackson. BP: Karagas. MR: Korrick. SA. (2020). Periconceptional
and prenatal exposure to metal mixtures in relation to behavioral development at 3 years of age.
Enviromnental Epidemiology 4: e0106. http://dx.doi.org/10.1097/EE9.00000000000001Q6.

Dominguez. S: Flores-Montova. MG: Sobin. C. (2019). Early chronic exposure to low-level lead alters total
hippocampal microglia in pre-adolescent mice. Toxicol Lett 302: 75-82.
http://dx.doi.Org/10.1016/i.toxlet.2018.10.016.

Donald. JM: Cutler. MG: Moore. MR. (1986). Effects of lead in the laboratory mouse: 1. Influence of pregnancy
upon absorption, retention, and tissue distribution of radiolabeled lead. Environ Res 41: 420-431.
http://dx.doi.org/10.1016/S0Q 13-9351(86)80136-0.

3-509

-------
Donald. JM: Cutler. MG: Moore. MR. (1987). Effects of lead in the laboratory mouse. Development and social
behaviour after lifelong exposure to 12 (iM lead in drinking fluid. Neuropharmacology 26: 391-399.
http://dx.doi.org/10.1016/0028-3908(87)90194-8.

Dong. HY: Feng. JY: Li. HH: Yue. XJ: Jia. FY. (2022). Non-parental caregivers, low maternal education,

gastrointestinal problems and high blood lead level: Predictors related to the severity of autism spectrum
disorder in Northeast China. BMC Pediatr 22: 11. http://dx.doi.org/10.1186/sl2887-021-03086-0.

Du. Y: Ge. MM: Xue. W: Yang. OO: Wang. S: Xu. Y: Wang. HL. (2015). Chronic lead exposure and mixed factors
of genderxagexbrain regions interactions on dendrite growth, spine maturity and NDR kinase. PLoS ONE
10: e0138112. http://dx.doi.org/10.1371/iournal.pone.0138112.

Duan. Y: Peng. L: Shi. H: Jiang. Y. (2017). The effects of lead on GABAergic interneurons in rodents. Toxicol Ind
Health 33: 867-875. http://dx.doi.org/10.1177/0748233717732902.

Dumkova. J: Smutna. T: Vrlikova. L: Le Coustumer. P: Vecera. Z: Docekal. B: Mikuska. P: Capka. L: Fictum. P:
Hampl. A: Buchtova. M. (2017). Sub-chronic inhalation of lead oxide nanoparticles revealed their broad
distribution and tissue-specific subcellular localization in target organs. Part Fibre Toxicol 14: 55.
http://dx.doi.org/10.1186/sl2989-017-0236-Y.

Ebrahimzadeh-Bideskan. AR: Hami. J: Alipour. F: Hagliir. H: Fazel. AR: Sadeghi. A. (2016). Protective effects of
ascorbic acid and garlic extract against lead-induced apoptosis in developing rat hippocampus. Metab Brain
Dis 31: 1123-1132. http://dx.doi.org/10.1007/sl 1011-016-9837-7.

Egan. KB: Cornwell. CR: Courtney. JG: Ettinger. AS. (2021). Blood lead levels in U.S. children ages 1-11 years,
1976-2016. Environ Health Perspect 129: 37003. http://dx.doi.org/10.1289/EHP7932.

Emer. LR: Kalkbrenner. AE: O'Brien. M: Yan. A: Cisler. RA: Weinhardt. L. (2020). Association of childhood blood
lead levels with firearm violence perpetration and victimization in Milwaukee. Environ Res 180: 108822.
http://dx.doi.Org/10.1016/i.envres.2019.108822.

Ercal. N: Treeratphan. P: Hammond. TC: Matthews. RH: Grannemann. NH: Spitz. DR. (1996). In vivo indices of

oxidative stress in lead-exposed C57BL/6 mice are reduced by treatment with meso-2,3-dimercaptosuccinic
acid or N-acetylcysteine. Free Radic Biol Med 21: 157-161. http://dx.doi.org/10.1016/Q891-
5849(96)00020-2.

Etliier. AA: Muckle. G: Bastien. C: Dewaillv. E: Avotte. P: Arfken. C: Jacobson. SW: Jacobson. JL: Saint-Amour.
D. (2012). Effects of enviromnental contaminant exposure on visual brain development: A prospective
electrophysiological study in school-aged children. Neurotoxicology 33: 1075-1085.
http://dx.doi.Org/10.1016/i.neuro.2012.05.010.

Etliier. AA: Muckle. G: Jacobson. SW: Avotte. P: Jacobson. JL: Saint-Amour. D. (2015). Assessing new dimensions
of attentional functions in children prenatally exposed to enviromnental contaminants using an adapted
Posner paradigm. Neurotoxicol Teratol 51: 27-34. http://dx.doi.Org/10.1016/i.ntt.2015.07.005.

Ettinger. AS: Egan. KB: Homa. DM: Brown. MJ. (2020). Blood lead levels in U.S. women of childbearing age,
1976-2016. Environ Health Perspect 128: 17012. http://dx.doi.org/10.1289/EHP5925.

Eum. KD: Korrick. SA: Weuve. J: Okereke. O: Kubzanskv. LP: Hu. H: Weisskopf. MG. (2012). Relation of

cumulative low-level lead exposure to depressive and phobic anxiety symptom scores in middle-age and
elderly women. Environ Health Perspect 120: 817-823. http://dx.doi.org/10.1289/ehp. 1104395.

Evens. A: Hrvhorczuk. D: Lanphear. BP: Rankin. KM: Lewis. DA: Forst. L: Rosenberg. D. (2015). The impact of
low-level lead toxicity on school performance among children in the Chicago Public Schools: A
population-based retrospective cohort study. Environ Health 14: 21. http://dx.doi.org/10.1186/sl2940-015-
0008-9.

Fan. G: Feng. C: Li. Y: Wang. CH: Yan. J: Li. W: Feng. JG: Shi. XL: Bi. YY. (2009). Selection of nutrients for
prevention or amelioration of lead-induced learning and memory impairment in rats. Ann Occup Hyg 53:
341-351. http://dx.doi.org/10.1093/annlivg/mepO 19.

3-510

-------
Fan. G: Feng. C: Wu. F: Ye. W: Lin. F: Wang. C: Yan. J: Zhu. G: Xiao. Y: Bi. Y. (2010). Methionine choline

reverses lead-induced cognitive and N-methyl-d-aspartate receptor subunit 1 deficits. Toxicology 272: 23-
31. http://dx.doi.Org/10.1016/i.tox.2010.03.018.

Fan. Y: Sheng. J: Liang. C: Yang. L: Liu. K: Wang. O: Zhang. D: Ma. Y: Li. X: Xie. S: Cao. H: Wang. S: Tao. F.
(2020). Association of blood lead levels with the risk of depressive symptoms in the elderly Chinese
population: Baseline data of a cohort study. Biol Trace Elem Res 194: 76-83.
http://dx.doi.org/10.1007/sl2011-019-Q1755-x.

Fang. F: Kwee. LC: Allea KD: Umbach. DM: Ye. W: Watsoa M: Keller. J: Oddone. EZ: Sandler. DP: Schmidt. S:
Kamel. F. (2010). Association between blood lead and the risk of amyotrophic lateral sclerosis. Am J
Epidemiol 171: 1126-1133. http://dx.doi.org/10.1093/aie/kwa063.

Fang. F: Peters. TL: Beard. JD: Umbach. DM: Keller. J: Mariosa. D: Allen. KD: Ye. W: Sandler. DP: Schmidt. S:

Kamel. F. (2017). Blood lead, bone turnover, and survival in amyotrophic lateral sclerosis. Am J Epidemiol
186: 1057-1064. http://dx.doi.org/10.1093/aie/kwxl76.

Farooaui. Z: Bakulski. KM: Power. MC: Weisskopf. MG: Sparrow. D: Spiro. A. Ill: Vokonas. PS: Nie. LH: Hu. H:
Park. SK. (2017). Associations of cumulative Pb exposure and longitudinal changes in Mini-Mental Status
Exam scores, global cognition and domains of cognition: The VA Normative Aging Study. Environ Res
152: 102-108. http://dx.doi.org/10.1016/i.envres.2016.10.007.

Faulk. C: Barks. A: Sanchez. BN: Zhang. ZZ: Anderson. OS: Peterson. KE: Dolinov. DC. (2014). Perinatal lead
(Pb) exposure results in sex-specific effects on food intake, fat, weight, and insulin response across the
murine life-course. PLoS ONE 9: el04273. http://dx.doi.org/10.1371/iournal.pone.01Q4273.

Feng. C: Liu. S: Zhou. F: Gao. Y: Li. Y: Du. G: Chen. Y: Jiao. H: Feng. .1: Zhang, Y: Bo. D: Li. Z: Fan. G. (2019).
Oxidative stress in the neurodegenerative brain following lifetime exposure to lead in rats: Changes in
lifespan profiles. Toxicology 411: 101-109. http://dx.doi.Org/10.1016/i.tox.2018.ll.003.

Fergusson. DM: Boden. JM: Horwood. LJ. (2008). Dentine lead levels in childhood and criminal behaviour in late
adolescence and early adulthood. J Epidemiol Community Health 62: 1045-1050.
http://dx.doi.org/10.il36/iech.2007.072827.

Fergusson. DM: Horwood. LJ: Lvnskev. MT. (1993). Early dentine lead levels and subsequent cognitive and
behavioural development. J Child Psychol Psychiatry 34: 215-227. http://dx.doi.org/10. Ill 1/i. 1469-
7610.1993.tb00980.x.

Fergusson. DM: Horwood. LJ: Lvnskev. MT. (1997). Early dentine lead levels and educational outcomes at 18
years. J Child Psychol Psychiatry 38: 471-478. http://dx.doi.org/10.Ill 1/i. 1469-7610.1997.tbO 1532.x.

Ferlemi. AY: Avgoustatos. D: Kokkosis. AG: Protonotarios. V: Constantinou. C: Margaritv. M. (2014). Lead-
induced effects on learning/memory and fear/anxiety are correlated with disturbances in specific
cholinesterase isofonn activity and redox imbalance in adult brain. Physiol Behav 131: 115-122.
http://dx.doi.Org/10.1016/i.phvsbeh.2014.04.033.

Fields. VL: Soke. GN: Reynolds. A: Tiaa LH: Wiggins. L: Maenner. M: DiGuiseppi. C: Krai. TVE: Hightshoe. K:
Schieve. LA. (2021). Pica, autism, and other disabilities. Pediatrics 147: e20200462.
http://dx.doi.org/10.1542/peds.2020-0462.

Fillion. M: Lemire. M: Philibert. A: Frenette. B: Weiler. HA: Davee Guimaraes. JR: Larribe. F: Barbosa. F. Jr:

Mergler. D. (2013). Toxic risks and nutritional benefits of traditional diet on near visual contrast sensitivity
and color vision in the Brazilian Amazon. Neurotoxicology 37: 173-181.
http://dx.doi.Org/10.1016/i.neuro.2013.04.010.

Filon. J: Ustvmowicz-Farbiszewska. J: Kraiewska-Kulak. E. (2020). Analysis of lead, arsenic and calcium content in
the hair of children with autism spectrum disorder. BMC Public Health 20: 383.
http://dx.doi.org/10.1186/sl2889-020-08496-w.

Flora. SJ: Gautam. P: Kushwaha. P. (2012). Lead and ethanol co-exposure lead to blood oxidative stress and

subsequent neuronal apoptosis in rats. Alcohol Alcohol 47: 92-101. http://dx.doi.org/10.1093/alcalc/agrl52.

3-511

-------
Flores-Montova. MG: Alvarez. JM: Sobin. C. (2015). Olfactory recognition memory is disrupted in young mice
with chronic low-level lead exposure. Toxicol Lett 236: 69-74.
http ://dx. doi.org/10.1016/i .toxlet.2015.04.013.

Flores-Montova. MG: Sobin. C. (2015). Early chronic lead exposure reduces exploratory activity in young
C57BL/6J mice. J Appl Toxicol 35: 759-765. http://dx.doi.org/10.1002/iat.3064.

Forns. J: Fort. M: Casas. M: Caceres. A: Guxens. M: Gascon. M: Garcia-Esteban. R: Julvez. J: Grimalt. JO: Sunver.
J. (2014). Exposure to metals during pregnancy and neuropsychological development at the age of 4 years.
Neurotoxicology 40: 16-22. http://dx.doi.Org/10.1016/i.neuro.2013.10.006.

Fox. DA: Campbell. ML: Blocker. YS. (1997). Functional alterations and apoptotic cell death in the retina following
developmental or adult lead exposure. Neurotoxicology 18: 645-664.

Fox. DA: Kala. SV: Hamilton. WR: Johnson. JE: O'Callaghan. JP. (2008). Low-level human equivalent gestational
lead exposure produces supernormal scotopic electroretinograms, increased retinal neurogenesis, and
decreased retinal dopamine utilization in rats. Environ Health Perspect 116: 618-625.
http://dx.doi.org/10.1289/ehp. 11268.

Fox. DA: Qpanashuk. L: Zharkovskv. A: Weiss. B. (2010). Gene-chemical interactions in the developing

mammalian nervous system: Effects on proliferation, neurogenesis and differentiation. Neurotoxicology
31: 589-597. http://dxidoi.Org/10.1016/i.neuro.2010.03.007.

Frank. AC: Huang. S: Zhou. M: Gdalvahu. A: Kastellakis. G: Silva. TK: Lu. E: Wen. X: Poirazi. P: Trachtenberg.
JT: Silva. AJ. (2018). Hotspots of dendritic spine turnover facilitate clustered spine addition and learning
and memory. Nat Commun 9: 422. http://dx.doi.org/10.1038/s41467-017-02751-2.

Froehlich. TE: Lanphear. BP: Auinger. P: Hornung. R: Epstein. JN: Braun. J: Kalin. RS. (2009). Association of
tobacco and lead exposures with attention-deficit/hyperactivity disorder. Pediatrics 124: E1054-E1063.
http://dx.doi.org/10.1542/peds.2009-0738.

Froehlich. TE: Lanphear. BP: Dietrich. KN: Corv-Slechta. DA: Wang. N: Kalin. RS. (2007). Interactive effects of a
DRD4 polymorphism, lead and sex on executive functions in children. Biol Psychiatry 62: 243-249.
http://dx.doi.Org/10.1016/i.biopsvch.2006.09.039.

Fruh. V: Rifas-Shiman. SL: Amarasiriwardena. C: Cardenas. A: Bellinger. DC: Wise. LA: White. RF: Wright. RO:
Oken. E: Henn. BC. (2019). Prenatal lead exposure and childhood executive function and behavioral
difficulties in project viva. Neurotoxicology 75: 105-115. http://dx.doi.Org/10.1016/i.neuro.2019.09.006.

Fruh. V: Rifas-Shiman. SL: Coull. BA: Devick. KL: Amarasiriwardena. C: Cardenas. A: Bellinger. DC: Wise. LA:
White. RF: Wright. RO: Oken. E: Henn. BC. (2021). Prenatal exposure to a mixture of elements and
neurobehavioral outcomes in mid-childhood: Results from Project Viva. Environ Res 201: 111540.
http://dx.doi.Org/10.1016/i.envres.2021.111540.

Galal. MK: Elleithv. EMM: Abdrabou. MI: Yasin. NAE: Shaheen. YM. (2019). Modulation of caspase-3 gene
expression and protective effects of garlic and spirulina against CNS neurotoxicity induced by lead
exposure in male rats. Neurotoxicology 72: 15-28. http://dx.doi.Org/10.1016/i.neuro.2019.01.006.

Gassowska. M: Baranowska-Bosiacka. I: Moczvdlowska. J: Frontczak-Baniewicz. M: Gewartowska. M:

Struzvnska. L: Gutowska. I: Chlubek. D: Adamczvk. A. (2016a). Perinatal exposure to lead (Pb) induces
ultrastructural and molecular alterations in synapses of rat offspring. Toxicology 373: 13-29.
http://dx.doi.Org/10.1016/i.tox.2016.10.014.

Gassowska. M: Baranowska-Bosiacka. I: Moczvdlowska. J: Tarnowski. M: Pilutin. A: Gutowska. I: Struzvnska. L:
Chlubek. D: Adamczvk. A. (2016b). Perinatal exposure to lead (Pb) promotes Tau phosphorylation in the
rat brain in a GSK-3(3 and CDK5 dependent manner: Relevance to neurological disorders. Toxicology 347-
349: 17-28. http://dx.doi.Org/10.1016/i.tox.2016.03.002.

GBP 2019 Mental Disorders Collaborators. (2022). Global, regional, and national burden of 12 mental disorders in
204 countries and territories, 1990-2019: A systematic analysis for the Global Burden of Disease Study
2019. Lancet Psychiatry 9: 137-150. http://dx".doi.org/10.10i6/S2215-0366(21)00395-3.

3-512

-------
Geier. DA: Kern. JK: Geier. MR. (2018). A cross-sectional study of the relationship between blood lead levels and
reported attention deficit disorder: An assessment of the economic impact on the United States. Metab
BrainDis 33: 201-208. http://dx.doi.org/10.1007/sll011-017-Q146-6.

Gerber. AJ: Peterson. BS: Giedd. JN: Lalonde. FM: Celano. MJ: White. SL: Wallace. GL: Lee. NR: Lenroot. RK.
(2009). Anatomical brain magnetic resonance imaging of typically developing children and adolescents
[Review]. J Am Acad Child Adolesc Psychiatry 48: 465-470.
http://dx.doi.org/10.1097/CHI.0b013e3i819f2715.

Gerson. AC: Butler. R: Moxev-Mims. M: Wentz. A: Shinnar. S: Lande. MB: Mendlev. SR: Waradv. BA: Furth. SL:
Hooper. SR. (2006). Neurocognitive outcomes in children with chronic kidney disease: Current findings
and contemporary endeavors [Review]. Ment Retard Dev Disabil Res Rev 12: 208-215.
http://dx.doi.org/10.1002/mrdd.20116.

Giddabasappa. A: Hamiltoa WR: Chanev. S: Xiao. W: Johnson. JE: Mukheriee. S: Fox. DA. (2011). Low-level
gestational lead exposure increases retinal progenitor cell proliferation and rod photoreceptor and bipolar
cell neurogenesis in mice. Environ Health Perspect 119: 71-77. http://dx.doi.org/10.1289/ehp. 1002524.

Gilbert. SG: Rice. DC. (1987). Low-level lifetime lead exposure produces behavioral toxicity (spatial discrimination
reversal) in adult monkeys. Toxicol Appl Pharmacol 91: 484-490. http://dx.doi.org/10.1016/0041-
008X(87)90070-6.

Gioia. GA: Isauith. PK: Retzlaff. PD: Espy. KA. (2002). Confirmatory factor analysis of the Behavior Rating
Inventory of Executive Function (BRIEF) in a clinical sample. Child Neuropsychol 8: 249-257.
http://dx.doi.Org/10.1076/cliin.8.4.249.13513.

Gipsoa PS: Wetherington. CE: Duquette. PJ: Hooper. SR. (2004). The nervous system and chronic kidney disease
in children [Review]. Pediatr Nephrol 19: 832-839. http://dx.doi.org/10.1007/sQ0467-004-1532-v.

Gittleman. R: Eskenazi. B. (1983). Lead and hyperactivity revisited: An investigation of non-disadvantaged

children. Arch Gen Psychiatry 40: 827-833. http://dx.doi.org/10.1001/archpsvc.1983.017900700170Q2.

Glenn. AL: Johnson. AK: Raine. A. (2013). Antisocial personality disorder: a current review [Review]. Curr
Psychiatry Rep 15: 427. http://dx.doi.org/10.1007/sll920-013-Q427-7.

Gomaa. A: Hu. H: Bellinger. D: Schwartz. J: Tsaih. SW: Gonzalez-Cossio. T: Schnaas. L: Peterson. K: Aro. A:
Hernandez-Avila. M. (2002). Maternal bone lead as an independent risk factor for fetal neurotoxicity: A
prospective study. Pediatrics 110: 110-118. http://dx.doi.org/10.1542/peds. 110.1.110.

