* x>
PRO^
Integrated Review Plan for the National
Ambient Air Quality Standards for Lead.
Volume 3: Planning Document for Quantitative
Exposure/Risk Analyses
-------
-------
EPA-452/R-22-003c
May 2023
Integrated Review Plan for the National Ambient Air
Quality Standards for Lead.
Volume 3: Planning Document for Quantitative
Exposure/Risk Analyses
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Health and Environmental Impacts Division
Research Triangle Park, NC
-------
DISCLAIMER
This document has been prepared by staff in the U.S. Environmental Protection Agency's
Office of Air Quality Planning and Standards and National Center for Environmental
Assessment. Any findings and conclusions are those of the authors and do not necessarily reflect
the views of the Agency. This document is being circulated to facilitate discussion with the
Clean Air Scientific Advisory Committee and for public comment to inform the EPA's
consideration of the current review of the ozone national ambient air quality standards. This
information is distributed for purposes of pre-dissemination peer review under applicable
information quality guidelines. It does not represent and should not be construed to represent any
Agency determination or policy. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
Questions or comments related to this document should be addressed to Dr. Deirdre
Murphy, U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
C504-06, Research Triangle Park, North Carolina 27711 (email: murphy.deirdre@epa.gov).
1
-------
TABLE OF CONTENTS
1 Introduction/Background 1-1
1.1 Current Review of Air Quality Criteria and Standards for Lead 1-3
1.2 Multimedia, Multisource Aspects of Environmental Lead and Population Exposure. 1-4
1.3 Temporal Trends in emissions, Ambient Air Concentrations and Population Blood
Lead 1-5
1.3.1 Air Pb Emissions and Concentrations 1-5
1.3.2 Human Exposure and Population Blood Pb Concentrations 1-6
2 Quantitative Analysis Planning For the Primary Standard 2-1
2.1 Assessments Informing the Last Review 2-5
2.1.1 Summary of Design Aspects of the 2007 Assessment 2-8
2.1.2 Characterization of Variability 2-13
2.1.3 Key Limitations and Uncertainties 2-13
2.2 Key Considerations 2-16
2.2.1 Newly Available Information Regarding Key Limitations or Uncertainties of
2007 REA 2-19
2.2.2 Consideration of Other Health Endpoints and/or Population Age Groups/Life
Stages 2-32
2.3 Initial Plans for the Current Review 2-38
3 Quantitative Analysis Planning for the Secondary Standard 3-1
3.1 Consideration of 2006 Assessment in Prior Reviews 3-2
3.2 Consideration of the Available Evidence In the Current Review 3-9
3.2.1 Linking Atmospheric Pb to Non-Air Media Concentrations 3-11
3.2.2 Exposure Assessment Tools and Factors Affecting Pb Bioavailability 3-15
3.3 Key Observations and Initial Plans for the Current Review 3-17
4 References 1
Appendix: Cumulative Exposure Estimates for Different Birth Cohorts
li
-------
TABLE OF FIGURES
Figure 1-1.Long-term trend in air Pb emissions (U.S. EPA, 2023b) 1-6
Figure 1-2. Prevalence of blood Pb concentrations at or above 5 |ig/dL in children, aged 1-5
years (from Egan et al. [2021] Table 4) 1-8
Figure 1-3.Blood Pb concentration estimates in the general population by age group from
NHANES (CDC, 2021) for median (upper panel) and 95th percentile (lower
panel) 1-9
Figure 1-4.Median (upper panel) and 95th percentile (lower panel) blood Pb estimates from
NHANES for young children of differing race. NH, non-Hispanic; Mex-Am,
Mexican American 1-11
Figure 2-1. Summary of health risk assessment approaches that have been employed in
NAAQS reviews 2-4
Figure 2-2. Simplified presentation of air-related Pb exposure pathways 2-6
Figure 2-3. Analytical approach for two case study categories in 2016 review 2-9
Figure 2-4. Conceptual model for 2007 Pb human health risk assessment 2-18
Figure 2-5. Analytical approach for the REA under consideration for the current review 2-41
TABLE OF TABLES
Table 1-1. Schedule for the review of ambient air quality criteria and NAAQS for Pb 1-4
Table 1-2. Changes in blood Pb concentrations from the first to second NHANES 1-7
Table 1-3. Age-specific cumulative lead exposures estimated for general population cohorts
born in the years 1945, 1970, 1990 and 2010 1-13
Table 1-4. Age-specific cumulative lead exposures estimated for non-Hispanic Black and
full population child cohorts born in the years 1990 and 2010 1-14
Table 2-1. Summary of approaches used to estimate case study media concentrations in
2007 REA (based on 2014 PA, Table 3-7) 2-7
Table 2-2. Summary of approaches used to estimate case study media concentrations in
2007 REA (based on 2014 PA, Table 3-7) 2-11
Table 2-3. Assessment of information (including methods and models) newly available in
this review related to quantitative assessment of exposure and risk of IQ
decrements in children 2-22
Table 2-4. Assessment of information (including methods and models) newly available in
this review related to quantitative assessment of Pb endpoints other than IQ
decrements in children 2-34
iii
-------
Table 3-1. Limitations and uncertainties of the exposure/risk analyses for the 2008 review,
and consideration of related newly available information. Drawn from the 2011
REA Planning Document, 2014 PA; 2015 and 2016 notices of proposed and final
decisions; and draft ISA 3-7
Table 3-2. Additional studies/surveys available since the 2008 review that may inform
characterization of non-air exposures associated with air Pb concentrations 3-10
Table 3-3. Studies describing temporal trends in Pb in soil, peat and terrestrial biota 3-13
Table 3-4. Studies describing temporal trends in Pb in sediments of freshwater lakes 3-14
Table 3-5. Studies describing temporal trends in Pb in marine biota 3-15
iv
-------
PREFACE
The planning phase of the U.S. Environmental Protection Agency's (EPA's) reviews of
the national ambient air quality standards (NAAQS) includes development of an integrated
review plan (IRP) which is made available for public comment and provided to the Clean Air
Scientific Advisory Committee (CASAC) for review or consultation. As a result of recent efforts
to improve the efficiency of the planning phase and to facilitate the receipt of timely input from
the CASAC and the public, the IRP for the current review of the lead NAAQS is comprised of
three volumes. Volume 1 provides background information on the air quality criteria and
standards for lead, and may serve as a reference for the public and the CASAC in their
consideration of the subsequent two volumes. Volume 2 addresses the general approach for the
review and planning for the integrated science assessment (ISA) and was the subject of a
consultation with the CASAC in 2022. Volume 3 (this document) is the planning document for
quantitative analyses to be considered in the policy assessment (PA), including exposure and risk
analyses, and will be the subject of a consultation with the CASAC in 2023. It will describe key
considerations in EPA's development of the PA and planning with regard to any quantitative
exposure/risk analyses to inform the review. In order that consideration of the availability of new
evidence in the review can inform these plans, the development and public availability of
Volume 3 has generally coincided with that of the draft ISA.
v
-------
1 INTRODUCTION/BACKGROUND
The U.S. Environmental Protection Agency (EPA) is conducting a review of the air
quality criteria and the national ambient air quality standards (NAAQS) for lead (Pb). Plans for
the review have been described in Volumes I and II of the Integrated Review Plan for this review
(U.S. EPA, 2022a, b; hereafter referred to as the IRP, volumes 1 and 2). This third volume (titled
Integrated Review Plan for the Lead National Ambient Air Quality Standards, 3. Planning
Document for Quantitative Exposure/Risk Analyses - hereafter referred to as the IRP, volume 3)
is the planning document for quantitative analyses to be considered in the policy assessment
(PA), including exposure and risk analyses. It describes key considerations in planning for any
quantitative exposure/risk analyses to inform the review. The purpose of this planning document
is to describe the consideration of the extent to which newly available scientific evidence and
tools or methodologies warrant the conduct of quantitative risk and exposure analyses that might
inform this review, and as warranted, the initial planning for such analyses. Also considered is
the extent to which newly available evidence may refine our characterization of exposure and
risk estimates available in the last review.
The current review of the Pb air quality criteria and standards builds on the substantial
body of work done during the course of prior reviews, represented both in comprehensive
science assessments (e.g., U.S. EPA, 2006a [2006 AQCD]; U.S. EPA, 2013a [2013 ISA]) and
past quantitative exposure and risk analyses.1 These different types of information, evaluated in
past policy assessments, provided the basis for decisions on the existing Pb NAAQS. This
planning document presents a critical evaluation of information related to Pb human and
ecological exposure and risk (e.g., data, modeling approaches) newly available in this review as
identified in the first draft Integrated Science Assessment for Lead (draft ISA; U. S. EPA, 2023a).
The focus of this evaluation is consideration of the risk and exposure analyses for health and
ecological risk warranted by the newly available evidence.
This document is intended to facilitate consultation with the Clean Air Scientific
Advisory Committee (CASAC), as well as public review, for the purpose of obtaining advice on
EPA's consideration of the recently available evidence (information, methods, etc.) as it relates
to its potential impact on quantitative exposure and risk analyses, both with regard to
consideration of the extent to which new assessments are warranted in this review and with
1 For example, in the 2008 review, EPA staff designed and conducted a complex multimedia, multipathway health
risk assessment involving case studies representing different ambient air Pb exposure circumstances, and an
assessment of the available information on ecological impacts of Pb, including the consideration of potentially
vulnerable ecosystems.
1-1
-------
regard to our consideration of prior assessments in evaluating risk and exposure-related
considerations in our Policy Assessment for this review. The discussion in this document is
intended to build upon the exposure and risk assessment approaches employed in past reviews,
and on Agency experience with Pb exposure and risk assessment since that time, while also
drawing from information presented in the March 2023 draft of the ISA for the current review.
Legislative Requirements
Sections 108 and 109 of the Clean Air Act (Act) govern the establishment and periodic
review of the NAAQS. These standards are established for pollutants that may reasonably be
anticipated to endanger public health and welfare, and whose presence in the ambient air
results from numerous or diverse mobile or stationary sources. The NAAQS are to be based
on air quality criteria, which are to accurately reflect the latest scientific knowledge useful in
indicating the kind and extent of all identifiable effects on public health or welfare which may
be expected from the presence of the pollutant in ambient air. The EPA Administrator is to
promulgate and periodically review, at five-year intervals, "primary" (health-based) and
"secondary" (welfare-based)1 NAAQS for such pollutants.1 Based on periodic reviews of the
air quality criteria and standards, the Administrator is to make revisions in the criteria and
standards, and promulgate any new standards, as may be appropriate. The Act also requires
that an independent scientific review committee advise the Administrator as part of this
NAAQS review process, a function now performed by the CASAC.
1 The legislative history of section 109 indicates that a primary standard is to be set at "the maximum permissible
ambient air level ... which will protect the health of any [sensitive] group of the population," and that for this
purpose "reference should be made to a representative sample of persons comprising the sensitive group rather
than to a single person in such a group." S. Rep. No. 91-1196, 91st Cong., 2d Sess. 10 (1970).
2Under CAA section 302(h) (42 U.S.C. § 7602(h)), effects on welfare include, but are not limited to, "effects on
soils, water, crops, vegetation, manmade materials, animals, wildlife, weather, visibility, and climate, damage
to and deterioration of property, and hazards to transportation, as well as effects on economic values and on
personal comfort and well-being."
1-2
-------
Additional background on this Pb NAAQS review is presented in the IRP, volume 1.
That document describes the preparation of key documents in the NAAQS review process
including an ISA and a PA. The ISA provides a critical assessment of the latest available
scientific information upon which the NAAQS are to be based, and the PA evaluates the policy
implications of the information contained in the ISA and of any policy-relevant quantitative
analyses, such as quantitative human and/or ecological risk and exposure assessments, that were
performed for the review or for past reviews. Based on this evaluation, the PA presents staff
conclusions regarding standard-setting options for the Administrator to consider in reaching
decisions on the NAAQS.2
1.1 CURRENT REVIEW OF AIR QUALITY CRITERIA AND
STANDARDS FOR LEAD
In July 2020, EPA announced the initiation of the current periodic review of the air
quality criteria for Pb and the Pb NAAQS and issued a call for information in the Federal
Register (85 FR 40641, July 7, 2020). Volumes 1 and 2 of the IRP were released in March 2022
(U.S. EPA, 2022a,b). The former provides background on the Pb NAAQS, and the latter is the
planning document for the review and the ISA. Volume 2 was the subject of a consultation with
CAS AC on_ April 8, 2022 (87 FR 15985, March 21, 2022; Shepard, 2022). With consideration of
input received during this consultation, EPA has separately developed a draft ISA, released on
March 31, 2023, for public comment and CASAC review (88 FR 19302, March 31, 2023).
With consideration of the newly available evidence identified in the draft ISA, the EPA
has developed this planning document for quantitative analyses, including exposure/risk
analyses, that might be warranted to inform decisions in the current review. This planning
document comprises the third volume of the IRP. With consideration of the CASAC review of
the draft ISA and consultation discussions on volumes 2 and 3 of the IRP, the EPA will develop
a draft of the PA (with associated policy evaluations and quantitative analyses) for public and
CASAC review. The current timeline projected for completing this Pb NAAQS review is
presented in Table 1-1. Completion of the final ISA and PA are projected in Spring 2024 and
Fall 2024, respectively. These will be followed by proposed and final decisions for the review,
projected in 2025 and 2026, respectively.
2 Decisions on the NAAQS involve consideration of the four basic elements of a standard: indicator, averaging time,
form, and level. The indicator defines the pollutant to be measured in the ambient air for the purpose of
determining compliance with the standard. The averaging time defines the time period over which air quality
measurements are to be obtained and averaged, considering evidence of effects associated with various time
periods of exposure. The form of a standard defines the air quality statistic that is to be compared to the level of
the standard in determining whether an area attains the standard.
1-3
-------
Table 1-1. Schedule for the review of ambient air quality criteria and NAAQS for Pb.
Stage of Review
Major Milestone
Target Dates
Planning
Federal Register Call for Information
July 7, 2020
Integrated Review Plan (IRP), volumes 1 and 2
March 2022
CASAC consultation on IRP, volume 2
April 2022
IRP, volume 3
May 2023
CASAC consultation on IRP, volume 3
June 2023
Science Assessment
External review draft of ISA
March 2023
CASAC public meeting for review of draft ISA
June 2023
Final ISA
Spring 2024
Quantitative
Exposure/Risk
Analyses and Policy
Assessment
External review draft of PA (including quantitative air
quality, exposure and/or risk analyses, as warranted)
CASAC public meeting for review of draft PA
Final PA
Winter 2023/24
2024
Fall 2024
Regulatory Process
Notice of proposed decision
2025
Notice of final decision
2026
1.2 MULTIMEDIA, MULTISOURCE ASPECTS OF ENVIRONMENTAL
LEAD AND POPULATION EXPOSURE
Unlike most other pollutants for which NAAQS are established, Pb is a multimedia,
multisource pollutant. Other sources of Pb in the environment include surface water discharges,
industrial and other historically contaminated sites, as well as Pb associated with past uses of Pb-
based paint. Other sources of Pb to human exposures include plumbing that includes Pb solder or
piping, as well as some imported dietary items and consumer products (draft ISA, Appendix 2).
While such exposures remain today, exposures related to many nonair sources that occurred in
the past, such as from the use of Pb solder in canned foods, have been dramatically reduced or
eliminated for today's populations (2013 ISA, section 3.1).
Lead emitted into ambient air may subsequently occur in multiple environmental media,
contributing to multiple pathways of exposure for humans and ecological receptors. This
multimedia distribution of, and multipathway exposure to, air-related Pb has a key role in the
Agency's consideration of the Pb NAAQS. Lead emitted into the air is predominantly in
particulate form, which can be transported long or short distances depending on particle size.
Exposure to Pb emitted into the ambient air (air-related Pb) can occur directly by inhalation or
indirectly by ingestion of Pb-contaminated food, water or other materials including dust and soil.
These exposures occur as Pb emitted into the ambient air is distributed to other environmental
media and can contribute to human exposures via indoor and outdoor dusts, outdoor soil, food
1-4
-------
and drinking water, as well as inhalation of air. Air-related exposure pathways are affected by
changes to air quality, including changes in concentrations of Pb in air and changes in
atmospheric deposition of Pb. Further, because of its persistence in the environment, Pb
deposited from the air may contribute to human and ecological exposures for years into the
future. Thus, the roles of both air concentration and air deposition in human exposure pathways,
and the persistence of Pb once deposited, influence the dynamics of the response of the various
Pb exposure pathways to changes in air quality.
1.3 TEMPORAL TRENDS IN EMISSIONS, AMBIENT AIR
CONCENTRATIONS AND POPULATION BLOOD LEAD
As background for the subsequent chapters, this section presents trends in air Pb
emissions and concentrations (section 1.3.1) and trends in population blood Pb concentrations
(section 1.3.2). Section 1.3.2 also summarizes patterns of blood Pb concentrations across
demographic groups, and estimates of the appreciable difference in Pb exposure history
experienced by population cohorts born at different times in the past.
1.3.1 Air Pb Emissions and Concentrations
Air Pb emissions have decreased substantially over the past 50 years, with the most
dramatic reductions occurring between 1970 and 1995 due to the removal of Pb from gasoline
used in highway vehicles (2014 PA, Figure 2-1). For example, total air Pb emissions in the U.S.
are estimated to have declined from over 200,000 tpy in 1970 to approximately 5,000 tpy in
1990. Lead emissions have continued to decrease since 1990 (Figure 1-1), with significant
reductions occurring in the metals industries as a result of national emissions standards for
hazardous air pollutants.
1-5
-------
5
0
Figure 1-1.
In response to reductions in air emissions, air Pb concentrations have also declined over
the past three to four decades. For example, as reported in the 2013 Pb ISA, from 1980 to 2010
the national median of maximum 3-month average Pb concentrations of 74 Pb-TSP monitors in
the U.S. declined from 0.87 [^g/m3 in 1980 to 0.03 [j,g/m3 in 2010, as a result of the phase-out of
leaded gasoline for automobiles and reductions of industrial use and processing of Pb (2013 ISA,
section 1.2.2). Although many of the network monitors on which these observations are based
are sited near major industrial sources, the 2013 Pb ISA also documented a strong decreasing
trend in ambient air Pb concentrations from five monitors in the vicinity of major roadways in
urban areas. The average of Pb-TSP concentrations from these monitors dropped from 0.90
|ig/m3 in 1980 to 0.18 |ig/m3 in 1986 to 0.01 |ig/m3 by 2010 (2013 ISA, section 2.5.1.2; 2014
PA, section 2.2.2.1). Air Pb concentrations have continued to decline since 2010 (draft ISA,
Appendix 1, section 1.5.1). These long-term national trends are generally reflected in long-term
local and regional observations of Pb concentrations in the U.S. and Canada for diverse locations
and conditions (draft ISA, Appendix 1, section 1.5.1).
1.3.2 Human Exposure and Population Blood Pb Concentrations
As described in the 2013 ISA and prior AQCDs, blood Pb is a commonly used exposure
metric that reflects human exposures across all routes and pathways. Upon absorption into the
body (via gastrointestinal tract or alveoli of the lungs, depending on pathway), individuals'
multimedia Pb exposures are reflected in the blood. The bloodstream provides distribution of Pb
to organs throughout the body with vast majority of Pb accumulating in bone which serves as a
1-6
Inventory Year
Long-term trends in air Pb emissions (U.S. EPA, 2023b).
Industrial and Other Processes
Stationary Fuel Combustion
Non-Road Mobile
Highway Vehicles
~
~
~
~
-------
long-term storage depot over a person's lifetime. Thus, bone Pb is another metric of exposure
(that is less easily monitored in a population). During times of physiological changes, certain
disease states and lifestages, such as during pregnancy and older age, Pb is also released from
bone back into blood (2013 ISA; 2006 AQCD).
1.3.2.1 Population Blood Pb Levels By Age and Demographic Group Over Time
Over the last four and a half decades, blood Pb levels (BLLs) in the U.S. population have
declined dramatically, with the most substantial reduction occurring between the 1970s and
1990s. Across ages, the age group with the highest levels is young children, aged one to five
years (as discussed below). The median level for this age group declined more than a factor of
four from 15 |ig/dL in the late 1970s to 3.5 |ig/dL by the late 1980s, and has declined by nearly
another factor of six since that time, to 0.6 |ig/dL in the 2017-2018 NHANES. Similar reductions
have also been documented in adults older than 19 years (Table 1-2). The reductions have also
been documented at the upper ends of the distribution. For example, the 95th percentile blood Pb
concentration for young children has declined from 29 |ig/dL in the late 1970s to 12 |ig/dL in the
late 1980s to 2 |ig/dL in the 2017-2018 NHANES, more than a factor of 10 in all (Table 1-2;
CDC, 2021). This is also illustrated by the decline in blood Pb concentrations in young children
that exceeded 5 |ig/dL through the 1990s, during which the prevalence of such levels dropped by
80 to 90% (Figure 1-2). This period, from the late 1970s through the 1990s corresponds to the
phase out of leaded gasoline in all cars and trucks.
Table 1-2. Changes in blood Pb concentrations from the first to second NHANES.
NHANES Years
Median Concer
1-5 years
itration (|jg/dL)
> 20 years
95th Percentile
1-5 years
Concentration (|jg/dL)
> 20 years
1976-1980
15
13
29
26
1988-1991
3.5
3.2
12
9.5
Sources: Pirkleetal., (1994) and U.S. EPA., (2013b).
The 1976-1980 NHANES blood Pb levels were strongly associated with usage of leaded
gasoline (metric tons per day) and population blood Pb levels (Schwartz and Pitcher, 1989).
Although national estimates of blood Pb concentrations are not available prior to the 1976-1980
NHANES, the relationship of blood Pb concentrations with leaded gasoline usage indicates the
blood Pb concentrations at that time may be a reasonable representation of levels during previous
years of leaded gasoline usage. Leaded gasoline was used in U.S. on-road vehicles from the
1930s through 1990s, with usage peaking in the 1970s (2013 ISA, Figure 2-7). Estimated U.S.
air Pb emissions in 1970 was more than 200,000 tons, with a subsequent declining trend
dominated by highway vehicle emissions through 1985, by which time the national estimate
from all sources had dropped by more than an order of magnitude from 1970 (2013 ISA, Figure
2-1).
1-7
-------
Figure 1-2. Prevalence of blood Pb concentrations at or above 5 jig/dL in children, aged 1-
5 years (from Egan et al. [2021] Table 4).
As individuals age, behaviors that influence exposure, such as crawling and hand-to-
mouth contact of babies and toddlers, change, such that exposures via some significant pathways
are reduced. This contributes to the lower blood Pb levels observed in primary school children
compared to pre-school age children, one to five years old (Figure 1-3).3 Factors that contribute
to adult blood Pb levels being higher than teenage blood Pb levels include higher exposures of
individuals in that age group during their earlier years (compared to those of the teenagers), bone
remodeling in older adults in the age group, and, potentially, contributions from occupational
exposures (Figure 1-3; 2013 ISA, 2006 AQCD; draft ISA, Appendix 2). As can be seen from
Figure 1-3, blood Pb levels (BLLs) have continued to decline in all ages (Figure 1-3).
3 Since 1976, the U.S. Centers for Disease Control and Prevention (CDC) has been monitoring blood Pb levels
nationally through the National Health and Nutrition Examination Survey (NHANES) and provides nationally
representative biomonitoring data for Pb. NHANES is designed to assess the health and nutritional status of the
civilian noninstitutionalized U.S. population and is conducted by the National Center for Health Statistics, part of
the CDC. The NHANES involves interviews and physical examinations with approximately 10,000 people in
each two-year year survey cycle. CDC's National Center for Enviromnental Health measures concentrations of
enviromnental chemicals in blood and urine samples collected from NHANES participants. Summaries of the
measured values for lead and more than 200 other chemicals since 1999-2000 are provided in the CDC's National
Report on Human Exposure to Enviromnental Chemicals (CDC, 2022).
1-8
-------
2.5
-o
20yr
~
A
¦
A
¦
¦
¦
¦
A
A
¦
A
¦
¦
~
A
A
A
~
x,x, x x x %
% % \ % % °%> % % %
_CD
1 7
~CT>
_o
Ma
CD
O
c
o
o 2
_Q
Q_
T3
° -1
O 1
CD
1-5 yr
A 6-11 yr
« 12-19 yr
¦
»
¦ >20yr
A
¦
¦
¦
i
~
~
A
~
1
¦
¦
_ _ ^
~
~
A
A
~
V
¦
A
A
~
4
A
A
~
%
^ ^ ^ ^ <5?/7 <5^ *b;
^ ^ ""%&> *°70 *°7j *°7
-------
Although blood Pb levels in U.S. children have declined, the mean blood Pb levels
reported in NHANES consistently differ among children of different ethnic backgrounds, and for
children from low-income households compared to the general population as a whole. More
specifically, blood Pb levels for non-Hispanic Black children have historically been higher than
levels for White children and for other minority populations, as well as for the U.S. population of
children as a whole (Figure 1-4; 2013 ISA, sections 3.4.1, 5.2.1.1 and 5.2.3). Similarly, blood Pb
levels in children from low-income households have also been historically higher than those for
other households. For example, among children, ages one to five years, in the 1991-94 NHANES,
the median blood Pb level for non-Hispanic Black children was nearly two-fold higher than the
median for White non-Hispanic children (4.3 |j,g/dL compared to 2.3 |j,g/dL), as seen in Figure 1-
4. Additionally, the median blood Pb level in young children from low-income households in
1991-94 NHANES was nearly 1.5 times higher than the median for the general population as a
whole (4.1 |j,g/dL compared to 2.6 |J,g/dL) (U.S. EPA 2022).
The reductions in blood Pb levels across the last several decades have occurred across
these demographic groups. Further, the gap between blood Pb levels from these populations
compared to the general population has also narrowed (Figure 1-4). For example, while the
1991-94 median Pb lead level in non-Hispanic Black children was nearly 2-fold higher than the
median for White children, by the 2013-2016 NHANES, the median blood Pb level for non-
Hispanic Black children had decreased to 0.9 ng/dl, less than 30% higher than the median of 0.7
Hg/dl for White children (Figure 1-4; U.S. EPA 2003, 2013b, 2019, 2022c). Similarly for low-
income children, the difference in median blood Pb level from that for the full population
narrowed from nearly a factor of 1.5 in 1991-94 to less than 30% (e.g., 0.9 |j,g/dL compared to
0.7 ng/dL) by the 2013-16 NHANES (U.S. EPA, 2022).
1-10
-------
ca
"O _
CD 5
33
*01 4,
ZL *
c
0
'2 3
¦»-'
c
1
o
B
0 2
o ^
J 1
CO
CD
o
©
o_
uo
a"J
cr>
¦2 3
CO
2013-2016
All races White-NH Black-NH Mex-Am
¦ 2007-2010 1 2009-2012 ¦ 2013-2016
Other
Figure 1-4. Median (upper panel) and 95th percentile (lower panel) blood Pb estimates
from NHANES for young children, aged one to five years, of differing race.
NH, non-Hispanic; Mex-Am, Mexican American.
1-11
-------
1.3.2.2 Cumulative Exposure for Different Birth Cohorts
Cumulative Pb exposure may be indicated by bone Pb concentration (e.g., tibia) or a
cumulative blood Pb metric derived from longitudinal BLL (2013 ISA, section 3.3.1). Blood Pb
is an indicator of Pb that is circulating through an individual at a point in time and, accordingly,
may best reflect recent exposures and to some extent past exposure depending on factors like age
and disease state, as well as differences of past and recent exposures. It cannot, however, fully
reflect an individual's exposure history. Because Pb is stored in the bone, bone Pb is considered
to provide a measure of an individual's cumulative lifetime exposure. Additionally, where
longitudinal blood Pb measurements are available, a cumulative blood Pb metric is also
informative regarding an individual's exposure across their lifetime. A cumulative blood Pb
metric may be derived in a manner conceptually similar to a cumulative exposure metric often
employed in worker studies (e.g., using ppm-days). The metric is derived essentially by
multiplying blood Pb measurement by years to which it is applied yielding an index in terms of
(|j,g/dL)-years (e.g., Nie et al., 2011 and Somervaille et al., 1988; 2013 ISA, section 3.3.1).
Given the backdrop of dramatic reductions in Pb exposures over the past four plus
decades, as evidenced in blood Pb levels of young children during that time, adults in the U.S.
population have significant Pb exposure histories. Further, adults of different ages may have
appreciably different exposure histories. This is especially true of adults who lived through
periods of the use of leaded on-road motor vehicle fuel compared to much younger adults. Such
differences in exposure history can be illustrated through use of a cumulative blood Pb index
(CBLI), a metric sometimes used as an exposure metric in epidemiological studies of adult
populations (2013 ISA, section 3.3.5; Nie et al., 2011; Hu et al., 2007; Somervaille et al., 1988).
Estimates of CBLI4 for different birth cohorts, the derivation of which is described in
more detail in the Appendix, can illustrate of the appreciably different Pb exposure histories of
adults in today's population and the difference of those histories from adults of tomorrow (Table
1-3). The limitations in the analysis, and its underlying datasets, and associated uncertainties
associated with the resultant CBLI estimates (as summarized in the Appendix) are important
considerations in their interpretation. Consistent with these considerations, the estimates in Table
1-3 provide a sense of the magnitude of the differences in exposure histories for several birth
cohorts of the U.S. general population that are represented among the adults of today and the
future.
