United States
Environmental Protection
Agency
EPA Handbook for Use of
Data from the National
Health and Nutrition
Examination Surveys
(NHANES): A Goldmine of
Data for Environmental
Health Analyses
-------
-------
EPA/600/R-02/044
March 2003
www.epa.gov/ncea
EPA Handbook for Use of Data from the National
Health and Nutrition Examination Surveys
(NHANES): A Goldmine of Data for Environmental
Health Analyses
National Center for Environmental Assessment-Washington Office
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
/T~V Recycled/Recyclable
^C/a-U Printed .with vegetable-based ink on
'
-------
DISCLAIMER j
This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for usp.
I
[ : . ' .
' i
ABSTRACT |
1 i
This Handbook provides descriptive background information and general guidance on
how to access and use data from the National Health and Nutrition Examination Surveys
(NHANES). This is an enormous human database that cai> be used to develop information
suitable for use in risk assessments, and to support regulatory and policy needs of EPA. For
more than 30 years, EPA has been one of many collaborating agencies that help plan and support
funding of data collection through NHANES. Because only a limited number of Agency
managers and staff are aware of the content and availability of this rich database, this Handbook
was developed to familiarize staffs with NHANES and footer increased use of the data to support
EPA needs. Despite the limitations and complex design of this survey, it is clear that NHANES
is a unique rich database that offers a tremendous amount of human health, nutrition, and
exposure information, and will continue to do so into the .future. It is hoped that by informing
staff about NHANES, this Handbook will encourage efforts to "mine" the data to support
Agency needs across the program offices. It is also hoped that innovative approaches (e.g., using
geographic information systems; linking NHANES to available databases such as the National
Death Index), will be developed to analyze the data in ne* ways that produce information that is
useful to the mission of the Agency. Now that the National Center for Health Statistics (NCHS)
has established their Research Data Center, it should be possible to conduct studies that were
impossible in the past because of lack of access to sensitive data. Finally, more thought should
be given to designing and conducting studies that make use of subjects' biological samples
(blood urine, saliva) stored by NCHS. These samples of fer a rare opportunity to study potential
biomarkers of exposure and/or effects on a national sample of the U.S. population and link the
data to health, nutrition, exposure and socioeconomic data collected in the baseline surveys.
!
uf Snvko^enS Protection Agency (EPA). (2003) Handbook for use of data from the national health
and nutrition examination surveys (NHANES). National Center for Environmental Assessment,
Washington, DC: EPA/600/R-02/044. Available from: National Technical Information Service,
Springfield, VA; PB2Q03-104276, and .
11
-------
TABLE OF CONTENTS
Authors; Contributors, and Reviewers
Acronyms and Abbreviations
. vi
viii
1.0. PURPOSE OF THIS HANDBOOK
2.0. EPA NHANES USERS GROUP ..
.1
2
3.0. OVERVIEW AND BACKGROUND OF NHANES 2
3.1. Followup Studies for the Baseline NHANES '.'.'.'.'.'. 6
3.2. How NHANES have been used to Improve Public Health 7
4.0. DETAILED INFORMATION ON NHANES CONTENT 8
5.0 HOW TO OBTAIN NHANES DATA AND ISSUES WITH FILE STRUCTURE AND
CONTENT '..., 1Q
5.1. Issues with Data File Structure and Content u
5.2. NCHS Research Data Center (RDC) 13
6.0. TYPES OF ANALYSES THAT CAN BE CONDUCTED WITH NHANES DATA 15
6.1. Studies using NHANES Data that have Environmental Relevance 17
6.2. Use ofNHANES Data at EPA '.'.'.'.'.'. 17
7.0. USE OF NHANES STORED BIOLOGICAL SPECIMENS 19
7.1. Serum Specimens i o
7.2. DNA Samples 19
8.0. LIMITATIONS AND PRECAUTIONS ON USE OF NHANES DATA ,... 20
9.0. ANALYSIS ISSUES
25
9.1. Weighting ,.; ~
9.2. Issue of Nonresponse Bias og
10.0. SUMMARY AND CONCLUSIONS 29
APPENDDTI
APPENDKII
APPENDIX HI
APPENDDCIV
NHANES-IH Sample Design and Analysis Guidelines I-l
NHANES-in Health Status Assessment '.'.'.'.'.'.'.'. H-l
NHANES-III Nutritional Status Assessment .!.... III-l
NHANES-III Summary of Examination & Clinical Biochemistry
Assessments and Dietary/Nutritional Evaluations IV-1
m
-------
APPENDIX V
APPENDIX VI
APPENDIX VH
APPENDIX VIE
APPENDIX DC
REFERENCES
TABLE OF CONTENTS (continued)
se;3
Comparison of Laboratory Analyse!
NHANES-I, -H, -HI, and Hispanic "
Comparison of Laboratory Analysi
Across NHANES-99, -00, -01, and -02
Comparison
Annotated Bibliography of Studies
Summaries of Recent EPA Studies
i of Environmental Date Collected Across
of Blood and Urine Across
HANES V-l
:, Examinations and Questionnaires
VI-1
allNHANES VH-1
UsingNHANES Data VHI-1
Using NHANES Data K-l
IV
R-l
-------
LIST OF TABLES
Table 1. Summary of Selected Design Features of the NHANES ;. 30
Table 2. Summary of the NHANES-I EpidemiologicalFollowup Studies 31
Table 3. Schedule of Survey Planning Activities and Release of Data for NHANES-99- 08 . .32
-------
AUTHORS, CONTRIBUTORS, Al^D REVIEWERS
The National Center for Environmental Assessment-Washington Office within EPA's
Office of Research and Development was responsible for 'the preparation of this Handbook.
AUTHOR
Susan Berlin, ScD.
U.S. Environmental Protection Agency
National Center for Environmental Assessment-W
Washington, DC 20460
CONTRIBUTORS TO APPENDIX IX
Elmer AMn
U.S. Environmental Protection Agency Region 4,
Waste Management Div.
Atlanta, GA
Ruth Allen
U.S. Environmental Protection Agency
Office of Pesticide Programs
Office of Prevention, Pesticides and Toxic Substances
Washingtoii, DC 20460 '
Denis Bonim
U.S. Environmental Protection Agency
Office of Science and Technology
Office of Water
Washington, DC 20460
Rebecca Calideron
U.S. Environmental Protection Agency
Human Studies Division
National Health & Environmental Effects Research Lab
Research Triangle Park, NC .'
Robert Chapman
U.S. Environmental Protection Agency
National Center for Environmental Assessment-RTF
Research Triangle Park, NC
Chuck French
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Office of Air and Radiation
Research Triangle Park, NC
Elizabeth Hilborn
U.S. En\|งonmentil Protection Agency
Human Studies Division
National|Health & Environmental Effects Research Lab
Research Triangle Park, NC
I
Helen Jacobs
U.S. Enwonmental Protection Agency
Office off Science and Technology
Office off Water
Washington, DC 20460
Thomas McCurdy
U.S. Environmental Protection Agency
National Exposure Research Laboratory
Research Triangle Park, NC
i
Ron Moirony
U.S. Environmental Protection Agency
Office of Pollution, Prevention and Toxics
Office of Prevention, Pesticides and Toxic Substances
Washington, DC 20460
Jacqueline Moya
U.S. Environmental Protection Agency
National! Center for Environmental Assessment-W
Washington, DC 20460
David Otto
U.S. Enyironmental Protection Agency
Human Studies Div.
Health &. Environmental Effects Research Lab
Researcli Triangle Park, NC
t
i
James Quackenboss
U.S. Environmental Protection Agency
National Exposure Research Laboratory
Office of Research and Development
Las Vegas, NV
VI
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Dina Schreinemachers
U.S. Environmental Protection Agency
Human Studies Division . .
National Health & Environmental Effects Research Lab
Research Triangle Park, NC
Brad Schultz
U.S. Environmental Protection Agency
Office of Pollution, Prevention and Toxics
Office of Prevention, Pesticides and Toxic Substances
Washington, DC 20460
John Schwemberger
U.S. Environmental Protection Agency
Office of Pollution, Prevention and Toxics.
Office of Prevention, Pesticides and Toxic Substances
Washington, DC 20460
Sherry Selevan
U.S. Environmental Protection Agency
National Center for Environmental Assessment-W
Washington, DC 20460
Marc Stifelman
U.S. Environmental Protection Agency, Region 10
Office of Environmental Assessment
Seattle, WA
Amina Wilkins
U.S. Environmental Protection Agency
National Center for Environmental Assessment-W
Washington, DC.20460
REVIEWERS
David Cleverly
U.S. Environmental Protection Agency
National Center for Environmental Assessment-W
Washington, DC 20460
Charles Dillon
National Center for Health Statistics
Hyattsville, MD 20782
Susan Schober
National Center for Health Statistics
Hyattsville, MD 20782
Wilbur Hadden
National Center for Health Statistics
Hyattsville, MD 20782
Sherry Selevan
U.S. Environmental Protection Agency
National Center for Environmental Assessment-W
Washington, DC 20460
Vll
-------
I
ACRONYMS AND ABBREVIATIONS
BP-Blood pressure j
Cd - Cadmium
CDC- Centers for Disease Control and Prevention
CNS- Central nervous system
CO - Carbon Monoxide. One of the six air pollutants the EPA regulates under NAAQS.
DEHLS - Division of Environmental Health Laboratory Sciences
DHHS - Department of Health and Human Services
EKG - Electrocardiogram
EPA - Envkonmental Protection Agency j
FEV1 - Forced Expiratory Volume in the first second j
FVC - Forced Vital Capacity
GIS - Geographical Information Systems
Hg - Mercury
HHANES - Hispanic Health and Nutrition Examination Siiirvey
1KB - Institutional Review Board at NCHS i
MEC - Mobile Examination Center j
NAAQS- National Ambient Air Quality Standard pollutants, which include ozone, sulfur
dioxide, nitrogen oxide, particulate matter, lead and carbon monoxide.
NCEA - National Center for Environmental Assessment. This is an EPA office.
NCEH-National Center for Environmental Health. This is a CDC lab.
NCHS - National Center for Health Statistics. This is a CDC office.
NDI-National Death Index : .
NH2MS- National Health and Nutrition Examination Survey-II Mortality Study
NHANES - National Health and Nutrition Examination Survey
NHDS - National Hospital Discharge Survey
NHEFS - NHANES Epidemiological Followup Study i.
NHES - National Health and Examination Survey ;
NHIS - National Health Interview Survey ,
NIOSH - National Institute of Occupational Safety and Health
NOx - Nitrogen oxides. ' i
NSFG-National Survey of Family Growth j
NTIS - National Technical Information Service I
O3 - Ozone. One of the six air pollutants the EPA regulates under the NAAQS.
OGGT - Oral Glucose Tolerance Test
ORD - Office of Research and Development. This is an EPA office.
Pb - Lead. One of the six air pollutants the EPA regulates under the NAAQS.
PbB - Blood lead , !
PM10 - PM10 is defined as particulate matter (PM) with a! mass median aerodynamic diameter
less than 10 micrometers (urn). One of the six air pollutants the EPA regulates under the
NAAQS. j
PSU - Primary Sampling Unit, usually a county !
i '
f: ::,
Vlll
-------
ACRONYMS AND ABBREVIATIONS (continued)
QA/QC - Quality Assurance/Quality Control
RDC- Research Data Center.. This is a facility in NCHS.
SEQN - Sample Sequence Number. This is the unique identification number given to each
NHANES subject.
SES - Socioeconomic status
SOx - Sulfur oxides
SP - Sample Person in NHANES
USDA - U.S. Department of Agriculture
VOCs - Volatile organic compounds
IX
-------
IIP]
, :
:! f? ;
II
1( ,;:'
1;; *"
1
1 i !'
-------
1.0 PURPOSE OF THIS HANDBOOK
The U.S. Environmental Protection Agency's (EPA's) National Center for Environmental
Assessment (NCEA), in the Office of Research and Development (ORD) has developed this
Handbook to provide descriptive background information and general guidance on how to access
and use data from the National Health and Nutrition Examination Surveys (NHANES). This is
an enormous human database that can be used to develop information suitable for use in risk
assessments, and to support regulatory and policy needs of EPA. For more than 30 years, EPA
has been one of many collaborating agencies that help plan and support funding of data
collection through NHANES. Because only a limited number of Agency managers and staff are
aware of the content and availability of this rich database, NCEA developed this Handbook to
familiarize staffs with NHANES and foster increased use of the data to support EPA needs. This
Handbook will be disseminated throughout the Agency via the EPA intranet.
This Handbook is not be a treatise on how to conduct an epidemiology study using
NHANES data. It is assumed that the audience is composed of epidemiologists, statisticians and
analysts familiar with appropriate research methods and how to interpret the findings, but
unfamiliar with the purpose, content, limitations and potential usefulness of NHANES for
supporting Agency risk assessment, policy and regulatory needs. This Handbook will also be
useful to EPA managers to provide them with an overview of the survey. This Handbook
provides the following: purpose of the Handbook and the EPA NHANES Users Group (Sections
1.0 and 2.0); summary of the history and content of ISfflLANES (Sections 3.0 and 4.0; Appendices
II through VII); overview of how the surveys are conducted and how to obtain the data (Sections
3.0,4.0 and 5.0); types of analyses that can be conducted with NHANES data and can support
EPA risk assessment, policy and regulatory needs (Section 6.0; Appendices VIII and IX);
availability and potential use of stored NHANES biological samples to support EPA needs
(Section 7.0); discussion of the major limitations to use of NHANES data (i.e., can not perform
local studies; issues of confidentiality) and some precautions on use and interpretation of study
results (Section 8.0); discussion of issues with analysis of NHANES data, such as the need for
weighting, and compensating for nonresponse bias (Section 9.0).
From this Handbook, the reader should gam a basic understanding of what data are
available through NHANES; how to obtain the data; if the data are potentially suitable for
supporting the needs of his/her office; key limitations of the data; and what types of analyses are
possible. As indicated in Sections 3 and 6, and Appendices VIII and IX, NHANES data can be
used to support a variety of environmental applications, such as: 1) evaluation of health effects
associated with exposure to certain pollutants; 2) estimation of the levels of certain pollutants
(e.g., lead, mercury, pesticides) hi the general population based on chemical biomarkers hi blood
and/or urine; 3) establishment of population reference standards for various physiological
parameters (e.g., height and weight, lung function) and; 4) examination of how personal risk
factors (e.g., smoking, poor nutrition, obesity, poverty) interact with chemical exposures to affect
health. Based on the information provided hi the summaries of EPA studies that have used
NHANES data (Appendix IX), the reader can also contact EPA scientists involved with specific
-------
studies to gain more insight into issues surrounding use of; these data.
2.0. EPA NHANES USERS GROUP
EPA has an NHANES Users Group for anyone in the Agency who is interested in using
the survey data. The Group has monthly conference calls ^vhere ideas are shared; members help
each other with analytical problems; and everyone is kept apprized of issues and data collection
activities of the on-going survey, NHANES99+. If you ate interested in being included on the e-
mail list and joining the Group, please contact Susan Perlin at perlin.susan@epa.gov.
3.0 OVERVIEW AND BACKGROUND OF NHANES
The National Center for Health Statistics (NCHS),: which is part of the Centers for
Disease Control and Prevention (CDC), has been collecting data on the health and nutrition
status of the U.S. population for many decades. In 1956 the National Health Survey Act was
passed, authorizrng establishment of a continuous survey to provide current statistical data on the
amount, distribution, and effects of illness and disability in the United States. Under this Act,
data are to be obtained from at least three sources: persomil interviews; clinical tests,
measurements, and physical examinations; and from medipal care facilities. Since passage of the
Act, NCHS has conducted seven major surveys, resulting in extensive, publicly available
databases containing varying amounts of information on hlealth effects, nutrition, and
environmental exposures from representative samples of the U.S. population.
.
i
The National Health Examination Survey (NHES); was the first survey resulting from the
Act. Within the first decade, three NHES surveys, each with approximately 7,500 subjects, were
conducted: I .
i!
NHES I: (1960-62). Focused on selected chronic diseases of adults 18 to 79 years old.
NHES II: (1963-65). Focused on growth and development of children 6 tol 1 years old.
NHES HI: (1966-70). Focused on growth and development of children 12 to 17 years old.
In the 1970s, with the discovery of the link between nutrition and certain diseases, the NHES
was expanded to include collection of nutritional information, and the name of the survey was
changed to the National Health and Nutrition Examination Survey (NHANES) to reflect this
expansion. The EPA is one of many federal agencies that collaborates with NCHS to support
NHANES. Other collaborating agencies include: Centers |for Disease Control and Prevention,
Food and Drug Administration., National Institutes of Health, National Institutes of Mental
Health, National Institute for Environmental Health Scienbes, Health Resources and Services
Administration, Agency for Toxic Substances and Diseas^ Registry, National Institute of
Occupational Safety and Health, and the Social Security Administration.
il
> I
i
! '
-------
In addition to the NHES and NHANES, NCHS also collects human data through a variety
of other surveys, including the National Health Interview Survey (NfflS), the Survey of Family
Growth (NSFG), National Hospital Discharge Survey, and the National Death Index (NDI). This
Handbook pertains only to the NHANES surveys. For additional information on the history of
NHANES, see . For more
information on other surveys conducted by NCHS, see .
Five major goals of NHANES are to provide the following:
1) national population reference distributions of selected health parameters (i.e., height,
weight, cholesterol levels);
2) national prevalence data on diseases, functional impairment, and risk factors (i.e., heart
disease, respiratory diseases, smoking, exposure to environmental pollutants);
3) information on secular changes in selected diseases and risk factors;
4) information to help understand disease etiology; and
5) information for investigating the natural history of selected diseases.
The last two goals of understanding disease etiology and the natural history of selected diseases
are to be met through planned follow-up surveys of cohorts of initial respondents from each of
the surveys (See Section 3.1) (NCHS, 1992b).
Since 1971 there have been four discrete NHANES (see Table 1). The latest survey,
NHANES99+, went into the field hi March 1999 and differs from the previous surveys in that it
will collect annual data continuously into the future. Since NHANES will be in the field
continuously, there is also a new naming convention to indicate the specific year of the survey
(NHANES-99, NHANES-00, NHANES-01, etc.) When referring to the current NHANES hi
general, without reference to a specific year, it is called NHANES99+.
NHANES uses a complex, stratified, multistage, probability cluster design to select a
representative sample of the noninstitutionalized, civilian U.S. population. A four-stage sample
design is used, as follows: 1) Primary Sampling Unite (PSUs) comprising mostly counties; 2)
area segments within PSUs, 3) households within area segments, and 4) persons within
households. It is beyond the scope of this Handbook to provide a detailed discussion of this
complicated survey design. See Appendix I for information on the sample design of NHANES-
III, which is similar to the design for all the NHANES. The reader also is referred to the
following references for more complete discussions of the NHANES sample design: for
NHANES-I, see NCHS (1973,1977,1978); for NHANES-II, see NCHS (1981); for HHANES
see NCHS (1985); and for NHANES-III see NCHS (1994b, 1996c).
As noted hi Table 1, each NHANES oversamples certain population subgroups.
Oversampling is conducted for people hi specific age/race/ethnicity/socioeconomic subgroups to
help ensure there will be sufficient numbers of subjects to support valid analyses. The following
excerpt from (NCHS, 1996c) helps, to explain why mere is a need to oversample certain
subgroups:'
-------
I "I
"Older persons, children, Mexican-Americans, and black persons were oversampled in
NHANES-in to insure a prespecified rninirnum sample size for each analytic domain so
that estimates of the health and nutrition status of persons hi each domain could be made
wiHi acceptable precision. The oversampling in MHANES-III was part of a pattern
established in the sample design. The population was decomposed into 52 subdomains: 7
age groups by sex for black and Mexican-American persons and 12 age groups by sex for
white persons and other racial groups combined, lifter defining these age-sex-
race/ethnicity subdomains, variable sampling rates; were derived to ensure the
achievement of sample sizes sufficient to permit analyses of the data for each
subdomain." I
:
NHANES is a valuable resource because it can support examination of public health
issues that can best be addressed through physical examinations and laboratory tests. Each
NHANES has a consistent core set of tests (e.g., height, weight, blood cholesterol) and questions
(e.g., annual family income, race/ethnicity of subject) that! are designed to assess the overall
health and nutritional status of the subject and evaluate certain variables (e.g., socioeconomic
status, SES) known to affect health and nutrition. Over th^ years, these core components have
increased in number. During the planning phase of each ^JHANES, the collaborating agencies
submit research proposals to NCHS requesting specific tests and/or questions to be administered
to the subjects hi addition to the core components. For example, hi NHANES-III, EPA and the
National Institute of Occupational Safety and Health (NIC>SH) requested and had approval for
lung function tests (e.g., spirometry) of subjects 8 years arid older. Unlike the core components,
agency-specific components will be retained hi the survey; only as long as there is mutual
agreement between NCHS and the requesting funding agency to do so, and funding is available.
Guidelines for preparing a proposal for submission to NCHS can be found on the following web
site: . These guidelines were for proposals for
the NHANES to be conducted hi 2002, but are appropriate for future years of the survey.
( , . i
All NHANES are cross-sectional surveys that use a complex, multistage, stratified design
with cluster sampling to obtain a probability sample of pebple that is representative of the U.S.
nonjnstitutionalized population. Although each NHANES provides a wealth of information on
the prevalence of health conditions and risk factors, the cross-sectional nature of the original
survey limits its usefulness for studying the effects of clinical, environmental, and behavioral
variables on the development of specific health conditions. Data from follow-up surveys of
original subjects are limited, but available, and can,be used for examining health outcomes,
primarily mortality (see Section 3.1). In addition, NHANJ3S99+ data can be linked to Medicare
and National Death Index records to permit longitudinal/tiistorical studies of disease
.
All NHANES are conducted hi a similar manner. Subjects, called sample persons or SPs,
are interviewed at home and this usually includes questions about the subject's personal health
history; the family (i.e., income, ethnic heritage); and household characteristics (i.e., number of
rooms in home, age of home). Subjects later go to a Mobile Examination Center (MEG) where
r 1
-------
standardized physical examinations are conducted by a doctor; blood is obtained by venipuncture
of subjects ages 1 year and older and urine specimens are collected for individuals ages 6 years
and older; and more questionnaires, including detailed questions about food preparation and
consumption, are administered. There are now three MECs, each consisting of four large
construction trailers, that travel around the country during the entire survey. At any point hi time,
only two MECs are set up and operating, while the third is either in transit or being set up. There
are two teams of examiners responsible for staffing the MECs that are hi operation in the field.
Because of weather issues, the survey is generally conducted hi northern areas during the summer
and southern areas during the whiter.
Each MEC is staffed with technicians, a doctor and a dentist and each is outfitted with
computers, standardized laboratory facilities, examination rooms, and all the necessary
equipment to conduct the various physiological tests. The clinical exams include evaluation of
numerous analytes in the blood and urine, a medical examination by a doctor (including
questions about past and current diseases and conditions), and other specialized tests (i.e.,
spirometry, computerized neurobehavioral testing). A limited number of tests are conducted on
blood and urine samples at the MEC. Primarily blood and urine samples are processed at the
MEC and then sent to specified labs for most of the analyses. The MECs stop at predetermined
locations called "stands"(see Table 1). Each MEC stays at a particular stand for about 4-6 weeks
during which time roughly 300- 600 subjects are evaluated. For more detailed information about
the MEC and to take a virtual tour of the facility, go to the NCHS website at
. Examples of the
types of data collected in NHANES are as follows:
demographics (e.g., age, sex, race/ethnicity, education, SES);
housing and family characteristics (e.g., type/condition of house, number of occupants);
risk factors (e.g., diet, physical activity, occupation, smoking habits);
. diseases of the subject and relatives (e.g., cardiovascular, respiratory, cancer, diabetes,
kidney);
reproductive history (e.g., number of children/pregnancies; age at onset of menses and
menopause);
detailed medical and dental examination;
. physiological tests (e.g., electrocardiogram (EKG), vision, hearing, neurobehavioral) ;
anthropometric measurements (e.g., height, weight, girth, skin fold);
clinical chemistries (assays of blood and urine, e.g., counts of various blood cell
components, cholesterol level, levels of different vitamins hi the blood, tests for kidney &
liver function), see Appendices IV and VI for summaries of clinical chemistries for
NHANES-IH and NHANES99+, respectively. Also see Appendix V, for a comparison of
clinical chemistries across NHANES-I, -II, -III and Hispanic HANES;
environmental biochemistries (assays of blood and urine, e.g., blood lead levels,
pesticides in urine, volatile organic compounds in blood, mercury in hah- and blood.) Not
all NHANES tested for the same chemicals, and the number of chemicals increased
significantly in NHANES99+), see Appendix VII for comparison of environmental data
-------
collected across allNHANES; I
* detailed nutrition questions (e.g., consumption of specific foods and dietary
supplements, water intake, frequency of consuming specific foods), see Appendix IV for
summary of dietary/nutritional evaluations from NHANES-III.
3.1. Followup Studies for the Baseline NHANES j
NCHS has planned to conduct epidemiological follow-up studies (NHEFS) for each of
the NHANES, but budget and staff limitations have severely curtailed these efforts. To date,
NHEFS have only been conducted for NHANES-I (see Table 2). NHANES are cross-sectional
studies that collect data on each subject at only one point in time. The objective of these
followup studies is to obtain longitudinal data on all subjects that will support investigation of
relationships among variables (e.g., demographic, clinical., nutritional, behavioral, and exposure)
assessed in the original baseline NHANES and subsequent morbidity and mortality. For
example, the data could.be used to examine the association between certain risk factors (e.g.,
smoking, blood pressure, cholesterol, weight) and subsequent morbidity/mortality or to study the
natural history of certain chronic diseases and functional impairment (e.g., why SPs with
radiological evidence of osteoarthritis do, or do not, go onito develop functional impairment).
Detailed information on the design, content, and operation of the NHEFS and access to the
public use data files and documentation can be found on the following NCHS website:
.
In addition to the NHEFS, NCHS is also conducting the NHANES-II Mortality Study
(NH2MS). This is a prospective cohort study designed tojpassively follow a subset of subjects
fromNHANES-H in order to investigate the association between factors measured at baseline
and overall mortality from specific causes. The NH2MS mortality data can be linked with the
baseline NHANES-H data to examine the relationship between any of the baseline variables (e.g.,
smoking, weight, blood lead, pesticides) and specific causes of death (NCHS, 1999a). The
NH2MS cohort contains adults who were 30-75 years of gge at the time of their NHANES-II
examination (N=9,252). During the baseline NHANES-I]:, some participants were interviewed
but not examined; however, only those examined at baseline were followed for mortality status.
During this first Phase of the study, mortality status was ascertained for years 1976-92. The
NH2MS cohort members were traced by searching national databases containing mortality and
cause-of-death information. The length of followup period ranges from 12 to 16 years.
Approximately 23 percent (n=2,145) of the NH2MS cohort were found to be deceased as of
December 31,1992. i
'i
The design of NH2MS differs substantially from that of the NHEFS, since the latter was
an active followup study with participants being recontacted and medical records and death
certificates being obtained. In contrast, the NH2MS is entirely passive, with participants not
being recontacted and not all death certificates being obtamed. Mortality status in NH2MS was
ascertained solely by computerized matching to national databases and evaluation of the resulting
matches. Matching to the National Death Index (NDI) and other national databases will continue
on a periodic basis, with resulting data being released to the public. One important limitation of
i
-------
the NH2MS design is that subjects not found to be deceased are assumed alive for analytic
purposes, and this could cause a misclassification of vital status.
Detailed information on the design, content, and operation of the 1992 NH2MS is in the
Vital and Health Statistics Series 1, Number 38 "Plan and Operation of the NHANES II
Mortality Study 1992"(NCHS, 1999a). This document, plus the public use data file and
documentation can be found at the NCHS website: .
3.2. How NHANES have been used to Improve Public Health
NHANES have provided much valuable data that have been used over the years to
improve public health. Some examples noted on the NCHS website
are as follows: a)
development of growth charts that are used nationally and internationally by physicians as
standards for assessing the growth of their young patients; b) assessment by the U.S. Department
of Agriculture (USD A) of the vitamin and mineral intake of the population and use of this
information to improve our diets. Earlier NHANES showed that low iron levels were a serious
problem for many people, including women of childbearing age, preschool children, and the
elderly. As a result, the government took steps to fortify grain and cereal with iron to correct this
deficiency; c) NHANES showed the need for folate to eliminate another dietary deficiency and
prevent birth defects; d) For years, NHANES has tracked the levels of cholesterol in adults. This
information has helped to establish the link between high cholesterol levels and the risk of heart
disease and to alert patients and doctors to the issue. When NHANES started testing, one-third
of adults had high cholesterol, but today fewer than 1 in 5 adults do. Changes hi diet and
lifestyle all built on information from the national survey have sharply reduced the risk of dying
from a heart attack; e) NHANES-II (1976-80) gave the first clear-cut evidence of the high levels
of blood lead (PbB) in the U.S. population, particularly hi children. This evidence led congress,
the EPA, and other agencies to phase out the use of lead (Pb) as an additive in gasoline and the
resulting reduction in PbB levels has been remarkable. These data were also used to support
federal policies to eliminate Pb from solder in food and soft drink cans. Monitoring of PbB
levels in NHANES-HI and NHANES99+ show a continued decrease in the body burdens of this
toxicant. Pb exposure remains a problem for certain groups, especially poor, inner city children
living in old houses with lead paint. NHANES99+ continues to monitor for PbB and this
information helps public health agencies pinpoint where Pb remains a problem; f) NHANES data
continue to indicate that undiagnosed diabetes is a significant problem in the U.S.. The data have
been used by federal and private agencies to increase public awareness, especially among
minorities, of this problem; and g) NHANES has produced information on the prevalence of
overweight and obesity that have led to the proliferation of programs emphasizing diet and
exercise and stimulated needed research. New measures of physical fitness used in
NHANES99+ will further our understanding of its role in health and enhance the analysis of
relationships among exercise, obesity and disease. For more information on how NHANES has
been, and will be used to improve public health, see the above-noted website. Also, see Section
-------
6.2 of this Handbook for more information on how NHANES has been used to support risk
assessment, poHcy needs and regulatory decisions at EPAi
4.0. DETAILED INFORMATION ON NHANES CONTENT
f
NHANES is the largest survey for gathering data qn the health, and nutritional status of
the U.S. population. It is designed to facilitate and encouiage subject participation, by providing
transportation to and from the MEC and compensating subjects for their time and effort. Each
participant also receives a report of their medical and dental findings. In addition, NCHS goes to
great lengths to study the reasons why potential subjects do, or do not, participate in the survey.
The information obtained is used to devise tactics for increasing response rates (e.g., increased
local media coverage; increased targeting of specific race/ethnicity groups with appropriately
tailored publicity). These efforts have paid off, as indicated by the steadily increasing response
rates with each successive survey. Thus, the overall response rates (including interview and
MEC examination) have increased as follows: NHANES-fl (73.1%), HHANES (73.3%),
NHANES-m (76.6%) andNHANES99+ (78%) (Source: iStHANES Consortium meeting, 2002).
ij: ' , ,,. :, . , ,
As noted in Section 3.0, all data are gathered through personal interviews, physical
examinations, and diagnostic and biochemical testing of a statistically representative sample of
the U.S. population. Also as noted in Section 3.0, there isja core set of components that is
administered in each of the NHANES and there are agency-specific components that are retained
as long as there is mutual interest between NCHS and the requesting agency to do so, and
funding is available. Over the years, the number of tests and questions administered to subjects
has greatly increased. Of the completed surveys (NHANES-I, -II, -III and HHANES), NHANES-
HI is the most recent and collected the largest amount of data. Although the field work for
NHANES-99 and -01 has been finished, analysis of bloodjand urine samples has not been
completed for many analytes, and data have not been relezlsed to the public (see Section 4.0 and
Table 3 for data release schedule for NHANES-99 - 08). For these reasons, this Handbook
presents detailed examples of the survey content from NHANES-III and then presents summary
information comparing specific components across the suiryeys. By becoming familiar with the
details of NHANES-m, the reader will have a sound understanding of the basic characteristics of
all the NHANES. ;
f !'
Appendix n provides an overview of the health status assessment component of
NHANES-IH by describing the public health objectives and data collected to support the
following: 1) evaluation of the health of specific population subgroups, such as children and
adolescents, the elderly, women and minorities; 2) assessment of environmental and occupational
health and exposures; 3) evaluation of specific diseases (etg., cardiovascular, respiratory,
diabetes); and 4) examination of specific risk factors (e.g..| smoking, alcohol and tobacco use).
More information on the design and data collection of NHLANES-III and public use data files can
be found on the NCHS website: .
This website also contains links to comparable information for all the surveys. Just click on the
name of the specific survey for access to manuals, documentation books and data files.
8
-------
Appendix HI provides an overview of the nutritional status assessment component of
NHANES-III by describing the data collected to support the following: 1) evaluation of alcohol
intake; 2) assessment of hunger; 3) examination of vitamin and mineral status through
biochemical testing of blood and urine; 4) evaluation of infant and child nutrition; 5) evaluation
of growth, overweight and obesity; and 6) examination of relationships between diet and health.
Appendix III also summarizes the methods employed for obtaining the nutritional data,
including questions about food frequency and. the 24-hour dietary recall; measures of
anthropometry; and laboratory determinations. More information on the design and data
collection of the nutritional component of NHANES-III and public use data files can be found at
. As noted above, this website
contains links to comparable information for all the surveys.
Appendix IV summarizes the tests and questions administered in NHANES-III. This
Appendix includes a description, by age of subject, of the following: 1) questionnaire topics; 2)
physical examination components; 3) analytes tested in blood and urine; 4) special studies,
including a list of volatile organic compounds (VOCs) tested hi the blood of a random sample of
adults; and 5) dietary and nutrition intake information.
Given the complexity of the NHANES, it is difficult to summarize information about the
entire content of individual surveys in order to compare components across all surveys.
Appendix V compares the clinical chemistries in blood and urine for NHANES-I, -II, -III and
HHANES. Appendix VI compares only the components of the current survey for the years 1999-
2002, as follows: Table 1 compares the analyses in blood, urine and hair, by year and subject age
Table 2 compares questionnaire content by year and subject age; Table 3 compares examination
components by year and subject age.
As indicated in Sections 3.0 and 4.0, each NHANES has collected a certain amount of
environmental and risk factor data that have potential interest to EPA. With each successive
survey, collection of these types of data has increased markedly. Appendix VII presents a
comparison of environmentally relevant data collected across all the NHANES, by age of
subjects. This includes questions asked about potential exposures (e.g., smoking history, use of
pesticides in garden and home) and identification of specific biomarkers hi blood, urine, hair and
environmental media. With increasing concerns about environmental exposures and the possible
link between exposures and health effects and/or decrements in physiological functioning, there
is increasing interest hi collecting needed exposure data through NHANES. Now, and into the
future, NHANES99+ will be collecting a tremendous amount of environmental biomarker data
that have potential interest to EPA. Because of the expense of the various tests and the finite
amount of bodily fluids that can be obtained from each subject, it is not possible to conduct all
analyses on every subject, so only subsamples may be tested. For all chemicals, the size of
subsamples is noted, such as "1/3 sample", which means that 1/3 of the age-eligible subjects
were tested. Analysis of many of the analytes listed in Appendix VII for NHANES99+ will
continue through at least 2002. Check the NCHS website for updates on specific chemicals to be evaluated hi the future years.
9
-------
Data from NHANES99+ have not yet been release^ to the public or collaborating
agencies, and none are expected for release until later in 2002. However, NCHS and CDC have
conducted some preliminary analyses on the limited data, primarily from NHANES-99. Reports
summarizing these limited analyses can be found on the fallowing NCHS website:
. Currently, the listed reports
include: Blood and Hair Mercury Levels in Young Children and Women ofChildbearingAge-
U.S. 1999; Blood Lead Levels in Young Children- United States and Selected States-1996-1999;
Folate Status in Women ofChildbearingAge- United States, 1999; Prevalence of Overweight
Among Children and Adolescents- United States, 1999; ayd Prevalence of Overweight and
Obesity Among Adults- United States, 1999. This website also contains the National Report on
Human Exposure to Environmental Chemicals, a new CDC publication started in March,
2001and described as providing an ongoing assessment of the U.S. population's exposure to
environmental chemicals using biomonitoring. This Report uses NHANES99+ blood and urine
biomonitoring data to provide statistical summaries of levels of environmental chemicals in the
people. Depending on the amount of raw data, these summaries may, or may not, be categorized
by age, race, SES, or other characteristics of the population. CDC plans to release ibis Report
every year based on new NHANES data. For some EPA (Offices these statistical summaries may
be sufficient to support regulatory and/or policy needs.
5.0 HOW TO OBTAIN NHANES DATA AND ISSUES WITH FILE STRUCTURE
AND CONTENT
r
NCHS works with the collaborating agencies to QA/QC the field data and then .generates
files for all the clinical, laboratory and questionnaire data 'for each subject. NCHS generates
various weighting factors and other documentation needed to analyze the data and develops data
sets that are periodically released to the collaborating agencies and then to the public. NCHS
does not release any confidential subject information that could be used to identify the subject or
his/her residential location. All of the publicly available data from all the NHANES, including
theNHEFS and the 1992 HANES-H Mortality Study, can! now be obtained either over the
Internet, through the NCHS Research Data Center (See Section 5.2) or ordered from the National
Technical Information Service (NTIS). The NCHS website provides links for obtaining jthe data and documentation on-line and
instructions for ordering materials from NTIS. Just click on the name of the specific survey for
access to manuals, documentation books and data files.
NCHS also provides survey-specific reports, manuals and other documentation that are
needed to understand the survey design, methodological issues, components of the survey and
how the different components were conducted. These documents are available for all the
NHANES through , or from NTIS,
and include the following: j
1) Plan and Operation Books contain a detailed overview of the specific survey and
provide descriptive information on specific components of the health status and
i
p
10
; !
>f
-------
nutritional assessments; sample design and analysis guidelines; data collection
procedures; a copy of the interview forms; and a list of exam components.
/
2) Documentation Books are provided with the data files. They contain a list of the
exam/interview components that are hi each data file; guidance on how to use the data
files; information on how to create datasets from the data files; data preparation and
processing procedures; descriptions of the variable names used and positional location of
each variable in the data files; and raw counts of subjects for each survey question, lab
component and test.
3) Interviewer Procedure Manuals were developed for training the interviewers and are
important references for NHANES as they detail the correct procedures, policies, and
standards for interacting with the subjects while conducting the interviews.
4) Exam Manuals detail the correct procedures and equipment used for administering
each physiological test and the questionnaires.
Now that NHANES is an annual survey that will be in the field continuously, NCHS has
had to deal with many new issues regarding frequency of data release. For example, since only
5,000 subjects are examined each year, there are questions about the number of subjects needed
to calculate statistically meaningful estimates from the data. With this small number of annual
subjects, there are also issues about protection of subject confidentiality. At the 2002
Consortium meeting of the federal agencies supporting NHANES, NCHS announced new plans
for scheduling not only raw data releases, but survey planning activities and report releases (see
Table 3). From now on, all these activities will be conducted on a 2-year cycle (1999-2000,
2001-2002,2003-2004, etc.). "Micro-data", identified as all the raw sociodemographic, exam,
and questionnaire data, will not be released for single years of the survey because of concerns
about protecting subject confidentiality. "Tabular Reports" will be prepared by NCHS and/or
CDC to summarize findings on a limited number of variables judged to be of significant public
health importance. These Tabular Reports, unlike the raw data, will be based one year's worth of
data and will be released annually. NCHS will now solicit proposals for new survey
components, and pilot test certain proposed components, on this 2-year cycle. Many data
elements (e.g., residential location, data collected on a small subset of subjects) will never be
made available to the public in order to protect subject confidentiality. These data are often very
valuable and critical to support studies of public, and/or environmental health importance, and
studies that can support EPA regulatory, policy and/or research needs. As indicated in Table 3,
NCHS is scheduled to make these data available only through their Research Data Center (RDC).
- See Section 5.2 of this Handbook for details on the RDC and how to access the data.
5.1. Issues with Data File Structure and Content
Each NHANES provides an incredibly rich source of human data for research and
analysis. However, the data set is large, complex and requires that the user be familiar with data
file manipulation and analysis. Thorough review of the extensive documentation provided by
11
-------
NCHS for the individual datasets should resolve most questions. All data users need to review
these reference materials and reports before analyzing any NHANES data. If you still have
questions after careful review of the documentation, try contacting the NCHS Data
Dissemination Branch at (301)458-4636.
The following examples demonstrate the complexity of the data file structure and content
and illustrate some of the many issues that need to be considered when designing a study using
data from any NHANES:
Data from eachNHANES are divided into separate files. Depending on the analysis to
be performed, one or more of these files will need to be merged in order to obtain all the
needed data on each subject. Users need to be familiar with the content of each file. For
example, the NHANES-HI survey design and demographic variables are in the Household
Adult Data File (subjects 18+ years old), Household Youth Data File (subjects up to 17
years old), Laboratory Data File (all ages), and Examination Data File (all ages). All of
the NHANES-ni public use data files are linked through the common survey participant
identification number (sample sequence number or; SEQN). Merging information from
multiple NHANES-m data files using the SEQN variable ensures that the appropriate
information for each subject is linked correctly. In; preparing a data set for analysis, other
data files should be merged with the Adult Household Data File and/or the Youth
Household Data File to obtain many important analytic variables.
i
NHANES public use data files do not have the same number of records on each file.
For example, inNHANES-III the Adult and Youth Household Questionnaire Files
contain more records than the Examination Data File because not everyone who was
interviewed went on to complete the examination, j The Laboratory Data File contains
data only for persons aged one year and older. The Individual Foods Data File based on
the dietary recall, the Prescription Medication Data File, and the Vitamin and Minerals
Data File, all have multiple records for each person rather than the one record per subject
contained in the other data files. :
With each data file, NCHS includes separate text files with SAS program code using
standard variable names and labels. This SAS program code can be used to create SAS
data sets from the data file. !
During the course of each NHANES, NCHS modifies items in the questionnaires,
laboratory, and/or examination components. As a result, data may not be available for
certain variables for all years of any one survey. In addition, variables may differ by the
phase of the survey because some changes were iniplemented between phases. For
example, NHANES-III was conducted in two phases from 1988-1991 and 1991-1994.
Because of the changing research needs of the sponsoring agencies, NCHS dropped a
limited number of questions and examination components after the first phase, and
replaced these with new questions and tests in the second phase. This process of survey
12
-------
modification will become more noticeable withNHANES99+, since the survey is now in
the field continuously, and proposals for new components will be on a two year cycle (see
Table 3). In general, if new components are added, then old components) need to be
deleted so the burden to subjects is not significantly increased. Furthermore, NCHS is
currently trying to determine how many years specific components should be retained in
order to obtain a statistically meaningful sample size. Because of these modifications,
users need to read the Notes sections of the file documentation carefully, as they provide
information about survey changes.
NCHS verifies extremely high and low test values whenever possible, and performs
numerous consistency checks on the data. Nonetheless, users need to examine the range
. and frequency of values before analyzing data.
Confidential arid administrative data are not available or released to the public. (See
Section 5.2 on how these data may be accessed through the NCHS Data Research
Center). Additionally, some variables have been receded to protect the confidentiality of
the survey participants. For example, inNHANES-III all age-related variables were
receded to 90+ years for persons who were 90 years of age or older.
See the following website for more information on guidelines for NHANES data users based on
NHANESJH: .
5.2. NCHS Research Data Center (RDC)
As already noted, there is no public release of any confidential data on NHANES
subjects; however, NCHS collects and maintains these data (such as residential address).
Without access to certain types of confidential information, it would be impossible to conduct
some types of studies that would be very useful to EPA. For example, if one just uses the
publicly available data, analyses are generally limited to a national or regional scale, or to
comparisons made for urban vs rural areas. It should be noted that while the design and release
of NHANES data limit analyses to these large geographic scales, hi many cases this is sufficient
and appropriate for EPA needs. A case in point was EPA's use of NHANES-II data to develop
the distribution of blood lead (PbB) levels in children for the whole country in support of
regulations to reduce Pb in gasoline (U.S.DHH, 1988; U.S. EPA, 1986).
NCHS recently established a Research Data Center (RDC) at its headquarters in
Hyattsville, Maryland. More detailed information about the RDC, its rules for prospective users,
and how to access it can be found at . Briefly, the RDC
was created to meet the continuing demand for analyses to be conducted on smaller geographical
scales (e.g., by state, county and below county) that require restricted data from NHANES.
Designed for the researcher outside of NCHS, the RDC allows access to data that would not be
permissible to analyze because of confidentiality/disclosure rules and regulations. These
sensitive data can not be publicly released, but potentially can be accessed through the RDC, for
13
-------
statistical studies. In addition, the potential exists for linking NHANES data to environmental
data, as noted above, but also to other human databases, siich as the Census data and the National
Death Index. By working through the RDC, it will be possible to link NHANES data to a variety
of other data bases in order to expand the usefulness of NIIANES and conduct studies not
possible if just based on NHANES data alone. Data sets for linking can be user-generated (e.g.,
exposure data generated from environmental monitors), oi| publicly available (e.g., Census data).
Only the RDC staff can link the data sets. Files and inforrtiation used to link any data sets are
destroyed after the merged file is created and are not be made available to the user.
The RDC has instituted restricted conditions to protect sensitive data but allow analyses
at a level not possible with public use data. These conditions are identified on the website
and summarized as follows:
1) Prospective researchers must submit a research proposal to the RDC for review and
approval. Approval will be based on the availabiliity of RDC staff and resources;
consistency of the proposal with the mission of NCHS; general scientific soundness of
the proposal; and the feasibility, of the project. It is jexpected that the user will develop the
research proposal with the RDC staff to minimise the time required.
2) Researchers will sign confidentiality agreements, but strict confidentiality protocols
require that researchers with approved projects must complete their work using the RDC
facilities. RDC facilities can be accessed on-site or remotely.
3) Researchers can supply their own data to be merged by RDC staff with NCHS data
sets. Merged files will be only available to the originating researcher unless written
permission is given to allow access to others. \
The RDC can be accessed on-site, in Hyattsville, MD or remotely. Another analysis
option for researchers with large, complex analytic projects may be to subcontract with the RDC
to have its staff run the necessary programs, etc. for the research project. Each mode of access
has associated costs, procedures, rules and limitations. Other details of the RDC operation are
discussed on the website as follows:
,i, , . . .
For on-site users: |
RDC staff will construct necessary data files, uicluding those linking NHANES data
with user data. i
PC SAS, SUDAAN, STATA, FORTRAN and HLM are available. Other statistical
packages are available with sufficient lead time for (RDC staff.
Output is subject to disclosure review by the RDC staff. Disclosure review guidelines
are published in the NCHS Staff Manual on Confidentiality.
Analyses (paper output) can be taken off site contingent on passing disclosure review
by RDC staff. \
The RDC is only open during normal working hours and requires RDC staff oversight.
I" '5
I
14
-------
For remote access users:
RDC staff will construct necessary data files, including those linking NHANES data
with user data.
Researchers can submit analytical computer programs via e-mail. Output is returned
by e-mail to users' registered address.
Only SAS programs can be run. Certain SAS procedures and functions are not
allowed, including: PROC TABULATE, PROCIML, LIST and PRINT (e.g., no listing of
individual cases), R_, FIRST., LAST., (e.g., no selection of individual cases). No cell
with fewer than five observations can be included; and if found, other cells will also be
suppressed.
The SAS job log will be scanned for conditions that result in case listings.
Specific costs:
To work on site costs a researcher $1,000 per week.
To remotely access any one data set costs $500 per month.
File construction and setup by RDC staff for either remote access or on-site use costs
$500 per day of effort.