Gottipolu. RR: Davuliigari. CB. (2014). Perinatal exposure to lead: Reduction in alterations of brain mitochondrial
antioxidant system with calcium supplement. Biol Trace Elem Res 162: 270-277.
http://dx.doi.org/10.1007/sl2011-014-Q112-7.

Graff-Radford. NR: Crook. JE: Lucas. J: Boeve. BF: Knopman. PS: Ivnik. RJ: Smith. GE: Younkin. LH: Petersen.
RC: Younkin. SG. (2007). Association of low plasma A beta 42/A beta 40 ratios with increased imminent
risk for mild cognitive impairment and Alzheimer disease. Arch Neurol 64: 354-362.
http://dx.doi.Org/10.1001/arclmeur.64.3.354.

Graham. PL: Grace. CE: Braun. AA: Schaefer. TL: Skelton. MR: Tang. PH: Vorhees. CV: Williams. MT. (2011).
Effects of developmental stress and lead (Pb) on corticosterone after chronic and acute stress, brain
monoamines, and blood Pb levels in rats. Int J Pev Neurosci 29: 45-55.
http://dx.doi.Org/10.1016/i.iidevneu.2010.09.008.

Grashow. R: Spiro. A: Taylor. KM: Newton. K: Shrairman. R: Landau. A: Sparrow. P: Hu. H: Weisskopf. M.

(2013). Cumulative lead exposure in community-dwelling adults and fine motor function: Comparing
standard and novel tasks in the VA Normative Aging Study. Neurotoxicology 35: 154-161.
http://dx.doi.Org/10.1016/i.neuro.2013.01.005.

Graves. AB: Van Puiin. CM: Chandra. V: Fratiglioni. L: Hevman. A: Jorm. AF: Kokmen. E: Kondo. K: Mortimer.
JA: Rocca. WA: Shalat. SL: Soininen. H: Hofrnan. A. (1991). Occupational exposures to solvents and lead
as risk factors for Alzheimer's disease: A collaborative re-analysis of case-control studies. Int J Epidemiol
2: S58-S61. http://dx.doi.org/10.1093/iie/20.Supplement 2.S58.

3-513

-------
Gu. H: Robison. G: Hong. L: Barrea. R: Wei. X: Farlow. MR: Pushkar. YN: Du. Y: Zheng. W. ei. (2012). Increased
(3-amyloid deposition in Tg-SWDI transgenic mouse brain following in vivo lead exposure. Toxicol Lett
213: 211-219. http://dx.doi.Org/10.1016/i.toxlet.2012.07.002.

Gu. X: Yuan. FF: Huang. X: Hou. YW: Wang. M: Lia J: Wu. J. (2018). Association of PIK3CG gene
polymorphisms with attention-deficit/hyperactivity disorder: A case-control study. Prog
Neuropsychophannacol Biol Psychiatry 81: 169-177. http://dx.doi.Org/10.1016/i.pnpbp.2017.10.020.

Guilarte. TR. (1997). Pb2+ inhibits NMDA receptor function at high and low affinity sites: Developmental and
regional brain expression. Neurotoxicology 18: 43-51.

Gulsoa BL: Mizon. KJ: Korsch. MJ: Palmer. JM: Donnelly. JB. (2003). Mobilization of lead from human bone

tissue during pregnancy and lactation - A summary of long-term research. Sci Total Environ 303: 79-104.
http://dx.doi.org/10.1016/S0048-9697(02)00355-8.

Gump. BB: Dvkas. MJ: Mackenzie. JA: Dumas. AK: Hruska. B: Ewart. CK: Parsons. PJ: Palmer. CD: Bendinskas.
K. (2017). Background lead and mercury exposures: Psychological and behavioral problems in children.
Environ Res 158: 576-582. http://dx.doi.Org/10.1016/i.envres.2017.06.033.

Haden. SC: Scarpa. A. (2007). The noradrenergic system and its involvement in aggressive behaviors. Aggression
and Violent Behavior 12: 1-15. http://dx.doi.Org/10.1016/i.avb.2006.01.012.

Han. XJ: Xiao. YM: Ai. BM: Hu. XX: Wei. O: Hu. OS. (2014). Effects of organic selenium on lead-induced

impairments of spatial learning and memory as well as synaptic structural plasticity in rats. Biol Phann Bull
37: 466-474. http://dx.doi.org/10.1248/bpb.bl3-00892. '

Haraguchi. T: Isliizu. H: Takehisa. Y: Kawai. K: Yokota. O: Terada. S: Tsuchiva. K: Ikeda. K: Morita. K: Horike. T:
Kira. S: Kuroda. S. (2001). Lead content of brain tissue in diffuse neurofibrillary tangles with calcification
(DNTC): The possibility of lead neurotoxicity. Neuroreport 12: 3887-3890.
http://dx.doi.org/10.1097/00001756-200112210-000Q6.

Hashemzadeh-Gargari. H: Guilarte. TR. (1999). Divalent cations modulate N-methyl-D-aspartate receptor function
at the glycine site. J Pharmacol Exp Ther 290: 1356-1362.

Havnes. EN: Sucharew. H: Kuhnell. P: Alden. J: Barnas. M: Wright. RO: Parsons. PJ: Aldous. KM: Praamsma. ML:
Beidler. C: Dietrich. KN. (2015). Manganese exposure and neurocognitive outcomes in rural school-age
children: The communities actively researching exposure study (Ohio, USA). Environ Health Perspect 123:
1066-1071. http://dx.doi.org/10.1289/ehp.1408993.

Henn. BC: Schnaas. L: Ettinger. AS: Schwartz. J: Lamadrid-Figueroa. H: Hernandez-Avila. M: Amarasiriwardena.
C: Hu. H: Bellinger. DC: Wright. RO: Tellez-Roio. MM. (2012). Associations of early childhood
manganese and lead coexposure with neurodevelopment. Environ Health Perspect 120: 126-131.
http://dx.doi.org/10.1289/ehp. 1003300.

Hilson. JA: Strupp. BJ. (1997). Analyses of response patterns clarify lead effects in olfactory reversal and

extradimensional shift tasks: Assessment of inhibitory control, associative ability, and memory. Behav
Neurosci 111: 532-542. http://dx.doi.org/10.1037/0735-7044.111.3.532.

Hong. SB: Im. MH: Kim. JW: Park. EJ: Shin. MS: Kim. BN: Yoo. HJ: Cho. IH: Bhang. SY: Hong. YC: Cho. SC.
(2015). Enviromnental lead exposure and attention-deficit/hyperactivity disorder symptom domains in a
community sample of South Korean school-age children. Environ Health Perspect 123: 271-276.
http://dx.doi.org/10.1289/ehp. 1307420.

Hong. T: Li. SM: Jia. B: Huang. Y: Shu. K: Yuan. KW: Chen. L: Li. LX: Liu. L: Liu. ZY. (2021). DNA methylation
changes in the hippocampus of learning and memory disorder offspring rats of lead exposure during
pregnant and lactation period. Ann Palliat Med 10: 1059-1069. http://dx.doi.org/10.21037/apm-19-421.

Hornung. RW: Lanphear. BP: Dietrich. KN. (2009). Age of greatest susceptibility to childhood lead exposure: A

new statistical approach. Environ Health Perspect 117: 1309-1312. http://dx.doi.org/10.1289/ehp.080Q426.

Horton. CJ: Weng. HY: Wells. EM. (2019). Association between blood lead level and subsequent Alzheimer's
disease mortality. Enviromnental Epidemiology 3: e045.
http://dx.doi.Org/10.1097/EE9.0000000000000b45.

3-514

-------
Horton. MK; Hsu. L: Henn. BC: Margolis. A: Austin. C: Svensson. K: Sclinaas. L: Gennings. C: Hu. H: Wright. R:
Tellez Roio. MM: Arora. M. (2018). Dentine biomarkers of prenatal and early childhood exposure to
manganese, zinc and lead and childhood behavior. Environ Int 121: 148-158.
http://dx.doi.Org/10.1016/i.envint.2018.08.045.

Hossain. S: Bhowmick. S: Jahan. S: Rozario. L: Sarkar. M: Islam. S: Basunia. MA: Rahman. A: Choudhurv. BK:
Shahialal. H. (2016). Maternal lead exposure decreases the levels of brain development and cognition-
related proteins with concomitant upsurges of oxidative stress, inflammatory response and apoptosis in the
offspring rats. Neurotoxicology 56: 150-158. http://dx.doi.Org/10.1016/i.neuro.2016.07.013.

Hsu. CY: Chuang. YC: Chang. FC: Chuang. HY: Cliiou. TTY: Lee. CT. (2021). Disrupted sleep homeostasis and
altered expressions of clock genes in rats with chronic lead exposure. Toxics 9: 217.
http://dx.doi.org/10.3390/toxics909Q217.

Hu. H: Hernandez-Avila. M. (2002). Invited commentary: Lead, bones, women, and pregnancy - The poison within?
[Comment]. Am J Epidemiol 156: 1088-1091. http://dx.doi.org/10.1093/aie/kwfl64.

Hu. H: Tellez-Roio. MM: Bellinger. D: Smith. D: Ettinger. AS: Lamadrid-Figueroa. H: Schwartz. J: Sclinaas, L:
Mercado-Garcia. A: Hernandez-Avila. M. (2006). Fetal lead exposure at each stage of pregnancy as a
predictor of infant mental development. Environ Health Perspect 114: 1730-1735.
http://dx.doi.org/10.1289/ehp.9067.

Huang. SY: Hu. H: Sanchez. BN: Peterson. KE: Ettinger. AS: Lamadrid-Figueroa. H: Sclinaas. L: Mercado-Garcia.
A: Wright. RO: Basu. N: Cantonwine. DE: Hernandez-Avila. M: Tellez-Roio. MM. (2016). Childhood
blood lead levels and symptoms of attention deficit hyperactivity disorder (ADHD): A cross-sectional study
of Mexican children. Environ Health Perspect 124: 868-874. http://dx.doi.org/10.1289/ehp. 1510067.

Iglesias. V: Steenland. K: Maisonet. M: Pino. P. (2011). Exposure to lead from a storage site associated with
intellectual impairment in Chilean children living nearby. Int J Occup Environ Health 17: 314-321.
http://dx.doi.org/10.1179/107735211799041841.

Isliitsuka. K: Yamamoto-Hanada. K: Yang. L: Mezawa. H: Konislii. M: Saito-Abe. M: Sasaki. H: Nisliizato. M:

Sato. M: Koeda. T: Ohva. Y. (2020). Association between blood lead exposure and mental health in
pregnant women: Results from the Japan enviromnent and children's study. Neurotoxicology 79: 191-199.
http://dx.doi.Org/10.1016/i.neuro.2020.06.003.

Jamesdaniel. S: Rosati. R: Westrick. J: Ruden. DM. (2018). Chronic lead exposure induces cochlear oxidative stress
and potentiates noise-induced hearing loss. Toxicol Lett 292: 175-180.
http://dx.doi.Org/10.1016/i.toxlet.2018.05.004.

Jedrvchowski. W: Perera. F: Jankowski. J: Mrozek-Budzvn. D: Mroz. E: Flak. E: Edwards. S: Skarupa. A:

Lisowska-Miszczvk. I. (2009a). Gender specific differences in neurodevelopmental effects of prenatal
exposure to very low-lead levels: The prospective cohort study in three-year olds. Early Hum Dev 85: 503-
510. http://dx.doi.Org/10.1016/i.earlhumdev.2009.04.006.

Jedrvchowski. W: Perera. FP: Jankowski. J: Mrozek-Budzvn. D: Mroz. E: Flak. E: Edwards. S: Skarupa. A:

Lisowska-Miszczvk. I. (2009b). Very low prenatal exposure to lead and mental development of children in
infancy and early childhood: Krakow prospective cohort study. Neuroepidemiology 32: 270-278.
http://dx.doi.org/10.1159/000203Q75.

Jeong. KS: Park. H: Ha. E: Hong. YC: Ha. M: Park. H: Kim. BN: Lee. SJ: Lee. KY: Kim. .TH: Kim. Y (2015).

Evidence that cognitive deficit in children is associated not only with iron deficiency, but also with blood
lead concentration: A preliminary study. J Trace Elem Med Biol 29: 336-341.
http://dx.doi.Org/10.1016/i.itemb.2014.08.014.

Ji. JS: Elbaz. A: Weisskopf. MG. (2013). Association between blood lead and walking speed in the National Health
and Nutrition Examination Survey (NHANES 1999-2002). Environ Health Perspect 121: 711-716.
http://dx.doi.org/10.1289/ehp. 1205918.

Ji. JS: Power. MC: Sparrow. D: Spiro. A. Ill: Hu. H: Louis. ED: Weisskopf. MG. (2015). Lead exposure and tremor
among older men: The VA normative aging study. Environ Health Perspect 123: 445-450.
http://dx.doi.org/10.1289/ehp.1408535.

3-515

-------
Ji. Y: Hong. X: Wang. G: Chatteriee. N: Rilev. AW: Lee. LC: Surkan. PJ: Bartell. TR: Zuckemiaa B: Wang. X.

(2018). A prospective birth cohort study on early childhood lead levels and attention deficit hyperactivity
disorder: New insight on sex differences. JPediatr 199: 124-13l.el28.
http://dx.doi.Org/10.1016/i.ipeds.2018.03.076.

Jiang. CB: Kao. CS: Cliien. LC: Chen. YJ: Liao. KW. (2022). Associations among prenatal and postnatal arsenic,
lead, and cadmium exposures and motor development in 3-year-old children: A longitudinal birth cohort
study in Taiwan. Environ Sci Pollut Res Int 29: 43191-43200. http://dx.doi.org/10.1007/sll356-Q21-
18321-5.

Joo. H: Choi. JH: Bunn. E: Park. H: Hong. YC: Kim. Y: Ha. F.H: Kim. Y: Kim. BN: Ha. M. (2018). Gender

difference in the effects of lead exposure at different time windows on neurobehavioral development in 5-
year-old children. Sci Total Environ 615: 1086-1092. http://dx.doi.Org/10.1016/i.scitotenv.2017.10.007.

Joo. H: Lim. MH: Ha. M: Kwon. HJ: Yoo. SJ: Choi. KH: Paik. KC. (2017). Secondhand smoke exposure and low
blood lead levels in association with attention-deficit hyperactivity disorder and its symptom domain in
children: A community-based case-control study. Nicotine Tob Res 19: 94-101.
http://dx.doi.org/10.1093/ntr/ntwl52.

Kamel. F: Umbach. DM: Munsat. TL: Shefner. JM: Hu. H: Sandler. DP. (2002). Lead exposure and amyotrophic
lateral sclerosis. Epidemiology 13: 311-319.

Kamel. F: Umbach DM: Stallone. L: Richards. M: Hu. H: Sandler. DP. (2008). Association of lead exposure with
survival in amyotrophic lateral sclerosis. Environ Health Perspect 116: 943-947.
http://dx.doi.org/10.1289/ehp. 11193.

Kang. GH: Uhm. JY: Choi. YG: Kang. EK: Kim. SY: Choo. WO: Chang. SS. (2018). Enviromnental exposure of
heavy metal (lead and cadmium) and hearing loss: Data from the Korea National Health and Nutrition
Examination Survey (KNHANES 2010-2013). AnnOccup Environ Med 30: 22.
http://dx.doi.org/10T186/s40557-018-Q237-9.

Kao. CS: Wang. YL: Chuang. TW: Jiang. CB: Hsi. HC: Liao. KW: Cliiea LC. (2021). Effects of soil lead exposure
and land use characteristics on neurodevelopment among children under 3 years of age in northern Taiwan.
Environ Pollut 286: 117288. http://dx.doi.Org/10.1016/i.envpol.2021.117288.

Karri. V: Ramos. D: Martinez. JB: Odena. A: Oliveira. E: Coort. SL: Evelo. CT: Mariman. ECM: Schuhmacher. M:
Kumar. V. (2018). Differential protein expression of hippocampal cells associated with heavy metals (Pb,
As, and MeHg) neurotoxicity: Deepening into the molecular mechanism of neurodegenerative diseases. J
Proteomics 187: 106-125. http://dx.doi.Org/10.1016/i.iprot.2018.06.020.

Kawamoto. EM: Vivar. C: Camandola. S. (2012). Physiology and pathology of calcium signaling in the brain. Front
Pharmacol 3:61. http://dx.doi.org/10.3389/lphar.2012.00061.

Khalil. N: Faulkner. KA: Greenspaa SL: Caulev. JA. (2014). Associations between bone mineral density, grip
strength, and lead body burden in older men. J Am Geriatr Soc 62: 141-146.
http://dx.doi.org/10. Ill 1/jgs. 12603.

Kicinski. M: Vriiens. J: Vennier. G: Hond. ED: Schoeters. G: Nelen. V: Bruckers. L: Sioen. I: Baevens. W: Van

Larebeke. N: Viaene. MK: Nawrot. TS. (2015). Neurobehavioral function and low-level metal exposure in
adolescents. Int J Hyg Environ Health 218: 139-146. http://dx.doi.Org/10.1016/i.iiheh.2014.09.002.

Kim. JI: Kim. JW: Lee. JM: Yun. HJ: Solin. CH: Shin. MS: Kim. B: Chae. J: Roh. J: Kim. BN. (2018a). Interaction
between DRD2 and lead exposure on the cortical thickness of the frontal lobe in youth with attention-
deficit/hyperactivity disorder. Prog Neuropsychophannacol Biol Psychiatry 82: 169-176.
http://dx.doi.Org/10.1016/i.pnpbp.2017.ll.018.

Kim. KN: Kwon. HJ: Hong. YC. (2016). Low-level lead exposure and autistic behaviors in school-age children.
Neurotoxicology 53: 193-200. http://dx.doi.Org/10.1016/i.neuro.2016.02.004.

Kim. S: Arora. M: Fernandez. C: Landero. J: Caruso. J: Chen. A. (2013a). Lead, mercury, and cadmium exposure
and attention deficit hyperactivity disorder in children. Environ Res 126: 105-110.
http://dx.doi.Org/10.1016/i.envres.2013.08.008.

3-516

-------
Kim. S: Eom. S: Kim. HJ: Lee. J J: Choi. G: Choi. S: Kim. S: Kim. SY: Cho. G: Kim. YD: Suh. E: Kim. SK: Kim. S:
Kim. GH: Moon. HB: Park. J: Kim. S: Choi. K: Eun. SH. (2018b). Association between maternal exposure
to major phthalates, heavy metals, and persistent organic pollutants, and the neurodevelopmental
performances of their children at 1 to 2 years of age-CHECK cohort study. Sci Total Environ 624: 377-
384. http://dx.doi.Org/10.1016/i.scitotenv.2017.12.058.

Kim. Y: Ha. EH: Park. H: Ha. M: Kim. Y: Hong. YC: Kim. EJ: Kim. BN. (2013b). Erratum to: "Prenatal lead and
cadmium co-exposure and infant neurodevelopment at 6 months of age: The Mothers and Children's
Enviromnental Health (MOCEH) study" [NeuroToxicology 35 (2013) 15-22] [Erratum], Neurotoxicology
37: 248-249. http://dx.doi.Org/10.1016/i.neuro.2013.04.009.

Kim. Y: Ha. EH: Park. H: Ha. M: Kim. Y: Hong. YC: Kim. EJ: Kim. BN. (2013c). Prenatal lead and cadmium co-
exposure and infant neurodevelopment at 6 months of age: The Mothers and Children's Enviromnental
Health (MOCEH) study. Neurotoxicology 35: 15-22. http://dx.doi.Org/10.1016/i.neuro.2012.ll.006.

Kim. Y: Kim. BN: Hong. YC: Shin. MS: Yoo. HJ: Kim. JW: Bhang. SY: Cho. SC. (2009). Co-exposure to

enviromnental lead and manganese affects the intelligence of school-aged children. Neurotoxicology 30:
564-571. http://dx.doi.Org/10.1016/i.neuro.2009.03.012.