4 The CBLI is calculated as: CBLI = £=i 0-5 x (PbBi+1 + PbBt) x (ti+1 - tj), where blood Pbi and blood Pb, i
are the blood lead concentrations at two consecutive times fc and fc+i.
1-12
-------
Table 1-3. Age-specific cumulative lead exposures estimated for general population
cohorts born in the years 1945,1970,1990 and 2010.
CBLI Estimate, (|Lxg/dL)-yrs, at Specified Age for Cohorts Born in:
Age (years)
1945
1970
1990
2010
5
74 a
74 a
89
4.3
10
CO
CO
>
CO
CO
>
23
7B
21
274 a
197
O
CO
CO
12B
30
395
212
39 c
16B
50
549
235°
53 c
24 b
75
587
256°
A Estimated CBLI for ages <25 for the 1945 cohort and <10 for the 1970 cohort occur in years
before the earliest NHANES (1970), which may contribute a bias low given higher Pb exposures
from the use of leaded gasoline. See Appendix for details.
B Estimated CBLI for ages at/above 10 years for the 2010 cohort are projected. In CBLI derivation,
the most recent NHANES estimates (2018) were used, which may contribute a bias high given
backdrop of declining Pb exposures. See Appendix for details.
c Estimated CBLI for ages at/above 20 years for the 1990 cohort are projected. In CBLI derivation,
the most recent NHANES estimates (2018) were used, which may contribute a bias high given
backdrop of declining Pb exposures. See Appendix for details.
D Estimated CBLI for ages at/above 48 years for the 1970 cohort are projected. In CBLI derivation,
the most recent NHANES estimates (2018) were used, which may contribute a bias high given
backdrop of declining Pb exposures. See Appendix for details.
As illustrated described in section 1.3.2.1 above, NHANES blood Pb estimates for low-
income and non-Hispanic Black populations have historically been appreciably greater than
those for the general population. The ramifications of this on estimates of cumulative childhood
exposure for cohorts born in 1990 and 2010 is illustrated by CBLI estimates presented in 0.
Notably for the 2010 birth cohort, while the CBLI estimate for the non-Hispanic Black
population is still slightly greater than that for the full cohort (all races), the gap between the two
is considerably narrower relative to its size for the 1990 birth cohort.5
5 Based on NHANES estimates and trends for low-income children, a similar pattern would be expected for low-
income cohorts.
1-13
-------
Table 1-4. Age-specific cumulative lead exposures estimated for non-Hispanic Black and
full population child cohorts born in the years 1990 and 2010.
CBLI Estimate, (|xg/dL)-yrs, at Specified Age* for Cohorts Born in:
Age (years)
iaan
onm
Full Population
Non-Hispanic
Black
Full Population
Non-Hispanic
Black
5
14
21
4.3
5.4
10
23
35
7.0
9.6
* CBLI for ages older than 10 has not been estimated in this table as NHANES blood Pb estimates are not readily
available for the 11-19 year old age range for non-Hispanic Black demographic group, although they are for the
full population. Accordingly, estimates for ages older than 10 would not be of comparable basis.
1-14
-------
2 QUANTITATIVE ANALYSIS PLANNING FOR THE
PRIMARY STANDARD
In reviews of primary NAAQS, quantitative risk and exposure assessments (REAs)6 are
generally designed to assess human exposure and health risk for air quality conditions associated
with the existing standards and, as appropriate, for conditions associated with potential
alternative standards. The objective for such assessments is to provide quantitative estimates of
impacts that inform judgments on the public health significance of exposures likely to occur
under air quality conditions reflective of the current NAAQS and, as appropriate, any alternative
standards under consideration. Risk estimates are also used in support of other quantitative
analyses of the evidence. For example, in the 2008 Pb NAAQS review, quantitative risk
estimates for specific case studies were employed in evaluation of the evidence-based framework
developed to inform the Administrator's judgments on a level for the revised standard.7
Accordingly, the assessments are also intended to provide a basis for judgments as to the extent
of public health protection afforded by such standards.
In reviews of primary NAAQS, quantitative exposure and health risk assessments are
generally intended to inform consideration of key policy relevant questions (see section 3.1),
such as the following:
What are the nature and magnitude of exposures and health risks associated with air
quality conditions just meeting the current standard and, as appropriate, alternate air
quality conditions?
To what extent are the estimates of exposures and risks to at-risk populations associated
with air quality conditions just meeting the current standard and, as appropriate, alternate
air quality conditions reasonably judged important from a public health perspective?
In considering exposure and risk estimates and their interpretation in this context, an
accompanying consideration is:
What are the important limitations and associated uncertainties associated with the
risk/exposure estimates?
6 While the term REA has in the past several NAAQS reviews referred to assessments presented in a stand-alone
REA document, in this review, we are also using this term, or the phrase "REA analyses" to simply refer to the
analyses which we intend to present in appendices or as supplemental materials to the PA.
7 In the 2008 review, an evidence based framework focused on air-related, Pb-attributable IQ decrements was
developed that related relationships of ambient air Pb with children's blood Pb and of children's blood Pb with IQ
decrements. General consistencies found between estimates derived from this evidence-based framework and risk
estimates generated using the case-study based approach informed the Administrator's judgments, providing
general support for the Administrator's decisions on revisions for the new standard (IRP, vl, section 3.3; 73 FR
67006, November 12, 2008).
2-1
-------
In developing REAs in a NAAQS review, we draw upon the currently available health
effects evidence that is characterized in the ISA. This includes information on atmospheric
chemistry, air quality, human and environmental exposures, dosimetry and mode of action, and
information on health effects associated with exposures considered likely to occur because of
pollutant concentrations in ambient air. We additionally employ current methods and tools to
support the quantitative modeling and assessment.
The REAs commonly rely on a case study approach, which involves quantitative analyses
focused on populations and pollutant concentrations in one or more specific exposure situations
reflecting air quality conditions that just meet the existing standards (and alternatives as
appropriate), often through simulations for different geographic areas.8 Reliance on this approach
is intended to provide assessments of the air quality scenario(s) of interest for a set of case
studies and associated exposed at-risk populations and ecosystems that will be informative to the
EPA's consideration of potential exposures and risks that may be associated with the stated air
quality conditions. For example, we are interested in the exposure and risk associated with air
quality conditions that just meet the current standard(s); such information is useful in interpreting
the degree of protectiveness given by the current standard(s), the adequacy of such standard(s),
and the need to consider alternatives.
Further, the REA analyses employ a case study approach that addresses practical
considerations, such as employing a tractable scale and considering resource constraints, while
providing estimates for populations and/or geographic areas of interest and also having broader
applicability (e.g., offering risk perspective for similar study areas that were not assessed). Thus,
REA analyses are not generally intended to provide a comprehensive national assessment of such
conditions, nor are they necessarily intended to provide such an assessment of existing air
quality. Rather, the purpose is to assess population exposure and risk for particular air quality
conditions based on currently available scientific information, modeling tools, and other
technical information. As a result, the REA can provide extended perspective on potential
exposures and risks in geographic areas across the U.S. not analyzed, but with similarity in the
attributes that primarily influence exposures and risks, such as ambient air concentrations,
population demographics, and the degree of correlation in their spatial distributions.
In planning any REA analyses that may be appropriate for a new NAAQS review, we
first consider the analyses conducted in the last review and the extent to which they provided
8 Although case studies often focus only on specific geographic locations (e.g., populations in specific urban areas),
in past Pb NAAQS reviews, we have also included a generic case study intended to estimate exposure and risk for
a generalized group of residential children, not linked to a specific geographic location. Rather it is characterized
by exposure associated with residence near a significant source of Pb emissions such that air quality conditions
just meet the standard (see section 2.1, below, for additional detail; 81 FR 71923-71924, October 18, 2016).
2-2
-------
important insights that were informative to the Agency's decision on the current standard.
Conclusions in this regard are generally influenced by an assessment of the uncertainties
associated with each type of analysis and the corresponding consideration of each type's relative
strength, as documented in the notice of the decision for the prior review and associated
assessment documents such as the PA and REA. In considering whether new analyses are
warranted for particular types of assessments, we evaluate the availability of new scientific
evidence and technical information in this review, as well as improved methods and tools, that
may provide support for conducting updates to address key limitations or uncertainties in
analyses from the last review or to provide additional insight beyond those provided by the prior
REA. Thus, we focus on identifying the new analyses that are warranted in consideration of
factors such as those raised here, while also bearing in mind practical and logistical
considerations such as available resources and timeline for the review.
The types of analyses performed in NAAQS reviews generally reflect the nature and
strength of the evidence in various aspects. For example, for the health effects pertaining to
exposures associated with the presence of the pollutant in ambient air, the availability and type of
information from the health effects literature on relationships between internal dose, exposure, or
ambient air concentration and health response influences the types of exposure assessment and
risk characterization that are performed. The health risk assessments focus on exposure metrics
that are appropriate for effects of concern for the subject pollutant and, along with available
ambient air concentration measurements and model estimates, where appropriate, are used to
generate estimates of exposure. Consistent with the health risk approaches that have been used in
NAAQS reviews (illustrated in Figure 2-1), assessments of air-related health risks of Pb in past
reviews have been based on estimating internal concentrations of the biomarker, blood Pb, and
then employing a concentration-response (C-R) function-based approach to estimate risk.
2-3
-------
Figure 2-1. Summary of health risk assessment approaches that have been employed in
NAAQS reviews.
The purpose of the sections below is to briefly summarize the comprehensive, complex,
and resource-intensive quantitative health and welfare risk assessments completed in past
reviews of the Pb NAAQS, giving attention to those analyses concluded to be most informative
to the decisions reached on the standards. In considering the issues raised above, we additionally
summarize key uncertainties and limitations of the analyses conducted for the last review and
consider the extent to which newly available information, tools or methodologies might address
those areas. For example, the scope of any analyses for this review would be informed by the
new scientific information characterized in the upcoming ISA; recent air quality data; the
availability of improved data, methods, tools, and models that can be used to address limitations
and uncertainties from the last review; and any constraints on resources and the review timeline.
Consideration is also given to a review of the way risk information has been used to address
specific science-policy questions in previous NAAQS reviews and whether that application
identifies specific areas of uncertainty or limitations that could be addressed to enhance the
utility of the risk assessment in informing NAAQS reviews going forward.
We are planning that the quantitative exposure and risk analyses newly developed in this
review will be described in the PA, and we plan to consider these analyses along with any
previously conducted analyses that remain pertinent and informative to consideration of the
adequacy of the current standard (and alternative standards, as appropriate). We intend to
provide associated technical details for any new exposure and risk analyses in appendices or
supplemental materials for the PA, while analyses from the prior reviews are described in the
documents for those reviews. Any quantitative assessments newly developed in this review
would then be made available for public comment and reviewed by the CASAC in the context of
2-4
-------
the draft PA. Public comments and CASAC advice on such REA-related analyses in the draft PA
would be considered in finalizing analyses for presentation in the final PA.
In this chapter, we summarize the types of analyses performed in the last review and
highlight some considerations for analyses in this review. The exposure and risk analyses
performed in past Pb NAAQS reviews are described in section 2.1 below, along with a summary
of key uncertainties and limitations. The roles of the risk analyses in informing both the 2008
and 2016 NAAQS reviews are also summarized in section 2.1 with a focus on identifying key
uncertainties and limitations that informed interpretation of that risk information. In section 2.2,
we discuss key considerations relative to the refinement and updating of the 2007 REA for use in
the current review with emphasis on the availability of new information that can address key
limitations and uncertainties identified in section 2.1. Based on the information presented in
sections 2.1 and 2.2, we then present our preliminary plans for developing an REA to support the
current review in section 2.3.
2.1 ASSESSMENTS INFORMING THE LAST REVIEW
The exposure and risk information available for consideration in the 2016 review was
drawn primarily from the 2007 REA (80 FR 300-305, January 5, 2015).9 The 2007 risk
assessment focused on assessment of Pb-attributable IQ decrements in young children residing in
areas with air quality just meeting a set of potential alternative primary standards. The
assessment, with its focus on Pb derived from sources emitting Pb to ambient air, employed
several different case studies, as described below. As the air quality scenarios included in the
2007 assessment had not included the existing primary standard, the 2007 information was
enhanced for the 2016 review by the inclusion of a limited new computation for one case study
focused on risk associated with the existing standard (2014 PA, section 3.4 and Appendix 3A).
Key aspects of the exposure and risk assessment considered in the 2016 review are described
below.
The conceptual model that informed planning for the 2007 REA identified sources,
pathways, routes, exposed populations, and health endpoints. The multimedia and persistent
nature of Pb, as well as the existence of many non-air sources of Pb to the environment,
contributed multiple complexities to the exposure and risk assessment for ambient air related Pb.
9 The information newly available in the 2016 review, with regard to designing and implementing a full REA for
that review, was such that performance of a new REA for that review was not warranted, a decision with which
the CASAC Pb Review Panel generally concurred. In that review, after careful consideration of the available
information regarding designing and implementing a full REA, the EPA determined that the available information
did not provide the means by which to develop an updated or enhanced risk model that would substantially
improve the utility of risk estimates in informing the current Pb NAAQS review (2011 REA Planning Document,
section 2.3).
2-5
-------
Sources of human Pb exposure include current and historical air emissions sources, as well as
miscellaneous non-air sources, which can contribute to multiple exposure media and associated
pathways (e.g., inhalation of ambient air, ingestion of indoor dust, outdoor soil/dust and diet or
drinking water). As illustrated in Figure 2-2 below, in addition to airborne emissions (recent or
those in the past), sources of Pb to these pathways also include old leaded paint, and associated
Pb mobilized indoors during renovation/repair activities, and in soils. Further, Pb in diet and
drinking water may have air pathway-related contributions as well as contributions from non-air
sources (e.g., Pb solder on water distribution pipes and Pb in materials used in food processing).
Pb in ambient air
*¦
\
Deposition
outdoors
Infiltration
Indoor air
Deposition
indoors
Resu3pension
viResuspension
\
\
\
"Hitchhiking" in
Indoor dust
Inhalation of
ambient air Pb
(outdoors or
indoors)
Ingestion of
indoordustPb
<-i r> soi"c'
/nistoric\
Ingestion of outdoor
soil/dust Pb
Ingestion of Pb in diet/
drinking water
*Green ovals indicate nonair or historic sources. Sources unrelated to
environmental pathways (e.g., jewelry, mini-blinds) not shown.
Figure 2-2. Simplified presentation of air-related Pb exposure pathways.
Although our focus for the assessment was on the ambient-air aspects of Pb exposure that
are most relevant to the standard, limitations in the available data and modeling tools affected
characterization of the various complexities of ambient air Pb pathways and handi capped our
ability to separate the non-air contributions to Pb exposure from estimates of air related Pb
exposure and risk. As a result, the assessment included several simplifying assumptions in a
number of areas, which produced estimates of air-related Pb risk that are approximate and are
characterized by upper and lower bounds. The lower bound is based on a combination of
pathway-specific estimates that do not completely represent all air-related pathways, while the
upper bound is based on a combination of pathway-specific estimates that includes pathways that
are not air-related but the separating out of which is precluded by modeling and data limitations
(81 FR 71925, October 18, 2016).
To inform our understanding of air-related exposure and risk in different types of air Pb
exposure situations, Pb exposure and associated risk were estimated in the 2008 review for
2-6
-------
multiple case studies that generally represent two categories of residential population exposures
to air-related Pb: (1) location-specific urban populations of children with a broad range of air-
related exposures, reflecting existence of concentration gradients, and (2) children residing in
localized areas with air-related exposures representing air concentrations specifically reflecting
the standard level being evaluated (see Table 2-1). Accordingly, the two categories of case
studies differed with regard to the extent to which they represented population variability in air-
related Pb exposure. The case studies developed included two point source-focused case studies.
One was a small rural community near a secondary Pb smelter (not shown in Table 2-110), and
the second was a community near a primary Pb smelter. For the latter, two different sets of
estimates were derived, one for the nearest residents for whom dust-related exposures were most
significant and a second set of estimates for somewhat more distant residents (e.g., falling into
the first of the two categories), using tools and data particular to each community's exposure
circumstances (2007 REA, section 2.2). Also in the first category were three location-specific
urban case studies in the metropolitan areas of Cleveland, Chicago and Los Angeles.
Table 2-1. Summary of approaches used to estimate case study media concentrations in
2007 REA (based on 2014 PA, Table 3-7).
Type of Population Exposure
Case Study
Broad range of
air-related
exposures
Part of metropolitan area with
spatially varying air concentrations,
inclusive of location at standard or
conditions being evaluated
Multiple exposure
zones, larger
populations
Location-specific urban:
Cleveland, Chicago, Los
Angeles
As above, with dominant, historically
active metals industry as ambient air
Pb source
Primary Pb smelter
(full study area)
Generalized,
high end of air-
related exposure
Localized residential area with air
concentrations generally representing
the standard or conditions evaluated
Single exposure
zone without
enumerated
population
Generalized local
A few exposure
zones with small
population
Primary Pb smelter (subarea)
Two case studies, one from each of these two categories, were considered to be most
useful in the 2016 review: (1) the location-specific urban case study for Chicago and (2) the
generalized local case study. The generalized local case study (referred to as general urban case
study in the 2008 and 2016 reviews) was not based on a specific geographic location and
10 Due to uncertainties associated with modeling ambient air impacts of Pb emitted from the secondary Pb smelter
and potential to underestimate exposure and risk impacts to residential populations within the study area (see
2007 REA, section 4.3.1), this case study did not weigh heavily in past reviews.
2-7
-------
reflected several simplifying assumptions in representing exposure, including uniform ambient
air Pb levels associated with the standard of interest across the hypothetical study area and a
uniform study population. Based on the nature of the population exposures represented by the
two categories of case study (see Table 2-1), the generalized local case study includes
populations that are relatively more highly exposed by way of air pathways to air Pb
concentrations near the standard level evaluated, compared with the populations in the location-
specific urban case. The location-specific urban case studies provided representations of urban
populations with a broad range of air-related exposures due to spatial gradients in both ambient
air Pb levels and population density. For example, the highest air concentrations in these case
studies (i.e., those closest to the standard being assessed) were found in very small parts of the
study areas, while a large majority of the case study populations resided in areas with much
lower ambient air concentrations.
2.1.1 Summary of Design Aspects of the 2007 Assessment
The approach to assessing exposure and risk for the two categories of case studies
identified above (urban location-specific and generalized local) was comprised of four main
analytical steps (Figure 2-3): (1) Estimation of ambient air Pb concentrations, (2) estimation of
Pb concentrations in other key exposure media, including outdoor soil and indoor dust, (3) use of
exposure media Pb concentrations, with other pathway Pb intake rates (e.g., diet), to estimate
blood Pb in children using biokinetic modeling, and (4) use of C-R functions derived from
epidemiological studies to estimate IQ decrements associated with the blood Pb. Concentrations
of Pb were estimated in media and indoor dust using a combination of empirical data and
modeling projections. The use of empirical data brings with it uncertainty related to the potential
inclusion of non-air source signals in these measurements (e.g., house paint contributions to
indoor dust and outdoor soil Pb). Conversely, the use of modeling tools introduces other
uncertainties (e.g., model and parameter uncertainties).
2-8
-------
U5
U5
LU
U5
U5
o
D.
X
Characterizing ambient air levels and
depositiontosoil
Ambient monitoring data
(general near-source and location-specific urban
case studies)
Stack and fugitive
emissions data
Air dispersion modeling
(primary Pb smelter)
Ambient air ;
concentrations }~-
for study areas I
Characterizing soil anqinaooraust
concentrations
Combination of (a) statistical (regression)
and/or (b) mechanistic (compartmental)
indoor dust prediction models
(all case studies)
Nationally-representative residential
soil Pb value selected from literature
(generalnear-source and location-
specific urban case studies)
E
Outdoor residential soil
concentrations
Site-specific soil
monitoring data
combined with
statistical extrapolation
(primary Pb smelter)
Indoor residential dust
concentrations
Characterizing blood Pb levels
Background Pb exposure levels
diet
drinking water
indoor paint (actually reflected in dust
modeling)
Biokinetic blood Pb
modeling (IEUBK)
(all case studies)
Ambient air !
J
concentrations
(see above) j
Demographics (distribution of children
within study areas)
(primary Pb smelterand location-specific
urban case studies)
Probabilistic modeling of blood Pb
levels for children within each case
study location
(all case studies)
GSD reflecting inter-
individual variability in
behavior related to Pb
exposure and biokinetics
<
!1J
cd
LU
I
S
o
C£
Characterizing risk (IQ loss)
Distribution of IQ loss for study
populations (partitioned
between policy-relevant
exposures and policy-relevant
background)
Jl
Risk estimation - IQ
change estimation
(all case studies)
Concentration-response functions based
on log-normal functios (Lanphear pooled
analysis), or on data from that study
.Note,
the greyed-out boxes reference study areas that ultimately did not receive as much focus in the 2016 NAAQS review.
Figure 2-3. Analytical approach for two case study categories in 2016 review.
2-9
-------
Characterization of Pb in ambient air relied on (1) the use of ambient air monitor data for
the location-specific urban case studies and (2) an assumption of uniform ambient air Pb levels
(just meeting the standard being considered) for the generalized local case study. The ambient air
monitors within each location-specific urban case study were used to characterize spatial
gradients. By contrast, the generalized local case study is designed to assess exposure and risk
for a smaller group of residents (e.g., neighborhood) exposed at the standard, and, therefore,
ambient air Pb concentration was fixed at the standard being assessed. For the generalized local
case study, which has a single exposure zone in which air Pb concentrations do not vary
spatially, we derived a single air Pb concentration estimate to meet the standard assessed.
Concentrations in the location-specific urban study areas, which relied on empirical (monitor-
based) data to define ambient air Pb concentrations, reflected contributions from all sources
affecting the concentrations in those locations, potentially including currently active stationary or
mobile sources, resuspension of previously deposited Pb or other sources.
The air quality scenarios assessed in the 2008 review included air quality associated with
a set of potential alternative standards more restrictive than the then-existing standard under
review (1.5 |j,g/m3 as a not to be exceeded calendar quarter average): a maximum calendar
quarter average of 0.2 |j,g/m3 and maximum monthly averages of 0.5, 0.2, 0.05 and 0.02 |j,g/m3
(2014 PA, Table 3-8). However, in estimating concentrations of Pb in media that might be
affected by reductions in the standard, the full multimedia impact was not simulated due to
limitations in the available data and modeling tools that precluded simulation of linkages
between some media and air Pb. Specifically, as shown in Table 2-2 below, while Pb
concentrations in indoor dust were simulated to differ among the different air quality scenarios
for which there were differing ambient air Pb concentrations (outdoors and indoors), dietary and
drinking water Pb concentrations, as well as soil Pb concentrations, were not varied across the air
quality scenarios in any case study (2014 PA, Table 3-7).
2-10
-------
Table 2-2. Summary of approaches used to estimate case study media concentrations in
2007 REA (based on 2014 PA, Table 3-7).
Simulation of air
quality impacts
Media
category
Generalized local case study
Location-specific urban case study
Concentrations
for these media
were varied
across air quality
scenarios
(see 2014 PA,
section 3.4.3.2).
Ambient
air Pb
Constant ambient air Pb just
meeting standard across entire
study area (single exposure zone)
Source and non-source monitors define
concentration gradient (6 to 11 exposure
zones per case study)
Indoor
dust Pb
a) Hybrid model with a dynamic aspect relating ambient air Pb concentrations to
indoor dust Pb, and empirical aspect representing other contributions (e.g., paint,
historical air, Pb carried indoors with people) (2007 REA, section 3.1.4)
b) Ambient air-only regression model from literature (2007 REA, section 3.1.4)
Concentrations
were constant
across air quality
scenarios
(data/modeling
limitations)
Outdoor
soil Pb
National dataset (HUD, for houses constructed between 1940 and 1998; 2007 REA,
section 3.1.3).
Dietary Pb
National datasets for Pb residue data (US FDA Total Diet Study) and food
consumption data (NHANES) (2007 REA, Appendix H, Table H-6)
Drinking
water Pb
US and Canada datasets for residential water Pb concentrations and ingestion rates
(2007 REA, Appendix H, Table H-6)
In estimating blood Pb using the Integrated Exposure and Uptake Biokinetic (IEUBK)
model,11 Pb concentrations in exposure media (e.g., ambient air, diet, water, indoor dust) were
held constant throughout the 7-year simulation period, while behavioral and physiological
variables were changed with age of child (2007 REA, sections 3.2.1.1 and 5.2.4).12 For all case
studies, population variability in Pb intake and uptake was simulated by Monte Carlo-based
population sampling around a mean blood Pb from IEUBK modeling and a geometric standard
deviation (GSD) representing children's blood Pb variability (2014 PA, section 3.4; 2007 REA,
Appendix H).13 The risk characterization step generated a distribution of IQ decrement estimates
for the set of children simulated in the assessment.
Specifically, blood Pb estimates for the concurrent blood Pb metric were combined with
four C-R functions for IQ decrements derived from the analysis by Lanphear et al. (2005) of a
nThe IEUBK model simulates exposure of children to Pb from multiple sources and through various routes
including inhalation and ingestion. Model inputs include soil-Pb concentration, air-Pb concentration, dietary-Pb
intake including drinking water, Pb-dust ingestion, human activity, and biokinetic factors (2013 ISA, p. 1-11).
12 Detail on methods used to characterize media Pb concentrations and all IEUBK inputs for each case study are in
the 2007 REA, sections 3.1, 3.2, 5.2.3 and 5.2.4, and Appendices C throughH.
13 This GSD reflects a number of factors which operate together to produce interindividual variability in blood Pb
levels, including: (a) biokinetic variability (differences in the uptake, distribution or clearance of Pb), (b)
differences in behavior related to Pb exposure (e.g., varying hand-to-mouth activity, tap water ingestion rates, and
time spent playing indoors) and (c) differences in environmental Pb exposure concentrations (e.g., spatial
gradients in ambient Pb levels of a resolution beyond that simulated in each case study, differences in
cleaning/vacuuming rates and air exchange rates) (see 2007 REA, section 3.2.3).
2-11
-------
pooled international dataset of blood Pb and IQ (see the 2007 REA, section 5.3.1.1).14 We used
the four different C-R functions to provide different characterizations of the blood Pb-IQ
decrements relationship at low exposures in recognition of uncertainty related to modeling this
endpoint, particularly at lower Pb levels for which there is limited or no representation in the
Lanphear et al. (2005) pooled dataset. Consideration of the risk estimates focused on estimates
for one of the four functions (referred to as the loglinear with low-exposure linearization C-R
function [2014 PA, section 3.4.3.3; 2007 REA, section 4.2.1]).15 The range of risk estimates
reflecting all four C-R functions provides perspective on the impact of uncertainty in this key
modeling step. Among the estimates based on the loglinear function, the focus in the 2016
review was on the median IQ decrement estimates (as in the 2008 review), due to increased
confidence in these estimates relative to the higher percentile estimates, for which significant
uncertainty is recognized (2014 PA, sections 3.4.5, 2.4.6 and 3.4.7; 2007 Staff Paper, p. 4-20).
As the 2007 REA did not include an air quality scenario simulated to just meet the
standard that was established by the 2008 decision, two different approaches were employed in
the 2016 review to estimate risk pertaining to conditions just meeting the standard set in 2008.
First, given the similarity of the then-current condition scenario for the Chicago case study
(among all the 2007 REA scenarios) to the current standard (set in 2008), the risk estimates for
that scenario were considered to be informative with regard to risk associated with the current
standard. To augment the risk information available in the 2016 review and in recognition of the
variation among specific locations and urban areas with regard to air quality patterns and
exposed population, estimates for an air quality scenario just meeting the current Pb NAAQS
were newly developed in the 2016 review in the context of the generalized local case study.
These estimates were derived based on interpolation from the risk estimates available for
scenarios previously assessed for the generalized local case study. Such interpolated estimates
were only developed for the generalized local case study due to its use of a single exposure zone
which greatly simplified the method employed.16
14 The four C-R functions each involved different approaches to estimating IQ decrement at and below the lowest
blood Pb measurements analyzed in Lanphear et al.(2005). Entry K in Table 2-3 provides more detail.
15 Of the four functions, this function is nonlinear in shape, overall, reflecting the differing response at different
blood Pb concentrations consistent with Lanphear et al. (2005); is based on a function fit to the entire polled
dataset, and for which sensitivity analyses indicate model coefficients to be robust (Lanphear et al., 2005); and
provides an approach for estimating IQ decrements at the lowest exposures simulated, which for some is below
those studied (2007 REA).
16 Interpolation of risk estimates to estimate risk for the current standard was not done for other case studies because
those case studies utilized a more complex, spatially-differentiated and population-based approach. Application of
the simple linear interpolation approach for those case studies would have introduced substantial additional
uncertainty (relative to the other estimates for the same case study). The simplicity of the generalized local study
area, however, with its single exposure zone, is amenable to the linear interpolation of risk described here.