6.0. TYPES OF ANALYSES THAT CAN BE CONDUCTED WITH NHANES DATA
NHANES can be a valuable resource for EPA to use in support of research, risk
assessment, policy decisions and regulations. This rich human database has several features that
make it attractive for EPA use, including: 1) potential to be linked with other databases,
including census, exposure and mortality data; 2) can be used to identify emerging health and
exposure issues; 3) can be used to examine associations between possible risk factors and
adverse health effects or other conditions; 4) can be used to help evaluate the effectiveness of
existing environmental regulations and the need for additional/new regulations; 5) can be used to
investigate issues based on race/ethnicity and SES (e.g., environmental equity issues) and
geographic location; and 6) potential for limited longitudinal studies as more follow-up data are
collected (see Section 3.1). .
Different methods can be used examine the NHANES data, including the following:
1) Development of population distributions of specific variables (e.g., height, weight,
PbB, blood or urine levels of environmental chemicals). These distributions can be
developed for the U.S. population as a whole, or for specific subgroups based on a variety
of characteristics such as age, race/ethnicity, SES, gender, etc. It should also be possible
to develop distributions for different parts of the country, but this may need to be
accomplished through the Research Data Center (see Section 5.2).
2) Estimation of the prevalence of selected diseases or chronic conditions (e.g.,
respiratory, cardiovascular, neurobehavioral). Prevalences can be calculated for the U.S.
population as a whole, or for various subgroups. Subgroups can be defined by
15
-------
demographic characteristics, or with assistance from, the RDC using characteristics of
their residential neighborhoods (see Section 5.2).
3) Use of inferential studies to examines possible jlssociations between various risk
factors and particular diseases or conditions. For example, based on NHANES-III data,
one can assess potential risk factors (e.g., active arid passive smoking, allergies to
cats/dogs, use of gas stove) for asthma in children,1 and assess the value of these risk
factors as predictors for asthma. i
Because of the design of NHANES, there are limitations on how estimates can be
expressed. For example, analyses can be developed based on:
1) National estimates (e.g., distribution of PbB levels for all children 1-5 years of age).
I r i
i
2) Estimates by broad geographic regions (NE, N\y, SE, etc.), (e.g., distribution of PbB
levels for all children 1-5 years of age and living in the south).
3) Estimates by degree of urbanization (e.g., distribution of PbB levels for children 1-5
years of age living in metropolitan areas vs rural areas).
4) Estimates by other major population subgroups.; such as race/etiinicity, income, sex, or
age (e.g., distribution of PbB levels in black children ages 1-5 years compared to
Hispanic children of the same age). !
i
With the advent of geographic information systems (GIS), it is now possible to
geographically locate point and area sources of pollution; inap plumes of pollutants in the air,
groundwater and soil; and model and map other environmiental exposures. It should also be
possible to link NHANES data, through appropriate geographical identifiers, with mapped
environmental data in order to evaluate possible relationsliips between variables in the human
data and exposures from the environmental data. For example, NCEA is collaborating with
NCHS to conduct a study of risk factors for respiratory ef fects in children by linking EPA air
monitoring data with NHANES-III children's respiratory data. The addresses of the children
have been geocoded to the Census block group level by NCHS and the monitoring data for
ozone and PM10 have been interpolated to the block group level by NCEA. -NCHS links both
sets of data by the common field of census block group arid then retains the linked data set to
protect subject confidentiality. The linked data are used tp evaluate relationships between
exposure to these air pollutants and respiratory effects, including decrements in lung function
(See APPENDIX DC, project summary entitled "Estimation of Risk Factors for Respiratory
Effects in Children: Use of NHANES-III Respiratory Heeilth and EPA Air Monitoring Data. Part
IV). The block group identification of the NHANES-III subjects is confidential and not released
to the public, but is part of the data set. Any publications resulting from this study will not
disclose the locations of the NHANES children. Moreover, it is important to note that given the
design of NHANES-in (and all NHANES), the results of this study can only be expressed as
16
t "
-------
whether or not ozone and/or PM10 have an effect on the respiratory health of children. The
design of NHANES will not support an analysis of whether children in specific parts (i.e., cities,
counties, states) of the country have poorer lung function because they live in an area with "high"
air pollution levels. The design of NHANES will support an analysis to determine if, in general,
higher levels of air pollution are associated with poorer respiratory health in children. Provided
that subject confidentiality is maintained and the type of analysis is consistent with the design of
NHANES, the possibility exists for conducting similar types of studies by linking NHANES data
to other environmental databases. See Section 8.0 for more precautions on appropriate use of
NHANES data. Now that the NCHS RDC (See Section 5.2) has been established, it should be
possible, with the appropriate data, to conduct these types of studies.
6.1. Studies using NHANES Data that have Environmental Relevance
Many papers have been published using NHANES data and these cover a wide variety of
topics, many of which are of potential interest to EPA. NCHS maintains an annotated
bibliography of the studies they are aware of; however, this list is not exhaustive. Annual
updates to this bibliography and copies of NCHS/CDC publications cited hi the bibliography
may be obtained from the Data Dissemination Branch, NCHS, 6525 Belcrest Road, Hyattsville,
Maryland 20782. A selective bibliography from 1997 -1999 can be found on the following
NCHS website: .
Based on the bibliography from the NCHS website, plus another selective bibliography
NCHS prepared for 1980-1996 but that is not on their website, we have assembled a list of
references (See APPENDIX VIII) that have potential environmental relevance. Note that
Appendix VIII contains a mix of annotated references, which were taken from the NCHS 1997-
1999 bibliography and references without annotation, which were taken from the NCHS 1980-
1996 bibliography. This compilation of references also provides a good overview of the broad
range of studies that can be conducted with NHANES data.
6.2. Use of NHANES Data at EPA
Over the years, EPA has successfully used NHANES data in research and to support
policy and regulatory decisions. Some of tibiese activities included:
1) Evaluation of the relationship between PbB, adverse health effects (e.g., cognitive
function, hypertension) and levels of Pb in the environment, particularly as a result of Pb
in gasoline. NHANES-II data were used to support EPA regulations to remove Pb from
gasoline. NHANES data also demonstrated decreases in PbB that paralleled decreases in
Pb in gasoline. EPA monitors Pb exposures by evaluating NHANES-ni and
NHANES99+ PbB data and uses these data to support continuing policy decisions.
(U.S.DHHS, 1988; U.S.EPA, 1986). Also see Appendix EX.
2) Evaluation of respiratory function and persistent respiratory symptom data from
NHANES-I and -II. EPA used these data to support regulatory decisions for the National
17
-------
Ambient Air Quality Standards (NAAQS) pollutants, which include ozone, sulfur oxides,
nitrogen oxides, particulate matter, lead and carbon monoxide.
r
3) Evaluation of NHANES99+ data on mercury (fig) levels in blood and hair of children
and women of child-bearing age. These data are used by several EPA offices to
determine exposure levels to methyl-mercury in these two susceptible subpopulations and
to support policy decisions on Hg exposures. EPA will continue to use these data to
support policy and regulatory decisions, as the levels for subjects tested in 2000 and
beyond are made publicly available (See Appendix; IK).
4) Use of HHANES urine and serum pesticide data to develop distributions of the
prevalence of pesticide exposures in Hispanic subpiopulations. These distributions were
used as reference standards for assessing pesticide exposures in children and adults in
studies of populations living along the U.S.- Mexico border.
5) Use of urine and serum pesticide data in combination with food consumption data to
support development of pesticide tolerance levels in food.
6) Evaluation of pulmonary function for different population subgroups based on age,
sex, and race. EPA used these data to develop predictive models for several spironietric
endpoints (e.g., FEVl5 FVC) for children, teens, and young adults. The models allow for
investigation of how pulmonary function differs by race and sex and how these
differences interact with growth and development patterns.
7) Examination of the relationship between blood pressure and the level of cadmium (Cd)
in urine, which is an indicator of the body burden of Cd. Work has also included
examining the correlation between Cd in urine and beta2-microglobulm (indicator of Cd-
induced renal damage) and using the information as benchmarks for measuring the effects
of environmental Cd exposure.
i
8) Use of surplus blood sera from NHANES-III subjects to determine if poorer drinking
water quality correlated with higher prevalence of antibodies to Cryptosporidium (see
Appendix IX).
Several EPA program offices currently use NHANES data in epidemiology studies and/or
to support policy and regulatory work. APPENDIX IX presents summaries of twenty two EPA
efforts we were aware of that use NHANES data. Again, note the broad diversity of issues being
addressed and different methods being applied to the data, i
18
-------
7.0. USE OF NHANES STORED BIOLOGICAL SPECIMENS
7.1. Serum Specimens:
Blood collected from NHANES-III subjects was separated into its components and sent to
the CDC laboratory or to one of the eight contract laboratories for testing (see Appendices IV and
VII). Because of the possibility of out-of-range results that need to be repeated, more specimen
volume was sent to the labs than was usually needed for the scheduled biochemical tests. Since
most subjects do not have out-of-range results, the.labs now have a large numbers of surplus
serum specimens. All specimens have been stored at -70 degrees and have been through at least
two freeze-thaw cycles. Stored sera from NHANES-III are available for research projects that
require a nationally representative sample of the population. Research proposals are accepted
and reviewed throughout the year by the NHANES-III Stored Sera Technical Review Panel,
which attempts to review the proposals within 30 days of receipt. See the NCHS website
for more details on the proposal
solicitation process; review process; proposal selection criteria; and guidelines for proposal
preparation.
Note that EPA has successfully used NHANES-in serum samples to conduct a study of
the relationship between serum antibody response to Cryptosporidium and the primary source
and treatment of drinking water. See Appendix IX, study entitled "Analysis of Serological
Responses to Cryptosporidium Antigen Among NHANES-III Participants" for a summary of this
work.
InNHANES99+, samples of blood, urine, and saliva (if applicable) for subjects 7 years
and older who give their written consent are being stored for future health studies
.
7.2. DNA Samples:
For NHANES-III subjects VI years of age and older, lymphocytes (white cells) were
isolated from blood samples and stored frozen in liquid nitrogen or as cell cultures immortalized
with Epstein-Barr virus. The cells have been stored and maintained at the Division of
Environmental Health Laboratory Sciences (DEHLS) at the National Center for Environmental
Health (NCEH), CDC. Cell cultures are available primarily from Phase 2 of the survey, which
was conducted 1991-1994. Though an extensive consent form was signed by participants in the
survey, specific mention of genetic testing was not included. Collection and storage of this
biological material was based on the promise offered by significant advances in the rapidly
evolving field of molecular biology that were occurring during the NHANES-III planning phase.
Technical advances now make it possible to use these specimens for genetic analysis. NCHS
and NCEH are making anonymized DNA from these specimens available to the research
community for such analyses, but no cell lines will be made available.
Given the scientific importance of this resource, NCHS developed a plan for making
DNA samples available to the research community for anonymized testing. This plan was
19
-------
approved by the NHANES Institutional Review Board (IE1B) September 16,1996. For more
details and guidelines concerning development, submission, review and acceptance of proposals
for studies using stored DNA samples, see the following NCHS web site:
. This website now indicates
that since the Fall'of 2001, NCHS is in the process of Devaluating their protocol for soliciting
and approving proposals for studies using the stored DNA: specimens. The website further
indicates that their ultimate goal is to allow reasonable access to the samples for important
scientific studies while assuring participant confidentiality and compliance with appropriate
ethical standards. For these reasons, NCHS has discontinued further review of applications
while they revise their protocol. If the revised protocol isj" approved by the NFIANES iRB, they
will place an announcement on the above-noted web site, and presumably start accepting
proposals again. ;
InNHANES99+, DNA samples (from blood or sdiva) from subjects ages 20 and over
who give their written consent are being stored for future genetic testing
.
8.0. LIMITATIONS AND PRECAUTIONS ON USE OF NHANES DATA
The analysis of NHANES data is not a straightforward task. "The analyst must consider
many issues to develop an appropriate and efficient strategy for conducting the analysis. Such
considerations should include ...the sample design, weights, or underlying assumptions hi the
analytic procedures to be applied..." (NCHS, 1982). |
Before starting any analysis using NHANES data, researchers need to be aware of several key
issues: j
A) Goals of NHANES vs goals of studies analyzing NE[ANES data a geographical
perspective.
Although there are many important issues to be considered when designing a study using
NHANES data, one of the most critical deals witibl the incorporation of geographical
information and the appropriateness of using NH/pJES data to examine health effects
and risk factors on a small scale. Remember that the goal of NHANES is to provide
information on the health and nutritional status of ithe U.S. population. The complex
design of NHANES has been developed to suppoirt that goal. Because this is a unique
survey that provides such a wealth of data on such a large number of people, it is easy to
overlook this goal and try to conduct analyses that are of interest to EPA but may be
inappropriate given the design of NHANES. A prune example of this potential pitfall is
based on the fact that NHANES is conducted hi numerous locations (e.g., stands), with
several hundred people examined at each location (See Section 3.0 and Table 1). NCHS
also publicly identifies the county of residence, provided the population is greater than
500,000. Because of this geographical distribution of subjects, you may be tempted to
identify stands that are located "near" hazardous waste sites or specific industrial point
sources of pollution that are of interest to EPA and then conduct an analysis to see if there
20
f :':
r is f,,'
11!,-;;,.i,.1!
-------
is a correlation between adverse health effects and living "near" a particular pollution
source. NHANES was not designed to support this type of study, because the people
examined at a particular stand are not necessarily representative of the population in that
immediate area. Although we now have techniques (e.g., GIS) that allow for creative
ways to incorporate geographic detail into analyses of NHANES data, these approaches
may not always be valid, given the design and underlying purpose of this survey. With
GIS it is now possible to access many envirojimentally-relevant databases (e.g., locations
of point and area sources of pollution) or generate your own (e.g., model air
concentrations of chemicals released from a specific industry or industries) and then link
these to the NHANES data in order to explore hypotheses about risk factors for adverse
health effects at specific locations. While many such interesting analyses are possible, are
they always appropriate? Remember, NHANES is designed to support estimates of
effects, etc. based on: a) a national level; b) a broad regional level (e.g., NE, NW, SE); c)
by degree of urbanization (e.g., metropolitan areas vs rural settings); d) by population
subgroup (e.g., race/ethnicity, income, sex, age). NHANES is not designed to provide
valid parameter estimates based on survey participants living in specific locations such as
local towns, counties or states. NHANES was not designed to support studies of the
health status of individuals in a specific location. Thus, when developing a study that
uses NHANES data, you need to ensure that the design of your study is consistent with
the goals and design of NHANES and makes appropriate use of the data (NCHS, 1982;
and the following website:
.
B) Complex sample design of NHANES.
NHANES uses a complex, stratified, multistage, probability cluster design to select a
representative sample of the noninstitutionalized, civilian U.S. population. Because of
this complicated design, one can riot use traditional statistical methods to analyze the
data.
C) Need for weighting.
Because the NHANES sample design is complex, sample weights must be used to
account for stratification, clustering, and the unequal probability of subject selection into
the survey. Each NHANES oversamples certain population subgroups (see Section 3.0
and Tablel), and this must be taken into account through appropriate weighting. Sample
weights are also used to adjust for possible bias resulting from subject nonresponse.
Weighting is used to bring the sample data up to the dimensions of the target population
totals and to reduce variances in estimations. The issue of using weights is critical to the
correct analysis of NHANES data. NCHS calculates the necessary weights for different
components of the survey and provides these with the public use datasets, along with
explanations of how the weights were derived and when to use them. It is the
responsibility of the user to select the appropriate weights and apply them correctly.
There is also an issues of whether or not to use certain weights, depending on the type of
analysis being conducted. Inappropriate use of weights may result in a flawed analysis
21
-------
and questionable study results. See Section 9.1 foil more information about weighting.
D) Limitations associated with small sample size. j
As noted elsewhere (see issue E, "Variable content; within surveys", of Section 8.0 and
Appendices VI and VTI) some variables are measured only in subsamples of the survey
and/or for a limited time. Moreover, because each NHANES is conducted over multiple
years, it may be enticing to try to examine tune trends by making and comparing annual
estimates within a particular survey. Any of these Conditions may result in having a small
sample size for one or more variables and this complicates analysis of the data and
interpretation of the results. This problem of small sample size is illustrated by the
CDC's analysis of the first year of the NHANES99+ biomonitoring data for their
National Report on Human Exposure to Environmental Chemicals (see Section 4.0 and
the following website: ). This Report indicates: ;
"Although the current NHANES is conducted using annual samples that are
nationally representative, the sample size iti any one year is relatively small,
resulting in large variability for estimates, especially those for detailed
demographic groups or other detailed analyses. The NHANES is designed to
increase precision by combining data across calendar years. Because of the small
sample size in 1999, a number of survey pa|rticipants have large sample weights,
and the potential exists that these sample weights may strongly influence
estimates. This is particularly important fot chemical results that were only -.
measured in subsamples." ;
i
"Another analytic limitation of the NHANES sample is that it is selected from a
relatively small number of sampling units (PSUs) or counties; the 1999 sample
was planned for only 12 PSUs. With a smzill number of PSUs, variance estimates
that account for the complex design will be| relatively unstable, a factor which
introduces a higher level of uncertainty in the annual estimates."
"Although the annual NHANES is nationally representative, it is not possible to
produce environmental exposure estimates by geographic region. Because the
number of geographic sites sampled each year is small and because environmental
exposure measures may vary geographically, national estimates of these
exposures, particularly those based on 1 year of data,, may be highly variable."
!' ' '
iji' ' " i i . .
"These limitations related to measuring environmental exposures from a single
year of NHANES will be addressed as more data become available from the
ongoing survey. More detailed analyses by demographic groups and other
variables will be possible with increased sejmple size and with a. larger number of
geographic locations."
i
22
-------
NCHS also raises the issue of the appropriateness of analyzing each phase of an
NHANES separately in order to look for time trends. As noted in Section 9.0, NHANES-III was
conducted in two consecutive phases. Each subject's participation in either the first or second
phase is identified in the public database by the variable names of sdppsul and sdppsu2, and each
phase has its own weighting factors. NCHS (1996b) indicates that:
"In NHANES III, 89 survey locations were randomly divided into two sets or
. phases, the first consisting of 44 and the other of 45 locations. One set of PSU's
was allocated to the first three-year siirvey period (1988-91) and the other set to
the second three-year period (1991-94). Therefore, unbiased national estimates of
health and nutrition characteristics can be independently produced for each phase
as well as for both phases combined. Computation of national estimates from
both phases combined, (i.e., total NHANES HI) is the preferred option; individual
phase estimates may be highly variable. In addition, individual phase estimates
are not statistically independent. It is also difficult to evaluate whether differences
in individual phase estimates are real or due to methodological differences. That
is, differences may be due to changes in sampling methods or data collection
methodology over time. At this time, there is no valid statistical test for
examining differences between Phase 1 and Phase 2. Therefore, although point
estimates can be produced separately for each phase, no test is available to test
whether those estimates are significantly different from each other."
E) Issues of confidentiality.
In order to protect subject confidentiality, NCHS will not provide certain information
(e.g., subject address or location, date of birth) with the publically available data. For
example, in NHANES-III, NCHS identified the county of residence, but only if the
county population exceeded 500,000. NCHS is even considering withholding some of
the NHANES99+ biomonitoring data that were obtained from partial samples of the
survey. Because of issues of confidentiality, it may be impossible to use the publicly
available data to conduct some analyses that are of interest to EPA. For example, this
would include studies in which it is necessary to know the location of the subjects and
studies that need to link NHANES to other databases, such as exposure and mortality
databases. Studies requiring confidential data should not be dismissed before contacting
the NCHS RDC to determine if they are feasible if conducted through the Center (See
Section 5.2 for details about conducting studies through the RDC).
F) Variable content within surveys.
In each of the NHANES, some variables (e.g., height, weight, age) are measured over the
entire study and some are measured during only part of the survey. For example in
NHANES-III, blood urea nitrogen was only measured in Phase II, but not hi Phase I.
Some variables are measured for all subjects (e.g., lead in blood was measured for all
persons aged 1-74+ years hi NHANES-III), and some variables are only measured for
certain subgroups (e.g., allergy skin reactivity tests were administered to all persons aged
23
-------
6-19 years and a random half-sample of the adults aged 20-59 years in NHANES-III).
Some variables are measured for only a random sainple within the survey (e.g., 1/3 of all
subjects ages 12+ years have dioxins measured in serum for NHANES99+). Different
data are collected for specific age groups and sometimes on the basis of gender (see
Appendices IV, VI and VII). Note that as subjects Iget older, the number of chemicals
tested in the blood and urine increases simply because the volume of body fluids that can
be obtained increases. It is important to understand the variability in the content of a
specific NHANES in order to design an analysis properly and to ensure that adequate
numbers of subjects will be in each class or cell, Note that NCHS cautions against
conducting an analysis with fewer than five subjects in a cell, as indicated by NCHS on
their RDC website . It is beyond the scope of
this Handbook to provide a detailed discussion of this complicated survey design. The
reader is referred to the following references for more complete discussions of the
NHANES sample design. References for NHANES-I: NCHS (1973,1977,1978); for
NHANES-H: NCHS(1981); for HHANES: NCHS |(1985); for NHANES-III: NCHS
(1994b, 1996c). The NCHS Documentation Books provide the counts of subjects for
each variable measured in any survey. During me preliminary phases of designing a
study, one can use these counts to determine if there are sufficient numbers of subjects to
support the desired analysis. Documentation books can be downloaded for each survey
from the following website .
G) Variable content and methods across surveys.
Given that data have been collected through NHAl^ES for about 30 years and there are
many variables that are measured across all the suiyeys, it may be intriguing to consider
conducting some kind of time trend analyses. There are many precautions to consider
before attempting such analyses, including the following: a) NHANES are cross-sectional
studies and new subjects are recruited for each survey; b) While many variables have .
been measured across surveys, there is no guarantee that the measurement methods and
equipment are the same and that test results are diilectly comparable. For example,
NHANES-I, -II and -ILL measured lung function b]r spirometry; however, each survey
used different kinds of spirometers so it is not certain that lung function test results can be
compared across surveys, or at least without some kind of adjustment. NCHS noted in
their website that data
collection for NHANES-E was made comparable to NHANES-I in order to establish a
baseline for assessing changes over time. This means that in both surveys many of the
same measurements were taken in the same way, on the same age segment of the U.S.
population. A brief discussion of all NHANES can be found at the above noted website;
c) Each survey is conducted at numerous location^ throughout the U.S. (see Section 3.0)
and the exact locations are not necessarily repeated from survey to survey. There are
many variables (e.g., allergies, respiratory effects) that probably are affected by
geographical location because of weather, altitude, humidity, degree of urbanization,
pollution, etc. If you want to examine time trends', you will need to know which variables
are sensitive to location of the subject and consideir if this is going to be a problem; d)
!': f-\
24
-------
Differences in sample sizes and designs for each survey (see Sections 3.0,4.0) must be
considered when comparisons are made across the various surveys. For example of the
first four surveys, NHANES-III was the only survey that included persons 75 years or
older, and NHANES-1 and -II did not include any oversampling of Hispanics. Although
the different surveys have similar analytic objectives, the differences in their sample sizes
and designs will cause estimates to differ in reliability across the surveys (NCHS, 1996c).
9.0. ANALYSIS ISSUES
When analyzing NHANES data, it is important to take into account the complex survey
design and the sample weights (see Sections 3,4, and 8). Sample weights are needed to estimate
means, medians, and other descriptive statistics. The weights must be used to produce correct
population estimates because each SP does not have an equal probability of selection into the
survey. The sample weights incorporate these differential probabilities of selection and include
adjustments for noncoverage and nonresponse. Further, with the large oversampling of specific
subgroups in each survey (e.g., young children, elderly, blacks, and Mexican-Americans in
NHANES-III, also see Table 1), it is essential that sample weights be used in analyses, otherwise
there is a risk that the results of the analysis will be incorrect. The data analysis must also take
into account the strata and PSUs from the survey design used to estimate variances and test for
statistical significance. In general, sampling variances will be underestimated if calculated
without incorporating the complex survey design into the analysis (NCHS, 1994b).
Although preliminary analyses may be performed on unweighted data and with standard
statistical packages that assume simple random sampling, the .final analyses should be done on "
weighted data using special computer programs (e.g.,, SUDAAN, PCCARP) that employ
appropriate methods for estimating variances from complex samples.
NCHS recently developed guidance for sample sizes needed to analyze complex survey
data, such as NHANES (see Table 1 in Appendix I). This guidance takes into consideration
several issues, including the survey design and the rarity of the event (e.g., disease, physical
condition) being studied.
9.1. Weighting
The purpose of weighting sample data is to permit analysts to produce estimates of
statistics that would have been obtained if the entire sampling frame had been surveyed (NCHS, .
1992b). The "sampling frame" is an operational definition of the target population, which in the
case of NHANES is the. civilian, non-institutionalized U.S. population. The specific definition of
"entire sampling frame" varies among the individual NHANES surveys. For example, in
.NHANES-III, the entire sampling frame was the civilian, noninstitutionalized U.S. population
ages two months and older. A sample weight can be thought of as the measure of the number of
persons that a particular sample observation represents.
25-
-------
Each NHANES has its o\vn set of weights and many are designed to accomplish the same
objectives across the surveys. The reader should refer to NCHS,(1975,1982,1992b and 1996c)
for much more thorough discussions of how weights were jcomputed for NHANES; how
adjustments were made for non-response and poststratification; and how weights were calculated
for specific survey components. To familiarize the readerjwith some of the issues involved in
developing and using weights, the following information is provided based on the weighting used
inNHANES-m; however, it is applicable to all of the surveys. Sample weighting inNHANES-
ni was used to accomplish the following objectives:
1) To compensate for differential probabilities of selection among subgroups. For
example members of different age-sex-race/ethnkjiry subgroups may be sampled at
different rates and even oversampled. Persons living in different geographic regions also
may be sampled at different rates;
2) To reduce biases arising from the fact that nonrespondents may be different from those
who actually participate in the survey; ;
3) To bring sample data up to the dimensions of tt[e target population.totals (e.g., it one
wants to use NHANES data to calculate national statistics, then the target population is
the whole U.S. population);
4) To compensate, to the extent possible, for inadequacies in the sampling frame. 1 nese
inadequacies may result from omissions of some housing units in the listing of area
segments; omissions of persons with no fixed address, etc.; and
5) To reduce variances in the estimation procedure by using auxiliary information that is
known with a high degree of accuracy.
(NCHS, 1996c and 1992b).
The procedure for weighting is summarized in the
(NCHS, 1996c):
following paragraphs as presented in
"The sample weighting was carried out in three stages. The first stage involved the
computation of weights to compensate for unequd probabilities of selection (Objective 1
above). The second stage adjusted for nonresponsje (Objective 2). The third stage used
poststratification of the sample weights to Census Bureau estimates of the U.S.
population to simultaneously accomplish the third, fourth, and fifth objectives. Due to
the form of estimators typically used with data from complex samples, extreme variability
in the weights may result in reduced reliability of the estimates. The NHANES-III
sample was designed to minimize the variability :in the weights, subject to operational and
analytic constraints. When analyzing NHANES
-------
estimates based on the data would be close to unbiased estimates of statistics for the total
U.S. population. However, nonresponse occurs in any survey operation, and nonresponse
bias may result. All persons selected in the sample were asked to participate in a personal
interview at their home, where medical history and socio-demographic information were
collected. After the initial interview, all interviewed persons were invited to the MEC for
a physical examination. Persons who were unable to come to the MEC were offered an
abbreviated physical examination at their home (Note: this was only done in NHANES-
III). Thus, nonresponse in NHANES can occur at several stages of the data collection
process: a) some of the SPs who were screened refused to be interviewed (interview
nonresponse); b) some of the interviewed SPs refused the medical examination (exam
nonresponse); and c) some of the SPs participated at the MEC but refused specific tests."
"The issue of weighting is further complicated by the procedure of poststratification.
Poststratification of sample weights to independent population estimates is used for
several purposes. In most household surveys, certain demographic groups in the U.S.
population (e.g., young black males) experience fairly high rates of undercoverage in
survey efforts. Poststratification to Census estimates partially compensates for such
undercoverage and for any differential nonresponse, and can help to reduce the resulting
bias in the survey estimates. Poststratification can also help to reduce the variability of
sample estimates as well as achieve consistency with accepted U.S. figures for various
subpopulations, and bring the weighted totals up to the level of the presumed total
civilian noninstitutionalized U.S. population."
As noted above fromNCHS, 1996c, one needs to be aware of the important issue of large
weights and particularly large weights in combination with extreme values (e.g., unusually high
concentrations of pesticides in blood or urine of a specific subject). Either one of these
conditions may unduly dominate the analysis and result in questionable, or inappropriate,
conclusions. A good practice during the initial exploratory data phase would be to screen the
weights to identify very large, influential values. It may also be useful to plot the values of the
weights against the values of the analytes (e.g. concentrations of specific chemicals in blood and
urine) to identify influential outliers. The results of these screening exercises should be used to
help make a determination of how the observations associated with these weights should be
handled in your analysis.
The following table, based on NHANES-III, illustrates the complexity of the survey
weighting scheme, which includes not only overall interview and exam weights, but also weights
for specific tests (e.g., allergy, and central nervous system (CNS)). NCHS survey
documentation, which is part of the publicly available information, identifies the various weights
and notes when it is appropriate to apply them.
Weight
Final interview weight
Application
Use only in conjunction with the sample interviewed at
home, and only with items collected during the
household interview.
27
-------
Final exam (MEC only) weight
Final MEC+Home exam weight
Final Allergy weight
Final CNS weight
Final Standard exam (MEC only) weight
Final Modified exam (MEC only) weight
Final Standard MEC+Home exam weight
Final Modified MEC+Home exam weight
(NCHS, 1996c)
Use only in ccmjunction with the MEC examined
sample, and only with interview and examination items
collected at the MEC.
Use only in cctojunction with the MEC+Home examined
sample, and only with items collected at both the MEC .
and home. ;
Use only in conjunction with the Allergy subsample, and
only with items collected as part of the allergy
component of the exam.
Use only hi cc>njunction with the CNS subsample, and
only with items collected as part of the CNS component
of the exam, j
Use only in conjunction with the MEC examined
persons assigned to the Standard subsample, and only
with items collected at the MEC exam. These weights
should be used to analyze tests such as the Oral Glucose
Tolerance Tests (OGTT), where overnight fasting is
preferred.
Use only in conjunction with the MEC examined
persons assigned to the Modified subsample, and only
with items collected at the MEC exam.
Use only in conjunction with the MEC and home
examined persons assigned to the Standard subsample,
and only with items collected during the MEC and home
examinationsJ
Use only in ccaijunction with the MEC and home
examined pensions assigned to the Modified subsample,
and only with items collected during the MEC and home
examinations.!
pi,;'
j," ~
9.2. Issue of Nonresponse, Bias
As noted above, nonresponse is a key issue with atiy survey, including NHANES. It is
important to assess nonresponse bias in the analyses of data from complex samples such as
NHANES. As indicated in NCHS (1994a), the assessment should include the following
sections that focus on item nonresponse: | ' .
I
i
a) A definition of item nonresponse and the level of nonresponse;
b) A description of the effect of item nonresponse on statistics of interest;
c) Assumptions used in the adjustment methodology; and
d) An assessment of the imputation procedure and [the impact of nonresponse adjustment
on survey estimates.
' ': i''li. " I "1" i. i i"; . . ' ' P ,'
NCHS (1994a) also notes that in addition to reporting nonresponse rates, to the extent possible,
any analysis of nonresponse should include information oil reasons for missing data (e.g., unable
28
,
in > 11
t" i
if "i f,
, 'H'li
i II
.$ i
-------
to contact subject, medical reasons, refusals, etc). Analysts should study the reasons for
nonresponse at each stage of sampling. The researcher also needs to determine if the non-
response is random or systematic. If there is systematic non-response by measured covariates
(e.g., by specific demographic variables such as race/ethnicity or age), then the analyst needs to
make an adjustment for it, for example, by reweighting the data. This information is valuable for
diagnostic purposes in the evaluation of nonresponse bias.
It is beyond the scope of this Handbook to provide a detailed discussion of how to
assesses and correct for subject nonresponse. The reader is referred to the following references
for more complete discussions on how nonresponse is denoted in the surveys and how to adjust
for nonresponse. For NHANES-I see NCHS(1973,1977,1978); NHANES-II see NCHS(1981);
HHANES see NCHS(1985); NHANES-HI see NCHS(1994a,b and 1996c). .
10.0. SUMMARY AND CONCLUSIONS
! This Handbook has provided a detailed overview of the purpose, history and content of
NHANES; how to access the data and documentation; how NHANES data have been used by
EPA and others to support public and environmental health studies and federal policies and
regulations; and suggestions on how the data can be used in studies to support risk assessment,
policy and regulatory needs at EPA. This document has also identified limitations and data
analysis issues that need to be considered when designing a study using NHANES data. Despite
the limitations and complex design of this survey, it is clear that NHANES is a unique, rich
database that offers a tremendous amount of human health, nutrition, and exposure information,
and will continue to do so into the future. Since the start of NHANES thirty years ago, EPA has
participated in the design and support of each survey, but has only used the data for limited
research, policy and regulatory needs. It is hoped that by informing staff about NHANES, this
Handbook will encourage efforts to "mine" the data to support Agency heeds across the Program
Offices. It is also hoped that innovative approaches (e.g., using GIS; linking NHANES to
available databases such as the National Death Index), will be tried in order to analyze the data in
new ways that produce information that is useful to the mission of the Agency. Now that NCHS
has established their Research Data Center, it should be possible to conduct studies that were
impossible in the past because of lack of access to sensitive data. Finally, more thought should
be given to designing and conducting studies that make use of the subjects' biological samples
(blood, urine, saliva) stored by NCHS. These samples offer a rare opportunity to study potential
biomarkers of exposure and/or effects on a national sample of the U.S. population and to be able
to link the data to health, nutrition, exposure and socioeconomic data already collected in the
baseline surveys. .
29
-------
TABLE 1: Summary of Selected Design Features of tie NHANES
SURVEY'"
NHANES-1^
NHANES-n
Hispanic
HANES
(HHANES)(0)
NHANES-IH(d)
NHANES99+(C)
DATES
of
SURVEY
1971-
1975
1976-
1980
1982-
1984
1988-
1994
started
March
1999
AGES
EXAMINED
1-74 years old
6 mo.-74 years
old
6 mo. -74 years
old
2 mo. +
all ages
# OF SlIBJECTS
INTERVIEWED/
(EXAMINED)
28,0437(20,739)
27,8017(20,322)
15,9317(11,672)
39,6957(30,818)
5,000/3|r.
NUMBER of SURVEY
LOCATIONS (Stands)
100
64
17 in the southwest;
9inNY,NJ,CT;
4inFL
89
15 per year
(Source: NCHS, 1994b; )
Different surveys oversampled different subgroups as follows: NHANES-I, II (low income, children, elderly).
NHANES-I also oversampled women ages 20-44 years. HHANES oversampled age groups 6mo.-19 yrs and 45-74
yrs old. NHANES ffl oversampled Black-Americans, Mexican-Americans, infants and young children (1-5 years)
and older persons (60+years). j
w The first segment of NHANES-I was conducted from 1971-1974 aid was followed by a 14 month period from
1974-1975 in which an additional national sample of people 25-74 yejars of age was examined to augment the size of
the original sample. This additional sample is called the NHANES-I Augmentation Survey of Adults 25-74 years.
(See for more information)
-------
Table 2. Summary of the NHANES-I Epidemiological Followup Studies
DATES of
FOLLOWUP
Phase-I:1982- 84
Phase-H: 1986
Phase-IH: 1987
Phase-IV: 1991
AGES EXAMINED
aUSPs 25-74 yrs old at
baseline survey
SPs 55-74 yrs old at
baseline survey & not
deceased at Phase-I
all non-deceased SPs
all non-deceased SPs
NUMBER
OF SPs
14,407
3,980
11,750
11,195
METHODS & DATA COLLECTED
face-to-face SP interview; pulse rate; weight;
BP; self-reported conditions; hospital &
nursing home records; death certificates
phone or mail interview; no physical
measurements; hospital & nursing home
records; death certificates
phone or mail interview; no physical
measurements; hospital & nursing home
records; death certificates
interviews primarily by telephone; no physical
measurements; health care facility abstracts;
death certificates
(Sources: NCHS, 1987,1990,1992a, 1997)
31
-------
-------
APPENDIXI
NHANES in Sample Design arid Analysis Guidelines
(Note: The material in this Appendix is taken verbatim from the NCHS source document. A list
of the references cited in the original NCHS document has not been included in this Appendix.
The reader is referred to the source document for the full citations.)
(Source: NCHS, 1994b, pp.20-22)
1-1
-------
:; f: :
SAMPLE DESIGN AND ANALYSIS GUIDELINES
Sample design ;
The general structure of the NHANES IE sample design is;the same as that of the previous
NHANES. Each of these surveys used a stratified multistage probability design. The major
design parameters of the two previous NHANES and the special Hispanic HANES, as well as
NHANES HI, have been previously summarized (17). i
The NHANES m sample was designed to be self-weighting^ within a primary sampling unit
(PSU) for subdomains and fairly close to self-weighting nationally for each of these subdomain
groups (but not for the total population). The NHANES ID. sample represents the total civilian
noninstitutionaUzed population, 2 months of age or over, in me 50 States of the United States.
The first stage of the design consisted of selecting a samplb of 81 PSU's, which, in the first stage,
are mostly individual counties. In a few cases, adjacent coiimties were combined to keep PSU's
above a minimum size. The PSU's were stratified and selected with probability proportional to
size (PPS). Thirteen large counties (strata) were chosen with certainty (probability of one). For
operational reasons, these 13 certainty PSU's were divided into 21 survey locations. After the 13
certainty strata were designated, the remaining PSU's in the United States were grouped into 34
strata, and 2 PSU's were selected per stratum (68 survey locations). The selection was done with
PPS and without replacement. The NHANES ffl sample therefore consists of 81 PSU's or 89
locations.
The 89 stands were randomly divided into 2 sets, 1 consisting of 44 sites and the other 45 sites.
One set of PSU's was allocated to the first 3-year survey period (1988-91) and the other set to
the second 3-yeaf period (1991-94). Therefore, unbiased estimates (from the point of view of
sample selection) of health and nutrition characteristics can be independently produced for both
Phase 1 and Phase 2 as well as for both phases combined.
For most of the sample, the second stage of the design consisted of area segments composed of
city or suburban blocks, combinations of blocks, or other area segments in places where block
statistics were not produced in the 1980 census. In the first phase of NHANES El, the area
segments were used only for a sample of persons who lived m housing units built before 1980.
For units built in 1980 and later, the second stage consisteld of sets of addresses selected from
building permits issued in 1980 or later. These are referred to as "new construction segments."
In the second phase, 1990 census data and maps were used to define the area segments. Because
the second phase followed within a few years of the 1990jcensus, new construction did not
account for a significant part of the sample and the entire [sample came from the area segments.
The third stage of sample selection consisted of households and certain types of group quarters,
. such as dormitories. All households and eligible group quarters in the sample segments were
listed, and a subsample was designated for screening in oMer to identify potential sample
persons. The subsampling rates enabled production of a national, approximately equal,
1-2
f :;
I
R, :!';!; ! I, : !!!! I
-------
probability sample of households in most of the United States, with higher rates for the
geographic strata with high Mexican-American populations. Within each geographic stratum,
there is an approximate equal-probability sample of households across all 89 stands. The
screening rate in each stratum was designed to produce the desired number of sample persons for
the rarest age-sex domain in the race and ethnic group defining the geographic stratum.
Persons within the sample of households or group quarters were the fourth stage of sample
selection. All eligible members within a household were listed, and a subsample of individuals
was selected based on sex, age, and race or ethnicity. The definitions of the sex, age, race or
ethnic classes, subsampling rates, and designation of potential sample persons within screened
households were developed to provide approximately self-weighting samples for each subdomain
within geographic strata and at the same time to maximize the average number of sample persons
per sample household. Experience in previous NHANES indicated that this increased the overall
participation rate. Although the exact sample sizes will not be known until data collection has
been completed, estimates have been made. A summary of the expected sample sizes at each
stage of the design is as follows:
Number of PSU's 81
Number of stands (survey locations) 89
Number of segments 2,138
Number of households to be screened 106,000
Number of households with sample persons 20,000
Number of sample persons 40,600
Number of interviewed sample persons 35,000
Number of examined sample persons. 30,100
A more detailed description of the sample design for NHANES m, including a description of the
research that resulted in the final design, has been previously published (17).
Analysis guidelines
Because of the complex survey design used in NHANES HI, traditional methods of statistical
analysis based on the assumption of a simple random sample are not applicable. Detailed
descriptions of this issue and possible analytic methods for analyzing NHANES data have been
described previously (7,79,134,135). These previously recommended guidelines are revised on a
periodic basis as new statistical procedures and analytic computer software are developed.
However, there are some important analysis considerations that have not changed over time.
First, there are the two aspects of the NHANES design that must be taken into account in data
analysis. One is the sample weights and the other is the complex survey design. Sample weights
are needed to estimate means, medians, and other descriptive statistics. They must be used to
produce correct population estimates because each sample person does not have an equal
probability of selection. The sample weights incorporate these differential probabilities of
selection and include adjustments for noncoverage and nonresponse. With the large
oversampling of young children, older persons, black persons, and Mexican-Americans in
1-3
-------
NHANES HI, it is essential that the sample weights be used in all analyses. Otherwise,
misinterpretation of results is highly likely. ;
The second aspect of the design that must be taken into account in data analysis is the strata and
PSU's from the sample design used to estimate variances and test for statistical significance. In
general, sampling variances will be underestimated if calculated without incorporating the
complex sample design. '
The effect of the complex sample design on variance estimates is referred to as the design effect,
which is the ratio of the variance of a statistic from a complex sample to the variance of the same
statistic from a simple random sample of the same size (3). A design effect of one indicates the
equality of the simple random sample variance and the complex sample variance.
Design effects in NHANES have traditionally been higher! than one, and the magnitude of the
design effects have been variable. In NHANES I and NHANES IT, the average design effect was
calculated to be about 1.5. Preliminary analyses from NHANES m indicate that the average
design effect might be lower (approximately 1.2 or 1.3).
Although preliminary analysescmay be performed on unweighted data with standard statistical
packages that assume simple random sampling, final analyses should be done on weighted data
using special computer programs that use an appropriate method for estimating variances from a
complex sample (e.g., SUDAAN(136) orPCCARP (137)). The calculation anduse of "average"
design effects (when unstable variances occur) along with the sample weights have been
suggested as an alternative NHANES analytic approach (135). Recently, NOES staff have
participated in an effort to establish guidelines for varianc^ estimation and statistical reporting
standards. In addition to delineating some of the previously mentioned issues, the staff produced
anomogram of recommended sample sizes for analyses of'complex survey data (table 1). For
means of fairly symmetric populations and proportions based on commonly occurring events
(where 0.25 < p < 0.75), a good rule of thumb is that the saimple size should be no smaller than
some broadly calculated "average design effect" times 30. The first column of the table
represents a simple random sample design and the other ccjlumns reflect the increased sample
size requirements for a more complex survey design. Thus] the minimum sample size for a
normal approximation increases for more rare events as well as for survey designs that result in
increased average design effects. Other criteria and approaches for estimating minimum sample
sizes exist; however, this is the approach currently proposed for NHANES IH analyses.
i
1-4
-------
1
_o
o
a
S
a.
o
a>
s=
o
ฃ
(0
ปo
(O
13
ฃ
"S
c
"w
CD
TJ
L.
0!
"55
o
T3
ฃ
10
S
a
for a conr
(0
0>
N
0)
(0
0)
a
in
0
CO
3
o
CM
a>
5
i
^
CO.
in
C"
CO
CM
c: '
o
Proportion
o
o
co_
o
I
o
o
o
CM
o
8.
t
o
s
CO
"-
8
o
o
CM_
o
CM
V-
O
sf
T
O
CO
en
g
CO
o
o
CO
en
d
ง
in
1
o
CO
o
CO
CO1
CM
CM
1
1-
i
CO
o
CM
CO1
CD
O
CO
cS
d
8
s
o
8
8
2
CD
T
ง
8
T-
S
8
CO
CO
8
o
d
'c5
8
CO
co
ฃ
0
CO
en
a>
g
8
in
8
S
co
in
S
s
d
o
S
o
o
8
CD
CM
CD
S
8
CO
in
S
9
'^
o
8
d
C\j
CO
en
8
S
CD
CO
in
,_
00
in
CM
R
10
CO
CM
d
in
o
o
en
12
8
In
in
in
CO
in
<*
CM
o>
9
CO
CO
o
d
d
ง
ง
12
o
CO
fe
in
in
CO
in
CM
CO
CO
CO
8
o
18
d
S
8
12
o
CD
in
S
in
9
IO
CM
CO
CO
CO
CO
g
o
o
d
s
ง
in
o
CO
In
S
in
eo
9
CM
co
CO
CO
S3
o
CO
in
d
ง
ง
m
8
te
8
in
9
CM
O
co
CO
CO
CO
CO
o
CO
| 0.26-0.44
CM
T
CO
CO
o
CO
s
CO
CO
in
5
,
in
CM
CO
CO
in
CO
CM
CO
in
CM
d
o
*
1
8
8
CO
CM
CO
s.
8
to
m
CM
in
CO
^.
o
o
CM
d
fe
o
CO
'I
.|
o
CD
CO
CO
in
CO
o
CO
m
r
S
in
S
in
d
8
O
1
O
CO
s
2
CO
T
00
CM
T
ฃ
o
eg
CO
CO
ง
o
d
s
S
5
o
s
co
i
CM
co
o
CM
CM
1
i
CD
r
8
g
d
0
o
CM"
CM"
1
o
CM
in
|
o
CD
T~
O
CO
0
o
(N
O
CM
T
1
O
CO
CO
1
1
T-
0
d
S
J_ซ
-------
These guidelines reflect a design-based approach to estimation and analysis. In some instances, a
model-based approach may be used. The use of an "average design effect" to estimate the
complex sample variances is one such instance. The use of model-based approaches is most
appropriate when maximizing use of all available data is preferable (138,139).
"l ' !'
It is important to remember that guidelines are just that, arid they are not absolutes. They
represent strategies that yield the most sound statistical conclusions. Violating the guidelines
introduces a greater degree of uncertainty about the soundness of the analytic conclusions but
does not necessarily mean that a particular analysis is invalid. Consideration of the survey design,
survey nonresponse, data collection and processing procedures, potential measurement errors,
and the subject matter being studied are all equally important and should be evaluated to judge
the merit of each analysis and interpretation of data from siny survey, including NHANES HI.
...
1-6
til
jl-jiS,!!-
;. 11
-------
APPENDIXII
NHANES JE Health States Assessment
CNote: The material in this Appendix is taken verbatim from the NCHS source document. A list
of ft* references dted m the original NCHS document has not been included in this Appendix.
The reader is referred to the source document for the full citations)
(Source: NCHS, 1994b, pp. 3-13)
H-l
-------
HEALTH STATUS ASSESSMENT !
Because of the variety and complexity of me data collected in NHANBS HI, the
information in this section is presented in different ways, The first part provides overviews of
fm^of the main areas of special interest addressed in NIEANES El and highlights some of the
pubHc healm and scientific issues covered by me survey in each of these areas. The second part
provides a more detailed account of the data collected by examination and interview for^each of
Se ma^r health topic areas included in the survey. The mird part describes me risk factorand
health behaviors measured in NHANES HI, and the fourth^ part describes three special studies.
Health of population subgroups and topics of special interest
In this section some of the major contributions made by NHANES ffl in assessing and
monitoring the health of population subgroups of interest, including children and adolescents, the
Sderiy, women, and minorities, are described. Also covered are contributions,m providing
information on special topics, including environmental and occupational health and the
Segment of health care coverage and needs. NHANESm is a national survey designed to
Sleet information to assess the health status of^ ^^.S. c^, no^^^
population. Within that framework, however, the survey was also designed to sample large
numbers of young children, older persons, black persons, and Mexican- Americans, so that
reliable estimates of health status can be produced for thefse population subgroups.