Kirrane. EF: Patel. MM. (2014). Identification and consideration of errors in Lanphear et al. (2005), "Low-level
enviromnental lead exposure and children's intellectual function: An international pooled analysis".
Available online at https ://www.regulations. gov/document/EPA-HO-QRD-2011 -0051 -0050 (accessed

Kordas. K: Ettinger. AS: Bellinger. DC: Schnaas. L: Tellez Roio. MM: Hernandez-Avila. M: Hu. H: Wright. RO.
(2011). A dopamine receptor (DRD2) but not dopamine transporter (DAT1) gene polymorphism is
associated with neurocognitive development of Mexican preschool children with lead exposure. J Pediatr
159: 638-643. http://dx.doi.Org/10.1016/i.ipeds.2011.03.043.

Koshv. B: Srinivasaa M: Zachariah. SM: Karthikevan. AS: Roshan. R: Bose. A: Mohan. VR: John. S: Ramanuiam.
K: Mulivil. J: Kang. G. (2020). Body iron and lead status in early childhood and its effects on development
and cognition: A longitudinal study from urban Vellore. Public Health Nutr 23: 1896-1906.
http://dx.doi.org/10.1017/S13689800190Q4622.

Krieg. EF. Jr: Butler. MA: Chang. MH: Liu. TB: Yesupriva. A: Lindegren. ML: Dowling. N. (2009). Lead and
cognitive function in ALAD genotypes in the Third National Health and Nutrition Examination Survey.
Neurotoxicol Teratol 31: 364-371. http://dx.doi.Org/10.1016/i.ntt.2009.08.003.

Krieg. EF. Jr: Butler. MA: MH. C: Liu. TB: Yesupriva. A: Dowling. N: Lindegrea ML. (2010). Lead and cognitive
function in VDR genotypes in the third National Health and Nutrition Examination Survey. Neurotoxicol
Teratol 32: 262-272. http://dx.doi.Org/10.1016/i.ntt.2009.12.004.

Kuang. W: Chen. Z: Shi. K: Sun. H: Li. H: Huang. L: Bi. J. (2020). Adverse health effects of lead exposure on
physical growth, erythrocyte parameters and school performances for school-aged children in eastern
China. Environ Int 145: 106130. http://dx.doi.Org/10.1016/i.envint.2020.106130.

Kuhlmann. AC: Mcglothan. JL: Guilarte. TR. (1997). Developmental lead exposure causes spatial learning deficits
in adult rats. Neurosci Lett 233: 101-104. http://dx.doi.org/10.1016/S0304-3940(97)00633-2.

Kumar. VL: Muralidhara. (2014). Ameliorative effects of ferulic Acid against lead acetate-induced oxidative stress,
mitochondrial dysfunctions and toxicity in prepubertal rat brain. NeurochemRes 39: 2501-2515.
http://dx.doi.org/10.1007/sllQ64-014-i451-7.

Lamoureux-Tremblav. V: Chauret. M: Muckle. G: Maheu. F: Suffren. S: Jacobson. SW: Jacobsoa JL: Avotte. P:
Lepore. F: Saint-Amour. D. (2021). Altered functional activations of prefrontal brain areas during
emotional processing of fear in Inuit adolescents exposed to enviromnental contaminants. Neurotoxicol
Teratol 85: 106973. http://dx.doi.Org/10.1016/i.ntt.2021.106973.

Lanphear. BP: Dietrich. K: Auinger. P: Cox. C. (2000). Cognitive deficits associated with blood lead concentrations
<10 microg/dL in US children and adolescents. Public Health Rep 115: 521-529.
http://dx.doi.Org/10.1093/plir/115.6.521.

3-517

-------
Lanphear. BP: Hornung. R: Khourv. J: Yolton. K: Baghurst. P: Bellinger. DC: Canfield. RL: Dietrich. KN:

Bornschein. R: Greene. T: Rothenberg. SJ: Needleman. HL: Sclinaas. L: Wasserman. G: Graziano. J:
Roberts. R. (2005). Low-level environmental lead exposure and children's intellectual function: An
international pooled analysis. Environ Health Perspect 113: 894-899. http://dx.doi.org/10.1289/ehp.7688.

Lanphear. BP: Hornung. R: Khourv. J: Yolton. K: Baghurst. P: Bellinger. DC: Canfield. RL: Dietrich. KN:

Bornschein. R: Greene. T: Rothenberg. SJ: Needleman. HL: Sclinaas. L: Wassennan. G: Graziano. J:
Roberts. R. (2019). Erratum: "Low-level enviromnental lead exposure and children's intellectual function:
An international pooled analysis" [Erratum], Environ Health Perspect 127: 099001.
http://dx.doi.org/10.1289/EHP5685.

Laughlia NK: Luck. ML: Laskv. RE. (2008). Postnatal lead effects on the development of visual spatial acuity in
rhesus monkeys (Macaca mulatta). Dev Psychobiol 50: 608-614. http://dx.doi.org/10.1002/dev.2Q315.

Laughlia NK: Luck. ML: Laskv. RE. (2009). Early lead exposure effects on an auditory threshold task in the rhesus
monkey (Macaca mulatta). Dev Psychobiol 51: 289-300. http://dx.doi.org/10.1002/dev.2Q364.

Leasure. JL: Giddabasappa. A: Chanev. S: Johnson. JE. Jr: Pothakos. K: Lau. YS: Fox. DA. (2008). Low-level
human equivalent gestational lead exposure produces sex-specific motor and coordination abnormalities
and late-onset obesity in year-old mice. Environ Health Perspect 116: 355-361.
http://dx.doi.org/10.1289/ehp. 10862.

Lee. H: Park. H: Ha. E: Hong. YC: Ha. M: Park. H: Kim. BN: Lee. SJ: Lee. KY: Kim. JH: Jeong. KS: Kim. Y
(2017). Stability of cognitive development during the first five years of life in relation to heavy metal
concentrations in umbilical cord blood: Mothers' and Children's Enviromnental Health (MOCEH) birth
cohort study. Sci Total Environ 609: 153-159. http://dx.doi.Org/10.1016/i.scitotenv.2017.07.074.

Lee. KS: Kim. KN: Ahn. YD: Choi. YJ: Clio. J: Jang. Y: Lim. YH: Kim. Jl: Shin. CH: Lee. YA: Kim. BN: Hong.
YC. (2021). Prenatal and postnatal exposures to four metals mixture and IQ in 6-year-old children: A
prospective cohort study in South Korea. Environ Int 157: 106798.
http://dx.doi.Org/10.1016/i.envint.2021.106798.

Lenroot. RK: Giedd. JN. (2006). Brain development in children and adolescents: Insights from anatomical magnetic
resonance imaging [Review]. Neurosci Biobehav Rev 30: 718-729.
http://dx.doi.Org/10.1016/i.neubiorev.2006.06.001.

Leviton. A: Bellinger. D: Allred. EN: Rabinowitz. M: Needleman. H: Schoenbaum. S. (1993). Pre- and postnatal
low-level lead exposure and children's dysfunction in school. Environ Res 60: 30-43.
http://dx.doi.org/10.1006/enrs. 1993.1003.

Li. HW: Deng. JG: Du. ZC: Yan. MS: Long. ZX: Pham Tlii. PT: Yang. KD. (2013). Protective effects of mangiferin
in subclironic developmental lead-exposed rats. Biol Trace Elem Res 152: 233-242.
http://dx.doi.org/10.1007/sl2011-013-961Q-2.

Li. N: Li. X: Li. L: Zhang. P: Qiao. M: Zhao. O: Song. L: Yu. Z. (2016a). Original research: The expression of

MMP2 and MMP9 in the hippocampus and cerebral cortex of newborn mice under maternal lead exposure.
Exp Biol Med 241: 1811-1818. http://dx.doi.org/10.1177/1535370216647808.

Li. N: Qiao. M: Zhao. O: Zhang. P: Song. L: Li. L: Cui. C. (2016b). Effects of maternal lead exposure onRGMa and
RGMb expression in the hippocampus and cerebral cortex of mouse pups. Brain Res Bull 127: 38-46.
http://dx.doi.Org/10.1016/i.braimesbull.2016.08.010.

Li. N: Yang. G: Wang. Y: Qiao. M: Zhang. P: Shao. J: Yang. G. (2016c). Decreased IDE and IGF2 expression but
increased A(U0 in the cerebral cortex of mouse pups by early life lead exposure. Brain Res Bull 121: 84-
90. http://dx.doi.Org/10.1016/i.braimesbull.2016.01.004.

Li. N: Yu. ZL: Wang. L: Zheng. YT: Jia. JX: Wang. O: Zhu. MJ: Liu. XL: Xia. X: Li. WJ. (2010). Increased tau

phosphorylation and beta amyloid in the liipocampus of mouse pups by early life lead exposure. Acta Biol
Hung 61:" 123-134. http://dx.doi.Org/10.1556/ABiol.61.2010.2.l.

Li. S: Xu. J: Liu. Z: Yan. CH. (2017). The non-linear association between low-level lead exposure and maternal
stress among pregnant women. Neurotoxicology 59: 191-196.
http://dx.doi.Org/10.1016/i.neuro.2016.07.005.

3-518

-------
Lilienthal. H: Lenaerts. C: Winneke. G: Hennekes. R. (1988). Alteration of the visual evoked potential and the
electroretinogram in lead-treated monkeys. Neurotoxicol Teratol 10: 417-422.
http://dx.doi.org/10.1016/0892-0362(88)90002-5.

Lilienthal. H: Winneke. G. (1996). Lead effects on the brain stem auditory evoked potential in monkeys during and
after the treatment phase. Neurotoxicol Teratol 18: 17-32. http://dx.doi.org/10.1016/0892-0362(95)02010-
1.

Lin. CC: Chen. YC: Su. FC: Lin. CM: Liao. HF: Hwang. YH: Hsieh. WS: Jeng. SF: Su. YN: Chen. PC. (2013). In
utero exposure to enviromnental lead and manganese and neurodevelopment at 2 years of age. Environ Res
123: 52-57. http://dx.doi.Org/10.1016/i.envres.2013.03.003.

Lin. D: Lutter. R: Rulun. C. (2016). Cognitive Performance and Labor Market Outcomes. National Bureau of
Economic Research, http://dx.doi.org/10.3386/w22470.

Lin. Y: Huang. L: Xu. J: Specht. AJ: Yan. C: Geng. H: Shen. X: Nie. LH: Hu. H. (2019). Blood lead, bone lead and
child attention-deficit-hyperactivity-disorder-like behavior. Sci Total Environ 659: 161-167.
http://dx.doi.Org/10.1016/i.scitotenv.2018.12.219.

Listos. J: Baranowska-Bosiacka. I: Talarek. S: Listos. P: Orzelska. J: Fidecka. S: Gutowska. I: Kolasa. A: Rvbicka.
M: Chlubek. D. (2013). The effect of perinatal lead exposure on dopamine receptor D2 expression in
morphine dependent rats. Toxicology 310: 73-83. http://dx.doi.Org/10.1016/i.tox.2013.05.007.

Liu. C: Huang. L: Huang. S: Wei. L: Cao. D: Zan. G: Tan Y: Wang. S: Yang. M: Tian L: Tang. W: He. C: Shen. C:
Luo. B: Zhu. M: Liang. T: Pang. B: Li. M: Mo. Z: Yang. X. (2022a). Association of both prenatal and early
childhood multiple metals exposure with neurodevelopment in infant: A prospective cohort study. Environ
Res 205: 112450. http://dx.doi.Org/10.1016/i.envres.2021.112450.

Liu. C: Yang. W: Ma. J: Yang. H: Feng. Z: Sun. J: Cheng. C: Jiang. H. (2018a). Dihydromyricetin inhibits lead-
induced cognitive impairments and inflammation by the adenosine 5'-monophosphate-activated protein
kinase pathway in mice. J Agric Food Chem 66: 7975-7982. http://dx.doi.org/10.1021/acs.iafc.8b02433.

Liu. CM: Tian. ZK: Zhang. YJ: Ming. OL: Ma. JO: Ji. LP. (2020). Effects of Gastrodin against Lead-Induced Brain
Injury in Mice Associated with the Wnt/Nrf2 Pathway. Nutrients 12: 1-13.
http ://dx. doi.org/10.3 3 90/nu12061805.

Liu. F: Xue. Z: Li. N: Huang. H: Ying. Y: Li. J: Wang. L: Li. W. (2014a). Effects of lead exposure on the expression
of amyloid (3 and phosphorylated tau proteins in the C57BL/6 mouse hippocampus at different life stages. J
Trace ElemMed Biol 28: 227-232. http://dx.doi.Org/10.1016/i.itemb.2014.01.002.

Liu. J: Li. L: Wang. Y: Yan. C: Liu. X. (2013a). Impact of low blood lead concentrations on IQ and school

performance in Chinese children. PLoS ONE 8: e65230. http://dx.doi.org/10.1371/iournal.pone.0065230.

Liu. J: Liu. X: Wang. W: Mccaulev. L: Pinto-Martin. J: Wang. Y: Li. L: Yan. C: Rogan. WJ. (2014b). Blood lead
concentrations and children's behavioral and emotional problems: A cohort study. JAMA Pediatr 168: 737-
745. http://dx.doi.org/10.1001/iamapediatrics.2014.332.

Liu. J: Portnov. J: Raine. A: Gladieux. M: McGarrv. P: Chen A. (2022b). Blood lead levels mediate the relationship
between social adversity and child externalizing behavior. Environ Res 204: 112396.
http://dx.doi.Org/10.1016/i.envres.2021.112396.

Liu. JA: Chen. YJ: Gao. DG: Jing. J: Hu. OS. (2014c). Prenatal and postnatal lead exposure and cognitive

development of infants followed over the first three years of life: A prospective birth study in the Pearl
RiverDelta region, China. Neurotoxicology 44: 326-334. http://dx.doi.Org/10.1016/i.neuro.2014.07.001.

Liu. JA: Gao. DJ: Chen. YM: Jing. J: Hu. OS: Chen. YJ. (2014d). Lead exposure at each stage of pregnancy and
neurobehavioral development of neonates. Neurotoxicology 44: 1-7.
http://dx.doi.Org/10.1016/i.neuro.2014.03.003.

Liu. MC: Liu. XO: Wang. W: Shen. XF: Che. HL: Guo. YY: Zhao. MG: Chen JY: Luo. WJ. (2012). Involvement
of microglia activation in the lead induced long-term potentiation impairment. PLoS ONE 7: e43924.
http://dx.doi.org/10.1371/iournal.pone.0043924.

3-519

-------
Liu. MC: Xu. Y: Chen. YM: Li. J: Zhao. F: Zheng. G: Jing. JF: Ke. T: Chen. JY: Luo. WJ. (2013b). The effect of
sodium selenite on lead induced cognitive dysfunction. Neurotoxicology 36: 82-88.
http://dx.doi.Org/10.1016/i.neuro.2013.03.008.

Liu. R: Bai. L: Liu. M: Wang. R: Wu. Y: Li. O: Ba. Y: Zhang. H: Zhou. G: Yu. F: Huang. H. (2022c). Combined
exposure of lead and high-fat diet enhanced cognitive decline via interacting with CREB-BDNF signaling
in male rats. Environ Pollut 304: 119200. http://dx.doi.org/10.1016/i.envpol.2022.119200.

Liu. SH: Bobb. JF: Claus Henn. B: Gennings. C: Schnaas. L: Tellez-Roio. M: Bellinger. D: Arora. M: Wright. RO:
Coull. BA. (2018b). Bayesian varying coefficient kernel machine regression to assess neurodevelopmental
trajectories associated with exposure to complex mixtures. Stat Med 37: 4680-4694.
http://dx.doi.org/10.1002/sim.7947.

Liu. W: Huo. X: Liu. D: Zeng. X: Zhang. Y: Xu. X. (2014e). S 100(3 in heavy metal-related child attention-deficit
hyperactivity disorder in an informal e-waste recycling area. Neurotoxicology 45: 185-191.
http://dx.doi.Org/10.1016/i.neuro.2014.10.013.

Liu. X: Dietrich. KN: Radcliffe. J: Ragan. NB: Rlioads. GG: Rogan. WJ. (2002). Do children with falling blood lead
levels have improved cognition? Pediatrics 110: 787-791. http://dx.doi.Org/10.1542/peds.110.4.787.

Liu. X: Wei. F: Cheng. Y: Zhang. Y: Jia. G: Zhou. J: Zhu. M: Shan. Y: Sun. X: Yu. L: Merzenich. MM: Lurie. DI:
Zheng. O: Zhou. X. (2019). Auditory training reverses lead (Pb)-toxicity-induced changes in sound-
azimuth selectivity of cortical neurons. Cereb Cortex 29: 3294-3304.
http://dx.doi.org/10.1093/cercor/bhYl99.

Liu. X: Zheng. G: Wu. Y: Shen. X: Jing. J: Yu. T: Song. H: Chen. J: Luo. W. (2013c). Lead exposure results in
hearing loss and disruption of the cochlear blood-labyrinth barrier and the protective role of iron
supplement. Neurotoxicology 39: 173-181. http://dx.doi.Org/10.1016/i.neuro.2013.10.002.

Liu. XO: Huang. R: Zhou. X: Cai. TJ: Chen. JJ: Shi. XW: Deng. HB: Luo. WJ. (2017). Presence of nano-sized

chitosan-layered silicate composites protects against toxicity induced by lead ions. Carbohydr Polym 158:
1-10. http://dx.doi.Org/10.1016/i.carbpol.2016.ll.084.

Liu. Y: Huo. X: Xu. L: Wei. X: Wu. W: Wu. X: Xu. X. (2018c). Hearing loss in children with e-waste lead and
cadmium exposure. Sci Total Environ 624: 621-627. http://dx.doi.Org/10.1016/i.scitotenv.2017.12.091.

Liu. YF: Chen. O: Wei. XP: Chea L: Zhang. X: Chen. K: Chen. J: Li. TY. (2015). Relationship between perinatal
antioxidant vitamin and heavy metal levels and the growth and cognitive development of children at 5
years of age. Asia Pac J Clin Nutr 24: 650-658. http://dx.doi.Org/10.6133/apicn.2015.24.4.25.

Long. X: Wu. H: Zhou. Y: Wan. Y: Kan. X: Gong. J: Zhao. X. (2022). Preventive effect of Limosilactobacillus
fennentum SCHY34 on lead acetate-induced neurological damage in SDrats. Front Nutr 9: 852012.
http://dx.doi.org/10.3389/fnut.2022.852012.

Lu. X: Jin. C: Yang. J: Liu. O: Wu. S: Li. D: Guan. Y: Cai. Y. (2013). Prenatal and lactational lead exposure

enhanced oxidative stress and altered apoptosis status in offspring rats' hippocampus. Biol Trace Elem Res
151: 75-84. http://dx.doi.org/10.1007/sl2011-Q12-9531-5.

Lucchini. RG: Zoni. S: Guazzetti. S: Bontempi. E: Micheletti. S: Broberg. K: Parrinello. G: Smith. DR. (2012).

Inverse association of intellectual function with very low blood lead but not with manganese exposure in
Italian adolescents. Environ Res 118: 65-71. http://dx.doi.Org/10.1016/i.envres.2012.08.003.

Luo. M: Xu. Y: Cai. R: Tang. Y: Ge. MM: Liu. ZH: Xu. L: Hu. F: Ruan. DY: Wang. HL. (2014). Epigenetic histone
modification regulates developmental lead exposure induced hyperactivity in rats. Toxicol Lett 225: 78-85.
http://dx.doi.Org/10.1016/i.toxlet.2013.ll.025.

Lvnam. PR: Charnigo. R: Moffitt. TE: Raine. A: Loeber. R: Stouthamer-Loeber. M. (2009). The stability of
psychopathy across adolescence. Dev Psychopathol 21: 1133-1153.
http://dx.doi.org/10.1017/S0954579409990Q83.

Lynch. G: Kessler. M: Arai. A: Larson. J. (1990). The nature and causes of hippocampal long-term potentiation
[Review], Prog Brain Res 83: 233-250. http://dx.doi.org/10.1016/s0079-6123(08)61253-4.