2-12
-------
In the 2016 review, the approach employed to develop estimates for the existing standard
(established in 2008) in the generalized local case study was to identify two potential alternative
standard scenarios that had been simulated in the 2007 REA for air quality conditions bracketing
those for the existing standard and then linearly interpolate an estimate of risk for the current
standard based on the slope created from the two bracketing estimates (2014 PA, section
3.4.3.3.2 and Appendix 3A). By this method, the air quality scenario for the existing standard
(0.15 ng/m3, as a not-to-be-exceeded 3-month average) was found to be bracketed by the
scenarios for the potential alternative standards of 0.20 |j,g/m3 (maximum calendar quarter
average) and 0.20 |j,g/m3 (maximum monthly average). Using interpolation between the median
risk estimates for these two scenarios, median risk estimates were developed for the existing
standard (2014 PA, Appendix 3A).
2.1.2 Characterization of Variability
With regard to the exposure/risk information, several important sources of variability in
air-related Pb exposures and associated risk were recognized; the approaches by which they were
addressed in the 2007 REA are summarized here (2014 PA, section 3.4.6; 80 FR, 303-304,
January 5, 2015).
Variation in distributions of potential residential exposure and risk across U.S. urban
residential areas is addressed by the inclusion of location-specific urban study areas
that reflect a diverse set of urban areas in the U.S.
Representation of a more highly exposed subset of urban residents potentially exposed
at the level of the standard is addressed by the inclusion of the generalized local study
area.
Variation in residential exposure to ambient air Pb within an urban area of the
location-specific case studies is addressed through the partitioning of these study areas
into exposure zones to provide some representation of spatial gradients in ambient air
Pb and their interaction with population distribution and demographics.
Inter-individual variability in blood Pb levels is addressed through the use of
empirically derived GSDs to develop blood Pb distribution for the child population in
each exposure zone, with GSDs selected particular to each case study population.
Inter-individual variability in IQ response to blood Pb is addressed through the use of
C-R functions for Pb-associated IQ decrement based on a pooled analysis reflecting
studies of diverse populations.
2.1.3 Key Limitations and Uncertainties
In characterizing risk associated with Pb from air-related exposure pathways in the 2008
review, we faced a variety of challenges and employed a number of methods in order to develop
characterizations. The challenges related to significant data and modeling limitations which
affected our ability to parse out the portion of total (all-pathway) blood Pb and IQ decrement
2-13
-------
attributable to air-related pathways, as well as our representation of key sources of variability
(summarized in section 2.1.2 above) and characterization of uncertainty.
With regard to parsing out the air-related blood Pb and risk estimates, we recognized that
Pb in diet and drinking water sources may in part be derived from Pb in ambient air, as well as
Pb from nonair sources, but limitations precluded explicit modeling of the contribution from air-
related aspects of these exposure pathways, such that the air-related component of these
exposures was not estimated. Although we separated total estimates into risk estimates for
diet/drinking water and two air-related categories ("recent air" and "past air"), significant
limitations in our modeling tools and data resulted in an inability to parse risk estimates specific
to the air-related pathways. Thus, we focused on estimates from the two air-related categories
("recent" and "past"), which we considered to under- and over-estimate air-related risk,
respectively, to create bounds within which we consider air-related risk to fall.
The "recent" air-related category included Pb exposure pathways tied most directly to
ambient air, which consequently have the potential to respond relatively more quickly to changes
in air Pb (i.e., inhalation and ingestion of indoor dust Pb derived from the infiltration of ambient
air Pb indoors). Importantly, media concentrations associated with the pathways in this category
were simulated to change in response to air concentrations (see section 2.1.1 and Table 2-2
above; 2014 PA, section 3.4.3.1; 80 FR 300-305, January 5, 2015). The air-related Pb exposure
pathways in the "past air" category, all of which are associated with atmospheric deposition,
included ingestion of Pb in outdoor dust/soil and ingestion of the portion of Pb in indoor dust
that after deposition from ambient air outdoors is carried indoors with humans. While there is
some potential for the "past" air-related category of exposures to be affected (over some time
frame) by changes in ambient air Pb concentrations (e.g., associated with an adjustment to the Pb
standard), limitations in our data and tools precluded simulation of that relationship.
Consequently, risk estimated for this category reflects media measurements available for the
2007 REA and is identical for all air quality scenarios. Further, although paint is not an air-
related source of Pb exposure, it may be reflected somewhat in estimates developed for the "past
air" category, due to modeling (and data) constraints (2007 Staff Paper, section 4.2.4). Thus, as
exposures included in the "recent" air-related category do not completely capture all air-related
pathways, we consider risk for this category to be an underestimate of air-related risk. Yet, as
exposures included in the "past" air-related category include pathways that are not air-related, we
consider the summed risk across both categories to include an overestimate of air-related risk.
In summary, because of limitations in the assessment design, data, and modeling tools,
the estimates of risk attributable to air-related exposure pathways are considered to be
approximate and to be bounded on the low end by the risk estimated for the "recent air" category
and on the upper end by the risk estimated for the "recent air" plus "past air" categories. With
2-14
-------
regard to the latter, modeling and data limitations reduce the extent to which the upper end of
these bounds reflects impacts of the alternative air quality conditions simulated. This limitation
will tend to contribute to estimates for the "past air" category representing relatively greater
overestimates for relatively lower air Pb air quality scenarios (80 FR 303, January 5, 2015). In
considering these risk estimates in the 2016 review, the EPA concluded that the "resultant,
approximate, air-related risk bounds, however, encompass estimates drawn from the air-related
IQ loss evidence-based framework, providing a rough consistency and general support, as was
the case in the last review (73 FR 67004, November 12, 2008)" (81 FR 71925, October 18,
2016). The Administrator's decision in 2016, that it was appropriate to retain the existing
standard, was based on consideration of the evidence, with support from the exposure/risk
information, "recognizing the uncertainties attendant with both," in addition to public comment
and supporting advice from the CAS AC (81 FR 71935, October 18, 2016).
In both the 2008 and 2016 reviews, a range of additional uncertainties, limitations, and
assumptions related to the 2007 REA modeling was described (see 2007 REA section Appendix
M, Exhibit M-l and 2011 REA Planning Document, Table 2-3). These sources of uncertainty
occur in the different modeling steps, as briefly summarized here, with some additional detail
provided in Table 2-3 below.
Characterizing ambient air Pb levels. Unlike the other sources of uncertainty
discussed in the bullets below that apply to both the generalized local and location-
specific case studies, uncertainty related to air quality characterization differs
between these two case studies. Uncertainty related to the characterization of ambient
air Pb levels is contributed by several elements of this modeling step including:
a) characterization of spatial gradients in ambient air Pb levels in the location-
specific urban context using monitoring data [location-specific element];
b) characterization of single exposure zone without spatial gradients in ambient
air Pb levels in the general near-source case study) [generalized local
element];
c) conversion of air quality metric associated with the standard and potential
alternatives of interest to an equivalent annual average Pb concentration for the
blood Pb modeling time step (e.g., maximum quarterly or monthly averages
were converted to annual average concentrations utilizing ratios reflecting
relationships in ambient air monitoring data) [elementfor both]; and
d) simulation of air quality patterns differing from what were current in the
location specific case studies (using a proportional rollback across monitor
sites, a pattern that may differ from what would occur based on actual
implementation strategies) [location-specific element].
Characterizing Pb concentrations in indoor residential dust. Given the potential
importance of the indoor residential dust pathway to ambient air Pb-related exposure,
it is addressed separately from the other media. The hybrid (mechanistic-empirical)
2-15
-------
indoor dust Pb model combines mechanistic (compartmental) modeling elements with
empirical (survey-statistical) data to simulate the relationship between outdoor
ambient air Pb and Pb in indoor dust (with consideration for non-air contributions
such as the tracking in of outdoor soil Pb, which may have derived from nonair
sources, and paint Pb). A number of aspects of the hybrid model are potential sources
of uncertainty (e.g., cleaning frequency and efficiency, Pb deposition and air exchange
rates, treatment of Pb resuspension indoors, handling of non-air related indoor dust Pb
fraction, empirical regression equations used to convert loadings to concentrations).
Characterizing Pb exposure from the multiple pathways. The approaches for
representing Pb exposures from sources not identified above were also subject to
various sources of uncertainty, including:
(a) estimates (and their supporting data) used for exposure related to drinking
water ingestion, dietary Pb, and incidental ingestion of indoor dust and outdoor
soil;
(b) the approach for deriving inhalation-exposure concentrations of ambient air
Pb;and
(c) the use of constant, unchanging Pb concentrations in all exposure media for
the 7-year exposure modeling period (e.g., assumption of same residence and
unvarying annual average media concentrations).
Estimating blood Pb levels for 7-year old children (resulting from seven years of
exposure since birth). The IEUBK model was used to estimate mean blood Pb levels
(with which an empirically derived GSD was used to derive population distribution of
blood Pb levels). The IEUBK model application was subject to a number of sources of
uncertainty including specification of absorption factors for dietary and drinking water
Pb and for indoor dust Pb. There was also uncertainty related to the limitations in the
extent to which IEUBK performance for the REA exposures (and the specific mix of
pathways) had been evaluated and assessed. Uncertainty may also be contributed by
the GSD used to characterize interindividual variability in blood Pb to the extent it is
not representative of the population being modeled in the REA.
Estimating risk (IQ decrement) for young children: There are a number of sources of
uncertainty associated with specification and application of the C-R functions derived
from the Lanphear et al. (2005) pooled analysis for the scenarios modeled. These
include: (a) the potential for covariates on the Pb effect to not have been fully
accounted for in the model; (b) potential differences between the Lanphear et al.
(2005) pooled study population and the populations in the REA case studies; and (c)
application of the C-R functions at levels at or below the lowest blood Pb levels in
Lanphear et al. (2005).
2.2 KEY CONSIDERATIONS
Our planning for an REA in this review involves a number of considerations related to
the design and application of the 2007 REA in supporting both the 2008 and 2016 Pb NAAQS
reviews. We begin in section 2.2.1 by considering sources of uncertainty impacting the 2007
2-16
-------
REA, noting those areas where newer information or data could result in refinements to the REA
to increase overall confidence in the analyses. Next in section 2.2.2, we consider the way in
which the REA was used in informing the 2008 and 2016 NAAQS reviews, including
implications for the design of the REA going forward. And finally, in section 2.2.3, we consider
the extent to which more recently available data and information support the assessment of
health effects associated with Pb, other than IQ decrements, and/or assessment of exposure/risk
to other age-groups and lifestages.
As discussed in section 2.1, there are a variety of complexities associated with the
assessment of air-related Pb exposure and health risk. In designing the REA to support the
current review, as was done in the 2008 review, we have attempted to focus effort on those
aspects that are most important and feasible to address within our scope and given the constraints
of time, pertinent data, models, etc. The conceptual model for current REA planning is presented
in Figure 2-4, with shaded boxes indicating items to include in modeling for the REA for which
ambient air has played a role (note, sources in the hatched boxes, while not directly linked to
ambient air Pb, would also be reflected in the modeling of exposure/risk). The conceptual model
presented in Figure 2-4 is intended to facilitate consideration of design aspects important to
planning for an REA, including in particular the discussion of uncertainties identified in the 2007
analysis and the degree to which more recent data and information allow us to address those
uncertainties, thereby supporting the design of an updated and refined REA.
2-17
-------
Figure 2-4. Conceptual model for 2007 Pb human health risk assessment.
2-18
-------
2.2.1 Newly Available Information Regarding Key Limitations or Uncertainties of 2007
REA
An important factor in considering an update to the exposure/risk assessment is the
degree to which new information available since the 2007 REA allows us to address key sources
of uncertainty, thereby resulting in an updated REA model with greater overall confidence. This
question was explored in the previous Pb NAAQS review completed in 2016 with the conclusion
being, at that time, that newly available information would not substantially improve the utility
of risk estimates in informing the current Pb NAAQS review (see 2011 EPA planning document,
section 2.3). Section 2.1.3 above identifies key limitations and uncertainties associated with the
exposure/risk estimates in the 2008 and 2016 reviews. In our consideration here of the currently
available information, while we focus on the period since the last Pb NAAQS review, we also
consider the full array of information available since development of the original REA in 2007.
We have performed a detailed assessment of the newly available information in the
context of its potential for addressing sources of uncertainty in the exposure and risk estimates.
That assessment is summarized in Table 2-3. The model design elements for which there is
newly available information that could contribute to more up-to-date estimates,, are described
below and it is these technical areas that would be updated or refined in the context of new
exposure and risk analyses as part of the current review.
Converting the air quality metric for the current standard to an annual-average
concentration reflecting recent patterns in daily air Pb monitoring data for use in
modeling exposure and risk: While information published since the 2007 REA does
not allow us to reduce uncertainty associated with this conversion step, more recent
monitoring data are available that can be used to generate more updated ratios for use
in these conversions.
Modeling relationship between outdoor ambient air Pb and indoor residential dust Pb
using hybrid (mechanistic-empirical) model. The hybrid indoor dust Pb model is a
critical element of the 2007 REA and allowed the impact of recent ambient air Pb on
indoor dust Pb to be evaluated through the associated fate and transport pathway (the
loading of indoor residential dust Pb by outdoor ambient air Pb). Newer information
may allow important components of the hybrid model to be updated, including the
approach used to convert dust Pb loadings to concentrations as well as the air
exchange rates (AERs) between outdoor air and indoor air. The availability of AER
data that may be specific to different cities or regions of the country and for specific
types of housing (e.g., those with and without air conditioning) may allow
customization of the model for different regions of the country and different types of
housing. This could increase overall confidence in the model by enhancing the
representativeness of the model for locations likely to experience disparities in indoor
dust Pb loading per unit of outdoor ambient air Pb.
Characterizing exposure to Pb through drinking water and dietary ingestion and
incidental outdoor soil ingestion: As noted earlier, assessment of air-related Pb
2-19
-------
impacts requires consideration for the multi-pathway nature of Pb exposure in order to
estimate total Pb exposure. In the development of an updated assessment, we would
anticipate updating the characterization of Pb in media with data published since the
2007 REA. This will include consideration of estimates for urban/residential soil and
diet and drinking water based on recent evidence described in sections 2.1.3.2 and
2.1.3.3, respectively of the draft ISA. In addition, we will consider the currently
available data on background Pb concentrations in drinking water and dietary sources
as well as updated exposure factors utilized in IEUBK 2.0 (U.S. EPA, 2021a).
Modeling blood Pb levels in young children given pathway-specific intake estimates
utilizing IEUBK: Presuming use of an IEUBK-based approach to modeling exposure
for children from birth up to age seven years, as in the 2007 assessment, we note that,
while IEUBK has not been significantly refined since the 2007 REA, some of the
exposure factors (intake rates) have been updated. In addition, performance
evaluations have been completed for IEUBK by the EPA since the 2007 REA that
have focused on the ability of the model to evaluate a target 95th percentile blood Pb
level of 5 |ig/dL (draft ISA, Appendix 2, section 2.6). Findings from these evaluations
may inform characterization of the range of exposures over which the IEUBK model
can be utilized without appreciable extrapolation and therefore the level of exposure
below which there is increased uncertainty in modeling exposure/risk. This type of
performance-based analysis of confidence represents an improvement over the
uncertainty characterization for modeled exposure and risk in the 2007 REA, and,
thus, could be an important refinement for an updated analysis.
Characterizing variability in blood Pb levels using empirically derived GSDs:
Presuming use of IEUBK to estimate a central-tendency blood Pb value for a study
population coupled with an empirically derived GSD to characterize variability in
blood Pb levels around that central tendency value, we recognize the availability of
additional blood Pb datasets that may provide for updating the GSD. For example,
blood Pb data from more recent studies for child populations of similar age to those
modeled in the REA may provide for estimation of variability reflecting generally
lower blood Pb levels common in the more recent years. We would anticipate that
GSDs for a smaller population of children in the residential context may have changed
from those utilized in the 2007 REA given the declining trends in overall blood Pb
levels that include smaller subsets of children continuing to experience relatively
higher blood Pb levels (section 1.3.2 above; draft ISA, Appendix 2, section 2.4.1).
Modeling IQ decrement in young children: Given the backdrop of declining blood Pb
levels in young children in the U.S., we anticipate that the IQ risk modeling would
focus on a lower range of blood Pb levels than those in the 2007 REA (i.e., in the
range of 5 |ig/dL and below).17 While there have been a number of studies published
since the 2007 REA assessing associations of children's IQ with children's blood Pb
17 Although the 2007 REA did not estimate blood Pb for an air quality scenario just meeting the standard that was
adopted in 2008, it estimated blood Pb for a somewhat similar scenario. In the generalized local (then called
general urban) case study, with its simplifying assumptions, the estimated blood Pb levels for the scenario based
on a mean maximum quarterly average of 0.14 |ig/m3 were 2 ng/dL (median total blood Pb) and 4 to 7 ng/dL
(total blood Pb at the 95th percentile) (see 2007 REA, section 3.4, Tables 3-7). Thus, a focus on risk associated
with blood Pb levels at and below 5 ng/dL is a reasonable expectation for this review.
2-20
-------
(e.g., Haynes et al., 2015, Martin et al., 2021), for reasons summarized in Table 2-3,
none of these studies provides support for C-R functions for this REA application.
However, given the current lower range of blood Pb levels, consideration of a
different approach from that used in the 2007 REA is appropriate. Rather than a focus
on the array of C-R extrapolations (to the blood Pb levels of interest) from the
nonlinear models presented in Lanphear et al. (2005), with corrections of Lanphear at
al. (2019), we are considering an approach focused on linear models for studies
involving narrower and lower blood Pb distributions, including the lower blood Pb
subset of studies included in the Lanphear et al. (2005) pooled dataset. For example,
the studies of child populations in Rochester, NY, and Boston, MA, contributed the
majority of lower blood Pb observations in the pooled dataset. We would look to
analyses of the association of IQ with blood Pb for these groups of children (e.g.,
Lanphear et al., 2005 with corrections of Lanphear at al. 2019, Figure 2; Canfield et
al., 2003; Bellinger, 2003) to identify C-R functions that may be appropriate for
assessing risk of IQ decrements at these lower blood Pb levels. In Lanphear et al.
(2005), non-linear functional forms were utilized to capture the steeper unit risk
associated with the lower blood Pb levels in the pooled dataset (e.g., below 7-10
|ig/dL) while also describing the less steep relationship across the higher Pb levels in
the dataset (e.g., 95th percentile of 35.7 |ig/dL, concurrent). For our focus on
characterizing risk for children with Pb levels only within the lower end of the range
(at/below approximately 5 |ig/dL), the nonlinear form is not needed and there would
be lower uncertainty with risk estimates based on the linear C-R functions from
analyses for that blood Pb range.
2-21
-------
Table 2-3. Assessment of information (including methods and models) newly available in this review related to quantitative
assessment of exposure and risk of IQ decrements in children.
Modeling Dimension
1 iiccrlaiiil> Kelaled (o Application
in 2011"7 RI.A
in Informing I pdaleol' RI.A Model
Characterizing Ambient Air Pb Levels (current conditions, current standard, and alternative standards)
A) Characterizing spatial gradients in ambient
air Pb levels in the urban context. For the
location-specific urban case studies (including
study areas based on Chicago, Cleveland and Los
Angeles), spatial variation in ambient air Pb levels
was characterized using ambient air Pb data for all
monitors within each study area, which were
defined as the portion of each city within one mile
of an existing urban monitor. U.S. Census blocks
were used as the basic unit of analysis with each
block being assigned the source-oriented monitor
closest to it (if that monitor was within a mile of the
block centroid) or the nearest non-source oriented
monitor. Interpolation of monitor data in this
fashion allowed a degree of spatial gradient in
monitored ambient air Pb levels to be reflected in
risk modeling (including the intersection of that
ambient air Pb gradient with demographics
characterized at the block-level).
Conversely, for the generalized local case study, a
simpler approach was used. Here, it was assumed
that a relatively smaller group of children (e.g., a
smaller neighborhood or residential area) existed in
a very small area within which there was a uniform
ambient air Pb level. This approach avoided the
need for any interpolation or direct use of
monitoring data.
Limitations in our monitoring network
and in studies of smaller scale spatial
gradients in ambient air Pb levels in the
urban setting introduced uncertainty
into our characterization of exposure
levels and risk for residential
populations modeled for the location-
specific urban study areas.
Similarly, limitations in our
characterization of ambient air Pb
levels more generally prevented us
from identifying areas across the U.S.
with the attributes associated with the
generalized local study area which
could have their exposure/risk
represented by the generalized local
case study scenario.
In the 2007 REA both of these sources
of uncertainty were assessed to have a
particularly significant impact on the
risk estimates generated (see 2007
REA Appendixes, Appendix M,
Exhibit M-l).
Evidence from studies published since the 2007 REA
suggests that there continues to be substantial intra-urban
variability in ambient air Pb levels with elevated levels being
associated with proximity to specific sources (and often
captured by source-oriented monitors), while lower levels
are often associated with more generalized urban areas.
However, the monitoring network in place is generally not
refined enough to provide comprehensive coverage for an
urban area such that the nature of the spatial gradient can be
well-characterized on a more localized (neighborhood) level
(draft ISA, section IS2.6; 2013 ISA, section 2.5.1.2). Lead
emissions and concentrations in the U.S. continue to steadily
decline with major industrial sources having either reduced
their emissions or closed, resulting in the emergence of
aviation gas as the dominant contemporary source (draft
ISA, Appendix 1, section 1.6).
This information suggests that the currently available
information does not provide the detailed information on
gradients that would be needed to reduce uncertainty
associated with this modeling element.
2-22
-------
Modeling Dimension
1 iiccrlaiiil> Kelaled lo Application
in 2011"7 RI.A
in Informing I pdalcof RI.A Model
B) Simulating air quality scenarios of interest
(e.g., meeting existing and, as appropriate,
potential alternate standards).
For the location-specific urban case studies, the
simulation of air quality scenarios of interest was
achieved using a proportional adjustment approach,
i.e., each monitor value for the exposure metric
under consideration was adjusted by the same
proportion needed for concentrations at the design
monitor to just meet the existing standard or
potential alternate of interest (in terms of the
standard averaging time, level and form) (2007
REA Appendices, Appendices C and 0).
For the generalized local case study, the constant
air concentration was set equal to the existing
standard or potential alternate of interest (in terms
of the standard averaging time, level and form).
There is uncertainty associated with the
use of proportional rollback to simulate
meeting the air quality scenarios of
interest since measures taken to meet
specific standards could result in
ambient air Pb reductions that diverge
from proportionality across the study
area.
This element remains a source of uncertainty for location-
specific case studies. The simplifying assumption of a lack
of gradients in the generalized local case study contributes a
different type of uncertainty (as noted earlier).
C) Converting air quality metrics (associated
with the standard of interest) to an equivalent
annual-average to use in modeling exposure and
risk.
Modeling of exposure and risk in the 2007 REA
utilized annual-average Pb levels and consequently
air metrics associated with standards of interest
(e.g., max quarterly, max monthly) had to be
converted to the annual averages associated with
the air quality scenarios for the various metrics.
This was done with ratios based on ambient air
monitoring data. These ratios were the mean or 95th
percentile ratio of the metric to the annual average
for the same location and time period for monitors
from urban areas with at least a million people, in
order to reflect ambient air Pb trends in larger urban
locations.
There is uncertainty introduced into the
analysis by these ratios (2007 REA
Appendices, Appendix O, Exhibit M-
1).
While information published since the 2007 REA does not
allow us to reduce uncertainty associated with the generation
or application of these ratios, the fact that ambient air Pb
concentrations have changed substantially since that time
suggests that at the very least, these ratios should be updated
to reflect more recent ambient air Pb concentrations across
urban areas.
2-23
-------
Modeling Dimension
I iiccrlaiiil> Kelaled (o Application
in 2011"7 RI.A
in Informing I pdaleol' RI.A Model
Characterizing Pb concentrations in indoor residential dust (given the importance of this exposure pathway in the context of ambient air Pb-related exposure
it is addressed here separately from the other media)
D) Modeling relationship between outdoor
ambient air Pb and indoor residential dust Pb
using hybrid (mechanistic-empirical) model.
The 2007 REA utilized the hybrid indoor dust Pb
model which combined (a) an empirical
(compartmental) modeling to relate outdoor
ambient air Pb to the loading of indoor dust Pb and
(b) empirical data obtained from the U.S. Housing
and Urban Development [HUD] National Survey of
Lead-based Paint in Housing (U.S. EPA, 1995) to
characterize the other contributions to indoor dust
Pb (e.g., tracking of outdoor soil/dust indoors,
indoor paint flaking, etc.). Empirical (regression)
relationships were then used to convert estimates of
indoor dust Pb loading to concentrations required
for IEUBK blood Pb modeling (see 2007 REA,
section 3.1.4.1 for additional detail on the hybrid
indoor dust Pb model).
A number of elements of the hybrid
indoor dust Pb model and its
application were identified in the 2007
REA as having particularly significant
impacts on risk estimates including: (a)
estimates of cleaning frequency and
efficiency, (b) potential variability in
the Pb deposition and air exchange
rates, (c) failure to consider Pb
resuspension indoors (which could bias
the indoor dust Pb levels high), (d)
characterization of non-air related
indoor dust Pb fraction using HUD
data, and (e) empirical (log-log
regression) equations used to convert
loadings to concentrations (2007 REA
Appendixes, Appendix M, Exhibit M-
1).
A number of new studies are available since the 2007 REA,
with data potentially addressing uncertainties related to the
hybrid indoor dust Pb model. These include: (a) Hunt et al.,
(2008) which examined the efficiency of vacuuming in
removing Pb dust from household surfaces and reported
updated cleaning efficiency estimates for vacuuming hard
floors (lack of information for other types of indoor surfaces,
including carpets, may limit its potential to reduce overall
uncertainty) and (b) Bevington et al. (2021) which addressed
the conversion of indoor dust Pb (wipe) loadings to
concentrations utilizing a large pooled dataset and
generating 17 regression models for converting loadings to
concentrations.
In addition, data now available on air exchange rates may
support an update to that aspect of the model. These data
(Cohen and Rosenbaum, 2012) have been most recently used
by EPA to update inputs to the Air Pollutant Exposure
(APEX) model for use in the exposure assessment in the
2020 ozone NAAQS review (U.S. EPA, 2020, Appendix 3D,
section 3D.2.6.1.1).
The availability of these new studies (particularly Bevington
et al., 2021) indicates the potential for updates that may
contribute to reduced uncertainty in estimates of dust Pb
concentration from that in the 2007 REA.
E) Modeling relationship between outdoor
ambient air Pb and indoor residential dust Pb
using an air-only regression model.
In addition to the hybrid indoor dust Pb model
described above, in the 2007 REA, indoor dust Pb
was also estimated using an air-only regression
model obtained from the literature (U.S. EPA,
2007 REA Appendixes, Appendix M,
Exhibit M-l addresses uncertainty
associated with the regression-based
(air-only) model noting that several of
the locations included in the data used
to generate the model were in urban
environments, but the data were
No new studies characterizing the empirical (statistical)
relationship between ambient air Pb and indoor residential
dust Pb were identified since the 2007 REA. EPA notes that
section 1.3.4 of Appendix 1 of the draft Pb ISA focuses on
the important issue of resuspension of legacy Pb in urban
soil as a source of ambient air Pb but does not discuss any
research addressing the empirical relationship between
2-24
-------
Modeling Dimension
1 ncerlainl> Kelaled lo Application
in 2011"7 RI.A
in Informing I pdaleol' RI.A Model
1989). That model estimates indoor dust Pb based
on (a) outdoor ambient air Pb (multiplied by an air-
related factor) and (b) an intercept which captures
other impacts besides air (e.g., indoor paint). The
air factor used in this equation is expected to
capture longer-term impacts of outdoor air Pb on
indoor dust, including the indirect effect of air Pb
on outdoor soil/dust Pb with subsequent impacts of
that outdoor soil/dust Pb on indoor dust Pb through
other mechanisms (U.S. EPA, 1989).
dominated by point sources. Thus, this
equation's application in urban
environments is limited by the
representativeness of the locations
included in the original pooled analysis
and the extent to which current
conditions are represented by
conditions in the 1980s when the data
were collected (noting in particular that
ambient air Pb levels have decreased
substantially as have direct emissions
of Pb into the ambient air). It is unclear
how these uncertainties may bias the
estimated indoor dust Pb
concentrations.
ambient air Pb (whether new sourced or entrained) and the
buildup of Pb in indoor dust.
Modeling multi-pathway Pb exposure (estimation of the indoor dust Pb covered separately - see above)
F) Characterizing exposure to Pb through
drinking water and dietary ingestion and
incidental outdoor soil ingestion. Note, indoor
dust and inhalation of outdoor/indoor air addressed
separately - see D and E for the former and G for
the latter.
Exposure estimates for these pathways used as
IEUBK inputs were estimates for the central
tendency of the U.S. population.
These inputs may contribute
uncertainty to the blood Pb estimates
associated with a number of factors:
(a) the potential for specific urban areas
to have media concentrations that
diverge from the central-tendency
values used in the analysis, (b) the
possibility that the exposure factors
utilized may not reflect the central
tendency for all children including,
particularly, disadvantaged children,
(c) regarding drinking water estimate,
the potential for the estimate to under-
represent Pb concentrations in drinking
water from older housing that could be
more impacted by Pb piping (2007
REA Appendixes, Appendix M,
Exhibit M-l).