Child and adolescent health
NKANES IH is the first NHANES to include children as young as 2 months of age. The
survey was designed to oversample children aged2monfc to 5yeirs so new growth charts
cSbe created for use in assessing children's growth afd development. To increase the
response rates among infants aged 2-11 months, fte option of a home examinations offered to
parents unwilling to bring veiy young children to the mobile examination center (MEG).
NHANES m data on child health are relevant to many key areas of public health for ^
children in the United States. Environmental lead exposure and progressm ซ^^^ *
lead exposure were assessed by questionnaire data and n easurements of blood lead levels for all
Srenl year of age and over. Irntoation on cMldrer's exposure to
glX questionnaire data on smoldng by household member
serum cotinine levels. Measurements of hepatitis B markers, tetanus antitoxm, diphtheria
antibodv and rubella antibody levels serve to assess and, monitor immunization levels in
!^SซrfNHAl^md^^
detail in the section "Nutritional heallh assessment." AJ so relevant to child health is the
knowledge gained from NHANES El about the health status of women of childbeanng age,
including folate status and susceptibility to the rubella virus.
n-2
E . .1
-------
The content of the survey varied for children of different ages. However, in general, for
infants and young children (2 months to 5 years of age), the survey included information on oral
health, growth, and motor and social development. For older children and adolescents,
NHANES IE collected data on varied conditions that include asthma and allergy, pulmonary
function, oral health, hearing, cognitive function, blood pressure, and stage of sexual maturation
as well as questionnaire data on physical activity and tobacco use and many laboratory
determinations. In addition, adolescents were asked in private interviews about tobacco, drug,
and alcohol use, reproductive history, and mental health.
Health of older persons
NHANES IE is the first NEANES to include persons 75 years of age and over. In order
to address scientific and policy issues pertinent to the older population in the United States,
NHANES ffl included an oversample of older persons (aged 60 years and over). To minimize
nonresponse in older persons, a home examination was developed for those persons who were
unable or unwilling to come to the MEC for a complete examination. This home examination
included an abbreviated set of measures parallel to those performed in the MEC.
The survey content of NHANES m is particularly useful for the study of the contribution
of multiple diseases to disability in old age. As covered in other sections of this report, this
content included nutritional status, cardiovascular disease, pulmonary disease, dental disease,
diabetes, retinopathy, osteoarthritis, and osteoporosis. Besides these specific diseases, the survey
included measures of functional status in older persons to ascertain the prevalence of disability
and limitations in function and the correlations of patterns of disease with functional health
status'.
The survey addressed three major areas of function: social, cognitive, and physical
function. Much of the content of NHANES HI in these areas was shaped by a special workshop,
"Innovations in the Measurement of Function for Older Persons: A Focus on National Surveys,"'
held in September 1985. . '
Minority health
NHANES HI is the first NHANES to include planned oversampling of the two largest
minority groups in the United States. The black and Mexican-American populations were
oversampled to obtain statistically reliable estimates for the two largest minority groups in the
United States. In previous national surveys, although these groups were included in the sample
according to their representation in the national population, sample sizes were often too small to
provide adequate estimates. As a result, it was decided to include planned oversampling of these
two groups in NHANES ffl.
The content of the examination is targeted to the national population as a whole and to
specific age ranges, rather than to specific minority groups. However, many health conditions
n-s
-------
studied in the survey occur at high rates in minority populations, including diabetes mellitus
among Mexican-Americans and hypertension among black persons. The survey provides
extensive data for minorities on chronic diseases, renal function, pulmonary function,
environmental exposures, immunization status, risk factors, and health behaviors. In addition,
becTs^e survey included oversampling of children and the elderly, NHANES HI provides
information on the health of black and Mexican-America*! children and older persons. In many
of these areas, NHANES HI provides the first comprehensive national data and reference
inta!l?K Further, NHANES ffl data allow for
valid comparisons among different race-ethnic groups because data were collected in a
standardized manner for all survey participants. Race-ethluc groups were define-lbased upon
combinations of the reported race and reported ethnicity o|f survey participants. These data can
be used to provide insight into the causes and concomitants of the disparities in health status
among race-ethnic groups ha the United States.
Women's health ;
Women's health has traditionally referred primarily to issues related to reproduction. In
recent years there has been aheightened awareness that women's h^en^mP^
-------
extensively to assess the extent of exposure to lead in the United States, to identify correlates of
exposure (e.g., urbanization, age), to monitor trends in exposure, to support policy and regulatory
decisions regarding lead in gasoline, and to identify health effects resulting from lead exposure
(e.g., blood.pressure elevations, diminished height in children) (19). It is expected that the lead
data from NHANES IE will serve a similar purpose during the 1990's. Cadmium, a tbxicologic
concern ranking close to that of lead (20), was measured for the first time in NHANES IE.
In 1985, the Meragency Committee for Indoor Air Quality identified NHANES HI as
providing an important opportunity to examine the relation of indoor pollutant exposures to
potential health effects on a national basis (21). Although levels of indoor air pollutants were not
directly measured in the NHANES HI homes, data were collected on housing characteristics,
water sources, and cooking and heating fuel systems. Information on health outcomes related to
the indoor environment included data on respiratory symptoms and smoking history,
measurements of allergic reactivity to mites (house dust), pulmonary function testing, and the
physician's assessment of bronchial sounds. Tobacco smoke is a significant indoor air pollutant,
and passive exposure to tobacco smoke has been determined to be a major health hazard by the
Environmental Protection Agency (22). hi NHANES IE, serum cotinine levels were measured to
determine active and passive exposure to tobacco smoke and other sources of nicotine.
A report by the House Committee on Government Operations entitled "Occupational
Health Hazard Surveillance12 Years Behind and Counting" (23) provided the impetus for the
NHANES ni collection of more data related to occupational health. Three examination
components were designed to assess potential health effects that may result from occupational
exposure: the neurobehavioral evaluation system (NES) for central nervous system testing, the
physician's examination for assessing hand blistering and redness, and the spirometry test for
measuring pulmonary function. Questions on current and longest held occupation, use of
protective equipment at work and passive exposure to smoke at work were also included.
Health care coverage and health care needs ,
NHANES HI provides a unique opportunity to assess the prevalence of unrecognized
disease and unmet health care needs. Because this is an examination survey, it allows for
objective determination by examination and measurement of many health conditions. Thus it is
possible to assess the degree to which these conditions are recognized and the implications for
health care needs.
For example, NHANES HI measurements of blood pressure, combined with information
on prior diagnosis and on the use of antihypertension medications or nonpharmacologic therapy
can be used to estimate the extent to which persons with high blood pressure are aware of then-
condition, the extent to which those who are aware are receiving treatment, and the extent to
which those receiving treatment have reduced their blood pressure to acceptable, levels. National
data on blood cholesterol levels from NHANES HI were used to estimate the numbers of people
requiring intervention and treatment under the Adult Treatment Panel guidelines of the National
n-5
-------
Cholesterol Education Program and to monitor changes since similar data were collected in
NHANES n. Dental examination data can be used to describe the extent of population needs for
dental care, the extent to which existing conditions have been treated, and, coupled with
information on dental care utilization, the degree of access to dental care for people with
differing needs. Immunization data obtained from blood samples can be used to assess the level
of protection in the population.
As part of NHANES EL, data on health care utilization, health insurance coverage for all
family members, and income assistance, including Medicaid, Social Security, and Supplemental
Security Income, were collected to assess health care needs and participation in public assistance
programs. Participants were asked detailed questions about coverage by Medicare, other forms
of health insurance or reasons for lack of coverage, and abput their use of health services and
medications, established relationships with providers, andjhistory of heaWh conditions and
hospitaUzations. These data can be used in conjunction With the other information collected in
the survey to determine the relationships between access to care and health status.
Health status components
This part provides a brief account of the data collected by examination and interview for
each of the major target conditions and physiological measurements in NHANES Hi. hi the
survey, data were collected on dietary intake and nutritional status (described in the section
"Nutritional health assessment"), anthropometric measurements (described in that same section),
reproductive history and sexual behaviors, use of vitamin jand mineral supplements and
medications, tobacco and alcohol use, physical activity, arid sociodemographic characteristics.
These data, although not mentioned specifically, are relevant to many.of the components. A list
of topics included in the questionnaires administered during the household interview and hi the
questionnaires and procedures administered in the examination can be found in appendix tables I
and II. Other summary information is included in appendix I and appendix tables m-XII and the
data collection forms are in appendixes m and IV. <
I
The examination teams, described more fully in the section "Data collection and reports
of findings," included a physician, a dentist, a certified ulttasound technician, health technicians,
medical technologists, a phlebotomist, a health interviewer, and dietary interviewers, as well as
other personnel. Except as noted under specific components, a health technician administered all
Jr -t i * j^ , ^^^
MEC examination procedures, and the health interviewer administered the MEC adult, youth,
and proxy questionnaires. Examinees were excluded from each of the examination components
for specific safety, health, or logistical reasons. These exclusion criteria are specified in
appendix table VOL i
Fasting instructions were common to all components. For morning examinations,
examinees aged 12-19 years were instructed to fast at leasit 8.5 hours preceding the examination,
and those 26 years of age and over were instructed to fast ^2 hours. For afternoon or evening
examinations, all examinees 12 years of age and over wens instructed to fast for 6 hours
E-6
' '!' ' it,:?!'1.! ('
'( t,]
pit-
|L 'I
-------
preceding the examination. Children under age 12 and persons of any age who reported using
insulin were instructed not to fast.
Many laboratory determinations were conducted on blood and urine specimens obtained
during the MEC examination. Most of these determinations are mentioned briefly under the
relevant topic headings. However, the laboratory methods are not described in any detail in this
section. Full procedural descriptions of the laboratory methods are available from NCHS, and a
summary of the assay methods is provided in appendix I. For the convenience of the reader, the
laboratory analyses conducted in all three of the NHANES and the Hispanic HANES are
provided in appendix table VI. A complete list of all the laboratory determinations on blood and
urine specimens can be found in appendix.table IV. The laboratories and diagnostic centers are
also listed in appendix table El.
Cardiovascular disease
Cardiovascular disease, including coronary heart disease and stroke, is the leading cause
of death in the United States for both men and women (18). Since the first National Health
Examination Survey in 1960-62, the assessment of cardiovascular disease-related risk factors
and, to a lesser extent, cardiovascular disease have been a central component of the NHANES
program (1). The main elements of the cardiovascular disease component in NHANES ffl were
measurements of blood pressure, measurements of blood lipid levels, and electrocardiograms
(ECG's).
For the first time in any NHANES, blood pressure was measured on two separate
occasions to reduce misclassification error. For adults 17 years of age and over, a total of six
seated blood pressure measurements were obtained: three by the interviewer in the household
interview and three by the physician during the MEC examination. For children 5-16 years of
age, three blood pressure measurements were made in the MEC by the physician. In the MEC,
the first, fourth, and fifth Korotkoff sounds (Kl, K4, and K5) were recorded for those 5-19 years
of age, and Kl and K5 were recorded for those 20 years of age and over. Blood pressure
measurements were conducted according to the standardized measurement protocols
recommended by the American Heart Association (24).
ECG's were done on all examinees 40 years of age.and over. The ECG's were
interpreted by computer using the Minnesota Code (25). The physician's examination in the
MEC included assessment of systolic and diastolic heart murmurs for all examinees.
Blood lipid levels were determined on a specimen obtained by venipuncrure during the
MEC or home examination. Serum total cholesterol, high-density lipoprotein (HDL) cholesterol,
and serum triglycerides were measured on all examinees 4 years of age and over. Measurements
of total and HDL cholesterol and fasting triglyceride levels permit low-density lipoprotein (LDL)
cholesterol levels to be calculated using the equation developed by Friedewald, Levy, and
E-7
-------
l ซ
Fredrickson (26). Phase 1 of the survey also included measurements of apolipoproteins Al and
B, and Phase 2 included measurements of Lp(a), both for jdl examinees 4 years of age and over.
i
The Household Adult Questionnaire, administered to adults aged 17 years and over,
included questions on family history of heart attack; histoiy, knowledge, and treatment of high
blood pressure and high blood cholesterol; and history of heart attack, stroke, transient ischemic
attacks, and congestive heart failure. The questionnaire also included three sections from the
London School of Hygiene Cardiovascular Questionnaire^?), including the Rose Angina,
Possible Infarction, and Intermittent Claudication questions. During the household interview, the
interviewer made three seated blood pressure measuremeEits (Kl and K5) for adults aged 17
years and over.
The Household Youth Questionnaire included questions on history of rheumatic fever and
heart disease for children aged 2 months to 16 years and questions on history of high blood
pressure and high blood cholesterol for children aged 4-1(5 years.
Respiratory disease,
Respiratory disease has a substantial effect on morbidity and mortality rates in the United
States. It is estimated that up to 20 percent of the adult population suffer from one of the chronic
obstructive pulmonary diseases (asthma, chronic bronchitis, or emphysema) (28).
i
i : " .
The respiratory disease component for NHANES in was designed to measure pulmonary
function and chronic obstructive pulmonary disease. The main element of the component was
assessment of pulmonary function by spirometry. Respondents were also tested for skin-test
reactivity to selected standardized allergens.
Spirometry was conducted for all examinees 8 yeate of age and over in the MEC or home
examinations. Procedures for testing were based on the cibnrent recommendations and standards
of the American Thoracic Society (29). A customized Ohio Censored 822 or 827 dry rolling seal
spirometer was used in the MEC and a portable spirometer was used in the home examination.
Examinees performed five to eight blows to obtain curves acceptable according to the protocol.
The National Institute for Occupational Safety and Health'(NIOSH) was responsible for training
the technicians, providing the equipment, and processing the spirometry data.
The Household Adult Questionnaire, administered to adults 17 years of age and over,
included questions on the medical history of respiratory and allergic symptoms and conditions.
Additional questions ascertained previous diagnosis of asthma, chronic bronchitis, or
emphysema. The Household Youth Questionnaire included a similar set of questions for
children aged 2 months to 16 years.
n-8
,., Ir .
if IN
-------
Diabetes mellitus
Diabetes mellitus is well recognized as a major public health problem in the United
States. The disease affects virtually every organ system in the body, and the rates of such
conditions as kidney disease, blindness, hypertension, ischemic heart disease, stroke, and
disability are significantly higher in persons with diabetes. The direct and indirect costs of
diabetes in the United States were estimated to be more than $20 billion in 1987 (30), and
diabetes ranks as the seventh leading cause of death (18).
The diabetes component was designed to assess glucose tolerance and diabetes. The
main elements of the component were an oral glucose tolerance test and other diabetes-related
laboratory determinations. Related information includes the data collected in the diabetic
retinopathy and vision component.
The MEC examination included a 2-hour 75-gram oral glucose tolerance test (OGTT).
Adults 40-74 years of age examined in the morning session were given the OGTT after being
instructed to fast for 12 hours prior to the examination. After a fasting blood specimen was
obtained by venipuncture, examinees were then administered a glucose challenge (Dextol- 75)
containing the equivalent of 75 grams of glucose. A second blood specimen was drawn 2 hours
after the fasting blood specimen. Measurements of fasting and 2-hour plasma glucose levels
permit identification of diabetes and impaired glucose tolerance according to World Health
Organization (WHO) criteria (31).
Adults 40-74 years of age who were examined in the afternoon or evening were given the
OGTT after being instructed to fast for 6 hours prior to the examination. This procedure does not
follow exactly the WHO recommendations. However, the National Diabetes Data Group of the
National Institute of Diabetes and Digestive and Kidney Diseases strongly recommended that all
adult participants 4074 years of age be screened for glucose tolerance in order to have sufficient
numbers of subjects to ascertain the natural history of glucose intolerance and to quantify risk
factors for the development of diabetes.
Fasting blood specimens obtained by venipuncture during the MEC examination from
adults 20 years of age and over were tested for glucose, levels of insulin, and C-peptide.
Glycated hemoglobin concentration (HbAic) was determined in all individuals 4 years of age and
over as a measure of glucose levels over time. In Phase 2, insulin and C-peptide levels were also
measured on the 2-hour blood specimens for adults 40-74 years of age. The Household Adult
Questionnaire, administered to adults aged 17 years and over, included questions designed to
ascertain those individuals with a medical history of diabetes. Information collected included
family history and age at diagnosis; use, frequency, and amount of insulin taken; use of oral
hypoglycemic agents or diet to lower blood glucose levels; and reported retinopathy. The
Household Youth Questionnaire included questions on diabetes and insulin use for safety
screening purposes only.
H-9
-------
Diabetic retinopathy and vision
The diabetic retinopathy and vision component of NEANES HI was designed to assess
diabetic retinopathy and macular degeneration, two of the jmajor causes of severe visual handicap
and blindness among adults in the United States (32). The main element of the component was
retinal photography carried out in the MEC. Related infoimation includes the data collected in
the diabetes component. ;
i ' ' '
The MEC examination included a nonmydriatic fundus photograph of either the right or
left eye for all examinees 40 years of age and over (33-35). The eye to be photographed was
randomly selected according to the last digit of the examinee's identification number.
Photographic fields were graded by masked, trained graders for macular degeneration, for the
presence and severity of retinopathy and for the presence of specified diabetic lesions using the
Modified. Airlie House Classification scheme (36). The training of photographers and the
grading of photographic slides was done by the staff of the Department of Ophthalmology,
University of Wisconsin Medical School.
hi the physician's examination, examinees aged 2 months to IS.years were evaluated for a
missing globe or blindness; children 2 months-4 years of age were examined for ability to track
light; and those 5-18 years of age were examined for strabismus. No funduscopic examination
was included in the physician's examination and no vision examination was included in the
survey.
i ' '
The Household Adult Questionnaire, administered^ to adults 17 years of age and over,
included questions on problems with vision and presence of blindness, cataracts, or retinopathy.
The Household Youth Questionnaire included an abbreviated set of vision questions for children
aged 2 months-16 years.
Thyroid function
The thyroid component of NHANES in was designed to provide information on the
prevalence of autoimmune thyroid disease, thyroid function, iodine intake, and population
estimates for normal hormone levels through laboratory measurements.
i, <
Measurement of serum thyroid-stimulating hormone (TSH), thyroxine (T4), and
antithyroglobulin and antimicrosomal antibodies were conducted on blood specimens obtained
by venipuncture during the MEC examination from examinees 12 years of age and over. Urinary
iodine and creatinine were also measured for examinees 12 years of age and over to evaluate the
relationship between iodine intake and thyroid dysfunction.
The Household Adult Questionnaire, administered to adults aged 17 years and over,
included questions regarding history of goiter or other thyroid diseases.
H-10
-------
Reproductive health
The reproductive health component of NHANES ffl was composed of questions on the
menstrual cycle, pregnancy history, menopause, use of contraception, and sexual experience
among women, as well as laboratory determinations of follicle-stimulating hormone (FSH),
luteinizing hormone (LH), and pregnancy and lactation status. Related information includes data
collected in the immunization and infectious disease component.
In the MEC Proxy Questionnaire, data were collected on age of menarche for girls 8-9
years of age. In the MEC Youth Questionnaire, girls 10-16 years of age were asked about age of
menarche and were asked to estimate the time since their last period. Girls aged 12-16 years
were asked about pregnancy history, breast feeding, use of oral contraceptives, and sexual
experience. In the MEC Adult Questionnaire, women 17 years of age and over were asked about
age at menarche, pregnancy history, breast feeding, natural and surgical menopause, use of
NORPLANT, use of estrogen, sexual experience, and whether they had ever had genital herpes.
A similar set of reproductive history questions was included in the home examination for women
20 years of age and over,
FSH and LH levels were determined on blood specimens obtained by venipuncture during
the MEC examination from women 35-60 years. In addition, a urine pregnancy test was
administered in the MEC to women 20-59 years of age.
Kidney disease
Kidney diseases constitute a major public health problem with rapidly increasing
visibility because of the fast-growing numbers of patients with end stage renal disease (ESRD).
The number of patients enrolled in the Medicare ESRD program increased from 113,542 in 1984
to more than 170,000 in 1989 (37). The annual costs of the ESRD program have continued to
increase since 1974. The annual expenditures for 1974 were reported to be $229 million and had
reached almost $3 billion in 1989 for this program (38). Based on both the escalating costs and
increasing numbers of patients being served by the ESRD program, cost-effective preventive
measures must be implemented.
The kidney disease component of the NHANES IE was designed to assess renal function.
The main elements of the component were laboratory determinations on blood and urine
specimens. Urinary albumin (microalbuminuria) and creatinine levels were measured in urine
specimens collected during the MEC examination from examinees aged 6 years and over.
Measurements of serum creatinine and blood urea nitrogen (BUN) were made on blood
specimens obtained by venipuncture from examinees 12 years of age and over. The Household
Adult Questionnaire, administered to adults aged 17 years and over, included questions on the
history of kidney and urologic disorders.
n-n
-------
Gallstone disease
Approximately 600,000 patients undergo cholecystectomy each year, making it the most
common abdominal surgical procedure. As a cause of hospitalization, gallstone disease is the
most common and most costly digestive disease, with an annual overall cost of well over $5
billion (39). ;
!"
The gallbladder component for the NHLANES HI was designed to determine the
prevalence of diagnosed and undiagnosed gallstone disease. The main element of the component
was real-time ultrasonography, a noninvasive technique for detecting gallstones. The ultrasound
examination of the gallbladder was administered by a certified abdominal ultrasound technician
to all examinees 20-74 years of age. Examinations were conducted with examinees in both
supine and left decubitus positions. A diagnosis of gallstone disease was made by commonly
used criteria of echoes within the gallbladder with shadowing in two views. Diagnoses were first
made by the ultrasound technician in the MEC and later confirmed by radiologists. If a right
upper quadrant or epigastric scar was observed and the gallbladder was not seen, it was
concluded that a cholecystectomy had been performed. Data for other abnormal pathologies
observed in the surrounding areas, such as the liver or the right kidney, were also recorded.
The Household Adult Questionnaire, administered! to adults 17 years of age and over,
included questions on previous diagnosis of gallstone disease, surgery for gallstones, or other
gallbladder disease. Further questions ascertained occurrence, frequency, and character of pain
in the abdomen or lower chest
Arthritis and related musculoskeletal conditions
Arthritis and related musculoskeletal disorders are frequently chronic, disabling, and
painful. It is estimated that the total economic cost to the United States of musculoskeletal
conditions was more than $126 bilh'on in 1988 (40). |
The arthritis component of NHANES ffi was designed to identify rheumatoid arthritis and
osteoarthritis in adults 60 years of age and over. The main elements of the component were
radiographs and physician's examination of joints. Related information includes data collected
in the osteoporosis and functional health status of the elderly components..
"!'
During the MEC examination, straight posterior-anterior x rays of the hands and wrists
and straight anterior-posterior non-weight-bearing views of the knees were obtained for
examinees aged 60 years and over. The knee position wasi selected because of safety
considerations related to the space limitations in the MEC Additional data were collected during
the physician's examination for those 60 years of age and over. Hand, knee, and great toe joints
were examined for tenderness, swelling, and pain on passive motion. The presence of hand and
foot deformities was also recorded. Abnormalities in gait Kvere evaluated by the physician for all
examinees 3 years of age and over. i
i
n-i2
-------
Serologic analyses of rheumatoid factor for examinees aged 60 years and over and of
C-reactive protein for examinees 4 years of age and over were conducted on blood obtained by
venipuncture during the MEG examination.
The Household Adult Questionnaire, administered to adults 17 years of age and over,
included questions on joint pain, stiffness and swelling in hands, wrists, and knees, back pain,
and medical history of arthritis.
Osteoporosis
The growing recognition of the public health significance of osteoporosis coupled with
the lack of prevalence estimates based on a nationally representative sample motivated the
inclusion of the osteoporosis component in NHANES IH. The cost of hip fractures was
estimated to be $3.5 billion per year in the United States (41). Those who survive hip fracture
are often permanently disabled and must be institutionalized. The extent of problems associated
with hip fracture is likely to increase in the future as the population ages, so that the number of
hip fractures may double or triple by the year 2050 (42). The NHANES osteoporosis component
was designed to assess many of the suspected risk factors for osteoporosis and hip fracture in a
nationally representative sample of adults over 20 years of age.
Although osteoporosis cannot currently be defined by bone density alone, low bone
density is a primary risk factor for osteoporotic fracture, with the risk of fracture increasing as
bone density decreases (43,44). The cornerstone of the osteoporosis component was the
measurement of bone density at the proximal femur of adults 20 years of age and over. Related
information includes data collected in the functional health status of the elderly and arthritis and
related musculoskeletal conditions components. Several bone-related biochemistries were also
measured.
Bone density measurements were made with dual-energy x-ray absorptiometry or DXA
(45). The equipment measured areal bone density (bone mass per unit of area scanned) in five
regions of interest in the proximal femur: femoral neck, trochanter, intertrochanter, Ward's
triangle, and total region. Scans were reviewed by consultants at the Mayo Clinic for the purpose
of quality control. .
The Household Adult Questionnaire, administered to adults aged 17 years and over,
included an extensive series of questions on history of falls and fractures and on maternal history
of fractures and osteoporosis. Data on historical mill:: intake and use of antacids and calcium
supplements were also collected as part of the dietary section. Data were also collected on
tobacco use, physical activity, reproductive health, medication use, and family history of
osteoporosis.
H-13
-------
I
Functional health status in the elderly
i
" i'
The functional health, status component was designed to ascertain the prevalence of
disability and limitations in function among the elderly, l^his component addressed three major
areas: cognitive, physical, and social function. !
Cognitive and physical function were assessed in the MEC and home examination for
persons 60 years of age and over. Cognitive assessment consisted of a short paragraph given as
an immediate and delayed recall task as part of the MEC Adult Questionnaire or the home
examination. Physical function was assessed with a short Battery of physical performance tests
chosen to test different aspects of physical function important in everyday life. The measures
included: range of motion of the shoulder, timed task of hand function (using a key to open a
lock), rising out of a chair without the use of arms and timied rising five times from a chair in
similar fashion, mobility of the hip and knee, timed task of balance (tandem stand), and timed
walk witli counting of steps on an 8-foot course.
Cognitive function among persons 60 years of age [and over was assessed in the
Household Adult Questionnaire through administration of a modified version of the Mini-
Mental State Examination (46). The questions included counting backward from 20 by 3's and
immediate and delayed recall of 3 items. In addition, all fiersons 17 years of age and oyer were
assessed for orientation to location and date in the Household Adult Questionnaire.
For persons aged 60 years and over, the Household Adult Questionnaire contained
standard questionnaire items on physical function derived from the NHANES I Epidemiologic
Followup Study and the Supplement on Aging to the 1984 National Health Interview Survey.
These items included questions on performance of activities of daily living and need for help and
several questions directed to instrumental activities of daily living and need for help. Questions
directed toward higher level function such as walking distances and climbing stairs were also
included.
For adults aged 17 years and over, the Household Adult Questionnaire contained
questions on social support. The questions included information on contact with friends and
family members, attendance at organized religious activities, and involvement in other types of
organizations.
Allergy
i
The primary element of the allergy component consisted of assessment of skin-test
reactivity to standardized allergens. Related information includes data collected in the
respiratory disease component.
Skin-prick tests were administered in the MEC to pll examinees 6-19 years of age and to
a random half-sample of examinees 20-5.9 years of age who were assigned to receive the allergy
E-14
-------
tests if their identification number ended in an even digit. Immediate hypersensitivity to any of
10 licensed commercially available allergens (mite, cat, short ragweed, perennial rye, alternaria,
Bermuda grass, cockroach, Russian thistle, white oak, peanut) was determined. Histamine
phosphate was used as a positive control and 50-percent glycerol saline was used as a negative
control. The skin reactions were read 15-20 minutes after the skin was punctured and the
allergens applied. Both the length and width of the wheal and flare were measured.
The Household Adult Questionnaire, administered to adults aged 17 years and over,
included an extensive series of questions on respiratory symptoms related to allergies. The
questions were designed to obtain information on trigger factors, severity, medication use, and
hospitalization. An additional question ascertained previous diagnosis of asthma. The
Household Youth Questionnaire included a similar set of questions for children aged 2
months-16 years.
Immunization and infectious diseases
Almost all infectious agentsbacteria, viruses, and parasiteselicit long-lasting and
detectable immunity in the host. Therefore, NELANESIH provides an important opportunity to
study the seroepidemiology of the following infectious diseases: hepatitis A, B, C, and delta,
herpes simplex I and n, human immunodeficiency virus (HIV), varicella, hantavirus, and
Toxoplasma gondii. In addition, antibodies to the following microbial agents have been
determined to assess the level of protective antibody in the population: tetanus, diphtheria, and
rubella. Finally, antibody to Cryptosporidia parvurn will be determined in sample persons from
selected communities to assess exposure to this water-borne pathogen based on water source.
Serologic tests for antibodies will provide national estimates of exposure to hepatitis A,
B, Cj delta, and E and will assist in validating surveys that are more localized or that involve
samples with potential sources of bias not found in NHANES. Because hepatitis A, B, and delta
were performed on NHANES H (1976-80) sera, trends over time in the prevalence of infection
can be determined (47). Hepatitis C virus is the name assigned to a newly detected virus that is
thought to be the primary cause of transfusion-associated non-A, non-B hepatitis in the United
States (48,49). Testing of the NHANES m sera provides the unique opportunity to produce a
baseline measure of the extent of infection in the U.S. population by this agent. The presence of
specific antibodies directed against herpes simplex I and n will also be determined by serology.
The population prevalence estimate will be used as a comparison for validating reporting systems
involving patient-based and other smaller studies. NHANES H surplus sera were also previously
tested for antibodies to these viral agents (50). Continuation of herpes serologic testing in
NHANES HI will produce trend data that will help to delineate the extent of a possible herpes
epidemic. Other information related to sexual behavior and history of genital herpes was
collected hi the MEC Youth and MEC Adult Questionnaires.
Human immunodeficiency virus (HIV) testing was performed on all sample people over
the age of 18 years using an anonymous protocol. Serum collected for the many other laboratory
H-15
-------
tests was separated into a vial that had been randomly numbered and not linked to the sample
person's identification number. The only demographic information attached to the HIV sample
was: age in 20-year groups, sex, race or ethnic group, and Sampling location, hi Phase 2 of the
survey, a basic sampling weight, an education variable, and the results of the urine drug testing
were also linked to the HIV result. Sample people were notified during the informed consent
process that blood samples would be tested for HIV. As \i^ith any other component of the survey,
the sample person had the right to refuse the test. The anonymous testing procedure was chosen
for the HTV antibody testing to provide the maximum safeguard of the sample person's
confidentiality. Anonymous testing was considered the orily feasible method to provide unbiased
estimates of seroprevalence of HTV antibody. The HTV prevalence estimate on a representative
sample of the U.S. population will contribute further to the knowledge of the epidemiology of the
disease previously obtained from select populations in the [Center for Disease Control and
Prevention's family of surveys (51,52) and the distribution of reported cases.
A candidate vaccine for varicella has been developed and is currently undergoing final
clinical trials prior to anticipated application for licensure for use in the United States. The
seroprevalence and risk factors for varicella infection need, to be established to better plan for
wide use of this vaccine. NHANES HI data will be used tฉ target at-risk populations in the
United States. ' ;
i
Because of the outbreak of adult respiratory disease syndrome caused by a newly
described hantavirus during the summer of 1993, sera from NHANES HI specimens were tested
to determine the geographical distribution and prevalence
-------
elicit a weaker immune response and provide reduced levels of protection against this bacterial
toxin. Because of recent outbreaks of measles and rubella, inclusion of a serologic test for
rubella antibody using NHANES in specimens will provide information on populations at risk
for these viruses.
Hearing
The principal elements in the hearing component were the measurements of pure tone air
conduction audiometric thresholds and tympanic compliance in children. These examinations,
pure tone audiometry and tympanometry, were conducted in a soundproof room in the MEC for
examinees aged 6-19 years. Pure tone air conduction audiometry thresholds were obtained in
both ears at 500,1000,2000, 3000,4000, 6000, and 8000 hz. A screening questionnaire
administered before the examination provided data on recent noise exposure and use of
headphones. Because pure tone screening by itself may not be sensitive enough to detect middle
ear disease,, tympanometry was conducted to provide an estimate of tympanic membrane
compliance. The Household Adult Questionnaire, administered to adults aged 17 years and over,
collected information on hearing status and use of hearing aids.
The Household Youth Questionnaire included questions on frequency and treatment of
ear infections, hearing status, and hearing aid use for children, aged 2 months-16 years.
Lead exposure
The lead exposure component was designed to assess environmental lead exposure through
measurement of blood lead levels. Blood lead levels were determined on examinees 1 year of age
and over on specimens collected by venipuncture during the MEC or home examination.
Analysis was performed by graphite furnace atomic absorption spectrophotometry. Erythrocyte
protoporphyrin, a screening test only sensitive to high lead levels, was also measured.
The Household Youth Questionnaire included questions on history of testing and treatment for
lead poisoning for children aged 2 months-16 years. Information on the age of the housing
structure was also collected in the Family Questionnaire.
Mental health and neurobehavioral function
The primary elements of the mental health and neurobehavioral function component were
conducted in the MEC and included assessment of depression and mania, cognitive function, and
functioning of the central nervous system. Supplemental data were collected in the household
interview.
The mental health and neurobehavioral function component of NHANES El included the
depression and mania subsections from the Diagnostic Interview Schedule (DIS), developed by
the National Institute of Mental Health (NIMH). Sections of the DIS have also been used in the
tt-17
-------
Hispanic HANES and in several community studies (53-55). Trained interviewers administered
the NHANES IE DIS questions in the MEC, using automated data entry, as part of the MEC
Youth Questionnaire for examinees 15-16 years of age and as part of the MEC Adult
Questionnaire for examinees 17-39 years of age. The data collected from the DIS permit
diagnoses based on the third edition of the Diagnostic and Statistical Manual of Mental Disorders
(DSM-EO(56). r . J
Intellectual function and academic performance were assessed for children aged 6-16
years with standardized cognitive tests. The examination included the block design and
digit-spaa subtests from the Wechsler Intelligence Scale for Children, Revised (WISC-R) (57)
and the reading and arithmetic sections from the Wide Rafitge Achievement Test, Revised
(WRAT-R) (58).
Central nervous system function was assessed witii a set of simple, nonverbal
performance tests designed to be minimally influenced by differences in language or education.
The tests were administered to a random half-sample of all adults 20-59 years of age, who were
assigned to receive these tests if their identification number ended in an odd digit. The
examination was composed of three testssimple reaction time, serial digit learning, and symbol
digit substitutionselected from the larger battery of Neurobehavioral Evaluation System (NES)
tests (59). Factors that might have affected performance such as motivation, use of drugs,
alcohol, or caffeine, and the temperature, humidity, and air flow in the testing booth were
recorded in a brief pre- and post-test questionnaire.
The Household Youth Questionnaire, administered to children aged 2 months-16 years,
included questions on attendance to special classes in school as a result of impairment and
diagnosis of mental retardation. Data were also collected on visits to a psychiatrist, psychologist,
or psychoanalyst for children 4-16 years of age and on school attendance and relationships with
friends for those 5-16 years of age. The Household Youth Questionnaire also included a series
of questions on motor and social development for childrerL aged 2 months-3 years. The
questions were modeled after the Denver Developmental Screening Test (60) and a similar
component used in the Child Health Supplement to the 1981 National Health Interview Survey.
Related information includes occupational history and the j cognitive, physical, and social function
data collected in the functional health status of the elderly Component.
Oral health
The main element of the oral health component walsi anoralexamination conducted in the
MEC. Methods used in this component were designed to 'be consistent with previous health
examination surveys conducted by NCHS and with previous national surveys of oral health
conducted by the National Institute for Dental Research (b)IDR). Related information was also
collected on selected risk factors such as diet, the use of srtiokeless tobacco, and the use of
fluoride supplements.
1- :
n-is
-------
During the MEC examination, the dentist performed oral examinations on all examinees
12 months of age and over. Oral soft tissue lesions were assessed for examinees aged 2 years
and over The assessment involved visual observation of the oral mucosa and laboratory
assessment of an oral mucosal smear for the presence of hyphae of Candida albicans. A dental
caries examination included an evaluation of coronal caries for those aged 2 years and over, root
surface caries for examinees 18 years and over, and baby-bottle tooth decay among children aged
12-23 months Examinees were questioned about history of injury to front teeth and then were
examined for evidence of traumatic injury to the four upper and four lower permanent incisors.
Occlusal characteristics were assessed in examinees aged 8-50 years and included measurement
of the alignment of teeth and assessment of posterior crossbite, overjet, overbite, and maxillary
diastema A periodontal examination was performed on two randomly selected quadrants of the
mouth for examinees 13 years of age and over. Restorations and tooth conditions for those
18-74 years were also evaluated.
The Household Adult Questionnaire, administered to adults 17 years of age and over,
included questions on utilization of dental health services and information needed to interpret fee
oral examination findings, such as history of cold sores and receipt of orthodontic treatment. The
Household Youth Questionnaire included similar questions for children aged 2-16 years, as well
as infant feeding practices contributing to baby bottle caries.
Risk factors and health behaviors
Risk factors and health behaviors associated wife many chronic diseases and conditions
were evaluated in NHANES ffl both in fee household interview and during the MEC
examination. The five primary behaviors assessed in fee survey were alcohol use tobacco use,
drug use, physical activity, and sexual experience. The component on drug use is described in
the special studies topic of this section.
Alcohol use
The MEC Youth and MEC Adult Questionnaires included questions on alcohol use for all
examinees 12 years of age and over. The questions were designed to identify nondrinkers, very
light drinkers, and former heavy drinkers; to ascertain quantity and frequency of use tor
quantifying alcohol intake; and to determine fee frequency of heavy drinking occasions. Data on
anv alcohol intake during the previous day were also recorded for all examinees as part of the
24-hour dietary recall. The section on "Nutritional health assessment" of this report also has
some information on alcohol. Standard liver function tests were performed. The Household
Adult Questionnaire, administered to adults 17 years of age arid over, included questions on fee
frequency of consumption of beer, wine, and liquor as part of fee food frequency section.
n-19
-------
".t-i
ill
I
Tobacco use and exposure !
To encourage honest reporting of tobacco use by youths aged 8-16 years, information on
use of cigarettes and smokeless tobacco (snuff or chewing; tobacco) was collected in the privacy
of the MEG as part of the MEG Youth Questionnaire. Data were collected on age of initiation,
frequency, duration, and amount of tobacco consumed. Recent use of tobacco or nicotine gum
within the past 5 days, for evaluation of laboratory results, was also assessed in the MEG Youth
and MEG Adult Questionnaires and the home examination for those aged 8 years and over.
The Household Adult Questionnaire, administeredjto adults 17 years of age and over,
included questions on the use of cigarettes, cigars, pipes, and smokeless tobacco (snuff or
chewing tobacco). Data were collected on age of initiation, frequency, duration, and amount of
tobacco consumed and on exposure to tobacco smoke at work. Data on passive smoke exposure
were collected in the Family Questionnaire. Family members who smoked cigarettes in the home
were identified and the amount smoked per day was estimated. The Household Youth
Questionnaire included questions on history of maternal smoking during pregnancy for children
aged 2 months-11 years. :
A biochemical determination of tobacco exposure was used to assess both passive
smoking and tobacco use through measurement of blood cqtinine levels from specimens obtained
by venipuncture in the MEG from examinees aged 4 years and over. Cotinine is a metabolite of
nicotine and is thus an indicator of primary or secondary exposure to tobacco. Cotinine was
detected using an isotope dilution, liquid chromatography, Itandem mass spectrometry method
developed by the National Center for Environmental Health., CDC, which conducted the
analyses. This was a newly developed method designed toj detect levels as low as 0.030
nanograms per milliliter. Related information includes data collected in the respiratory disease
component and identification of oral soft tissue lesions in the oral health component.
.| , :
Physical activity . i
For children 8-16 years of age, data on frequency of exercibe and physical activity were collected
during the MEG interview. The Household Adult Questionnaire, administered to adults aged 17
years and over, contained questions on leisure-time physical activity adapted from thel985
National Health Interview Survey and included information on types of activity, frequency, and
assessment of level of activity compared with others. Participants were also asked to compare
current levels of physical activity with those of the past year and 10 years ago. Related
information includes data collected in the functional health status of the elderly component.
Sexual experience \
Questions on sexual experience were included in the MEG Youth and MEC Adult
Questionnaires for examinees aged 15-59 years. Information on age at first sexual intercourse,
total number of partners, and number of partners in the past year was collected for those 17-59
n-20
-------
years of age. Males 17-59 years of age were asked about the numbers of male and female
partners. Age at first sexual intercourse was obtained from youths aged 15-16 years. Related
information includes data collected in the immunization and infectious disease and reproductive
health components.
Special studies
Four special studies requiring an additional collection of blood or urine during the MEC
examination were carried out in conjunction with NHANES m. Planned and sponsored with
other agencies, their unique status was warranted either by the confidential or experimental
design of the research. The two highly sensitive studi.es, HIV testing and drug testing, were
conducted using a rigorous protocol that maximized anonymity and confidentiality. The fflV
testing is described in the immunization and infectious diseases section. The results from these
analyses can only be linked to a limited set of demographic.and medical information collected in
the survey. The priority toxicant range study and the establishment of a deoxyribonucleic acid
(DNA) storage bank for genetic research were both designed, in part, to explore new and
experimental laboratory techniques. Also, a portion of the sera was placed in a bank for
unanticipated future research projects.
Drug use
All examinees aged 12 years and over were questioned in the MEC Adult and Youth
Questionnaires about lifetime and past-month use of marijuana and cocaine. In Phase 2 of
NHANES HI (1991-94), anonymous urine testing was included in the MEC examination in order
to detect the presence of marijuana, cocaine, phencyclidine (PCP), opiates (morphine and
codeine), and stimulants (amphetamine and methamphetamine) among examinees 18-59 years of
age. Urine specimens were randomly numbered so they could not be linked with the examinee
identification numbers. Limited demographic data including age (in 20-year categories), sex,
race or ethnicity, sampling location, and educational level were included with the random
numbers on protected data files. The identical random numbers and the associated demographic
variables were assigned to the fflV serum, so that the association between drug use and HIV
status could be examined.
Specimens were screened using an enzyme multiplied immunoassay technique with
cutoff concentrations lower than those generally used in drug screening. Positive specimens
were confirmed and then quantified using gas chromatography mass spectrometry. Urine
analyses for drug screening were performed in a National Institute of Drug Abuse (NIDA)
certified laboratory according to NIDA guidelines.
Priority toxicant reference range study
The purpose of the priority toxicant reference range study was to assess the levels of
common pesticides and volatile organic compounds (VOC's) in a large sample of the U.S.
H-21
-------
population and to evaluate laboratory analytic methods in the process. Two groups of organic
compounds were measured in the priority toxicant study: selected pesticides and their
metabolites at the low parts-per-billion levels in urine and VOC's at the low parts-per-trillion
levels in whole blood (see appendix table YD). The Division of Environmental Health
Laboratory Sciences, National Center for Environmental Health, CDC, conducted the Priority
Toxicant Reference Range Study on approximately 3,600;persons aged 20-59 years examined in
NHANES IE. Participants volunteered to complete a brief chemical exposure questionnaire and
to provide an additional 20 ml of blood and 40 ml of urinง. No formal sampling procedures were
instituted; 45 volunteers participated at each survey location. A $10 remuneration was awarded
for participation. Demographic and medical information obtained from NHANES HI can be
linked to the resulting laboratory measurements. :
Serum bank and DNA specimen bank
NHANES in provided an opportunity to establish two nationally representative specimen
banks, a serum specimen bank and a DNA bank of preserved, viable cells. Serum specimens are
stored at -70 ฐC or less and will be used for unanticipated future research projects.
For the DNA analyses, new molecular genetic techiniques make it possible to examine
substantial portions of the DNA sequence and its variation in the population using small samples
of nucleated cells obtained by venipuncture. The development of transformation and
immortalization procedures to maintain active cultures of cells means that small samples
collected from a large population can be maintained and amplified to provide specimens for
future studies. It is anticipated that this endeavor will help establish a new era of health research
that integrates genetics and environmental factors in the understanding of human disease.
11ป
During the MEC examination, a 6-cm3 specimen of venous blood was collected in a
vacutainer tube containing a Ficoll heavy-density layer overlaid with a thixotropic gel followed
by ACD anticoagulant from examinees 12 years of age and over. The specimen was then
prepared by the National Center for Environmental Health following one of two methods. Either
the nucleated cells were separated from the blood sample and.then separated in several aliquots
or the cells were separated from me blood sample and viraily transformed to yield an
immortalized culture. In both instances, multiple aliquots [from each subject were frozen,
following a controlled freezing procedure. The frozen aliq[uots are maintained in liquid nitrogen.
i ป
n-22
-------
APPENDIX III
NHANES m Nutritional Status Assessment
(Note: The material in this Appendix is taken verbatim from the NCHS source document. A list
of the references cited in the original NCHS document has not been included in this Appendix.
The reader is referred to the source document for the full citations.)
(Source: NCHS, 1994b, pp. 12-17.)
m-i
-------
NUTRITIONAL STATUS ASSESSMENT '
Nutrition data from the National Health and Nutrition Examination Survey (NHANES)
are vital to nutrition monitoring and public health. As the cornerstone of the ' 'nutrition and
related health measuremenfcomponent of the Federal Government's National Nutrition
Monitoring and Related Research Program (NNMRRP) (61), NHANES nutrition data are used in
a variety of settings to enhance the health and nutritional status of the Nation. NHANES in data
will be used to track progress toward the Nation's health Jind nutrition objectives (62,63,16) for
diet, serum cholesterol, hypertension, iron deficiency anemia, overweight, and infant feeding
practices. Additionally, NHANES provides reference data for nutritional biochemistries (64-72),
anthropometric measures (73-76,8), and nutrient intakes (77-78); and provides information for
policymakers to set nutrition policy (16,62,79-82) and research agendas (79,83,84). NHANES
El was also designed to demonstrate relationships between diet and health. The nutritional
assessments were designed to complement and link to NHANES HI health components such as
cardiovascular disease, diabetes, hypertension, osteoporosis, and dental caries to maximize data
utility. A longitudinal design was added to the traditional NHANES cross-septional design and
studies of the relationship between present diet and future disease will be possible.
Food and nutrient consumption
Dietary factors are associated with 5 of the 10 leading causes of death and are associated
with other conditions such as obesity (16). Deficiencies of nutrients and minerals, such as iron
and some vitamins, remain a problem in selected population subgroups. Inadequate food intake
and undemutrition are problems in high-risk subgroups such as low-income populations.
Overconsumption of food components such as fat, cholesterol, and salt and underconsumption of
fruits, vegetables, and complex carbohydrates are significaint problems in the general population.
Measurement of nutrient intake is important in evaluating food fortification, nutrition education,
and intervention programs aimed at improving the population's dietary intake. Measurement of
foods as they contribute to nutrient intake and as they comprise dietary patterns are important for
evaluating and developing dietary guidance (81). Recognizing the importance of measuring both
nutrient intake and food intake to meet current nutrition monitoring data needs, the NHANES HI
dietary component was developed to estimate total nutrient intake, nutrient intake from foods,
intake of specific foods, and problems and factors related to insufficient food and nutrient intake.
Vitamin and mineral supplement data will be used py NCHS and the Food and Drug
Administration (FDA) to determine the prevalence of very low and very high total nutrient intake
levels in the population and for assessing the contribution of supplements to total nutrient intake
and nutritional status (83). Total nutrient intake is also imjportant for evaluating diet-health
relationships such as the association between total calciuni intake, blood pressure, and
hypertension risk (84), and total calcium intake and bone density (85-86).
IH-2
-------
Alcohol intake
Alcohol problems and associated health risks are prevalent problems in adolescents and
adults (16). Excessive alcohol intake is associated with cirrhosis of the liver as well as accidents
and suicides (16). Moderate amounts of alcohol have been related to both increased risk of
certain cancers and decreased risk of coronary heart disease (87). Information on alcohol was
collected in NHANES in to quantify the contribution of alcohol to total caloric intake for
population reference data, to assess the proportion of the population who typically consume
larger amounts of alcohol than recommended in the Dietary Guidelines for Americans (81), and
to investigate the relationship between alcohol intake and health outcomes (87,15).