3-520

-------
Ma. J: Yan. L: Guo. T: Yang. S: Guo. C: Liu. Y: Xie. O: Wang. J. (2019). Association of typical toxic heavy metals
with schizophrenia. Int J Environ Res Public Health 16: 4200. http://dx.doi.org/10.3390/iierphl6214200.

Macedoni-Luksic. M: Gosar. D: Biorklund. G: Orazem. J: Kodric. J: Lesnik-Musek. P: Zupancic. M: France-Stiglic.
A: Sesek-Briski. A: Neubauer. D: Osredkar. J. (2015). Levels of metals in the blood and specific
porphyrins in the urine in children with Autism Spectrum Disorders. Biol Trace Elem Res 163: 2-10.
http://dx.doi.org/10.1007/sl2011-014-Q121-6.

Magzamen. S: Amato. MS: Imm. P: Havlena. JA: Coons. MJ: Anderson. HA: Kanarek. MS: Moore. CF. (2015).

Quantile regression in enviromnental health: Early life lead exposure and end-of-grade exams. Environ Res
137: 108-119. http://dx.doi.org/10.1016/i.envres.2014.12.004.

Maiti. AK: Saha. NC: More. SS: Panigrahi. AK: Paul. G. (2017). Neuroprotective efficacy of mitochondrial
antioxidant MitoQ in suppressing peroxynitrite-mediated mitochondrial dysfunction inflicted by lead
toxicity in the rat brain. Neurotox Res 31: 358-372. http://dx.doi.org/10.1007/sl2640-Q16-9692-7.

Mani. MS: Joslii. MB: Shettv. RR: Dsouza. VL: Swatlii. M: Kabekkodu. SP: Dsouza. HS. (2020). Lead exposure
induces metabolic reprogramming in rat models. Toxicol Lett 335: 11-27.
http://dx.doi.Org/10.1016/i.toxlet.2020.09.010.

Mansouri. MT: Naghizadeh. B: Lopez-Larrubia. P: Cauli. O. (2012). Gender-dependent behavioural impairment and
brain metabolites in young adult rats after short term exposure to lead acetate. Toxicol Lett 210: 15-23.
http ://dx. doi.org/10.1016/i .toxlet. 2012.01.012.

Mansouri. MT: Naghizadeh. B: Lopez-Larrubia. P: Cauli. O. (2013). Behavioral deficits induced by lead exposure
are accompanied by serotonergic and cholinergic alterations in the prefrontal cortex. Neurochem Int 62:
232-239. http://dx.doi.Org/10.1016/i.neuint.2012.12.009.

Marcus. DK: Fulton. JJ: Clarke. EJ. (2010). Lead and conduct problems: A meta-analysis. J Clin Child Adolesc
Psychol 39: 234-241. http://dx.doi.org/10.1080/153744110Q3591455.

Marks. DF. (2010). IQ variations across time, race, and nationality: An artifact of differences in literacy skills.
Psychol Rep 106: 643-664. http://dx.doi.Org/10.2466/pr0.106.3.643-664.

Marques. RC: Bernardi. JV: Dorea. JG: de Fatima Ramos Moreira. M: Malm. O. (2014). Perinatal multiple exposure
to neurotoxic (lead, methylmercury, ethylmercury, and aluminum) substances and neurodevelopment at six
and 24 months of age. Environ Pollut 187: 130-135. http://dx.doi.Org/10.1016/i.envpol.2014.01.004.

Martia KV: Sucharew. H: Dietrich. KN: Parsons. PJ: Palmer. CD: Wright. R: Amarasiriwardena. C: Smith. PR:
Havnes. EN. (2021). Co-exposure to manganese and lead and pediatric neurocognition in East Liverpool,
Ohio. Environ Res 202: 111644. http://dx.doi.Org/10.1016/i.envres.2021.111644.

Maughan. B: Stafford. M: Shah. I: Kuli. D. (2014). Adolescent conduct problems and premature mortality: follow-
up to age 65 years in a national birth cohort. Psychol Med 44: 1077-1086.
http://dx.doi.Org/10.1017/S0033291713001402".

Mazumdar. M: Bellinger. DC: Gregas. M: Abanilla. K: Bacic. J: Needleman. HL. (2011). Low-level enviromnental
lead exposure in childhood and adult intellectual function: A follow-up study. Environ Health 10: 24.
http://dx.doi.org/10.1186/1476-069X-10-24.

McCall. RB: Hogartv. PS: Hurlburt. N. (1972). Transitions in infant sensorimotor development and the prediction of
childhood IQ [Review]. Am Psychol 27: 728-748. http://dx.doi.org/10.1037/h0033148.

Mccollister. KE: French. MT: Fang. H. (2010). The cost of crime to society: new crime-specific estimates for policy
and program evaluation. Drug Alcohol Depend 108: 98-109.
http://dx.doi.Org/10.1016/i.drugalcdep.2009.12.002.

McFarlane. AC: Searle. AK: VanHooff. M: Baghurst. PA: Sawyer. MG: Galletlv. C: Sim. MR: Clark. LS. (2013).
Prospective associations between childhood low-level lead exposure and adult mental health problems: the
Port Pirie cohort study. Neurotoxicology 39: 11-17. http://dx.doi.Org/10.1016/i.neuro.2013.08.003.

Mcgivern. RF: Sokol. RZ: Bennan. NG. (1991). Prenatal lead exposure in the rat during the third week of gestation:
Long-term behavioral, physiological and anatomical effects associated with reproduction. Toxicol Appl
Pharmacol 110: 206-215."http://dx.doi.org/10.1016/S0041-008X(05)80003-l.

3-521

-------
Mcmichael. AJ: Baghurst. PA: Vimpani. GV: Robertson. EF: Wigg. NR: Tong. SL. (1992). Sociodemographic

factors modifying the effect of environmental lead on neuropsychological development in early childhood.
Neurotoxicol Teratol 14: 321-327. http://dx.doi.org/10.1016/0892-0362(92)90038-0.

Menezes-Filho. JA: Carvalho. CF: Rodrigues. JLG: Arauio. CFS: Dos Santos. NR: Lima. CS: Bandeira. MJ:

Marques. BLS: Anios. ALS: Bah. HAF: Abreu. N: Pliilibert. A: Mergler. D. (2018). Enviromnental co-
exposure to lead and manganese and intellectual deficit in school-aged children. Int J Environ Res Public
Health 15: 2418. http://dx.doi. org/10.3390/ii erphl 5112418.

Meng. H: Wang. L: He. JH: Wang. ZF. (2016). The Protective Effect of Gangliosides on Lead (Pb)-Induced
Neurotoxicity Is Mediated by Autophagic Pathways. Int J Environ Res Public Health 13: 365.
http://dx.doi.org/10.3390/iierphl3040365.

Meramat. A: Raiab. NF: Shahar. S: Sharif. RA. (2017). DNA damage, copper and lead associates with cognitive

function among older adults. J Nutr Health Aging 21: 539-545. http://dx.doi.org/10.1007/sl2603-016-Q759-
1.

Merced-Nieves. FM: Chelonis. J: Pantic. I: Schnass. L: Tellez-Roio. MM: Braun. JM: Paule. MG: Wright. RJ:

Wright. RO: Curtia P. (2022). Sexually dimorphic associations between prenatal blood lead exposure and
performance on a behavioral testing battery in children. Neurotoxicol Teratol 90: 107075.
http://dx.doi.Org/10.1016/i.ntt.2022.107075.

Min. JY: Min. KB: Cho. SI: Kim. R: Sakong. J: Paek. D. (2007). Neurobehavioral function in children with low
blood lead concentrations. Neurotoxicology 28: 421-425. http://dx.doi.Org/10.1016/i.neuro.2006.03.007.

Min. KB: Lee. KJ: Park. JB: Mia JY. (2012). Lead and cadmium levels and balance and vestibular dysfunction

among adult participants in the National Health and Nutrition Examination Survey (NHANES) 1999-2004.
Environ Health Perspect 120: 413-417. http://dx.doi.org/10.1289/ehp. 1103643.

Min. MO: Singer. LT: Kirchner. HL: Minnes. S: Short. E: Hussain. Z: Nelsoa S. (2009). Cognitive development
and low-level lead exposure in poly-drug exposed children. Neurotoxicol Teratol 31: 225-231.
http://dx.doi.Org/10.1016/i.ntt.2009.03.002.

Miranda. ML: Kim. D: Galeano. M: Paul. CJ: Hull. AP: Morgan. SP. (2007). The relationship between early

childhood blood lead levels and performance on end-of-grade tests. Environ Health Perspect 115: 1242-
1247. http://dx.doi.org/10.1289/ehp.9994.

Miranda. ML: Kim. D: Reiter. J: Galeano. MAO: Maxson. P. (2009). Enviromnental contributors to the achievement
gap. Neurotoxicology 30: 1019-1024. http://dx.doi.Org/10.1016/i.neuro.2009.07.012.

Mohammad. IK: Mahdi. AA: Raviraia. A: Naimul. I: Iqbal. A: Thuppil. V. (2008). Oxidative stress in painters

exposed to low lead levels. Arh Hig Rada Toksikol 59: 161-169. http://dx.doi.org/10.2478/100Q4-1254-59-
2008-1883.

Molendiik. ML: de Kloet. ER. (2015). Immobility in the forced swim test is adaptive and does not reflect

depression. Psychoneuroendocrinology 62: 389-391. http://dx.doi.Org/10.1016/i.psvneuen.2015.08.028.

Molina. RM: Phattanarudee. S: Kim. J: Thompson. K: Wessling-Resnick. M: Maher. TJ: Brain. JD. (2011).

Ingestion of Mn and Pb by rats during and after pregnancy alters iron metabolism and behavior in
offspring. Neurotoxicology 32: 413-422. http://dx.doi.Org/10.1016/i.neuro.2011.03.010.

Monday. HR: Younts. TJ: Castillo. PE. (2018). Long-term plasticity of neurotransmitter release: Emerging

mechanisms and contributions to brain function and disease [Review]. Annu Rev Neurosci 41: 299-322.
http://dx.doi.org/10.1146/annurev-neuro-080317-062155.

Moore. CF: Gaiewski. LL: Laughlin. NK: Luck. ML: Larson. JA: Schneider. ML. (2008). Developmental lead

exposure induces tactile defensiveness in rhesus monkeys (Macaca mulatta). Environ Health Perspect 116:
1322-1326. http://dx.doi.org/10.1289/ehp.11203.

Moreira. EG: Vassilieff. I: Vassilieff. VS. (2001). Developmental lead exposure: Behavioral alterations in the short
and long term. Neurotoxicol Teratol 23: 489-495. http://dx.doi.org/10.1016/S0892-0362(01)00159-3.

3-522

-------
Munoz. C: Garbe. K: Lilicnlhal. H: Winneke. G. (1989). Neuronal depletion of the amygdala resembles the learning
deficits induced by low level lead exposure in rats. Neurotoxicol Teratol 11: 257-264.
http://dx.doi.org/10.1016/0892-0362(89)90068-8.

Mufioz. MP: Rubilar. P: Valdes. M: Mufioz-Ouezada. MT: Gomez. A: Saavedra. M: Iglesias. V. (2020). Attention
deficit hyperactivity disorder and its association with heavy metals in children from northern Chile. Int J
Hyg Environ Health 226: 113483. http://dx.doi.Org/10.1016/i.iiheh.2020.113483.

Murphy. PL: Patel. M: Kirrane. E: Vinikoor-Imler. L. (2013). Comments on: Chari, R.; Burke, T.A.; White, R.H.;
Fox, M.A. Integrating susceptibility into enviromnental policy: An analysis of the National Ambient Air
Quality Standard for lead. Int. J. Environ. Res. Public Health 2012, 9, 1077-1096 [Comment]. Int J Environ
Res Public Health 10: 712-716. http://dx.doi.org/10.3390/iierphlQ020712.

Naicker. N: Richter. L: Mathee. A: Becker. P: Norris. SA. (2012). Enviromnental lead exposure and socio-

behavioural adjustment in the early teens: The birth to twenty cohort. Sci Total Environ 414: 120-125.
http://dx.doi.Org/10.1016/i.scitotenv.2011.ll.013.

Nam. S: Seo. J: Nalun. SS: Chang. BJ. (2019a). Effects of ascorbic acid on osteopontin expression and axonal
myelination in the developing cerebellum of lead-exposed rat pups. Int J Environ Res Public Health 16:
983. http://dx.doi.org/10.3390/iierphl6060983.

Nam. SM: Aim. SC: Go. TH: Seo. JS: Nalun. SS: Chang. BJ: Lee. JH. (2018a). Ascorbic acid ameliorates

gestational lead exposure-induced developmental alteration in GAD67 and c-kit expression in the rat
cerebellar cortex. Biol Trace Elem Res 182: 278-286. http://dx.doi.org/10.1007/sl2011-Q17-1086-z.

Nam. SM: Chang. BJ: Kim. JH: Nalun. SS: Lee. JH. (2018b). Ascorbic acid ameliorates lead-induced apoptosis in
the cerebellar cortex of developing rats. Brain Res 1686: 10-18.
http://dx.doi.Org/10.1016/i.braimes.2018.02.014.

Nam. SM: Cho. IS: Seo. JS: Go. TH: Kim. JH: Nalun. SS: Chang. BJ: Lee. JH. (2019b). Ascorbic acid attenuates
lead-induced alterations in the synapses in the developing rat cerebellum. Biol Trace Elem Res 187: 142-
150. http://dx.doi.org/10.1007/si2011-Q18-1354-6.

Nam. SM: Choi. SH: Cho. HJ: Seo. JS: Choi. M: Nalun. SS: Chang. BJ: Nah. SY. (2020). Ginseng gintonin

attenuates lead-induced rat cerebellar impairments during gestation and lactation. Biomolecules 10: 385.
http://dx.doi.org/10.3390/biomlQ030385.

Nan. A: Zhou. X: Chen. L: Liu. M: Zhang. N: Zhang. L: Luo. Y: Liu. Z: Dai. L: Jiang. Y. (2016). A transcribed
ultraconserved noncoding RNA, Uc. 173, is a key molecule for the inhibition of lead-induced neuronal
apoptosis. Onct 7: 112-124. http://dx.doi.org/10.18632/oncotarget.6590.

Needleman. HL: Gatsonis. CA. (1990). Low-level lead exposure and the IQ of children: A meta-analysis of modern
studies. JAMA 263: 673-678. http://dx.doi.org/10 1001 /iama. 1990.03440050067035.

Needleman. HL: Gunnoe. C: Leviton. A: Reed. R: Peresie. H: Maher. C: Barrett. P. (1979). Deficits in psychologic
and classroom performance of children with elevated dentine lead levels. N Engl J Med 300: 689-695.
http://dx.doi.org/10.1056/NEJM1979032930Q1301.

Needleman. HL: Schell. A: Bellinger. D: Leviton. A: Allred. EN. (1990). The long-term effects of exposure to low
doses of lead in childhood. An 11-year follow-up report. N Engl J Med 322: 83-88.
http://dx.doi.org/10.1056/NEJM1990011132202Q3.

Neelima. A: Raianna. A: Bhanuprakash. RG: Chettv. CS: Suresh. C. (2017). Deleterious effects of combination of
lead and (3-amyloid peptides in inducing apoptosis and altering cell cycle in human neuroblastoma cells.
Interdiscip Toxicol 10: 93-98. http://dx.doi.org/10.1515/intox-2017-0015.

Nelson. MM: Espy. KA. (2009). Low-level lead exposure and contingency-based responding in preschoolers: An
exploratory study. Dev Neuropsychol 34: 494-506. http://dx.doi.org/10.1080/875656409Q2964565.

Neugebauer. J: Wittsiepe. J: Kasper-Sonnenberg. M: Schoneck. N: Scholmerich. A: Wilhelm. M. (2015). The

influence of low level pre- and perinatal exposure to PCDD/Fs, PCBs, and lead on attention performance
and attention-related behavior among German school-aged children: Results from the Duisburg Birth
Cohort Study. Int J Hyg Environ Health 218: 153-162. http://dx.doi.Org/10.1016/i.iiheh.2014.09.005.

3-523

-------
Neuwirth. LS: Emenike. BU: Barrera. ED: Hameed. N: Rubi. S: Dacius. TF: Skeen. JC: Bonitto. JR: Khairi. E:

Igbal. A: Ahmed. I; Jose. TJ: Lynch. K: Khan. M: Alvira. AL: Mathew. N: Kaur. S: Masood. S: Tranguilee.
B: Tliiruverkadu. V. (2019a). Assessing the anxiolytic properties of taurine-derived compounds in rats
following developmental lead exposure: A neurodevelopmental and behavioral pharmacological pilot
study. In J Hu; F Piao; SW Schaffer; A Elldrissi; JY Wu (Eds.), Taurine 11 (pp. 801-819). Singapore,
Singapore: Springer, http://dx.doi.org/10.1007/978-981-13-8023-5 69.

Neuwirth. LS: Kim. Y: Barrerra. ED: Jo. C: CMsphonte, JM: Hameed. N: Rubi. S: Dacius. TF. Jr: Skeen. JC:
Bonitto. JR: Khairi. E: Iqbal. A: Ahmed. I: Masood. S: Tranguilee. B: Tliiruverkadu. V. (2019b). Early
neurodevelopmental exposure to low lead levels induces fronto-executive dysfunctions that are recovered
by taurine co-treatment in the rat attention set-shift test: Implications for taurine as a
psychophannacotherapy against neurotoxicants. In J Hu; F Piao; SW Schaffer; A El Idrissi; JY Wu (Eds.),
Taurine 11 (pp. 821-846). Singapore, Singapore: Springer. http://dx.doi.org/10.1007/978-981-13-8Q23-
5 70.

Neuwirth. LS: Masood. S: Anderson. DW: Schneider. JS. (2019c). The attention set-shifting test is sensitive for
revealing sex-based impairments in executive functions following developmental lead exposure in rats.
Behav Brain Res 366: 126-134. http://dx.doi.Org/10.1016/i.bbr.2019.03.022.

Nguyen. HP: Oh. H: Hoang. NHM: Jo. WH: Kim. MS. (2022). Enviromnental science and pollution research role of
heavy metal concentrations and vitamin intake from food in depression: A national cross-sectional study
(2009-2017). Environ Sci Pollut Res Int 29: 4574-4586. http://dx.doi.org/10.1007/sll356-021-15986-w.

Nicolescu. R: Petcu. C: Cordeanu. A: Fabritius. K: Schlumpf. M: Krebs. R: Kramer. U: Winneke. G. (2010).

Enviromnental exposure to lead, but not other neurotoxic metals, relates to core elements of ADHD in
Romanian children: Performance and questionnaire data. Environ Res 110: 476-483.
http://dx.doi.Org/10.1016/i.envres.2010.04.002.

Nigg. J: Knottnerus. G: Martel. M: Nikolas. M: Cavanagh. K: Kannaus. W: Rapplev. M. (2008). Low blood lead
levels associated with clinically diagnosed attention-deficit/hyperactivity disorder and mediated by weak
cognitive control. Biol Psychiatry 63: 325-331. http://dx.doi.Org/10.1016/i.biopsvch.2007.07.013.

Nigg. JT: Elmore. AL: Nataraian. N: Friderici. KH: Nikolas. MA. (2016). Variation in an iron metabolism gene

moderates the association between blood lead levels and attention-deficit/hyperactivity disorder in children.
Psychol Sci 27: 257-269. http://dx.doi.org/10.1177/0956797615618365.

Nigg. JT: Nikolas. M: Knottnerus. GM: Cavanagh. K: Friderici. K. (2010). Confirmation and extension of

association of blood lead with attention-deficit/hyperactivity disorder (ADHD) and ADHD symptom
domains at population-typical exposure levels. J Child Psychol Psychiatry 51: 58-65.
http://dx.doi.Org/10.llll/i.1469-7610.2009.02135.x.