Newer study data characterizing Pb concentrations in
relevant media have been published since the 2007 REA.
For example, for newer estimates for urban/residential soil
(draft ISA, Appendix 2, section 2.1.3.2), and newer data for
dietary items including drinking water (draft ISA, Appendix
2, section 2.1.3.3). These more recent data across both Pb
media concentrations and exposure factors would be
consulted to update the method for modeling multi-pathway
Pb exposure.
2-25
-------
Modeling Dimension
I iicerlaiiil> Kelaled lo Application
in 2011"7 RI.A
in Informing I pdalcof RI.A Model
(G) Characterizing Pb exposure concentrations.
The estimation of the Pb exposure concentrations
for children that reflect time spent indoors versus
outdoors was completed using location-specific
adjustment factors also used in the exposure
modeling component of EPA's 1999 national-scale
air toxics assessment (U.S. EPA 2006c) (see 2007
REA section 3.1.2).
The adjustment factors used in the
2007 REA for exposures from birth
through age 7, which modeled
exposure from birth through age 7,
were those used in the exposure
modeling for EPA's 1999 air toxics
assessment for the 0-4 year-old age
groups, which introduced uncertainty
into these estimates. Additionally, the
penetration factor utilized in the
derivation of the adjustment factorto
estimate fraction of outdoor air Pb that
reaches indoor air is based on another
particulate metal compound,
hexavalent chromium, which is more
reactive than Pb, potentially
introducing uncertainty into use of the
factors for Pb. Furthermore, the mean
for the penetration factor was used for
all modeled individuals which masks
potential variability in this outdoor-to-
indoor Pb ratio across individuals (for
further discussion of these sources of
uncertainty see 2007 REA, Appendix
M, Exhibit M-l).
No more recent analyses of this issue or outdoor to indoor
Pb concentrations is available.
H) Pathway apportionment of modeled blood Pb
levels (and risk) for different population
percentiles of interest at a given study area
(reflecting assumption regarding potential
correlation of Pb exposure across exposure
pathways). A critical step in modeling risk for the
2007 REA was the pathway apportionment of both
exposure and risk for specific percentiles of a
simulated population. Given the relative lack of
data characterizing correlations among exposure
pathways at the population level to inform a more
There is appreciable uncertainty
associated with this element as, in
reality, the relative roles of different
pathways might be expected to shift
with higher exposure percentiles (e.g.,
Pb paint and/or drinking water
exposures may increase in importance,
with air-related contributions
decreasing as an overall percentage of
blood Pb levels) The 2007 REA
identifies this source of uncertainty as
While several studies have used data from the EPA-
sponsored National Human Exposure Assessment Survey
(NHEXAS) dataset to characterize potential correlations
between Pb in potential media of interest including indoor
air, drinking water, indoor dust and outdoor soil (draft ISA,
Appendix 2, section 2.1.3; Clayton et al., 1999; Egeghy et
al., 2005), these analyses focus on media concentrations and
do not provide correlation information regarding percentiles
of relevant intake rates (exposure factors) for children which
would be needed to characterize potential correlations
2-26
-------
Modeling Dimension
I iicerlaiiil> Kelaled lo Application
in 2011"7 RI.A
in Informing I pdaleol" RI.A Model
refined approach, the central-tendency pathway
apportionment was applied to all modeled blood Pb
percentiles (i.e., the fraction of total blood Pb
associated with recent air-source indoor dust
ingestion for the average child was also the fraction
for the 90th and 95th percentile modeled child).
assessed to have a particularly
significant impact on the risk estimates
generated in the 2007 REA, Appendix
M, Table M-l).
between exposure pathways across percentiles of a modeled
child population.
Estimating Blood Pb Levels for Children (0-7 years of age)
I) Modeling blood Pb levels in young children
given pathway-specific intake estimates utilizing
IEUBK. In the 2007 REA, the IEUBK model was
used to estimate mean blood Pb levels for a given
scenario. A subsequent step (addressed below)
involved the use of an empirically-derived GSD to
characterize population-level variability in blood Pb
levels around that mean blood Pb estimate.
Although limited data were available in
2007 for performance evaluation, the
IEUBK estimates were observed to be
similar to data reported in the 2006
AQCD for children with known Pb
exposure (see 2007 REA, Appendix J,
section J. 1 for additional detail). The
2007 REA further identifies potential
uncertainty associated with the
soil/dust and indoor dust GI absorption
factors that were used in IEUBK
modeling (i.e., diet and drinking water
= 0.50, outdoor soil/dust and indoor
dust = 0.30) in light of the wide
variation across the estimates (2007
REA, Appendix M, Table M-l and
Appendix H, Table H-6). The 2007
REA concludes that these estimates
may either over- or under-estimate the
actual GI absorption for a child in these
study areas, with this factor having a
particularly significant impact on risk
estimates (2007 REA, Appendix M,
Table M-l).
The information newly available since the 2007 REA
includes an updated IEUBK model (v2.0) which has been
evaluated in the context of estimating exceedance of target
95th percentile blood Pb of 5 (ig/dL (draft ISA, Appendix 2,
section 2.6). This performance evaluation was completed
using children's blood Pb data paired with Pb concentrations
for yard soil and indoor dust (U.S. EPA, 2021b). A
consideration in an updated application of the IEUBK model
would be the extent to which blood Pb estimates in the
application fall into lower concentrations than those assessed
in the performance evaluation, representing an extrapolation
beyond the evaluated performance of the model and
accordingly contributing increased uncertainty to the
estimates.
The current version of IEUBK (2.0) includes updated
exposure factors for some of the exposure pathways
involved on total blood Pb modeling (U.S. EPA 2021a,
section 2.3.1.6; U.S. EPA, 2018a; U.S. EPA, 2018b; U.S.
EPA, 2018c; U.S.EPA, 2018d). Those updated factors, and
the associated evidence, will be considered in any future
simulations of exposure and blood Pb levels conducted using
the model.
J) Characterizing variability in blood Pb levels
using empirically derived GSDs. The IEUBK-
generated mean blood Pb estimate (for each
The extent to which the population on
which the GSD is based differs from
the population that is the focus of risk
Since the 2007 REA, blood Pb levels in children have
continued to decrease over time (draft ISA, Appendix 2,
section 2.4.1), although the extent to which population
2-27
-------
Modeling Dimension
1 ncer(;iin(\ Rohikd (o Application
in 2011"7 RI.A
in 1 nl'tinning I pdiKcol' RI.A Model
scenario) was combined with a GSD representing
variability around that mean estimate for children
from a hypothetical population whose Pb exposures
are similar. The GSD encompasses biological and
behavioral differences, measurement variability
from repeat sampling, variability as a result of
sample locations, and analytical variability (U.S.
EPA, 2021a, section 2.3.8).
modeling in a particular study area
contributes uncertainty to the estimates
(2007 REA, Appendix M, Table M-l).
At the time of the 2007 REA, the data
seemed to indicate a trend of higher
population-level GSDs in more recent
datasets even as overall blood Pb levels
in children decreased (possibly
reflecting the fact that a subset of
individuals were being left higher up
on the blood Pb distribution perhaps
due to artifact Pb exposure even as
general Pb exposure levels decreased),
also potentially contributing
uncertainty to the estimates (2007
REA, Appendix M, Table M-l).
The GSDs for the 2007 REA were
obtained from studies of blood Pb
variability in specific populations and
can display considerable variation
reflecting a number of factors which
can introduce uncertainty (e.g.,
population and geographic size, spatial
variation in Pb levels for key media,
differences in housing attributes or
behavior of the children).
variability (as summarized by a GSD) has changed is not
clear. More detailed analysis of the currently available
information may indicate whether GSDs associated with
more recent child population blood Pb datasets have shifted
(e.g., along with the general decreasing trend in blood Pb
levels).
Since the 2016 review, an additional population-level blood
Pb modeling approach has been developed by EPA that
utilizes a reduced form of IEUBK in combination with the
Stochastic Human Exposure and Dose Simulation (SHEDS)
multimedia model. This approach, rather than applying a
GSD to the mean blood Pb level estimated for a specific age
(based on a multi-year exposures to mean concentrations for
all relevant exposure media/pathways), applies stochastic
sampling to media concentrations and exposure parameters
to develop pathway-specific Pb intake rates for each
simulated child which are in turn translated into total blood
Pb estimates (Zartarian et al., 2017).
Although the SHEDS-IEUBK approach is valuable for
multiple applications, several aspects of the analytical design
described in this document lead us to continue to identify the
GSD-based approach for simulating exposure variability as
appropriate to this assessment. For example, the current
analytical design for this REA is not focused on a large-scale
population assessment. Rather, it focuses on small
(neighborhood-scale) groups of children living in localized
areas with ambient air concentrations equal to the standard
of interest. Also, the GSD approach proposed here avoids
the potential for added uncertainty associated with
characterizing potential correlations between exposure factor
distributions (which is required for the accurate simulation
of high-end exposure using probabilistic methods such as the
SHEDS-IEUBK approach).
2-28
-------
Modeling Dimension
1 ncerl;iinl> Kchilcd lo Application
in 200"7 RI.A
in Informing I pdiileol' RI.A Model
Estimating Risk for Young Children
K) Modeling IQ decrement in young children.
For the 2007 REA, IQ decrement was estimated
using several C-R functions derived from statistical
models of IQ association with blood Pb reported in
Lanphear et al. (2005). These functions each
involved different approaches to estimating IQ
decrement at and below the lowest blood Pb
measurements in the Lanphear et al. (2005) pooled
dataset: (a) log-linear function with a cutpoint at 1.0
Hg/dL concurrent, (b) log-linear function with low-
dose linearization, (c) dual-linear with stratification
at 10 ng/dL peak BLL and (d) dual-linear with
stratification at 7.5 ng/dL peak BLL (for additional
detail on these C-R functions see 2007 REA,
section 5.3.1.1).
While any effects of covariates on the
Lanphear et al. (2005) model
predictions are unknown, given that the
IQ change functions used in this
analysis were derived from this
Lanphear study, any inherent
differences between the Lanphear et al.
(2005) populations and the children
simulated in the 2007 REA case studies
contribute uncertainty to the associated
IQ estimates (2007 REA Appendix M,
Table M-l). The 2007 REA also
highlights as a particularly significant
source of uncertainty, in terms of its
potential impact on risk, estimates of
the degree of health decrement
associated with lower exposure levels
(i.e., blood Pb levels less than 5
(ig/dL). This magnitude of blood Pb
levels were represented in only a small
minority of the Lanphear et al (2005)
pooled dataset. Accordingly, estimates
for these blood Pb levels based on the
nonlinear function (which was intended
to capture the steeper unit risk
associated with the lower blood Pb
levels in the dataset (e.g., below 7-10
Hg/dL) while also describing the less
steep relationship across the higher Pb
levels in the dataset (e.g., 95th
percentile of 33.2 ng/dL) have
increased uncertainty. The array of
estimates produced by the four blood
Pb-IQ functions used provide an
Consideration of the currently available information for this
modeling element focuses on risk of IQ decrement impacts
at the generally lower blood Pb levels associated with
current exposures and the existing standard. As at the time of
the 2013 review, the recently available evidence does not
include new studies of blood Pb and IQ that fit our needs
with regard to a study of blood Pb and IQ in U.S. children
with blood Pb levels similar to those common in the U.S.
today. The current draft ISA makes this point, stating in the
executive summary (draft ISA, sectionES.7.1.3) that: "The
evidence assessed in the 2013 Pb ISA found that cognitive
effects in children were substantiated to occur in populations
with mean BLLs between 2 and 8 (ig/dL. Recent studies
generally include somewhat older children or employ
modelling strategies designed to answer relatively narrow
research questions and consequently do not have the
attributes of the studies on which the conclusion of the 2013
Pb ISA was based (i.e., early childhood BLLs, consideration
of peak BLLs, or concurrent BLLs in young children).
Therefore, the recently available studies were not designed
and may not have the sensitivity to detect the effect or
hazard at these very low BLLs, nor do they provide evidence
of a threshold for the effects across the range of BLLs
examined." For that reason, in considering evidence
available for modeling IQ decrement in younger children,
we look back to epidemiology studies utilized as the basis
for risk modeling in the 2007 REA.
Although the set of C-R functions derived from Lanphear et
al. (2005) and used in the 2007 REA provided different
approaches for extrapolating from the loglinear model based
on the pooled dataset to blood Pb levels generally below
those in the dataset, the vast majority of the observations in
the pooled analysis occur at exposures above our range of
potential interest, such that the fitting of the initial log-linear
2-29
-------
Modeling Dimension
I ncer(;iin(\ Rohikd lo Application
in 2011"7 RI.A
in 1 nl'tinning I pdiiiool" RI.A Model
illustration of this uncertainty (2007
REA Appendix M, Table M-l).
form may be dominated by data outside (above) our range of
interest. Several errors were identified in the original pooled
analysis of Lanphear et al. 2005 and subsequently corrected
(Lanphear et al. 2019, Crump et al. 2013, Kirrane and Patel
2014). In addition, Crump et al. (2013) extended the original
analysis by employing different approach to the log
transformation of BLL and a modelling strategy that was
designed to use more of the available data. The alternative
analysis of Crump et al. (2013) supported the findings of the
original analysis.
However, rather than relying on the analyses of the
international pooled dataset of Lanphear et al. (2005),
corrected in Lanphear et al. (2019), with its heavy
representation of blood Pb levels higher than those of
interest, another approach would be to consider a C-R
function based on analyses of the association of IQ with
blood Pb for the cohorts that had contributed to the majority
of lower blood Pb levels in the Lanphear et al. (2005
dataset), as corrected in Lanphear et al. (2019), i.e., the
Rochester, NY and Boston, MA cohorts (e.g., Lanphear et
al., 2005, Figure 2; Canfield et al., 2003; Bellinger 2003).
This focus on C-R functions from studies with Pb exposure
closer to the range of interest in this review (5-7 ng/dL and
below) will increase confidence in the associated risk
estimates.
We have focused on U.S studies to reduce uncertainty that
can result from a mismatch between attributes of the
epidemiology study population (e.g. potential confounders,
effect modifiers and behavioral factors related to Pb
exposure) and the U.S. population that is the focus of the Pb
NAAQS risk assessment. The other studies of children in the
U. S. released since completion of the 2007 REA focusing
on association of IQ with Pb exposure vary in terms of
specific factors that may compromise their appropriateness
as a basis for C-R functions to predict IQ decrements
2-30
-------
Modeling Dimension
I ncer(;iin(\ Rohikd lo Application
in 2011"7 RI.A
in 1 nl'tinning I pdiiU'ol' RI.A Model
associated with Pb in the context of risk modeling. For
example, while Chiodo et al. (2007) examines IQ
associations with blood Pb of a child cohort in Detroit, the
results may less generalizable than the Rochester or Boston
cohorts due to high prevalence of prenatal alcohol and drug
use (2013 ISA, Table 4-3). A number of more recent studies
have considered the effects of Pb co-exposure with
manganese (Mn) on children near an industrial facility in the
vicinity of East Liverpool Ohio (Haynes et al., 2015; Martin
et al., 2021). The study population is considerably older than
the 0-7 yr age range that is the focus of IEUBK-based blood
Pb modeling used in the 2007 REA. Modeling of exposure
and risk for such older child populations (e.g., children older
than seven years, including teenagers) is subject to increased
uncertainty both with regard to the timing, and other aspects
of the exposures that may be influencing the outcomes with
which blood Pb levels are associated and regarding the
simulation of exposure patterns and associated blood Pb
levels for these older cohorts. In addition, while these studies
have found associations between Pb exposure and IQ the
presence of elevated Mn exposure reduces the
generalizability of these results to other populations.
2-31
-------
An overarching limitation of the 2007 REA concerned parsing out blood Pb and risk
estimates specific to air-related Pb. As noted earlier, in light of this, we presented air-related
estimates in terms of upper and lower bounds. We note that the newly available information does
not provide for a reduction in the uncertainty associated with parsing out blood Pb and risk
estimates specific to air-related Pb.
2.2.2 Consideration of Other Health Endpoints and/or Population Age Groups/Life Stages
Another aspect to evaluation of information newly available since the 2007 REA is the
consideration of support for quantitative assessment of risk of health effects other than IQ
decrements and/or risk to other ages or life stages. In the planning phase of the 2016 review, the
2011 REA Planning Document considered specific endpoints other than IQ decrements and ages
or lifestages, for which the evidence documented a role for Pb exposures (2011 REA Planning
Document, Table 2-3). Similarly, in planning for the current review, we consider the current
evidence (including data and models) and associated limitations and uncertainties with regard to
quantitative analyses for effects other than IQ decrements in children (Table 2-4). The areas
considered here include: (a) nervous system outcomes in children other than IQ decrements, (b)
health outcomes in children other than neurobehavioral outcomes, (c) cardiovascular outcomes in
adults and (d) health outcomes in adults other than cardiovascular outcomes.
The potential for development of quantitative exposure/risk analyses in these areas is
considered in light of the information now available, e.g., with regard to the degree to which
newly available data and information would support development of quantitative exposure/risk
estimates (summarized in Table 2-4). As outlined in Table 2-4, the available evidence does not
support estimation with confidence of exposures and risks for the outcomes and age groups listed
above that can inform Pb NAAQS decision making. Key considerations in these evaluations are
summarized here.
With regard to Pb-attributable endpoints for children other than IQ decrements, as
summarized in Table 2-4, a number of factors affect our ability to quantify risks of these effects
for Pb exposures of interest. For endpoints involving exposure extending to an age older than
seven years, e.g., into the teenage years, we note the much greater uncertainty associated with
predicting blood Pb in teenagers relative to younger, preschool-age children due to the greater
complexity of both the timing and nature of Pb exposures during those years. Further, the studies
supporting conclusions regarding cognitive endpoints other than IQ, specifically academic
performance measures, rely on assessment metrics specific to a particular location (e.g.,
decrements in the passing rate for proficiency tests in a particular school system) challenging our
ability to estimate risk relevant to the broader population of children in the United States
(because exams are typically specific to each state, data cannot be directly compared across
2-32
-------
states - draft ISA, Appendix 3, p.3-69). Similarly, other endpoints may involve particular at-risk
populations (e.g., children born to mothers with increased rates of alcohol or drug use or children
with coexposure to Pb and other metals) which again can impact the generalizability of risk
estimates generated beyond the particular study population involved. As a whole, we find the
information in support of risk estimates for IQ to remain the most robust among the endpoints for
children, and such estimates have historically provided a strong foundation for NAAQS
decision-making.
Regarding the development of quantitative risk estimates for health endpoints for adults,
appreciable uncertainties in the evidence base affect our ability to reliably model those
endpoints. While the evidence is sufficient to conclude that Pb exposure is causally related to
several categories of effects in adults, including cardiovascular mortality, as in the last review,
uncertainty remains with regard to the frequency, duration and magnitude of Pb exposures
associated with these adult outcomes (draft ISA, Appendix 4, sections 4.10.1 and 4.10.2). In
addition to these uncertainties, we note that the Pb exposure histories of the adult study
populations in the epidemiology studies involve substantially higher Pb exposures, particularly
during their childhood years due to the contemporary use of Pb in gasoline, than young adults of
recent times, as is illustrated by the estimates of cumulative exposures in section 1.3.2.2 above.
Accordingly, it is not currently possible to disentangle the influence of the substantial early and
prolonged Pb exposures from the influence that much lower lifetime Pb exposure might exert on
the associations observed in the available epidemiologic evidence between adult blood Pb and
these health outcomes. Thus, the evidence does not support quantitative predictions of risk for
adult health outcomes.
2-33
-------
Table 2-4. Assessment of information (including methods and models) newly available in this review related to quantitative
assessment of Pb endpoints other than IQ decrements in children.
Ncwlv Available Information and Consideration of Potential Utility and Impact
on Quantitative Exposure/Risk Assessment by Modeling Dimension
A) Neurobehavioral endpoints in children other than IQ
In the 2008 review, while the 2006 AQCD described the evidence for a range of cognitive effects including deficits in academic achievement, effects in
attention and executive function, behavioral effects and delinquent behavior, the REA focused on decrements in IQ for younger children due to the strength of
the evidence for association with blood Pb levels below 10 ng/dL, and the strength of the dose-response information at these exposure levels (2007 REA,
section 2.1.5). The REA Planning Document for the last review additionally concluded that the studies available at that time did not support quantitative risk
assessment for neurodevelopmental endpoints besides IQ (2011 REA Planning Document, p. 2-32 and Table 2-3).
The draft ISA identified five nervous system outcomes in children as being causally or likely causally associated with Pb exposure (draft ISA, Table IS-1).
These include effects on cognitive function (including IQ), externalizing behaviors (attention, impulsivity and hyperactivity), internalizing behaviors
(depression and anxiety), conduct disorders and motor function. In determining whether the scope of the REA should be expanded to incorporate additional
endpoints in any of these outcome categories, we first consider the evidence with regard to the extent of the available information that might support
estimating risk for the endpoints associated with Pb exposures of interest, and the associated limitations and uncertainties. We then consider the degree to
which they could substantially enhance the assessment of the potential public health impact related to Pb exposure beyond that afforded by the current focus
on IQ effects in younger children. In making this determination, we contrast overall uncertainty and other limitations associated with modeling risk for these
endpoints with a similar calculus for the modeling of Pb-associated IQ effects in younger children.
Careful consideration of the evidence available as a basis for modeling risk for endpoints in the above outcome categories results in a conclusion that the REA
would not be significantly enhanced by expanding the scope to incorporate any of the above nervous system endpoints. Examples of the sources of uncertainty
and limitations leading us to this conclusion regarding such expansion of the REA scope are presented below.
With regard to cognitive function endpoints other than IQ, several epidemiology studies involved populations and/or endpoint metrics not readily
generalizable outside of the specific context of the study, thereby limiting the degree to which risks based on those studies could be interpreted in the broader
public health context for U.S. children. Specifically, a number of studies focus on standardized test performance (Evens et al., 2015, Blackowicz et al., 2016,
and Shadbegian et al., 2019). However, as noted in Draft Pb ISA Appendix 3, section 3.5.1.6.4, because exams are typically specific to each state, data cannot
be directly compared across states. This would make the generalizability of risk estimates associated with academic performance endpoints in the national-
context more challenging.. Furthermore, limitations of each of these three studies would introduce uncertainty into risk estimates based on these studies'
analyses. These limitations include (a) use of surrogates rather than direct measures for HOME inventory and parental IQ, both identified as strong
confounders for blood Pb exposure (Evens et al., 2015, Blackowicz et al., 2016) and (b) failure to collect data on parental occupation and household income,
both of which are determinants of academic performance for children (Shadbegian et al., 2019).
A number of the studies across these outcome categories would require modeling of Pb exposures from birth through the teenage years resulting in blood Pb
levels for older children (teenagers). Such estimation is associated with increased uncertainty compared to modeling blood Pb levels in younger children (0-7
years of age). This increased uncertainty results from the greater complexity in the pattern of Pb exposure for older children and the increasing role played by
endogenous Pb in influencing changes in blood Pb which is a critical element of modeling changes in Pb exposure and hence risk. Most of the studies
associated with academic performance, externalizing and internalizing behaviors fall into this category since they typically involve older children.
2-34
-------
V'\\l\ A\:iiliil)lo In To nil a I ion ;ind Consideration of Potential I IiIil> iiiul Impact
on ()iianlilali\c l-'.\posu re/Risk Assessment l)\ Modeling Dimension
A number of the studies also present challenges in terms of specifying C-R functions that might be appropriate for risk modeling. For example, studies
investigating associations of effects on executive function with Pb exposures (e.g., Ruebner et al., 2019; Fruh et al., 2019; Braun et al., 2018) display a number
of design elements that would pose obstacles to extracting a C-R function that could be used in predictive risk assessment. These include a focus on relating
health outcomes to changes in dust Pb (but not to changes in total Pb exposure or in blood Pb levels), a focus on relating outcomes to maternal blood Pb
(rather than child blood Pb), or a focus on specific subpopulations (e.g., children with chronic kidney disease). Similarly, regarding internalized behavioral
endpoints (depression and anxiety), we note that some studies (e.g., Fruh et al., 2019) investigated health outcome associations with maternal (rather than
childhood) exposure or blood Pb levels which complicates the specification and application of a C-R function.
Additionally, a number of the studies involve child cohorts from earlier periods when blood Pb levels were elevated relative to today which reduces the
representativeness of C-R functions based on these studies for our purpose of estimating risk for today's relatively lower Pb exposures. For example, in the
context of impulsivity, a more recent study while focusing on younger children (Winter and Sampson, 2017) involves a group recruited while blood Pb levels
were appreciably higher than they are today. Specifically, the cohort is from the late 1990s to early 2000s, with mean blood Pb level of 6.14 ng/dL, roughly
three times higher than the 95th percentile for young children (1-5 yrs) in the 2017-18 NHANES (and ten times the mean), contributing uncertainty to
consideration in the context of risk for today's children with relatively lower blood Pb levels.
We also note the potential for greater subjectivity and hence bias in utilizing parent-measured metrics for some of the studies examining effects of Pb exposure
on externalizing behaviors (e.g., ADHD). This is in contrast with more formally administered tests (e.g., of IQ) which would accordingly be expected to have
reduced bias in the measurement of endpoint decrements. More recent studies focusing on ADHD have attempted to address these limitations, in particular Ji
et al., 2018 (Draft Pb ISA, Appendix 3, section 3.5.2.7). Other limitations of the study by Ji et al. (2018) would introduce uncertainty into risk estimates
generated based on that study's analyses. In particular, while the study adjusted for multiple major risk factors for ADHD identified in previous studies, the
authors acknowledge that data related to multiple family-related factors such as poor parenting, maltreatment, conflict/parent-child hostility, and severe early
deprivation were not available for the study cohort which introduces the potential for confounding.
As with studies of externalized behavioral effects, we note that a number of studies examining internalized behavioral effects (anxiety and/or depression)
involved non-US populations (e.g., Liu et al., 2014; Joo et al., 2018; and Horton et al., 2018) and while these studies may be relevant in assessing overall
causality, they are not as relevant in deriving C-R functions for use in modeling risk to characterize public health impacts in the U.S.
B) Other health endpoints in children (beyond those for nervous system)
In the 2008 review, the 2006 AQCD described the evidence for health outcomes in children other than nervous system effects, including hematological and
immune effects, the strongest evidence for generating quantitative risk estimates was for neurodevelopmental effects and specifically, IQ decrements.
In the current review, in addition to nervous system effects, the draft ISA identified several health outcome categories, including cardiovascular, renal,
hematological and immunological, developmental and reproductive effects as being causally or likely causally related with Pb exposure (draft ISA, Table IS-
1). Many of the challenges identified above for quantitative risk analyses of Pb effects on nervous system endpoints besides IQ (e.g., challenges in modeling
Pb levels for older children, challenges in specifying C-R functions) also apply to other categories of health endpoints. Together these challenges affect the
suitability of studies focused on other health outcomes for quantitative assessment in the context of this review. Furthermore, given the status of IQ in children
as an endpoint with clear public health significance and with epidemiologic data that supports derivation of C-R functions, the increased value in attempting to
model non-neurological endpoints in children with their increased uncertainty is not clear.
2-35
-------
V'\\l\ A\ailal)lc In To nil a I ion ;ind (onsidcralion of Potential I IiIil> iiiul Impacl
on ()iianlilali\e l-'.\posu re/Risk Assessment l)\ Modeling Dimension
C) Cardiovascular-related health endpoints in adults
As discussed in the draft ISA, the evidence supports a causal relationship between cardiovascular effects, including cardiovascular mortality, and Pb exposure
(draft ISA, Appendix 4, section 4.12). In planning for the 2016 NAAQS review, we considered the extent of support in the evidence for quantitative risk
analyses for cardiovascular endpoints. In so doing, we noted the substantial uncertainty associated with the interpretation of the exposures eliciting the effects
analyzed in adult epidemiology studies due to the extensive historical exposures experienced by the study cohorts, which influenced the study cohort blood
and bone Pb measurements, and the cohort susceptibility. Additionally, at that time, ongoing research had not substantially reduced uncertainty related to
predicting changes in blood Pb or bone Pb associated with reductions in ambient air Pb related exposures (2011 REA Planning Document, Table 2-3). Such
uncertainties remain in the current review. As described in the draft ISA, adult blood Pb levels may be representative of contributions from both recent Pb
exposures and mobilization of legacy Pb from bone (draft ISA, Appendix 2, section 2.3.5.2). Additionally, the risk of cardiovascular health effects may be
influenced by biological responses to the elevated, earlier-in-life exposures. Accordingly, it remains unclear to what extent either recent, past, or cumulative
Pb exposures contribute to the observed associations with cardiovascular mortality (draft ISA, section 4.10.2). Due to the appreciably greater ambient air Pb
concentrations and population blood Pb levels prior to the phase out of leaded gasoline, the larger majority of study populations in the epidemiologic studies
of Pb and cardiovascular mortality (and all of the older individuals) would be expected to have had significantly elevated blood Pb levels during earlier stages
of their lives than the time of the blood Pb measurements analyzed in the epidemiologic studies, further complicating the determination of blood lead levels
that might contribute to the observed associations reported in these studies (see Table 1-3 in section 1.3.2.2 and draft ISA, Appendix 4, section 4.10.1).