Food program participation and food security
In the 1980's there were several reports that indicated that hunger was a serious problem
in the United States on a national level and for certain subgroups of the population (88-93).
However, accurately estimating the prevalence and severity of hunger is complex and historically
has been controversial, hi 1987, the University of California at Berkeley sponsored a workshop
to bring together hunger researchers working at the local, State, and national levels. The
workshop concluded that "of all the relevant Federal surveys, NHANES is probably the best
equipped to look at the interrelationships between diet, food shortages, and health indicators"
(94).
In addition to questions for families and individuals about having enough food or money
to buy food, data were collected in NHANES HI about the use of food stamps, participation in
the Special Supplemental Food Program for Women, Infants, and Children (WIC), school
breakfast and school lunch programs, and elderly feeding programs. Assessment of the dietary
status of participants of such programs is important hi aiding the study of Federal food programs
and their effect on the dietary intake of low-income and high-risk subgroups.
NHANES m will enable researchers to link food security and program participation with
other nutrition and health indicators, including cognitive function. This is especially important
for children. It has been documented that hungry children can be irritable, apathetic, or lethargic,
which can interfere with learning (95). The National Education Goals were established in 1990,
the first of which states that by the year 2000, "all children in America will start school ready to
learn" (96). Having adequate food is a large part of this.
Vitamin and mineral status
Biochemical and hematological indicators of nutritional status are an essential part of the
NHANES m nutrition component. Blood assessments have been included in past NHANES to
determine the prevalence of compromised vitamin and mineral status at both the high and low
ends of the population distribution (64-72); and the prevalence of nutrition-related risk factors,
such as elevated serum cholesterol (10,11,97,98) or serum albumin (68).
m-3
-------
Table 1. Comparison of Laboratory Analyses of Blood and Urine Across
Lab Test
Whole Blood Assessments:
Sedimentation rate
Complete blood count
Platelets
3-cell differential
Differential smear
Red cell distribution width
Lead
Protoporphyrin
Red blood, cell folate
Glycated hemoglobin (HbAiC)
Carboxyhemoglobin
Serum Biochemistry Assessments:
Folate
Iron & total iron-binding capacity
Ferritin
Vitamin C
Vitamin D
Vitamin E
Zinc & copper
Vitamin A (retinol)
Carotenoids
Retinyl esters
Vitamin B 12
Methyl malonic acid
Homocysteine
Selenium
Total cholesterol
High density lipoprotein cholesterol
Triglycerides
Apolipoproteins A, & B
Lp(a)
Total & ionized calcium
Cotinine
Bile salts
Pesticides00
Syphilis
Hepatitis A, B & delta
Hepatitis C
, Hepatitis E
Tetanus
Diphtheria
Polio
Herpes simplex I & n
SURVEY(1)
I
X
X
X
X
X
X
X
X
X
X
X
n
X
X
X
X
X
X
X
X
X
X
X
Xฎ
X
X
X
X
X
X
X
X
X
HH
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
m
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
x(3)
x<3)
x(3) .
X
X
X
X
x(4)
XP)
X
X
X
X
X
X
X
X
X
V-2
-------
Infant and child nutrition
Infants and children are particularly vulnerable to poor nutrition. Childhood and
adolescence are important periods for establishing nutrition and health habits for later life.
Whether or not an infant is breastfed, the type of milk, or infant formula an infant is fed, and the
types of solid foods first introduced are all critical infant feeding practices. Adequate dietary
intake during infancy and childhood is necessary for proper growth and development and the
prevention of future health problems. Of particular concern at this early age are iron deficiency,
poor dietary habits, breastfeeding, and inadequate intake in high-risk populations (95). Also,
overconsumption of certain foods is related to the development of obesity and dental caries in
children.
Growth
Anthropbmetric measurements have been included in the National Health Examination
Surveys since the first National Health Examination Survey (NHES I) was conducted in 1960-62
(1,104). These measurements were the basis for the NCHS growth charts, which were
constructed with data from the earlier health examination surveys. The.charts are used nationally
in hospitals, health departments, and physicians' offices and have been adopted for international
use by the World Health Organization (105,106). The production of these original growth charts,
however, was affected by some inherent limitations. Because data were not available in previous
NCHS surveys for the very youngest age group (age under 1 year), the data were supplemented
with data from the Pels Research Institute (8). This resulted in growth curves for recumbent
length (for children from birth through 3 years of age) based on Pels data and for stature (for
children aged 2-18 years) based on NCHS data. Because the median statures for the Pels data
were greater than the median statures in the NCHS data, there was a disjuncture in the curves for
children between 24 and 36 months of age (107). NHANES HI was specifically designed to
resolve both of these limitations; children 2 months of age and over were included in the survey
(108), and more sophisticated curve-smoothing techniques that have evolved since the first
NCHS growth charts will be used.
Overweight and obesity
/ . .
Overweight and obesity are current public health issues and prevalent risk factors for
chronic disease. NHANES m anthropometric data will be used to estimate the prevalence.of
overweight and severe overweight in the United States for various age, race or ethnicity, and
gender subgroups (16,74). NHANES H data showed that 26 percent of adults were overweight
(16). Assessment of body fat distribution has been shown to be related to chronic disease
development (87). Body measurement data indicative of overweight or obesity will be used as
control or explanatory variables in epidemiological analyses of many other examination items,
including blood pressure, glucose intolerance, gallbladder ultrasound, and a battery of other
indicators for cardiovascular disease.
ffl-5
-------
Anthropometric measures can be utilized in many Ways; for example, Jo estimate body
composition, to develop various reference standards, to establish baseline data for future
longitudinal studies, to monitor trends over time in the population, and to evaluate risks for
adverse health outcomes (109). ] j
! ' " ' i ! " '
Diet-health relationships
i . -I
With the growing understanding of the role of nutrition in health promotion and disease
prevention, nutritional status assessment has assumed greater prominence and has been integrally
linked with other aspects of NHANES ffl. As the Surgeon General remarked upon the release of
The Surgeon General's Report on Nutrition and Health (15), for the majority of adults who don't
smoke and don't drink excessively, what they eat is the most significant controllable risk factor
affecting their long-term health. The NHANES HI has been designed to capture as many
nutrition risk factors as feasible related to the major chronic diseases affecting Americans, while
continuing to provide a comprehensive assessment of the population's nutritional status for
nutrition monitoring purposes.
The dietary information will be useful for studying1 the relationship between dietary habits
and health. For example, sodium-intake data from the dietary interview, more specific than in
past NHANES, can be linked with blood pressure data, saturated fat can be linked to blood
cholesterol, and intake of antioxidants such as vitamins A, C, E, and carotenoids can be joined
with followup information on cancer and heart disease (15). Past NHANES datk have been used
to relate the number of meals and snacks eaten to dental caries (110); and information on the
number of meals eaten away from home can be used to plan and evaluate nutritipn education
programs targeting overweight and obese clients. ;
' i
,1 , ' '
, Osteoporosis and calcium intake |
i . i | ..
. Osteoporosis is a debilitating disease of reduced boiae mass that causes fijactures of the
vertebrae, hip, forearm, and other bones. Intake of calcium, phosphorus, vitamin D, protein, and
alcohol, as well as a sedentary lifestyle, may all be related to the development of osteoporosis
(111). NHANES ffl measures intake of these nutrients, including total calcium and frequency of
consumption of calcium-rich foods. In addition, a question on historical intake of milk was
included in the household interview to investigate the relationship between past Calcium intake
and current bone densitometry results. Interest in past consumption of dairy foods has been
raised by findings suggesting that the level of calcium intake by young adults may be related to
peak bone mass (85,86). ' ''
Nutritional status methods
i , | t , , , ,,
The nutritional assessment component of NHANES: El was designed to include several
data sources (dietary intake interviews, nutrition-related interviews, anthropbmetric data,
m-6
-------
hematological and nutritional biochemistries, and nutrition-related clinical assessments) in order
to provide a comprehensive assessment of nutritional status (112).
Methodologies for the nutritional assessment were developed with input from experts and
data users from government agencies, academic research institutions, and industry. In 1986, an
NHANES IE Nutrition Methodology Working Group was established. In addition to NCHS
planning staff, it included Federal staff with specific expertise in the topics under discussion
and/or who were primary data users with a nutrition policy need for the data. The Nutrition
Methodology Working Group reviewed the literature and discussed survey methods, operational
issues, and specific details that needed to be determined for the NHANES m nutrition
component. Planning sessions included discussion of the following topics:
General issues. Household versus mobile examination center (MEC) administration of
the dietary interview; automated data collection; nutrition monitoring and comparability
to other national surveys, primarily the food consumption surveys conducted by the U.S.
Department of Agriculture (USDA).
24-hour dietary recall method, Automated versus manual data collection; number of
days of observation; location of interview; number of interviews per individual; adults
versus children in the household.
Food frequency questionnaire (FFQ). Review of FFQ's used in other surveys,
including past NHANES and the 1987 National Health Interview Survey; appropriate
uses of FFQ data.
ซ Longitudinal study issues relative to nutrition.
Children's issues. Proxy rules; use of food models; data retrieval for day care and
school lunch.
Dietary questions. Interview information needed on food security (hunger); water
intake; dietary practices.
. Vitamin and mineral supplement usage. Level of detail required; method of data
collection.
ป Anthropometry. Measurements; bioelectrical impedance; supplemental interview
questions (e.g., self-assessment of overweight).
Based on the study objectives, consideration of outside input, and the Nutrition
Methodology Working Group discussions, a comprehensive nutrition proposal was developed by
NCHS staff. The proposal was reviewed by NHANES ffi Research Consortium members and
served as the basis for planning the nutrition component.
m-7
-------
i! I
i E
Dietary methodology . . ',
Prior NHANES conducted between 1971 and 1984 included 24-hour dietary recall and
food frequency components as parts of the dietary interview (77,113). On March 16-18,1986,
NCHS sponsored the Dietary Methodology Workshop to ibview dietary methodologies and
obtain recommendations for selection of methods for NHJ^NES m (114). Experts in the fields of
dietary survey methodology, epidemiology, nutrition, public health, and biostatistics presented
papers that addressed statistical-issues unique to the interpretation and use of dietary survey data,
selection of dietary methods appropriate for nutrition monitoring activities, and approaches
useful for assessing the relationship of diet to energy balance and three diet-related, chronic
diseases (cancer, cardiovascular disease, and osteoporosis) (114).
The overall workshop recommendation, based on the major aims of the NHANES IE
nutrition component, was that NCHS should continue to tlse the 24-hour recall as the principal
methodology to provide detailed quantitative food and nutrient intake data for the U.S.
population. Use of a food frequency was recommended t
-------
administration of the 24-hour dietary recall and the FFQ in NHANES IH by age of the sample
person, respondent (self and/or proxy), place of interview (sample person's home or MEC), and
interview type (nondietary, dietary, and telephone).
Food frequency
In a major innovation for the NHANES, a FFQ was incorporated into the household
interview to provide general qualitative dietary information for individuals aged 17 years and
over. The FFQ used a 1-month reference period and was not quantitative, i.e., did not collect
portion sizes. It was not designed to produce population nutrient intake estimates, and use of
food frequency data for this purpose is not appropriate (109,117119).
The FFQ food list was developed to be comparable to food lists used in past NHANES
for trend assessment but was expanded to capture more detailed intake of foods containing
specific nutrients of interest. Foods containing nutrients related to risk for cancer, cardiovascular
disease, and osteoporosis (87), such as vitamins A and C, caffeine, and calcium, respectively,
were added to a general food list. In addition, the instrument was modified to be appropriate for
use with the population subgroups sampled in NHANES ffi by including foods high in these
nutrients that were reported by white and black persons in NHANES n and by
Mexican-Americans in Hispanic HANES.
Because the FFQ was collected during the household interview, information on food
intake is available for all interviewed persons and can. also be linked with reported health
conditions. Collection of the FFQ in the household for all interviewed persons will also allow
assessment of potential nonresponse to the 24-hour recall, which was collected in the MEC for
interviewed and examined persons. Another important use of the food frequency data is to
provide baseline dietary data for followup analysis. E&ecause all NHANES ffi sample persons are
followed for mortality, a larger sample of interviewed, persons with dietary intake information is
available for followup analysis.
To complement the osteoporosis component, adults were asked to report their milk
consumption during five age periods: 5-12 years, 13-17 years, 18-35 years, 36-65 years, and 65
years of age arid over. Responses were recorded as "more than once per day," "once per day,"
"less than once per day, but more than once per week," "once per week," "less than once per
week," or "never." Although several researchers have found that the recall of past diet was
strongly influenced by present dietary habits (120-124) and that it is difficult to quantify the
amount of calcium consumed during periods of peak bone growth using retrospective dietary
data, most people can probably retrospectively report whether or not they consumed milk.
products during these time periods in a qualitative sense.
m-9
-------
Nutrition-related interview
A complete picture of dietary intake is not possible with a single 24-hour recall and food
frequency. Therefore, additional interview questions were asked about water intake, usage of
vitamin and mineral supplements, meal and snack patterns}, infant feeding practices, alcohol
intake, and food sufficiency. Appendix table X shows the nutrition-related interview information
collected in NHANES in by age of individual. . i
Questions related to periodic or chronic food shortages were asked for both families and
individuals to study the impact of food insecurity on dietary intake, nutritional status, and health
(88). At the family level, questions were asked about the number of days per month on which
mere was no food or money to buy food and the reasons for the problem. The questions for
individuals, modeled after those developed and used in th
-------
selected from a public health perspective. Because of heightened awareness and the emergence
of evidence of associations between fat distribution and health outcomes, the number of
circumference measurements was expanded. Additionally, to be on the cutting edge of new
technology amenable to the survey environment, bioelectrical impedance analysis (BIA) was
included for those age groups for which stable prediction equations were expected to become
available from empirical research (127).
The selection of both procedures and equipment was influenced by constraints inherent to
the unique setting of this survey and by the need to ensure reasonable comparability with the past
while collecting data to meet current needs, ha order to include an optimal number of measures
in a limited time frame, a considerable amount of planning and experimentation was devoted to
modifying and refining applications of the equipment, the procedures, and the facility, including
the automated data recording system. The final array of body measures was distilled to several
sets that are variably administered, dependent upon the age of the sample persons. These
measures are shown in appendix IV and may be categorized as weight, height, length,'
circumference, breadth, skihfold, and bioelectrical impedance.
As with all other components of this survey, the primary objective was to maximize
validity and reliability. Because one major end product of the anthropometric component is to
produce reference values, accuracy was emphasized through standardized training and a
multifaceted quality control system. Related to this, reproducibility is also a paramount concern,
not only within and between individual data collectors and trainers for NHANES ffl, but also to
facilitate comparisons between the NHANES IE and other surveys and studies. Specific
consideration was given to selecting methods that incorporated, to the extent possible, objective
procedures. For example, bony landmarks were selected to identify anatomical sites for
placement of the instruments and proper positioning of the sample persons; marks were made on
the measurement sites to locate midpoints and anchor the measuring devices; and where feasible,
measures were taken directly on the skin, hi general, the guidelines of the Anthropometric
Standardization Reference Manual (128) were followed, although modifications were made for
selected procedures. Documentation of complete details of the NHANES m anthropometric
procedures will be disseminated in a separate publication.
Laboratory determinations
When selecting nutritional biochemistry and hematological indicators to include in
NHANES m, first priority was given to scientific merit. An NCHS survey planning group was
charged with developing a list of blood determinations for NHANES HI, including priorities by
age group. The planning group used recommendations from an ad hoc panel convened by the
Life Sciences Research Office, Federation of American Societies for Experimental Biology, at
the request of the FDA (100), as well as other important sources, such as the first report of the
Joint Nutrition Monitoring Evaluation Committee (129) and The Surgeon General's Report on
Nutrition and Health (15). Individual agencies and institutes within the NTH also developed
specific proposals for biochemical and hematological measures to be included in NHANES m.
m-ii
-------
t ; '
For example, the National Heart, Lung, and Blood Institute developed and funded the lipid
analysis for NHANES HI (see the part "Health status assessment"). The scientific merit of each
measure was evaluated hi the context of the goals of the siirvey, i.e., which NHANES IE health
conditions and examination measures were available for linkage with the laboratory data.
[ . , j | , ,1
If deemed to have sufficient scientific merit, the next criterion considered was feasibility
of measuring the indicator hi a national survey. This inchided whether the state-of-the-art
analytic methodology currently accepted by the scientific community was practical for a large
field survey lasting 6 years. Specimen size requirements and staff and monetary costs also had a
bearing on feasibility (130). Laboratory protocols and analytical methods for several nutritional
determinations were developed at the National Center for Environmental Health (CDC). A list
of the nutritional biochemistry and hematological variables assayed hi NHANES HI is shown by
age in appendix table IV. In previous NHANES, blood was collected from children by
fingerstick. However, because of problems in performing^ fingerstick without creating
contamination or causing hemodilution by "milking," it was decided to collect blood from
children aged 1 year and over hi NHANES m by vempuneture only (130,131). Because a lesser
amount of blood can be collected from children than from adults, it was not possible to assay the
full battery of nutritional biochemistries in children. For young children, only the most critical
nutritional biochemistries were assayed (appendix table ISO-
One new indicator, red cell distribution width, was added because it may become
abnormal earlier in the development of iron deficiency thsin other blood cell cotmt indicators,
such as hemoglobin or mean cell volume, but after the fall of iron stores (132). ;Many iron-status
indicators are affected by inflammation as well as by iron deficiency (72). The ability to assess
iron status in NHANES HI has been enhanced by the addition of a biochemical measure of
inflammation, C-reactive protein. The C-reactive protein measure, which will also be useful hi
the arthritis component, will aid determination of the prevalence of true iron dejiciency without
confounding by inflammatory conditions. This will be particularly useful when assessing iron
status of older persons, in whom the prevalence of abnorniial iron status indicators as a result of
inflammation is high (133). I j
Serum and red blood cell folate were assessed on all examined persons '4 years of age and
over. For Phase 2 of the survey (1991-94), assessment of homocysteine, methyhnalonic acid,
and vitamin B12 were added to provide population reference data on vitamin B12 status. This
information will be critical to assessing the population's fplate status and folate-vitamin B12
relationships. ; .1
NHANES in provides the most comprehensive assessment of fat-soluble vitamin status
available from an NHANES. In addition to vitamins A and E, which have been measured in at
least one previous NHANES, vitamin D, retinyl esters (which may increase in vitamin A
toxicity), and a profile of five different carotenoids are being assessed.
ffl-12
f: '..I
-------
Clinical assessments related to nutrition.
Unlike previous NHANES, the physician's examination component of NHANES in did
not screen for overt clinical signs of nutritional deficiencies such as keratomalacia, pellagrous
dermatitis, or follicular hyperkeratosis, which are uncommon in the United States. Instead, a
number of nutrition-related health conditions were assessed in NEANES in (see the section
"Health status assessment"), including cardiovascular disease and related risk factors, diabetes,
osteoporosis, dental conditions, and gallbladder disease. Dietary and nutrition-related interview
information (appendix table X) supplement the physical examination findings and allow for
farther study of the interrelationships between nutrition and health in the population and
subgroups at increased risk.
m-13
-------
; I:
i I.
1 i
i :
;,i ;
I,:"1:')
-------
APPENDIX IV
NHANES-in Summary of Examination & Clinical Biochemistry Assessments and
Dietary/Nutritional Evaluations
(Note: With the exception of the footnote in bold on page IV-8, the material in this Appendix is
taken verbatim from the NCHS source document.)
(Source: NCHS, 1994b, pp. 45,46,48,49, 53, 56)
IV-1
-------
-------
I
I
ctt
H
I
s; S
'
li!
1?
ง
p
I* r^t fll WJ *"
5ซSซSซ
JV-3
-------
> I :
' !!! I
tj
ฃ>
I
s
si
I
1
1
o
o
u
I
0\
g
w
I-
'ง
ซ
ง543'
ง w 53
: fe w PL, !
I
.a-
in -
>>
ง I
d,
d I
o Q
a 2
8C-* ca
?"* O
alllti
!n>'E
to 6} ^ *5 *^ o
u .a jj a p,.a
t&m
%J ^ ง
u ^ s
.. '55 "2
S 53 I
:!ง ง
as!
o o iJ
: o n p
HaS
*.!H
, pq
t^
^&
^> s
'ili^
"ฐ i
JlSss
31
.1!
a
I
^
.5 l.g
lillv.
!*filill!
O .ซ> -< J3 ฃ P. "2 ^
ll^l^
S
B ffi
S 1
ฃ> s
a-2.
ง|!
:ซpi
IS
t)p-CJ O W) O q
<3 u &Q K* w ,
O ^3 tw O Cl* !
1g *o ฎ W S '
i3 J e a 1
Is "i! "^tS i
o
-------
I
\
1
oป
a
13
I
<=>
ป
งs
o:
a ง
II
.8-a
s-a
1
@
ll
gl
il
^ O
g.a
>K Mป
JJ ฃ:
lillf:
ฃง
&.i
i -Ma g ซ 5 .
?g
>1=
i ฃ ฃ ฃ ^H g ff 3(
;s?<
?53i3i
1
M
I H S
IV-5'
-------
I
I
C3
O
1
a
ra
I
to
I
I
s,
Os
i H W
BlllWSas^^a
\ -ll ปฐ .
ili;
!
I
I
3
-------
1.
1
v
f
a
a
S
0
1
61
-------
53
.2
CO
O
I
*o *o w ซ
g s -a-a
mi
of cs M rf
IV-8
-------
a
o
s
5
o
&
ample Person and Ty
02
<3
o
.a
-8
=
a
53
3
a
O
'5
S
a
a
.a
>
Table 5. Nutrition-Related Inter
Information
2 months and over
2 months and over
2 months and over
2 months and over
*>^
JS1
24-hour dietary recall
Food security1
Food program participation1
Drinking water source and quant
2 months and over
2 months and over
1
12 years and over
12 years and over
Food frequency
Alcohol use
17 years and over
20 years and over
Antacids use
Lifetime milk frequency
2 months and over
2 months and over
t"
^> ji
Self-(or proxy-) reported height i
Self-(-or proxy) assessed weight
2 months- 11 years ,
1 year and over
en
Birth weight
Weight loss practices and reason
12 years and over
25 years and over
Desired weight
Weight history
j
JB
^
r3
1 Also collected at the household (
rv-9'
-------
1 1
-------
APPENDKV
Comparison of Laboratory Analyses of Blood and Urine Across NHANES-I, -II, -HI, and
Hispanic HANES
(Note: With the exception of minor changes to the table footnotes, the material in this Appendix
is taken verbatim from the NCHS source document.)
(Source: NCHS, 1994b)
V-l
-------
Table 1. Comparison of Laboratory Analyses of Blood and Urine Across
NHAJNES-I, -n, -m and Hispanic HANES
Lab Test
Whole Blood Assessments:
Sedimentation rate
Complete blood count
Platelets
3-cell differential
Differential smear
Red cell distribution width
Lead
Protoporphyrin
Red blood cell folate
Glycated hemoglobin (HbAic)
Carboxyhemoglobin
Serum Biochemistry Assessments:
Folate
Iron & total iron-binding capacity
Ferritin
Vitamin C
Vitamin D
Vitamin E
Zinc & copper
Vitamin A (retinol)
Carotenoids
Retinyl esters
Vitamin B 12
Methyl malonic acid
Homocysteine
Selenium
Total cholesterol
High density lipoprotein cholesterol
Triglycerides
Apolipoproteins A, & B
Lp(a)
Total & ionized calcium
Cotinine
Bile salts
Pesticides(7)
Syphilis
Hepatitis A, B & delta
Hepatitis C
, Hepatitis E
Tetanus
Diphtheria
Polio
Herpes simplex I & II
SURVEY(1)!
I
X
X
X
X
X
X
X
X
X
X
X
n
X
X
X
X
X
X
X
X
X
X
X
x<2)
X
X
X
X
X
X
X
X
X
HH
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
TTT
X
x
X
x
X
X
X
X
X
X
X
x
X
X
X
X
Jf
X
x(3)
x(3)
x(3)
X
X
X
X
x(4)
!x<3>
X
X
X
X
X
X
X
X
X
L:
\r o '!
-------
Lab Test
Serum Biochemistry Assessments:
Human, immunodeficiency virus I
Rubella antibody
Varicella antibody
Toxoplasmosis antibody
Helicobacter pylori
Hantavirus
C-reactive protein
Rheumatoid factor
FSH/LH(5)
Thyroxine (T4)
Thyroid-stimulating hormone (TSH)
Antithyroglobulin antibodies
Antimicrosomal antibodies
Insulin
C-peptide
Biochemistry Profile:
Total carbon dioxide
Blood urea nitrogen
Total bilirubin
Alkaline phosphatase
Total cholesterol
Aspartate aminotransferase (serum
glutamic-oxaloacetic transaminase)
Alanine aminotransferase (serum
glutamate pyruvate transaminase)
Lactate dehydrogenase
Gamma glutamyl transpeptidase
Total protein
Albumin
Creatinine
Glucose
Calcium
Chloride
Uric acid
Phosphorus
Sodium
Potassium
Plasma Assessments:
Plasma fibrinogen
Glucose (oral glucose tolerance test)
SURVEY(1)
I
X
X
X
X
X
x(6)
X
X
X
X
X
X
X
X
n
xซo
x(6)
x(ซ>
X
X
X
HH
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
m
X
X
X
X
x(4)
x(4)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
x
X
X
X
X
X
X
X
X
X
X
V-3
-------
Lab Test
Urinary Assessments:
Urinalysis
Pesticides^
Riboflavin
TMamine
Cadmium
Creatinine
Albumin (micro)
Iodine
Cocaine
Opiates
Phencych'dine
Amphetamines
Marijuana
Sodium
Pregnancy test
Excess and Reserve Vials
Serum
White blood cells for DNA banking
SURVEY(1)
I
X
X
X
X
X
X
X
n
X
X
X
HH
X
X
X
m
X
X
X
X
x<3>
x(3)
x(3)
x(3)
x<3)
X
X
X
X
(1) I is NHANES-I, H is NHANES-H, m is NHANES-IH and HH is 'the Hispanic Health and Nutrition Survey
(2) Children only
(3) Phase 2 subjects only
(4) Phase 1 subjects only
(5) Follicle-stimulating hormone/Luteinizing hormone
(6) Bile salts subset
(7) See Table 1 in Appendix VII of this Handbook
i;;,;ป
V-4
-------
APPENDIX VI
Comparison of Laboratory Analyses, Examinations and Questionnaires Across
NHANES-99,-00,-01,and-02 ,
(Note: The material in this Appendix is taken verbatim from the NCHS source document.
Footnotes were added for clarification as indicated in each table.)
(Source: 1SIHANES Consortium Meeting, 2002)
VI-1
-------
,
Table 1. Laboratory Analyses of Blood, Urine and Hair for NHANES-99 to- 02
Lab Test
Lead<ซ>
Cadmium ^
Erythrocyte protoporphyrin
Red blood cell folate
Serum folate
Glycohemoglobin
Mercury (hair)w
Mercury (blood)(1)
Mercury (urine)(1)
CD4ป
White Wood cells/DNA
Iron
TEBC^
Ferritiri
Vitamin B 12
C-Reactive protein
Helicobacter pylori
Cryptosporidium
Vitamin A/E/Carotenoids
Vitamin C
Measles/Varicella/Rubella
Cotinine
Chemistry panel
Bone alkaline phosphatase
Toxoplasma<4)
Total cholesterol
High density lipoprotein
Low density lipoprotein
Triglycerides
HIV antibody
Insulin/c-peptide
Herpes. 1 & 2 antibody
Syphilis
Human papiloma virus
Prostate specific antigen
FSH/LHฎ
Latex
Vitamin D
TSH/EHฎ
EPA persistent pesticides(I)
surplus sera.
Homocysteine
Methyl malonic acid
4,
Age Range
1 year & older !
1 year & older
1 year & older
3 years & older
3 years & older
12 years & older
1- 5 years & females 16-49
1- 5 years & females 16-49
1- 5 years & females 16-49
18 -49 years . "
20 years & older
1 year & older
1 year & older
1 year & older
3 years & older
3 years & older
3 years & older
6-49 years
3 years & older
6 years & older
6-49 years
3 years & older
12 years & older
8 years & older
6-49 years
3 years & older
3 years & older
Subsample 3 years & older
Subsample 3 years & older
18 -49 years
Subsample 12 years & older
14 - 49 years
18 - 49 years
14 - 59 years
males 40 years & older
females 35 - 60 years
12 - 59 years
6 years & older ,
1/3 subsample 12 years & older
1/3 subsample 6 years & older
6 years & older
3 years & older
3 years & older
YEAR OF SURREY
1999
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2000
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2601
X'
X
x;
X
X;
x :
>x i
X '
X
X
X
X
X
X
X
X
x ;
x i
X
X :
X
x '.
X
x :
X
X '
X
X
X
X
X
X '
X ,
X
X
X .
X
X
X
X
X
2002
x .
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
VI-2
! ', .Mil
n r-- " I
-------
Lab Test
Glucose (plasma )
Fibrinogen
Hepatitis B antibodies
Hepatitis A, B, C, D HbSAg
Complete blood count
Selenium
Chlamydia (urine)
Gonorrhea (urine)
Albumin (urine)
Creatinine .(urine)
NTX(7)
Iodine (urine)
BV/Trich(8)
MRSA<9>
VOC(IO) exposure monitor(1)
Pthalates(1)
OPmetabolites(1'H)
Metals(1)
Nonpersistent pesticides-LC(1)
Nonpersistent pesticides-GC(1)
Phytoestrogens(1)
PAHs'1'12'
Dioxins(1)
Lead dust^
VOC(blood)(1-10>
Age Range
Subsample 12 years & older
40 years & older
2-5 years
6 years & older
lyear & older
3-11 years
14 - 39 years
14 - 39 years
6 years & older
6 years & older
8 years & older
1/3 sample 6 years & older
Females 14 - 49 years
1 year & older
Subsample 20 - 59 years
1/3 subsample 6 years & older
1/3 subsample 6 years & older
1/3 subsample 6 years & older
1/3 subsample 12 - 19 years
1/3 subsample 12 years & older
1/3 subsample 6 years & older
1/3 subsample 6 years & older
1/3 subsample 12 years & older
Households with 1-5 year olds
Subsample 20 - 59 years
YEAR OF SURVEY
1999
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2000
X
X
X .
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2001
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2002
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
(Sources: Table was presented at the 2002 NCHS Consortium meeting of the NHANES sponsoring agencies; several
footnotes were generated based on information in NCHS, 2001.) ' .
(1) See Table 1 from Appendix VII of this Handbook
(2) La addition to HTV testing in NHANES-99+, whole blood samples will be collected and stored for ftrture CD4
testing once the HIV status of the subject is known. This will allow for the determination of the distribution of CD4
cells in a random sample pf HIV positive individuals.
(3) TIBC is Total Iron Binding Capacity.
(4) Surplus sera used for years 1999 and 2000.
(5) FoUicle-stimulating hormone/Luteinizing hormone
(6) Thyroid-stimulating hormoneTThyroxine
(7) NTX is a marker in the urine indicative of bone mineral resorption. It is used in conjunction with bone alkaline
phosphatase (a marker in the serum, for bone formation) to evaluate a subject's bone mineral status.
(8) BV/Trich are bacterial vaginosis and Trichomoniasis. Both are tested by using vaginal swabs.
(9) MRSA is metMcillin-resistant Staph aureus. Nasal swabs will be tested for this bacterium.
(10) VOCs are volatile organic compounds.
(11) OP is organophosphate pesticide.
(12) PAH is polyaromatic hydrocarbon.
VI-3
-------
111:'!" I liSi '1 'I
Table 2. Questionnaire Matrix for NHANES-99 to -Ok
Component
Subject Questionnaire:
Acculturation
Audiometry
Balance
Hospital utilization & access to health care
Blood pressure
Cardiovascular diseases
Demographics
Dermatology
Diabetes
Diet behavior & nutrition
Dietary supplements & medications
Digit Symbol Substitution Test<2)
Early childhood
Immunizations
Introduction & verification
Kidney conditions
Medical conditions
Miscellaneous pain
Occupation
Oral health
Osteoporosis
Physical activity & physical fitness
Physical functioning
Respiratory health & disease
Smoking & tobacco use
Social support
Tuberculosis
Vision
Weight history
Family Questionnaire:
Demographic background/Occupation
Food security
Health insurance
Housing characteristics
Income
Pesticide use
Smoking
Tracking & tracing
Age Range(1)
12 years & older
20 years & older
40 years & older
0 years & older
16 years & older
40 years & older
0 years & older
6 years & older
1 year & older
0 years & older
0 years & older
60 years & older
0-15 years
0 years & older
0 years & older
20 years & older
1 year & older
20 years & older
12 years & older
2 years & older
20 years & older
2 years & older
1 year & older
1 year & older
20 years & older
60 years & older
1 year & older
20 years & older
16 years & older
0 years & older
0 years & older
0 years & older
0 years & older
0 years & older
0 years & older
0 years & older
0 years & older
,
YEAR<
1999
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
3FSURV
2000
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
EY:
2001
' 1
X
X ;
X
X i
X ,
X
X
xl
x i
X '
x i
X
x
x '
X
X
X :
X
X ;
X
x:
X :
X
X i
X ''
X
X ;
X
X
X
X
X
X
X
x
X
2002
X
X
X
X .
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
(Sources: Table was presented at the 2002 NCHS Consortium meeting of the NHANES sponsoring agencies;
footnote #2 was generated based on information in NCHS, 200 Ib.)
(1) Age range of subjects varies for specific components within each of the specified sections,
(2) Digit Symbol Substitution Test requires the subject to correctly code a series of symbols in 120 seconds. This
exercise is used as a sensitive measure of dementia, and requires response speed, sustained attention, visual spatial
skills, associative learning, and memory.
VI-4
-------
Table 3. Examination Matrix for NHANES-99 to -02
Component
Audiometry
Balance
BIA<2>
Body measures (anthropometry)(3)
Cardiovascular fitness
DXA<4>
Dietary^
Lower Extremity Disease
Peripheral vascular disease
Peripheral neuropathy
Muscle strength
Vision
Oral health
Physician's exam
Blood pressure
Sexually transmitted diseases
BV/Trich(6)
Syphilis
PSA (prostate-specific antigen)
HPV(human papiloma virus)
TB (tuberculin) testing
VQC badge
Hair mercury
MRSA(7>
Hepatitis C follow-up
Dermatology exam(8)
MEC interview
CAPI<9>
Physical activity
Current health status
Tobacco use
Alcohol use
Reproductive health
Kidney-urogenital health
ACASI(10)
Sexual behavior
Drug use
Tobacco use
Alcohol use
Youth conduct disorder
Urologic health & PSA
Mental health questions
Age Range
half sample 20-69 years
see footnote (1)
g-49 years
all ages
12 - 49 years
see footnote (4)
all ages
40 years .& older
40 years & older
40 years & older
50 years & older
12 years & older
2 years & older
all ages
8 years & older
14 - 49 years
14 _ 49 years
18-49 years
males 40 years & over
females 14 -59 years
1 year & older
subsample 20 - 59 years
1-5 years; females 16-49
1 year & older
all subjects w/Hepatitis C
20 - 59 years
12 years & older
12 years & older
.12-15 years
12 years & older
20 years & older
20 years & older
12 years & older
20 years & older
12 years & older
14 - 59 years
14 - 59 years
12 - 19 years
12 - 19 years
12 - 19 years
males 20 years & older
8-39 years
YEAR OF SURVEY
1999
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
,
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2000
X
X
X
X
X
X
X
X
X
X
X
X
x-
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2001
X
X
X.
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2002
X
X
X
X
X
X
x(5)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
fg\
1C- '
X
X
X
X
X
X
X
X '
X
X
X
X
X
X
X
X
o. a o . *. . - , . .
footnotes were generated based on information in NCHS, 2001and 2001a.)
(1) MEC locations 101-124 tested a half sample of subjects 40-69 years. Locations 125 and beyond will test
VI-5
-------
individuals 40 years and older.
(2) BIA is bioelectrical impedance analysis and is a clinical method for measuring impedance of body tissues. The
resulting information provides data on body composition, including lean and fat tissue. !
(3) Body measures include weight, stature, recumbent length, lengths and circumferences of body parts, and s'Jdnfold
measures. j ; :
(4) DXA is dual-energy X-ray absorptiometry and is a clinical method for measuring body composition (bone, lean
and fat tissue). DXA provides information on bone density and soft tissue composition. MEC locations 101-112
tested males 8 years and older and females 18 years and older. Locations 113 and beyond will test subjects 8 years
and older. j ; j
(5) A new dietary interview was started in 2002 and included a 21"1 i3ay recall by phone interview.
(6) BV/Trich are bacterial vaginosis and Trichomoniasis. Both arejtested by using vaginal swabs.
(7) MRSA is methicilUn-resistant Staph aureus. Nasal swabs will be tested for this bacterium.
(8) The dermatology exam will be pilot tested in 2002, and will go into field in 2003.
(9) CAPI is Computer-Assisted Personal Interviewing. Randomly selected people are invited to participate in
NHANES-99-l- by first being interviewed in their homes. Household interview data are collected, via CAPI and
include demographic, socioeconomic, dietary, and health-related questions. After completion of the interview,
respondents are asked to participate in a physical examination at th<5 MEC. ',
(10) ACASI is Audio-Computer-Assisted-Self-Interviewing (CASr)system, and is generally used for sensitive topic
areas. Subjects listen to a recorded voice though a headset, as well as reading the questions on the screen, and then
indicate their response by touching the computer screen. ACASI is conducted at the MEC.
VI-6
-------
APPENDIX VII
Comparison of Environmental Data Collected Across all NHANES
(Sources: NCHS, 1973,1977,1981,1985,19941?; NCHS website:
for Survey Questionnaires,
Examination Components and Laboratory Components for NHANES 99+)
VIM
-------
I .
Table 1. Comparison of Key Environmental Data Collected Across the NHANES Surveys
.'.,,'
NHANESI
1971-1975
NHANESH
1976-1980
HHANES
198J2-1984
NHANESin
1988-1994 ;
NHANES 99+
(1999-
indefinitely)
BLOOD AND SERDM ;
Cadmium
Carboxyhemoglobin
Cotinine1
Dioxins, Furans,
Coplanar PCBs2
Lead
Mercury3
PCBs4
Pesticide residues &
metabolites5
Phytoestrogens6
Volatile Organic
Compounds
(VOCs)7
No
No
No
No
No
No
No
No
No
No
No
SPs3-74yrs
0/4 sample)
No
No
all children
6 mo - 6 yrs; &
SPs7-74yrs
04 sample)
No
No
SPs 12-19 yrs
04 sample) &
all SPs 20-74 yrs
No
No
No
SPs
04!
inti
3-74 yrs
ample
le first yr)
No
No
SPs
yrs
No.
6 mo-74
all SPs 12-19
yrsdfc
SPs 20-74 yrs
06 sample )
all Sps 12-19
yrs<&
SPs 20-74 yrs
04 sianple)
No
No
No
No
SPs 4-74+ yrs
No
SPs 1-74+ yrs
No
No
No
No ;
non-random
subsample of
1000 SPs
20-59 yrs '
1+yrs
No
SPs 3+ yrs
SPs 12+ yrs
(1/3 sample)
SPs 1+yrs
All SPs 1-5 yrs
& women 16-
49 yrs
SPs 12+ yrs
(1/3 sample)
SPs 12+ yrs
(1/3 sample)
SPs 12+ yrs
(1/3 sample)
random
subsample each
year of 1000
SPs 20-59 yrs
URINE
Hadmium
No
No
No
VII-2
SPs 6-74+ yrs
SPs 6+ yrs
(1/3 sample)
,: . , "; ;
-------
Heavy Metals (other
than Cadmium)8
Organophosphates
Pesticide Screen9
Pesticide residues &
metabolites10
Phthalates11
Phytoestrogens6
Polyaromatic
Hydrocarbons
(PAHs)12
Lead dust hi
homes"
Household water
samples14
Hair15
Personal Smoking
& Environmental
Tobacco Smoke
Questions16
Pesticide Exposure
(Self-reported)17
No
No
No
No
No
No
No
No
SPs 12-74 yrs
(l/2 sample)
No
No
No
No
No
SPs 12-74 yrs
0/2 sample)
No
No
No
OTHER
No
trace
minerals
drinking
water
3059 SPs
25-74 yrs
No
SPs 25-74 yrs
No
No
No
No
SPs 12-74 yrs
SPs 12-74 yrs
No
No
All SPs ages
12-74 for
selected trace
metals for a
CDC pilot
study
SPs 12-74 yrs
SPs 12-74 yrs
No
No
1000 SPs ages
20-59 yrs were
to be tested for
several
pesticides but
were not
No
No
No
No
No
No
SPs 8+ yrs.
No
SPs 6+ yrs
(1/3 sample)
SPs 6- 11 yrs
0/2 sample) &
12+ yrs (1/4
sample)
SPs 6- 11 yrs
0/2 sample) &
12+ yrs (1/4
sample)
SPs 6+ yrs
(1/3 sample)
SPs 6+ yrs
(1/3 sample)
SPs 6+ yrs
(1/3 sample)
Households
with SPs 1-5
yrs
VOCs from
home tap water
of a random
subsample of
1000 SPs
20-59 yrs
All SPs 1-5 yrs
& women 16-
49 yrs, for total
mercury
All SPs
Use of
pesticides in
home & yard
VII-3
-------
'Cotinine is a metabolic byproduct of nicotine metabolism. Coriniie levels and other data (i.e., self-reported
smoking and exposure to others who smoke) gathered in NHANES-m and NHANES99+ caft be used to help assess:
1) prevalence and extent of tobacco use in subjects; 2) extent of exposure to environmental tbbacco smoke (ETS),
and determine trends in exposure to ETS; and 3) relationships between tobacco use (and/or ETS) and chronic health
conditions (e.g., respiratory and cardiovascular diseases). j j .
2 NHANES99+ tested for the following dioxins, furans and coplaniir PCBs in serum: 2,3,7,8-Wachloro-dibenzo-p-
dioxin (tcdd); 1,2,3,7,8-Pentachloro-dibenzo-p-dioxin (pncdd); 1,2,3 ,4,7,8-Hexachloro-dibenzo-p-dioxin (hxcdd);
1,2,3,6,7,8-Hexachloro-dibenzo-p-dioxin (hxcdd); 1,2,3,7,8,9-Hexachloro-dibenzo-p-dioxin jChxcdd); 1,2,3 4678-
HeptacbJoro-dibenzo-p-dioxin (hpcdd); 1,2,3,4,6,7,9-Heptachloro-idibenzo-p-dioxin (hpcdd);' 1,2,3,4,6,7,8,9-
Octachloro-dibenzo-p-dioxin (ocdd); 2,3,7,8-Terrachloro-dibenzoftean (tcdf); l,2,3,7,8-Pentachloro-diben;zo&ran
(pncdf). | ; i
:, " ! !.
; .'. ' /, ...... , . , 1 ' ; :'',, ' '. ..
In NHANES99+, mercury was tested in the blood, hair and urine of this subsample of women and children.
'"' ' ''
4 HHANES did not identify specific PCBs that were, tested. NHANJES99+ tested for the following noncoplanar
PBCs in serum: PCB-19,28,44,49,52, 56, 60, 66,74, 87, 99,101,
105, 110,116,118, 128,|l38, 146,149.
Pesticide residues and metabolites in blood or serum for NHANES-n, HHANES, and NHANES99+: aldrin, beta-
BHC, gamma-EHC, op'-DDD, pp'-DDD, op'-DDE, pp'-DDE, op'-[DDT, dieldrin, endrin, heptachlor epoxide,
hexachlorobenzene, mirex, oxychlordane, ftww-Nonachlor. NHANES-n and HHANES (but not NHANES99+) also
tested for a/JpAa-BHC,delta-BHC,pp'-DDT,heptachlor. f ' !|
Additional pesticide residues and metabolites tested in blood or senim only in NHANES99+: a/^a-Chlordane, cis-
Nonachlor, gamma-Cbloidane, Lindane, Nitrobenzene. j , j
6 Phytoestrogens measured in both serum and urine in NHANES99-I- were: Coumestrol, Daidzein, Enterodiol,
Enterolactone, Equol, Genistein, Matakesinol, o-Desmethylangolensin (O-DMA). ;
I1, . i ]. . ! i . .:.. L
7 In NHANES-m, only blood samples were analyzed for selected ViOCs. In NHANES99+, personal exposures to
VOCs were characterized by testing for chemicals in personal air saimples, blood, and drinking water of selected
subjects. Air samples were obtained by small lightweight passive silinpling badges worn by subjects for 48-hours.
Badges were returned to the MEC 48-72 hours later along with a sanple of tap water from their homes. Blood
samples were taken when the badges were returned. Both NHANES-IH and NHANES99+ measured blood levels of
the following VOCs: 1,1,1-Trichloroethane; l,l,2,2-TetrachloroethฃJne; 1,1,2-Trichloroethane- 1,2-Dichloroethane;
1,2-Dichloropropane; 1,3-Dichlorobenzene; 1,4-Dichlorobenzene; 2j-Butanone; Acetone; Benzene;
Bromodichloromethane; Bromoform; Carbon Tetrachloride; Chlorobenzene; Chloroform; Dibromochloromethane;
Dibromomethane; Ethylbenzene; Hexachloroethane; Xylenes; Methylene chloride; Styrene; Tetrachloroethene;
Toluene; Trichloroethene. NHANES99+ also tested blood for the following VOCs that were not tested in
NHANES-in: 1,1-Dichloroethane; 1,1-Dichloroethene; l,2-Dichlon>benzene; cis-l,2-Dichloroethene; Methyl
tertiary-butyl ether; trans- 1,2-Dichloroethene. ! 1 '
i i . !
5 [
: [
NHANES99+ VOCs measured in both air and blood were: benzene,; carbon tetrachloride, chloroform,
1,4-dichlorobenzene, ethyl benzene, methylene chloride, methyl tertiary-butyl ether, styrene, toluene, xylenes,
tetrachloroethylene, trichloroelhylene. VOCs measured in air but n
-------
Tungsten, Uranium.
9 Organophosphate pesticides in this screen were: Dimethylphosphate, Diethylphosphate, Dimethylthiophosphate,
Diethylthiophosphate, Dimethyldithiophosphate, Diethyldithiophosphate
ป Pesticide residues and metabolites measured in urine of subjects from NHANES-H, HHANES and NHANES99+
were: 1-Naphthol; 2,4,5-T; 2,4,5-Tricholorphenol; 2,4-D; 2-Isopropoxyphenol; 3,5,6-Tnchloropyridrnol;
Carbofuranphenol; Dicambajpara-nitrophenol, Pentachlorophenol.
Additional pesticide residues and metabolites measured in urine from NHANES-H subjects only were: alpha-
Monocarboxylic acid; Dicarboxylic acid; DMTP, DETP, DMDTP, DEDTP, DMP, and DEP. .
Additional pesticide residues and metabolites measured in urine from HHANES subjects only were: Malathion
Dicarboxylic Acid andMalathion Monocarboxylic Acid.
Additional pesticide residues and metabolites measured in urine from NHANES99+ subjects only were:
2 4 6-TrichLophenol; 2,4-Dichlorophenol; 2,5-Dichlorophenol; 2-Naphthol; 3,4-Dichloroamlme; 3-Phenoxy
Benzoic Acid; Ilachlor Mercapturate; Atrazine Mercapturate; DEBT, Glyphosate, Malatmon di-acid; Metolachlor
Mercapturate; Oxypyrimidine; o-Phenyl phenol.
ป Phthalates tested in urine were: Bisphenol A; Mono.(2-ethyl)-hexyl phthalate; Mono-benzyl phthalate; Mono-
cyclohexyl phthalate; Mono-ethyl phthalate; Mono-isodecyl phthalate; Mono-isononyl phthalate; Mono-n-butyl
phthalate; Mono-n-octyl phthalate; Nonyl-phenol; Octyl-phenol.