Niu. RY: Sun. ZL: Cheng. ZT: Li. ZG: Wang. JD. (2009). Decreased learning ability and low hippocampus
glutamate in offspring rats exposed to fluoride and lead. Environ Toxicol Pharmacol 28: 254-258.
http://dx.doi.Org/10.1016/i.etap.2009.04.012.

Nkomo. P: Mathee. A: Naicker. N: Galpin. J: Richter. LM: Norris. SA. (2017). The association between elevated
blood lead levels and violent behavior during late adolescence: The South African Birth to Twenty Plus
cohort. Environ Int 109: 136-145. http://dx.doi.Org/10.1016/i.envint.2017.09.004.

Nkomo. P: Naicker. N: Mathee. A: Galpin. J: Richter. LM: Norris. SA. (2018). The association between

enviromnental lead exposure with aggressive behavior, and dimensionality of direct and indirect aggression
during mid-adolescence: Birth to Twenty Plus cohort. Sci Total Environ 612: 472-479.
http://dx.doi.Org/10.1016/i.scitotenv.2017.08.138.

Nordberg. M: Winblad. B: Fratiglioni. L: Basua H. (2000). Lead concentrations in elderly urban people related to
blood pressure and mental performance: Results from a population-based study. Am J Ind Med 38: 290-
294. http://dx.doi.org/10. 1002/1097-0274(200009)38:3<290::AID-AJIM7>3.6.CO:2-T.

Nozadi. SS: Li. L: Luo. L: Mackenzie. D: Erdei. E: Du. R: Roman. CW: Hoover. J: O'Donald. E: Burnette. C:

Lewis. J. (2021). Prenatal metal exposures and infants' developmental outcomes in a Navajo population. Int
J Environ Res Public Health 19: 425. http://dx.doi.org/10.3390/iierphl901Q425.

3-524

-------
Nvanza. EC: Bernier. FP: Martin. JW: Manvama. M: Hatfield. J: Dewev. D. (2021). Effects of prenatal exposure
and co-exposure to metallic or metalloid elements on early infant neurodevelopmental outcomes in areas
with small-scale gold mining activities in Northern Tanzania. Environ Int 149: 106104.
http://dx.doi.Org/10.1016/i.envint.2020.106104.

Okesola. MA: Aiibove. BO: Ovinlove. BE: Oio. OA. (2019). Neuromodulatory effects of ethyl acetate fraction of
Zingiber officinale Roscoe extract in rats with lead-induced oxidative stress. J Integr Med 17: 125-131.
http://dx.doi.Org/10.1016/i.ioim.2019.01.002.

Qpler. MGA: Buka. SL: Groeger. J: McKeague. I: Wei. C: Factor-Litvak. P: Bresnahan. M: Graziano. J: Goldstein.
JM: Seidman. LJ: Brown. AS: Susser. ES. (2008). Prenatal exposure to lead, 5-aminolevulinic acid, and
schizophrenia: Further evidence. Environ Health Perspect 116: 1586-1590.
http://dx.doi.org/10.1289/ehp. 10464.

Qppenheimer. AY: Bellinger. DC: Coull. BA: Weisskopf. MG: Korrick. SA. (2022). Prenatal exposure to chemical
mixtures and working memory among adolescents. Environ Res 205: 112436.
http://dx.doi.Org/10.1016/i.envres.2021.112436.

Ouvang. L: Zhang. W: Du. G: Liu. H: Xie. J: Gu. J: Zhang. S: Zhou. F: Shao. L: Feng. C: Fan. G. (2019). Lead
exposure-induced cognitive impairment through RyR-modulating intracellular calcium signaling in aged
rats. Toxicology 419: 55-64. http://dx.doi.Org/10.1016/i.tox.2019.03.005.

Paier. KA. (1998). What happens to "bad" girls? A review of the adult outcomes of antisocial adolescent girls
[Review]. Am J Psychiatry 155: 862-870. http://dx.doi.org/10.1176/aip. 155.7.862.

Palaniappan. K: Roy. A: Balakrishnan. K: Gopalakrishnan. L: Mukheriee. B: Hu. H: Bellinger. DC. (2011). Lead
exposure and visual-motor abilities in children from Chennai, India. Neurotoxicology 32: 465-470.
http://dx.doi.Org/10.1016/i.neuro.2011.03.011.

Paraiuli. RP: Fuiiwara. T: Umezaki. M: Watanabe. C. (2013). Association of cord blood levels of lead, arsenic, and
zinc with neurodevelopmental indicators in newborns: A birth cohort study in Chitwan Valley, Nepal.
Environ Res 121: 45-51. http://dx.doi.Org/10.1016/i.envres.2012.10.010.

Paraiuli. RP: Fuiiwara. T: Umezaki. M: Watanabe. C. (2015a). Home enviromnent and cord blood levels of lead,
arsenic, and zinc on neurodevelopment of 24 months children living in Chitwan Valley, Nepal. J Trace
Elem Med Biol 29: 315-320. http://dx.doi.Org/10.1016/i.itemb.2014.08.006.

Paraiuli. RP: Umezaki. M: Fuiiwara. T: Watanabe. C. (2015b). Association of cord blood levels of lead, arsenic, and
zinc and home enviromnent with children neurodevelopment at 36 months living in Chitwan Valley, Nepal.
PLoS ONE 10: e0120992. http://dx.doi.org/10.1371/iournal.pone.0120992.

Park. JH: Sep. JH: Hong. YS: Kim. YM: Kang. JW: Yoo. JH: Chueh. HW: Lee. JH: Kwak. M.T: Kim. J: Woo. HP:
Kim. DW: Bang. YR: Choe. BM. (2016). Blood lead concentrations and attention deficit hyperactivity
disorder in Korean children: A hospital-based case control study. BMC Pediatr 16: 156.
http://dx.doi.org/10.1186/sl2887-016-0696-5.

Park. SK: Elmarsafawv. S: Mukheriee. B: Spiro. A. Ill: Vokonas. PS: Nie. H: Weisskopf. MG: Schwartz. J: Hu. H.

(2010). Cumulative lead exposure and age-related hearing loss: The VA Normative Aging Study. Hear Res
269: 48-55. http://dx.doi.Org/10.1016/i.heares.2010.07.004.

Paul. KC: Horvath. S: Del Rosario. I: Bronstein. JM: Ritz. B. (2021). DNA methylation biomarker for cumulative
lead exposure is associated with Parkinson's disease. Clinical Epigenetics 13: 59.
http://dx.doi.org/10.1186/sl3148-021-01051-3.

Paulsen. AJ: Schubert. CR: Johnson. LJ: Chen. Y: Dalton. PS: Klein. BEK: Kleia R: Pinto. A: Cruickshanks. KJ.
(2018). Association of cadmium and lead exposure with the incidence of contrast sensitivity impairment
among middle-aged adults. JAMA Ophthalmol 136: 1342-1350.
http://dx.doi.org/10.1001/iamaophthalmol.2018.3931.

Pawlas. N: Broberg. K: Olewinska. E: Kozlowska. A: SkerfVing. S: Pawlas. K. (2015). Genetic modification of

ALAP and VPR on lead-induced impairment of hearing in children. Environ Toxicol Pharmacol 39: 1091-
1098. http://dx.doi.Org/10.1016/i.etap.2015.03.008.

3-525

-------
Pedroso. TF: Oliveira. CS: Fonseca. MM: Oliveira. VA: Pereira. ME. (2017). Effects of zinc and n-acetylcysteine in
damage caused by lead exposure in young rats. Biol Trace Elem Res 180: 275-284.
http://dx.doi.org/i0.1007/sl2011-017-10Q9-z.

Peters. JL: Kubzanskv. LP: Ikeda. A: Spiro. A. Ill: Wright. RO: Weisskopf. MG: Kim. D: Sparrow. D: Nie. LH:
Hu. H: Schwartz. J. (2011). Childhood and adult socioeconomic position, cumulative lead levels, and
pessimism in later life: The VA Normative Aging Study. Am J Epidemiol 174: 1345-1353.
http://dx.doi.org/10.1093/aie/kwr269.

Peters. S: Broberg. K: Gallo. V: Levi. M: Kippler. M: Vineis. P: Veldink. J: van Den Berg. L: Middleton. L: Travis.
RC: Bergmann. MM: Palli. D: Grioni. S: Tumino. R: Elbaz. A: Vlaar. T: Mancini. F: Kuhn. T: Katzke. V:
Agudo. A: Gofii. F: Gomez. JH: Rodriguez-Barranco. M; Merino, S; Barricarte, A; Trichopoulou, A; Jenab,
M; Weiderpass, E; Vermeulen, R. (2020). Blood Metal Levels and Amyotrophic Lateral Sclerosis Risk: A
Prospective Cohort. Ann Neurol 89: 125-133. http://dx.doi.org/10.1002/ana.25932.

Phvu. MP: Tangpong. J. (2013). Protective effect of Thunbergia laurifolia (Linn.) on lead induced

acetylcholinesterase dysfunction and cognitive impairment in mice. BioMed Res Int 2013: 186098.
http://dx.doi.org/10.1155/2013/186098.

Pilsner. JR: Hu. H: Wright. RO: Kordas. K: Ettinger. AS: Sanchez. BN: Cantonwine. D: Lazarus. AL: Cantoral. A:
Schnaas. L: Tellez-Roio. MM: Hernandez-Avila. M. (2010). Maternal MTHFR genotype and haplotype
predict deficits in early cognitive development in a lead-exposed birth cohort in Mexico City. Am J Clin
Nutr 92: 226-234. http://dx.doi.org/10.3945/aicn.2009.28839.

Pocock. SJ: Smith. M: Baghurst. P. (1994). Enviromnental lead and children's intelligence: A systematic review of
the epidemiological evidence [Review]. Br Med J 309: 1189-1197.
http://dx.doi.org/10.1136/bmi.309.6963.1189.

Power. MC: Korrick. S: Tchetgen Tchetgen. EJ: Nie. LH: Grodstein. F: Hu. H: Weuve. J: Schwartz. J: Weisskopf.

MG. (2014). Lead exposure and rate of change in cognitive function in older women. Environ Res 129: 69-
75. http://dx.doi.Org/10.1016/i.envres.2013.12.010.

Przvbvla. J: Houseman. EA: Smit. E: Kile. ML. (2017). A path analysis of multiple neurotoxic chemicals and
cognitive functioning in older US adults (NHANES 1999-2002). Environ Health 16: 19.
http://dx.doi.org/10.1186/sl2940-017-0227-3.

Oin. YY: Jian. B: Wu. C: Jiang. CZ: Kang. Y: Zhou. JX: Yang. F: Liang. Y. (2018). A comparison of blood metal
levels in autism spectrum disorder and unaffected children in Shenzhen of China and factors involved in
bioaccumulation of metals. Environ Sci Pollut Res Int 25: 17950-17956. http://dx.doi.org/10.1007/sl 1356-
018-1957-7.

Rabinowitz. MB: Wang. JD: Soong. WT. (1992). Children's classroom behavior and lead in Taiwan. Bull Environ
Contain Toxicol 48: 282-288. http://dx.doi.org/10.1007/BF0Q194385.

Rahbar. MH: Ibrahim. SH: Azam. SI: Hessabi. M: Karim. F: Kim. S: Zhang. J: Ali. NG: Loveland. KA. (2021).

Concentrations of lead, mercury, arsenic, cadmium, manganese, and aluminum in the blood of Pakistani
children with and without autism spectrum disorder and their associated factors. Int J Environ Res Public
Health 18: 8625. http://dx.doi.org/10.3390/iierphl8168625.

Rahbar. MH: Samms-Vaughan. M: Dickerson. AS: Loveland. KA: Ardiomand-Hessabi. M: Bressler. J:

Shakespeare-Pellington S: Grove. ML: Pearson. DA: Boerwinkle. E. (2015). Blood lead concentrations in
Jamaican children with and without Autism Spectrum Disorder. Int J Environ Res Public Health 12: 83-
105. http://dx.doi.org/10.3390/iierphl20100Q83.

Rahman. A: Al-Awadi. AA: Khan. KM. (2018). Lead affects vitamin D metabolism in rats. Nutrients 10: 264.
http://dx.doi.org/10.3390/nulQ030264.

Ralunan. A: Khan. K: Al-Khaledi. G: Khan I: Attur. S. (2012a). Early postnatal lead exposure induces tau
phosphorylation in the brain of young rats. Acta Biol Hung 63: 411 -425.
http://dx.doi.Org/10.1556/ABioi.63.2012.4.l.

3-526

-------
Rahman. A: Khan. KM: Al-Khaledi. G: Khan. I; Al-Shemarv. T. (2012b). Over activation of hippocampal

serine/threonine protein phosphatases PP1 and PP2A is involved in lead-induced deficits in learning and
memory in young rats. Neurotoxicology 33: 370-383. http://dx.doi.Org/10.1016/i.neuro.2012.02.014.

Raian. P: Kelsev. KT: Schwartz. JD: Bellinger. DC: Weuve. J: Sparrow. D: Spiro. A. Ill: Smith. TJ: Nie. H: Hu. H:
Wright. RO. (2007). Lead burden and psychiatric symptoms and the modifying influence of the delta-
aminolevulinic acid dehydratase (ALAD) polymorphism: The VA Normative Aging Study. Am J
Epidemiol 166: 1400-1408. http://dx.doi.org/10.1093/aie/kwm22Q.

Raian. P: Kelsev. KT: Schwartz. JD: Bellinger. DC: Weuve. J: Spiro. A. Ill: Sparrow. D: Smith. TJ: Nie. H:

Weisskopf. MG: Hu. H: Wright. RO. (2008). Interaction of the delta-aminolevulinic acid dehydratase
polymorphism and lead burden on cognitive function: The VA Normative Aging Study. J Occup Environ
Med 50: 1053-1061. http://dx.doi.org/10.1097/JQM.0b013e3181792463.

Rao Barkur. R: Bairv. LK. (2016). Comparison of the developmental milestones and preweaning neurobehavioral
parameters in rat pups exposed to lead (Pb) during gestation, lactation and pregestation period. Drug Chem
Toxicol 39: 248-255. http://dx.doi.org/10.3109/01480545.2Q15.1082136.

Rasnick. E: Ryan. PH: Bailer. AJ: Fisher. T: Parsons. PJ: Yolton. K: Newman. NC: Lanphear. BP: Brokamp. C.

(2021). Identifying sensitive windows of airborne lead exposure associated with behavioral outcomes at
age 12. Enviromnental Epidemiology 5: el44. http://dx.doi.org/10.1097/EE9.000000000000Q144.

Rawat. PS: Singh. S: Mahdi. AA: Mehrotra. S. (2022). Enviromnental lead exposure and its correlation with
intelligence quotient level in children. J Trace Elem Med Biol 72: 126981.
http://dx.doi.Org/10.1016/i.itemb.2022.126981.

Rea. P. (2015). Overview of the nervous system. In P Rea (Ed.), Essential clinically applied anatomy of the
peripheral nervous system in the limbs (pp. 1-40). London, United Kingdom: Academic Press.
http://dx.doi.org/10.i016/B978-0-12-803062-2.00Q01-2.

Reuben. A: Caspi. A: Belskv. DW: Broadbent. J: Harrington. H: Sugden. K: Houts. RM: Ramrakha. S: Poulton. R:
Moffitt. TE. (2017). Association of childhood blood lead levels with cognitive function and socioeconomic
status at age 38 years and with IQ change and socioeconomic mobility between childhood and adulthood.
JAMA 317: 1244-1251. http://dx.doi.org/10 1001/iama.2017.1712.

Reuben. A: Elliott. ML: Abraham. WC: Broadbent. J: Houts. RM: Ireland. D: Knodt. AR: Poulton. R: Ramrakha. S:
Hariri. AR: Caspi. A: Moffitt. TE. (2020). Association of childhood lead exposure with MRI measurements
of structural brain integrity in midlife. JAMA 324: 1970-1979. http://dx.doi.org/10.1001/iama.202Q.19998.

Reuben. A: Schaefer. JD: Moffitt. TE: Broadbent. J: Harrington. H: Houts. RM: Ramrakha. S: Poulton. R: Caspi. A.
(2019). Association of childhood lead exposure with adult personality traits and lifelong mental health.
JAMA Psychiatry 76: 418-425. http://dx.doi.org/10.1001/iamapsvchiatrv.2018.4192.

Reves. JW. (2015). Lead exposure and behavior: Effects on antisocial and risky behavior among children and
adolescents. Economic Inquiry 53: 1580-1605. http://dx.doi.org/10. Ill 1/ecin. 12202.

Reynolds. CR: Kamphaus. RW. (2015). BASC-3: Behavior assessment system for children (3rd ed.). Bloomington,
MN: NCS Pearson.

Rice. DC. (1990). Lead-induced behavioral impairment on a spatial discrimination reversal task in monkeys exposed
during different periods of development. Toxicol Appl Pharmacol 106: 327-333.
http://dx.doi.org/10.1016/0041-008X(90)9025 l-O.

Rice. DC. (1992). Lead exposure during different developmental periods produces different effects on FI
performance in monkeys tested as juveniles and adults. Neurotoxicology 13: 757-770.

Rice. DC. (1997). Effects of lifetime lead exposure in monkeys on detection of pure tones. Fundam Appl Toxicol
36: 112-118. http://dx.doi.org/10.1006/faat. 1996.2268.

Rice. DC. (1998). Effects of lifetime lead exposure on spatial and temporal visual function in monkeys.
Neurotoxicology 19: 893-902.

3-527

-------
Rice. DC: Gilbert. SG. (1990a). Lack of sensitive period for lead-induced behavioral impairment on a spatial
delayed alternation task in monkeys. Toxicol Appl Pharmacol 103: 364-373.
http://dx.doi.org/10.1016/0041-008X(90)90236-N.

Rice. DC: Gilbert. SG. (1990b). Sensitive periods for lead-induced behavioral impairment (nonspatial discrimination
reversal) in monkeys. Toxicol Appl Pharmacol 102: 101-109. http://dx.doi.org/10.1016/0Q41-
008X(90)90087-B.

Rice. DC: Karpinski. KF. (1988). Lifetime low-level lead exposure produces deficits in delayed alternation in adult
monkeys. Neurotoxicol Teratol 10: 207-214. http://dx.doi.org/10.1016/0892-0362(88)90019-0.

Ris. MP: Dietrich. KN: Succop. PA: Berger. OG: Bornschein. RL. (2004). Early exposure to lead and
neuropsychological outcome in adolescence. J Int Neuropsychol Soc 10: 261-270.
http://dx.doi.org/10.1017/S1355617704102154.

Rivenbark. JG: Odgers. CL: Caspi. A: Harrington. H: Hogan. S: Houts. RM: Poulton. R: Moffitt. TE. (2018). The
high societal costs of childhood conduct problems: evidence from administrative records up to age 38 in a
longitudinal birth cohort. J Child Psychol Psychiatry 59: 703-710. http://dx.doi.org/10. Ill 1/icpp. 12850.

Roda. SM: Greenland. RD: Bornschein. RL: Hammond. PB. (1988). Anodic stripping voltammetry procedure
modified for improved accuracy of blood lead analysis. Clin Chem 34: 563-567.
http://dx.doi.Org/10.1093/clinchem/34.3.563.

Rodrigues. ALS: Rocha. JBT: Mello. CF: Souza. DO. (1996). Effect of perinatal lead exposure on rat behaviour in
open-field and two-way avoidance tasks. Basic Clin Pharmacol Toxicol 79: 150-156.
http://dx.doi.Org/10.llil/i.1600-0773.1996.tb00259.x.

Rodrigues. EG: Bellinger. DC: Valeri. L: Hasan. MPS. I: Ouamruzzaman. O: Golam. M: Kile. ML: Christian! DC:
Wright. RO: Mazumdar. M. (2016). Neurodevelopmental outcomes among 2- to 3-year-old children in
Bangladesh with elevated blood lead and exposure to arsenic and manganese in drinking water. Environ
Health 15: 44. http://dx.doi.org/10.1186/sl2940-016-0127-Y.