A number of studies of cardiovascular health endpoints have employed NHANES study populations to assess the association of the outcomes with Pb
exposure. These studies (and their NHANES study population and follow-up period) include: Cook et al. (2022), NHANES 1988-1994, 18 yr follow up;
Menke et al. (2006), NHANES III 1988-1994, 13 yrs; Akoi et al. (2016), NHANES 1999-2010, 13 yr; Ruiz-Hernandez et al., (2017), NHANES III, 1988-
1994 NHANES 1999-2004, 8-9 yrs; Lanphear et al. (2018), NHANES III 1988-1994, 24 yrs. Consideration of the blood Pb levels recorded across the
NHANES samples indicates the appreciable lifetime exposures experienced by the majority of these cohorts, and the extent to which they are greater than
those of today's young adults. For example, as illustrated by the cumulative Pb exposure estimates for different cohorts summarized in section 1.3.2.2 above,
cohorts born in 1945 or 1970 would reasonably be estimated to experience, by age 50, as much as five to ten times the Pb exposures of a cohort born in 1990.
Even the cohort born in 1990 would, by age 50, be reasonably estimated to experience twice the Pb exposure of a cohort born in 2010 (section 1.3.2.2 above).
Data from NHANES indicates that the older individuals (e.g., older than 50 years) included in the cardiovascular mortality studies who might be expected to
comprise the majority of the cardiovascular-related deaths analyzed, and who spent a significant portion of their earlier life during the period of elevated Pb
exposure in the U.S., had quite high blood Pb levels in their younger years. For the most recent NHANES cohort studied (1999-2010 in Akoi et al. [2016]), the
older individuals (>50 yrs of age) would have been at least ~20 to 25 years old in 1980, after having experienced a full childhood during the substantially
elevated Pb exposure period of leaded gasoline. Similarly, regarding the study by Duan et al. (2020) that evaluated associations of cardiovascular mortality
and blood Pb for NHANES participants enrolled in cycles between 1999 and 2014, the draft ISA notes that "[although some members of this population may
have had lower Pb exposures due to the phaseout of leaded gasoline, especially when compared with studies assessing adults in NHANES II (1976-1980) and
NHANES III (1988-1994), the vast majority of the participants were born well before the phaseout" (draft ISA, Appendix 4, p. 4-78). This observation is even
more pronounced for studies of earlier NHANES populations, who would have spent substantially larger fractions of their earlier lives during the U.S. period
of appreciably elevated Pb exposure.
While the epidemiologic evidence of a statistical association between blood Pb and cardiovascular effects including mortality is supported by toxicological
evidence, leading to the ISA conclusion of a causal relationship of cardiovascular effects (including mortality) with Pb exposure, challenges in identifying the
2-36
-------
V'\\l\ A\ailal)lc In To nil a I ion ;ind (onsidcralion of Potential I IiIil> iiiul Impacl
on ()iianlilali\e l-'.\posu re/Risk Assessment l)\ Modeling Dimension
specific nature and pattern of Pb exposures that elicit these cardiovascular effects introduces substantial uncertainty into predictive risk estimation for a
simulated population without such appreciable exposure histories. Although, recent NHANES-based analyses that evaluate cardiovascular mortality
associations with more recent adult blood Pb levels continue to observe strong associations, "these analyses still contain populations greatly influenced by
high historic Pb exposure" (draft ISA, Appendix 4, section 4.10.2). As concluded in the draft ISA, "given the appreciable history of exposure in decades past
(see Appendix 2, Section 2.4.1), and that Pb accumulates in the body over a lifetime, the extent to which past Pb exposures contribute to the BLLs and
positive associations reported in epidemiologic studies remains uncertain" (draft ISA, Appendix 4, p. 4-76).
D) Health endpoints in adults other than cardiovascular effects
The evidence supports causal or likely causal relationships for Pb exposure and several adult health outcomes other than cardiovascular outcomes. The
limitations and uncertainties for cardiovascular outcomes, as discussed above, also pertain to these outcomes. More specifically, uncertainty in specifying the
nature and pattern of Pb exposure that is eliciting the cardiovascular endpoints in cohorts with substantial Pb exposure histories also exists for other categories
of adult health endpoints. In particular the reality of earlier periods of substantially higher Pb exposures for older-adult study populations complicates our
understanding of the extent to which the associations observed in the studies are influenced by the earlier in life exposures, introducing significant uncertainty
into any risk estimates generated for a simulated population without such significant exposure histories.
2-37
-------
2.3 INITIAL PLANS FOR THE CURRENT REVIEW
In planning regarding a new exposure and health risk assessment for this review, we have
considered how the exposure/risk information has supported judgments and decisions in the past
two Pb NAAQS reviews, as well as the availability of new information that could reduce
uncertainties, address limitations, and appreciably impact a revised or updated assessment.
Together, these considerations contribute to our conclusions regarding the extent to which a new
or substantially revised REA is warranted by the currently available information and, as
warranted, the general scope and approach for such an assessment.
In considering the role of exposure/risk estimates drawn from the 2007 REA in informing
decisions in the 2008 and 2016 NAAQS reviews, we note the associated key limitations and
uncertainties of the estimates broadly as well as increased uncertainties for estimates from some
specific case studies. While the Administrator gave primary consideration to estimates based on
an evidence-based framework in the 2008 review, the risk estimates provided a perspective on
the potential magnitude of air-related IQ decrements associated with the then-current standard in
support of the conclusion that the then-current standard did not provide the requisite protection
of public health (73 FR 66987, November 12, 2008). Given associated limitations and
uncertainties, the risk estimates were somewhat less informative to identifying a specific revised
standard. Rather, the risk estimates provided support to the evidence-based framework which
guided the Administrator's decision on a specific level for the revised standard (73 FR 67006,
November 12, 2008). In considering the risk estimates as to support for estimates based on the
evidence-based framework, the EPA placed greater emphasis on estimates for the generalized
local case study and the primary Pb smelter subarea case study, the two case studies that
represent population exposures for more highly air-pathway exposed children residing in small
neighborhoods or localized residential areas with air concentrations nearer the standard level
being evaluated (a construct most comparable to that of the evidence based framework). By
contrast, estimates from the location-specific (urban) case studies were for populations having a
broader range of air related exposures including many with air-related Pb exposures well below
the standard level being evaluated. Although these case studies provide risk estimates
representative of populations in urban areas more broadly, they are less relevant to the single
exposure zone construct of the evidence-based framework.
In the 2016 review of the Pb NAAQS established in 2008, EPA focused on both the risk
estimates generated for the generalized local case study (then called the general urban case
study), due to its simplified design and more direct comparability to the evidence-based
framework, and the location-specific case study for Chicago due to the similarity of the current
conditions scenario for that case study to expectations for air quality associated with the current
2-38
-------
standard (81 FR 71925, October 18, 2016). Our planning regarding a new exposure/risk
assessment in the current review has been informed by these past judgments regarding
limitations, uncertainties and designs of the 2007 case studies.
Based on (a) the potential for addressing key limitations and uncertainties in the 2007
REA using new information/data (see section 2.2.1 above) and (b) the role played by the 2007
REA in informing the 2008 and 2016 NAAQS reviews (summarized immediately above), we
have identified key considerations for preliminary planning with regard to conducting an REA as
part of the current Pb NAAQS review. The analytical approach being considered for the current
REA, which reflects integration of the elements detailed in the bullets below, is presented in
Figure 2-5. We emphasize that the approach outlined here is subject to ongoing modification and
refinement as we continue to consider new data/information (including the evaluation of that
information in development of the ISA for this review) and consider comments provided by the
public and CASAC in response to this document. Key elements of an analytical approach for a
REA that might inform the current review include:
Health endpoint, age cohort and pathway apportionment: The current information
supports a continued focus on modeling risk of IQ decrements for young children
exposed from birth up to seven years of age. As in the 2007 REA, we would expect to
model total Pb exposure and risk, with pathway apportionment allowing us to
apportion the estimates to 'recent air' and 'recent plus past air' pathways (with this
pair of estimates providing the bounding estimates for ambient air Pb-related exposure
and risk). We may consider modeling ADHD (change in probability of child
developing ADHD as a result in change in Pb exposure). However, given limitations
in the available study data supporting modeling of this endpoint (see Table 2-4), this
endpoint would be included as a sensitivity analysis.
IEUBK blood Pb modeling and application of the GSD: As in 2007, we expect to
model central tendency exposures using IEUBK and pathway-specific Pb intakes to
generate central-tendency blood Pb estimates to which an updated GSD(s) would be
applied to characterize the variability (due to exposure-related behavioral and
biokinetic factors, see section 2.1.1 above) in blood Pb levels for the study population.
Case study selection: We would expect to include a case study similar to the
generalized local case study. This simplified construct provides exposure and risk
estimates for a single very near-source exposure zone, rather than estimates across the
gradients common for air-related Pb that would be based on a detailed spatial analysis
of exposure. While this approach provides estimates for what is expected to be an
extremely small portion of population (children living immediately near sources
where the standard is just met), it may be sufficient to provide insight on the adequacy
of the current standard. Depending on findings for this case study, it may be
appropriate to consider other case studies providing a broader representation of
population exposure and risk. As discussed in Table 2-3 above, the limited spatial
coverage of air Pb monitors is a source of uncertainty to estimates of spatial gradients
2-39
-------
in air Pb concentrations in urban areas, as was recognized for the location specific
(city-level) case studies completed for the 2007 REA.
Use of an updated hybrid indoor dust Pb model: New information is available to
support updates to the hybrid indoor dust Pb model, particularly the loading-to-
concentration conversion step and AER data (as summarized in Table 2-3).
Consideration for the prevalence of hardwood versus carpeted flooring can also affect
the buildup of indoor dust Pb since flooring type can impact cleaning efficiency.
Given non-linearities associated with the hybrid model (as noted in the 2007 REA)
and the role of this model in linking ambient air Pb to indoor dust Pb (a critical air-
related exposure pathway), refinements/updates to this model may have substantial
impacts on the modeling of exposure and risk.
Specifying the C-R function for estimating Pb-associated IQ decrements in children:
To reduce the extent of extrapolation beyond the blood Pb levels on which the C-R
functions are based, we would intend to utilize functions derived for the subset of
study populations in the Lanphear et al. (2005) and Lanphear et al. (2019) pooled
dataset that comprised the majority of the low-exposure observations, which included
the Boston and Rochester cohorts. In applying these C-R functions, the resulting risk
estimates will be characterized with regard to the extent to which they involve
extrapolation beyond the range of the observations on which the C-R functions are
based, with attention to the increased uncertainty associated with that extrapolation.
Characterization of uncertainty: A range of sensitivity analyses may be considered for
important elements of the REA including potentially: (a) the factors converting from
the form of the standard to an annual average ambient air Pb level, (b)
parameterization of the hybrid indoor dust Pb model, (c) GSD representing variability
in blood Pb around a predicted central tendency value, and (d) C-R function for IQ
decrement. In addition to these sensitivity analyses, we also will recognize uncertainty
associated with exposure/risk estimates based on ranges of exposure beyond those
supported by the overall REA model. Identification of such exposure ranges will
likely include consideration of the performance ranges evaluated for the three main
components of the REA (hybrid indoor dust model, IEUBK blood Pb modeling
including application of the GSD and the C-R functions for IQ decrement). Each of
these has an associated range of exposure for which the component has been validated
(hybrid dust model and IEUBK) or which is reflected in the underlying observational
data (interindividual variability GSD, C-R function). To the extent instances where
exposure and risk estimates fall outside of one or more of these modeling component
ranges can be identified, associated reduction in overall confidence for those
exposure/risk estimates will be noted. Such an assessment of confidence related to
application of the three components of the REA model may be informative to
interpretation of exposure/risk estimates for the current standard and, as applicable,
any alternatives that may be considered. In addition, as with previous NAAQS-related
REAs, we will complete a characterization of uncertainty utilizing the WHO
framework (WHO, 2008). The WHO's four-tiered approach matches the
sophistication of the assessment of uncertainty to the overall complexity of the risk
assessment, while also considering the potential magnitude of the impact that the risk
assessment can have from a regulatory/policy perspective (e.g., risk assessments that
2-40
-------
are complex and are associated with significant regulatory initiatives would likely be
subjected to more sophisticated uncertainty analysis). The WHO framework includes
the use of sensitivity analysis both to characterize the potential impact of sources of
uncertainty on risk estimates and to generate an array of reasonable risk estimates.
<
X
o
Figure 2-5. Analytical approach for the REA under consideration for the current review.
2-41
-------
3 QUANTITATIVE ANALYSIS PLANNING FOR THE
SECONDARY STANDARD
In reviews of secondary NAAQS, quantitative exposure/risk analyses are generally
intended to assess ecosystem exposure and welfare risk for air quality conditions associated with
the existing standards and, as appropriate, for conditions associated with potential alternative
standards. The objective for such assessments is then to provide quantitative estimates of impacts
that inform judgments on the public welfare significance of exposures likely to occur under air
quality conditions reflective of the current NAAQS and, as appropriate, any alternative standards
under consideration. Accordingly, the assessments are also intended to provide a basis for
judgments as to the extent of ecological welfare protection afforded by such standards. Thus, the
assessments are generally intended to inform consideration of key policy relevant questions (See
IRP, volume 2), such as the following:
What are the nature and magnitude of exposures and ecological/welfare risks associated
with air quality conditions just meeting the current standard?
To what extent are the estimates of exposures and ecological risks associated with air
quality conditions just meeting the current standard reasonably judged important from a
public welfare perspective?
In considering exposure/risk estimates in this context, an accompanying consideration is:
What are the important limitations and uncertainties associated with any risk/exposure
estimates?
In this planning document, our evaluation of information available in this review is
intended to inform judgments regarding the extent to which the available information supports
development of a new quantitative risk assessment likely to contribute to substantively new
conclusions regarding the risk to welfare associated with Pb in the ambient air to inform the
Administrator's judgment of the adequacy of protection against adverse environmental effects
afforded by the current NAAQS. This evaluation is structured around consideration of two key
questions:
Is there newly available information relevant to critical uncertainties or limitations
associated with the ecological risk information in the last review?
To what extent does the currently available information support the development of an
updated or new quantitative risk assessment that would generate results providing
more specific or more certain estimates of ecological risk associated with the current
Pb secondary standard?
In designing and implementing an assessment of exposure and risk associated with Pb,
we face a level of complexity substantially greater than for similar assessments typically
3-1
-------
performed for other criteria pollutants. Unlike most other pollutants for which NAAQS are
established, Pb is a multimedia and persistent pollutant. Risk associated with exposure to Pb
originally released to ambient air is multimedia in nature, with exposure and risk associated with
a range of pathways associated with deposition of ambient air Pb. Exposure of terrestrial animals
and vegetation to air-related Pb can occur by contact with ambient air or by contact with soil,
water or food items that have been contaminated by Pb from ambient air (ISA, section 6.2).
Transport of Pb into aquatic systems similarly provides for exposure of biota in those systems,
and exposures may vary among systems as a result of differences in sources and levels of
contamination, as well as characteristics of the systems themselves. Further, the persistent nature
of Pb means that exposure and risk can be associated with Pb originally emitted into the ambient
air recently or at some point in the past, from current or historic sources, under conditions
associated with previous Pb NAAQS or prior to the existence of any Pb NAAQS. Such
historically emitted Pb can also then be transported through other media (e.g., contributing to Pb
in a water body via runoff from soils near historical air sources). Additionally, ecological
exposures and risk also result from uses of Pb that contribute Pb to the environment without it
passing through ambient air, such as land and water disposal of wastes, weathering of structures
with lead-based paint, leaching of solder used in water distribution systems into water that flows
through wastewater treatment facilities and land uses such as mining.
In considering the extent to which the currently available information warrants
development of updated quantitative exposure/ risk assessments in this review - and, as
warranted, in considering approaches and design aspects for any such assessments - we will
assess both the availability of new air quality data and data or estimates for other media that
might inform consideration of the current Pb standard, as well as any newly available scientific
evidence that indicates a more refined understanding of the effects of deposited ambient air Pb
on ecosystems and organisms. More specifically, we focus on 1) the ability of current data sets
and tools to characterize exposure of ecosystems to ambient air Pb deposited in association with
air quality meeting the existing standard and 2) the availability of new evidence that would allow
the current review to develop more robust assessments of associated welfare risk than available
in the last review.
3.1 CONSIDERATION OF 2006 ASSESSMENT IN PRIOR REVIEWS
The 2011 REA Planning Document (for the 2016 review) described our consideration of
the assessment conducted in 2006 to inform the 2008 review, and particularly the extent to which
its limitations and uncertainties could be addressed by information newly available at that time.
The conclusion, with which the CASAC Pb Review Panel in that review generally concurred,
was that the information available since the preceding (2008) review, with regard to designing
3-2
-------
and implementing a full REA that would appreciably address limitations and uncertainties of the
prior assessment, was such that performance of a new REA was not warranted. Accordingly, the
risk assessment information available for consideration in the 2016 review was based on the
screening-level risk assessment which informed decisions in the prior, 2008, review ((U.S. EPA,
2006b [referred to as the 2006 REA]; 2007 Staff Paper; 73 FR 66964, November 12, 2008).
The 2006 assessment focused on estimating the potential for ecological risks associated
with ecosystem exposures to Pb emitted into ambient air (2014 PA), section 5.2; 2006 REA,
section 7). Both a national-scale screen and a case study approach were used to evaluate the
potential for ecological impacts that might be associated with atmospheric deposition of Pb
(2014 PA, section 5.2; 2006 REA, section 7.1.2).18 The risk information from the 2006 REA
analyses informed to varying extents, and with associated limitations, the risk-based
considerations in the 2008 review regarding the potential for adverse welfare effects to result
from levels of air-related Pb that would meet the existing standard at that time.
The national-scale screen in the 2006 REA evaluated surface water and sediment
monitoring locations across the U.S. for the potential for ecological impacts that might be
associated with atmospheric deposition of Pb (2006 REA, section 7.1.2). In addition to the
national-scale screen (2006 REA, section 3.6), the assessment involved a case study approach,
with case studies for areas surrounding a primary Pb smelter (2006 REA, section 3.1) and a
secondary Pb smelter (2006 REA, section 3.2), as well as a location near a non-urban roadway
(2006 REA, section 3.4). An additional case study focused on consideration of atmospherically
derived Pb effects on an ecologically vulnerable ecosystem, the Hubbard Brook Experimental
Forest (HBEF), in the White Mountain National Forest near North Woodstock, New Hampshire
(2006 REA, section 3.5). The HBEF case study was a qualitative analysis focusing on a
summary review of the literature, without new quantitative analyses (2006 REA, Appendix E).19
For the other three case studies, exposure concentrations of Pb in soil, surface water, and/or
sediment were estimated from available monitoring data or modeling analysis and then compared
to ecological screening benchmarks (2006 REA, section 7.1). In the discussion below, we
summarize these various components of the 2006 assessment, with a focus on limitations and
associated uncertainties in interpretation of their findings (see also Table 3-1).
18 Detailed descriptions of the location-specific case studies and the national screening assessment, key findings of
the risk assessment for each, and an interpretation of the results with regard to past air quality conditions are
presented in the 2006 REA, summarized in section 5.2 of the 2014 PA and in section III.C of the 2015 proposal
(80 FR 317-319, January 5, 2015).
19 The HBEF was selected as a case study because of its location and its extensive data record on trends of Pb
concentrations in three media (air or deposition from air, soil, and surface water).
3-3
-------
In the 2006 screening assessment, the non-air media exposure estimates for the first three
case studies, and the national-scale screen for surface water and sediment, indicated a potential
for adverse effects from environmental Pb to multiple ecological receptor groups in terrestrial
and aquatic locations. However, the extent to which non-air sources or past air emissions of Pb,
under air quality conditions that did not meet the existing standard, had contributed to the surface
water or sediment Pb concentrations assessed is unclear.
For example, at some of the locations in the national-scale surface water and sediment
screen, non-air sources likely contributed significantly to the surface water Pb
concentrations. For other locations, a lack of nearby non-air sources indicated a
potential role for air sources to contribute to observed surface water Pb concentrations.
However, these concentrations may have been influenced by Pb in resuspended
sediments and may reflect contribution of Pb from erosion of soils with Pb derived
from historic as well as current air emissions. Given the appreciable changes in
ambient air Pb concentrations prior to the 2016 review, these uncertainties limited the
extent to which these national screens were informative to the 2016 consideration of
the existing standard. Given the more restrictive standard being reviewed in 2016
(than that in 2008), this limitation loomed larger in 2016 than in the 2008 review of
the standard set in 1978. With regard to the case studies, the specific circumstances of
the point source and near-roadway case studies also contributed uncertainty which
limited conclusions with regard to ambient air Pb concentrations of interest.
For the primary Pb smelter case study, while the contribution to Pb concentrations
from air, as compared to non-air sources, is not quantified, air emissions from the
smelter facility were substantial (2006 REA, Appendix B). In addition, this facility,
which closed in 2013, had been emitting Pb for many decades, including some seven
decades prior to establishment of any Pb NAAQS, such that it is likely air
concentrations associated with the facility were substantial relative to the 1978
NAAQS, which it exceeded at the time of the last review. At the time of the 2016
review and also since the adoption of the current standard, concentrations monitored
near this facility have exceeded the level of the applicable NAAQS (2007 Staff Paper,
Appendix 2B-1; 2014 PA, Appendices 2D and 5A). Accordingly, the 2016 conclusion
regarding this case study was that it was not informative for considering the likelihood
of adverse welfare effects related to Pb from air sources under air quality conditions
associated with meeting the existing Pb standard.
At the time of the 2016 review, the secondary Pb smelter case study location
continued to emit Pb, and the county where this facility is located did not meet the
current Pb standard (2014 PA, Appendices 2D and 5A). Given the exceedances of the
current standard, which likely extend back over four to five decades, this case study
was also concluded to not be informative for considering the likelihood of adverse
welfare effects related to Pb from air sources under air quality conditions associated
with meeting the existing Pb standard.
As to the near-roadway non-urban case study, its locations were recognized to be
highly impacted by past deposition of gasoline Pb. Given evidence from the past of Pb
concentrations near highways that ranged above the previous (1978) Pb standard
3-4
-------
(U.S., 1986 [1986 AQCD], section 7.2.1), it was concluded in the 2016 review that
conditions at these locations during the time of leaded gasoline very likely exceeded
the existing standard. Similarly, those conditions likely resulted in Pb deposition
associated with leaded gasoline that exceeds that being deposited (or resuspended and
redeposited20) under air quality conditions that would meet the existing Pb standard.
Given this legacy, it was concluded that the potential for environmental risks from
levels of air-related Pb associated with meeting the existing Pb standard in these
locations is highly uncertain.
The most useful case study to the 2016 review was that of the Vulnerable Ecosystem
Case Study located in the HBEF. This case study was focused on consideration of information
which included a long record (from 1976 through 2000) of available data on concentration trends
of Pb in three media (air, soil, and surface water) and on air deposition. While no quantitative
analyses were performed, a summary review of literature published on HBEF was developed.
This review indicated: (1) atmospheric Pb inputs do not directly affect stream Pb levels at HBEF
because deposited Pb is almost entirely retained in the soil profile; and (2) soil horizon analysis
results showed Pb to have become more concentrated at lower soil depths over time, with the soil
serving as a Pb sink, appreciably reducing Pb in pore water as it moves through the soil layers to
streams (dissolved Pb concentrations were reduced from 5 [j,g/L to about 5 ng/L from surface soil
to streams). As a result, this case study, and the HBEF studies on which it was based, concluded
that the contribution of dissolved Pb from soils to streams was insignificant (2006 REA,
Appendix E). Further, atmospheric input of Pb, based on bulk precipitation data, was estimated
to decline substantially from the mid-1970s to 1989; forest floor soil Pb concentrations between
1976 and 2000 were also estimated to decline appreciably (2006 REA, sections E.l and E.2). In
considering HBEF and other terrestrial sites with Pb burdens derived primarily from long-range
atmospheric transport, the 2006 AQCD found that "[djespite years of elevated atmospheric Pb
inputs and elevated concentrations in soils, there is little evidence that sites affected primarily by
long-range Pb transport have experienced significant effects on ecosystem structure or function"
(2006 AQCD, p. AX7-98). The explanation suggested by the 2006 AQCD for this finding is
"[l]ow concentrations of Pb in soil solutions, the result of strong complexation of Pb by soil
organic matter" (2006 AQCD, p. AZX7-98). While more recent soil or stream data on Pb
concentrations were not available, it was considered unlikely in the 2016 review, given the
general evidence for air Pb emissions and concentration declines over the past several decades
(e.g., 2014 PA, Figures 2-1, 2-7 and 2-8), that conditions would have worsened from those on
which these conclusions were drawn (e.g., soil data through 2000). Therefore, this information
20 Air Pb concentration trends at a limited set of near-roadway monitors in the 2013 PA provides evidence of the
decline in concentrations from the time of on-road leaded gasoline usage, and the current concentrations, at such
sites (2014 PA, Figure 2-9).
3-5
-------
was found to suggest that the lower ambient air concentrations associated with meeting the
standard set in 2008 would not be expected to directly impact stream Pb levels.
In summary, the nationwide surface water screen was not particularly informative to
evaluation of the adequacy of the 2016, or 2008, secondary Pb standard because potential
confounding by both non-air inputs and resuspension of Pb related to historic sources was not
easily accounted for. Further, the circumstances assessed in all but one of the case study
locations likely include a history of ambient air Pb concentrations that exceeded the NAAQS set
in 1978. For example, Pb deposited before that standard was enacted remains in soils and
sediments, complicating interpretations regarding the impact of the current standard; historic Pb
emitted from leaded gasoline usage continues to move slowly through systems along with more
recently deposited Pb and Pb derived from non-air sources (2014 PA, section 1.3.2).
Consequently, in the 2016 review, these analyses were not considered informative for predicting
effects at the far lower concentrations associated with the current NAAQS.
Salient aspects and associated limitations and uncertainties of the national screen and
case study approach employed in the 2006 REA (screening assessment) for the 2008 review are
summarized, along with important limitations and uncertainties, in Table 3-1 below. As indicated
above and in Table 3-1, a critical limitation of some of these analyses related to a lack of clarity
with regard to contributions from sources other than atmospheric deposition. Uncertainty
associated with that limitation is reduced for case studies, such as HBEF, for which non-air Pb
sources are unlikely. While the HBEF case study proved informative to judgments regarding air
contributions to non-air media, uncertainty remains with regard to quantitative characterization
of the associated ambient air Pb concentrations.
3-6
-------
Table 3-1. Limitations and uncertainties of the exposure/risk analyses for the 2008 review, and consideration of related newly
available information. Drawn from the 2011 REA Planning Document, 2014 PA; 2015 and 2016 notices of proposed and
final decisions; and draft ISA.
Analysis
Element
Summary of National Screens and Case Studies in 2006 REA (Screening Assessment), including Salient
Observations Focused on Recognized Limitations/Uncertainty
Newly Available Information
for Current Review
Approach for characterizing non-air exposures and linkages to air concentrations
National-
surface water
& sediment
screen
The limited availability of locations for which recent Pb measurements were available, and uncertainty regarding
potential for non-air sources to have contributed to non-air media concentrations, limited our ability to assess media
concentrations, nationally, likely to be associated with recent atmospheric Pb conditions and those associated with
existing standard.
The largest uncertainties are associated with our ability to apportion concentrations between air and non-air
sources, with additional limitations in our ability to quantitatively characterize associated air Pb concentrations.
The limitations identified for the
national screen and case
studies relate to the connection
between air concentration and
ecosystem exposures, which
continues to be poorly
characterized for Pb and to the
lack of information on the
contribution of atmospheric Pb
to specific sites, such that the
limitations in application of a
national screen and these case
studies remain.
Primary and
secondary
smelter case
studies
Exposure concentrations in soil, surface water, and/or sediment were estimated for the three case studies from
available monitoring data or modeling analysis (2006 REA, sections 3.1-3.3), however air quality conditions for the
modeling analyses as well as the monitoring data were well in exceedance of the existing standard, and the
historical records also indicated contributions from non-air sources.
Exceedances of the existing standard in these locations, which likely extend back over 4 to 5 decades, mean that
these case study datasets are not informative to considering the likelihood of adverse welfare effects related to Pb
from air sources under air quality conditions meetinq the current Pb standard.
Near-roadway
case studies
These case study locations near established highways for which soil Pb data are available, are highly impacted by
past deposition of gasoline Pb, providing information potentially useful for the standard existing in the 2008 review,
although with uncertainty in estimating the associated air Pb concentrations.
Given evidence from the past of Pb concentrations near highways that ranged above the previous (1978) Pb
standard (1986 CD, section 7.2.1), conditions at these locations during the time of leaded gasoline very likely
exceeded the existing standard, limiting their usefulness for estimating soil Pb concentrations associated with the
existing standard.
Hubbard
Brook
Experimental
Forest Case
Study
As a strength, this case study included a remote location, making atmospheric deposition the most likely contributor
to environmental Pb, and the was a long record of available data on concentration trends of Pb in three media: air or
deposition from air, soil, and surface water (2006 REA, Appendix E)). However, limitations in the availability of air
concentration data hindered quantitative characterization of Pb air quality conditions.