" Polyaromatic hydrocarbons tested in urine were: 1-Hydro^-aniline; 1^^
chrysene; 1-Hydroxy-naphthalene; 1-Hydroxy-pyrene; 2-Hydroxy-amhne; 2-Hydroxy-naphthalene, 2-Hydroxy-
Er^ene 3-Hydroxy-benzManthracene; 3-Hydroxy-benzo[3] pyrene; 3-Hydroxy-benzota] pyrene; 3-Hydroxy-
beSotS^-e; 3?Hydroxy! dibenz [a^antoacene; 3-Hydroxy-fluorene; 3-Hydroxy-ph -anthrene; ^dro xy-
acenaphthalene; Hydroxy-benzo[a]fluoranthene; Hydroxy-benzoMphenanthrene; Hydroxy-benzofeWperylene,
Hydroxy-benzoQlfluoranthene; Hydroxy-benzo[k]fluoranthene; Hydroxy-fluoranthene;
Hydroxy-indeno[l,2,3-cd]pyrene; Hydroxy-perylene.
ป Si NHANES99+ , window sill and floor dust samples are collected in the home at fte time of the household
intervie^ This information can be used to examine the relationship between dust and blood levels, along with other
risk factors and housing characteristics.
14 As oart of the NHANES-I Augmentation Survey of Adults 25-74 years of age, water samples were collected from
sutec^tap^r weUs and from^blic water distribution supplies and tested for hardness, afcahmty, total amount of
solute present, and levels of trace minerals. Subjects also were asked numerous questions about personal water
conSU : sources of drinking water, and water supplied to the household. This study was conducted to examme
possible relationships between hardness and trace minerals in the water and cardiovascular disease.
As part of the special VOC study in NHANES99+ (see footnote #7) VOCs measured in home water were:
Chloroform, Bromodichloromethane, Bromoform, Methyl tertiary-butyl ether, Chlorodibromomethane.
"The objective of the NHANES99+ hair mercury component was to document total mercury levels in hair, because
relationships have been established between the concentrations of mercury in human scalp hair and dietary
SySw exposure. Hair samples were collected during flu, first three years of NHANES99H- from males^and
females aged 1-5 years and females aged 16-49 years.
16 Personal smoking- NHANES-I, -H asked the same several questions about smoking habits (cigarettes cigars, pipe,
chewing tobacco) of the subjects'. NHANES-HI subjects ages 8+ yrs , were asked similar questions. All
VII-5
-------
more
MIANES99+ subjects were asked about personal smoking and those 20+ years old were asked additional
detauedquestions. \ i \ .'
E^mnmental tobacco smoke (ETS): NHANES-HI and NHANESJW- inventoried the number of people who smoke
in each household and how much each one smokes per day. Additionally in NHANES-ffl, youths up to 16 yrs old
were asked questions about the number of people who smoke cigaiettes in the home and how much each one smokes
and subjects 17+ years were also asked about ETS in the workplac,. All NHANES99+ subjects were also asked '
about exposure to ETS at home and at work and in-utero ETS exposure among children. ',
17 NHANES-n asks questions about use of pesticides, such as weedkillers, insecticides, fungicides and other
chemicals, at me home, in the yard or at work. HHANES asks many questions about subject?s work in the farming
industry and use of pesticides as part of subject's employment. NEIANES99+ asks questions! about me use of
pesticides in the home and yard.
vn-6
-------
APPENDIX VIII
ANNOTATED BIBLIOGRAPHY OF STUDIES USING NHANES DATA
(Sources: NCHS, 1996d, 1999b)
Vffl-1
-------
Albanes, D; Jones, DY; Micozzi, MS; Mattson ME (1987). Associations between smoking and
body weight in the US population: analysis of NHANES E. Am. J. Pub. Hlth.; 77(4, Apr), 439-
444. [adult; aged; body height; body weight; caloric intake; NHANES-E; smoking]'.
Albanes, D; Jones, DY; Schatzkin, A; Micozzi, MS; Taylor, PR (1988). Adult stature and risk of
cancer. Cancer Res. 48(6, Mar 15), 1658-1662. [adult; jaged; body height; caloric intake; middle
age; neoplasms, epidemiology; NHANES-I; nutrition; risk-factors; smoking]. ;
Albanes, D; Blair, A; Taylor, PR (1989). Physical activity and risk of cancer in the NHANES I
population. Am. J. Pub. Hlth. 79(6, Jun), 744-750. [adult; aged; body constitution; body weight;
cohort studies; exercise; health surveys; life style; middle age; neoplasms, epidemiology;
NHANES-I; nutrition surveys; recreation; risk-factors]. ;
i" , .. [ .
Anda, RF; Williamson, DF; Escobedo, LG; Remington, PL (1990). Smoking |and the risk of
peptic ulcer disease among women in the United States, jlrch. Intern. Med. 150(7, Jul), 1437-
1441. [adult; aged; cohort studies; follow-up studies; incidence; middle age; NHANES-I; peptic
ulcer, epidemiology; peptic ulcer, etiology; prevalence; risk-factors; sex factors; smoking,'
adverse effects; United States, epidemiology]. ' i ' !
I _ , ' ; j
Anderson, LA Jr (1995). A review of blood lead results feom the Third National Health and
Nutrition Examination Survey (NHANES IE). Am. Ind. Hyg. Assoc. J. 56(1, Jan), 7-8.
[environmental exposure; health surveys; lead poisoning] prevention and control; lead, blood-
NHANES M; United States]. ! i
, |... , ,'... j I i ";, : , j
Anon. (1989): Jjron nutrition and risk of cancer. Nutr. Re^. 47(6, June), 176-178 (Includes 1 1
references.) [iron; nutritional state; nutrient intake; iron binding capacity; carcinoma; risks;
epidemiology; nutrition surveys; NHANES-rj. ; [
Ametz, BB; Nicolich, MJ (1990). Modeling of environmental lead contributors to blood lead in
human. Int. Arch. Occup. Env. Hlth. 62(5), 397-402. [enyironmental exposure; lead poisoning
epidemiology; NHANES E]. \ , V &
Arrighi, HM; Boyd, BK; Casty, FE (1995). Childhood asthma and adult height: Data from
SHANES E.Pediatrtc Asthma, Allergy and Immunology 9(3), 157-163.
[adult; aged; asthma; body height; Caucasian; data analysis; education; female; high school:
human; major clinical study, male; negro; NHANES E; scicial status]. :
, ,
Ashley, DL; Bonin, MA; Cardinali, FL; McCraw, JM; We
oten, JV (1994). Blood concentrations
, , .
of volatile organic compounds in a non-occupational expc sed US population and in groups with
suspected exposure. Clin. Chem. 40(7 Pt 2, Jul), 1401-1404. [environmental exposure-
environmental pollutants, blood; NHANES ETj. '
Ballard-Barbash, R; Schatzkin, A; Taylor, PR; Kahle, LL (l990). Association of change in body
VHI-2
-------
mass with breast cancer. CancerRes. 50(7, Apr 1), 2152-2155. [alcohol drinking; body weight;
breast neoplasms, etiology; NHANES I; NHEFS].
Ballew C; Khan, LK; Kaufmann, R; Mokdad, A; Miller, DT; Gunter, EW (1999). Blood lead
concentration and children's anthropometric dimensions in the Third National Health and
Nutrition Examination Survey (NHANES 111), 1988-1994. /. Pediatr. 134 (5, MayV623-630.
[blood lead levels; BMI; children; height; Mexican Americans; NHANES-H; NHANES-ffl; non-
Hispanic Blacks; non-Hispanic Whites]. ;
OBJECTIVE: To assess the association between lead exposure and children s physical growtn.
DESIGN- Cross-sectional analysis of data from the Third National Health and Nutrition
Examination Survey, 1988-1994. PARTICIPANTS: A total of 4391 non-Hispanic white, non-
Hispanic black, and Mexican-American children age 1 to 7 years. Measurements and Results:
We investigated the association between blood lead concentration and stature, head
circumference, weight, and body mass index with multiple regression analysis adjusting for sex,
ethnic group, iron status, dietary intake, medical history, sociodemographic factors, and
household characteristics. Blood lead concentration was significantly negatively associated with
stature and head circumference. Regression models predicted reductions of 1.57 cm rn stature
and 0.52 cm in head circumference for each 0.48 micromol/L (10 micrograms/dL) increase rn
blood lead concentration. We did not find significant associations between blood lead
concentration and weight or body mass index. CONCLUSIONS: The significant negative
associations between blood lead concentration and stature and head circumference among
children age 1 through 7. years, similar in magnitude to those reported for the Second National
Health and Nutrition Examination Survey, 1976-1980, suggest that although mean blood lead
concentrations of children have been declining in the United States for 2 decades, lead exposure
may continue to affect the growth of some children.
Bang KM- Gergen, PJ; Carroll, M (1990). Prevalence of chronic bronchitis among US
Hispanics from the Hispanic Health and Nutrition Examination Survey, 1982-84. Am,J. Pub.
Hlfh. 80(12, Dec), 1495-1497. [adolescence; bronchitis, epidemiology; HHANES; Hispanic
Americans; DHES]
Bang KM (1993). Prevalence of chronic obstructive pulmonary disease in blacks. J. Natl. Med.
Assoc. 85(1, Jan), 51-55. [adult; lung diseases, obstructive, epidemiology; negroid race;
NHANESrj
Bang KM; Gergen, PJ; Kramer, R; Cohen, B (1993). The effect of pulmonary impairment on
all-cause mortality in a national cohort. Chest 103(2, Feb), 536-540. [adult; aged; forced
expiratory volume; mortality; NHANES-I; NHEFS; vital capacity; DHES].
Basu, S; Landis JR (1995). Model-based estimation of population attributable risk under cross-
sectional sampling. Am. J. Epid. 142(12, Dec 15), 1338-1343.
[adolescence; adult; cohort studies; models, statistical; NHANES-IIJ
VUI-3
-------
! " . i | " . :
Berney, B (1993). Round and round it goes: the epidemiology of childhood lead poisoning,
1950-1990. Milbank Q. 71(1), 3-39. [attitude to health! child; child welfare, legislation and
jurisprudence; child welfare, trends; health policy, trends!' lead poisoning, epidemiology;
NHANES-n; public health, trends].
Block ,G (1992). Dietary assessment issues related to cancer for NHANES in. Vital Health Stat.
4(27, Mar), 24-31. [case-control studies; cohort studies; cross-sectional studies; data collection,
methods; diet records; diet surveys; diet, adverse effects;
questionnaires]
neoplasms, etiology; NHANES-IH;
Briefel, RR (1994). Assessment of the US diet in national nutrition surveys: national
collaborative efforts and NHANES. yim. J. Clin. Nutr. 59(1 Suppl, Jan), 164S-167S.
[adolescence; adult; aged; child; child, preschool; diet; eljhnic groups; infant; middle age;
NHANES-HT; nutrition assessment; nutrition surveys; nutritional status; research design; United
States; DHES] : |
i - j 1
Brody, DJ; Pirkle, JL; Kramer ,RA; Flegal, KM; Matte, TD; Gunter, EW; Paschal, DC (1994).
Blood lead levels in the US population. Phase 1 of the third National Health and Nutrition
Examination Survey (NHANES HI, 1988 to 1991) [see comments]. JAMA 272(4, Jul 27), 277-
283. [adolescence; adult; age factors; aged; child; child, preschool; cross-sectional studies;
health surveys; infant; lead, blood; NHANES HI; DHES] | ;
Brody, DJ; Pirkle, JL; Kramer, RA; Flegal, KM; Matte, ID; Gunter, EW; Centers for Disease
Control and Prevention (1995): Erratum: Blood lead levels in the US population: Phase 1 of the
Third National Health and Nutrition Examination Survey (NHANES JJJ, 1988 to 1991, JAMA
(1994) 272,277-283). JAMA 274(2), 130. [erratum; erroi; NHANES IE; DHES]
! :' ! :
Buckley, TJ; Liddle, J; Ashley, DL; Paschal, DC; Burse, yW; Needham, LL; Akland, G (1997).
Environmental and biomarker measurements in nine homfes in the Lower Rio Grande Valley:
Multimedia results for pesticides, metals, PAHs, and VOCs. Envir. International 23, 705-732.
[environmental biomarker; NHANES JJDQ | :
Residential environmental and biomarker measurements were made of multiple pollutants during
two seasons (spring and summer, 1993) in order to assess human exposure for a purposeful
sample of 18 nonsmoking adults residing within nine homes (a primary and secondary subject in
each home) in the Lower Rio Grande Valley (LRGV) near Brownsville, TX. Pesticides, metals,
PAHs, VOCs, and PCBs were measured in drinking water, food, air, soil, end house dust over a
one- to two-day period in each season. Biomarker measurements were made in jblood, breath, and
urine. A total of 375 measurements across five pollutant classes (227 pesticides. 44 trace
elements, 78 VOCs, 18 PAHs, and 8 PCBs) was possible for each home in one or more media. A
large percentage of the measurements was below the mettled limit of detection ranging from 0-
37% for pesticides, 22-61% for metals,. 6% and 90% for VOCs in water and ami respectively, and
0-74% for PAHs. The total number of analytes measurable in blood, urine, or breath was
considerably less, i.e., 58 (21 pesticides, 1 PCB, 4 metalsjsi VOCs, and 1 PAH) with the
i;':
'. i I; : t
VJH-4
-------
percentage above the method limit of detection for pesticides and metals ranging from 40 to
100%, while for VOCs, PAHs, and PCBs, this percentage ranged from 2 to 33%. A significant
seasonal difference (p < or = 0.10) was found in the biomarker levels of two of seven
nonpersistent pesticides (3,5,6- trichloro-2-pyridinol end 2,5-dichlorophenol) and 3 of 3 metals
(arsenic, cadmium, and mercury) and the pyrene metabolite, 1-hydroxypyrene, measured in urine.
In all cases levels were higher in the summer relative to the spring. For the persistent pesticides
and PCBs in blood serum, a seasonal effect could be evaluated for 5 of 10 analyzes; a significant
difference (p* 0.10) was observed only for hexachlorobenzene, which like the urine biomarkers,
was higher in the summer. In contrast to the urine metals, blood-Pb concentrations did not
change significantly (p< or = 0.05) from spring to summer. Biological results from the current
study are compared to the reference range furnished by the Third National Health and Nutation
Examination Survey (NHANES ffl). Comparisons are only suggestive due to limitations in
comparability between the two studies. Based on the percentage of measurements above me
detection limit, a significant elevation (psO.10) in 2 of 12 nonpersistent pesticides (4-mtrophenol
and 2 4-D) was observed for the LRGV study subjects,, The VOC carbon tetrachlonde was found
in the'blood (monitored only in spring) with greater prevalence (p^O.10) than would be expected
from NHANES ffl results. Blood serum levels of two persistent pesticides (4,4'-DDE, and trans-
nonachlor) and PCB exceeded median and/or 95th percentile reference levels as did arsenic in
urine Where seasonal differences were identified or for compounds exceeding reference levels,
environmental monitoring results were investigated to identify potential contributing pathways
and sources of exposure. However, because environmental sampling did not always coincide
with the biological sampling and because of the high frequency of analytes measured below the
limit of detection, sources and pathways of exposure in many cases could not be explained,
Chlorpyrifos was an exception where urine metabolite (3,5,6-TCP) levels were found to be
significantly correlated with air (R2=0.55; p^O.Ol) and dust (R2=0.46; p<0.01) concentrations.
Based on the results of biomarkers and residential environmental measurements over two
seasons this seeping study shows a seasonal effect for some analytes and suggests where
exposures may be high for others. This information may be useful in considering future studies in
the region.
Burt VL- Harris T (1994) The third National Health and Nutrition Examination Survey:
contributing data on aging and health. Gerontologist 34(4, Aug), 486-490. [activities of daily
living- aged- aged, 80 and over; cardiovascular diseases, epidemiology; health status; health
surveys; longitudinal studies; lung diseases, obstructive, epidemiology; musculoskeletal diseases,
epidemiology; NHANES-IH; nutrition surveys; nutritional status; DHES]
Caraballo RS; Giovino, GA; Pechacek, TF; Mowery, PD; Richter, PA; Strauss, WJ; Sharp, DJ;
Eriksen, MP; Pirkle, JL; Maurer, KR (1998). Racial and ethnic differences in serum cotinme
levels of cigarette smokers: Third National Health and Nutrition Examination Survey, 1988-1991
[see comments]: JAMA 280 (2, 8 Jul), 135-139. [NHANES ffl; serum creatinine; smoking]
CONTEXT- Cotinine, a metabolite of nicotine, is a marker of exposure to tobacco smoke.
Previous studies suggest that non-Hispanic blacks have higher levels of serum cotinine than non-
Hispanic whites who report similar levels of cigarette smoking. OBJECTIVE: To investigate
Vffl-5
-------
III. II . , , ... I I
differences in levels of serum cotinine in black, white, and Mexican American cigarette smokers
in the US adult population. DESIGN: Third National Health and Nutrition Examination Survey,
1988-1991. PARTICIPANTS: A nationally representative sample of persons aged 17 years or
older who participated in the survey. OUTCOME MEASURES: Serum cotinine levels by
reported number of cigarettes smoked per day and by rac|e and ethnicity. RESULTS: A total of
7182 subjects were involved in the"study; 2136 subjects reported smoking at least 1 cigarette in
the last 5 days. Black smokers had cotinine concentrations substantially higher at all levels of
cigarette smoking than did white or Mexican American smokers (p<.001). Serum cotinine levels
for blacks were 125 nmol/L (22 ng/mL) (95% confident? interval [CI], 79-176 nmol/L [14-31
ng/mL]) to 539 nmol/L (95 ng/mL) (95% CI, 289-630 nmol/L [51-111 ng/mL]) higher man for
whites and 136 nmol/L (24 ng/mL) (95% CI, 85-182 nm|i/L [15-32 ng/mL]) to 641 nmol/L (113
ng/mL) (95% CI, 386-897 nmol/L [68-158 ng/mL]) highibr than for Mexican Americans. These
differences do not appear to be attributable to differences in environmental tobacco smoke
exposure or in number of cigarettes smoked. CONCLUSIONS: To our knowledge, this study
provides the first evidence from a national study that senlm cotinine levels are higher among
black smokers than among white or Mexican American sjnokers. If higher cotinine levels among
blacks indicate higher nicotine intake or differential pharjjnacokinetics and possibly serve as a
marker of higher exposure to cigarette carcinogenic components, they may help explain why
blacks find it harder to quit and are more likely to experience higher rates of lung cancer man
white smokers. j
' 1 i : ' ,:: :
I ' I I ' i '' " '
Carter, CL; Jones, DY; Schatzkin, A; Brinton, LA (1989). A prospective study of reproductive,
familial and socioeconomic risk factors for breast cancer using NHANES I data. Pub.Hlth.Rep.
104(1, Jan-Feb), 45-50. [adult; aged; breast neoplasms, epidemiology; breast neoplasms,
etiology; cross-sectional studies; educational status; memifche; menopause; middle age;
NHANES-I; prospective studies; risk-factors; socioeconohiic factors; United States]
!. ' !; ! :'"
Carter-Pokras, O; Pirkle, J; Chavez, G; Gunter, E (1990). j Blood lead levels of J4-11-year-old
Mexican American, Puerto Rican, and Cuban children. Pub. Hlth. Rep. 105(4, M-Aug), 388-
393. [child; child, preschool; HHANES; Hispanic Ameripans; lead, blood] ! ;
i- , . i j
Carter-Pokras, OD; Gergen, PJ (1993). Reported asthma &mong Puerto Rican, Mexican-
American, and Cuban children, 1982 through 1984. Am. ฃ Pub. Hlth. 83(4, Apr), 580-582.
[absenteeism; activities of daily living; age factors; asthm^, epidemiology; HHANES; Hispanic
Americans, statistical and numerical data; Mexican Americans, statistical and numerical data-
NHANES II; DHES]. . ]
:;: : : ] i "":;: :! : :
Casty, F; Arrighi, HM (1994). Asthma and adult height attained - data from NHANES E. /.
Allergy Clin. Immunol 93 (1), 246. [NHANES II]. ; ]
, '"': '. 1 i "i; , ':'
CDC (1997a). Update: blood lead levels-United States, 1991-1994 [published erratum appears
in MMWR Morb Mortal Wkly Rep 1997 Jul 4;46(26):60' ]. MMWR 46, 141-146. [blood lead
levels; NHANES
vm-6
M
-------
Lead is an environmental toxicant that may deleteriously affect the nervous, hematopoietic,
endocrine, renal, and reproductive systems. Lead exposure hi young children is a particular
hazard because children absorb lead more readily than do adults and because the developing
nervous systems of children are more susceptible to the effects of lead. Blood lead levels (BLLs)
at least as low as 10 micrograms/dL can adversely affect the behavior and development of
children CDC's National Health and Nutrition Examination surveys (NHANES), an ongoing
series of national examinations of the health and nutritional status of the civilian
noninstitutionalized population, have.been the primary source for monitoringiBLLs:in the U.b.
, population. From NHANES H (conducted during 1976-1980) to Phase 1 of NHANES HI
(conducted during October 1988-September 1991), the geometric mean (GM) BLL for persons
aged 1-74 years declined from 12.8 micrograms/dL, and the prevalence of elevated BLLs (BLLs
> or = 10 micrograms/dL) decreased from 77.8% to 4,4%. This report updates national BLL
estimated with data from Phase 2 of NHANES m (conducted during October 1991-September
1994) which indicate that BLLs in the U.S. population aged > or = 1 year continued to decrease
and that BLLs among children aged 1-5 years were more likely to be elevated among those who
were poor, non-Hispanic black, living in large metropolitan areas, or living in older housing.
CDC (1997b) Update: Blood lead levels - United States, 1991-1994 (Reprinted from MMWR,
vol 46, pg 141-146,1997). JAMA 211,1031-1032. [blood lead levels; NHANES].
CDC (1997c). Use of unvented residential heating appliances-United States, 1988-1994.
MMWR 46,1221-1224. [heating; NHANES ffl]. .
Many heating appliances rely on combustion of carbon-based fuels and therefore are potential
sources of health-threatening indoor air pollution. Most combustion heating appliances, are
vented to the outside of buildings to facilitate removal of the products of combustion, which
include carbon monoxide (CO), carbon dioxide, nitrogen dioxide, and water vapor. However,
some combustion heating devices maybe unvented (e.g., kerosene- and propane-fueled space
heaters, some gas-fueled log sets, and cooking devices used improperly for heating), and the use
of such unvented devices in closed settings may be associated with nsks for exposure to toxic
eases and other emissions. This report presents an analysis of data from the Third National
Health and Nutrition Examination Survey (NHANES ffl) to estimate the number and regional
distribution of adults using unvented residential heating appliances and stoves or ovens misused
as heating devices in the United States during 1988-1994. The findings indicate that the
percentage of adults using these devices was higher in the South, among low-income groups
among blacks, and among rural residents, and underscore the need for public education about the
health risks associated with exposure to elevated levels of combustion by-products. NHANES ffl
collected data from approximately 20,000 adults about household characteristics, including the
prevalence of various types of residential heating appliances, the use of unvented combustion
space heaters, and use of stoves or ovens specifically for heating during the previous year.
NHANES weights were used to obtain national estimates based on these responses. Because
responses by race/ethnicity other than for whites and blacks were too small for reliable estimates,
responses from all others were combined.
vm-7
-------
8
Centers for Disease Control and Prevention (1993): Preliminary data: exposure of persons aged >
or = 4 years to tobacco smoke-United States, 1988-91. i\fMWR 42(2, Jan 22)? 37-39.
[adolescence; adult; aged; aged, 80 and over; child; child, preschool; cotinine, blood; middle age;
NHANES-ni; population surveillance; tobacco smoke pollution, analysis; DHES].
Centers for Disease Control and Prevention (1994). Bk>ij?d lead levels-United States, 1988-
1991. MMWR 43(30, Aug 5), 545-548. [adolescence; 4ult; aged; child; child, preschool; lead
poisoning, epidemiology; lead, blood; NHANES-ni; population surveillance; DHES].
Chestnut, LG; Schwartz, J; Savitz, DA; Burchfiel, CM (1991). Pulmonary function and ambient
particulate matter: epidemiological evidence from NHANES I. Arch. Env. Hlfy. 46(3, May-Jun),
135-144. [adult; aged; air pollutants, environmental, adverse effects; air pollutants,
environmental, analysis; anthropometry; forced expiratory volume; health status indicators;
health surveys; maximum permissible exposure level; middle age; NHANES-I; predictive value
of tests; regression analysis; respiratory tract diseases, epidemiology; respiratory tract diseases,
physiopathology; socioeconomic factors; United States, epidemiology; vital capacity].
Coate, D; Grossman, M (1988). Carbon monoxide in ttop ambient air and blood pressure
evidence from NHANES H and the SAROAD system. National Bureau of Economic Research,
Cambridge, MA (1050 Massachusetts Avenue, Cambridge, Mass. 62138). pSlHANES H]
Coate, D; Fowles, R (1989). Is there statistical evidence for a blood lead-blood pressure
relationship? J. Health Econ. 8(2), 173-184. [blood pressure; hypertension, epidemiology; lead
blood level; NHANES-H; statistical analysis; United States]. '
': , , 'I ' ' : I
Cohen, BB; Friedman, DJ; Mahan, CM; Lederman, R; Munoz, D (1993). Ethnicity, maternal
risk, and birth-weight among Hispanics in Massachusetts,! 1987-89. Pub. Hlth. Rep 108(3) 363-
371. [HHANES].
Collins, JW;.Shay, DK (1994). Prevalence of low-birth-v
eight among Hispanic infants with
United States-born and foreign-born mothers -the effect of urban poverty. Am. J. Epid. 139
(2), 184-192. [acculturation; birth weight; Black; depressive symptoms; health; ^HHANES 1982-
84; Hispanic Americans; Mexican-American population; mortality; poverty] ' ! .
Cooper, RS; Ford, E (1992). Comparability of risk factor^ for coronary heart disease among
blacks and whites in the NHANES-I Epidemiologic Follow-up Study. Ann. Epid. 2(5), 637-645.
[adult; African-American; body mass; Caucasian; cholesterol, endogenous compound; coronary
risk; female; follow-up; hospitalization; human; incidence]; ischemic heart disease, epidemiology;
life style; major clinical study; male; mortality; NHANES U; NHEFS; race difference; smoking; '
systolic blood pressure; United States] '
Cowie CC; Harris MI; Silverman RE; Johnson EW; Rust JCF (1993): Effect of multiple risk
factors on differences between blacks and whites in the prevalence of non-insulin-dependent
vm-8
-------
diabetes mellitus in the United States. Am. J. Epidemiol. 137(7, Apr 1), 719-732. [BLACKS;
DIABETES MELLITUS, NON-INSULIN-DEPENDENT, ETHNOLOGY; NHANESII].
Crocetti AF;Mushak,P; Schwartz, J (1990). Determination of numbers of lead-exposed women
of childbearing age and pregnant women: an integrated summary of a report to the U.S. Congress
on childhood lead poisoning. Env. Hlth. Persp. 89(Nov), 121-124. [adolescence; adult;
epidemiologic methods; fetus, drag effects; lead poisoning, blood; lead poisoning, complications;
lead poisoning, epidemiology; lead, blood; NHANES-H; pregnancy complications,
epidemiology]. .
DeBaun, MR; Sox, HC Jr (1991). Setting the optimal erythrocyte protoporphyrin screening
decision threshold for lead poisoning: a decision analytic approach. Pediatrics 88(1, Jul), 121-
131. [child; decision support techniques; erythrocytes, chemistry; lead poisoning, diagnosis;
NHANES-n; protoporphyrin, blood].
Dreon DM; John, EM; DiCiccio, Y; Whittemore, AS (1993). Use of NHANES data to assign
nutrient densities to food groups in a multiethnic diet history questionnaire. Nutrition and Cancer
20(3), 223-230. [adult; aged; data interpretation, statistical; diet surveys; food analysis; middle
age; NHANES-n; nutrition assessment; Whites].
Eck LH;Hackett-Renner,C;Klesges,LM(1992). Impact of diabetic status, dietary intake,
physical activity, and smoking status on body mass index in NHANES H. Chn. Nutr. 56(2,
Aug), 329-333. [body mass index; diabetes mellitus, non-insulin-dependent, metabolism; eating,
physiology; NHANES HJ.
Engel, A; Johnson, ML; Haynes, SG (1988). Health effects of sunlight exposure in the United
States Results from the first National Health and Nutrition Examination Survey, 1971-1974.
Arch. Dermatol 124(1, Jan), 72-79. [NHANES I; skrn diseases, epidemiology; sunlight,
adverse effects].
Enstrom JE (1999). Smoking cessation and mortality trends among two United States
populations./ Clin. Epidemiol. 52(9, Sep), 813-825. [mortality;NHEFS; smoking].
The long-term impact of smoking cessation on mortality is assessed among two U.S.
populations: a large cohort of U.S. veterans aged 55-64 at entry and followed from 1954 through
1979 and the NHANES I Epidemiologie Followup Study (NHEFS) cohort of a national sample
of U S adults aged 55-74 at entry and followed from 1971 through 1992. Direct and indirect
survey data indicate that 50-70% of those who were current cigarette smokers at entry had quit
smoking during the 19- to 26- year follow-up periods. The impact of smoking cessation on
mortality among the cigarette smokers as a whole has been assessed by determining the time
trend of the relative risk (RR) of death and 95% confidence interval (CI) for the cigarette
smokers compared with never-smokers over the entire follow-up period in both cohorts. The total
death rates for the 1954/57 U.S. veteran smokers as a whole (63,159 males) have converged only
slightly toward'those of never-smokers, from RR= 1.65 (1.58-1.72) during 1954-1959 to RR =
vm-9
-------
! '
1.61 (1.58-1.63) during 1954-1979. The lung cancer death rates for 1954/57 smokers as a whole
have not converged toward those of never-smokers, with RR = 10.89 (7.70-is'.41) during 1954-
1959 andRR = 11.10 (9.78-12.61) during 1954-1979. The total death rates for the 1971-1975
NHEFS smokers as a whole (694 males and 1116 females) have not converged toward those
of never-smokers. For males, RR = 1.92 (1.46-2.52) during 1971-1982 and RR = 1.96 (1.63-
2.36) during 1971-1992; for females, RR = 1.79 (1.31-2.^6) during 1971-1982 and RR = i.79
(1.47-2.17) during 1971- 1992. .The lung cancer death rates have diverged, based on small
numbers of deaths. For males, RR = 15.76 (2.06-120.61)' during 1971-1982 and RR = 22.20
(5.31-92.92) during 1971-1992; for females, RR = 2.92 (iO.57-15.06) during 1971-1982 and RR =
4.74 (1.94-11.59) during 1971-1992. These trends are contrary to the substantial convergence
predicted by the death rate trends among U.S. veterans wjho were former smokers at the
beginning of follow-up. While these results confirm that |those former smokers who survive for at
least 5 years experience death rates that converge toward those of never-smokers, they also
indicate that a cohort of cigarette smokers that undergoeSj substantial cessation experiences a
death rate that does not converge toward the death rate of never-smokers. The tact that there has
been no convergence for lung cancer is quite surprising, as this is the disease niost strongly
linked to smoking and smoking cessation and less likely to be influenced by other lifestyle
factors. Further investigation is needed for a complete understanding of the impact of smoking
cessation. ' . .' ' ' '
' M '
Escobedo, LG; Remington, PL (1989). Birth cohort analysis of prevalence of cigarette smoking
among Hispanics in the United States. JAMA 261(1, Jan J6), 66-69. [adolescence; adult; aged;
cohort studies; Cuba, ethnology; HHANES; Hispanic ArrLericans, psychology; smoking,
ethnology] .
. ' . i i
Escobedo, LG; Remington, PL; Anda, RF (1989). Long-term age-specific prevalence of cigarette
smoking among Hispanics in the United States. J. Psychdpctive Drugs 21(3, Jul-Sep), 307-318.
[adolescence; adult; age factors; aged; Cuba, ethnology; HHANES; Hispanic Americans,
psychology; smoking, epidemiology] I '
Escobedo, LG; Remington, PL; Anda, RF (1989). Long-l|erm secular trends in initiation of
cigarette smoking among Hispanics in the United States.. JPttb. Hlth. Rep. 104(6, Nov-Dec), 583-
587. [adult; HHANES; Hispanic Americans; smoking, ethnology]
1 , ' ' , :
Espino, DV; Moreno, C; Katerndahl, D; Wood, R (1990). Serum lead and hypertension in older
Mexican-Americans - the HHANES. J. Am. Geriatr. Soc. 38(8), 16. [HHANES; hypertension;
serum lead].
Ezzati-Rice, TM; Murphy, RS (1995). Issues associated with the
sample for human exposure assessment. Env. Hlth. Perspl 103 Suppl 3(Apr), 55-59.
[data collection; environmental exposure; epidemiologic methods;
Hazardous substances; national health programs; NHANES El; Probability;
assessment; DHES].
Vffi-10
design of a national probability
research design; risk
-------
Fanelli-Kuczmarski, MT; Johnson, CL; Elias, L; Najjar, MF (1990) Folate
American, Cuban, and Puerto Rican women. Am. J. Clin. Nutr. 52(2, Aug), 368-372. [adult,
c^hrkogy; Educational status; erythrocytes, analysis; folic acid, blood; HHANES; Hispanic
Americans; DHES].
Farfel M (1987) Evaluation of health and environmental effects of two methods for residential
lead paint removal. SC.D Thesis, The Johns Hopkins University, 315 p. [health sciences, public
health; NHANES-TI; thesis].
Flesal KM (1990). Ratio of actual to predicted weight as an alternative to a power-type weight-
height'mdex(Benn index). Clin.Nutr. 51(4, Apr), 540-547. [adult; age factors; body height;
body mass index; body weight; NHANES H; DHES]
AS (1991). Decreased stature associated with moderate blood lead
^-Amicanchildre,. din. Nutr. 54(3, Sep), 516-519. [agefactors;
body height; HHANES; Hispanic Americans; lead, blood]
Garg R-Madams, JH; Kleinman, JC (1992). Regional variation in ischemic heart disease
incidence./. Clin. Epid. 45(2, Feb), 149-156. [aged; body mass index; coronary disease
epidemiology; coronary disease, mortality; diabetes mellitus, complications; educational status,
follow-up studies; NHANES-I; NHEFS; residence chsiractenstics]
Gartside PS (1988). The relationship of blood lead levels and blood pressure in NHANES H:
additional calculations. Env. Hlth. Persp. 78(Jun), 31-34. [adult; aged; blood pressure, drug
effects; lead, blood; NHANES n].
Gartside, PS; Wang, P; Glueck CJ (1998a). Prospective assessment of cancer
morbidity/mortality risk factors: The NHANES 116 year EP^molo^c^lo^PA^yT:
meeting abstract. J. Investig. Med. 46 (N3, Mar), A212-A212. [cancer; CHD; NHANES I,,
NHEFS].
Gartside, PS; Glueck, CJ (1993). Relationship of dietary .intake to.hospital admi ssion for
coronWheaVtandvasculardiseasettheNHANESHNationalProbabihtyStudy. J. Am. Coll.
Nutr. 12(6, Dec), 676-684. [alcohol drinking; caffeine; cholesterol, blood; coronary disease,
diet; hospitalization; NHANES-H; nutrition; vascular diseases].
Gartside, PS; Glueck, CJ (1994). Inverse association of milk intake with cancer morbidity and
mortality, the prospective NHANES I Epidemiologic Follow-up-study. Chn. Res. 42(3), A38U.
[NHANES I; NHEFS] ..
Gersen PJ- Mullally, DI; Evans, R 3d (1988). National survey of prevalence of asthma among
Se^S'the United States, 1976 to 1980. Pediatrics 81(1, Jan), 1-7. [adolescence; asthma,
diagnosis; astona, epidemiology; asthma, immunology;.child; NHANES-E; skin tests; United
vm-ii
-------
States; DHES]
1.. . ' it
Gergen, PJ; Turkeltaub, PC (1988). The association of percutaneous immediate hyperserisitivity
and respiratory symptoms in the United States population - data from the second National Health
and Nutrition Examination Survey -1976-80 (NHAMES -IT). J. Allergy Clin. Immuhol 81(1)
174. [NHANES H; DHES] " ;. "
Gergen, PJ; Weiss, KB (1990). Changing patterns of aslfoma hospitalization among United
C^/i^^C* rt Mป 1*4 <**A**ซ f\ 1 *7 *********, _ ^ _ . ! f\f~lf\ f^r^ T J t r 1 f*. ji* * . .*-.ซ._ J' '"' ' L . ' '. I'M
States children 0-17 years of age: 1979-87. JAMA 264,
Gergen, PJ; Turkeltaub, PC (1991). The association of
688-1692. [NHANES E; DHES]
:' '. ..'...'..',' "U ', '.; -
llergen skin test reactivity and
respiratory disease among whites in the US population, ibata from the second Rational Health
and Nutrition Examination Survey, 1976 to 1980. Arch. Intern. Med. 151(3, Mar), 487-492.
[adolescence; adult; aged; child; cross-sectional studies; health surveys; middle age; NHANES-
H; nutrition surveys; prevalence; regression analysis; respiratory hypersensitivity, epidemiology;
sampling studies; skin tests; United States, epidemiology; DHES]
i, ,'.' i ,i
Gergen, PJ; Turkeltaub, PC (1992). The association of individual allergen reactivity with
respiratory disease in a national sample; data from the second National Health and Nutrition
Examination Survey, 1976-80 (NHANES H). J. Allergy tlin.Immunol 90(4 Pt 1, Oct), 579-588.
[adolescence; adult; allergens, immunology; health surveys; NHANES-E; respiratory
hypersensitivity, immunology; DHES] "' ' : i
Gergen, PJ; Turkeltaub, PC; Kramer, RA (1992). Age of onset in childhood asthma: data from a
national cohort. Ann. Allergy 68(6, Jun), 507-514. [adolescence; adult; age factors: asthma
etiology; NHANES U; DHES]
Gergen, PJ; Fowler, JA; Maurer, KR; Davis, WW; Overpjeck, MD (1998). The burden of
environmental tobacco smoke exposure on the respiratory health of children 2 months through 5
years of age in the United States: third National Health arid Nutrition Examination Survey, 1988
to 1994. Pediatrics 101 (2, Feb), E8. [asthma; children; ^|HA^S-in; smoidng; tobacco]
OBJECTIVE: To measure the effect of environmental tobacco smoke (ETS) on respiratory health
in a national sample of young children. METHODS: The
through 5 years of age participating in the Third National
study evaluated children 2 months
Health and Nutrition! Examination
Survey, 1988 to 1994. The group was a representative sample of the US population (N = 7680).
A parental report of household smoking or maternal smoking during pregnancy ascertained ETS
exposure. Respiratory outcomes were based on parental rpport of wheezing, cough, upper
respiratory infection, or pneumonia in the last 12 months imd chronic bronchitis orphysician-
diagnosed asthma at any time. Logistic regression was useld to adjust for age, sex, race/ethnicity,
birth weight, day care, family history of allergy, breast-fee'ding, education level of head of
household, and household size. RESULTS: Approximately 38% of children were presently
exposed to ETS in the home, whereas 23.8% were expose!! by maternal smoking during
pregnancy. ETS exposure increased chronic bronchitis and three or more episodes of wheezing
- VJJI-12
-------
among children 2 months to 2 years old and asthma among children 2 months to 5 years old. For
household exposure, a consistent effect was seen only at >/=20 cigarettes smoked per day.
Adjusted odds ratios for increased risk (95% confidence interval) for household exposures
(>/=20 cigarettes smoked per day vs none smoked) and maternal prenatal exposure (prenatal
smoking vs no smoking), respectively, for children 2 months to 2 years old were chronic
bronchitis, 2.5 (1.6,4.1); 2.2, (1.6, 3); three or more episodes of wheezing, 2.7 (1.7,4.2), 2.1 (1.
5, 2.9); and for children 2 months to 5 years old were asthma, 2.1 (1.4, 3.2); 1.8 (1.3,2.6).
Reported use within the past month of prescription medications for asthma (beta-agonists or
inhaled steroids) was not different between those with asthma reporting ETS exposure and those
reporting no exposure; percent of patients with asthma reporting use of medication by household
exposure was 0,25. 7%; 1 to 19 cigarettes smoked per day, 32.9%; and >/=20 cigarettes smoked
per day, 23.1%; percent of patients with asthma reporting use of medication by maternal smoking
during pregnancy was no, 28.9%; yes, 22.7%. Among children 2 months to 2 years of age
exposed to ETS, 40% to 60% of the cases of asthma, chronic bronchitis, and three or more
episodes of wheezing were attributable to ETS exposure. For diagnosed asthma among children 2
months through 5 years old, there were 133 800 to 161 600 excess cases. Among exposed
children 2 months through 2 years of age, there were 61 000 to 79 200 excess cases of chronic
bronchitis and 126 700 to 172 000 excess cases of three or more episodes of wheezing.
CONCLUSIONS: ETS exposure is common among children in the United States. The reported
prevalence of asthma, wheezing, and chronic bronchitis was increased with ETS exposures. No
statistically significant increase in the prevalence of upper respiratory infection, pneumonia, or
cough was associated with ETS exposure. ETS exposure has little effect on the respiratory health
of children between 3 and 5 years of age, with the exception of asthma. ETS appears to increase
the prevalence of asthma rather than the severity as measured by medication use. These findings
reinforce the need to reduce the exposure of young children to ETS.
Geronimus, AT; Hillemeier, MM (19'92). Patterns of blood lead levels in US black^ and white
women of childbearing age. Ethn. Dis. 2(3,Summer), 222-231. [adolescence; adult; age factors;
blacks, statistical and numerical data; environmental exposure; lead poisoning, epidemiology;
NHANES-II; whites, statistical and numerical data]
Gottlieb, DJ; Beiser, AS; O'Connor, GT (1995). Poverty, race, and medication use are correlates
of asthma hospitalization rates - a small-area analysis in Boston. Chest 108 (1), 28-35. [asthma;
changing patterns; childhood asthma; children; epidemiology; exposure; health; particulate air-
pollution; medication; NHANES; poverty; risk-factors; small-area analysis; smoking; United
States].
Goyer, RA (1990). Lead toxicity: from overt to subclinical to subtle health effects. Env. Hlth.
Persp' 86(Jun), 177-181. [behavior, drug effects; lead, adverse effects; NHANES H]
j
Guendelman, S; Abrams, B (1994). Dietary, alcohol, and tobacco intake among Mexican-
American women of childbearing age: Results from HANES data. Am. J. Health Promot. 8(5),
363-372. [adolescent; adult; alcohol consumption; Caucasian; comparison; demography; dietary
Vffl-13
-------
'!' , I
intake; ethnic group; feeding behavior; female; fluid intiake; health behavior; health survey;
HHANES; human; lactation; major clinical study; maternal nutrition; pregnancy; prenatal period;
puerperium; smoking; social status] ;
; ,::'t | l! i " I. 'I
Hahn, RA; Baker, E; Barker, ND; Teutsch, SM; Sosniai, W; Krieger; N (1995). Poverty and
death in the United States - 1973 and 1991. Epid. 6(5), $90-497. [cholesterol blood level; death;
health statistics; NHANES-I; poverty; risk assessment; lisk factor; smoking; statistical analysis;
survival] I ' . I I -
\ ' ". ' \ \
Hankinson, JL; Odencrantz, JR; Fedan, KB (1999). Spitometric reference values from a sample
of the general U.S. population. Am. J. Respir. Crit. Card Med. 159 (1, Jan), 179-87. [Black
Americans; Mexican Americans; NHANES-m; spirome^ry; White Americans]
Spkometric reference values for Caucasians, Anican-Arhericans, and Mexican-Americans 8 to
80 yr of age were developed from 7,429 asymptomatic, lifelong nonsmoking participants in the
third National Health and Nutrition Examination Survey (NHANES IE). Spirometry
examinations followed the 1987 American Thoracic Society recommendations, and the quality of
the data was continuously monitored and maintained. Caucasian subjects had higher mean FVC
and FEV1 values than did Mexican-American and African-American subjects across the entire
age range. However, Caucasian and Mexican-American .subjects had similar FVC and FEV1
values with respect to height, and African-American sub; ects had lower values. These differences
maybe partially due to differences in body build: observed Mexican-Americaras were shorter
than Caucasian subjects of the same age, and African-Americans on average have a smaller
trunk:leg ratio than do Caucasians. Reference values and lower limits of normal were derived
using a piecewise polynomial model with age and heightjas predictors. These reference values
encompass a wide age range for three race/ethnic groups
and research purposes.
Harlan, WR (1988). The relationship of blood lead levels to blood pressure in
population. Env. Hlth. Persp. 78(Jun), 9-13. [adolescence; blood pressure, drug
blood; NHANES
and should prove useful for diagnostic
the U.S.
effects; lead,
Huseman, CA; Varma, MM; Angle, CR (1992). Neuroendocrine effects of toxic
lead levels in children. Pediatrics 90(2 Pt 1, Aug), 186-189. [Insulin-like Growth
analysis; lead, blood; NHANES H]
and low blood
Factor I,
Istvan, JA; Cunningham, TW (1992). Smoking rate, cartoxyhemoglobin, and body mass in the
second National Health and Nutrition Examination Survey (NHANES IT). J. Behav. Med. 15(6,
Dec), 559-572. [body weight; carboxyhemoglobin, analysis; NHANES-JJ; nutrition surveys]
i I
Istvan, JA; Nides, MA; Buist, AS; Greene, P; Voelker, HS(1994). Salivary cqtinine, frequency of
cigarette-smoking, and body-mass index - findings at bastp-line in the lung health study Am. J.
Epid. 139(6), 628-636. [2nd National Health; body weight; cotinine; energy-expenditure;
nutrition examination survey; NHANES-II; nicotine; obesity; population; smokers; smoking;
VJJI-14
L
h ....... I
t
-------
weight; women]
Jenkins, RA; Counts, RW (1999). Personal exposure to environmental tobacco smoke: salivary
cotinine, airborne nicotine, and nonsmoker rnisclassification. /. Expo. Anal. Environ. Epidemiol.
9 (4, Jul-Aug), 352-363. [cotinine; EPA; lung cancer; NHANES-HI; smoking]
A large study was conducted to assess exposure to environmental tobacco smoke (ETS) in a
geographically dispersed study population using personal breathing zone air sampling and
salivary cotinine levels. Approximately 100 self-reported nonsmoking subjects in each of 16
metropolitan areas were recruited for this investigation. Cumulative distributions of salivary
cotinine levels for subjects in smoking and nonsmoking homes and workplaces exhibited a
general trend of decreasing salivary cotinine levels with decreasing time spent in smoking
environments. Median salivary cotinine levels for the four experimental cells in the study
(product of smoking and nonsmoking home and workplaces) were comparable to those reported
for a large national study of serum levels of cotinine (Third National Health and Nutrition
Examination Survey, NHANES ID), when the latter was corrected for expected differences
between serum and saliva concentrations. However, the most highly exposed group in this study
had a median salivary cotinine concentration approximately a factor of 2 greater than that of the
comparable group in the NHANES ffl study. Misclassification rates, both simple (for self-
reported nonsmokers) and complex (self-reported lifetime never smokers), were near the median
of those reported for other studies. Estimated misclassification rates for self-reported lifetime
never-smoking females are sufficiently high (2.95% using a discrimination level of 106 ng/ml)
that, if used in the Environmental Protection Agency (EPA) risk assessment related to ETS and
lung cancer, would place the lower 90% confidence interval (CI) for relative risk at nearly 1.00,
i.e., no statistically significant increased risk. For the 263 most highly exposed subjects in the
study whose self-reported nonsmoking status was accurate, the correlation between airborne
exposure to nicotine and average salivary cotinine is so small, on an individual basis, that it
makes the relationship useless for estimating exposure on a quantitative basis. When subjects are
grouped according to likely categories of nicotine exposure, correlation between group median
airborne nicotine exposure and salivary cotinine level increases dramatically. The comparison
improves for the most highly exposed subjects, suggesting that such quantitative comparisons are
useful for only those subjects who are exposed, to the higher levels of ETS. However, airborne
nicotine exposure for most of the subjects does not account for estimated systemic levels of,
nicotine, based on salivary cotinine levels.