Rodrigues. JLG: Arauio. CFS: Dos Santos. NR: Bandeira. MJ: Anios. ALS: Carvalho. CF: Lima. CS: Abreu. JNS:
Mergler. D: Menezes-Filho. JA. (2018). Airborne manganese exposure and neurobehavior in school-aged
children living near a ferro-manganese alloy plant. Environ Res 167: 66-77.
http://dx.doi.Org/10.1016/i.envres.2018.07.007.

Rokoff. LB: Shoaff. JR: Coull. BA: Enlow. MB: Bellinger. DC: Korrick. SA. (2022). Prenatal exposure to a mixture
of organochlorines and metals and internalizing symptoms in childhood and adolescence. Environ Res 208:
112701. http://dx.doi.Org/10.1016/i.envres.2022.fl2701.

Roonev. JPK: Woods. NF: Martin. MP: Woods. JS. (2018). Genetic polymorphisms of GRIN2A and GRIN2B
modify the neurobehavioral effects of low-level lead exposure in children. Environ Res 165: 1-10.
http://dx.doi.Org/10.1016/i.envres.2018.04.001.

Rossi-George. A: Virgolini. MB: Weston. D: Thiruchelvam. M: Corv-Slechta. DA. (2011). Interactions of lifetime
lead exposure and stress: Behavioral, neurochemical and HPA axis effects. Neurotoxicology 32: 83-99.
http://dx.doi.Org/10.1016/i.neuro.2010.09.004.

Rothenberg. SJ: Schnaas. L: Salgado-Valladares. M: Casanueva. E: Geller. AM: Hudnell. HK: Fox. DA. (2002).
Increased ERG a- and b-wave amplitudes in 7- to 10-year-old children resulting from prenatal lead
exposure. Invest Ophthalmol Vis Sci 43: 2036-2044.

Rov. A: Hu. H: Bellinger. DC: Mukheriee. B: Modali. R: Nasaruddin. K: Schwartz. J: Wright. RO: Ettinger. AS:
Palaniapan. K: Balakrishnan. K. (2011). Hemoglobin, lead exposure, and intelligence quotient: Effect
modification by the DRD2 Taq IA polymorphism. Environ Health Perspect 119: 144-149.
http://dx.doi.org/10.1289/ehp.0901878.

Ruebner. RL: Hooper. SR: Parrish. C: Furth. SL: Fadrowski. JJ. (2019). Enviromnental lead exposure is associated
with neurocognitive dysfunction in children with chronic kidney disease. Pediatr Nephrol 34: 2371-2379.
http://dx.doi.org/10.l607/s00467-019-043Q6-7.

3-528

-------
Ruff. HA: Biiur. PE: Markowitz. M: Ma. YC: Rosen. JF. (1993). Declining blood lead levels and cognitive changes
in moderately lead-poisoned children. JAMA 269: 1641-1646.
http://dx.doi.org/] 0.1001 /iama. 1993.03500130055032.

Rygiel. CA: Dolinov. DC: Bakulski. KM: Aung. MT: Perng. W: Jones. TR: Solano-Gonzalez. M: Hu. H: Tellez-
Roio. MM: Sclinaas. L: Marcela. E: Peterson. KE: Goodrich. JM. (2021). DNA methylation at birth
potentially mediates the association between prenatal lead (Pb) exposure and infant neurodevelopmental
outcomes. Environ Epigenet 7: dvab005. http://dx.doi.org/10.1093/eep/dvab005.

Sadeghi. A: Khordad. E: Ebrahimi. V: Raoofi. A: Alipour. F: Ebraliimzadeh-Bideskan. A. (2021). Neuroprotective
effects of vitamin C and garlic on glycoconjugates changes of cerebellar cortex in lead-exposed rat
offspring. J Chem Neuroanat 114: 101948. http://dx.doi.Org/10.1016/i.ichemneu.2021.101948.

Saleh. HA: Abd El-Aziz. GS: Mustafa. HN: El-Fark. M: Mai. A: Aburas. M: Deifalla. AH. (2019). Thymoquinone
ameliorates oxidative damage and histopathological changes of developing brain neurotoxicity. J
Histotechnol 42: 116-127. http://dx.doi.org/10.1080/01478885.2Q19.1619654.

Saleh. HA: Abdel El-Aziz. GS: Mustafa. HN: Saleh AHA: Mai. AO: Deifalla. AHS: Aburas. M. (2018). Protective
effect of garlic extract against maternal and foetal cerebellar damage induced by lead administration during
pregnancy in rats. Folia Morphol (Warsz) 77: 1-15. http://dx.doi.org/10.5603/FM.a2017.0Q63.

Salkever. PS. (1995). Updated estimates of earnings benefits from reduced exposure of children to enviromnental
lead. Environ Res 70: 1-6. http://dx.doi.org/10.1006/enrs. 1995.1038.

Sanchez-Martin. FJ: Lindauist. DM: Landero-Figueroa. J: Zhang. X: Chen. J: Cecil. KM: Medvedovic. M: Puga. A.
(2015). Sex- and tissue-specific methylome changes in brains of mice perinatally exposed to lead.
Neurotoxicology 46: 92-100. http://dx.doi.Org/10.1016/i.neuro.2014.12.004.

Sanchez. BN: Hu. H: Litman. HJ: Tellez-Roio. MM. (2011). Statistical methods to study timing of vulnerability
with sparsely sampled data on enviromnental toxicants. Environ Health Perspect 119: 409-415.
http://dx.doi.org/10.1289/ehp.1002453.

Sandhir. R: Gill. KD. (1995). Effect of lead on lipid peroxidation in liver of rats. Biol Trace Elem Res 48: 91-97.
http://dx.doi.org/10.1007/BF02789Q81.

Sandin. S: Hultman. CM: Kolevzon. A: Gross. R: MacCabe. JH: Reichenberg. A. (2012). Advancing maternal age is
associated with increasing risk for autism: A review and meta-analysis. J Am Acad Child Adolesc
Psychiatry 51: 477-486. http://dx.doi.Org/10.1016/i.iaac.2012.02.018.

Sasaki. N: Carpenter. DO. (2022). Associations between metal exposures and cognitive function in American older
adults. Int J Environ Res Public Health 19: 2327. http://dx.doi.org/10.3390/iierphl9042327.

Saxena. R: Gamble. M: Wassennan. GA: Liu. X: Parvez. F: Navas-Acien. A: Islam. T: Factor-Litvak. P: Uddin.

MN: Kioumourtzoglou. MA: Gibson. EA: Shahriar. H: Slavkovich. V: Ilievski. V: Loiacono. N: Balac. O:
Graziano. JH. (2022). Mixed metals exposure and cognitive function in Bangladeshi adolescents.
Ecotoxicol Environ Saf 232: 113229. http://dx.doi.Org/10.1016/i.ecoenv.2022.113229.

Schell. LM: Denliam. M: Stark. AD: Gomez. M: Ravenscroft. J: Parsons. PJ: Avdennir. A: Samelson. R. (2003).

Maternal blood lead concentration, diet during pregnancy, and anthropometry predict neonatal blood lead
in a socioeconomically disadvantaged population. Environ Health Perspect 111: 195-200.
http://dx.doi.org/10.1289/ehp.5592.

Sclinaas. L: Rothenberg. SJ: Flores. MF: Martinez. S: Hernandez. C: Osorio. E: Velasco. SR: Perroni. E. (2006).
Reduced intellectual development in children with prenatal lead exposure. Environ Health Perspect 114:
791-797. http://dx.doi.org/10.1289/ehp.8552.

Schneider. JS: Kidd. SK: Anderson. DW. (2013). Influence of developmental lead exposure on expression of DNA
methyltransferases and methyl cytosine-binding proteins in hippocampus. Toxicol Lett 217: 75-81.
http://dx.doi.Org/10.1016/i.toxlet.2012.12.004.

Schneider. JS: Mettil. W: Anderson. DW. (2012). Differential effect of postnatal lead exposure on gene expression
in the hippocampus and frontal cortex. J Mol Neurosci 47: 76-88. http://dx.doi.org/10.1007/sl2031-011-
9686-0.

3-529

-------
Schwartz. J. (1994a). Low-level lead exposure and children's IQ: A meta-analysis and search for a threshold.

Environ Res 65: 42-55. http://dx.doi.org/10.1006/enrs. 1994.1020.

Schwartz. J. (1994b). Societal benefits of reducing lead exposure. Environmental Progress 66: 105-124.
http://dx.doi.org/10.1006/enrs. 1994.1048.

Schwartz. J: Otto. D. (1987). Blood lead, hearing thresholds, and neurobehavioral development in children and
youth. Arch Environ Health 42: 153-160. http://dx.doi.org/10.1080/00Q39896.1987.9935814.

Schwartz. J: Otto. D. (1991). Lead and minor hearing impairment. Arch Environ Health 46: 300-305.
http://dx.doi.org/10.1080/00Q39896.1991.9934391.

Sepeliri. H: Ganii. F. (2016). The protective role of ascorbic acid on hippocampal CA1 pyramidal neurons in a rat
model of maternal lead exposure. J Chem Neuroanat 74: 5-10.
http://dx.doi.Org/10.1016/i.ichemneu.2016.01.005.

Shadbegian. R: Guignet. D: Klemick. H: Bui. L. (2019). Early childhood lead exposure and the persistence of
educational consequences into adolescence. Environ Res 178: 108643.
http://dx.doi.Org/10.1016/i.envres.2019.108643.

Shargorodskv. J: Curhan. SG: Henderson. E: Eavev. R: Curhan GC. (2011). Heavy metals exposure and hearing
loss in US adolescents. Arch Otolaryngol Head Neck Surg 137: 1183-1189.
http://dx.doi.org/10.1001/archoto.2011.2Q2.

Shekhawat. PS: Janu. VC: Singh. P: Shanna. P: Singh. K. (2021). Association of newborn blood lead concentration
with neurodevelopment outcome in early infancy. J Trace Elem Med Biol 68: 126853.
http://dx.doi.Org/10.1016/i.itemb.2021.126853.

Sliiue. I. (2013). Urinary enviromnental chemical concentrations and vitamin D are associated with vision, hearing,
and balance disorders in the elderly. Environ Int 53: 41-46. http://dx.doi.Org/10.1016/i.envint.2012.12.006.

Shuttleworth-Edwards. AB. (2016). Generally representative is representative of none: commentary on the pitfalls of
IQ test standardization in multicultural settings. Clin Neuropsychol 30: 975-998.
http://dx.doi.org/10.1080/13854046.2016.12Q4011.

Shvachiv. L: Geraldes. V: Amaro-Leal. A: Rocha. I. (2018). Intermittent low-level lead exposure provokes anxiety,
hypertension, autonomic dysfunction and neuroinflammation. Neurotoxicology 69: 307-319.
http://dx.doi.Org/10.1016/i.neuro.2018.08.001.

Shvachiv. L; Geraldes. V: Amaro-Leal. A; Rocha. I. (2020). Persistent effects on cardiorespiratory and nervous

systems induced by long-term lead exposure: Results from a longitudinal study. Neurotox Res 37: 857-870.
http://dx.doi.org/l6.1007/sl2640-020-0Q162-8.

Silbergeld. EK. (1991). Lead in bone: Implications for toxicology during pregnancy and lactation [Review]. Environ
Health Perspect 91: 63-70. http://dx.doi.org/10.2307/3430984.

Silva. PA: Hughes. P: Williams. S: Faed. JM. (1988). Blood lead, intelligence, reading attaimnent, and behaviour in
eleven year old children in Dunedin, New Zealand. J Child Psychol Psychiatry 29: 43-52.
http://dx.doi.org/10.1111/i. 1469-7610.1988.tb00687.x.

Silver. MK: Li. XO: Liu. YH: Li. M: Mai. XO: Kaciroti. N: Kilenv. P: Tardif. T: Meeker. JD: Lozoff. B. (2016).
Low-level prenatal lead exposure and infant sensory function. Environ Health 15: 65.
http://dx.doi.org/10.1186/sl2940-016-0148-6.

Singh. PK: Nath. R: Ahmad. MK: Rawat. A: Babu. S: Dixit. RK. (2016). Attenuation of lead neurotoxicity by
supplementation of polyunsaturated fatty acid in Wistar rats. Nutr Neurosci 19: 396-405.
http://dx.doi.org/10.1179/1476830515Y.0000000Q28.

Singh. PK: Singh. MK: Yadav. RS: Dixit. RK: Mehrotra. A: Nath. R. (2017). Attenuation of lead-induced
neurotoxicity by omega-3 fatty acid in rats. Ann Neurosci 24: 221-232.
http://dx.doi.org/10.1159/0004818Q8.

3-530

-------
Singh. PK: Singh. MK; Yadav. RS: Nath. R: Melirotra. A: Rawat. A: Dixit. RK. (2019). Omega-3 fatty acid
attenuates oxidative stress in cerebral cortex, cerebellum, and hippocampus tissue and improves
neurobehavioral activity in chronic lead-induced neurotoxicity. NutrNeurosci 22: 83-97.
http://dx.doi.org/10.1080/1028415X.2Q17.1354542.

Sioen. I: Den Hond. E: Nelen. V: Van De Mieroop. E: Croes. K: Van Larebeke. N: Nawrot. TS: Schoeters. G.
(2013). Prenatal exposure to enviromnental contaminants and behavioural problems at age 7-8 years.
Environ Int 59: 225-231. http://dx.doi.Org/10.1016/i.envint.2013.06.014.

Skalnv. AV: Simashkova. NV: Klvushnik. TP: Grabeklis. AR: Biorklund. G: Skalnava. MG: Nikonorov. AA:
Tinkov. AA. (2017). Hair toxic and essential trace elements in children with autism spectrum disorder.
Metab Brain Dis 32: 195-202. http://dx.doi.org/10.1007/sll011-Q16-9899-6.

Skerfving. S: Lofmark. L: Lundh. T: Mikoczv. Z: Stromberg. U. (2015). Late effects of low blood lead

concentrations in children on school performance and cognitive functions. Neurotoxicology 49: 114-120.
http://dx.doi.Org/10.1016/i.neuro.2015.05.009.

Skogheim. TS: Wevde. KVF: Engel. SM: Aase. H: Suren. P: 0ie. MG: Biele. G: Reichborn-Kiennerud. T:

Caspersen. IH: Hornig. M: Haug. LS: Villanger. GD. (2021). Metal and essential element concentrations
during pregnancy and associations with autism spectrum disorder and attention-deficit/hyperactivity
disorder in children. Environ Int 152: 106468. http://dx.doi.Org/10.1016/i.envint.2021.106468.

Sobin. C: Flores-Montova. MG: Gutierrez. M: Parisi. N: Schaub. T. (2015). 5-Aminolevulinic acid dehydratase

single nucleotide polymorphism 2 (ALAD2) and peptide transporter 2*2 haplotype (hPEPT2*2) differently
influence neurobehavior in low-level lead exposed children. Neurotoxicol Teratol 47: 137-145.
http://dx.doi.Org/10.1016/i.ntt.2014.12.001.

Sobin. C: Montova. MGF: Parisi. N: Schaub. T: Cervantes. M: Anniios. RX. (2013). Microglial disruption in young
mice with early chronic lead exposure. Toxicol Lett 220: 44-52.
http://dx.doi.Org/10.1016/i.toxlet.2013.04.003.

Sobolewski. M: Abston. K: Conrad. K: Marvin. E: Harvey. K: Susiario. M: Corv-Slechta. DA. (2020). Lineage- and
sex-dependent behavioral and biochemical transgenerational consequences of developmental exposure to
lead, prenatal stress, and combined lead and prenatal stress in mice. Environ Health Perspect 128: 27001.
http://dx.doi.org/10.1289/EHP4977.

Sobolewski. M: Vanna. G: Adams. B: Anderson. DW: Schneider. JS: Corv-Slechta. DA. (2018). Developmental
lead exposure and prenatal stress result in sex-specific reprograming of adult stress physiology and
epigenetic profiles in brain. Toxicol Sci 163: 478-489. http://dx.doi.org/10.1093/toxsci/kfV046.

Soderstrom. H: Siodin. AK: Carlstedt. A: Forsman. A. (2004). Adult psychopathic personality with childhood-onset
hyperactivity and conduct disorder: A central problem constellation in forensic psychiatry. Psychiatry Res
121:271-280. http://dx.doi.org/10.1016/S0165-1781(03)00270-1.

Soetrisno. FN: Delgado-Saborit. JM. (2020). Chronic exposure to heavy metals from informal e-waste recycling
plants and children's attention, executive function and academic performance. Sci Total Environ 717:
137099. http://dx.doi.Org/10.1016/i.scitotenv.2020.137099.

Song. H: Zheng. G: Shen. XF: Liu. XO: Luo. WJ: Chen. JY. (2014). Reduction of brain barrier tight junctional
proteins by lead exposure: Role of activation of nonreceptor tyrosine kinase Src via chaperon GRP78.
Toxicol Sci 138: 393-402. http://dx.doi.org/10.1093/toxsci/kfu007.

Souza-Talarico. JN: Marcourakis. T: Barbosa. F. Jr: Barros. SBM: Rivelli. DP: Pompeia. S: Caramelli. P:

Plusauellec. P: Lupien. SJ: Catucci. RF: Alves. AR: Suchecki. D. (2017). Association between heavy metal
exposure and poor working memory and possible mediation effect of antioxidant defenses during aging. Sci
Total Environ 575: 750-757. http://dx.doi.Org/10.1016/i.scitotenv.2016.09.121.

Souza Lisboa. SFD: Goncalves. G: Komatsu. F: Oueiroz. CAS: Almeida. AA: Moreira. EG. (2005). Developmental
lead exposure induces depressive-like behavior in female rats. Drug Chem Toxicol 28: 67-77.
http://dx.doi.org/10.1081/DCT-39696.

3-531

-------
Stangle. DE: Smith. PR: Beaudin. SA: Strawderman. MS: Levitskv. DA: Strupp. BJ. (2007). Succimer chelation
improves learning, attention, and arousal regulation in lead-exposed rats but produces lasting cognitive
impairment in the absence of lead exposure. Environ Health Perspect 115: 201-209.
http://dx.doi.org/10.1289/ehp.9263.

Stansfield. KH: Ruby, KN: Spares, BP: McGlothan. JL: Liu. X: Guilarte. TR. (2015). Early-life lead exposure

recapitulates the selective loss of parvalbumin-positive GABAergic interneurons and subcortical dopamine
system hyperactivity present in schizophrenia. Transl Psychiatry 5: e522.
http://dx.doi.org/10.1038/tp.2014.147.

Stavner. L: Steenland. K: Posemeci. M: Hertz-Picciotto. I. (2003). Attenuation of exposure-response curves in
occupational cohort studies at high exposure levels. Scand J Work Environ Health 29: 317-324.
http://dx.doi.org/10.5271/siweh.737.

Sterling. G: O'Neill. K: McCaffertv. M: O'Neill. J. (1982). Effect of chronic lead ingestion by rats on glucose
metabolism and acetylcholine synthesis in cerebral cortex slices. J Neurochem 39: 592-596.
http://dx.doi.Org/10.illl/i.147i-4159.1982.tb03989.x.

Stewart. PW: Blaine. C: Cohen. M: Burright. RG: Ponovick. PJ. (1996). Acute and longer term effects of meso-2,3
dimercaptosuccinic acid (PMSA) on the behavior of lead-exposed and control mice. Physiol Behav 59:
849-855. http://dx.doi.org/10.1016/0031-9384(95)02185-X.

Stiles. KM: Bellinger. PC. (1993). Neuropsychological correlates of low-level lead exposure in school-age children:
A prospective study. Neurotoxicol Teratol 15: 27-35. http://dx.doi.org/10.1016/0892-0362(93)90042-M.

Su. P: Zhang. J: Wang. S: Aschner. M: Cao. Z: Zhao. F: Wang. P: Chen. J: Luo. W. (2016). Genistein alleviates
lead-induced neurotoxicity in vitro and in vivo: Involvement of multiple signaling pathways.
Neurotoxicology 53: 153-164. http://dx.doi.Org/10.1016/i.neuro.2015.12.019.