While this case study provided stream, soil and pore water Pb concentrations likely largely associated with air
deposition sources, and supported qualitative conclusions regarding a lack of impacts of atmospheric Pb on stream
Pb, either through direct deposition or transport of deposited Pb from surface soils, and with regard to patterns of Pb
across soil depths under air quality conditions that may have met the existing standard, quantitative analysis with
regard to air Pb concentrations was not feasible.
There are no recent publications
on Pb in the HBEF. The most
recent Pb air concentration data
for this area comes from a
monitor in Camp Dodge
(3-7pprox.. 35 miles away)
covering the years 1995-2021.
3-7
-------
Analysis
Element
Summary of National Screens and Case Studies in 2006 REA (Screening Assessment), including Salient
Observations Focused on Recognized Limitations/Uncertainty
Newly Available Information
for Current Review
Approach for characterizing potential for ecological impacts
National-
surface water
& sediment
screen
Media concentrations were compared to ecological screening
benchmarks (e.g. Ecological Soil Screening Levels (Eco-
SSLs) and Ambient Water Quality Criteria (AWQC)), which
did not accommodate a more rigorous or detailed
consideration of bioavailability characteristics influential to Pb
toxicity in these media.
With regard to the extent of lead-related effects in
the natural environment, it remains difficult to
assess the concentrations at which Pb elicits
specific effects in terrestrial and aquatic
ecosystems, due to the influence of other
environmental variables on both Pb bioavailability
and toxicity and also due to substantial species
differences in Pb susceptibility.
There is some new evidence
with regard to bioavailability and
its role in influencing the
potential for effects in the
environment. (See draft ISA,
Appendix 11, Sections 11.2,
11.3, and 11.4)
There is little new information
that would facilitate
extrapolation of evidence on
individual species, generally
from laboratory-controlled
exposure studies, to
conclusions regarding adversity
to populations or ecosystems
within the natural environment.
(See draft ISA, Appendix 11,
Sections 11.2,11.3, and 11.4)
Primary and
secondary
smelter and
near-roadway
case studies
Hubbard
Brook
Experimental
Forest Case
Study
Approach for characterizing risk for adverse effects associated with existing NAAQS
Consideration of the environmental risks and potential welfare effects associated with the current standard is
complicated by the environmental burden associated with air Pb concentrations, predominantly in the past, that
exceeded both the previously existing standard and the current standard.
The largest uncertainties are identified to be associated with our ability to determine the potential risk/extent of lead-
related effects in the natural environment and to apportion effects between air and non-air sources, and with the
relationship between effects and environmental conditions associated with the standard
Overall, these remain the major sources of uncertainty in our evaluation of effects and exposure evidence for Pb in
ecosystems.
3-8
-------
3.2 CONSIDERATION OF THE AVAILABLE EVIDENCE IN THE
CURRENT REVIEW
In planning for any new assessments or evaluations in this review, we consider
limitations and uncertainties of the prior assessment so as to direct any new analyses toward
reducing these to improve upon the usefulness of the exposure/risk information considered in
this review. Accordingly, critical at this planning stage is consideration of the extent of newly
available evidence that may address these limitations or uncertainties or otherwise augment or
improve upon the assessment from the last review. Thus, as in any review, key considerations in
planning for new analyses that may be appropriate in this review include:
Identification of aspects of the prior assessment for which updates are available and
feasible and that may reduce uncertainty or address limitations, thus improving
appropriateness of the assessment for its purpose; and
Availability of new information or tools that have potential to address key areas of
uncertainty or to provide new approaches to inform our understanding of
environmental exposures and associated risk associated with air quality under the
current standard.
In considering the extent to which the information available in this review warrants the
conduct of new exposure/risk analyses and, as warranted, in identifying the types of assessments
or evaluations to be developed or updated in this review, we give particular attention to the
assessment approaches for which limitations and uncertainties were fewer and accordingly which
were found to be most informative in the last review. As discussed in section 3.1 above, the
evaluation found to be most informative in the last review focused on a multimedia dataset for a
setting for which the source of environmental Pb was considered to be primarily atmospheric Pb
under air quality conditions considered likely to have met the existing standard (the HBEF case
study). This dataset was considered most relevant to our focus of characterizing Pb ecosystem
exposures likely to reflect predominantly atmospheric sources under air quality conditions that
would meet the current standard.
There are several areas of limited information from the last review which we consider
below with regard to the availability of new information described in the draft ISA in this
review. Our focus in these sections is on the extent to which the now available information
addresses key limitations in developing quantitative analyses that would substantively and
quantitatively inform consideration of the adequacy of the current secondary standard. Further,
we note that with regard to ecosystem services, the draft ISA concluded that "Although the
ecosystem services literature has expanded since the 2013 ISA, there are few publications that
specifically link an ecological effect attributed to Pb to a change in an ecosystem service" (draft
ISA, Appendix 11, section 11.1.5).
3-9
-------
Key aspects of the assessment developed for the 2008 review, aspects of which also
informed the last review, are briefly summarized in Table 3-1 above. As summarized there, key
limitations generally relate challenges in characterizing the role of atmospheric Pb in non-air
media exposures and associated quantitative relationships. In the 2016 review, our understanding
of the complexities in relating non-air media concentrations to air concentrations that might be
associated with the existing standard was also further informed by aspects of the evidence
available at that time (e.g., Table 3-2).
Table 3-2. Additional studies/surveys available since the 2008 review that may inform
characterization of non-air exposures associated with air Pb concentrations.
Study
Type
Summary of Information Available in the 2016 Review, including
Salient Observations Focused on Recognized Limitations/Un
Certainty
Information Newly
Available in the
Current Review
Deposition
Studies
Surveys newly reported in the 2013 ISA indicate mid-twentieth century peak in
U.S. Pb deposition, with declines documented in locations remote from
industrial areas, with stationary sources, and in industrial areas since the
1970s-80s (2013 ISA, sections 2.3.1; 2006 AQCD, section 2.3.1; Jackson et
al., 2004; Watmough and Dillon, 2007; Sabin and Schiff, 2008). Such surveys
provided little if any information on associated air Pb concentrations.
No new studies
identified in draft ISA.
Soil
Surveys
Studies in more remote forested areas continue to indicate gradual migration
of deposited Pb into mineral soils since the 70s/80s (e.g., Miller and Friedland,
1994; Zhang, 2003), without corresponding air Pb measurements.
In locations nearer point sources, historically deposited Pb (during periods
existing standard was not met) complicates estimates of soil Pb
concentrations that might be expected under conditions meeting the current
standard.
Findings to date indicate that many of those systems less influenced by
current point sources have and may still be responding to reduced Pb
deposition rates associated with reduced atmospheric emissions of Pb,
including those associated with the phase-out of leaded gasoline for on-road
vehicles, while potential responses of soils near point sources and those
involving historically deposited Pb near roadways are less well characterized,
but might be expected to have longer time horizons (2014 PA).
Recently available
national and regional
surveys of soil Pb
concentrations (draft
ISA, Appendix 11
section 11.1.3 and
Table 11-1), and also
peat core study (draft
ISA, Appendix 1,
section 1.3.3.4). See
details in Table 3-3
Terrestrial
Biota
Surveys
Limited measurements of Pb in some biota indicate reductions in biologically
available Pb over the past 30 years or more in some remote locations, such as
Pb in lichen in Mount Ranier and Sequoia/Kings Canyon National Parks (2013
ISA, section 2.6.6; Landers et al., 2008); and Pb in teeth of juvenile and adult
moose in Isle Royale National Park in northern Michigan (2013 ISA, section
2.6.8; Vucetich et al., 2009). Moss Pb concentrations near a hauling road for a
zinc and Pb mine mining increased over the decade subsequent to initiation of
the mine, with uncertainty associated with relative contribution of atmospheric
distributed Pb versus truck fallout. Across these surveys, a lack of linkages to
air Pb concentrations across the relevant temporal period limits more precise
consideration of associated air Pb concentrations.
Recently available
studies of Pb in tree
rings (draft ISA,
Appendix 11, Section
11.2.2.2) may be useful
in informing our
understanding of air Pb
and associated uptake
in the trees, to the
extent information is
also available regarding
air Pb concentrations.
See details in Table 3-3
Surface
Water
Surveys
Surveys in the 2006 AQCD were the focus of the National Surface Water
Screen in 2008 review (see above). Updated information was not available in
2016 review.
No new studies
identified in draft ISA.
3-10
-------
Study
Type
Summary of Information Available in the 2016 Review, including
Salient Observations Focused on Recognized Limitations/Un
Certainty
Information Newly
Available in the
Current Review
Sediment
Surveys
Several studies of cores or surface concentrations have observed increases in
concentration in the mid-nineteenth century continuing thru mid-twentieth
century and more recent declining Pb in sediments usually attributed to leaded
gasoline phaseout and industrial emissions reductions of 1970s-1990s.
Included in these surveys are more remote areas, including lakes in several
western U.S. National Parks (2013 ISA, section 2.6.2; Landers et al., 2010).
The declining pattern since the 1970s-80s was also reported for cores in areas
closer to metropolitan areas, with uncertainty regarding relative contribution of
changes in surface water discharges versus air deposition, although upstream
sites may be more reflective of atmospheric Pb influences (2014 PA).
Recently available study
of sediment cores in
Adirondack Mountain
lakes remote from non-
air Pb sources may be
useful in informing our
understanding of air Pb
and associated
sediment
concentrations, to the
extent information is
also available that might
inform an understanding
of linkages to air Pb
concentrations, (draft
ISA, Appendix 1,
Section 1.3.3.3). See
details in Table 3-4
Aquatic
Biota
Surveys
Shellfish monitoring across a recent 20-year period provides varying patterns
of temporal trends and is limited by combined contribution of site-specific
environmental releases as well as other ecosystem influences on Pb fate and
transport (2013 ISA, section 2.6.7; Kimbrough et al., 2008).
Recently available
studies of Pb in shellfish
and Crustacea,
depending on potential
influence of non-air
sources such as surface
water discharges, may
be informative to
conclusions regarding
air Pb and associated
aquatic biota exposures,
to the extent information
is also available to
assess linkages to air
Pb concentrations.
(draft ISA, Appendix 11,
Section 11.4.2). See
details in Table 3-5
3.2.1 Linking Atmospheric Pb to Non-Air Media Concentrations
As summarized in section 3.1 above, a critical challenge in assessing exposure and risk to
ecological receptors relates to characterization of concentrations in non-air media associated with
air quality conditions of interest. Contributing to this challenge are the multimedia aspects of Pb
and its environmental persistence, coupled with the complexity of its movement among and
within ecosystem compartments, as well as the historical environmental legacy of air
3-11
-------
concentrations well in excess of the current Pb standard and the often obscuring environmental
legacy of Pb that has not passed through ambient air, as recognized at the top of this chapter.21
Although the concept of critical loads has been described in the literature in the context
of considering potential for ecosystem impacts from atmospheric deposition of some pollutants,
many such approaches, including the very few publications that relate to Pb, do not account for
bioavailability within ecosystem compartments, fluxes within systems and other non-air
contributions.22 Such limitations were noted in the 2006 AQCD with regard to Pb and remained
in the 2018 ISA (2013 ISA, section 7.2.6).23 The draft ISA for the current review identifies no
advances to address these limitations. Accordingly, as concluded in the last review, substantial
limitations remain in the data needed to incorporate the necessary components (flux,
bioavailability, speciation, and source) at ecosystem levels in a way that could support a large-
scale application of a model that might inform evaluation of ecosystem Pb loading to inform the
NAAQS review.
Given the complexity of the temporal aspects of the distribution of Pb from air to other
media, and in light of the many past years of appreciably greater air concentrations than those
allowed by the current NAAQS, in our consideration of the currently available evidence, we
consider the potential usefulness of studies that characterize temporal trends in various
environmental media. Trends observed in such studies and the extent to which they provide a
basis for relating some characterization of non-air media concentrations to air Pb concentrations
could prove informative to judgments concerning potential for impacts to the public welfare
under air quality conditions allowed by the current standard. Currently available studies of this
type available for consideration in this light are summarized in Tables 3-3 through 3-5 below.
21 This includes Pb entering the environment from land and water disposal of wastes, weathering of structures with
lead-based paint, leaching of solder used in water distribution systems into water that flows through wastewater
treatment facilities and land uses such as mining.
22 There also significant limitations on the information that might inform the Administrator's judgments on
appropriate endpoints and targets on which to base identification of critical loads appropriate to inform decision-
making on a secondary standard protective of the public welfare.
23 The 2013 ISA noted that there were many uncertainties inherent in a critical load approach to Pb risk assessment,
such as soil type, critical concentration of dissolved metals, and the absorption coefficients of exposed soils, and
that despite a greater understanding of critical load modeling in other reviews (such as with regard to critical
loads for acidic nitrogen and sulfur compounds in vulnerable freshwater systems), we are still strongly limited
with regard to data to quantify these processes and influences adequately in such modeling of U.S. ecosystems for
Pb. Since the 2013 ISA, no recently published loading analyses for Pb for terrestrial ecosystems or for
atmospheric deposition of Pb to inland or coastal water in the U.S. have been identified (draft ISA, Section
11.1.7.1).
3-12
-------
Table 3-3. Studies describing temporal trends in Pb in soil, peat and terrestrial biota
Study Description
Location/Time
Period
Description of
Temporal Trends
Notes
Regional survey of forest
floor soils sampled in
Northeastern US (PA, NY,
NH, MA, VT, and CT)
(Richardson et al. 2014;
draft ISA, Appendix 1,
section 1.3.2.3)
16 sites across
deciduous and mixed
deciduous/coniferous
forests 1980-2011
with individual year
sampling in 1980,
1990, 2002, and
2011.
Across all sites,
mean soil Pb
concentrations
decreased from 151
± 29 (SE) mg Pb/kg
in 1980 to 68 ± 13
(SE) mg Pb/kg in
2011.
Isotopic analysis of Pb samples indicates
gasoline Pb as the dominant source of
soil Pb.
Many study site locations are in rural
areas which may have limited Pb input
from media sources other than air
deposition.
Dendrochronology study
of spruce, beech, white
pine, and cedar tree
species in Eastern
Canada
(Doucet et al. 2012; draft
ISA, Appendix 11, Section
11.2.2.2)
Reconstructed Pb
trends from 1880-
2007
Beech trees
recorded decline in
concentrations after
a 1970-1985 peak.
Long-term Pb composition of pine and
red spruce species was interpreted to be
influenced by internal physiological
mechanisms (radial mobility) and was not
used to infer air pollution history of
respective sites. Beech species seemed
to exhibit less Pb radial mobility and were
consistent with the paleolimnological
profile. The authors suggest combining
several tree species to study past air
pollution history.A
Dendrochronology study
of white spruce trees in
Northern Athabasca Oil
Sands of Western Canada
(Dinis et al. 2016; draft
ISA, Appendix 11, Section
11.2.2.2)
1878-2009
Relative Pb
concentrations
increased beginning
in 1922, peaked in
1968-1973, then
decreased until
2009.
Samples of Pb in Lichen
from Golden Lake in
Mount Rainier National
Park (2005) and Emerald
Lake Basin in
Sequoia/Kings National
Parks (2004)
(Landers et al. 2008; 2013
ISA, Section 2.6.6)
Initial sampling in
1984 at both MRNP
and Sequoia/Kings
NP, then resampled in
2005 and 2004
respectively.
Later sampling
revealed significant
reductions in Pb
measurements in
lichen (about 3-5
fold) compared to
initial samples
Air monitors within both National Parks,
which report Pb-PM2.5 (EPA AirData
website). This monitoring network
(IMPROVE) has generally operated from
1985-present data, although data may
not be available until 2004-05 for these
sites.
Six peat core survey sites
were sampled in 2013 and
2014 in Northern Alberta,
Canada
(Shotyk et al. 2016; draft
ISA, Appendix 11, Section
11.3.2)
1910-2014
Peak accumulation
rates between 1950
and 2000 in each
sample, and overall
decreasing rates of
Pb accumulation
were observed
since 1980.
Pb accumulation rates coincide with the
introduction of unleaded gasoline in the
US and Canada in the mid-1970's. The
uppermost, most recent, peat layers
show near-zero modern atmospheric Pb
deposition in the Alberta peat bogs.
Sampling sites are in largely rural areas.
A As stated by the authors, "Surprisingly, we have observed that the lead concentrations were influenced by some
physiological mechanisms (radial mobility) in red spruce and pine trees, whereas their cadmium and zinc concentrations were
useful to infer air pollution chronology. Opposite patterns, integrity for lead concentrations but translocation for cadmium and
zinc, were observed in the two studied beech stands. The results obtained clearly show that all analyzed indicators cannot be
used together in any of the investigated tree species. For that reason, we recommend combining several tree species to study
past air pollution history, as this approach can provide complementary information to reconstruct the environmental
perturbations." (Doucet et al., 2012).
3-13
-------
Table 3-4. Studies describing temporal trends in Pb in sediments of freshwater lakes.
Study Description
Time Period
Covered
Description of
Temporal Trends
Notes
Sediment core
samples in small lake
near Williamsburg,
VA
(Balascio 2019; draft
ISA, Appendix 1,
Section 1.2.7)
Core samples
estimated to
represent sediments
accumulated since
approximately 1700,
when the lake was
formed.
Pb concentrations
continued to increase
during the 1900s to a
peak maximum in
1975 followed by a
sharp decline
Pb deposition rates reflect the rise and
fall of coal combustion, smelting, and
leaded gasoline.
Study location is in relatively urban
area, having originally been used as a
mill pond, and the site likely has
received Pb inputs from media aside
from air deposition.
Sediment core
samples of 14 lakes
in 8 National Parks in
the US
(Landers et al. 2010;
draft ISA, Appendix
11, Section 11.1.3
and 2013 ISA,
Section 2.6.2)
1850-2000
In parks in the
conterminous US, Pb
showed significant
enrichment in all
sediment profiles, and
most lakes showed a
peak between 1960
and 1980, with a
decrease afterwards.
In 4 Alaska lakes, Pb
only showed modest
enrichment starting in
the 1920s.
Air monitors are present within the
bounds of the National Parks surveyed
in this study and are available on the
AirData EPA website.
Sediment cores from
two high-altitude
lakes in the
Adirondack
Mountains in NY
(Sarkar et al. 2015;
draft ISA, Appendix
1, Section 1.2.7)
1880-2007
Concentration of Pb
peaked in 1973 in
both lake cores, and
have since dropped
64% (West Pine Pond
site) and 50% (Clear
Pond site) since
-1980 at
Despite a 94% drop in national lead
emissions since 1980s, authors theorize
the less dramatic decline in Pb flux data
could be release of legacy Pb stored in
watershed soils that bias recent
sedimentary Pb concentrations.
Study sites are in remote locations and
are likely to have deposition from
ambient air as primary source of Pb
input.
Sediments in lakes in
multiple lakes across
New York, New
England and Nova
Scotia
(Dunnington, 2020;
draft ISA, Appendix
1, Section 1.3.3.4)
1800-2000
Pb dating reveals
anthropogenic Pb
concentrations began
increasing first in the
Adirondack region in
1859 followed by the
VT-NH-ME region in
1874, and finally in
Nova Scotia in 1901.
Pb declines sharply
after 1980.
Lakes sampled have long history of
anthropogenic uses, and may have Pb
inputs from multiple media pathways.
3-14
-------
Table 3-5. Studies describing temporal trends in Pb in marine biota.
Study Description
Time
Period
Covered
Description of
Temporal Trends
Notes
Long term sampling in
California coast from as
part of National Mussel
Watch (NMW) and CA State
Mussel Watch (CSMW)
(Melwani et al. 2014; draft
ISA, Appendix 11, Section
11.4.2)
1977-
2010
In NSW, 11 of 35 sites
saw decreasing Pb
levels and in CSMW, 8
of 21 sites saw
decreasing Pb levels
in metal concentration.
No significant trends
found in remaining
sites.
Declines likely reflect changes in multiple
sources, including surface water discharges
and atmospheric Pb from historic gasoline
and industrial sources.
Study locations are in highly urbanized
areas and likely received Pb input from
numerous media pathways.
Shell collection off coastal
NC in 1980, 1982, 2002,
and 2003
(Gillikin et al. 2005; draft
ISA, Appendix 11, Section
11.4.2)
1942-
2002
Pb/Ca rations in shells
peaked near 1980 and
decreased until the
conclusion of sampling
in 2003.
Annual average Pb/Ca ratios were
estimated from 1942-2002 using
concentration measurements milled
between the mollusk shell growth lines,
which provide corresponding chronological
measurements.
Study locations include both urban and rural
areas.
Study examined Pb
concentrations in horseshoe
crab eggs
(Burger and Tsipoura 2014;
draft ISA, Appendix 11,
Section 11.4.2)
1993,
1994,
1995 and
1999,
2000,
and 2012
General decline in Pb
over time in a
comparison of
compiled data from
earlier surveys (the
1993-1995) and the
later surveys (1999-
2012).
Some of the individual resampled sites
showed a clear temporal decrease in Pb
from earlier surveys to later ones, while at
other locations the temporal Pb
concentration trend was more variable.
Samplings sites in the Delaware Bay likely
have multiple pathways of Pb input.
Bivalve shell analysis off
coastal VA
(Krause-Nehring et al. 2012;
draft ISA, Appendix 11,
Section 11.4.2)
1775-
2006
General patter of
continuous increase in
Pb concentration after
1910, peaking in 1979,
and then declining to
pre-1930 values after
2000.
The elevated shell Pb corresponded to the
period of peak Pb gasoline use in the US,
with Pb deposition to the offshore site
including atmospheric transport by easterly
winds. Sampling sites in VA likely have
multiple pathways of Pb input.
3.2.2 Exposure Assessment Tools and Factors Affecting Pb Bioavailability
In considering the currently available information with regard to evaluation of potential
risk associated with air-related ecosystem exposures, we give attention to the following question:
Are there updates to or additional information regarding available risk screening tools,
such as soil screening levels, or ambient water and sediment quality criteria?
Current research on bioavailability and speciation of Pb continues to describe the
complexity of Pb bioavailability in ecosystems and the associated challenges to describing
toxicity (draft ISA, Appendix 11, sections 11.2.2, 11.3.2, and 11.4.2). As recognized in prior
3-15
-------
reviews, a wide range of environmental factors affects the distribution of Pb in the environment
and Pb bioavailability and, accordingly, Pb-induced toxicity and associated ecological risk (draft
ISA, Appendix 11, section 11.1.6; 2013 ISA, section 2.6.1; 2006 AQCD, summarized in Section
8.7).
In regard to soil screening levels, the 2006 REA (screening assessment) had relied on
Eco-SSLs, which are maximum contaminant concentrations in soils that are predicted to result in
little or no quantifiable effect on terrestrial receptors (draft ISA, Section 11.1.7.2). In that
application of the Eco-SSLs, that assessment noted appreciable uncertainty regarding the avian
Eco-SSLs. The draft ISA for the current review identifies a study proposing an update to the
avian Eco-SSL based on re-analysis of the study used in the initial 2005 derivation (Sample
2019).
The current and past evidence documents that a wide range of environmental factors
affects the distribution of Pb in the environment and Pb bioavailability and, accordingly, Pb-
induced toxicity and associated ecological risk (draft ISA Appendix 11, Section 11.1.6; 2013
ISA, Section 6.2.1; 2006 AQCD, Section 8.7). The draft ISA discusses the current evidence
regarding various aspects of Pb bioavailability and impacts on toxicity. In terrestrial and
freshwater systems, complexities related to the differences in the many interacting determinants
of bioavailability and the difficulty of identifying and quantifying those interactions have
complicated capabilities for predictive response modeling (draft ISA, section 11.1.6). Since the
2006 screening assessment, there has been some development and refinement of models that
predict toxicity by incorporating factors affecting bioavailability, which have advanced the field
of metals risk assessment.
With regard to aquatic ecosystems, the draft ISA notes that the "physicochemical
composition of the receiving water determines the bioavailability and thus the toxicity of metals
to aquatic organisms" (draft ISA, section 11.1.6). Accordingly, bioavailability models for aquatic
systems incorporate water chemistry. For example, the Biotic Ligand Model (BLM), which
predicts both the bioaccessible and bioavailable fraction of Pb in the aquatic environment, is
described as predicting toxicity of metals, including Pb, for a large range of water quality
conditions and with the consideration of metal-ligand complexation and as having potential for
estimating the importance of environmental variables such as dissolved organic carbon in
limiting uptake of metals by aquatic organisms, (draft ISA, section 11.1.6). Other approaches to
describing and predicting bioavailability and subsequent toxicity of metals in aquatic
environments include empirically based multiple linear regression models, which take into
consideration a wide range of endpoints and water chemistry parameters from large empirical
toxicity datasets (draft ISA, section 11.1.6; Brix et al., 2020). Such models are suggested to have
3-16
-------
some advantages over the BLM in adjusting for water chemistry in the context of ambient water
quality criteria (draft ISA, section 11.1.6).
There continue to be challenges affecting our ability to predict Pb exposure responses in
terrestrial systems, primarily due to much of the experimental toxicity data for terrestrial
organisms utilizing soil spiked with soluble Pb salts for dosing. In this regard, the draft ISA notes
a recent study has suggested applying empirically derived adjustments to results from laboratory
experiments using soils spiked with Pb in soluble laboratory salts to account for variation in
toxicity observed in the field due to percolation and aging, as well as differences in toxicity that
arise from differing soil properties (draft ISA, section 11.1.6). Such an approach, if evaluated
further, may be useful for addressing a limitation of Eco-SSLs, some of which are lower than
background soil Pb concentration (draft ISA, section 11.2.5).
3.3 KEY OBSERVATIONS AND INITIAL PLANS FOR THE CURRENT
REVIEW
In considering the question as to whether the currently available evidence provides
support for new evaluations or reduced uncertainty in past analysis so as to inform our
understanding of welfare risks associated with air quality under the current Pb secondary
standard, we reflect on the information discussed in the preceding sections. As an initial matter,
we look to the case study that was most informative in the last two reviews (HBEF).
Additionally, we consider the potential for additional analyses or investigations of information
available in this review to also prove informative to judgments regarding the existing standard in
the current review. We conclude that limitations remain in available datasets and in multimedia
fate, transport, exposure and toxicity modeling tools for Pb that preclude the conduct of a
comprehensive predictive risk assessment for ecological effects of air-related Pb without
substantial uncertainty. Rather for the purposes of quantitative analyses in this review, we judge
it appropriate to continue to consider the findings of the HBEF case study and to also investigate
long-term non-air media datasets in more remote areas (with little influence of non-air sources)
with regard to potential for quantitative relationships of non-air media concentrations with air Pb
concentrations that may inform consideration of multimedia responses to air quality conditions
under the current Pb standard.
With regard to the HBEF case study, we note that it continues to be provide informative
evidence regarding the consideration of atmospherically derived Pb effects on an ecologically
vulnerable ecosystem. Conclusions regarding this case study included that there was "little
evidence that sites affected primarily by long-range Pb transport [such as this one] have
experienced significant effects on ecosystem structure of function" (2006 AQCD, p AX-98). At
the time of the last review, it was considered unlikely that conditions would have appreciably
3-17
-------
changed from that of the data available through 2000 on which the previous conclusions were
based, such that the ambient air concentrations under air quality conditions associated with
meeting the standard were judged unlikely to directly impact stream Pb levels (81 FR 71938). In
planning so far in this review, we have conducted searches of the ambient air monitors24 for air
Pb monitors that could provide a historical record of ambient air lead concentrations in the areas
near the HBEF. The search did not, however, reveal any nearby monitors operating during the
time period in which HBEF Pb deposition data was collected. In further consideration of this
case study data set in this review, it may be appropriate to investigate the potential for a more
distant Pb monitor to act as surrogate in a way that may be informative to consideration of
protection offered by the existing standard.
As evidenced by the discussions above, in this review, as in the last review, we face
complications in estimating environmental Pb concentrations and exposures associated with the
current standard and associated welfare risks. However, in the discussion above we have
identified some datasets that we plan to investigate that may update our understanding in these
areas in a way that will contribute to conclusions regarding public welfare protection afforded by
the current standard. More specifically, we plan to investigate the potential for discerning status
of current environmental concentrations with regard to dependence on (or influence of) current
ambient air concentrations through evaluation of trend studies.
The temporal trends studies listed in Tables 3-3 through 3-5 presented little data on
ambient air concentrations of Pb concurrent with the non-air media measurements, limiting our
ability to conduct a full quantitative analysis of the relationship between deposition rates and
ambient air concentrations. Investigation of the availability of air Pb monitors near locations of
these trends' studies, however, may yield data that could inform our understanding of
relationships between ambient air concentrations of Pb, deposition and concentrations in other
media. A preliminary search for air monitors within the vicinity of these study locations reveals
potential availability of this type of data. Where such data may exist, consideration will be
needed on a number of factors, including distance of the air Pb monitor from a study location,
monitoring time period, sample frequency and the type of air Pb monitor (i.e., Pb-TSP, Pb-
PM2.5).
A preliminary investigation was conducted using the EPA AirData Air Monitoring
website to explore the possibility of looking at relationships between relevant proximity air
monitor concentrations and study sites for which historical Pb concentration data are available.