John, EM; Schwartz, GG; Dreon, DM; Koo J (1999): Vitamin D and breast cancer risk: the
NHANES lEpidemiologic follow-up study, 1971-1975 to 1992. National Health and Nutrition
Examination Survey. Cancer Epidemiol Biomarkers 8 (5, May), 399-406. [breast cancer;
NHEFS; Vitamin D]
We analyzed data from the first National Health and Nutrition Examination Survey
Epidemiologic Follow-up Study to test the hypothesis that vitamin D from sunlight exposure,
diet, and supplements reduces the risk of breast cancer. We identified 190 women with incident
breast cancer from a cohort of 5009 white women who completed the dermatological
examination and 24-h dietary recall conducted froml971-1974 and who were followed up to
VJJJ-15
-------
1992. Using Cox proportional hazards regression, we estimated relative risks (RRs) for breast
cancer and 95% confidence intervals, adjusting for age, education, age at menarche, age at
menopause, body mass index, alcohol consumption, and! physical activity. Several measures of
sunlight exposure and dietary vitamin D intake were associated with reduced risk of breast
cancer, with RRs ranging from 0.67-0.85. The associations with vitamin D exposures, however,
varied by region of residence. The risk reductions were liighest for women who lived in United
States regions of high solar radiation, with RRs ranging from 0.35-0.75. No reductions in risk
were found for women who lived in regions of low solari radiation. Although limited by the
relatively small size of the case population, the protectivp effects of vitamin D observed in this
prospective study are consistent for several independent pleasures of vitamin D. These data
support the hypothesis that sunlight and dietary vitamin D reduce the risk of breast cancer.
___ ' ' . ' ' ' i
Jones, CP (1995). Methods for comparing distributions development and application exploring
"race"-associated differences in systolic blood pressure. PH.D. Thesis, The Johns Hopkins
University, 356 p. [biology, biostatistics; health sciences!, public health; NHANES I; NHANES
,
Jones, DY; Schatzkin, A; Green, SB; Block, G; Brinton, LA; Ziegler, RG; Hoover, R; Taylor, PR
(1987). Dietary fat and breast cancer in the National Health and Nutrition Examination Survey I
Epidemiologic Follow-up Study. J. Natl. Cancer Inst. 7< | (3, Sep), 465-471. [adult; age factors;
aged; breast neoplasms, epidemiology; breast neoplasms, etiology; caloric intake; dietary iats,
administration and dosage; dietary fats, adverse effects; follow-up studies; menarche- NHANES
I;NHEFS] i
' , i: " ' i ! " "
Kant, AK; Schatzkin, A; Block, G; Ziegler, RG; Nestle, p (1991). Food group intake patterns
and associated nutrient profiles of the US population. J. Am. Diet. Assoc. 91(12, Dec), 1532-
1537. [adult; aged; anthropometry; ascorbic acid, blood;^lacks; diet; diet records; diet surveys;
eating; food preferences; middle age; NHANES-II; nutriton; United States; whites].
|
Kennedy, A (1987). Serum cholesterol and cancer in the NHANES I Epidemiological Follow up
esterol, blood; neoplasms,
Study [letter]. Lancet 2(8567, Nov 7), 1096. [adult; chol
epidemiology; NHANES I; NHEFS]
Kirnmel, CA; Neumann, DA (1997). Accounting for susceptibility m risk assessment: Current
practice and new directions. Env. Tox. cmdPharm. 4, 189r194. [chemical exposures; NHANES;
risk assessments] i i !
ซ4 I ,,,,,, | i | ,ซ,,
Differences hi susceptibility between individuals can leadj to variability in respdnse to chemical
exposures which in turn modify the risk of illness. As a npans of exploring thei basis for such
differences in susceptibility, a project was undertaken to determine what data were available on
the range of response variability for several health effects; neurotoxicity, .!
reproductive/developmental toxicity, pulmonary toxicity, and cancer, hi addition, modeling
approaches for characterizing response variability were examined and evaluated. The main goal
of this effort was to determine whether human response variability was adequately accounted for
VTJI-16
> is; iii:
I !j||,|lป if.
t , i
I i'l f
-------
in the current risk assessment procedures for human health effects. The conclusions of the
project were that few data are available, both because variability has rarely been the primary
focus of study, and because data are not usually reported in such a way that response variability
can be determined. Several recommendations were made to facilitate better characterization of
interindividual variability,' including the study of variability in available human data (e.g. the
NHANES database) and allowing greater access to raw data from epidemiologic studies. In
addition, the identification of relevant biomarkers, improved understanding of sources of
variability, interaction of chemical effects with other exposures or pre-existing disease, and
retrospective evaluations of risk assessments were recommended. It is hoped that these
recommendations will stimulate research on susceptibility and response variability and encourage
the reporting of data in a way that facilitates analysis of interindividual variability in response.
Klein, BE; Cruickshanks, KJ; Klein, R (1995). Leisure time, sunlight exposure and cataracts.
Doc: Ophthalmol. 88(3-4), 295-305. [adult; aged; cataract, epidemiology; cataract, etiology;
environmental exposure, adverse effects; leisure activities; lens, crystalline, radiation effects;
middle age; NHANES-I; radiation injuries, epidemiology; radiation injuries, etiology; risk-
factors; sunlight, adverse effects; time factors]
Klein, R; Rowland, ML; Harris, MI (1995), Racial/ethnic differences in age:related
maculopathy. Third National Health and Nutrition Examination Survey. Ophthalmology 102(3,
Mar), 371-381.
[adult; aged; aged, 80 and over; caucasoid race; ethnic groups, statistical and numerical data;
macula lutea, pathology; macular degeneration, ethnology; macular degeneration, pathology;
Mexican-Americans; middle age; negroid race; NHANES-III; photography; prevalence; racial
stocks; United States, epidemiology; DHES]
Klesges, RC; Klesges, LM; Meyers, AW (1991). Relationship of smoking status, energy
balance, and body weight: analysis of the second National Health and Nutrition Examination
Survey. J. Comult.Clin. Psychol. 59(6, Dec), 899-905. [body weight, physiology; energy
metabolism, physiology; NHANES H]
Klesges, RC; Eck, LH; Ray, JW (1995). Who under reports dietary intake in a dietary recall?
Evidence from the second National Health and Nutrition Examination Survey, /. Consult. Clin.
Psychol. 63(3, Jun), 438-444. [caloric intake; NHANES-E; nutrition surveys; recall; truth
disclosure]. :
Kowal, NE (1988). Urinary cadmium and beta 2-microglobulin: correlation with nutrition and
smoking.history. J. Tox. Env. Hlth. 25(2), 179-183. [beta 2-microglobulin, urine; cadmium,
urine; NHANES H]
' Kritchevsky, SB (1992). Dietary lipids and the low blood cholesterol-cancer association. Am. J.
Epid. 135(5, Mar 1), 509-520. [age factors; cholesterol, blood; diet surveys; dietary fats,
administration and dosage; follow-up studies; interviews; logistic models; neoplasms,
Vm-17 .
-------
; t unsf
epidemiology; NHANES I; NHEFS]
i ' , j i "
Kurtin, D; Therrell, BL Jr; Patterson, P (1997). Demographic risk factors associated with
elevated lead levels in Texas children covered by Medioaid. Env. Hlth. Persp. 105, 66-68 [lead
levels; NHANES E; NHANES m] ;
This is the first large population-based study of demographic risk factors for elevated lead in
Texas children. It summarizes data on 92,900 children clovered by Medicaid screened for blood
lead during the first 6 months of 1993 in Texas. The highest percentage of elevated lead levels
(14.3%) was in children 25-36 months of age, with slightly lower percentages; in those younger
(13% of 19-24 months) and older (12% of 37-48 months) with blood lead levels greater than 10
micrograms/dl. The group with the highest percentage of elevated blood lead Bevels was 2-4-
year-old Afiican American males (17.3%) making this subgroup 3.5 tunes higher than the group
with the lowest percentage-white girls over age 4 (4.8%). Males had higher blood lead levels for
all ages and ethnic groups. Three principal risk factors vfere found for excessive blood lead in
children: ethnicity, gender, and age; this is consistent with the second National Health and
Nutrition Examination Survey (NHANES n) and Phase I of the NHANES HI results
demonstrating ethnicity and income association with lea^in children in the United States.
f ' ' : ' i j f ' f .'- '
Kutz, FW; Cook, BT; Carter-Pokras, OD; Brody, DJ; Murphy, RS (1992). Selected pesticide
residues and metabolites in urine from a survey of the UiS. general population. J. Tox. Env. Hlth.
37(2, Oct), 277-291. [adolescence; adult; age factors; aged; chlorophenols, urine; NHANES-n,
pentachlorophenol, urine; pesticide residues, urine; DUE S].
Landis, JR; Flegal, KM (1988). A generalized Mantel-Haenszel analysis of the regression of
blood pressure on blood lead using NHANES H data. Env. Hlth. Persp. 78(Jun), 35-41.
[adolescence; blood pressure, drug effects; lead, blood; iJlHANES H; DHES]
' ii
Lee, DJ; Markides, KS (1991). Health behaviors, risk factors, and health indicators associated
with cigarette use in Mexican Americans: results from the Hispanic HANES, km, J. Pub. Hlth.
81(7, Jul), 859-864. [activities of daily living; adult; aged; alcohol drinking, epidemiology;
health behavior; health status indicators; HHANES; Hispanic Americans, psychology; smoking,
ethnology] ; I
; - ' ( I : ; ; : '
Leidy, L (1998): Menarche, menopause, and migration: Inplications for breast cancer research
Am. J. Human Biology 10 (N4), 451-457. [cancer; HHANES; menarche; menopause]
A multi generational delay in the rise of breast cancer incidence rates has been documented
among immigrants to the United States. Prompted by thisj observation, this study examines three
breast cancer risk factors, age at menarche, parity, and agfe at menopause, in relation to each other
and in relation to migration status and language most oftein used in U.S. Hispanic populations.
Mexican American (n = 1,502), Cuban American (n = 53.4), and Puerto Rican(n = 700) women,
aged 30-74 years, were drawn from the Hispanic Health and Nutrition Examination Survey
(HHANES), 1982-1984. Mean recalled age at menarche was significantly later among first
generation compared to second generation immigrants in both Mexican Americans (13.3 ys 12.8
Vm-18
-------
years) and Puerto Ricaris (12.8 vs 11.9 years). Among Mexican. Americans, more children were
reported by first generation immigrants than women of the third or more generations (4.9 vs 4.0
children) and by Spanish speakers compared to women who used English more frequently (4.5 vs
3.3 children). Mean and median .ages at menopause were later among second generation Mexican
American women than first generation women. There was a small, significant, positive
correlation between recalled ages at menarche and menopause within each of the first generation
Hispanic subgroups. The unique positive correlation between ages at menarche and menopause
among first generation immigrants may relate to having spent early years in the country of origin
and later years in the United States. .
Leske, MC; Ghylack, LT; Wu, SY (1991). The lens opacities case-control study - risk-factors for
cataract. Arch. Ophthalmol.lQ9(2), 244-251. [allopurinol therapy; american diet; classification-
system; NHANES-II; nuclear; nutrient sources; population; quantitative data; senile cataract;
sunlight]
Linn, S; Fulwood, R; Carroll, M; Brook, JG; Johnson, C; Kalsbeek, WD; Rifkind, BM (1991).
Serum total cholesterol: HDL cholesterol ratios in US white and black adults by selected
demographic and socioeconomic variables (HANES II). Am. J. Pub. Hlth. 81(8, Aug), 1038-
1043. [adult; aged; alcohol drinking; body mass index; cholesterol, blood; Framingham study;
lipoproteins, hdl cholesterol, blood; negroid race; NHANES H; DHES]
Madans, JH; Reuben, CA; Rothwell, ST; Eberhardt, MS (1995). Differences in morbidity
measures and risk factor identification using multiple data sources: the case of coronary heart
disease [see comments]. Stai. Med. 14(5-7, Mar 15-Apr 15), 643-653. [aged; algorithms;
coronary disease, epidemiology; databases, factual; death certificates; epidemiologic methods;
follow-up studies; health surveys; incidence; middle age; models, statistical; NHANES-I;
regression analysis; risk assessment; risk-factors; sex distribution; United States, epidemiology]
Madigan, MP; Ziegler, RG; Benichou, J; Byrne, C; Hoover, RN (1995). Proportion of breast
cancer cases in the United States explained by well-established risk factors. J. Natl. Cancer Inst.
87(22, Nov 15), 1681-1685. [adult; age factors; aged; breast neoplasms, epidemiology; breast
neoplasms, etiology; breast neoplasms, genetics; income; middle age; NHNAES-I; NHEFS;
parity; pregnancy]
Mannino, DM; Petty, TL (1998). Obstructive lung diseases and low lung function in the United
States population, 1988-94: Results from NHANES IE - meeting abstract. Am. J. Epid. 147
(N11.S, 1 Jun), 53-53. (Ill Market Place, Ste 840, Baltimore, MD 21202-6709) [lung diseases;
NHANES nrj.
Marcus, AH; Schwartz, J (1987). Dose-response curves for erythrocyte protoporphyrin vs blood
lead: effects of iron status. Env. Res. 44(2, Dec), 221-227. [erythrocytes, metabolism; iron,
deficiency; NHANES UJ
Vffl-19
-------
Marks, G; Garcia, M; Sotis, JM (1990). Health risk behaviors of Hispanics in the United States:
Findings fromHHANES, 1982-84. Am. J. Pub. Hlth. SC^Suppl), 20-26. [adult; aged; alcohol;
cigarette smoking; diet; ethnic group; female; health behalvior; HHANES; human; male; mass
screening; normal human; risk] ~>:~ . ! '
!
Matanoski, G; Kanchanaraksa, S; Lantry, D; Chang, Y (1995). Characteristics of nonsmoking
women in NHANESI and NHANES I Epidemiologic Follow-up Study with exposure to spouses
who smoke. Am. J. Epid. 142(2, Jul 15), 149-157. [aduljt; follow-up studies; NHANES I;
NHEFS; nutrition; tobacco smoke pollution],
McWhorter, WP (1988). Allergy and risk of cancer. A prospective study usingNHANES I
follow-up data. Cancer 62(2, Jul 15), 451-455. [adult; aged; follow-up studies; hypersensitivity,
complications; middle age; neoplasms, epidemiology; NHANES-I; NHEFS; prospective studies]
! . , A ! i i , " ,i
McWhorter, WP; Polis, MA; Kaslow, RA (1989). Occur] ence, predictors, and consequences of
adult asthma in NHANES I and Follow-up Survey. Am. $<3V. Respir. Dis. 139(3, Mar), 721-724.
[adult; aged; aged, 80 and over; asthma, epidemiology; NIEANES I; NHEFS]
1 ' ' \
Morabia, A; Sorenson, A; Kumanyika, SK; Abbey, H; Cohen, BH; Chee, E (1989): Vitamin A,
cigarette smoking, and airway obstruction. Am. Rev. Respw. Dis. 140(5, Nov), 1312-1316.
[adult; airway obstruction; NHANES-I; smoking; Vitamiit A, pharmacology]
"' ' ' [ i
Nash, D; Silbergeld, E; Magder, L; Stolley, P (1998). Meaopause, hormone replacement therapy
(HRT), and blood lead levels among adult women from NHANES IE, 1988-1994 - abstract. Am.
J. Epid. 147 (Nl 1,S, 1 Jun), L7-L7. [blood lead levels; NHANES IE].
i i
i |
Neas, LM; Schwartz, J (1998a): Pulmonary function levels as predictors of mortality in a national
sample of US adults. Am. J. Epid. 147 (11,1 Jun), 1011-1018. [mortality; NHANES-I;
pulmonary function]
Single breath pulmonary diffusing capacity for carbon mo:ioxide (DL(CO)) was examined as a
predictor of all-cause mortality among 4,333 subjects who were aged 25-74 years at baseline in
the First National Health and Nutrition Examination Survey (NHANES I) conducted from 1971
to 1975. The relation of the percentage of predicted DL(CO) to all-cause mortality was examined
in a Cox proportional hazard model that included age, sexj race, current smoking status, systolic
blood pressure, serum cholesterol, alcohol consumption, bbdy mass index, percentage of
predicted forced vital capacity (FVC), and the ratio of forced expiratory volume at 1 second
(FEV1) to FVC. Mortality had a linear association with tlie percentage of predicted FVC (rate
ratio (RR) = 1.12,95% confidence interval (Cl) 1.08-1.17J for a 10% decrement) and a
significantly nonlinear association with the percentage of predicted DL(CO) with an adverse
effect that was clearly evident for levels below 85% of those predicted (RR = 1.24, 95% CI 1.12-
1.37 for a 10% decrement). The relative hazard for the percentage of predicted DL(CO) below
85% was not modified by sex, smoking status, or exclusion of subjects with clinical respiratory
disease on the initial examination. This association with the percentage of predicted DL(CO) was
VJJI-20
if, I1 '
f ,
-------
present among 3,005 subjects with FEV1 levels above 90% of those predicted. Thus, pulmonary
diffusing capacity below 85% of predicted levels is a significant predictor of the all-cause
mortality rate within the general US population independent of standard spirometry
measures and even in the absence of apparent clinical respiratory disease.
Neas, LM; Schwartz, J (1998b). Pulmonary function levels as predictors of mortality in a
national sample of US adults. Am J Epidemiol 147 (11,1 Jun), 1011-1118. [alcohol; BMI;
cholesterol; mortality; NHANES-I; pulmonary function; smoking; systolic blood pressure]
Single breath pulmonary diffusing capacity for carbon monoxide (DL(CO)) was examined as a
predictor of all-cause mortality among 4,333 subjects who were aged 25-74 years at baseline in
the First National Health and Nutrition Examination Survey (NHANES I) conducted from 1971
to 1975. The relation of the percentage of predicted DL(CO) to all-cause mortality was examined
in a Cox proportional hazard model that included age, sex, race, current smoking status, systolic
blood pressure, serum cholesterol, alcohol consumption, body mass index, percentage of
predicted forced vital capacity (FVC), and the ratio of forced expiratory volume at 1 second
(FEV1) to FVC. Mortality had a linear association with the percentage of predicted FVC (rate
ratio (RR) = 1.12, 95% confidence interval (CI) 1.08-1.17, for a 10% decrement) and a
significantly nonlinear association with the percentage of predicted DL(CO) with an adverse
effect that was clearly evident for levels below 85% of those predicted (RR = 1.24, 95%' CI 1.12-
1.37 for a 10% decrement). The relative hazard for the percentage of predicted DL(CO) below
85% was not modified by sex, smoking status, or exclusion of subjects with clinical respiratory
disease on the initial examination. This association with the percentage of predicted DL(CO) was
present among 3,005 subjects with FEV1 levels above 90% of those predicted. Thus, pulmonary
diffusing capacity below 85% of predicted levels is a significant predictor of the all-cause
mortality rate within the general US population independent of standard spirometry
measures and even in the absence of apparent clinical respiratory disease.
Needham, LL; Hill, RH Jr; Ashley, DL; Pirkle, JL; Sampson, EJ (1995). The priority toxicant
reference range study: interim report. Env. Hlth. Persp. 103 Suppl 3(Apr), 89-94.
[adult; hazardous substances; middle age; NHANES-DI; reference values].
Nordenberg, D; Yip, R; Binkin, NJ (1990). The effect of cigarette smoking on hemoglobin
. levels and anemia screening. JAMA 264(12, Sep 26), 1556-1559. [anemia, diagnosis;
hemoglobins, analysis; NHANES ITj
Paschal, DC (1995). Blood lead levels in the US population - Phase-1 of the third National
Health and Nutrition Examination Survey. JAMA 274(2), 130.
[blood lead levels; NHANES HI; DHES].
Perez-Stable, EJ; Marin, BV; Marin, G; Brody, DJ; Benowitz, NL (1990). Apparent under
reporting of cigarette consumption among Mexican American smokers [see comments]. Am. J.
Pub. Hlth. 80(9, Sep), 1057-1061. [adult; aged; health surveys; HHANES; Hispanic Americans,
psychology; self disclosure; smoking, psychology; DHES].
vm-2i
-------
Perez-Stable, EJ; Marin, G; Marin, BV; Benowitz, NL (1992). Misclassification of smoking
status by self-reported cigarette consumption. Am. Rev. R!esp. Dis. 145(1, Jan)j 53-57. [adult;
aged; cotinine, blood; HHANES; self disclosure; smoking]
Perez-Stable, EJ; Benowitz, NL; Marin, G (1995). Is serum cotinine a better measure of cigarette
smoking than self-report? Prev. Med. 24(2, Mar), 171-179. [adult; aged; body mass index;
cotinine, blood; data collection, methods; HHANES; self disclosure; smoking, blood]
Perloff, BP; Rizek, RL; Haytowitz, DB; Reid, PR (1990).
Dietary intake methodology, n.
USDA's Nutrient Data Base for Nationwide Dietary Intake Surveys. J. Nutr. 120(S1 l)(N6v)5
1530-1534. [agriculture; databases, factual; diet surveys; [eating; NHANES II; NHANES HT|
Perry, GS; Byefs, T; Yip, R; Margen, S (1992). Iron nutrition does not account for the
hemoglobin differences between blacks and whites. J. Nu$r. 122(7, Jul), 1417-1424. [caucasoid
race; hemoglobins, analysis; iron, blood; NHANES EQ. ' :
1 n- ' ' J"i - h
Pirkle, JL; Brody, DJ; Gunter, EW; Kramer, RA; Paschal, DC; FlegaX KM; Matte, TD (1994):
The decline in blood lead levels in the United States. The National Health and Nutrition
Examination Surveys (NHANES) [see comments]. JAMA 272(4, Jul 27), 284-291.
[adolescence; adult; age factors; aged; child; child, preschool; cross-sectional studies; ethnic
groups; health surveys; infant; lead, blood; middle age; N13ANES-II; racial stocks; sex factors;
DHES].
if r
i \ ' ' i' .
Pkkle, JL; Kaufinann, RB; Brody, DJ; Hickman, T; Gunter, EW; Paschal, DC (1998). Exposure
of the U.S. population to lead, 1991-1994. Environ Health Perspect 106(11):745-50 (11, Nov),
745-750. [children; lead; NHANES m] '' |
Blood lead measurements were obtained on 13,642 persor s aged 1 year and older who
participated in Phase 2 of the Third National Health and Nutrition Examination Survey
(NJHANES HO) from 1991 through 1994. NHANES HI is ii national representative survey of the
civilian, noninstitutionalized U:S. population. The overall [mean blood lead level for the U.S.
population aged 1 year and older was 2.3 microgram/dl, with 2.2% of the population having
levels >=10 microgram/dl, the level of health concern for iphildren. Among U.S.' children aged 1-
5 years, the mean blood lead level was 2.7 microgram/dl, iind 890,000 of these children (4.4%)
had elevated blood lead levels. Sociodemographic factors associated with higher blood lead
levels in children were non-Hispanic black race/ethnicity, low income, and residence in older
housing. The prevalence of elevated blood lead levels wasj2L9% among non-Hispanic black
children living in homes built before 1946 and 16.4% amoiig children .in low-income families
who lived in homes built before 1946. Blood lead levels continue to decline in the U.S.
population, but 890,000 children still have elevated levels. Public health efforts have been
successful in removing lead from population-wide sources such as gasoline and lead-soldered
food and drink cans, but new efforts must address the difficult problem of leaded paint,
especially in older houses, as well as lead in dust and soil. Lead poisoning prevention programs
should target high-risk persons, such as children who live in old homes, children of minority
VTJI-22
|;'
I:
u
-------
groups, and children living in families with low income.
Pletsch, PK (1991). -Prevalence of cigarette smoking in Hispanic women of childbearing age.
Nurs. Res. 40(2, Mar-Apr), 103-106. [adolescence; adult; age factors; HHANES; Hispanic
Americans; smoking, epidemiology, smoking, ethnology]
Pletsch, PK (1994). Environmental tobacco smoke exposure among Hispanic women of
reproductive age. Pub. Hlth. Nurs. 11(4, Aug), 229-235. [adolescence; adult; age factors; child;
environmental exposure; HHANES; Hispanic Americans; population surveillance; tobacco
smoke pollution, statistical and numerical data; women's health].
Pocock, SJ; Shaper, AG; Ashby, D; Delves, HT; Clayton, BE (1988). The relationship between
blood lead, blood pressure, stroke, and heart attacks in middle-aged British men. Env. Hlth.
Persp. 78(Jun), 23-30. [adult; blood pressure, drag effects; cerebrovascular disorders,
epidemiology; cerebrovascular disorders, etiology; Great Britain; health surveys; lead, blood;
myocardial infarction, epidemiology; NHANES IT]
Posner, BM; Cupples, LA; Franz, MM; Gagnon, DR (1993). Diet and heart disease risk factors
in adult American men and women: The Framingham Offspring-Spouse nutrition studies. Int. J.
Epid. 22(6), 1014-1025. [adult; aged; blood pressure; carbohydrate; carbohydrate intake;
cardiovascular disease, epidemiology; cholesterol; cholesterol blood level; diet; dyslipidemia,
epidemiology; eating habit; fat; fat intake; female; food intake; heart disease, epidemiology;
human; hypertension, epidemiology; hypertension, rehabilitation; male; NHANES-H; obesity,
epidemiology, risk factor; saturated fatty acid; sodium; sodium intake; United States; weight
reduction]
Preston, AM (1991). Cigarette smoking-nutritional implications. Progress in food and nutrition
science'. 15(4), 183-217. [disease, etiology; minerals; NHANES-H; nutritional status;
pregnancy; smoking, adverse effects; vitamins].
Roche, AF; Guo, S; Baumgartner, RN; Chumlea, WC; Ryan, AS; Kuczmarski, RJ (1990).
Reference data for weight, stature, and weight/stature in Mexican Americans from the Hispanic
Health and Nutrition Examination Survey (HHANES 1982-1984). Clin. Nutr. 51(5, Suppl,
May), 917S-924S.(Includes 22 references.) [NHANES H; nutrition surveys; health; body weight;
height; height-weight tables; children; adolescents; Mexican-Americans; DHES]
Romero-Gwynn, E; Gwynn, D (1991). Differential patterns of food-consumption among
Mexican born and Mexican-Americans in the California HHANES sample. FASEB J. 5(6), 1665.
[HHANES]
Russell, LB; Carson, JL; Taylor, WC; Milan,, E; Dey A; Jagannathan R (1998). Modeling all-
cause mortality: projections of the impact of smoking cessation based on the NHEFS. NHANES
I Epidemiologic Follow-up Study. Am J Public Health 88 (4, Apr), 630-636. [mortality;
VEI-23
-------
NHANES-I; NHEFS; smoking]
OBJECTIVES: A model that relates clinical risk factors to subsequent mortality was used to
simulate the impact of smoking cessation. METHODS: Sjurvivor functions derived from
multivariate hazard regressions fitted to data from the'firs t National Health and Nutrition
Examination Survey (NHANES I) Epidemiologic Followiip Study, a longitudinal survey of a
representative sample of US adults, were used to project deaths from all causes! RESULTS:
Validation tests showed that the hazard regressions agreed with the risk relationships reported by
others, that projected deaths for baseline risk factors closely matched 'observed mortality, and that
the projections attributed deaths to the appropriate levels of important risk factors. Projections of
the impact of smoking cessation showed that the number of cumulative deaths would be 15%
lower after 5 years and 11% lower after 20 years. CONCLUSIONS: The modej produced
realistic projections of the effects of risk factor modification on subsequent mortality in adults,
Comparison of the projections for smoking cessation with estimates of the risk attributable to
smoking published by the Centers for Disease Control and Prevention suggests that cessation
could capture most of the benefit possible from eliminating smoking.
. ' '..', .. .! J
Sandier, RS; Lyles, CM; McAuliffe, C; Woosley, JT; Kupper LL (1993). Cigarette-smoking,
alcohol, and the risk of colorectal adenomas. Gastroenter{ology 104(5), 1445-1451. [NHAlslES
ITj
Sandier, RS; Lyles, CM; Peipins, LA; McAuliffe, CA; Woosley, JT; Kupper, LL (1993). Diet
and risk of colorectal adenomas - macronutrients, cholesterol, and fiber. J. Natl. Cancer Inst. 85
(11), 884-891 .[adenoma; cholesterol; diet; fiber; NHANES].
Schlussel, YR; Schnall, PL; Zimbler, M; Warren, K; Pickering, TG (1990). The effect of work
environments on blood pressure: evidence from seven New York organizations. J. Hypertens.
8(7, Jul), 679-685. [adult; blood pressure, physiology; cross-sectional studies; hypertension,
epidemiology; middle age; NHANES-II; occupational diseases, epidemiology; occupations;
work]
Schocken, DD; Arrieta, MI; Leaverton, PE; Ross, EA (19S^2). Prevalence and mortality rate of
congestive heart failure in the United States. J. Am. Coll. Cardiol. 20(2, Aug), 301-306. [adult;
age factors; heart failure, congestive, epidemiology; NHAJSfES I].
Schwartz, J (1988). The relationship between blood lead sind blood pressure in the NHANES n
survey. Env. Hlth. Per.sp.78(Jun), 15-22'. [adult; aged; blood pressure, drug effects; lead, blood;
NHANES EJ , ; ;
! i '. '
Schwartz, JD; Katz, SA; Fegley, RW; Tockman, MS (1988). Analysis of spiroriietric data from a
national sample of healthy 6- to 24-year-olds (NHANES EJ). Am. Rev. Respir. Dis. 138(6, Dec),
1405-1414. [adolescence; adult; lung, physiology; NHANES-II; spirometry]
Schwartz, J; Katz, SA; Fegley, RW; Tockman, MS (1988)
VHI-24
Sex and race differences in the
-------
development of lung function [see comments]. Am. Rev. Respir. Dis. 138(6, Dec), 1415-1421.
[adolescence; adult; child; forced expiratory volume; forecasting; lung, physiology; NHANES-II;
racial stocks; sex characteristics]
Schwartz, J (1989). Lung function and chronic exposure to air pollution: a cross-sectional
analysis of NHANES H. Env. Res. 50(2, Dec), 309-321. [adolescence; adult; aged; air pollutants,
adverse effects; air pollution, adverse effects; child; child, preschool; lung diseases,
epidemiology; lung, physiology; NHANESII].
Schwartz ,J; Weiss, ST (1990). Dietary factors and their relation to respiratory symptoms. The
second National Health and Nutrition Examination Survey. Am. J. Epid. 132(1, Jul), 67-76.
[bronchial spasm, etiology; bronchitis, etiology; NHANESITJ.
Schwartz, J (1991). Lead, blood pressure, and cardiovascular disease in men and women. Env.-
Hlth Pers. 91(Feb), 71-75. [adult; aged; environmental pollutants, poisoning; heart enlargement,
chemically induced; hypertension, chemically induced; lead poisoning, complications; NHANES
m .
Schwartz, J; Otto, D (1991). Lead and minor hearing; impairment. Arch. Env. Hlth. 46(5, Sep-
Oct), 300-305. [hearing disorders, blood; Hispanic Americans; NHANES H]
Schwartz, J; Weiss, ST (1991). Host and environmental factors influencing the peripheral blood
leukocyte count. Am. J. Epid. 134(12, Dec 15), 1402-1409. [leukocyte count; leukocytosis,
epidemiology; NHANES H]
Schwartz, J; Weiss, ST (1992). Caffeine intake and asthma symptoms. Ann. Epid. 2(5), 627-
635. [adult; asthma, epidemiology; coffee; human; methylxanthine; NHANES-H; normal
human; tea; United States; wheezing]
Schwartz, J (1993). Particulate air pollution and chronic respiratory disease. Env. Res. 62(1,
Jul), 7-13. [adolescence; adult; aged; air pollutants, toxicity; child; NHANES-I; respiratory tract
diseases, chemically induced]
Schwartz, J; Weiss, ST (1993). Peripheral blood leukocyte count and respiratory symptoms.
'Ann. Epid. 3(1, Jan), 57-63. [bronchitis, blood; cough, blood; leukocytes; NHANES H]
Schwartz, J; Weiss, ST (1993). Prediction of respiratory symptoms by peripheral blood
neutrophils and eosinophils in the First National Nutrition Examination Survey (NHANES I).
Chest 104(4, Oct), 1210-1215. [adult; aged; cross-sectional studies; eosinophils; leukocyte
count; neutrophils; NHANES-I; respiratory tract diseases, epidemiology].
Schwartz, J; Weiss, ST (1994). Cigarette smoking and peripheral blood leukocyte differentials.
Ann. Epidemiol. 4(3, May), 236-242. [adult; aged; eosinophils, immunology; leukocyte count;
VHI-25
-------
NHANES-II; smoking cessation; smoking, immunology]
1 ' ' !
Schwartz, J; Weiss, ST (1994). Relationship between dietary vitamin C intake and pulmonary
function in the first National Health and Nutrition Examination Survey (NHANES I). Am. J.
Clin. Nutr. 59(1, Jan), 110-114. [adult; aged; ascorbic acfd, administration and dosage; health
surveys; lung, drug effects; lung, physiology; NHANES I\
"i :, !
Schwartz, J; Weiss, ST (1994). The relationship of dietary fish intake to level of pulmonary
function in the first National Health and Nutrition Survey (NHANES I). Eur. Respir. J. 7(10,
Oct), 1821-1824. [adult; aged; diet; fishes; forced expiratpry volume; health surveys; middle
age; NHANES-I, regression analysis; spirometry; United States] '
Schwartz, J; Weiss, ST (1995). Relationship of skin test inactivity to decrements in pulmonary
function in children with asthma or frequent wheezing. Atia. J. Respir. Crit. Care. Med. 152(6 Pt
1, Dec), 2176-2180. [allergens; asthma, physiopathology;
respiratory sounds, physiopathology; skin tests].
NHANES-II; respiratory mechanics;
Shahar, E; Folsom, AR; Melnick, SL; Tockman, MS; Comstock, GW; Shimakawa, T; Higgins,
MW; Sorlie, PD; Szklo, M (1994). Does dietary vitamin A protect against airway obstruction?
The Atherosclerosis Risk in Communities (ARIC) Study Investigators. Am. J. Respir. Crit. Care.
Med. 150(4, Oct), 978-982. [airway obstruction, epidemiblogy; airway obstruction, prevention
and control; atherosclerosis, epidemiology; NHANES-I; Vitamin A, administration and dosage]
' ' 'I ' '
Shoff, SM; Newcomb, PA (1998): Diabetes, body size, and risk of endometrial cancer. Am. J.
Epid. 148 (3,1 Aug), 234-240. [cancer; diabetes; NHANES; IT]
Data from a population-based case-control study of Wisconsin women were used to evaluate the
relation of diabetes to the risk of endometrial cancer on tbi basis of body mass index (BMT).
Cases (n=723) were identified from a statewide tumor registry; controls (n=2,291) were selected
randomly from population lists. Diabetes status, weight, height, and other factors were
ascertained by telephone interview. Subjects were categorized as not overweight (BMI, <29.1),
overweight (BMI, 29.1-31.9), or obese (BMI, >31.9) according to the BMI distribution of
middle-aged white women in the second National Health a^id Nutrition Examination Survey.
Joint associations between diabetes status, BMI, and endometrial cancer were evaluated using
unconditional logistic regression models that controlled for age, parity, use of hormone
replacement therapy, education, and smoking. Compared vidth persons without diabetes, those
with diabetes had an adjusted odds ratio of 1.86 (95% conJidence interval (CI) 1.37-2.52) for.
endometrial cancer. This association was modified by BMI (p interaction=0.04). Compared with
non-overweight non-diabetic subjects, non-overweight and. overweight women who reported
diabetes had nonsignificant elevated risks of endometrial cancer (non-overweight, odds ratio
(OR)=1.10, CI 0.66-1.86; overweight, OR=1.58, CI 0.81-3!.05). Iri contrast, elevated risk
estimates were observed for obese diabetic women (OR=2,
95, CI 1.60-5.46). These data
contradict earlier reports and suggest that diabetes confers no additional risk of endometrial
cancer in women who are neither overweight nor obese.
VTH-26
-------
Silbergeld, EK; Schwartz, J; Mahaffey, K (1988). Lead and osteoporosis: mobilization of lead
from bone in postmenopausal women. Env. Res. 47(1, Oct), 79-94. [bone and bones,
metabolism; lead, metabolism; NHANES ifj.
Smith, SA; Campbell, DR; Elmer, PJ; Martini, MC; Slavin, JL; Potter, JD (1995). The
University of Minnesota Cancer Prevention Research Unit vegetable and fruit classification
scheme (United States). Cancer Causes & Control 6(4), 292-302. [anierican diet; biological
markers; beta-carotene; carcinogenesis; constituents; consumption; diet; epidemiologic methods;
fruit; indoles; inhibition; juices; NHANES-II survey; metabolism; neoplasms; nutrition
assessment; United States; vegetables].
Sorel, JE; Heiss, G; Tyroler, HA; Davis, WB; Wing, SB; Ragland, DR (1991). Black-white
differences in blood pressure among participants in NHANES E: the contribution of blood lead.
Epid. 2(5, Sep), 348-352. [blood pressure, physiology; caucasoid race; NEANES ITj.
Steenland, K; Sieber, K; Etzel, RA; Pechacek, T; Maurer, K (1998). Exposure to environmental
tobacco smoke and risk factors for heart disease among never smokers in the third National
Health and Nutrition Examination Survey. Am JEpid. 147 (10,15 May), 932-939. [CHD;
NHANES; smoking]
The relative risk of coronary artery disease among never smokers exposed to environmental
tobacco smoke (ETS) versus never smokers not exposed to ETS is approximately 1.2 based on
more than a dozen epidemiologic studies. Most of these studies have controlled for the major
heart disease risk factors, but residual or uncontrolled confounding remains a possible
explanation for the epidemiologic findings. The authors studied 3,338 never-smoking adults aged
17 years or older, who are representative of all US never smokers, in the 1988-1991 third
National Health and Nutrition Examination Survey (NHANES HQ to determine whether selected
risk factors for heart disease differ between ETS-exposed and nonexposed persons. Both self-
reported ETS exposure (at home and at work) and serum cotinine levels were available, the latter
reflecting recent ETS exposure. After adjustments were made for age, sex, race, and education
among adults aged 17 years or older, no significant differences were found between the ETS
exposed and the nonexposed for any of 13 cardiovascular risk factors, with the exception of
dietary carotene, which was lower among the exposed. On the other hand, significant positive
linear trends were found between serum cotinine and two risk factors (body mass index and
alcohol consumption), and significant inverse trends were found with dietary carotene. There
were also few differences between exposed and nonexposed never smokers among adults aged
40 years or older, who are most at risk of heart disease. In this group, however, there was an
inverse linear trend between serum cotinine and high density lipoprotein cholesterol (p < 0.001).
This finding could result from ETS exposure rather than be an indication of confounding; a
similar inverse trend was found for children, confirming other results in the literature. Overall,
these data suggest little potential for confounding by the heart disease risk factors studied here
when ETS exposure is determined by self-report.
vm-27
-------
II,
Stehr-Green, PA; Wohlleb, JC; Royce, W; Head, SL (198J8). An evaluation of serum pesticide
residue levels and liver function in persons exposed to dairy products contaminated with
heptachlor. JAMA 259(3, Jan 15), 374-377. [adolescence; adult; aged; child; dairy products;
food contamination; heptachlor, adverse effects; heptachlor, analysis; liver, drug effects;
NHANES-II; pesticide residues, analysis].
Stehr-Green, PA (1989). Demographic and seasonal influences on human serum pesticide
residue levels. J. Toxicol. Env. Hlth. 27(4), 405-421. [adolescence; adult; demography;
NHANES-TJ; pesticide residues, analysis] :
1 ' !
Stevens, RG; Jones, DY; Micozzi, MS; Taylor, PR (1988^. Body iron stores and the risk of
cancer [see comments]. N. Engl. J. Med. 319(16, Oct 20), rl047-1052. [ferritin, blood; follow-up
studies; iron, metabolism; neoplasms, etiology; neoplasm!!, metabolism; NHANES-I; nutrition;
risk-factors; serum albumin, analysis; sex factors].
Stevens, RG (1990). Iron and the risk of cancer. Med. On?ol. .
177-181. [iron, adverse effects; liver neoplasms, epidemiology, NHANES I].
Tumor Pharmacother. 7(2-3),
Subar, AF; Block, G; James, LD (1989). Folate intake and food sources in the u
Clin. Nutf. 50(3, Sep), 508-516. [adult; aged; folic acid; ijniddle age; NHANES
nutrition surveys; United States]
US population.
-II; nutrition;
Subar, AF; Harlan, LC; Mattson, ME (1990). Food and nutrient intake differences between
smokers and non-smokers in the United States. Am. J. Pub. Hlth. 80(11, Nov), 1323-1329.
[adult; age factors; aged; blacks; caloric intake; diet; NHANES-H; smoking]
Swanson, CA; Jones, DY; Schatzkin, A; Brinton, LA; Ziejjler, RG (1988). Breast cancer risk
assessed by anthropometry in the NHANES I Epidemiolog|ical Follow-up Study. Cancer Res.
48(18, Sep 15), 5363-5367. [adult; aged; anthropometry; breast neoplasms, epidemiology;
NHANES I; NHEFS] \
1 i. i
Symanski, E; Hertz-Picciotto, I (1995). Blood lead levels m relation to menopause, smoking,
and pregnancy history. Am. J. Epid. 141(11, Jun 1), 1047-31058. [adult; age factors; alcohol
drinking, blood; lead, blood; menopause, blood; Mexican-Kmericans; NHANES-I; pregnancy,
blood; reproductive history; smoking, blood]. : ;
Trout, D; Decker, J; MueUer, C; Bernert, JT; Pirkle, J (199|8). Exposure of casino employees to
environmental tobacco smoke. J. Occup. Env. Med. 40 (3,
smoking]
Environmental and medical evaluations were performed to
Vfar), 270-276. [NHANES JJJ;.
i i
evaluate occupational exposure to
environmental tobacco smoke (ETS) among casino employees. Ah- concentrations of both
nicotine and respirable dust were similar to those published in the literature for other non-
industrial indoor environments. The geometric mean seruir
VTH-28
cotinine level of the 27 participants
. . M it::
;"! a
, I- ' I J H I
;, i IE"
j |:
i ! It
i
,
1! IS I
':! I ii
-------
who provided serum samples was 1.34 nanograms per milliliter (ng/mL) (pre-shift) and 1.85
ng/mL (post-shift). Both measurements greatly exceeded the geometric mean value of 0.65
ng/mL for participants in the Third National Health and Nutrition Examination Survey
. (NHANES ffl) who reported exposure to ETS at work. This evaluation demonstrates that a
sample of employees working in a casino gaming area were exposed to ETS at levels greater than
those observed in a representative sample of the US population, and that the serum and urine
cotinine of these employees increased during the work shift.
Turkeltaub, PC; Gergen, PJ (1988). The prevalence of allergic and non-allergic respiratory
symptoms in the United States population- data from the secondNationaLHeaMi and.Nutaton
Examination Survey, 1976-80 (NHANES-E). J. Allergy Clm. Immunol 81(1), 305. [NHANES
HDHES]
Turkeltaub, PC; Gergen, PJ (1991). Prevalence of upper and lower respiratory conditions in the
US population by social and environmental factors: data from the second National Health and
Nutrition Examination Survey, 1976 to 1980 (NHANES TJ). Ann.Allergy 67(2 Pt 1, Aug), 147-
154. [age factors; NHANES-H; respiratory tract diseases, epidemiology; DHESJ.
Wagstaff, DJ (1993). Assessment of human exposure to toxic substances in food. Tax. Subst. J.
4(3), 184-198. [dietary intake; NHANES I]
Wallace LA (1997). Human exposure and body burden for chloroform and other
trihalomethanes. Crit. Rev. in Env. Sci. and Tech. 27,113-194. [chloroform; NHANES;
SstmTmtaation on human exposure to chloroform and other trihalomethanes (THMs) in air,
Water, and food is summarized. Three major surveys have collected data on chloroform levels in
finished water at treatment plants. EPA's TEAM Studies have measured concentrations of THMs
in residential drinking water and in personal, indoor, outdoor, and expired air from about 800
participants in eight cities. The Food and Drug Admmistration has surveyed chloroform levels in
food and beverages. Recently, the Centers for Disease Control (CDC) have completed measuring
blood levels of THMs in about 1000 participants in the National Health and Nutation
Examination Survey (NHANES). Exposure occurs through ingestion (drinking tap water and
soft drinks and eating certain dairy foods), inhalation (breathing peak amounts of chloroform
emitted during showers or baths, and lower levels in indoor air from other indoor sources), and
dermal absorption (during showers, baths, and swimming). Each of these routes of exposure
appear to be potentially substantial contributors to total exposure. The major source of exposure
to chloroform is chlorination of water supplies. This results in exposure through ingestion of
drinking water, but also through inhalation and skin absorption as a result of the myriad other
uses of chlorinated water in the home: showers, bams, washing clothes and dishes, etc. Because
chlorinated water supplies are used by bottling plants of soft drink manufacturers, even the
chloroform found in beverages may be partially due to the chlorination of water supplies. Other
sources of exposure, which can be important for specific groups of people, include chlorination
of swimming pools, industrial production and use, and use of bleach during clothes washing.
VIH-29
-------
Wax, Y (1992). Collinearity diagnosis for a relative risk regression analysis: an application to
assessment of diet-cancer relationship in epidemiological studies. Stat. Med. 11 (10, Jul), 1273-
1287. [breast neoplasms, etiology; diet; NHANES-I; NHEFS; regression analysis]
Whittemore, AS; DiCiccio, Y; Provenzano, G (1991). Urinary cadmium and blood pressure:
results from the NHANES E survey. Env. Hlfh Persp. 91(Feb), 133-140. [cadmium, urine;
environmental pollutants, urine; NHANES H]
Whittemore, AS; Perlin, SA; DiCiccio, Y (1995). Chroni
lifelong nonsmokefs: results from NHANES. Am. J. Pub.
|| r
obstructive pulmonary disease in
Hlth. 85(5, May), 702-706. [aged;
lung diseases, obstructive, epidemiology; NHANES I; NHANES H; NHANEiS HI; smoking].
Wolff, CB; Portis, M; Wolff, H (1993). Birth weight and smoking practices during pregnancy
among Mexican-American women. Hlth. Care Women Jwl 14(3, May-Jun), 271-279.
[birth weight; HHANES; Mexican-Americans; pregnancy,!ethnology]
Wood, PR; Hidalgo, HA; Prihoda, TJ; Kromer, ME (1993). Hispanic children with asthma -
morbidity. Pediatrics 91(1), 62-69. [HHANES]. !
'' i . ' i "
1 ' ! B;
Yong, LC; Brown, CC; Schatzkin, A; Dresser, CM; Slesinski, MJ; Cox, CS; Taylor, PR (1997).
Intake of vitamins E, C, and A and risk of lung cancer - The NHANES I Epidemiologic
Followup Study. Am. J. Epid. 146,231-243. [lung cancer;!NHANES I; NHEFS; vitamins]
The relation between the dietary intake of vitamins E, C, and A (estimated by a 24-hour recall)
and lung cancer incidence was examined in the first National Health ajid Nutrition Examination
Survey Epidemiologic Followup Study cohort of 3,968 men and 6,100 women!, aged 25-74 years.