Sumner. SA: Mercy. JA: Pahlberg. LL: Hillis. SP: Klevens. J: Hourv. P. (2015). Violence in the United States:
Status, Challenges, and Opportunities. JAMA 314: 478-488. http://dx.doi.org/10.1001/iama.2015.8371.

Sun. H: Chen. W: Wang. P: Jin. Y: Chen. X: Xu. Y: Huang. L. (2015). Inverse association between intelligence
quotient and urinary retinol binding protein in Chinese school-age children with low blood lead levels:
Results from a cross-sectional investigation. Chemosphere 128: 155-160.
http://dx.doi.Org/10.1016/i.chemosphere.2015.01.036.

Sun. L: Zhou. XL: Yi. HP: Jiang. SJ: Yuan. H. (2014). Lead-induced morphological changes and amyloid precursor
protein accumulation in adult rat hippocampus. Biotech Histochem 89: 513-517.
http://dx.doi.org/10.3109/10520295.2014.9Q4926.

Surkan. PJ: Schnaas. L: Wright. RJ: Tellez-Roio. MM: Lamadrid-Figueroa. H: Hu. H: Hernandez-Avila. M:

Bellinger. PC: Schwartz. J: Perroni. E: Wright. RO. (2008). Maternal self-esteem, exposure to lead, and
child neurodevelopment. Neurotoxicology 29: 278-285. http://dx.doi.Org/10.1016/i.neuro.2007.ll.006.

Surkan. PJ: Zhang. A: Trachtenberg. F: Paniel. PB: Mckinlav. S: Bellinger. PC. (2007). Neuropsychological
function in children with blood lead levels <10 (ig/dL. Neurotoxicology 28: 1170-1177.
http://dx.doi.Org/10.1016/i.neuro.2007.07.007.

Suszkiw. J: Toth. G: Murawskv. M: Cooper. GP. (1984). Effects of Pb2+ and Cd2+ on acetylcholine release and
Ca2+ movements in synaptosomes and subcellular fractions from rat brain and Torpedo electric organ.
Brain Res 323: 31-46." http://dx.doi.org/10.1016/0006-8993(84)90262-2.

Takahashi. A: Miczek. KA. (2014). Neurogenetics of aggressive behavior: Studies in rodents. In K Miczek; A
Meyer-Lindenberg (Eds.), Neuroscience of aggression (pp. 3-44). Berlin, Germany: Springer.
http://dx.doi.org/10.1007/7854 2013 263.

Tartaglione. AM: Serafini. MM: Raggi. A: Iacoponi. F: Zianni. E: Scalfari. A: Minghetti. L: Ricceri. L: Cubadda. F:
Calamandrei. G: Viviani. B. (2020). Sex-dependent effects of developmental lead exposure in Wistar rats:
Evidence from behavioral and molecular correlates. International Journal of Molecular Sciences 21: 2664.
http://dx.doi.org/10.3390/iims21082664.

3-532

-------
Tassiopoulos. K: Huo. Y: Braun. J: Williams. PL: Smith. R: Aschengrau. A: Nichols. S: Hazra. R: Mever. WA. Ill;
Knapp. K; Devgoo. NS: Seage. GR. III. (2017). Blood lead levels and neurodevelopmental function in
perinatally HIV-exposed, uninfected children in a U.S.-based longitudinal cohort study. AIDS Res Hum
Retroviruses 33: 919-928. http://dx.doi.org/10.1089/aid.2016.0265.

Tatsuta. N; Nakai. K; Kasanuma. Y; Iwai-Shimada. M; Sakamoto. M; Murata. K; Satoh. H. (2020). Prenatal and
postnatal lead exposures and intellectual development among 12-year-old Japanese children. Environ Res
189: 109844. http://dx.doi.Org/10.1016/i.envres.2020.109844.

Tatsuta. N; Nakai. K; Murata. K; Suzuki. K; Iwai-Shimada. M; Kurokawa. N; Hosokawa. T; Satoh. H. (2014).

Impacts of prenatal exposures to polychlorinated biphenyls, methylmercury, and lead on intellectual ability
of 42-month-old children in Japan. EnvironRes 133: 321-326.
http://dx.doi.Org/10.1016/i.envres.2014.05.024.

Tatsuta. N: Nakai. K: Murata. K: Suzuki. K: Iwai-Shimada. M: Yaginuma-Sakurai. K: Kurokawa. N: Nakamura. T:
Hosokawa. T: Satoh. H. (2012). Prenatal exposures to enviromnental chemicals and birth order as risk
factors for child behavior problems. Environ Res 114: 47-52.
http://dx.doi.Org/10.1016/i.envres.2012.02.001.

Tavakoli-Nezhad. M: Barron. AJ: Pitts. DK. (2001). Postnatal inorganic lead exposure decreases the number of
spontaneously active midbrain dopamine neurons in the rat. Neurotoxicology 22: 259-269.
http://dx.doi.org/10.1016/S0161-813X(01)00010-9.

Taylor. CM; Emond. AM; Lingam. R: Golding. J. (2018). Prenatal lead, cadmium and mercury exposure and

associations with motor skills at age 7 years in a UK observational birth cohort. Environ Int 117: 40-47.
http://dx.doi.Org/10.1016/i.envint.2018.04.032.

Taylor. CM; Humpliriss. R; Hall. A; Golding. J; Emond. AM. (2015). Balance ability in 7- and 10-year-old children:
Associations with prenatal lead and cadmium exposure and with blood lead levels in childhood in a
prospective birth cohort study. BMJ Open 5: e009635. http://dx.doi.org/10.1136/bmiopen-2015-009635.

Taylor. CM; Kordas. K; Golding. J; Emond. AM. (2017). Effects of low-level prenatal lead exposure on child IQ at
4 and 8 years in a UK birth cohort study. Neurotoxicology 62: 162-169.
http://dx.doi.Org/10.1016/i.neuro.2017.07.003.

Thangaraian. S: Vedagiri. A; Somasundaram. S: Sakthimanogaran. R; Murugesan. M. (2018). Neuroprotective

effect of morin on lead acetate- induced apoptosis by preventing cytochrome c translocation via regulation
of Bax/Bcl-2 ratio. Neurotoxicol Teratol 66: 35-45. http://dx.doi.Org/10.1016/i.ntt.2018.01.006.

Tian. Y; Green. PG: Stamova. B; Hertz-Picciotto. I: Pessah. IN; Hansen. R; Yang. X; Gregg. JP: Ashwood. P:

Jickling. G: Van de Water. J; Sharp. FR. (2011). Correlations of gene expression with blood lead levels in
children with autism compared to typically developing controls. Neurotox Res 19: 1-13.
http://dx.doi.org/10.1007/sl2640-009-9126-x.

Tlotleng. N; Naicker. N; Mathee. A; Todd. AC; Nkomo. P; Norris. SA. (2022). Association between bone lead
concentration and aggression in youth from a sub-cohort of the birth to twenty cohort. Int J Environ Res
Public Health 19: 2200. http://dx.doi.org/10.3390/iierphl904220Q.

Tong. S: Mcmichael. AJ; Baghurst. PA. (2000). Interactions between enviromnental lead exposure and
sociodemographic factors on cognitive development. Arch Environ Health 55: 330-335.
http://dx.doi.org/10.1080/00039890009604Q25.

Totsika. V: Svlva. K. (2004). The home observation for measurement of the enviromnent revisited. Child and
Adolescent Mental Health 9: 25-35. http://dx.doi.Org/10.1046/i.1475-357X.2003.00073.x.

Trombini. TV; Pedroso. CG: Ponce. D: Almeida. AA; Godinho. AF. (2001). Developmental lead exposure in rats: Is
a behavioral sequel extended at F2 generation? Pharmacol Biochem Behav 68: 743-751.
http://dx.doi.org/10.1016/S0091-3057(01)00473-7.

Tu. YO; Fan. GR: Wu. N; Wu. H; Xiao. HJ. (2021). Association of plasma lead, cadmium and selenium levels with
hearing loss in adults: National Health and Nutrition Examination Survey (NHANES) 2011-2012. Br J Nutr
128: 1100-1107. http://dx.doi.org/10.1017/S00071145210Q4335.

3-533

-------
Tung. PW: Burt. A: Karagas. M: Jackson. BP: Punshon. T: Lester. B: Marsit. CJ. (2022). Association between

placental toxic metal exposure and NICU Network Neurobehavioral Scales (NNNS) profiles in the Rhode
Island Child Health Study (RICHS). Environ Res 204: 111939.
http://dx.doi.Org/10.1016/i.envres.2021.111939.

Turnbull. DH: Mori. S. (2007). MRI in mouse developmental biology [Review]. NMR Biomed 20: 265-274.
http://dx.doi.org/10.1002/nbm. 1146.

U.S. EPA (U.S. Enviromnental Protection Agency). (1986). Air quality criteria for lead [EPA Report]. (EPA/600/8-
83/028aF-dF). Research Triangle Park, NC. http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=32647.

U.S. EPA (U.S. Enviromnental Protection Agency). (2006). Air quality criteria for lead [EPA Report].
(EPA/600/R-05/144aF-bF). Research Triangle Park, NC.
https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=158823.

U.S. EPA (U.S. Enviromnental Protection Agency). (2013). Integrated science assessment for lead [EPA Report].
(EPA/600/R-10/075F). Washington. DC. https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockey=P100K82L.txt.

U.S. EPA (U.S. Enviromnental Protection Agency). (2015). Preamble to the Integrated Science Assessments [EPA
Report]. (EPA/600/R-15/067). Research Triangle Park, NC: U.S. Enviromnental Protection Agency, Office
of Research and Development, National Center for Enviromnental Assessment, RTP Division.
https://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=310244.

Valeri. L: Mazumdar. MM: Bobb. JF: Henn. BC: Rodrigues. E: Sharif. OIA: Kile. ML: Ouamruzzaman. O: Afroz.
S: Golam. M: Amarasiriwardena. C: Bellinger. DC: Christian! DC: Coull. BA: Wright. RO. (2017). The
joint effect of prenatal exposure to metal mixtures on neurodevelopmental outcomes at 20-40 months of
age: Evidence from rural Bangladesh. Environ Health Perspect 125: 067015.
http://dx.doi.org/10.1289/EHP614.

Van Landingham. C: Fuller. WG: Schoof. RA. (2020). The effect of confounding variables in studies of lead
exposure and IQ [Review]. Crit Rev Toxicol 50: 815-825.
http://dx.doi.org/10.1080/10408444.2020.1842851.

van Wiingaarden. E: Winters. PC: Corv-Slechta. DA. (2011). Blood lead levels in relation to cognitive function in
older U.S. adults. Neurotoxicology 32: 110-115. http://dx.doi.Org/10.1016/i.neuro.2010.ll.002.

Vega-Dienstmaier. JM: Salinas-Pielago. JE: del Rosario Gutierrez-Campos. M: Mandamiento-Ayquipa. RD: del
Carmen Yara-Hokama. M: Ponce-Canchihuaman. J: Castro-Morales. J. (2006). Lead levels and cognitive
abilities in Peruvian children. Braz J Psychiatry 28: 33-39. http://dx.doi.org/10.1590/S1516-
44462006000100008.

Verma. M: Schneider. JS. (2017). Strain specific effects of low level lead exposure on associative learning and
memory in rats. Neurotoxicology 62: 186-191. http://dx.doi.Org/10.1016/i.neuro.2017.07.006.

Verma. SK: Dua. R: Gill. KD. (2005). Impaired energy metabolism after co-exposure to lead and ethanol. Basic Clin
Pharmacol Toxicol 96: 475-479. http://dx.doi.org/10.1111/i. 1742-7843.2005.pto 96611.x.

Vigeh. M: Yokovama. K: Matsukawa. T: Shinohara. A: Ohtani. K. (2014). Low level prenatal blood lead adversely
affects early childhood mental development. J Child Neurol 29: 1305-1311.
http://dx.doi.org/10.1177/0883073813516999.

3-534

-------
Vigueras-Villasefior. RM: Chavez-Saldafia. MP: Landero-Huerta. DA: Montes. S: Rios. C: Roias. P: Molina-

Obregon. HA: Duran. P: Roias-Castafieda. JC. (2021). Chronic lead exposure alters photic entrainment of
locomotor activity rhythm and neuronal photoactivation in the suprachiasmatic nucleus of the adult rat. J
Chem Neuroanat 117: 101991. http://dx.doi.Org/10.1016/i.ichemneu.2021.101991.

Villa-Cedillo. SA: Nava-Hernandez. MP: Soto-Dominguez. A: Hernandez-Ibarra. JA: Perez-Truiillo. JJ: Saucedo-
Cardenas. O. (2019). Neurodegeneration, demyelination, and astrogliosis in rat spinal cord by chronic lead
treatment. Cell Biol Int 43: 706-714. http://dx.doi.org/10.1002/cbin. 11147.

Vinceti. M: Filippini. T: Mandrioli. J: Violi. F: Bargellini. A: Weuve. J: Fini. N: Grill. P: Michalke. B. (2017). Lead,
cadmium and mercury in cerebrospinal fluid and risk of amyotrophic lateral sclerosis: A case-control study.
J Trace Elem Med Biol 43: 121-125. http://dx.doi.Org/10.ldl6/i.itemb.2016.12.012.

Vinceti. M: Guidetti. D: Bergomi. M: Caselgrandi. E: Vivoli. R: Olmi. M: Rinaldi. L: Rovesti. S: Solime. F. (1997).
Lead, cadmium, and selenium in the blood of patients with sporadic amyotrophic lateral sclerosis. Ital J
Neurol Sci 18: 87-92. http://dx.doi.org/10.1007/BF01999568.

Virgolini. MB: Rossi-George. A: Weston. D: Corv-Slechta. DA. (2008). Influence of low level maternal Pb

exposure and prenatal stress on offspring stress challenge responsivity. Neurotoxicology 29: 928-939.
http://dx.doi.Org/10.1016/i.neuro.2008.09.010.

Volman. V: Behrens. MM: Seinowski. TJ. (2011). Downregulation of parvalbumin at cortical GABA synapses
reduces network gamma oscillatory activity. JNeurosci 31: 18137-18148.
http://dx.doi.org/10.1523/JNEUR6sCI.3041-ll.2011.

Wan. C: Pan. S: Lin. L: Li. J: Dong. G: Jones. KC: Liu. F: Li. D: Liu. J: Yu. Z: Zhang. G: Ma. H (2021). DNA
methylation biomarkers of IQ reduction are associated with long-term lead exposure in school aged
children in Southern China. Environ Sci Technol 55: 412-422. http://dx.doi.org/10.1021/acs.est.0c01696.

Wang. DH: Xu. H: Zheng. YH: Gu. PS: Zhu. YJ: Ren. Y: Wang. SC: Yang. L: Xu. LW. (2020). Environmental
exposure to lead and cadmium and hearing loss in Chinese adults: A case-control study. PLoS ONE 15:
e0233165. http://dx.doi.org/10.1371/iournal.pone.0233165.

Wang. FT: Hu. H: Schwartz. J: Weuve. J: Spiro. AS. Ill: Sparrow. D: Nie. HL: Silvennan. EK: Weiss. ST: Wright.
RO. (2007). Modifying effects of the HFE polymorphisms on the association between lead burden and
cognitive decline. Environ Health Perspect 115: 1210-1215. http://dx.doi.org/10.1289/ehp.9855.

Wang. R: Wu. Z: Bai. L: Liu. R: Ba. Y: Zhang. H: Cheng. X: Zhou. G: Huang. H. (2021a). Resveratrol improved
hippocampal neurogenesis following lead exposure in rats through activation of SIRT1 signaling. Environ
Toxicol 36: 1664-1673. http://dx.doi.org/10.1002/tox.23162.

Wang. R: Wu. Z: Liu. M: Wu. Y: Li. O: Ba. Y: Zhang. H: Cheng. X: Zhou. G: Huang. H. (2021b). Resveratrol
reverses hippocampal synaptic markers injury and SIRT1 inhibition against developmental Pb exposure.
Brain Res 1767: 147567. http://dx.doi.Org/10.1016/i.brainres.2021.147567.

Wang. T: Guan. RL: Liu. MC: Shen. XF: Chen. JY: Zhao. MG: Luo. WJ. (2016). Lead exposure impairs

hippocampus related learning and memory by altering synaptic plasticity and morphology during juvenile
period. Mol Neurobiol 53: 3740-3752. http://dx.doi.org/10.1007/sl2035-Q15-9312-l.

Wang. X: Miller. G: Ding. G: Lou. X: Cai. D: Chen. Z: Meng. J: Tang. J: Chu. C: Mo. Z: Han. J. (2012). Health risk
assessment of lead for children in tinfoil manufacturing and e-waste recycling areas of Zhejiang Province,
China. Sci Total Environ 426: 106-112. http://dx.doi.Org/10.1016/i.scitotenv.2012.04.002.

Wang. XM: Liu. WJ: Zhang. R: Zhou. YK. (2013). Effects of exposure to low-level lead on spatial learning and

memory and the expression of mGluRl, NMDA receptor in different developmental stages of rats. Toxicol
Ind Health 29: 686-696. http://dx.doi.org/10.1177/0748233712436641.

Wang. Y: Wang. Y: Yan. C. (2022). Gender differences in trace element exposures with cognitive abilities of

school-aged children: A cohort study in Wujiang City, China. Environ Sci Pollut Res Int 29: 64807-64821.
http://dx.doi.org/10.1007/sll356-022-2Q353-4.

Wassennan. G. (2003). The relationship between blood lead, bone lead and child intelligence [Erratum], Child
Neuropsychol 9: 303-304. http://dx.doi.Org/10.1076/cliin.9.4.303.23520.

3-535

-------
Wasserman. G: Graziano. JH: Factor-Litvak. P: Popovac. D: Morina. N: Musabegovic. A: Vrenezi. N: Capuni-
Paracka. S: Lekic. V: Preteni-Rediepi. E: Hadzialievic. S: Slavkovich. V: Kline. J: Slirout. P: Stein. Z.
(1992). Independent effects of lead exposure and iron deficiency anemia on developmental outcome at age
2 years. J Pediatr 121: 695-703. http://dx.doi.org/10.1016/S0022-3476(05)81895-5.

Wassennan. GA: Factor-Litvak. P. (2001). Methodology, inference and causation: Enviromnental lead exposure and
childhood intelligence [Comment]. Arch Clin Neuropsychol 16: 343-352. http://dx.doi.org/10.1016/SQ887-
6177(00)00085-8.

Wassennan. GA: Factor-Litvak. P: Liu. X: Todd. AC: Kline. JK: Slavkovich. V: Popovac. D: Graziano. JH. (2003).
The relationship between blood lead, bone lead and child intelligence. Child Neuropsychol 9: 22-34.
http://dx.doi.org/10.1076/chin.9.1.22.14497.

Wassennan. GA: Liu. X: Pine. PS: Graziano. JH. (2001). Contribution of maternal smoking during pregnancy and
lead exposure to early child behavior problems. Neurotoxicol Teratol 23: 13-21.
http://dx.doi.org/10.1016/S0892-0362(00)00116-1.

Wassennan. GA: Musabegovic. A: Liu. X: Kline. J: Factor-Litvak. P: Graziano. JH. (2000). Lead exposure and
motor functioning in 4 1/2-year-old children: The Yugoslavia prospective study. J Pediatr 137: 555-561.
http://dx.doi.org/10.1067/mpd.200Q. 109111.

Wassennaa GA: Staghezza-Jaramillo. B: Slirout. P: Popovac. D: Graziano. J. (1998). The effect of lead exposure on
behavior problems in preschool children. Am J Public Health 88: 481-486.
http://dx.doi.Org/10.2105/AJPH.88.3.481.

Webster. R. (2001). Neurotransmitters, drugs and brain function. New York, NY: Wiley, https://www.wilev.com/en-
us/Neurotransmitters%2C+Drugs+and+Brain+Function-p-9780471978190.

Weiss. B. (1988). Neurobehavioral toxicity as a basis for risk assessment [Review]. Trends Phannacol Sci 9: 59-62.
http://dx.doi.org/10.1016/0165-6147(88)90118-6.