Air monitoring data available on the AirData website include from monitors within the SLAMS,
24 The EPA maintains a mapping application that facilitates searching for all ambient air monitors for which air
monitoring data is available in the EPA Air Quality System (https://www.epa.gov/outdoor-air-quality-data).
3-18
-------
CSN, IMPROVE, and NAATS networks (as described in IRP, VI, Appendix A). For this
investigation we focused on studies that included rural sites with limited potential for
environmental Pb releases other than atmospheric. The following studies were identified as
involving generally rural locations, with potentially relevant air monitors available. In further
analysis for consideration in the PA, we intend to investigate the availability of data from these
monitors for the time period of the study, and the possibility of Pb data from the monitors
proving useful to the evaluation of potential relationships between the air Pb and non-air media,
including with regard trend comparisons.
Richardson et al. (2014). This study described a regional survey of forest floor soils
sampled in Northeastern US (PA, NY, NH, MA, VT, and CT).
Sarkar et al. (2015). This study presented sediment core data, with dating, from two
high-altitude lakes in the Adirondack Mountains in NY.
Landers et al. (2010). This study collected sediment core data of 14 lakes in 8 National
Parks in the U.S.
Landers et al. (2008). This study reported on Pb concentrations in Lichen from Golden
Lake in Mount Rainier National Park (2005) and Emerald Lake Basin in
Sequoia/Kings National Parks (2004).
Analyses developed in this review will be described in the PA, with details documented
in appendices or accompanying volumes, as appropriate. We expect to also consider in the PA
any analyses from the last review that we do not update in this review but that are still
informative to this review when viewed in the context of the currently available evidence as
characterized in the ISA and of updated air quality and other analyses performed for this review.
Accordingly, the PA will include description and discussion of all analyses being considered of
in this review, both those newly developed in this review as well as analyses from the last review
for which an updated assessment was not performed but that are still informative for this review.
The draft PA will be released for public comment and provided to the CASAC for its review.
Advice and comments received will be considered in completing the final version of the
quantitative analyses and information considered in the policy evaluations presented in the final
PA.
3-19
-------
4 REFERENCES
Aoki, Y, Brody, DJ, Flegal, KM, Fakhouri, THI, Axelrad, DA, and Parker, JD (2016). Blood
lead and other metal biomarkers as risk factors for cardiovascular disease mortality.
Medicine (Baltimore) 95(1): e2223. HERO ID: 4643410.
http://dx.doi.org/10.1097/MD.00000000000Q2223.
Aoki, Y, Brody, DJ, and Flegal, K (2017). Blood lead and other metal biomarkers as risk factors
for cardiovascular disease mortality: Erratum. Medicine 96: e7997. HERO ID: 7113738.
http://dx.doi.org/10.1097/MD.00000000000Q7997.
Balascio, NL, Kaste, JM, Meyer, MG, Renshaw, M, Smith, K, and Chambers, RM (2019). A
high-resolution mill pond record from eastern Virginia (USA) reveals the impact of past
landscape changes and regional pollution history. Anthropocene 25: 100190. HERO ID:
6735952. https://doi.Org/10.1016/i.ancene.2019.100190.
Bellinger, DC (2003). Perspectives on incorporating human neurobehavioral end points in risk
assessments. Risk Analysis 23(1): 163-174. HERO ID: 51898.
http://dx.doi.org/10.1111/0272-4332.00045.
Blackowicz, MJ, Hryhorczuk, DO, Rankin, KM, Lewis, DA, Haider, D, Lanphear, BP, and
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(8): 774. HERO ID: 4856035. http://dx.doi.org/10.3390/iierphl3080774.
Bevington, C, Gardner, HD, Cohen, J, Henning, C, and Rasmussen, PE (2021). Relationship
between residential dust-lead loading and dust-lead concentration across multiple North
American datasets. Build Environ 188: 107359. HERO ID: 7332562
http://dx.doi.Org/10.1016/j.buildenv.2020.107359.
Braun, JM, Hornung, R, Chen, A, Dietrich, KN, Jacobs, DE, Jones, R, Khoury, JC, Liddy-Hicks,
S, Morgan, S, Vanderbeek, SB, Xu, Y, Yolton, K, and Lanphear, BP (2018). Effect of
residential lead-hazard interventions on childhood blood lead concentrations and
neurobehavioral outcomes: A randomized clinical trial. JAMAPediatr 172(10): 934-942.
HERO ID: 5017788. http://dx.doi.org/10.1001/iamapediatrics.2Q18.2382.
Brix, KV, DeForest, DK, Tear, L, Peijnenburg, W, Peters, A, Middleton, ET, and Erickson, R
(2020). Development of empirical bioavailability models for metals. Environ Toxicol and
Chem 39(1): 85-100. HERO ID: 6708125. http://dx.doi.org/10.1002/etc.4570.
Burger, J, and Tsipoura, N (2014). Metals in horseshoe crab eggs from Delaware Bay, USA:
Temporal patterns from 1993 to 2012. Environ Monit Assess 186(10): 6947-6958. HERO
ID: 2347606. http://dx.doi.org/10.1007/s 10661-014-3901-8.
Canfield, RL, Henderson, CR, Jr, Cory-Slechta, DA, Cox, C, Jusko, TA and Lanphear, BP
(2003). Intellectual impairment in children with blood lead concentrations below 10 jag
per deciliter. N Engl J Med 348: 1517-1526. HERO ID: 42310
http: //dx. doi. or g/10.105 6/NE JMoa022848.
4-1
-------
CDC (2021) National Report on Human Exposure to Environmental Chemicals (Report).
http s: //www, cdc. gov/nceh/1 ead/data/nhanes. htm.
Chiodo, LM, Covington, C, Sokol, RJ, Hannigan, JH, Jannise, J, Ager, J, Greenwald, M, and
Delaney-Black, V (2007). Blood lead levels and specific attention effects in young
children. Neurotoxicol Teratol 29(5): 538-546. HERO ID: 460552.
http://dx.doi.Org/10.1016/i.ntt.2007.04.001.
Clayton, CA, Pellizzari, ED, Whitmore, RW, Perritt, RL, and Quackenboss, JJ (1999). National
Human Exposure Assessment Survey (NHEXAS): Distributions and associations of lead,
arsenic, and volatile organic compounds in EPA Region 5. J Expo Anal Environ
Epidemiol 9(5): 381-392. HERO ID: 14003. http://dx.doi.org/10.1038/si.iea.7500Q55.
Cohen, J, Mallya, H, and Rosenbaum, A (2012). Updated Analysis of Air Exchange Rate Data.
Appendix 5E of U.S. EPA. 2014. Health Risk and Exposure Assessment for Ozone. Final
Report. Chapter 5 Appendices. Research Triangle Park, NC. Available at:
https://www.epa.gov/naaqs/ozone-o3-standards-risk-and-exposure-assessments-review-
completed-2015.
Cook, MK, Zhang, J, and Wei, Y (2022). Blood lead levels and risk of deaths from
cardiovascular disease. Am J Cardiol 173: 132-138. HERO ID: 10294133.
http: //dx. doi. or g/10.1016/i. ami card .2022.03.002.
Crump, KS, Van Landingham, C, Bowers, TS, Cahoy, D, and 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. HERO ID: 3838553. http://dx.doi.org/10.3109/10408444.2Q13.832726.
Dinis, L, Savard, MM, Gammon, P, Begin, C, and Vaive, J (2016). Influence of climatic
conditions and industrial emissions on spruce tree-ring Pb isotopes analyzed at ppb
concentrations in the Athabasca oil sands region. Dendrochronologia 37: 96-106. HERO
ID: 3458987. http://dx.doi.Org/10.1016/i.dendro.2015.12.011.
Doucet, A, Savard, MM, Begin, C, Marion, J, Smirnoff, A, and Ouarda, TBMJ (2012).
Combining tree-ring metal concentrations and lead, carbon and oxygen isotopes to
reconstruct peri-urban atmospheric pollution. Tellus B Chem Phys Meteorol 64: 19005.
HERO ID:1537188. http://dx.doi.org/10.3402/tellusb.v64i0.19005.
Duan, W, Xu, C, Liu, Q, Xu, J, Weng, Z, Zhang, X, Basnet, TB, Dahal, M, ND Gu, A. (2020).
Levels of a mixture of heavy metals in blood and urine and all-cause, cardiovascular
disease and cancer mortality: A population-based cohort study. Environ Pollut 263:
114630. HERO ID: 6717646. http://dx.doi.Org/10.1016/i.envpol.2020.114630.
Dunnington, DW, Roberts, S, Norton, SA, Spooner, IS, Kurek, J, Kirk, JL, Muir, DCG, White,
CE, and Gagnon, GA (2020). The distribution and transport of lead over two centuries as
recorded by lake sediments from northeastern North America. Sci Total Environ 737:
140212. HERO ID: 6858566. https://www.ncbi.nlm.nih.gov/pubmed/32783842.
4-2
-------
Egan, KB; Cornwell, CR, Courtney, JG, and Ettinger, AS (2021). Blood lead levels in U.S.
children ages 1-11 years, 1976-2016. Environ Health Perspect 129: 37003. HERO ID:
10197483 http://dx.doi.org/10.1289/EHP7932.
Egeghy, PP, Quackenboss, JJ, Catlin, S, and Ryan, PB (2005). Determinants of temporal
variability in NHEXAS-Maryland environmental concentrations, exposures, and
biomarkers. J Expo Anal Environ Epidemiol 15: 388-397. HERO ID: 89369
http://dx.doi.org/10.1038/si.iea.7500415.
Evens, A, Hryhorczuk, D, Lanphear, BP, Rankin, KM, Lewis, DA, Forst, L, and 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. HERO ID: 6771303 http://dx.doi.org/10.1186/sl2940-015-00Q8-9.
Frey, HC (2011) Letter from Dr. H. Christopher Frey, Chair, Clean Air Scientific Advisory
Committee Lead Review Panel, to Administrator Lisa P. Jackson. Re: Consultation on
EPA's Review of the National Ambient Air Quality Standards for Lead: Risk and
Exposure Assessment Planning Document. October 14, 2011. Available at:
https://www. regulations. gov/document/EP A-HQ-OAR-2010-0108-0026.
Fruh, V, Rifas-Shiman, SL, Amarasiriwardena, C, Cardenas, A, Bellinger, DC, Wise, LA, White,
RF, Wright, RO, Oken, E, and Henn, BC (2019). Prenatal lead exposure and childhood
executive function and behavioral difficulties in project viva. Neurotoxicology 75: 105-
115. HERO ID: 6680639. http://dx.doi.Org/10.1016/i.neuro.2019.09.006.
Gillikin, DP, Dehairs, F, Baeyens, W, Navez, J, Lorrain, A, and Andre, L (2005). Inter- and
intra-annual variations of Pb/Ca ratios in clam shells (Mercenaria mercenaria): A record
of anthropogenic lead pollution? Mar Pollut Bull 50: 1530-1540. HERO ID: 474485.
http://dx.doi.Org/10.1016/i.marpolbul.2005.06.020.
Haynes, 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.
Horton, MK, Hsu, L, Henn, BC, Margolis, A, Austin, C, Svensson, K, Schnaas, L, Gennings, C,
Hu, H, Wright, R, Tellez Rojo, MM, and Arora, M (2018). Dentine biomarkers of
prenatal and early childhood exposure to manganese, zinc and lead and childhood
behavior. Environ Int 121: 148-158. HERO ID: 5047408.
http://dx.doi.Org/10.1016/i.envint.2018.08.045.
Hunt, A., Johnson, D. L., Brooks, J., & Griffith, D. A. (2008). Risk remaining from fine particle
contaminants after vacuum cleaning of hard floor surfaces. Environmental Geochemistry
and Health, 30(6), 597-611. http://dx.doi.org/10.1007/slQ653-008-9183-8
4-3
-------
Hu, H, Shih, R, Rothenberg, S, and Schwartz, BS (2007). The epidemiology of lead toxicity in
adults: measuring dose and consideration of other methodologic issues. Environ Health
Persp 115: 455-462. HERO ID: 481462.
https://www.ncbi.nlm.nih.gov/pubmed/17431499.
Joo, H, Choi, JH, Burm, E, Park, H, Hong, YC, Kim, Y, Ha, EH, Kim, Y, Kim, BN, and 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.
HERO ID: 4914827. http://dx.doi.Org/10.1016/i.scitotenv.2017.10.007.
Jackson, BP, PV, Winger, and Lasier, PJ (2004), Atmospheric lead deposition to Okefenokee
Swamp, Georgia, USA. Environ Poll. 130: 445-451. HERO ID: 72533.
https://www.ncbi.nlm.nih.gov/pubmed/15182975.
Kimbrough, KL, Lauenstein, GG, Christensen, JD, and Apeti, DA (2008). An assessment of two
decades of contaminant monitoring in the nation's coastal zone. National Centers for
Coastal Ocean Science, Silver Spring, MD. NOAA Technical Memorandum NOS
NCCOS 74. HERO ID: 671524.
https://repositorv.librarv.noaa.gov/view/noaa/2499/noaa 2499 DS1 .pdf.
Kirrane EF, and Patel, MM (2014). Identification and consideration of errors in Lanphear et al.
(2005), "Low-level environmental lead exposure and children's intellectual function: An
international pooled analysis". Available at: https://www.regulations.gov/document/EPA-
HO-QRD-2011-0051-0050.
Krause-Nehring, J, Brey, T, and Thorrold, SR (2012). Centennial records of lead contamination
in northern Atlantic bivalves (Arctica islandica). Mar Pollut Bull 64: 233-240. HERO ID:
6701336. http://dx.doi.Org/10.1016/i.marpolbul.2011.11.028.
Landers, DH, Simonich, SL, Jaffe, DA, Geiser, LH, Campbell, DH, Schwindt, AR, Schreck, CB,
Kent, ML, Hafner, WD, Taylor, HE, Hageman, KJ, Usenko, S, Ackerman, LK, Schrlau,
JE, Rose, NL, Blett, TF, and Erway, MM (2008). The fate, transport, and ecological
impacts of airborne contaminants in western national parks (USA). (EPA/600/R-07/138).
Corvallis, OR: U.S. Environmental Protection Agency, NHEERL, Western Ecology
Division. Available at: https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P 100EJ34.txt.
Landers, DH, Simonich, SM, Jaffe, D, Geiser, L, Campbell, DH, Schwindt, A, Schreck, C, Kent,
M, Hafner, W, Taylor, HE, Hageman, K, Usenko, S, Ackerman, L, Schrlau, J, Rose, N,
Blett, T, and Erway, MM (2010). The Western Airborne Contaminant Assessment
Project (WACAP): An interdisciplinary evaluation of the impacts of airborne
contaminants in western U.S. National Parks. Environ Sci Technol 44: 855-859. HERO
ID: 677566. http://dx.doi.org/10.1021/es901866e.
Lanphear, BP, Hornung, R, Khoury, J, Yolton, K, Baghurst, P, Bellinger, DC, Canfield, RL,
Dietrich, KN, Bornschein, R, Greene, T, Rothenberg, SJ, Needleman, HL, Schnaas, L,
Wasserman, G, Graziano, J, and Roberts, R (2005). Low-level environmental lead
exposure and children's intellectual function: An international pooled analysis. Environ
Health Perspect 113: 894-899. HERO ID: 73026. http://dx.doi.org/10.1289/ehp.7688.
4-4
-------
Lanphear, BP, Rauch, S, Auinger, P, Allen, RW, and Hornung, RW (2018). Low-level lead
exposure and mortality in US adults: A population-based cohort study. The Lancet Public
Health 3: el77-el84. HERO ID: 4260470. http://dx.doi.org/10.lQ16/S2468-
2667(18^30025-2.
Lanphear, BP, Hornung, R, Khoury, J, Yolton, K, Baghurst, P, Bellinger, DC, Canfield, RL,
Dietrich, KN, Bornschein, R, Greene, T, Rothenberg, SJ, Needleman, HL, Schnaas, L,
Wasserman, G, Graziano, J, and Roberts, R (2019). Erratum: "Low-level environmental
lead exposure and children's intellectual function: An international pooled analysis"
[Erratum], Environ Health Perspect 127: 099001. HERO ID: 6687219.
http://dx.doi.org/10.1289/EHP5685.
Liu, J; Liu, X; Wang, W; Mccauley, L; Pinto-Martin, J; Wang, Y; Li, L; Yan, C; Rogan, WJ.
(2014). Blood lead concentrations and children's behavioral and emotional problems: A
cohort study. JAMA Pediatrics 168 (8):737-745.
Martin, KV, Sucharew, H, Dietrich, KN, Parsons, PJ, Palmer, CD, Wright, R, Amarasiriwardena,
C, Smith, DR, and Haynes, EN (2021). Co-exposure to manganese and lead and pediatric
neurocognition in East Liverpool, Ohio. Environ Res 202: 111644. HERO ID: 10276134
http://dx.doi.Org/10.1016/i.envres.2021.l 11644.
Melwani, AR, Gregorio, D, Jin, Y, Stephenson, M, Ichikawa, G, Siegel, E, Crane, D, Lauenstein,
G, and Davis, JA (2014). Mussel Watch update: long-term trends in selected
contaminants from coastal California, 1977-2010. Mar Pollut Bull 81: 291-302. HERO
ID: 2551525. http://dx.doi.Org/10.1016/i.marpolbul.2013.04.025.
Menke, A, Muntner, P, Batuman, V, Silbergeld, EK, and Guallar, E (2006). Blood lead below
0.48 [j,mol/l (10 (J,g/dl) and mortality among US adults. Circulation 114: 1388-1394.
HERO ID: 194179. http://dx.doi.Org/10.l 161/circulationaha. 106.628321.
Miller, EK, and Friedland, AJ (1994). Lead migration in forest soils: response to changing
atmospheric inputs. Environ. Sci. Technol. 28: 662-669. HERO ID: 46732.
http://dx.doi.org/10.1021/esQ0053a020.
Nie, LH, Wright, RO Bellinger, DC, Hussain, J, Amarasiriwardena, C, Chettle, DR, Pejovic-
Milic, A, Woolf, A, and Shannon, M (2011). Blood lead levels and cumulative blood lead
index (CBLI) as predictors of late neurodevelopment in lead poisoned children.
Biomarkers 16: 517-24. HERO ID: 690017. http://dx.doi.org/10.1088/0Q31-
9155/56/3/N01.
Richardson, J, Friedland, AJ, Kaste, JM, and Jackson, BP (2014). Forest floor lead changes from
1980 to 2011 and subsequent accumulation in the mineral soil across the Northeastern
United States. J Environ Qual 43: 926-935. HERO ID: 2632817.
http://dx.doi.org/10.2134/ieq2013.10.Q435.
Ruebner, RL, Hooper, SR, Parrish, C, Furth, SL, and Fadrowski, JJ (2019). Environmental lead
exposure is associated with neurocognitive dysfunction in children with chronic kidney
disease. Pediatr Nephrol 34: 2371-2379. HERO ID: 6725824.
http://dx.doi.org/10.1007/s00467-019-043Q6-7.
4-5
-------
Ruiz-Hernandez, A, Navas-Acien, A, Pastor-Barriuso, R, Crainiceanu, CM, Redon, Guallar, E,
and Tellez-Plaza, M (2017). Declining exposures to lead and cadmium contribute to
explaining the reduction of cardiovascular mortality in the US population, 1988-2004. Int
J Epidemiol 46: 1903-1912. HERO ID: 4642774. http://dx.doi.org/10.1093/iie/dvxl76.
Sabin, LD; Schiff, KC. (2008). Dry atmospheric deposition rates of metals along a coastal
transect in southern California. Atmos Environ 42: 6606-6613. HERO ID: 516933
http://dx.doi.Org/10.1016/i.atmosenv.2008.04.042.
Sample, BE; Beyer, WN; Wentsel, R. (2019). Revisiting the avian Eco-SSL for lead:
Recommendations for revision. Integr Environ Assess Manag 15: 739-749.
http: //dx. doi. org/10.1002/ieam.4157
Sarkar, S, Ahmed, T, Swami, K, Judd, CD, Bari, A, Dutkiewicz, VA, and Husain, L (2015).
History of atmospheric deposition of trace elements in lake sediments, similar to 1880 to
2007. J Geophys Res Atmos 120: 5658-5669. HERO ID: 3019076.
http://dx.doi.org/10.1002/2015JD0232Q2.
Sarkar, A, Asaeda, T, Wang, Q, Kaneko, Y, and Rashid, MH (2018). Arbuscular mycorrhiza
confers lead tolerance and uptake in Miscanthus sacchariflorus. Chem Ecol 34: 454-469.
HERO ID: 4598835. http://dx.doi.org/10.1080/02757540.2018.143715Q.
Schwartz, J, and Pitcher, H (1989) The relationship between gasoline lead and blood lead in the
United States. J Official Statistics 5: 421-431. HERO ID: 625648.
https://www.scb.se/contentassets/ca21efb41fee47d293bbee5bf7be7fb3/the-relationship-
between-gasoline-lead-and-blood-lead-in-the-united-states.pdf.
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.
HER O ID: 6721887. http://dx.doi.Org/10.1016/i.envres.2019.108643.
Shepard, EA (2022) Letter from Dr. Elizabeth A. Shepard, Chair, Clean Air Scientific Advisory
Committee, to Administrator Michael Regan. Re: Consultation on the EPA's Integrated
Review Plan for the National Ambient Air Quality Standards for Lead. Volume 2:
Planning for the Review and the Integrated Science Assessment (March 2022). EPA-
CASAC-22-003. April 22, 2022.
Shotyk, W, Appleby, PG, Bicalho, B, Davies, L, Froese, D, Grant-Weaver, I, Krachler, M,
Magnan, G, Mullan-Boudreau, G, Noernberg, T, Pelletier, R, Shannon, B, van Bellen, S,
and Zaccone, C (2016). Peat bogs in northern Alberta, Canada reveal decades of
declining atmospheric Pb contamination. Geophys Res Lett 43: 9964-9974. HERO ID:
3475235. http://dx.doi.org/10.1002/2016GL07Q952.
Somervaille, LJ, Chettle, DR, Scott, MC, Tennant, DR, McKiernan, MJ, Skilbeck, A, and
Trethowan, WN (1988). In vivo tibia lead measurements as an index of cumulative
exposure in occupationally exposed subjects. British J Ind Med 45: 174-181. HERO ID:
57784. https://www.ncbi.nlm.nih.gov/pubmed/3348993.
U.S. EPA. (1986). Air Quality Criteria for Lead (Final Report, 1986). Washington, DC.
EPA/600/8-83/028AF. Available at: https://www.epa.gov/naaqs/lead-pb-air-qualitv-
standards-documents-prior-review-completed-2008.
4-6
-------
U.S. EPA (1989). Review of National Ambient Air Quality Standard for Pb: Exposure Analysis
Methodology and Validation. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA-450/2-89-011. Available at:
http://www.epa.gOv/ttn/naaqs/standards/pb/s pb pr td.html.
U.S. EPA (1995). Report on the National Survey of Lead-Based Paint in Housing: Appendix I:
Design and Methodology. Office of Pollution Prevention and Toxics, Washington, DC.
EPA 747-R95-004. Available at: https://www.epa.gov/lead/report-national-survev-lead-
based-paint-housing-appendix-i-design-and-methodology.
U.S. EPA (2003) America's Children and the Environment: Measures of Contaminants, Body
Burdens, and Illnesses. 2nd ed. EPA-240-R-03-001. February 2003. Available at:
https://www.epa.gov/americaschildrenenvironment/publications
U.S. EPA (2006a). Air Quality Criteria for Lead: Volume I of II. Office of Research and
Development, Research Triangle Park, NC. EPA/600/R-05/144aF. Available at:
https://cfpub. epa.gov/ncea/risk/recordisplav. cfm?deid=l 58823.
U.S. EPA (2006b). Lead Human Exposure and Health Risk Assessments and Ecological Risk
Assessment for Selected Areas. Pilot Phase. External Review Draft Technical Report.
Research Triangle Park, NC. EPA/600/R-05/144aF. Available at:
https://www.epa.gov/naaqs/lead-pb-air-qualitv-standards-documents-review-completed-
2008.
U.S. EPA (2006c). 1999 National-Scale Air Toxics Assessment. Available at:
https://archive.epa.gov/airtoxics/natal999/web/html/index.html.
U.S. EPA (2007). Review of the National Ambient Air Quality Standards for Lead: Policy
Assessment of Scientific and Technical Information OAQPS Staff Paper. Office of Air
Quality Planning and Standards, Research Triangle Park, NC. EPA452/R-07-013.
Available at: https://www.epa.gov/naaqs/lead-pb-air-qualitv-standards-documents-
review-completed-2008.
U.S. EPA. (2011). Review of the National Ambient Air Quality Standards for Lead: Risk and
Exposure Assessment Planning Document. Office of Air Quality Planning and Standards,
Research Triangle Park, NC. EPA/452/P-11-003. Available at:
http://www.epa.gOv/ttn/naaqs/standards/pb/s pb 2010 pd.html
U.S. EPA (2013a). Integrated Science Assessment (ISA) for Lead (Final Report, Jul 2013).
Office of Research and Development, Research Triangle Park, NC. EPA/600/R-10/075F,
2013. Available at: https://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=255721.
U.S. EPA (2013b). America's Children and the Environment, 3rd ed. EPA-240-R-13-001.
January 2013. Available at: https://www.epa.gov/americaschildrenenvironment/basic-
information-about-ace
4-7
-------
U.S. EPA (2014). Policy Assessment for the National Ambient Air Quality Standards for Lead.
Office of Air Quality Planning and Standards. Research Triangle Park, NC. EPA-452/R-
14-001. Available at: https://www.epa.gov/sites/default/files/2020-
07/documents/140501 pa pb fin.pdf.
U.S. EPA (2018a). Estimation of inhalation rates for U.S. children: update to the default values
for the integrated exposure uptake biokinetic model for lead in U.S. children. OLEM
Directive 9200.1-147. February 17. Available at:
https://semspub.epa.gov/work/HQ/400703.pdf.
U.S. EPA (2018b). Estimation of dietary lead exposure: update to the default values for the
integrated exposure uptake biokinetic model for lead in U.S. children. OLEM Directive
9200.1-149. February 17. Available at: https://semspub.epa.gov/work/HO/4007Q6.pdf.
U.S. EPA (2018c). Estimation of daily water consumption: update to the default values for the
integrated exposure uptake biokinetic model for lead in U.S. children. OLEM Directive
9200.1-151. February 17. Available at: https://semspub.epa.gov/work/HO/4007Q5.pdf.
U.S. EPA (2018d). Estimation of lead exposure from water sources: update to the default values
for the integrated exposure uptake biokinetic model for lead in U.S. children. OLEM
Directive 9200.1-153. February 17. Available at:
https://semspub.epa.gov/work/HO/4007Q4.pdf.
U.S. EPA (2019). America's Children and the Environment. October 2019. Publication number
100K19004. Available at: https://www.epa.gov/americaschildrenenvironment/americas-
children-and-environment-october-2019
U.S. EPA (2020). Policy Assessment for the Review of the Ozone National Ambient Air Quality
Standards. U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC. EPA-452/R-20-001, May 2020. Available at:
https://www.epa.gov/sites/default/files/2020-05/documents/o3-final pa-05-29-
20compressed.pdf.
U.S. EPA (2021a). User's Guide for the Integrated Exposure Uptake Biokinetic Model for Lead
in Children (IEUBK) Version 2.01 (FINAL), May 2021. Available at:
https://semspub.epa.gov/work/HO/40070Q.pdf.
U.S. EPA (2021b) Advancing Pb Exposure and Biokinetic Modeling for U.S. EPA Regulatory
Decisions and Site Assessments Using Bunker Hill Mining and Metallurgical Complex
Superfund Site Data. Office of Research and Development, Washington, DC,
EPA/600/R-21/017F, 2021, April 2021. Available at:
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=351563.
U.S. EPA (2022a) Integrated Review Plan for the National Ambient Air Quality Standards for
Lead. Volume 1: Background Document. U.S. Environmental Protection Agency, Office
of Air Quality Planning and Standards and Office of Research and Development.
Research Triangle Park, NC. EPA-452/R-22-003a. March 2022. Available at:
https://www.epa.gov/naaqs/lead-pb-standards-planning-documents-current-review.
4-8
-------
U.S. EPA (2022b). Integrated Review Plan for the National Ambient Air Quality Standards for
Lead. Volume 2: Planning for the Review and the Integrated Science Assessment. U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards and
Office of Research and Development. Research Triangle Park, NC. EPA-452/R-22-003b.
March 2022. Available at: https://www.epa.gov/naaqs/lead-pb-standards-planning-
documents-current-review
U.S. EPA (2022c). Update to Lead Biomonitoring Indicator of America's Children and the
Environment. Available at: https://www.epa.gov/americaschildrenenvironment/basic-
information-about-ace
U.S. EPA (2023a). Integrated Science Assessment (ISA) for Lead (Pb) (External Review Draft).
U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-23/061, 2023.
U.S. EPA (2023b). Overview of Lead (Pb) Air Quality in the United States. Updated: February
07, 2023. Office of Air Quality Planning and Standards, Research Triangle Park, NC.
Available at: https://www.epa.gov/air-qualitv-analvsis/lead-naaqs-review-analvses-and-
data-sets.