During a median follow-up period of 19 years (from 1971-11975 to 1992), 248 persons developed
lung cancer. Adjusted for potential confounders using Cox proportional hazards regression
methods with age as the underlying time variable, the relative risk of lung cancer for subjects in
the highest quartile of vitamin C intake compared with those in the lowest quartile was 0.66
(95% confidence interval (CI) 0.45-0.96). For vitamin A m'take, a protective effect was observed
only for its fruit and vegetable component (carotenoids) among current smokers (relative risk =
0.49, 95% CI 0.29-0.84), but this was modified by the intensity of smoking (a statistically
significant effect (relative risk = 0.33, 95% CI 0.13-0.84) was observed only for those in the
lowest tertile of pack-years of smoking). The vitamin E intake-lung cancer relation was modified
by the intensity of smoking with a significant protective effect confined to current smokers in the
lowest tertile of pack-years of smoking (relative risk = 0.3ej, 95% CI 0.16-0.83). Overall, there
was no additional protective effect of supplements of vitanjins E, C, and A beyond that provided
through dietary intake. When vitamin E, vitamin C, and carotenoid intakes were examined in
combination, a strong protective effect was observed for those in the highest compared with
those in the lowest quartile of all three intakes (relative risi[ = 0.32, 95% CI 0.14-0.74). These
data provide support for a protective role of dietary vitamMs E and C and of carotenoids against
lung cancer risk but with a modification in effects by the intensity of cigarette exposure. While
smoking avoidance is the most important behavior to reduce lung cancer risk, the daily
VHI-30
i
-------
consumption of a variety of fruits and vegetables that provides a combination of these nutrients
and other potential protective factors may offer the best dietary protection against lung cancer.
Ziegler, RG; Ursin, G; Craft, NE; Subar, AF; Graubard, BI; Patterson, BH (1993). Does beta-
carotene explain why reduced cancer risk is associated with vegetable and fruit intake?
Pennington Cen. Nutr. Ser. 3, 352-371.(Literature review.) [beta-carotene; blood; carcinoma;
carotenoids; chromatography; diet; disease prevention; food groups; food intake; fruit; literature
reviews; NHANES-I; nutrient intake; nutrition research; risk; vegetables]
vm-31
-------
I
i
I
i!
if:
'i1"1',!',!
Mi-
K\
isi
m
f" l1
-iw
i t ,i.|.
11: 1* i
t ': i
t "
i . . :t
;. : i 'i;H 11
'! I i: si'1!
-------
APPENDIX IX
Summaries of Recent EPA Studies Using NHANES Data
IX-1
-------
Use of NHANES Blood and Urine Biochemical Data in Evalusiting Exposure for Risk Assessments and Response
Actions at Superfund Sites i . , '
Name: Elmer Akin
Phone Number: 404-562-8634
iif* ***** '? ซ
Office Affiliation: Offices of Technical Services,
Waste Management Div. (WMD), USEP A Region 4, Atlanta, GA
1) Which NHANES did you use? DNHANES-I DNHANES-H D NHANES-m DHHANES
HNHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
All the NHANES99+ subjects, plus any breakouts of the population data that can be developed just for the southeastern part
of the U.S.. Subjects are classified by age, race, gender, etc.
3) Detailed description of the goals/purpose and approach of the study:
Blood and urine levels of environmental chemicals (e.g., pesticides, organoplorines, metals) in U.S. random population
samples from NHANES are used as screening levels for comparison with data from local populations exposed to hazardous
waste sites (Superfund sites). Blood and urine levels of these chemicals in local populations are compared to the mean levels
derived framNHANES99-K In the future, as data permit, this analysis willjbe refined by stratifying both the local populations
and the NHANES subjects by age, race, sex, etc. and then comparing the blood and urine levels by these categories.
! ' ' ;
4) What statistical methods and models are you using? Include a brief isummary of how you present the results. If
appropriate, attach sample tables and/or charts.
We calculate mean values and distribution statistics by age, race, gender, et:. for the blood and urine levels of the
environmental chemicals evaluated in NHANES99+.
5) If you use NHANES data without publishing documents, include a brief description of these cases, (e.g., use of raw
NHANES data and/or results from published NHANES studies to supp t>rt setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
The NHANES values are used as an input parameter in determining potential risk to local populations around hazardous
waste sites and Superfund sites and in determining the need for regulatory action. In the future, these comparisons of
environmental chemicals in blood and urine of local population and NHANJES subjects will be used to determine the urgency
and priority of response actions.
lap*'- ,iJ":j-,..iJiiซ.iTSMij^K';)w;!iil;jBpt lakTji.^'^i-r.rrl
{Citation and Abstract/Executive Summary .-'.
JE^ ^_j ^Sas " '' < "'"' ' '"*" ' ' ' '
Citation:
>ฃ jjupjipietirf Studies: * ^ _"?",, '^ ^_ l^ \;
Starting Date:
Ending Date:
ISEEi.ฃ
_.^b^.....^^.^k^'" ซ--^
;::i:\:3i;lias!!Viia;l!ll:::ilil!g^^^^^^ I'ICT if'R !i:ii'flP'1 i111: F'i'.lii!B!',r;r.;:v.* ;* . ".>:, " -^i^^
Anticipated Product(s):
Tentative Completion Date(s):
IX-2
I
1:1-
!::l
-------
**
NHANES-m/HHANES Pesticide Epidemiology Study
ftafc
'A Contact Person
Name: Ruth Allen Office Affiliation: Office of Pesticide Programs (OPP), Health Effects Div (HED), Office of
Prevention, Pesticides and Toxic Substances (OPPTS), Washington, DC
Phone Number: 703-305-7191
~
|f|iMption of Study
-TT
1) Which NHANES did you use? DNHANES-I DNHANES-II XDNHANES-in XD HHANES
13 NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
NHANES subjects are classified by the following variables: socio-dernographic (e.g., age, education, gender, race/ethnicity,
poverty-income ratio, place of residence), personal life style (e.g., tobacco & alcohol use, # rooms in'home, # people in home,
year house was built), medical/reproduction (e.g., general health status, access to health care, blood group parameters),
environmental (e.g., source of drinking water), occupational (e.g., usual & current occupation & business or industry), diet
(e.g., foods eaten), and regional variation (e.g., state, county, census region, urban/urban fringe or rural residence). Analysis
of pesticides groups focused only on adults 20-59 years old,.& included 978 subjects fromNHANES-m & 2008 Mexican-
Americans from HHANES who had urinary pesticide measurements.
3) Detailed description of the goals/purpose and approach of the study: Determine prevalence of pesticide exposure
biomarkers among subgroups of NHANES-HI and HHANES, as noted above, & determine if there are statistically significant
risks of exposure associated with these subgroup characteristics. OPP uses NHANES data to: a) validate aggregate and
cumulative risk models; b) characterize exposures to single & multiple pesticide parent compounds, or their metabolites, in
various human subgroups at various points in time; c) improve understanding of how exposure to pesticide metabolites affects
human health; d) examine probable routes of pesticide exposures, including through residues in food, drinking water & in
occupational & residential settings, for both aggregate and cumulative exposures. These analyses will provide OPP with
valid & reliable information on environmental exposures and body burdens for the set of pesticides assessed in NHANES-in,
HHANES, and NHANES 99+ so that risk managers and assessors can make decisions based on actual human data.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts.
Simple frequency tables, univariate analysis, odds ratio, t-tests and maximum likelihood methods.
5) If you use NHANES data without publishing documents, include a brief description of these cases. (e.g., use of raw
NBANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
Based on urinary pesticide levels hi NHANES-IE, HHANES, and eventually NHANES 99+ subjects, OPP will estimate the
underlying distribution of possible pesticide exposures in the U.S. population. This underlying distribution will be compared
to EPA's oral Reference Dose (RfD) values for specific pesticides to evaluate the adequacy of protection under current
regulations. In instances where pesticide metabolites are elevated above the minium detectable level, OPP will use the
NHANES subpopulation characteristics to help identify segments of the U.S. population potentially at high risk of exposure
to these pesticides.
Results support OPP on-going efforts to use more real human data in models for aggregate and cumulative risk assessments.
Conclusions drawn from NHANES-ffl, HHANES, and eventually NHANES99+ pesticide data can be used to cross-check and
validate results from other NHANES surveys (i.e., NHANES 03+) and from published biomonitoring data from other large
population-based environmental epidemiology and environmental health studies.
^Citation and Abstracf^xeStive SunMary; ' ,
JX-3
-------
Citation:
1) Mage, D., Allen, R., Gondy, G., Smith, W., Barr, D., Needham, L., (20(il). Estimating Pesticide Exposures of NHANES-
III participants. Presentation given at International Society for Exposure Assessment (ISEA) Conference in South Carolina,
2001. ,
2) Allen, R., Werner, E., Gondy, G., Mage, D., (2000). The NHANES-IWJHHANES Pesticide Epidemiology (PEPI) Study:
Examination of High End Exposure. Presentation given at American Public Health Association (APHA) Conference in
Boston 2000. i
3) Allen, R., Werner, E.,(l 999). The NHANES IH/HHANES Pesticide Epi demiology Study. Presentation given at American
Public Health Association (APHA) Conference in Chicago 1999.
!Sfcf fort;omp!etetl'itu3!es:" *" ""*'';""" ''^'m v";-,-/"*;-f^"^-^r-"; _
i.gg!!;.l,:f!!li:iiP!i.jinii'!iii /I^NAiiiiiUNinr^s.fK ^i;ilhlS'li^llJliS"!liir''iliilll^lM:^::LJL.^ v,a,":^r:.,-..wj.^s;%:-;,,
Ending Date:
Anticipated Product(s): Reports on NHANES-ffl and HHANES chemical s-specific statistics for organophosphates,
carbamates, phenoxyacetic.acids, organochlorines and fungicides. Reports Ion comparison of chemical groups and specific
chemicals of interest in'NHANES data, plus comparisons with data in the open literature, population-based analyses,
unresolved statistical and methodological issues for population- based risk ibssessment and recommendations for future
NHANES analyses. i
Tentative Completion Date(s): Chemical specific reports for 2,-4 D, lindane and methyl parathion, completed in 2001.
Reports for chlorpyrifos, propoxur, carbofuran, carbaryl planned for compltjtion in 2002. In 2002 the NHANES analysis
team will complete papers on maximum likelihood estimation methods and Jon reclassification of NHANES occupation codes.
Abstracts:
1) Mage, D., Allen, R., Gondy, G., Smith, W., Barr, D., Needham, L.., (2001). Estimating Pesticide Exposures of NHANES-m
participants. Presentation given at International Society for Exposure Assessment (ISEA) Conference in South Carolina, 2001.
The Third National Health and Nutrition Examination Survey (NBt^NES-BCQ collected a single urine sample for
pesticide analysis from almost 1000 subjects during the period 1988-1994. "Ijhe subjects were male and female volunteers from a
national probability sample, with ages 20 to 59. We assume that the choice to donate or not donate urine was independent of the
subject's pesticide exposure variables, so that the urine samples are equivalent to a national probability sample. The urine was
analyzed for 12 pesticide residues that correspond to more than 30 different pjossible parent compounds. These data, reported as
mg residue/liter (mg/L) are only indicative of an exposure if the values are ab&ve the minimum detectable; level (MDL) for each
residue. Because volumetric urine production depends upon the variable .quantity of fluids people ingest prior to the sample
collection, it is necessary to normalize the urinary pesticide excretion rates by dividing by the urinary excretion rate of creatinine
(gCr/L) to provide a value of mg/gCr.
Each individual excretes creatinine at then: own constant daily rateja function of their gender, age, height and weight,
muscularity, and renal-related health status (gCr/day). From the subject questionnaire data we estimate the individual's daily
creatinine excretion rate and multiplying it by mg/gCr we obtain an estimate of the daily excretion rate of the pesticide residue.
Using stoichiometry and moiety data, we analyze and report the constant daily intake of the most likely parent pesticide of the
residue that would lead to a constant daily urinary excretion rate of residue equal to the measured values. We divide by the
subject's recorded body weight to obtain the equivalent dose rate in mg/kg/daly for comparison to the reference dose (RฃD)
established by EPA for each pesticide. s .
We fit the distribution of estimated dose rates by a Johnson SB (4-parameter lognormal) model. A major methodological
difficulty in fitting these data is the large number of below MDL values (BMDL). Each BMDL (mg/L) corresponds to a value
less than a variable amount of residue per gram creatinine. We fit the SB model using the exact and MDL values by the method
of maximum likelihood estimation (MLE) to predict the fraction of the population of the U.S. that is expected to be exposed to
the pesticide at or above the RfD. People are exposed to pesticides in the air,jwater and diet by ingestion and dermal contact.
However, when pesticides degrade in the environment fromrnicrobial activity or sunlight, these residues remain in the
environment and can also be inhaled or ingested in water and food. Given thtsse considerations, the model results show that less
than 1 person in 1000 is expected.to be exposed to the parent pesticide at or above the RfD. We recommend that the ongoing
NHANES surveys continue to report, at least, the same 12 pesticide residues ,so that changes in estimated exposures over time
can be obtained to evaluate the effect of regulation. We also recommend thai! a 24-hour total urine collection be instituted, to
replace the estimates of daily creatinine and pesticide-residue excretion with measured values. '
LX-4
-------
The overall significance of this approach is that we make use of the complete data set, including BMDL values, to
estimate the underlying distribution of possible pesticide exposures in the U.S. We compare these exposures to the RfD to
evaluate the adequacy of protection. . .
2) Allen, R., Werner, E., Gondy, G., Mage, D., (2000). The NHANES-in/ HHANES Pesticide Epidemiology (PEPI) Study:
Examination of High End Exposure. Presentation given at American Public Health Association (APHA) Conference in Boston
2000.
The MHANES Hi/Hispanic HANES Pesticide Epidemiology (PEPI) Study is designed to analyze the prevalence of
pesticide biomarkers among samples of the survey populations from the National Health and Nutrition Examination Survey m
(NHANES) and the Hispanic Health and Nutrition Examination Survey (HHANES). The distribution percentiles of urinary
metabolites of more than 20 pesticides have been determined. These distributions enabled us to identify subjects with high serum
and urinary levels of pesticide analytes who are at potential risk for pesticide-related illnesses. This analysis focuses a detailed
examination of individuals at the high end of exposure both at the 95th and 99th percentiles. Among the 1,018 people in the
NHANES HE pesticide sample, six had high values for more than one analyte: two females, one aged 40-49 who is a non-
Hispanic white, and one aged 20-29 who is .Mexican-American. The four males include two non-Hispanic blacks between 40-49
years of age and two non-Hispanic whites between 30-39 years. The results are useful for efforts underway to use more real data
in support of models for aggregate and cumulative risk assessments. The conclusions also contribute to comparisons across
NHANES surveys and with published biomonitoring data from other large population-based environmental epidemiology and
environmental health studies.
3) Allen, R., Werner, E.,(1999). The NHANES m/HHANES Pesticide Epidemiology Study. Presentation.given at American
Public Health Association (APHA) Conference in Chicago 1999.
The NHANES-in/Hispanic HANES Pesticide Epidemiology (PEPI) Study is designed to analyze the prevalence of pesticide
biomarkers among samples of the survey populations from the National Health and Nutrition Examination Survey III (NHANES)
and the Hispanic Health and Nutrition Examination Survey (HHANES). The distribution percentiles of urinary metabolites of
more than 20 pesticides will be determined for the NHANES ffi and HHANES sample populations, and will be analyzed in
relation to the sample population' sociodemographic, geographic, occupational, and other personal characteristics. The
pesticides examined include carbamates; phenoxyacetic acid herbicides; fungicides; and organochlorines. These compounds
have been prioritized to meet EPA's regulatory obligations hi the implementation of its Office of Pesticide Programs. Elevated
detectable levels of metabolites associated with sample populations' characteristics will enable the EPA to identify segments of
the U.S. population at high risk exposure to pesticides.
rx-5
-------
I \
I
| ijjijJIIjiJJIL!!1 il1 ;; I,, ,]j|'
| fyijMtei Estimates of body weight for water program exposure and risk assessment
Diปiipiiiai|!i.miiijjM. pซ . _i mil.!. ซ .'.ซ i.iji ,s ! i ..... ..i.iihi(ii.i|ป,..~ ...y. '"^slfe..^';,;.;!; ,-, S'^-fi,,' . x,;. :ป,.; ;;:. Ss./\/'^i'4;i'iiiiSS^>ซrfปr^sl%!
4ipi^i i."2 .W*ซ4*""ซ i .*ซ mm , ซ...i-,..,i i,.ซ..u.t. ,,,.i,,,,,,.,.-..l ซ.-.vi*;,.frh,,..i.i.,ซi.,,a6ซ'^;y*ds.'-.*ซ'.*i;!|*i *-ซR;to;ii. 'vffj-Sii'Wftts.-A'w.1**^!"--*-
Name: Denis Borum and Helen Jacobs Office Affiliation: Office of Science & Technology (OST), Office of
Water (OW), vlfashington, DC :
Phone Number: 202-260-8996 (Denis); 202-260-5412 (Helen)
j^escriptipn of Study , ..^ ,. ;';/,';;_ ( ...; ;;;/,,; , !Li ,~; l"j ;>;.';. ... -. . j
1) Which NHANES did you use? DNHANES-I DNHANES-H H
D NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, re.
Groups classified by age and gender
3) Detailed description of the goals/purpose and approach of the study:
OW revised its "Methodology for Deriving Ambient Water Quality Criteria
many scientific advances since its original publication in 1980. The NHAb
body weight recommendations for various age and gender categories based
4) What statistical methods and models are you using? Include a brief !
appropriate, attach sample tables and/or charts.
Tabulations of national estimates of body weight in kilograms by age and g
The statistical sampling weights used in NHANES-m for the national estin
These weights apply to all persons examined in the Mobile Examination Cซ
5) If you use NHANES data without publishing documents, include a bi
NHANES data and/or results from published NHANES studies to suppi
support setting air quality standards for ambient pollutants.):
Incorporated into guidance document and will be used to develop Water Qi
ง1 i .^ L ~^~ il .' *3iakii*St*StA'$A;Jl aft. l;::a:*'!V.:S;v: iSซf!:SJ[
;ซtion and Abstract/Executive Summary . ; : I
iisi&Siwsi^
lfe=m;j;teป^^
NHANES-ra DHHANES
gion of country):
for the Protection of Human Health" reflecting
IBS analysis was conducted in order to provide
on the most recent available data.
ummary of how you present the results. If
snder are provided for selected age categories.
ates are the "examined' sample final weights".
nters or at home.
ief description of these cases, (e.g., use of raw
>rt setting pesticide tolerances on food, or to
laliry Criteria.
i^ilii^fflifc ;
Citation: USEPA. 2000. Methodology for Deriving Ambient Water Quality Criteria for the Protection of Human Health
(2000). Office of Science and Technology, Office of Water. EPA-822-B-00-004. October
t '- - ' ' ^ ' * - ' ' ป *"Bv;ซ--i ' ;, ;,.^.isr~ซ4v*^
J^!|egF]pr Completed Studies: i;t, ^ _ ,i;_ ,.,_,:, ;..;,'.,,,;:,,,,,; ...-...,;,,.-,-;.J
Starting Date: Ending Date: October 2002
- JiซiSS iiiiiiii iiซii,i,!,i 'i -s ".-S -iii.. iS i-.i.fe :; . '.:.: \vS> s/Sl"; ' i-t S-'i ,ISปi liis'iSSSiK" ~--tl ^SSf'iS^M
Iilformation for. Work ,in Progress: , _v , , ., ,.,.; .;,.; .....-:-.,' x..;,;^
Anticipated Product(s): A Technical Support Document on Exposure Assi
Methodology, including finer age categories for body weight.
Tentative Completion Date(s): September 2002.
DC-6
^^g^'^S^^^t^(^;^S"l^y,,^^t^:
liilKafe&^iiwCiEi^
sssment to provide greater details on the
; i
i
;j ',,..'
': ', |
f
, ' ' i - '
i
!
i , ,
If ' . '
^ 11
! ; 4
|, .-(
.
i IF "I,1:
-------
Analysis of Serological Responses to Cryptosporidium Antigen Among NHANES m Participants
Name: Rebecca Calderon
Phone Number: 919-966-0617
Office Affiliation: Human Studies Division (HSD), National Health &
Environmental Effects Research Laboratory (NHEERL), Research Triangle Park, NC
description of. Study
1) Which NHANES did you use? DNHANES-I DNHANES-]! HNHANES-IH DHHANES
D NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
Surplus sera from subjects within seven of the NHANES-IH primary sampling units (PSUs) were used. These seven
geographic areas were selected because they differed in their sources and treatment of drinking water. First, the PSUs were
selected, then sera were selected randomly from subjects living in each PSU.
3) Detailed description of the goals/purpose and approach of the'study: The objective of the study was to see if intensity
of serum antibody response to Cryptosporidium was related to a PSU and therefore the primary source and treatment of
drinking water. Each PSU was composed of one or more adjacent counties. The PSUs were chosen based on their source
water characteristics.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts. Multivariate analysis of the observed intensity of serological response to
the Cryptosporidium antigens was conducted using a Tobit model. Because of the multistage design of the NHANES survey,
the analyses were stratified first by the PSU and then by family within the PSU. Data are presented in tabular form.
5) If you use NHANES data without publishing documents, include a brief description of these cases, (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to .
support setting air quality standards for ambient pollutants.):
-^
vQtatton and Abstract/Executive Summary
* -r
Citation: Analysis of Serological Responses to Cryptosporidium Antigen Among NHANES El Participants
_ -
fDates, for Completed Studies:
Starting Date: October 1993
Ending Date: January 31,2002
* Jt* -i --1, T** "-*"
J^formatton for Work in Progress:
Anticipated Product(s): Journal article
Tentative Completion Date(s): Draft journal article being revised based on reviews for clearance.
Abstract
This study related the intensity of serological responses to two Cryptosporidium antigen groups with the city of residence, family,
age, sex and other characteristics of study participants for 1356 NHANES HI participants from seven of the Survey's primary
sampling units (PSUs). The PSUs differed in their primary source and treatment of drinking water. The mean intensity of
serological responses differed by PSU but not by family unit within the PSU. Increasing age, female sex, Black race, serological
response to Toxoplasma and larger family size were associated with a more intense response to both antigen markers (p<0.05).
Other race, residing in a major metropolitan area and seropositivity of Hispanics for hepatitis A were also associated with a more
intense response to the 15/17-kDa marker (p<0.05). Results suggest that Cryptosporidium transmission occurs within
households, especially in large families. No associations were detected between the intensity of serological response and health
status.
rx-7
-------
"1 1
I! 11
||lT|He: . Examination of Risk Factors for Respiratory Effects in Childn
HT "" ' " "i Monitoring Data. Part I: Effect of Asthma Status and Househi
|J [ " Children and Adolescents.
n: Use of Respiratory Health and EPA Air
>ld Environmental Exposures on Lung Function of
LSI .;Y'L ,,.,,-.. : s ' .*>ซ*<**..-. *ป(ป, -~ ซ * .- w **ปปซ*' (, ,ซ-?i^, 5
pfeACoplift person , _,. T( , t( ( t>i ;- -*t - . ^ - ~< A * < (
1 Name: Robert Chapman Office Affiliation: National Center for
Research & Development. Researc
Phone Number: 9 19-541-4492
fSTETWCTTBE fT'!ปTl!?'S tW "" SrSCrS^^
pfSSriPt^nsCSIs* . -. ::;..;;. i ; .; ;4, .*: ;.; : .,*: :'>:,;.ป* .a,;,;: r ,-,,;;;;J
1) Which NHANES did you use? DNHANES-I DNHANES-n E
ONHANES99+
^
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, n
All 8-16 year-old subjects who had spirometric lung function testing.
3) Detailed description of the goals/purpose and approach of the study:
1. Ascertain effects of active and inactive asthma, and of the household en
a gas stove, and presence of a dog or cat, on children's and adolescents' lui
2. Compare sensitivities of different lung function metrics (i.e., FEV1, FV
household environmental factors.
4) What statistical methods and models are you using? Include a brief
appropriate, attach sample tables and/or charts.
Linear regression models using SUDAAN software. This gives the same e
regression, but also employs generalized estimating equations to adjust star
arising from clustering in the data.
5) If you use NHANES data without publishing documents, include a b
NHANES data and/or results from published NHANES studies to supp
support setting air quality standards for ambient pollutants.):
I'tfi ! v::a ' i ' a * A " ' "3- ' : '' '* t , 3,1, ซS5H,;;;: W;%ป? ?s^ซ";Tฃfe?ft;|
reflation and Abstract/Executive Summary ;
Hi 1 1 IIIVUI , 'u ill" ;, .ill", Jซ *,r d.Mh, 'Ill ,vl, i .rf,-,, ' JW> ^--^- ..9 i VK. ' :'.<
Citation:
f jf,,,,!:1 . i. .' . iv 1 ,fl n!:,,' '" 11, 1' ! , r. "( ' ' it1.;11! i,, in ซJni''!|i I1" -,,:,'. ' '!' Ivi'*1' > r ' '*' ; .|i " / , ' \- '^^ - '- '>^A
jlIO|lllldN1iliiii|i''i|n 'ir giiihrjjlll 111! U lrป ni Jli'i; ป j; ,11, ' ill,,11!1 ซ m ' :r,i i nun, ijilln, 'Jt ii,,ปn ซ MI,: ,u< ,L line j' Hhft 'VIBi'm, L * 'fHa !'i'"',i iปi,m. *VJ-1SiiTi1
For Completed Studies, Provide the Following Dates: 1
?Vfm::l v '" ,.r !ซ ' ', !iih, ' i i' i- - >&, s'i', ^',M *->'ซ, ;p,-:. :.H fi;-*ji=^-^ft.-^vy
Starting Date: Ending Date:
Environmental Assessment (NCEA), Office of
i Triangle Park, NC
' [
iS'^fl^^rf,^,M;-^C^:/;*::?S^%i.'SSWiซre^^;,r,?^;^4S;
fc^:>f^:;^*T.'r*:/^T^M^ฎ*^^:y;.S'.:.l;i
NHANES-m DHHANES
gion of country):
oronmental factors passive smoking, cooking with
ig function.
3, MMEF) in detecting effects of asthma and
summary of how you present the results. If
ffects estimates (betas) as does simple linear
dard errors and p-values for non-independence
ief description of these cases, (e.g., use of raw
>rt setting pesticide tolerances on food, or to
b^l^i^:fe'^^;SiS*''w*i^iป..^B!b^;-iff^^^- . ,'* j^^wif^'^iii^&i^^rlfc'^fl1:
! ; '
lliiiiiiifiifei^^
! 1 !
1
Ityปซiii ijt "ซ>*i '-. (^4*5* "ป; ,Jt'i/i*t #*
Klork in Progress. Provide the Following Information: : -." iElZปS-:i,iwL'-; '" *ฅ~:: ' k :\-^^-^-'ft''.-^.^'&V
Ill-ill flPii. ., 'ป(, . ..ป ,ป ,fSf, .,,. ป>'>*!. ',:,' ,,'t; I'lf i--ft;. ซซ. j'...'ia!' ."n-n.[-' ".::ซ'. - 1.: , >-"? !liSSS3?BBSSMi8* fr-s-"-- '.ซ-tป-s ซiiii"Vif!i!!:.ite.^'ซ'Sc*'ftป*^'ป'f'ซf':
it i - ', - -
Anticipated Produces): Peer-reviewed journal article. ,. ;
Tentative Completion Datc(s): June 2002.
LX-8
,
(-* !(; rt
'
-------
Examination of Risk Factors for Respiratory Effects in Children: Use of Respiratory Health and EPA Air
Monitoring Data. Part IT. Effects of Ethnicity on Lung Function in Children and Adolescents
l|f|$L |3(>ntact Person,
-* * i
Name: Robert Chapman
Phone Number: 919-541-4492
Office Affiliation: National Center for Environmental Assessment (NCEA), Office of
Research & Development. Research Triangle Park, NC
i- ^jfU. < "* >fKZ
;escriptiott of St
1 i*:
**
1) Which NHANES did you use? DNHANES-I DNHANES-II 0NHANES-HI DHHANES
DNHANES99+ (please check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
All 8-16 year-old subjects who had spirometric lung function testing.
3) Detailed description of the goals/purpose and approach of the study:
Ascertain effects of ethnicity on lung function in children and adolescents. Model-adjusted lung function will be compared
among non-Hispanic whites, non-Hispanic blacks, and Mexican-Americans. Comparative assessment of Puerto-Rican
Americans' lung function will also be conducted, to the extent that the NHANES m data allow.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts. .
Linear regression models using SUDAAN software. This gives the same effects estimates (betas) as does simple linear
regression, but also employs generalized estimating equations to adjust standard errors and p-values for non-independence
arising from clustering in the data.
5) If you use NHANES data without publishing documents, include a brief description of these cases, (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.): '_
; Citation and Abstract/Executive: Summary
Citation:
yBtor Completed Studies, Provide tne following Dates:^
Starting Date:
Ending Date:
IWork in Progress, Provide the Following Information:
Anticipated Prpduct(s): Peer-reviewed journal article.
Tentative Completion Date(s): June 2002.
LX-9
-------
1 : , '1 1
',,' !
" : . /"! 1 " :
i
. ' ' !
i :
;|||!!e? Examination of Risk Factors for Respiratory Effects in Children: Use of Respiratory Health and EPA Air
BiwฃBE i Monitoring Data. Part HI: Risk Factors for Asthma in Children and Adolescents
Sfpr -fk-ฑ-'r J.SSS - -'-&;
iSM ^"SSSiSP?, .,s i i- ซ; i ป ;:; ;ป,- ,H&.,;,n:.ซ^-'-^^Kk~.ฃ ~ SSiiVslIi's^llf&^iyiSf^
Name: Robert Chapman Office Affiliation: National Center for .Environmental Assessment (TSfCEA), Office of
Research & Development. Researaji Triangle Park, NC
Phone Number: 919-541-4492 i
(itjl ' ,1,1 thS* ' 1 H i *ซ- l31 itaBwifct * fe A i l||jjt fafcjk.rfSf* k ซ^ ป_*ปrttj,ซ JferisS ป* Sซf ซ, *J*f
JOdcrlpmm of Study JET ( *
tir r J IHtMSMuurmL. -^ - ^ r< tt*,.*
1) Which NHANES did you use? DNHANES-I DNHAPflES-H E
D NHANES99+ (please check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, re.
All subjects ages 0-16 years.
3) Detailed description of the goals/purpose and approach of the study:
Ascertain effects of host and environmental factors on prevalence of asthn
interest include age, gender, ethnicity, and atopic-allergic status. Environrr,
cooking with a gas stove, and presence of pets in the household.
4) What statistical methods and models are you using? Include a brief ;
appropriate, attach sample tables and/or charts.
Logistic regression models, adjusted for cluster effects with generalized esl
GENMOD procedure).
5) If you use NHANES data without publishing documents, include a bi
NHANES data and/or results from published NHANES studies to supp<
support setting air quality standards for ambient pollutants.):
SiHU'Ub irfl ' ' life ' !' i iif 1" "''"Hi ' '*** **' SS' :'!' "* *'*'f K "i IKTili1*?;!,1;?*!!! tfirs*- fr$
Sl?llfiM?^.lA^l|?c^ecu^|;iSum^ M,.'..U>^
Citation:
NHANES-ra DHHANES
f
gion of country):
la in children and adolescents. Host factors of
ental factors of interest include passive smoking,
ummary of how you present the results. If
imating equations (SUDAAN soflware or SAS
ief description of these cases, (e.g., use of raw
>rt setting pesticide tolerances on food, or to
.'
li^iSff -^SM- >^&:
KraV^ปWte^ifปv^.S ซS-i:,ซ*iซSliSsfc!SrfJ3liJ|ig:iBf! !
.Sr-^'.'.'ICj fe ; _; _, L _ ' ,,,||KPr'~*" *^ ' ' "* ^sJfK',, "J
^ c^gietM Studies, Fro\ide the Following Dates: j[_
Starting Date: Ending Date:
i1 r1'" "ป !"""!"'? ',l MIM IJI 'ซ' "Ill 1 W" ""f ซt.WfSiKSS;i|ซ^^StteSS-:S*ป/Ss^4S
\Y or(k in progress, Provide the Following Informatum- ^ ^^ ,: .Jj^^^^^-^-^.^^^^^-^^i
Anticipated Produces): Peer-reviewed journal article.
Tentative Completion Date(s): April 2002.
DC-10
;
" 1
I"'
-------
,|jTitie:
Mercury (Hg)
eVf *6,~j*~ <_.
'A Contact Person
Name: .ChuckFrench
Phone Number: 919-541-0467
Office Affiliation: Office of Air Quality Planning and Standards (OAQPS), Emission
Standards Div. (BSD), Office of Air and Radiation (OAR), Research Triangle Park,
NC '
^Description ofStudy
l)WMchNHANESdidyouuse? DNHANES-I DNHANES-H aNHANES-in DHHANES
El NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
Women of childbearing age (16-49 years old) and young children (1-5 years old).
3) Detailed description of the goals/purpose and approach of the study: Blood and hair Hg levels fromNHANES 1999
were used to help support a regulatory determination for Hg emissions from electric utilities as well as several other Hg
projects (e.g., Mercury Action Plan, PBT Monitoring Strategy). These data have proven to be very useful and informative
and support our technical and policy efforts with Hg.
4) What statistical methods and models are you using? Include a brief summary of how you" present the results. If
appropriate, attach sample tables and/or charts. Data and statistical results presented in the publication on March 2,2001
in the CDC's Morbidity & Mortality Weekly Report (MMWR) (see below) are used in this work. The MMWR presents the
data as geometric mean and percentiles (10th, 25th, 50th, 75th, 90th) for both the women and children for blood Hg and for hair
Hg. .
5) If you use NHANES data without publishing documents, include a brief description of these cases, (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.): NHANES results (as reported in the MMWR article) were
used with lots of other .information to support the general conclusion that Hg in the environment and Hg exposures for the
U.S. human population are of concern, and there is a need for emissions reductions. The amount of emissions reductions
needed to decrease environmental exposures was not quantified, nor were estimate was made of Hg environmental exposure
due to utility emissions. The NHANES results have been used in various briefings, posters, draft reports, etc.... to support
various Hg efforts and to present information on Hg exposures.
-^ * s* ** ** ^ s
Citation and Abstract/Executive Sujnmary
Citation: Blood and Hair Mercury Levels in Young Children and Women of Childbearing Age United States, 1999.
MMWR, March 2, 2001, Vol 50, No 08; 140. (Contributing authors from EPA included Kate Mahaffey and Chuck French;
http://www.cdc.gov/mmwr/preview/mmwrhtml/mrn5008a2.htrn
l)afes for Completed Studies:
*>3tfl
' f. ^
Starting Date:
Ending Date:
information for Woriyn Progress:
^S^SS ~*V."F* "3**-.^
Anticipated Produces): Various
Tentative Completion Date(s): Uncertain
Abstract:
Blood and Hair Mercury Levels in Young Children and Women of Childbearing Age United States, 1999.
LX-11
-------
!! ;i
Mercury (Hg), a heavy metal, is widespread and persistent in the environmer t. Exposure to hazardous Hg levels can cause
permanent neurologic and kidney impairment (13). Elemental or inorganic! Hg released into the air or \rater becomes
methylated in the environment where it accumulates in animal tissues and inisfeases in concentration through the food chain. The
U.S. population primarily is exposed to methyl mercury by eating fish. Mettlyl mercury exposures to women of childbearing age
are of great concern because a fetus is highly susceptible to adverse effects, 'this report presents preliminary estimates of blood
and hair Hg levels from the 1999 National Health and Nutrition Examination' Survey (NHANES 1999) arid compares them with
a recent toxicologic review by the National Research Council (NRC). The findings suggest that Hg levels in young children and
women of childbearing age generally are below those considered hazardous, j These preliminary estimates show that
approximately 10% of women have Hg levels within one tenth of potentially [hazardous levels indicating a narrow margin of
safety for some women and supporting efforts to reduce methyl mercury exposure.
! I
CDC's NHANES is a continuous survey of the health and nutritional status of the U.S. civilian, noninstitiitionalized population
with each year of data constituting a representative population sample. A household interview and a physical examination were
conducted for each survey participant. During the physical examination, blood was collected by venipuncture for all persons
aged >1 year and hair samples, consisting of approximately 100 strands, were cut from the occipital position of the head of
children aged 15 years and women aged 16-49 years. Whole blood specimens were analyzed for total Hg and inorganic Hg for
children aged 1-5 years and women aged 16-49 years by automated cold vapor atomic absorption spectrophotometry in CDC's
trace elements laboratory. The detection limit was 0.2 parts per billion (ppb) for total Hg and 0.4 ppb for inorganic Hg (4).
Hairs of 0.6 inches (1.5 cm) closest to the scalp (approximately 1 month's grciwth) were analyzed for total Hg concentration using
cold vapor atomic fluorescence spectroscopy (5). The limit of detection for total Hg in hair varied by analytic batch; the
maximum limit of detection (0.1 parts per million [ppm]) was used in these analyses. Blood Hg levels less than the limit of
detection were assigned a value equal to the detection limit divided by the sqjiare root of two for calculation of geometric mean
values. The geometric mean total blood Hg concentration for all women age
-------
Iffe:
Methyl Ethyl Ketone (MEK) (also called 2-butanone)
Corilacf Person
Name: Oiuck French Office Affiliation: Office of Air Quality Planning and Standards (OAQPS), Emission Standards
Div. (BSD), Office of Ak and Radiation. (OAR), Research Triangle Park, NC
Phone Number: 919-541-0467 .
**'"" -Jf~> - * !ปป"
Description of Study
1) Which NHANES did you use? DNHANES-I DNHANES-IE ENHANES-HI DHHANES
DNHANES99+ (check to the left of 'each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
Unstratified sample of adults 20-59 years of age. Variables such as age and race were not used in the analysis.
3) Detailed description of the goals/purpose and approach of the study: Blood MEK measurements from NHANES El
were used to determine background levels in humans due to environmental exposures plus normal metabolism. These data
were considered along with other information to help evaluate a petition submitted to EPA by the Chemical Industry to delist
MEK from the Clean Air Act List of Hazardous Air Pollutants.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts. Raw NHANES-DI data were not used. Instead, data and statistical results
were taken from 2 publications (Ashley, et al., 1994 and Churchill, et al. 2001, see below for full citations). The results were
as follows: n = 1101 adults; median MEK blood level = 5.4 ppb; mean = 7 ppb; 5th %ile = 2 ppb; 95th %ile = 17 ppb. A
positive association was observed for alcohol, smoking, & exposure to moth balls & pressure-treated wood, but, these factors
do not appear to account for average exposures.
5) If you use NHANES data without publishing documents, include a brief description of these cases. (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.): Published MEK data were used in the evaluation of the
delist petition. When a final decision on the petition has been made, it will be published in the Federal Register and will
reference the NHANES data and studies. These NHANES results are one of several important factors in the overall review
and analysis of the petition. Various calculations were made to compare the NHANES blood levels with those levels one
might expect to see in a population from chronic exposure to MEK at the level of EPA's Inhalation Reference Concentration
(RfC). The mean blood levels from NHANES were about 2 times higher than the levels expected from chronic exposure at
the RfC. This was useful information for the analysis and raised questions about the RfC, natural levels of MEK, etc.
t,Citation>andAbstract?ExecutLve/Summaryx
Citation:
tStSf*^- .,-*_-"- * -"ซ " X < * ^
vFor Completed Studies, Provide the Following Tjates:
Starting Date:
Ending Date:
Kirk in Progress, Provide the Following Information:
Anticipated Produces): Federal Register Notice
Tentative Completion Date(s): Year 2002
References Used to Estimate MEK Background Levels for this Work:
1. Ashley, et al. 1994. Blood Concentrations of Volatile Organic Compounds in a Non-occupationally Exposed US Population
and in Groups wifli Suspected Exposures. Clin. Chem. 40/7, 1401-1404 (1994).
rx-13
-------
r:"
it
2. Churchill, I.E., Kaye, W.B. 2001. Recent Exposures and Blood Volatile Organic Compound Levels in a Large Population-
Based Sample. Archives of Environmental Health. March/April 2001. Vol. 56 (No. 2).
rx-14
' ! l'ป
f Ij
.; ill
i,;;
" I | S
i i''ป. I
\-"-.\\ ?l
uli
i 14"'
L;;|
, i' -t" i:,
i:.'
-------
Environmental Risk Factors for Presence of Cryptosporidium parvum Antibodies in Human Serum
jGJontact Person
Name- Elizabeth HUborn
Name. Jiiizaoeui mioo
Phone Number: 919- 966-0658
Office Affiliation: Epidemiology & Biomatkers Branch, Human Studies Division,
^^ ^^ & Environmeatal Effects Researcn Laboratory (NHEERL), Research
Triangle Park, NC
^ ^ V_ ** / i. ^ ^f^ J3, ซ#*.^ป-A Jf Hซl * * f A^ ป4t
Jescfiption of Study . - . - '.""" . r ; _^_2
1) Which NHANES did you use? DNHANES-I DNHANES-H BNHANES-in HIHHANES
d NHANES99+ (check to the left of each appropriate box)
i Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country): Surplus sera from subjects within
JSrfSSSSSp^samplifg units (PSU's) were used. These seven geographic areas were selected because
they differed in their sources and treatment of drinking water. First, the PSU's were selected, then sera were selected
randomly from subjects living in each PSU. Subjects included both sexes, various ages and race/ethnic subgroups.
description of the goals/purpose and approach of the study: A subset of NHANES m participants' serum was
^~. . me presence of anti- Cryptosporidium antibodies. Environmental risk factors, such as water source and preferred
diet, were assessed as risk factors for seropositivity.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. H
appropriate, attach sample tables and/or charts. Univariate and Bivariate analysis were conducted. Logistic regression was
used to evaluate the contribution of multiple risk factors.
5) If you use NHANES data without publishing documents, include a brief description of these cases, (eg., use of raw
NBA^ESTdatelmd/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.): N/A
Citation and Abstract/Executive Summary
Citation:
2000.
I Frost F Neas L, Muller T, Calderon R. Environmental Risk Factors for the presence of Cryptosporidium parvum
representation at: International Society for Environm:ental Epidemiology 2000. Buffalo, New York. August,
forjCpmnleted Studies:
Starting Date:
^ -ง*f--f~
Sr"
Ending Date:
^Information for Work in Progress:
Anticipated Produces): Journal publication
Tentative Completion Date(s); December 2003
Environmental Risk Factors for Presence of Cryptosporidium parvum Antibodies in Human Serum
E. Hilborn, F. Frost*, L. Neas, T. Muller*, R. Calderon
USEPA/HSD; *Southwest Center for Managed Care Research, Albuquerque, NM
TX-15
-------
I
I "I
il
Background- Ctyptosporidium parvum is aparasite that infects a variety of vertebrate hosts. Sources of infection include
contaminated raw produce and drinking water. Outbreaks of food borne and katerborne infections are infrequently reported, yet
serological evidence suggests that infection may commonly occur. !
Methods - We analyzed serum and data gathered from a convenience sample of National Health and Nutrition Examination
Survey III participants to identify risk factors for the presence of anti-cryptoiiporidial antibodies. Seropositivity was defined as:
those samples determined to contain IgG antibodies to 15/17 kDa proteins by the enzyme-linked immunoelectro-transfer blot
assay with an intensity of response greater than 10% of a positive control. D|ata were characterized by univariate analysis.
Bivariate statistical analysis and logistic regression were used to determine tlie association between gender, age, race, city of
residence, salad and water consumption, serum carotenoids (biomarkers of exposure to produce), pet ownership, antacid and
bottled water usage, home water source, and seropositivity. . t
* i J .,
Results- Of 1356 participants, 558 (41%) were seropositive. Persons resided m seven cities and ranged in age from 1-90 years
(median: 38 years); 50% were female, 50% white race, 25% African-Americijui race, and 22% Hispanic ethnicity. Seropositivity
varied by city (range: 29% - 57%). Older participants were significantly moije likely to be seropositive (trend p< 0.0001). The
full regression model included gender, age, race, city, and serum alpha-carotene, all significantly associated with seropositivity;
inclusion of an increased water consumption variable in the model improved [the fit, but was not statistically significant (p=0.09).
No other variables listed above were significantly associated with seropositivity.
"' ' : "' !' | ''''
Conclusion - These data suggest that significant differences in seropositivity sxist among populations of different cities, and that
prevalence of seropositivity increases with age. Increased serum alpha-carotene is significantly associated with seropositivity.
Although serum alpha-carotene levels have been correlated with consumption of produce, the biomarker is not specific to the raw
produce items which may be potential vehicles for oocysts. Although exposiire to municipal water (by city) and serum alpha-
carotene appear to be independent risk factors for seropositivity, unmeasuredjdifferences in food consumption, behavior, or
cultural practices among persons living in various geographic areas may contribute to these findings. This is an abstract of a
proposed presentation and does not necessarily reflect EPA policy. !
1! ''
,),.
If
i
i .
!i:
i
rx-16
i
-------
Physical Activity and Nutritional Status of US Children
Contact Person
Name: Thomas McCurdy
Phone Number: 919-541-0782
Office Affiliation: National Exposure Research Laboratory (NERL),
Research Triangle Park, NC
Description ofStudy
1) Which NHANE'S did you use? 0NHANES-I HNHANES-H HNHANES-m OHHANES
DNHANES99+ (please check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
Children cohorts that are defined variously hy age & gender classes.
3) Detailed description of the goals/purpose and approach of the study:
Description of .the amount of physical activity and exercise that children participate in; description of the total caloric intake
of children. These data are used to evaluate the representativeness of human activity information that we use to estimate
human exposure to environmental contaminants. "*
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts.
Various tests of statistical hypotheses (mostly 2-sample K-S tests), analysis of variance, analysis of covariance.
5) If you use NHANES data without publishing documents, include a brief description of these cases. (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
These analyses generally are not published, but are used for evaluating data sets that are used as inputs in in-house modeling
evaluation efforts. Some of these evaluation studies will be published, but most are not.
, , .
^Citation and Abstract/Executive Summary
Citation:
ijfoj,Completed Studies, Provide the FollowinJ'patesT
Starting Date:
Ending Date:
^
Wof k iiSTProgress, Provide thf Following Information:
Anticipated Product(s): Journal article on comparing exercise participation rates determined from the literature with those
derived from NERL's Consolidated Human Activity Database.
Tentative Completion Date(s): FY2002
Example of the types of NHANES secondary studies that are used by this Office:
Albanes, d., Blair, A., and Taylor, P.R. (1989). "Physical activity and. risk of cancer in the NHANES I population." Amer. J.
Pub. Health 79: 744-750.
Blair, D., Haabicht, J.P., and Alekel, L. (1989). "Assessment of body composition, dietary patterns, and nutritional status in the
National Health Examination Surveys and National Health and Nutrition Examination Surveys," pp. 79-104 in: T. Drury (ed.).
Assessing Physical Fitness and Physical Activity in Population-Based Surveys. Washington DC: U.S. Government Printing
Office.
LX-17
-------
- I ill"
!. ! II;
Brcslow, R.A., et al. (2001). "Long-term recreational physical activity and breast cancer in the National Health and Nutrition
Examination Survey I epidemiologic follow-up study." Cancer Epidem. Bioimark. Prey. 10:805-808.
' ^,,,..,:, ::::! , , I , | .- . ,. ,
Briefel, R.R., McDowell, M.A., Alaimo,K., Caughman, C.R., Bischof, A.L., Carroll, M.D., and Johnson, C.L. (1995). "Total
energy intake of the US population: The .third National Health and Nutrition Examination Survey, 1988-1991." Amer. J. Clin.
Nutr. 62(Supp.): 1072S-1080S. :
Clark, G. and Whittemore, A.S. (2000). "Prostate cancer risk in relation to anthropometry and physical activity: The National
Health and Nutrition Examination Survey I Epidemiological Follow-Up Studly." Cancer Epidem. Biomarkers Prev. 9: 875-881.
Crespo, C.J., Smit, E., et al. (2001). "Acculturation arid leisure-time physica
fromNHANES m. 1988-1994." Amer. J. Pub. Health 91:1254-1257.
inactivity in Mexican American adults: Results
I ,, i |
Eck, L.H., Hackett-Renner, C., and Klesges, L.M. (1992). "Impact of diabetic status, dietary intake, physical activity, and
smoking status on body mass index in NHANES H." Amer. J. Clin. Nutr. 5(.