Weisskopf. MG: Proctor. SP: Wright. RO: Schwartz. J: Spiro. A. Ill: Spanow. D: Nie. HL: Hu. H. (2007).

Cumulative lead exposure and cognitive perfonnance among elderly men. Epidemiology 18: 59-66.
http://dx.doi.org/10.1097/01.ede.000Q248237.35363.29.

Weisskopf. MG: Weuve. J: Nie. H: Saint-Hilaire. MH: Sudarskv. L: Simon. DK: Hersh. B: Schwartz. J: Wright.
RO: Hu. H. (2010). Association of cumulative lead exposure with Parkinson's Disease. Environ Health
Perspect 118: 1609-1613. http://dx.doi.org/10.1289/ehp.1002339.

Weisskopf. MG: Wright. RO: Schwartz. J: Spiro. A. Ill: Spanow. D: Aro. A: Hu. H. (2004). Cumulative lead

exposure and prospective change in cognition among elderly men: The VA Nonnative Aging Study. Am J
Epidemiol 160: 1184-1193. http://dx.doi.org/10.1093/aie/kwh333.

Weston. HI: Westoa DP: Allen. JL: Corv-Slechta. DA. (2014). Sex-dependent impacts of low-level lead exposure
and prenatal stress on impulsive choice behavior and associated biochemical and neurochemical
manifestations. Neurotoxicology 44: 169-183. http://dx.doi.Org/10.1016/i.neuro.2014.06.013.

Westreich. D: Greenland. S. (2013). The table 2 fallacy: presenting and interpreting confounder and modifier
coefficients. Am J Epidemiol 177: 292-298. http://dx.doi.org/10.1093/aie/kws412.

Weuve. J: Kelsev. KT: Schwartz. J: Bellinger. D: Wright. RO: Raian. P: Spiro. A. Ill: Spanow. D: Aro. A: Hu. H.
(2006). Delta-aminolevulinic acid dehydratase polymorphism and the relation between low level lead
exposure and the Mini-Mental Status Examination in older men: The Nonnative Aging Study. Occup
Environ Med 63: 746-753. http://dx.doi.org/10.1136/oem.2006.027417.

Weuve. J: Korrick. SA: Weisskopf. MA: Ryan. LM: Schwartz. J: Nie. HL: Grodstein. F: Hu. H. (2009). Cumulative
exposure to lead in relation to cognitive function in older women. Environ Health Perspect 117: 574-580.
http://dx.doi.org/10.1289/ehp. 11846.

Weuve. J: Press. DZ: Grodstein. F: Wright. RO: Hu. H: Weisskopf. MG. (2013). Cumulative exposure to lead and
cognition in persons with Parkinson's disease. Mov Disord 28: 176-182.
http://dx.doi.org/10.1002/mds.25247.

3-536

-------
Whitcomb. S: Merrell. KW. (2012). Behavioral, social, and emotional assessment of children and adolescents (4th
ed.). New York, NY: Routledge. http://dx.doi.org/10.4324/9780203818596.

WHO (World Health Organization). (1948). Preamble to the Constitution of the World Health Organization as

adopted by the International Health Conference. New York, 19-22 June, 1946; signed on 22 July 1946 by
the representatives of 61 States (Official Records of the World Health Organization, no. 2, p. 100) and
entered into force on 7 April 1948. In Constitution of the World Health Organization (pp. 2). Geneva,
Switzerland, http://whqlibdoc.who.int/liist/official records/constitution.pdf.

Winter. AS: Sampson. RJ. (2017). From lead exposure in early childhood to adolescent health: A Chicago birth
cohort. Am J Public Health 107: 1496-1501. http://dx.doi.org/10.2105/AJPH.2017.303903.

Wright. JP: Dietrich. KN: Ris. MP: Hornung. RW: Wessel. SD: Lanphear. BP: Ho. M: Rae. MN. (2008).

Association of prenatal and childhood blood lead concentrations with criminal arrests in early adulthood.
PLoS Med 5: elOl. http://dx.doi.org/10.1371/iournal.pmed.0050101.

Wright. JP: Lanphear. BP: Dietrich. KN: Bolger. M: Tullv. L: Cecil. KM: Sacarellos. C. (2021). Developmental lead
exposure and adult criminal behavior: A 30-year prospective birth cohort study. Neurotoxicol Teratol 85:
106960. http://dx.doi.Org/10.1016/i.ntt.202L106960.

Wright. RO: Tsaih. SW: Schwartz. J: Spiro. A. Ill: McDonald. K: Weiss. ST: Hu. H. (2003). Lead exposure
biomarkers and mini-mental status exam scores in older men. Epidemiology 14: 713-718.
http://dx.doi.org/10.1097/01.EDE.0000081988.85964.db.

Wu. J: Basha. MR: Brock. B: Cox. DP: Cardozo-Pelaez. F: Mcpherson CA: Harry. J: Rice. DC: Malonev. B: Chen.
D: Laliiri. DK: Zawia. NH. (2008). Alzheimer's disease (AD)-like pathology in aged monkeys after
infantile exposure to enviromnental metal lead (pb): Evidence for a developmental origin and
enviromnental link for AD. J Neurosci 28: 3-9. http://dx.doi.org/10.1523/ineurosci.4405-07.2008.

Wu. S: Liu. H: Zhao. H: Wang. X: Chen. J: Xia. D: Xiao. C: Cheng. J: Zhao. Z: He. Y. (2020a). Enviromnental lead
exposure aggravates the progression of Alzheimer's disease in mice by targeting on blood brain barrier.
Toxicol Lett 319: 138-147. http://dx.doi.Org/10.1016/i.toxlet.2019.lL009.

Wu. Z: Bai. L: Tu. R: Zhang. L: Ba. Y: Zhang. H: Li. X: Cheng. X: Li. W: Huang. H. (2020b). Disruption of

synaptic expression pattern and age-related DNA oxidation in a neuronal model of lead-induced toxicity.
Environ Toxicol Pharmacol 76: 103350. http://dx.doi.Org/10.1016/i.etap.2020.103350.

Xiao. J: Wang. T: Xu. Y: Gu. X: Li. D: Niu. K: Wang. T: Zhao. J: Zhou. R: Wang. HL. (2020). Long-term probiotic
intervention mitigates memory dysfunction through a novel H3K27me3-based mechanism in lead-exposed
rats. Transl Psychiatry 10: 25. http://dx.doi.org/10.1038/s41398-020-Q719-8.

Xiao. L: Zan. G: Oin. J: Wei. X: Lu. G: Li. X: Zhang. H: Zou. Y: Yang. L: He. M: Zhang. Z: Yang. X (2021).

Combined exposure to multiple metals and cognitive function in older adults. Ecotoxicol Environ Saf 222:
112465. http://dx.doi.Org/10.1016/i.ecoenv.2021.112465.

Xiao. Y: Fu. H: Han X: Hu. X: Gu. H: Chen. Y: Wei. O: Hu. O. (2014). Role of synaptic structural plasticity in

impairments of spatial learning and memory induced by developmental lead exposure in Wistar rats. PLoS
ONE 9: ell5556. http://dx.doi.org/10.1371/iournal.pone.0115556.

Xiao. Y: Ma. B: McElhenv. D: Parthasarathv. S: Long. F: Hoslii. M: Nussinov. R: Isliii. Y. (2015). A(3(l-42) fibril
structure illuminates self-recognition and replication of amyloid in Alzheimer's disease. Nat Struct Mol
Biol 22: 499-505. http://dx.doi.org/10.1038/nsmb.2991.

Xu. L: Huo. X: Liu. Y: Zhang. YL: Oin. OL: Xu. XJ. (2020). Hearing loss risk and DNA methylation signatures in
preschool children following lead and cadmium exposure from an electronic waste recycling area.
Chemosphere 246: 125829. http://dx.doi.Org/10.1016/i.chemosphere.2020.125829.

Y Ortiz. MT: Tellez-Roio. MM: Treio-Valdivia. B: Schnaas. L: Osorio-Valencia. E: Coull. B: Bellinger. D: Wright.
RJ: Wright. RO. (2017). Maternal stress modifies the effect of exposure to lead during pregnancy and 24-
month old children's neurodevelopment. Environ Int 98: 191-197.
http://dx.doi.Org/10.1016/i.envint.2016.ll.005.

3-537

-------
Yang. W: Tian. ZK: Yang. HX: Feng. ZJ: Sun. JM: Jiang. H: Cheng. C: Ming. OL: Liu. CM. (2019). Fisetin

improves lead-induced neuroinflammation, apoptosis and synaptic dysfunction in mice associated with the
AMPK/SIRT1 and autophagy pathway. Food Chem Toxicol 134: 110824.
http://dx.doi.Org/10.1016/i.fct.2019.li0824.

Yang. X: Wang. B: Zeng. H: Cai. C: Hu. O: Cai. S: Xu. L: Meng. X: Zou. F. (2014). Role of the mitochondrial
Ca2+ uniporter in Pb2+-induced oxidative stress in human neuroblastoma cells. Brain Res 1575: 12-21.
http://dx.doi.Org/10.1016/i.brainres.2014.05.032.

Yang. Y: Ma. Y: Ni. L: Zhao. S: Li. L: Zhang. J: Faa M: Liang. C: Cao. J: Xu. L. (2003). Lead exposure through
gestation-only caused long-term learning/memory deficits in young adult offspring. Exp Neurol 184: 489-
495. http://dx.doi.org/10.1016/80014-4886(03)00272-3.

Yang. YW: Liou. SH: Hsueh. YM: Lvu. WS: Liu. CS: Liu. HJ: Chung. MC: Hung. PH: Chung. CJ. (2018). Risk of
Alzheimer's disease with metal concentrations in whole blood and urine: A case-control study using
propensity score matching. Toxicol Appl Pharmacol 356: 8-14.
http://dx.doi.Org/10.1016/i.taap.2018.07.015.

Yassa. HA. (2014). Autism: A form of lead and mercury toxicity. Environ Toxicol Pharmacol 38: 1016-1024.
http://dx.doi.Org/10.1016/i.etap.2014.10.005.

Ye. F: Li. X: Li. F: Li. J: Chang. W: Yuan. J: Chen. J. (2016a). Cyclosporin A protects against lead neurotoxicity
through inhibiting mitochondrial permeability transition pore opening in nerve cells. Neurotoxicology 57:
203-213. http://dx.doi.org/10.1016/i.neuro.26l6.10.004.

Ye. F: Li. X: Li. L: Yuan. J: Chea J. (2016b). t-BHQ provides protection against lead neurotoxicity via Nrf2/HO-l
pathway. Oxid Med Cell Longev 2016: 2075915. http://dx.doi.org/10.1155/2016/2075915.

Ye. F: Li. XY: Liu. YW: Jiang. AL: Li. XT: Yang. LY: Chang. W: Yuan. J: Chen. J. (2020). CypD deficiency

confers neuroprotection against mitochondrial abnormality caused by lead in SH-SY5Y cell. Toxicol Lett
323: 25-34. http://dx.doi.Org/10.1016/i.toxlet.2019.12.025.

Yin. JZ: E. M: Chao. H. (2021). Population-based study of enviromnental lead exposure and hearing loss: A
systematic review and meta-analysis [Review]. Public Health 197: 63-67.
http://dx.doi.Org/10.1016/i.puhe.2021.06.009.

Yorifuii. T: Debes. F: Weihe. P: Grandiean. P. (2011). Prenatal exposure to lead and cognitive deficit in 7- and 14-
year-old children in the presence of concomitant exposure to similar molar concentration of
methylmercury. Neurotoxicol Teratol 33: 205-211. http://dx.doi.Org/10.1016/i.ntt.2010.09.004.

You. YY: Sun. LG: Peng. B: Li. Y: Ben. SB: Gao. S. (2012). Increased hippocampal Disrupted-In-Schizophrenia 1
expression in mice exposed prenatally to lead. Neural Regen Res 7: 1939-1945.
http://dx.doi.Org/10.3969/i.issn.1673-5374.2012.25.003.

Yousef. AO: Fahad. AA: Abdel Moneim. AE: Metwallv. DM: El-Khadragy. MF: Kassab. RB. (2019). The
neuroprotective role of coenzyme Q10 against lead acetate-induced neurotoxicity is mediated by
antioxidant, anti-inflammatory and anti-apoptotic activities. Int J Environ Res Public Health 16: 2895.
http://dx.doi.org/10.3390/iierphl6162895.

Yu. YL: Tliiis. L: Saenen. N: Melgareio. JD: Wei. D: Yang. W: Yu. C: Roels. HA: Nawrot. T: Maestre. GE:

Staessen. J: Zhang. ZY. (2021). Two-year neurocognitive responses to first occupational lead exposure.
Scand J Work Environ Health 47: 233-243. http://dx.doi.org/10.5271/siweh.3940.

Yuan. W: Holland. SK: Cecil. KM: Dietrich. KN: Wessel. SD: Altave. M: Hornung. RW: Ris. MP: Egelhoff. JC:
Lanphear. BP. (2006). The impact of early childhood lead exposure on brain organization: A functional
magnetic resonance imaging study of language function. Pediatrics 118: 971-977.
http://dx.doi.org/10.1542/peds.2006-Q467.

Yun. S: Wu. Y: Niu. R: Feng. C: Wang. J. (2019). Effects of lead exposure on brain glucose metabolism and insulin
signaling pathway in the hippocampus of rats. Toxicol Lett 310: 23-30.
http ://dx. doi.org/10.1016/i .toxlet.2019.04.011.

3-538

-------
Yun. SW: Hover. S. (2000). Effects of low-level lead on glycolytic enzymes and pyruvate dehydrogenase of rat
brain in vitro: Relevance to sporadic Alzheimer's disease? Journal of Neural Transmission 107: 355-368.
http://dx.doi.org/10.1007/s00702005003Q.

Zhang. J: Cai. T: Zhao. F: Yao. T: Chen. Y: Liu. X: Luo. W: Chen. J. (2012). The role of a-synuclein and tau

hyperphosphorylation-mediated autophagy and apoptosis in lead-induced learning and memory injury. Int J
Biol Sci 8: 935-944. http://dx.doi.org/10.7150/iibs.4499.

Zhang. J: Yan. C: Wang. S: Hou. Y: Xue. G: Zhang. L. (2014). Chrysophanol attenuates lead exposure-induced
injury to hippocampal neurons in neonatal mice. Neural RegenRes 9: 924-930.
http://dx.doi.org/10.4103/1673-5374.133141.

Zhang. NH: Baker. HW: Tufts. M: Raymond. RE: Salihu. H: Elliott. MR. (2013). Early childhood lead exposure and
academic achievement: Evidence from Detroit public schools, 2008-2010. Am J Public Health 103: E72-
E77. http://dx.doi.org/10.2105/AJPH.2012.301164.

Zhang. RB: Huo. X: Ho. GY: Chen XJ: Wang. HW: Wang. TY: Ma. L. (2015a). Attention-deficit/hyperactivity

symptoms in preschool children from an e-waste recycling town: Assessment by the parent report derived
fromDSM-IV. BMC Pediatr 15: 51. http://dx.doi.org/10.1186/sl2887-015-0368-x.

Zhang. XL: Guariglia. S. R.: Mcglothan. JL: Stansfield. KH: Stanton PK: Guilarte. TR. (2015b). Presynaptic
mechanisms of lead neurotoxicity: Effects on vesicular release, vesicle clustering and mitochondria
number. PLoS ONE 10: e0127461. http://dx.doi.org/10.1371/iournal.pone.0127461.

Zhao. ZH: Du. KJ: Wang. T: Wang. JY: Cao. ZP: Chen. XM: Song. H: Zheng. G: Shen. XF. (2021). Maternal lead
exposure impairs offspring learning and memory via decreased GLUT4 membrane translocation. Front Cell
Dev Biol 9: 648261. http://dx.doi.org/10.3389/fcell.2021.648261.

Zhao. ZH: Zheng. G: Wang. T: Du. KJ: Han. X: Luo. WJ: Shen. XF: Chen. JY. (2018). Low-level gestational lead
exposure alters dendritic spine plasticity in the hippocampus and reduces learning and memory in rats. Sci
Rep 8: 3533. http://dx.doi.org/10.1038/s41598-018-21521-8.

Zheng. W: Shen. H: Blaner. WS: Zhao. O: Ren. X: Graziano. JH. (1996). Chronic lead exposure alters transthyretin
concentration in rat cerebrospinal fluid: The role of the choroid plexus. Toxicol Appl Pharmacol 139: 445-
450. http://dx.doi.org/10.1006/taap.1996.0186.

Zhou. CC: Gao. ZY: Wang. J: Wu. MO: Hu. S: Chen. F: Liu. JX: Pan. H: Yan CH. (2018). Lead exposure induces
Alzheimers's disease (AD)-like pathology and disturbes cholesterol metabolism in the young rat brain.
Toxicol Lett 296: 173-183. http://dx.doi.Org/10.1016/i.toxlet.2018.06.1065.

Zhou. F: Du. G: Xie. J: Gu. J: Jia. O: Fan. Y: Yu. H: Zha. Z: Wang. K: Ouvang. L: Shao. L: Feng. C: Fan. G.

(2020a). RyRs mediate lead-induced neurodegenerative disorders through calcium signaling pathways. Sci
Total Environ 701: 134901. http://dx.doi.Org/10.1016/i.scitotenv.2019.134901.

Zhou. L: Xu. J: Zhang. J: Yan. C: Lin. Y: Jia. Y: Hu. W. (2017). Prenatal maternal stress in relation to the effects of
prenatal lead exposure on toddler cognitive development. Neurotoxicology 59: 71-78.
http://dx.doi.Org/10.1016/i.neuro.2017.01.008.

Zhou. T: Guo. J: Zhang. J: Xiao. H: Oi. X: Wu. C: Chang. X: Zhang. Y: Liu. O: Zhou. Z. (2020b). Sex-specific

differences in cognitive abilities associated with childhood cadmium and manganese exposures in school-
age children: A prospective cohort study. Biol Trace ElemRes 193: 89-99.
http://dx.doi.org/10.1007/sl2011-019-dl7Q3-9.

Zhu. G: Dai. B: Chen. Z: He. L: Guo. J: Dan. Y: Liang. S: Li. G. (2019a). Effects of chronic lead exposure on the
sympathoexcitatory response associated with the P2X7 receptor in rat superior cervical ganglia. Auton
Neurosci 219: 33-41. http://dx.doi.Org/10.1016/i.autneu.2019.03.005.

Zhu. G: Fan. G: Feng. C: Li. Y: Chen. Y: Zhou. F: Du. G: Jiao. H: Liu. Z: Xiao. X: Lin. F: Yan. J. (2013). The
effect of lead exposure on brain iron homeostasis and the expression of DMT1/FP1 in the brain in
developing and aged rats. Toxicol Lett 216: 108-123. http://dx.doi.Org/10.1016/i.toxlet.2012.ll.024.

3-539

-------
Zhu. G: Peng. T: Peng. C: Li. H. (2019b). Chronic lead exposure decreases the expression of Huntingtin-associated
protein 1 (HAP1) through Repressor element-1 silencing transcription (REST). Toxicol Lett 306: 1-10.
http://dx.doi.Org/10.1016/i.toxlet.2019.02.003.

Zhu. X: Liu. X: Wei. F: Wang. F: Merzenich. MM: Schreiner. CE: Sun. X: Zhou. X. (2016). Perceptual training
restores impaired cortical temporal processing due to lead exposure. Cereb Cortex 26: 334-345.
http://dx.doi.org/10.1093/cercor/bhu258.

Zou. Y: Feng. W: Wang. W: Chen. Y: Zhou. Z: Li. O: Zhao. T: Mao. G: Wu. X: Yang. L. (2015). Protective effect
of porcine cerebral hydrolysate peptides on learning and memory deficits and oxidative stress in lead-
exposed mice. Biol Trace Elem Res 168: 429-440. http://dx.doi.org/10.1007/sl2011-015-0329-Q.

3-540

-------