Vucetich. JA. Outridge. PM. Peterson. R.O. Eide. R. and Isrenn. R (2009). Mercury, lead and
lead isotope ratios in the teeth of moose (Alces alces) from Isle Royale, US Upper
Midwest, from 1952 to 2002. J Environ Monit 11: 1352-1359. HERO ID: 532646.
http://dx.doi.org/10.1039/b903417i.
Watmough, SA, and Dillon, PJ (2007). Lead biogeochemistry in a central Ontario forested
watershed. Biogeochemistry 84: 143-159. HERO ID: 534034.
http://dx.doi.org/10.1007/sl0533-007-911Q-6.
Winter, AS and Sampson, RJ (2017). From lead exposure in early childhood to adolescent
health: A Chicago birth cohort. Am J Public Health 107: 1496-1501. HERO ID: 4591734.
http://dx.doi.org/10.2105/AJPH.2017.3039Q3.
World Health Organization (WHO). 2008. Harmonization Project Document No. 6. Part 1:
Guidance Document on Characterizing and Communicating Uncertainty in Exposure
Assessment. Available at:
http://www.who.int/ipcs/methods/harmonization/areas/exposure/en/.
Zartarian, V, Xue, J, Tornero-Velez, R, and Brown, J. (2017). Children's Lead Exposure: A
Multimedia Modeling Analysis to Guide Public Health Decision-Making. Environmental
Health Perspectives Vol 125, No 9.
Zhang, Yanhong. (2003). 100 Years of Pb Deposition and Transport in Soils in Champaign,
Illinois, U.S.A. Water, Air, and Soil Pollution. 146:197-210. HERO ID: 47833.
http://dx.doi.Org/10.1023/A:1023957226204
4-9
-------
Appendix: Cumulative Exposure Estimates for Different
Birth Cohorts
A-l
-------
SUMMARY
Blood Pb is an indicator of Pb that is distributing through an individual at a point in time,
and, accordingly, may best reflect recent exposures. It cannot, however, reflect an individual's
exposure history. Given the role of bone as a repository for Pb once in the body, bone Pb is
considered to provide a measure of an individual's cumulative lifetime exposure. Additionally,
where longitudinal blood Pb measurements are available, a cumulative blood Pb metric is also
informative regarding an individual's exposure across their lifetime. A cumulative blood Pb
metric may be derived in a manner conceptually similar to a cumulative exposure metric often
employed in worker studies (e.g., using ppm-days). The metric is derived essentially by
multiplying blood Pb measurement by years to which it is applied yielding an index in terms of
(|ig/dL) yrs (e.g., Nie et al., 2011 and Somervaille et al., 1988).
Given the backdrop of dramatic reductions in Pb exposures over the past four plus
decades, as evidenced in blood Pb levels of young children during that time, adults in the U.S.
population have significant Pb exposure histories. Further, adults of different ages may have
appreciably different exposure histories. This can be illustrated through use of a cumulative
blood Pb index (CBLI), a metric sometimes used as an exposure metric in epidemiological
studies of adult populations (2013 ISA, section 3.3.5; Nie et al., 2011; Hu et al., 2007;
Somervaille et al., 1988).
The CBLI is calculated as:
Where blood Pbi and blood Pbi+i are the blood Pb concentrations at two consecutive times ti and
ti+i and ti and tN are the beginning and ending time for the blood lead measurements (Nie et al.,
2011). Using NHANES blood Pb, CBLI was estimated for four birth cohorts (1945, 1970, 1990
and 2010). The CBLI estimates for different birth cohorts (See Tables 1-2 and 1-3 above)
illustrate the appreciably different Pb exposures that adults in today's population have
experienced and the difference of those exposures from adults of tomorrow.
NHANES data (Table A-l) are used to develop age-specific CBLI estimates for four
birth cohorts: 1945, 1976, 1990 and 2010 (Table A-2). Since 1976, the CDC has been monitoring
blood Pb levels nationally through the NHANES that provides nationally representative
biomonitoring data for Pb. NHANES is designed to assess the health and nutritional status of the
civilian noninstitutionalized U.S. population and is conducted by the National Center for Health
Statistics, part of the Centers for Disease Control and Prevention. NHANES conducts interviews
71
CBLI = 0.5 x (PbBj+1 + PbBi) x (ti+1 t{)
i=1
ANALYTICAL APPROACH
A-2
-------
and physical examinations with a nationally representative sample of approximately 10,000
people in each two-year year survey cycle. CDC's National Center for Environmental Health
measures concentrations of environmental chemicals in blood and urine samples collected from
NHANES participants. Summaries of the measured values are provided in the Fourth National
Report on Human Exposure to Environmental Chemicals (CDC, 2021).
Because the blood Pb monitoring began in 1976, there are data gaps (particularly for
years prior to the first NHANES) and surrogate assignments are employed. CBLIs are estimated
for all cohorts up to 2018 using the most recently available NHANES (2017-2018). The
projected CBLI estimates were also calculated for the years beyond 2028 for the 1990, 2010 and
1970 cohorts. In general, the following approaches were used to assign blood Pb levels each year
for the four birth cohorts:
For years prior to the 1st NHANES, the results for the relevant age(s) from the 1st
survey are used.
For years subsequent to the last NHANES, the results for the relevant age(s) from the
last survey are used.
For years intermediate to existing NHANES, there's a bridging of the results for the
two closest NHANES are used (see notes in Table A-2 below for specific details).
A similar approach used NHANES data (Table A-3) to estimate the CBLI for three Non-
Hispanic Black cohorts, born in 1990, 2000 and 2010 (Table A-4).
KEY LIMITATIONS AND UNCERTAINTIES
There are a number of inherent limitations and associated uncertainties in the analytical
approach described above (and in the notes section of Table A-2). Several of importance are
summarized here.
Reliance on the NHANES blood Pb level estimates presumes those surveys
appropriately reflect national average blood Pb measurements.
The use of first NHANES in assigning blood Pb levels in years before 1976
contributes uncertainty to the yearly blood Pb estimates and associated CBLI
estimates. The blood Pb assignments, and associated CBLI, for the 1945 and 1970
cohorts may be underestimated, to the extent exposures were higher in years prior to
1976 from the widespread use of leaded gasoline and other sources.
The use of readily available NHANES estimates, which were limited with regard to
separate age groups for adults, resulted in the use of NHANES reported estimates for
all adults 20 years or older, rather than for adult age subgroups. This will have
contributed uncertainty, with different bias for different ages.
The recently available publication by Wang et al. (2021) reports geometric means
from NHANES from 1999 to 2016 for birth cohorts of the 1940s, 50s, 60s, 70s , 80s,
and 90s. These reported values for the 1940s cohort indicate that the estimates applied
to the 1945 cohort in Table A-2 may be slight underestimates. This would contribute
A-3
-------
to a low bias for the 1945 cohort CBLI estimate for ages older than 50 years. Use of
Wang et al. (2021) in future iterations of this analysis may reduce some uncertainties
in CBLI estimates associated with assignments of blood Pb levels to older ages.
Given the current trajectory of declining blood Pb levels, the use of 2017-2018
NHANES blood Pb levels in assigning blood Pb levels for subsequent years in the
1970, 1990, and 2010 cohorts may contribute to overestimates in estimates of
exposure (and accordingly, CBLI estimates) for those years.
A-4
-------
Table A-l. Geometric Mean Blood Pb Levels (BLL) for the general population.
Year
1-2 yr
3-5 yr
1-5 yr
6-8 yr
9-11yr
6-11 yr
12-
14yr
15-
17yr
12-
19yr
18-
24yr
>20
20-
49yr
20-64yr
25-
34yr
35-
44yr
45-54yr
55-64yr
65-
74yr
>65
50-
69yr
>70
Women
15-49yrs
All
ages
1976-1980
median
15.7
14.9
15.2
13.1
12.3
12.7
11.4
12.1
13.1
13.7
14.6
15.3
14.6
14.4
10.37
1988-1991
4
3.3
3.6
2.5
2.3
2.4
1.6
2.6
4
4
1.85
1991-1994
3.1
2.5
2.7
2.1
1.8
1.9
1.5
2.1
3.4
3.4
1.53
1995-1998
2.7
2.2
2.3
1.8
1.6
1.7
1999-2002
2.2
1.8
1.9
1.4
1.3
1.4
1.05
1999-2000
2.23
1.51
1.1
1.75
1.66
2001-2002
1.7
1.25
0.942
1.56
1.45
2003-2006
1.8
1.5
1.6
1.2
1.1
1.1
0.91
2003-2004
1.77
1.25
0.946
1.52
1.45
2.07
1.43
2005-2006
1.46
1.02
0.797
1.41
1.31
2.1
1.29
2007-2010
1.5
1.2
1.3
1
0.8
0.9
0.81
2007-2008
1.51
0.988
0.8
1.38
1.3
1.91
1.27
2009-2010
1.17
0.838
0.68
1.23
1.16
1.71
1.12
2011-2016
0.9
0.8
0.8
0.7
0.6
0.6
0.61
2011-2012
0.97
0.681
0.554
1.09
1.02
1.6
0.973
2013-2014
0.782
0.567
0.506
0.967
0.858
2015-2016
0.758
0.571
0.467
0.92
0.82
2013-2016
2017-2018
0.67
0.475
0.411
0.855
0.753
GM (Geometric Mean) for all ages from NHANES and published reports. BLL values are color coded to match the corresponding reference.
Interpolated GMs averaged values for 1991-94 and 1999-2002.
GM for ages 1-5, 6-11,12-19, and 20+for the year 2017-2018 obtained from CDC Early Release of BLL from NHANES
GMs forage 1 thru 11 (4-yrperiods: 1976-80,1988-91,1991-94,1999-2002, 2003-05, 2007-10, 2011-16 6) from Egan eta!., 2021
GMs (2 yrperiods: 2003-2016) forages 20-64 and >=65 from Jain, 2016a and GM 1-19 yrs (2003-2012) Jain 2016b
Medians for ages 12-74 (1976-1980) from Mahaffey et a!., 1982
GMs forage 1-19 and > 20 (2-yrperiods) from CDC, 2018
GMs forage 12 to > 70 (1991-1994) from Pirkle eta!., 1998
GMs forage 12 to > 70 (1988-1991) from Brody et a!., 1994
GMs for women, ages 15-49 (4-yr periods) from Ettinger et al., 2020
A-5
-------
Table A-2. Estimated Cumulative CBLI based on NHANES GM estimates for four
general population cohorts.
2010 Cohort
1990 Cohort
1970 Cohort
1945 Cohort
Age
Year Age
GM
BLL
ug/dL
CBLI
ug/dL
¦yrs
Year Age
GM
BLL
ug/dL
CBLI
ug/dL
¦yrs
Year Age
GM
BLL
ug/dL
CBLI
ug/dL
¦yrs
Year Age
GM
BLL
ug/dL
CBLI
ug/dL
¦yrs
2010
0
0.81
1990
0
1.85
1970
0
10.37
1945
0
10.37
2011
1
0.97
0.9
1991
1
4
2.9
1971
1
15.7
13.0
1946
1
15.7
13.0
2012
2
0.97
1.9
1992
2
3.1
6.5
1972
2
15.7
28.7
1947
2
15.7
28.7
2013
3
0.8
2.7
1993
3
2.5
9.3
1973
3
14.9
44.0
1948
3
14.9
44.0
2014
4
0.8
3.5
1994
4
2.5
11.8
1974
4
14.9
58.9
1949
4
14.9
58.9
5
2015
5
0.8
4.3
1995
5
2.5
14.3
1975
5
14.9
73.8
1950
5
14.9
73.8
2016
6
0.6
5.0
1996
6
2.1
16.6
1976
6
13.1
87.8
1951
6
13.1
87.8
2017
7
0.475
5.6
1997
7
1.8
18.5
1977
7
13.1
100.9
1952
7
13.1
100.9
2018
8
0.475
6.1
1998
8
1.8
20.3
1978
8
13.1
114.0
1953
8
13.1
114.0
2019
9
0.475
6.5
1999
9
1.3
21.9
1979
9
12.3
126.7
1954
9
12.3
126.7
10
2020
10
0.475
7.0
2000
10
1.3
23.2
1980
10
12.3
139.0
1955
10
12.3
139.0
2021
11
0.475
7.5
2001
11
1.3
24.5
1981
11
12.3
151.3
1956
11
12.3
151.3
2022
12
0.411
7.9
2002
12
0.942
25.6
1982
12
11.4
163.2
1957
12
11.4
163.2
2023
13
0.411
8.3
2003
13
0.946
26.5
1983
13
11.4
174.6
1958
13
11.4
174.6
2024
14
0.411
8.7
2004
14
0.946
27.5
1984
14
6.5
183.5
1959
14
11.4
186.0
2025
15
0.411
9.2
2005
15
0.797
28.4
1985
15
1.6
187.6
1960
15
12.1
197.7
2026
16
0.411
9.6
2006
16
0.797
29.2
1986
16
1.6
189.2
1961
16
12.1
209.8
2027
17
0.411
10.0
2007
17
0.8
30.0
1987
17
1.6
190.8
1962
17
12.1
221.9
2028
18
0.411
10.4
2008
18
0.8
30.8
1988
18
1.6
192.4
1963
18
13.1
234.5
2029
19
0.411
10.8
2009
19
0.68
31.5
1989
19
1.6
194.0
1964
19
13.1
247.6
2030
20
0.411
11.2
2010
20
0.68
32.2
1990
20
1.6
195.6
1965
20
13.1
260.7
21
2031
21
0.411
11.6
2011
21
0.68
32.9
1991
21
1.6
197.2
1966
21
13.1
273.8
2032
22
0.411
12.0
2012
22
0.68
33.5
1992
22
1.6
198.8
1967
22
13.1
286.9
2033
23
0.411
12.4
2013
23
0.68
34.2
1993
23
1.6
200.4
1968
23
13.1
300.0
2034
24
0.411
12.9
2014
24
0.68
34.9
1994
24
1.6
202.0
1969
24
13.1
313.1
2035
25
0.411
13.3
2015
25
0.68
35.6
1995
25
1.6
203.6
1970
25
13.7
326.5
2036
26
0.411
13.7
2016
26
0.68
36.3
1996
26
1.6
205.2
1971
26
13.7
340.2
2037
27
0.411
14.1
2017
27
0.68
36.9
1997
27
1.6
206.8
1972
27
13.7
353.9
2038
28
0.411
14.5
2018
28
0.68
37.6
1998
28
1.6
208.4
1973
28
13.7
367.6
2039
29
0.411
14.9
2019
29
0.68
38.3
1999
29
1.6
210.0
1974
29
13.7
381.3
30
2040
30
0.411
15.3
2020
30
0.68
39.0
2000
30
1.6
211.6
1975
30
13.7
395.0
2041
31
0.411
15.7
2021
31
0.68
39.7
2001
31
1.56
213.2
1976
31
13.7
408.7
2042
32
0.411
16.1
2022
32
0.68
40.3
2002
32
1.56
214.7
1977
32
13.7
422.4
2043
33
0.411
16.6
2023
33
0.68
41.0
2003
33
1.45
216.2
1978
33
13.7
436.1
2044
34
0.411
17.0
2024
34
0.68
41.7
2004
34
1.45
217.7
1979
34
13.7
449.8
2045
35
0.411
17.4
2025
35
0.68
42.4
2005
35
1.31
219.1
1980
35
14.6
464.0
2046
36
0.411
17.8
2026
36
0.68
43.1
2006
36
1.31
220.4
1981
36
14.6
478.6
2047
37
0.411
18.2
2027
37
0.68
43.7
2007
37
1.3
221.7
1982
37
14.6
493.2
2048
38
0.411
18.6
2028
38
0.68
44.4
2008
38
1.3
223.0
1983
38
14.6
507.8
A-6
-------
2010 Cohort
1990 Cohort
1970 Cohort
1945 Cohort
Age
GM CBLI
BLL kig/dL
Year Age ug/dL yrs
GM CBLI
BLL kig/dL
Year Age ug/dL yrs
GM CBLI
BLL kig/dL
Year Age ug/dL yrs
GM CBLI
BLL kig/dL
Year Age ug/dL yrs
2049 39 0.411 19.0
2029 39 0.68 45.1
2009 39 1.16 224.2
1984 39 8.6 519.4
2050 40 0.411 19.4
2030 40 0.68 45.8
2010 40 1.16 225.4
1985 40 2.6 525.0
2051 41 0.411 19.8
2031 41 0.68 46.5
2011 41 1.02 226.5
1986 41 2.6 527.6
2052 42 0.411 20.3
2032 42 0.68 47.1
2012 42 1.02 227.5
1987 42 2.6 530.2
2053 43 0.411 20.7
2033 43 0.68 47.8
2013 43 0.967 228.5
1988 43 2.6 532.8
2054 44 0.411 21.1
2034 44 0.68 48.5
2014 44 0.967 229.4
1989 44 2.6 535.4
2055 45 0.411 21.5
2035 45 0.68 49.2
2015 45 0.92 230.4
1990 45 2.6 538.0
2056 46 0.411 21.9
2036 46 0.68 49.9
2016 46 0.92 231.3
1991 46 2.1 540.3
2057 47 0.411 22.3
2037 47 0.68 50.5
2017 47 0.855 232.2
1992 47 2.1 542.4
2058 48 0.411 22.7
2038 48 0.68 51.2
2018 48 0.855 233.0
1993 48 2.1 544.5
2059 49 0.411 23.1
2039 49 0.68 51.9
2019 49 0.855 233.9
1994 49 2.1 546.6
50
2060 50 0.411 23.5
2040 50 0.68 52.6
2020 50 0.855 234.8
1995 50 2.1 548.7
2021 51 0.855 235.6
1996 51 2.1 550.8
2022 52 0.855 236.5
1997 52 2.1 552.9
2023 53 0.855 237.3
1998 53 2.1 555.0
2024 54 0.855 238.2
1999 54 1.75 557.0
2025 55 0.855 239.0
2000 55 1.75 558.7
2026 56 0.855 239.9
2001 56 1.56 560.4
2027 57 0.855 240.7
2002 57 1.56 561.9
2028 58 0.855 241.6
2003 58 1.45 563.4
2029 59 0.855 242.4
2004 59 1.45 564.9
2030 60 0.855 243.3
2005 60 1.45 566.3
2031 61 0.855 244.2
2006 61 1.45 567.8
2032 62 0.855 245.0
2007 62 1.45 569.2
2033 63 0.855 245.9
2008 63 1.45 570.7
2034 64 0.855 246.7
2009 64 1.405 572.1
2035 65 0.855 247.6
2010 65 1.36 573.5
2036 66 0.855 248.4
2011 66 1.36 574.9
2037 67 0.855 249.3
2012 67 1.36 576.2
2038 68 0.855 250.1
2013 68 1.36 577.6
2039 69 0.855 251.0
2014 69 1.36 578.9
2040 70 0.855 251.9
2015 70 1.36 580.3
2041 71 0.855 252.7
2016 71 1.36 581.7
2042 72 0.855 253.6
2017 72 1.36 583.0
2043 73 0.855 254.4
2018 73 1.36 584.4
2044 74 0.855 255.3
2019 74 1.36 585.7
__7§| _Age 75
2045 75 0.855 256.1
2020 75 1.36 587.1
A-7
-------
* Approach:
Age 0 is assigned maternal BLL (women age 15-49 from Ettingeretai., 2020) for relevant time period
For ages less than 20, cohorts are generally assigned BLLs for their age group and year.
Red font indicates variation from this approach. Variations include the following:
A - For ages 20 through 30, for years subsequent to 1989, the NHANES value for ages 12-19 yr is used. This is because
the next age group (>=20) includes ALL ages greater than 20 (with multiple gas Pb exposure years) and thus would overestima
BLLs for twenty-somethings. Given the declining exposures backdrop, this approach is not likely to bias estimates.
B - When year is prior to NHANES, the cohort is are assigned BLLs from 1st NHANES according to age group.
C - For ages less than 20 occurring in years subsequent to the most recent NHANES (2018), the last NHANES value
for the relevant age group is assigned (also see note A).
D - Forages >=20 occurring in years after the most recent NHANES (2018), the last NHANES for that age is retained
for subsequent years. Given the current downward BLL trend, this approach is not believed to underestimate BLLs
and may lend an upward bias.
E - For years during the NHANES sampling period (1976-2018) for which age-specific values are uncertain, interpolation (red itc
is applied consistent with understanding that with constant (or declining) exposure, an individual's BLL would not
increase as they age (we note that NHANES generally shows higher adult [vs teen] BLLs due to older individuals with
higher past BLLs in that group). Where values are needed to bridge years between two NHANES, the central year
between the two estimates is assigned the average of the two values, the prior value is retained for prior years, and
the subsequent value is assigned for subsequent years. More specifically, in time periods between NHANES for which
we lack estimates for age/year combo (e.g., 1945 birth cohort, ages 36-43), we retain the last estimate for
the first half of period, take next match for latter half of period and use average of two for midpoint.
F - For 1970 birth cohort, BLLs for ages 22-30 (years 1992-2000) is set to 1.6, as interpolation toward the 2001 (age 31)
BLL from 2001-02 NHANES.
G - For 1945 birth cohort, BLLs for ages 60-68, method of note E is applied to bridge from BLL for >=20yr-olds in 2003-4
to >=61yr-olds in 2013-2016.
A-8
-------
Table A-3. Geometric Mean Blood Pb Levels (BLL) for Non-Hispanic Black Populations
(NHB) from NHANES.
NHB
NHB
NHB
NHB
NHB
NHB,
NHB
NHB
Childre
Childre
Childre
Childre
NHB,
NHB,
NHB,
NHB,
women
All
Men
Women
Years
n, 6 mo-
n, 1-5
n, 6-11
n,6-
18-
20-
> 65yr
> 15yr
,15-49
Age
> 15yr
> 15yr
5yr
yr
yr
16yr
74yr
64yr
yr*
s
1976-80
20
20.3
16.2
14
15
11.52
15.7
(Med)
.9
.8
.5
1988-1991
5.2
3.9
2.16
1991-1994
4.3
3
1.75
2.8
1999-2002
2.8
2
1.21
1999-2000
3.08
1.87
2001-2002
2.59
1.65
2003-2006
2.4
1.6
1.03
2003-2004
2.73
1.89
1.63
2.8
1.69
2005-2006
2.12
1.35
1.39
2.46
1.39
2007-2010
1.8
1.2
0.92
2007-2008
1.94
1.31
1.39
2.33
1.39
2009-2010
1.59
1.15
1.27
2.02
1.24
2011-2016
1.1
0.8
0.61
2011-2012
1.29
0.9
1.04
1.85
0.998
2013-2014
0.98
0.871
2015-2016
0.89
0.856
2013-2016
2017-2018
0.766
2015-2018
GMs for all ages from NHANES Reports and BLL values are color coded to match the corresponding reference.
GMs forage 1 thru 11 (4-yrperiods: 1976-80,1988-91,1991-94,1999-2002, 2003-05, 2007-10,2011-16) fromEgan eta!.,
2021
GMs (2 yrperiods: 2003-2016) forage 20-64 & >=65 from Jain 2016a and GM1-19 yro (2003-2012) Jain 2016b
Medians for age 12-74 (1976-1980) from Mahaffey et a!., 1982 NEJM
GM for age 1-5, 6-11,12-19, and 20+ for the year 2017-2018 obtained from CDC Early Release of BLL from NHANES
(CAS# 7439-92-1)
GMs for women, 15-49 (4-yr periods) from Ettinger et al., 2020
GMs forage 1-19 & >=20 (2-yrperiods) from CDC, 2018
GMs forage 1-5 from 1999-2016 from Teye etal., 2022
GMs forage 12 to GE70 (1991-1994) from Pirkle etal 1998
A-9
-------
Table A-4.Estimated Cumulative Blood Lead Index (CBLI) based on Non-Hispanic Black
Populations NHANES GM estimates in Table 3.
2010 Cohort
2000 Cohort
1990 Cohort
Age
GM CBLI
Year age BLL (ug/dL)yrs
Year
age
GM CBLI
BLL (ug/dL)-yrs
Year
age
GM
BLL
CBLI
(ug/dL)yrs
0
2010
0
0.92
2000
0
1.21
1990
0
2.16
2011
1
1.1
1.0
2001
1
2.59
1.9
1991
1
5.2
3.7
2012
2
1.1
2.1
2002
2
2.59
4.5
1992
2
4.3
8.4
2013
3
1.1
3.2
2003
3
2.73
7.2
1993
3
4.3
12.7
2014
4
1.1
4.3
2004
4
2.73
9.9
1994
4
4.3
17.0
5
2015
5
1.1
5.4
2005
5
2.12
12.3
1995
5
4.3
21.3
2016
6
0.8
6.4
2006
6
1.89
14.3
1996
6
3.2
25.1
2017
7
0.8
7.2
2007
7
1.2
15.9
1997
7
3.2
28.2
2018
8
0.8
8.0
2008
8
1.31
17.1
1998
8
2
30.8
2019
9
0.8
OO
CO
2009
9
1.15
18.3
1999
9
2
32.8
10
2020
10
0.8
9.6
2010
10
1.15
19.5
2000
10
2
34.8
A-10
-------
REFERENCES
Brody, DJ, Pirkle, JL, Kramer, RA, Flegal, KM, Matte, TD, Gunter, EW, and Paschal, DC (1994). Blood
lead levels in the US population: Phase 1 of the third National Health and Nutrition Examination
Survey (NHANES III, 1988 to 1991). JAMA 272: 277-283. HERO ID: 50642.
http://dx.doi.org/10.1001/iama.1994.03520040039Q38.
CDC (2018). Fourth National Report on Human Exposure to Environmental Chemicals, Updated Tables,
March 2018, Volume One. Available at: https://www.cdc.gov/exposurereport/data_tables.html
CDC (2021) National Report on Human Exposure to Environmental Chemicals (Report). Available at:
http s: //www, cdc. gov/nceh/1 ead/data/ nhanes. htm.
Egan, KB, Cornwell, CR, Courtney, JG, and Ettinger, AS (2021). Blood lead levels in U.S. children ages 1-
11 years, 1976-2016. Environ Health Perspect 129: 37003. HERO ID: 10197483.
http://dx.doi.org/10.1289/EHP7932
Ettinger, AS, Egan, KB, Homa, DM, and Brown, MJ (2020). Blood lead levels in U.S. women of
childbearing age, 1976-2016. Environ Health Perspect 128: 17012. HERO ID: 6686085.
http://dx.doi.org/10.1289/EHP5925.
Jain, RB (2016a). Trends and variability in blood lead concentrations among US adults aged 20-64 years and
senior citizens aged >65 years. Environ Sci and Pollut Res 23: 14056-14067. HERO ID: 3985736.
http ://dx. doi. org/10.1007/s 11356-016-6583-7
Jain, RB (2016b). Trends and variability in blood lead concentrations among US children and adolescents.
Environ Sci and Pollut Res 23: 7880-7889. HERO ID: 4683874 http://dx.doi.org/10.10Q7/sll356-
016-6039-0
Mahaffey, KR, Annest JL, Roberts, J, and Murphy, RS (1982). National estimates of blood lead levels:
United States, 1976-1980. NEJM 307: 573-579. HERO ID: 7110203.
http://dx.doi.org/10.1056/NEJM198209023071Q01
Nie, LH, Wright, RO Bellinger, DC, Hussain, J, Amarasiriwardena, C, Chettle, DR, Pejovic-Milic, A,
Woolf, A, and Shannon, M (2011). Blood lead levels and cumulative blood lead index (CBLI) as
predictors of late neurodevelopment in lead poisoned children. Biomarkers 16: 517-24. HERO ID:
787265. http://dx.doi.org/10.3109/1354750X.2011.6Q4133
Pirkle, JL, Kaufmann, RB, Brody, DJ; Hickman, T, Gunter, EW, and Paschal, DC (1998). Exposure of the
U.S. population to lead, 1991-1994. Environ Health Perspect 106: 745-750. HERO ID: 51085.
http://dx.doi.org/10.1289/ehp.98106745
Somervaille, LJ, Chettle, DR, Scott, MC, Tennant, DR, McKiernan, MJ, Skilbeck, A, and Trethowan, WN
(1988). In vivo tibia lead measurements as an index of cumulative exposure in occupationally
exposed subjects. British J Ind Med 45: 174-181. HERO ID: 57784.
https://www.ncbi.nlm.nih.gov/pubmed/3348993
Wang, T; Zhou, YP; Sun, Y; Zheng, YX. (2021). Trends in blood lead levels in the U.S. from 1999 to 2016.
Am JPrevMed 60: el79-el87. http://dx.doi.org/10.1016/j.amepre.2020.10.024.
A-ll
-------
Teye, SO, Yanosky, JD, Cuffee, Y, Weng, X, Luquis, R, Farace, E and Wang, L (2021). Exploring persistent
racial/ethnic disparities in lead exposure among American children aged 1-5 years: Results from
NHANES 1999-2016. Int Arch Occup and Environ Health 94: 723-730. HERO ID: 7301982.
http ://dx. doi .org/10.1007/s00420-020-01616-4
A-12
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-452/R-22-003c
Environmental Protection Health and Environmental Impacts Division May 2023
Agency Research Triangle Park, NC
A-13
------- |