: 329-333.
Farmer, M.E., Locke, B.Z., MbScicki, E.K., Dannenberg, A.L., Larson, D.B., and Radloff, L.S. (1988). "Physical activity and
depressive symptoms: the NHANES I epidemiologic follow-up study." Ameir. J. Epidem. 128: 1340-1351.
= ..' i [
, ' ' i
Ford, E.S. (1998). "Characteristics of survey participants with and without a telephone: Findings from the Third National Health
and Nutrition Examination Survey." J. Clin. Epidem. 51:55-60. '',.',.
i
!, ' '
Gillum, R.F., Mussolino, M.E., and Ingram, D.D. (1996). "Physical activity [and stroke incidence in women and men. The
NHANES I epidemiologic follow-up study." Amer. J. Bpidem. 143: 860-869. ' ! s
Klesges, L.M., Klesges, R.C., and Cigrang, J.A. (1992). "Discrepancies between self-reported smoking and carboxyhemoglobin:
an analysis of the Second National Health and Nutrition Survey." Amer. J. Pab. Health 82:1026-1029.
McDowell, A.J. (1989). "Cardiovascular endurance, strength, and lung func'iion tests in the National Health and Nutrition
Examination Surveys," pp. 21-77 in: T. Drury (ed.). Assessing Physical Fitness and Physical Activity in Population-Based
Surveys. Washington DC: U.S. Government Printing Office. ;
': , ... '..'::' . T. ;,. , .
Norris,J., Harnack,L., Carmichael, S., Pouane,T., Wakimoto.P., and Block, G. (1997). "US trends in nutrient intake: the
1987 and 1992 National Health Interview Surveys." Amer. J. Pub. Health 87: 740-746.
Troiano, R.P. et al (2000). "Energy and fat intakes of children and adolescent's in the United States: Data from the National
Health and Nutrition Examination Surveys." Amer. J. Clin. Nutr. 72(Supp.): T343S-I453S!
Schectman, G., McKinney, P., Pleuss, J., andHof&nan, R.G. (1990). "Dietar
cholesterol diet (NHANES n)." Amer. J. Public Health 80: 698-703. |
intake of Americans reporting adherence to a low
LX-18
, I",, : I
u
1 : .-i if',;;
; | 13:
i ! I iป!'
-------
Exposure Factors Handbook
E~PA Contact Person
Name: Jacqueline Moya
Phone Number: 202-564-3245
Office Affiliation: National Center for Environmental Assessment (NCEA), Office of
Research and Development (ORD); Washington, DC
Qgescription of Study
1) Which NHANES did you use? DNHANES-r BNHANES-II ENHANES-m DHHANES
D NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
We looked at all ages and both M/F.
3) Detailed description of the goals/purpose and approach of the study:
To derive body weights and body surface areas based on body weight that can be used for risk and exposure assessments The
data are presented in tabular form as part of the Exposure Factors Handbook, which contains numerous tables of data on
various socioeconomic and exposure-related variables derived from a variety of sources including NHANES.
4) What statistical methods and models are you using? Include a brief summary of how you present the results If
appropriate, attach sample tables and/or charts.
Mean and median values of body weight and body surface area were calculated for subjects by age; by gender- and by
combined age and gender. Age was classified as 2-6 months, 7-12 months; then yearly until age 17; thereafter as 18-24 25-
34,35-44,45-54,55-64, 65-74,75 years and older. '
5) If you use NHANES data without publishing documents, include a brief description of these cases. (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food or to
support setting air quality standards for ambient pollutants.):
We used analysis done by others on the NHANES H and HI data
^Citation and Abstract/Executive Summary
Citation: Exposure Factors Handbook EPA/600/P-95/002Fa-c
Child-Specific Exposure Factors Handbook EPA/600/P/00/002F
fPatestor Completed Studies:
Starting Date:
Ending Date: EFH - 1997; CS-EFH external review draft June 2000
Interim final may be released in 2002.
Kformation for Work in Progress:
Anticipated Produces):
Tentative Completion Date(s):
Abstract:
The Exposure Factors Handbook provides a summary of the available statistical data on various factors used in assessing human
exposure. This Handbook is addressed to exposure assessors inside the Agency as well as outside, who need to obtain data on
standard factors to calculate human exposure to toxic chemicals. These factofs include: drinking water consumption- soil
uigestion; inhalation rates; dermal factors including skin area and soil adherence factors; consumption of fruits and vegetables
fash, meats, dairy products, homegrown foods; breast milk intake; human activity factors; consumer product use- and residential
characteristics. Recommended values are for the general population and also for various segments of the population who may
rx-19
-------
have characteristics different from the general population. NCEA has strived to include full discussions of the issues that
assessors should consider in deciding how to use these data and recommendations. The document is in final form, hut as new
data become available updates will be posted on this home page. . [ ; ,
,,',''' i t 'i .,:,, ' ,
NCEA has also produced a CD-ROM that contains an interactive version of t|he Exposure Factors Handbook. The CD-ROM has
word search capabffities, downloadable tables, hypertext links to various chapters in the document, and key references.
A limited number of paper copies and CD-ROM version of the Handbook are available from the National Service Center for
EnSmenS Publications (NSCEP) in Cincinnati, Ohio (phone 1-800-490L9198; 513-489-8190; fax 513-489-8695). Please
providethetitteandEPAniimberwhenorderingfromNSCEP.Documentsinayalsobeorderedon-lineat
www eoa.gov/NCTPTTiome/orderDub.htrnl. Paper copies may be purchased from the National Technical Information Service
(NTIS) in Springfield, VA (phone l-800-553-NTIS[6847] or 703-605-6000;! fax 703-321-8547).
:C'; I
IX-20
-------
The performance of human subjects on three neurobehavioral tests included in the Third National Health and
Nutrition Examination Survey
Contact Person
Name: David Otto
Phone Number: (919) 966-6226
Office Affiliation: Clinical Research Branch, Human Studies Division (HSD), National
Health & Environmental Effects Research Laboratory (NHEERL), Research Triangle
Park,NC
Description of Study
1) Which NHANES did you use? DNHANES-I DNHANES-II ENHANES-m DHHANES
D NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region.of country):
Half-sample of adults (20-59 years old) with odd-numbered survey identification numbers who participated in NHANES-IE.
The sample size is 5,662 and is a representative national sample. Blacks and Mexican-Americans were over-sampled to
obtain more precise estimates for these groups.
3) Detailed description of the goals/purpose and approach of the study: The purpose of the study was to obtain a
representative national sample of the performance of working-age adults on three computer-assisted neurobehavioral tests:
simple reaction-time, symbol-digit substitution (coding) and serial digit learning (short-term memory). The tests used were
from the Neurobehavioral Evaluation System (NES2). Demographic variables assessed included gender, age, educational
level, family income, and race-ethnicity.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts. The sample design was a stratified, multistage probability design.
Performance estimates were calculated using the NHANES-m central nervous system (CNS) sample weights (WTPFCNS6)
calculated with SUDAAN (release 7.5.2). Variances were estimated by SUDAAN using a linear approximation method.
Means and standard errors for performance measures are presented for gender (M/F), age groups (20-29,30-39,40-49 50-
59), education levels (0-8, 9-11,12,13+), family income (<10,000,10,000-29,999, 30,000-49,999, >50,000) and race^
ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, other).
5) If you use NHANES data without publishing documents, include a brief description of these cases. (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
Citation and Abstract/Executive Summary.
. .*ซ,.
Citation: Krieg, EF; Chrislip, DW; Letz, RE; Otto, DA; Crespo, CJ; Brightwell, WS; Ehrenberg, RL. The performance of
luman subjects on three neurobehavioral tests included in the Third National Health and Nutrition Examination Survey.
Neurotoxicology and Teratology (2001), 23:569-589.
Sates for Completed Studies?
Starting Date: 1988
Ending Date: 1994
Information for Work in Progress:
-"ซ~~-
Anticipated Produces):
Tentative Completion Date(s): NA
Abstract:
rx-21
-------
:!'!
H I
KRffiG JR., E. F., D. W. CHRISLIP, R. E. LETZ, D. A. OTTO, C. J. CRESPO, W. S. BRIGHTWELL AND R. L.
EHMBNBERG. Neurobehavioral Test Performance in the third National He'alth and Nutrition Examination Survey.
NEUROTOXICOL TERATOL (2001), 23:569-589. The third National Health and Nutrition Examination Survey (NHANEb
ffl) contained three computerized neurobehavioral tests from the Neurobehavioral Evaluation System: simple reaction tune,
symbol-digit substitution, and serial digit learning. The neurobehavioral date that were collected came from a nationally
representative sample of adults 20 to 59 years old. Performance on the tests was related to sex, age, education level, family
income, and race-ethnicity. Performance decreased as age increased, and increased as education level and family income
increased. Differences in performance between sexes, levels of education, atild racial-ethnic groups tended to decrease as family
income increased. The relationship between age and performance on the symbol-digit substitution test varied by education level
and by racial-ethnic group. The relationship between age and performance oti the serial digit learning test varied by racial-ethnic
group Questionnaire variables that were related to performance on one or more of the tests included the reported amount of last
night's sleep, energy level, computer or video game familiarity, alcoholic beverages within the last three hours, and effort.
Persons who took the tests in English or Spanish performed differently on the symbol-digit substitution and .serial digit learning
tests. Performance on all the tests decreased as test room temperature increased.
a
'
! i, |
qi'
I
f- '
I" I
1 M
K-22
-------
Examination of Risk Factors for Respiratory Effects in Children: Use of NHANES-ffl Respiratory Health and
EPA Air Monitoring Data. Part IV: Estimation of Ambient Air Pollutant Concentrations and Their Effects on
Children's Respiratory Health
Name: Susan Perlin
Phone Number:202-564-3248
Office Affiliation: National Center for Environmental Assessment (NCEA); Office of
Research and Development (ORD), Washington, DC
Description of Study
* 5- * ^ *% i. ?
Tซ. ; * t *--^
1. I
Sj- v*.
1) Which NHANES did you use? DNHANES-I DNHANES-II E NHANES-m DHHANES
D NHANES99+ (please check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country): All children and adolescents 0-
16 years of age. This is a data set of 13,944 subjects and includes both sexes and all ethnic groups. In these children, we
evaluate outdoor and indoor environmental effects on asthma prevalence, respiratory symptom prevalence, and spirometric
lung function (e.g., FVC, FEVJ. NHANES-HI collected data on asthma arid respiratory symptoms for all children, regardless
of age, but performed lung function tests only on subjects 8 years and older.
3) Detailed description of the goals/purpose and approach of the study: This is a multi-part study to evaluate risk factors
for respiratory effects (i.e., lung function and symptomatology) in children. Parts 1-3 are described in preceding project
summaries (Contact Person: Robert Chapman) and address risk factors based only on data collected in NHANES-IH. The
remainder of the work builds on models developed in Parts 1-3 to examine risk factors associated with ambient air quality and
urbanization. Geographic information system (GIS) technology is used with non-NHANES databases to develop measures of
ambient air pollution, population density and road density for each year of the Survey for all locations with NHANES-m
subjects. Environmental variables are estimated at the census block group level (BG) and linked at that level to the
NHANES-m subjects. To maintain subject confidentiality, the National Center for Health Statistics does all the data Unking
and runs all models on the linked data. EPA's national air monitoring data are used to estimate levels of atmospheric
pollution (O3, PM10, SOx, NOx, CO and Pb). Monitoring data are interpolated to estimate ambient air pollutant
concentrations for all BGs in the NHANES-IH counties. As yet, there is no scientific consensus as to the interpolation
method of choice. Therefore, so several different interpolation methods are used to derive air concentration levels and
determine the conditions under which different methods produce significantly different concentration values. Method-
specific concentration values are linked with each NHANES-m subject to ascertain and compare the impact of alternative
methodologic approaches on the evaluation of environmental health effects.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts. Kriging, spatial averaging, nearest neighbor and inverse distance
weighting are the interpolation methods used to estimate concentration levels of pollutants (O3, PM10, SOx, NOx, CO and
Pb) at the BG level for NHANES-in counties. Population density is calculated from 1990 census data by dividing the total
population by the land area for each BG. A measure of road density has tentatively been calculated using 1990 Census
TIGER data and estimating the length of road segments for each BG.
Regression models developed in Parts 1-3 of this study incorporate variables such as active and passive smoking, presence of
home combustion devices, allergy status of subjects, parental history of asthma and allergies, and socioeconomic and
demographic characteristics of subjects and their families to determine their effects on respiratory health of children. As a
starting point, interpolated air concentration values, other measures of urbanization, and meteorologic variables will be added
to these regression models to evaluate the impact of ambient environmental exposures on children's respiratory health.
5) If you use NHANES data without publishing documents, include a brief description of these cases.
(e.g., use of raw NHANES data and/or results from published NHANES studies to support setting pesticide
tolerances on food, or to support setting air quality standards for ambient pollutants.):
TX-23
-------
i. ,M,i:
::
'
Citation:
Starting Date:
Ending Date:
^m^,^^^^-i^^^^^o^^^^or^liti^f*'f * **ซ 1Bฃ*^
|l|li| II I i B!I| I II jr^ If ^ I i" lull ijill W'lii" i"1'! rii'm^iigirH n !i" . .TS
nS report "ExSa'tion of Risk Factors for Respiratory Effects in Chilldren: Use of NHANES-HI Respiratory Health and
EPA Air Monitoring Data" (tentative title) describes the four interpolation methods and results across all BGs in all
NHANES-m counties for O3 and PM10 for the year 1990. j
2) Several journal articles to be written for publication detailing evaluation^ of the linked health and exposure data.
Tentative Completion Date(s): ,
1) EPA report: External review of draft report completed. Finalized report should be completed by 3/(XJ.
i '
2) Journal articles will be written and submitted for publication during 2002 - 2004. ,
IX-24
-------
Development of Methods, Data, and a Model of Human Cumulative Exposure, Dose, and Health Risks for
Pyrethroid Insecticides - Task 4: Evaluation of dietary exposure model
AjContact Person
. . *
Name: James-Quackenboss Office Affiliation: National Exposure Research Laboratory (NERL), Human Exposure &
Atmospheric Sciences Division, Office of Research and Development (ORD), Las Vegas, NV
Phone Number; (702) 798-2442 -
fSeseripfion of Study * 1 ^ ~? - "" " ""^-" ฃ*ป '#* * ^* } -" ** (" " -^1
^v^^^ -r~, .. t* &* ~ป >- > jlT ^ & *? '^'f' "* ซ- ^* s
LX-25
-------
1) Which NHANES did you use? D NHANES-I
(check to the left of each appropriate box)
DNHANES-n DNHANES-in DHHANES 0NHANES99+
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country): Sample of adults and children
whose urine samples were analyzed for pesticide metabolites, and who provided food diaries.
3) Detailed description of the goals/purpose and approach of the study: the objective of the overall study is to develop an
exposure-dose-response model that will allow for more well-informed decisions on managing the potential cumulative risks
associated with aggregate exposures to pyrethroids. The purpose of this task is to evaluate a dietary exposure model which is
based on individual food consumption patterns collected in NHANES 99+ iirid currently available food residue databases.
Approach: Estimates of the subjects' dietary exposure to several pyreihroids will be modeled Using their food diaries and
information on food pesticide residues. A dose-estimating model will then 'be used to derive the corresponding urine-
metabolite concentrations. These urine-metabolite estimates will be compared with the subjects' urine-metabolite
concentrations measured hi NHANES99+ to provide a "reality check" on indirect estimates of dietary exposure, and to
identify the relative contribution of dietary exposure to the total eliminated flose. The SHEDS (Stochastic Human Exposure
and Dose Simulation) model dietary module currently uses the 1994-96 and' 1998 Continuing Survey of Food Intake by
Individuals (CSFII) consumption data, and available food pesticide residue databases, including USDA's Pesticide Data
Program (PDP) and FDA's Total Diet Study (TDS). SHEDS will be adapte[j to use food consumption records for the same
individuals who have urine-metabolite results hi NHANES99+. The metabolite currently available for NHANES99+, 3-
phcnoxybenzoic acid (3PBA), is the common metabolite of several pyrethroids including permethrin, cyperrnethrin,
dcltamethrin, tralomethrin, fenvalerate, cyhalothrin, fluvalinate, and esfenvaierate. Multiple simulations will be conducted for
each individual's consumption record to characterize both variability and uncertainty in the exposure estimates. (SHEDS uses
a two-stage Monte Carlo approach to estimate uncertainty distributions). Within-individual variability will include
differences in residue concentrations. Model uncertainties include assumptions made for food items and pesticides not
represented in the available residue databases, differences between food consumption and residue databases (e.g., relating
CSFII consumption information to pesticide concentrations for TDS foods),! and methods for handling observations below the
detection limits. The Exposure Related Dose Estimating Model (ERDEM) will use published PK parameters, together with
absorption and elimination coefficients determined hi this Project (Task 1), to estimate blood and urine-metabolite
concentrations. I
4) What statistical methods and models are you using? Include a brief s immary of how you present the results. If
appropriate, attach sample tables and/or charts. '
A distribution of metabolite concentrations will be calculated for each individual. The 95%-confidence intervals for the
individual's mean will be compared with the urine-metabolite concentration! measured by CDC for the NHANES 99+. This
comparison will be grouped by residential insecticide usage during the previous month, as reported in the NHANES Pesticide
Use Questionnaire, to help distinguish dietary from other sources of these e?;posures.
The measured metabolite concentrations provide an upper limit for defining [acceptable model performance. Thus, if the
model is performing adequately then we should expect dietary exposure to account for some percentage (<100%) of the
measured urine-metabolite concentrations, plus or minus the variability and uncertainty bounds. The remaining excreted dose
provides an estimate of the contribution from non-dietary exposures that arejpresent in the US population.
S) If you use NHANES data without publishing documents, include a brief description of these cases. (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
This task is part of a proposal which was submitted in response to a "Call fo:r Pre-Proposals for Collaborative Research on
Intermittent Exposure/Risk in Support of ORD Safe Food Research Program." The project was approved in 1/02.
I \ { I | I I J L | t | j II
Cation .md Abstract/Executive Summary
A
"
Citation:
DC-26
: !"i'"
* l'"':
1
I
'.Him I
-------
^aJ^Wt.Completedlltuefiest *,
t~ J ='
Starting Date:
Ending Date:
!*rtVfrti*iYto'i'i'nปป ^ft^"Vtn>*^"irt":JlPrft^V^"cc^r''S?'"rS^'i?'^-^^-i^^::^vJ^^;~-^^^^$^-5S^
Anticipated Produces): manuscript on comparison of SHEDS-EKDEM (exposure-dose) model with pesticide metabolites
Tentative Completion Date(s): Task 4 is planned for FY2003; the manuscript will be prepared in FY04
IX-27
-------
'i- ,''
'' . 1 ''! ' '":
. ;> : :\->'\ : V: '!. ,;i : _'.
HjJTltJe:,^, ., Hispanic Health and Nutrition Examination Survey (HHANES), 1982-1984. Pesticides in urine and blood.
pEj^iisSBfti^ ^,s^
Name: Dina M. Schreinemachers Office Affiliation: Epidemiology an
(HSD), National Health & Environ]
Research Triangle Park, NC
Phone Number: (919) 966-5875
i ! r ' " i 1
Inscription of Study ^ j
d Biomarkers iBranch, Human Studies Division
nental Effects Research Laboratory (NHEERL),
m^ p,^-^ ^ ^ -,, w , ^.
<# 0* * Y- -
ฃ Si_ r 7
1) Which NHANES did you use? DNHANES-I DNHANES-H DNHANES-IH H HHANES
nWBANES99+ (check to the left of each appropriate box) i
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
HHANES (1982 - 1984) examined three distinct U.S. Hispanic subpopulations: Mexican-Americans living in CA, AZ, NM,
CO.andTX; Cuban-Americans living in Dade county, FL; and Puerto Ricans living in the New York City area. We studied
both men and women, ages 12-74, for these three subpopulations.
3) Detailed description of the goals/purpose and approach of the study:
Report on pesticide residue and metabolite levels in the blood of all 3 Hispanic subpopulations and the urine of the Mexican-
Americans. ' .
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables arid/or charts. _
Means and frequencies of pesticide residues and metabolites. Pesticide levels in this report are determined by gender for
children aged 12-17, and adults aged 18-74. The percentage of subjects anc estimated population at or above the level of
detection, and the percentile distributions are determined for each pesticide residue. The pesticide and metabolite data
themselves should not be interpreted in relation to one another (e.g., the ratio of metabolites to a specific compound to each
other), since no individual data are provided in this report. The proportion of exposed people and the average exposure levels
can be compared across populations. . ;
5) If you use NHANES data without publishing documents, include a br ief description of these cases, (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
trigii^
Citation: Schreinemachers DM, Gonzales M, Everson RB, Mendola P, Ray BM, Lewis DR. .Hispanic Health and Nutrition
Examination Survey, 1982-1984. Pesticides in urine and blood. EPA Report, in press.
P'1"1!"*!118 "'ป '"' T ' ! >T I!"11!1""1'!!" '" I!1*!11"!!! "" M- '-1'1 ' ' l|";llllll!'!l"i;i'. ili!Jl|ป! ซ"' ' ''*"H'" *'K , if li^iSHiilFSfiw"1 "w'Wr.n'^1; "^y'r^^Wi^-f-^^^'--'^^^^^^'^^^^^.^^
tSjSjtfjs.fbf Completed' Studies: ? ,:...,..,-,,. v>'.vjr; ./X^'^/dl
Starting Date: 1998 Ending Date: 2000
: .!: . -
inlmrn)anoh''for"Workin,Progress: ": : " ''''".' ''- ';';": ''^''^^^^^^^^^Js^:^^i^^^^^^^^-&'f^
"hit~ 11 jiii: iii'i Hi1;1,11: ,,J1,,I '^i, ^ M,:. ";,,iC?,,;jL,: ' , ,,::,ป.;,!ii':ซi :. . r '^A i,J.^\ซ-:ri.iii,l ii.'n,". ,IM 'i!w; - .; iซ,r. >:~^\famฎ&vim$#ii&^ฎ*!ฎf!'-!':^^: :?^^^ฎ^^t-v/*:".:*':<^**^!m&'f:-ซ?!"'*
Anticipated Produces):
Tentative Completion Date(s):
Abstract:
, ! 1
; . : . ^ \ !
The Hispanic Health and Nutrition Examination Survey, conducted during 1982 - 1984 by the National Center for Health
Statistics, examined the nutritional and health-status of three distinct U.S. Hispanic subpopulations: Mexican-Americans living
in California, Arizona, New Mexico, Colorado, and Texas; Cuban-Americans! living in Dade county, Florida; and Puerto Ricans
living in the New York City area. i
LX-28
'1 "
,,:,u
;! 'Ill
I 'jj!;!
-------
The Environmental Protection Agency sponsored the measurement of pesticide residues and metabolites in serum for a
subset of each of the three subpopulations, and in urine for a subset only of the Mexican-Americans. The following pesticides
were included in the analyses: urinary malathion metabolites (malathion mono- and dicarboxylic acid); urinary multiphenols
(2,4-dichloro phenoxy acetic acid, dicamba, para-nitrophenol, pentachlorophenol, 2,4,5-trichloro phenoxy acetic acid, 2,4,5-
tri'chlorophenol, Silvex, 3,5,6-trichloro-2-pyridinol); and serum chlorinated pesticides (aldrin; a-, P-, y-> and 5-
hexachlorocyclohexane, o,p'-DDD, pjp'-DDD, op'-DDE, pp'-DDE, op'-DDT, pp'-DDT, dieldrin, endrin, heptachlor epoxide,
heptachlor, hexachlorobenzene, mirex, oxychlordane, polychlorinated biphenyls, trans-nonachlor). Laboratory methods for the
pesticide analyses included flame photometric chromatography for the malathion urinary metabolites, and electron capture gas
chromatography for both the urinary multiphenols and chlorinated pesticides in serum. Laboratory methods and QA/QC
procedures are described in detail. .
This report presents the percent of subjects with quantifiable levels of the pesticides residues and metabolites and their
distribution percentiles for two age groups (12-17,18-74), by gender and subpopulation. For each pesticide, use, route of
exposure, and potential health effects are discussed. These data provide an estimate of the general prevalence of exposure in
these three Hispanic subpopulations, and will contribute to studies involving comparison of the general U.S. population and the
Hispanic-American subpopulations.
LX-29
-------
Hง2iฉe' Aspirin use and lung, colon, and breast cancer incidence in a p
RtK!ฑ:ฑฑl^ 11 K2Sฑ2S!!ฑHi'KSU
V&& Cflsvaytf Person ,. ,.,
Name: Dina M. Schreinemachers
Phone Number: (919) 966-5875
iffiEiMlsisCs!!^, ". " "
respective study
::p::;!;:E:;3::::;:,S^
- i, !,., .-: ^;,r,,vs,i4.,i-;>jfaf,,|^,Jp|=s.rji:Jst::*: :J^',-r; t'/s?**:;V^: iซf*'t:.!.ijฅfi5iiS!'J.|i!i|||i
Office Affiliation: Epidemiology at
(HSD), National Health & Environ:
Research Triangle Park, NC
^
d Biomarkers Branch, Human Studies Division
ijental Effects Research Laboratory (NHEERL),
- - '
1) Which NHANES did you use? ENHANES-I DNHANES-H nMHANES-in DHHANES DNHANES99+
(please check to the left of each appropriate box) ' ;
We used NHANES-I including the follow-up survey
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
Men and women, white and nonwhite, USA., ages 25-74
3) Detailed description of the goals/purpose and approach of the study:
Study the of association between aspirin use and cancer risk ]
i
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts.
Mantcl-Hasnszel analyses and proportional hazards models
i
5) If you use NHANES data without publishing documents, include a brief description of these cases, (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
and Abstract/Executive Summary
Citation: \ . \ I
Schrcinemnchcrs DM, Everson RB. Aspirin use and lung, colon, and breast! cancer incidence in a prospective study.
Epidemiology 5:138-146,1994.. j
j|3[55!;::^;i;iF;,ijii,ii ,'i ฃ i;:sjS gf!,(.rti; J i;;;:v1"!,;1 i;'4:VV'irfiI1" WfV-V'.ป!.'Sit-rat}-!!" .tiJ'V"i Pi""*'*?i
[Tor Completed Studies. Provide the Following Dates:
IHii liilillllili. j ffl ป i illh imi II i Iti; < M'i,ฅviiii > j ^^r,;n.j',^tr a n w^rfnai.;*.'1^. t-'m,ป!i:aii,^!:,w';-.^i,!.>!'.ป.,ii,hin.'d
Starting Date: 1992
Ending Date: 1994
Work in Progress, Pro^
'isMflM; "it '!f>,v
,', ft p
iCJWKJ '^-*-> '4'^:i.- rZ
Anticipated Product(s):
Tentative Completion Date(s):
Abstract:
A large body of experimental data and several recent epidemiologic studies suggest that aspirin use may decrease cancer risk.
The experimental studies found effects.at many anatomic sites, while the epid<>rniologic studies saw the greatest effect on
mortality from digestive cancers. To provide further human data, we examine!! the association between aspirin use and cancer
risk using data from the National Health and Nutrition Examination Survey I (NHANES I) and the NHANES I Epidemiologic
Followup Studies (NHEFS). Characterization of aspirin use was based on questions in the baseline interview asking whether
subjects used aspirin during the previous 30 days. Data were available from 12,668 subjects aged 25-74 at time of initial
examination for NHANES I, who were followed for an average of 12.4 years.' Among these subjects 1,257 were diagnosed with
cancer more than two years after their NHANES I exam. Incidence of several Jcancers was lower among persons who reported
aspirin use: the incidence rate ratio and 95% CI for all sites combined were 0.! 3 [0.74-0.93], lung cancer 0.68 [0.49-0.94], breast
cancer in women 0.70 [0.50-0.96], and colorectal cancer in younger men 0.35
IX-30
[0.17-0.73]. These findings were not readily
fl
1= >
-------
explained by potentially confounding factors. The data suggest an association between aspirin consumption and decreased
cancer incidence at several cancer sites.
DC-31
-------
| jijptfe: Risk Assessment for Toxic Substances Control Act, Section 4()3, Lead Standards
fer ^Mti!;!11"!;1!::1 ii, Aniio IEOI: i: asBrsiK;^^^ ss:sr:rr::te;OTfc"S^
Name: Brad Schultz and Ron Morony Office Affiliation: Office of Pollution, Prevention and Toxics (OPPT), National
Program Chemicals Div. (NPCD), Office of Prevention, Pesticides and
Toxic Substances (OPPTS), Washington, DC
Phone Number: 202-260-3896 (Brad) 202-260-0282 (Ron)
rui 1 i if H IHSIH 1 Nli n*-~ , v - i, t" r* ijtaaniw A.vasU,^ '-' "^ >',' ' '^*rg^p:xa*!
,%'scTlptronofStud} lKSfr'"~ ^Sirs.;?*'
"ซ ' ^ J i * 1KT% f ป- , fesSBiB-sKsSstt
1) Which NHANES did you use? DNHANES-I DNHANES-H IB
D NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, re
We used the nationally representative sample of children 1-5 years of age, >
age of housing, and economic status.
3) Detailed description of the goals/purpose and approach of the study:
We used the blood lead (PbB) level data to establish a baseline for lead exj
represents the level of risk to this subgroup before there are any reductions
exposure standards in dust, soil, and paint.
4) What statistical methods and models are you using? Include a brief i
appropriate, attach sample tables and/or charts.
PbB data, adjusted by NHANES sampling weights, was used as input to ris
effects. The models link PbB levels to IQ deficits as described in the two n
levels were used, as appropriate, as an approximation for the purpose of ma
5) If you use NHANES data without publishing documents, include a bi
NHANES data and/or results from published NHANES studies to supp
support setting air quality standards for ambient pollutants.):
Not applicable.
NHANES-DJ DHHANES
gion of country):
lid also looked at subgroups separately by race,
osure in children ages 1-5 years. This baseline
in exposures due to the promulgation of new lead
''" ; . r- :;
ummary of how you present the results. If
ฃ models linking blood-lead levels with health
sports below. Also, lognormal distribution of PbB
king presentations.
i i :
ief description of these cases, (e.g., use of raw
>rt setting pesticide tolerances on food, or to
^ISSISI!^^^^M^^f''^1'ly^T;^ "
ฃฎ*m W^m^x^twSumm^ _ ,;::, :i,f;i,,ii;s;;;^
Citations:
Risk Analysis to Support Standards for Lead in Paint, Dust, and Soil (E
?ull Report Available at http://www.epa.gov/lead/403risk.htm
Executive Summary available at http://www.epa.gov/lead/raexsumm.pdf
Xisk Analysis to Support Standards for Lead in Paint, Dust, and Soil: S
December 2000
Full Report Available at http://www.epa.gov/lead/403risksupp.htrti
1 !
PA747-R-97-006), June 1998.
upplemental Report (EPA 747-R-00-004),
ixccutive Summary Available at http.7/www.epa.gov/lead/rasuppexsumm.pdf
i!!"!" ' ' :"?" "r; !*: rn* 11 ,iป ซ i * i * i r jปปซปi*tt >*-%'ซ!
D|jg^o|r Completed Studies: ^ 1
jjf**^-*?^ ซsjฃ % * -f " *ปป ,jJT* f;
(jfeปt*-* ' " Sซ -" *
Starting Date: 1992 Ending Date: December 2000
it *;.,'( }.*: i ' -t:-;t!i;, ,; :i-, ,, , t',-v. ,!;::'.ป.;! ;':'* *-.;i: nw?1. r'n-r^wft
MsmsSm^^&s^m&m^^ ^^K,,*,^,, ***Ww.vi->w^"i
Anticipated Produces):
Tentative Completion Datc(s):
TXT" OO
|l^ปซ*JiJspfepe.ซ5:**4*'ฅte"V"><ป^^^^^
!" '
,i 1
-------
Abstract:
Risk Analysis to Support Standards for Lead in Paint, Dust, and Soil:
Lead poisoning in children is recognized as a major health problem in the United States. While there are many sources of lead in
the human environment, lead-based paint hazards in residential housing are considered the primary source of lead exposure for
children. To help develop a national strategy to eliminate lead-based paint hazards, the President of the United States signed into
law the Residential Lead-Based Paint Hazard Reduction Act of 1992 (42 U.S.C. 4851). This legislation included an amendment
to the Toxic Substances Control Act (Title
IV: Lead Exposure Reduction), requiring the Administrator of the U.S. Environmental Protection Agency (EPA) to enact a
variety of activities to identify and reduce environmental exposure to lead hazards. Specifically, ง403 of TSCA (15 U.S.C. 2683)
states: "... the Administrator shall promulgate regulations which shall identify, for purposes of this title and the Residential Lead-
Based Paint Hazard Reduction Act of 1992, lead-based paint hazards, lead-contaminated dust, and lead-contaminated soil."
Under ง403, the Agency is required to identify what constitutes a lead-based paint hazard (i.e., conditions that cause exposure to
lead-contaminated dust, soil, or paint that would result in adverse health effects to humans) and what constitutes lead
contamination of dust and soil (i.e., the presence of lead levels which can pose a threat of adverse health effects). In particular,
the ง403 rule to be established by the Agency will set standards for lead levels in dust and soil to determine 1) whether a
residential environment has lead-contaminated dust and soil, and 2) whether a lead-based paint hazard is present in a residential
environment.
This report presents the methods and findings of a risk analysis, which provides a scientific foundation for the regulatory
standards that the Agency will establish in response to ง403. This risk analysis consists of two parts. Part I (Chapters 2 through
5) constitutes the risk assessment, or EPA's assessment of the health risks to young children from exposures to lead-based paint
hazards, lead-contaminated dust, and lead-contaminated soil in the nation's housing. Part H (Chapter 6) constitutes an analysis of
risk management options, which includes the Agency's approach to estimating how these risks are reduced following
promulgation of the ง403 rule and illustrates use of this methodology for a broad range of example options for the ง403
standards. The objective of the risk assessment is to characterize baseline health risks to young children from specific residential
exposures to lead. The term baseline (or "pre-ง403") refers to conditions in 1997, prior to promulgating any rule in response to
ง403. The objectives of risk management are to develop and apply methodology to determine how risks are expected to be
reduced from baseline levels because of interventions conducted in response to the ง403 rule (or "post ง403"), and to develop an
approach to estimate numbers of children and housing units that would be directly impacted by the rule. Information presented
in this risk analysis will ultimately be used to consider various standards for rulemaking and as input to the Regulatory Impacts
Analysis (RIA) for the proposed rule, as well as any interim economic cost-benefit analyses.
LX-33
-------
'Itle:
The Decline in Blood Lead Levels in Children 1-5 Years of Age
PA Von tact Person
l-S
''"
: John Schwemberger
hone Number: 202-260-7195
Office Affiliation: Office of Pollution, Prevention and Toxics (OPPT), National
Program Chemicals Div.j(NPCD), Office of Prevention, Pesticides and
Toxic Substances (OPPTS), Washington, DC
1) \Vhich NHANES did you use? DNHANES-I ENHANES-H EJNHANES-HI DHHANES
D NHANES99+" (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
The study group was children 1 to 5 years old.
3) Detailed description of the goals/purpose and approach of the study;!
The goal of the study was to develop a graph showing the decline in children's blood lead (PbB) levels over time and to
compare the graph with time points when federal actions designed to lower exposure to lead went into effect.
4) What statistical methods and models are you using? Include a briefjsummary of how you present the results. If
appropriate, attach sample tables and/or charts. j
Geometric mean (ng/dl) of PbB calculated for children ages 1-5 years either for a year or for a three year period in the case of
NHANES m.
5) If you use NHANES data without publishing documents, include a I rief description of these cases. (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
Not applicable.
Citations; ,
Goldman, LR, Linking Research and Policy to Ensure Children's Environmental Health, Environmental Health
Perspectives 106 (Supplements), 1998, pp. 857-862. Available at:
http-y/ehpnetl.niehs.nih.gov/members/1998/Suppl-3/857-862goldman/goldman-Ml.htnil
Office of Pollution Prevention and Toxics Program Activities for Fiscal Years 1998 and 1999, (EPA 745-K-99-003),
December 1999, p. 19. Available at http://www.epa.gov/oppt/ar98-99/opptreport.pdf.
,, ,,.
for Completed Studies:
1 M
Starting Date: 1997
Ending Date: 1998 (Goldroan Article), 1999 (OPPT Report)
;.,. .. . ,. . ,,..,
information for Work in Progress:
Anticipated Productฎ: Graph will be updated when NHANES 2000 and 2001 blood lead statistics are available.
Tentative Completion Date(s): April 2001 and April 2002 . '
Abstract from-Goldman et al, 1998: . i .' \
Prevention of Childhood Lead Exposure ! . -,
Blood lead levels previously considered safe are now known to be associated with adverse health effects in children, bmce
the level of concern for blood lead has been revised downward, from 60 to 10 ung/dl. Research with more sensitive measures and
LX-34
it:,LI,
I1 .' t:' i-ilii i I'1 iff I ,'iii" I
-------
.
IS-
IS-
14-
12-
10.
8-
o.
T
1371 1315 1SB
t 1 1
1 2 S
i 11-
1S3?
i I i i
1991
ฅear
1395
1 1
56
IX-35 .
-------
f|i! I VmabmtymPubertyMeasuresmCrdldreninResponse to Classes of Chemicals-NHANES-in
tpHllllliiiiliTilli'iiiiiil'1'^ _ ; . - ^^^^^_^^^_^^ M. 111111 ...HIM ^ i |A , ,_"^^^^?T^^:^^^^:'': '^ .: r,. ,,. ,. -. -i.,... .a- M- _ . - -, -i: ^'^ii- ^,-" T|- 7- '-'-"-'"':-; "j^!: ;""-'"
"- - rrr* i!!|*!|ii!i!s:|^
Name: Sherry Selevan
Phone Number: 202-564-3403
Office Affiliation: National Center for Enyironmental Assessment (NCEA-W),
Office of Research and Development (ORD), Washington; DC .
1) Which NHANES did you use? DNHANES-I D NHANES-II H NHANES-IH DHHANES D
NHANES99+ (check to the left of each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region-of country):
8-18 year olds; male and female; non-Hispanic whites, non-fiispanic blacks, Mexican^Americans; all regions
i '
3) Detailed description of the goals/purpose and approach of thซ study:
To provide an overview of the variability in response of children to Exposure to certain classes of chemicals using
NKANES-m data. The study examines the effects of these classes of chemicals on timing of puberty onset in
adolescent males and females, and the variability of response based ;on age, race, sex, and socioeconomic status
(SES) Measures of timing of puberty available in NHANES-HI include Tanner stages for pubic hair
development and breast development, and age at onset of menarche Assessment of the variability in these
responses includes evaluating the proportion of individuals who have the effect or event (for example a particular
SLer stege) for a particular exposure, the average age of each eveiit, and the impact of the identified exposures
ซ*. average ages for these events. The exposure information available in NHANES-m falls into 2 primary
categories: those measured clinically (e.g., serum levels of lead, cacbrium, cotmine), and information recalled by
subjects via questionnaire (e.g., active and passive smoking).
I '
4) What statistical methods and models are you using? Include a brief summary of how you present the
results. If appropriate, attach sample tables and/or charts.
All analyses include the sampling weights addressing the complex ^ultilevel design of NHANES-IH, through the
useofSUDAAN. : . ,
Statistical methods include descriptive statistics (means for measurements and ages, proportions for dichotomous
data such as "ever smoked, yes/no") and significance testing for differences in the descriptive statistics between
demographic groups. !
Statistical models allowing for consideration of related influences (Such as body weight) or simultaneous,
confounding exposures (for which data are available in NHANES-ffl) include proportional hazards analysis for
time-related outcomes, and ordinal logistic regression for categorical outcomes (i.e., Tanner- stage attainment at
specific ages). j
5) If you use NHANES data without publishing documents, include a brief description of these cases.
fe g use of raw NHANES data and/or results from published ]raANES studies to support setting pesticide
tolerances on food, or to support setting air quality standards iFor ambient pollutants.):
Not applicable. \ __
K-36
-------
Igitatfon and Abstjr%ct/Exejeutiv& Summary"
Citation:
|P.ates for Completed Studies:
Starting Date:
Ending Date:
formation for Work in "Progress?
Anticipated Product(s): Currenay anticipate at least 3 reports:
1. Environment and Puberty in Girls: Examination of association of pubertal timing with blood lead and passive
MeasS! (A?^lSati0nal ^^ ^ Nutritiฐn Examinati Survey m - report to satisfy Annual Performance
2. An examination of puberty in girls and the remaining exposures measured in NHANES-IH
3. At least one report studying the above associations in boys aged 8-18 years
Tentative Completion Date(s):
1. December 2001, currently under NCEA management clearance process
2. FY2002
3. FY2003
IX-37
-------
Blood Lead Concentrations of U.S. Adult Females: Summary Statistics from Phases 1 and 2 of the National
Health and Nutrition Evaluation Survey ( NHANES m)
L Con! act Person
........
' .
.'"->-v-~-*:i ;n":- ^-^ ........ * ..... -,
Nam MarcStifelman '' Office Affiliation: Office of Environmental ^ss^
Phone Number: (206)553-6979
ft 1C::, 1 :. ::::; ':;'" ", :>;,,;: V,., 'S*i , i'-,,iฃ -t iv'i**y-fe 4 iซif:. sw J-> 4f'
fl|g|cription of Study , .''-'' :.:'" '''.-'"'!":
DNHANES-I DNHANES-H E;NHANES-m dHHANES DNHANES99+
JL I TT UtVU AT A**"** i*Jfcrf *-*- j *~~* A ! i :
(check to the left of each appropriate box) j
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
Females, 17-45 years old analyzed together and by race/ethnicity and region of country.
3) Detailed description of the goals/purpose and approach of the study:
based on NHANES HI, Phase 1 only; this analysis used Phases 1 and 2 of NHANES m.
SesensiUvity of me parameter estimates to the treatment of non-detects ,s also addressed.
remediation goals calculated firom.theALM are presented in a table.
Estimates of the GM and GSD for Mexican-Americans in the northeast relgion were highly uncertain due to the small sample
size f24^ *
support setting air quality standards for ambient pollutants.): ^_ _ _ ^^
*- ...... " ......
ar , . . - .- ;
5T*Si ....... : ....... '*fg.w:. ......... 9*- ...... " ......
itutioi; and Abstract/Executive aummary
S,,is.ii .Et;,s; >i ;*is,:;i.
iitjllii ; it }; ; ;;; ajji. ; ;;jt:3 li1,.*, -!!!;
^ ' i for Completed Studies:
:;g^aq:ate^jyg^
Starting Date:
Ending Date:
Anaeipated Prlduct(s^ Technical Support Document (httpV/www.epaJgov/r.rerfimH/pro^nis/lead/prods.htm)
Tentative Completion Date(s): November, 2001
K-38
-------
||le|g^ Sociodemographic Data For Use In Identifying Potentially Highly Exposed Children's Populations
Name: Amina Wilkins
Phone Number: 202564-3256
Office Affiliation: National Center for Environmental Assessment (NCEA), Office of
Research and Development (ORD), Washington, DC
* it '
sDjescription of Study (
1) Which NHANES did yon use? DNHANES-I DNHANES-n ENHANES-m DHHANES DNHANES99+
(check to the left of 'each appropriate box)
2) Characteristics of the study group (e.g., age, gender, race/ethnicity, region of country):
Children and adolescents ages 1-19 years of age from Phase 2 of NHANES-m, which was conducted 1991-1994. We looked
at many variables, including body weight, blood lead levels (PbB), race/ethnicity, housing characteristics, and various
demographic and socioeconomic characteristics.
3) Detailed description of the goals/purpose and approach of the study:
Derive the distribution of overweight children and adolescents (ages 6-17 years old) by gender, race, and age. Also derive the
distribution of children and adolescents (1-19 years old) with PbB levels >10|ig/dl by age, gender, race/ethnicity, housing
characteristics, income and urban-rural status. These data can be used for risk and exposure assessments and will be
presented in tabular form as part of a guidance document, Sociodemographic Data For Use In Identifying Potentially Highly
Exposed Children's Populations. This document will contain numerous tables of data on various socioeconomic and
exposure-related variables relevant to children and adolescents and derived from a variety of sources including NHANES.
4) What statistical methods and models are you using? Include a brief summary of how you present the results. If
appropriate, attach sample tables and/or charts.
Data are presented as simple distributions of overweight individuals and individuals with PbB levels >10ng/dl by age, race,
etc.
5) If you use NHANES data without publishing documents, include a brief description of these cases. (e.g., use of raw
NHANES data and/or results from published NHANES studies to support setting pesticide tolerances on food, or to
support setting air quality standards for ambient pollutants.):
^Citation and Abstract/Executive Summary.
**
Citation:
Pates for Completed Studies:
Starting Date:
Ending Date:
Information for Work in Progress:
Anticipated Product(s): EPA guidance document, Sociodemographic Data Used For Identifying Potentially Highly Exposed
Children's Populations, and possible website/weblinks.
Tentative Completion Date(s): 2005
Abstract:
The specific goals of the document Sociodemographic Data for Use in Identifying Potentially Highly Exposed Children's
Populations will be to (1) help assessors identify potentially highly exposed children's populations and (2) help assessors
estimate the size of these populations. It will provide information, when available, on the number of individuals, or the percent
of the general population, associated with dietary preferences, cultural practices, geographic location and setting (i.e., urban vs.
rural), and other activities that target populations and individuals as being possibly highly exposed. Data will be presented as
they appear in the original studies/reports. Research on children's exposures/risks due to environmental contaminants has been
IX-39
-------
rninimal prior to the mid-1990s (before the Executive Order on the Protection of children from Environmental Risks and safety
Risks, was signed on April 21,1997). Thus the literature summaries to be provided will not be all inclusive, but will be meant to
provide the reader with a general overview of available children's population data. In most cases, data will be from government
publications, peer-reviewed literature, and trade associations. j
The Sociodemographic Data document is intended to be used in conjunction with the Child-Specific Exposure Factors
Handbook currently being developed. The Handbook will provide statistical data on human characteristics and behaviors (e.g.,
ingestion rates of foods, activity duration and frequency, soil ingestion rates;jbody weight, skin surface area) used in assessing
exposure. Where possible, data for specific ages, gender, and race/ethnic groups will be presented. The procedure for using these
two documents in combination will be as follows: ;
Use the Sociodemographic Data document to help determine jif potentially highly exposed populations may exist
in the area of interest and to estimate the number of children of concern. I
Identify the suspected potentially highly exposed populations^ then use the Child-Specific Exposure Factors
Handbook to select the exposure factor values specific to the population of interest. These exposure factor values would then be
combined with site-specific information on environmental contaminant concentrations to estimate exposure levels.
i
,,
f"!i {
f-vt
I "
fr ,ฃ,[
1,1!
rx-40
r; ;
*"
IK:
ii H
-------
REFERENCES
NHANES Consortium Meeting, 2002. NCHS conducts a Consortium meeting once or twice a
year for all tne federal agencies to update them on the status of the survey in the field. This
meeting was held at NCHS headquarters in Hyattsville, MD on February 28.
NCHS (National Center for Health Statistics), 1973. Plan and Operation of the Health and
Nutrition Examination Survey, United States, 1971-1973. Hyattsville, MD Vital & Health
Statistics, 1(1 Oa).
NCHS, 1975. Data Evaluations and Research Methods: Distribution of Variance and Properties
Estimators for Complex Multistage Probability Samples An Empirical Distribution Hyattsville
MD. Vital & Health Statistics 2(65).
NCHS, 1977. Plan and Operation of the Health and Nutrition Examination Survey United
States, 1971-1973. Hyattsville, MD. Vital & Health Statistics l(10b).
NCHS, 1978. Plan and Operation of the NHANES I Augmentation Survey of Adults 25-74
Years, United States, 1974-1975. Hyattsville, MD. Vital & Health Statistics 1(14).
NCHS, 1981. Plan and Operation of the Second National Health and Nutrition Examination
Survey, 1976-80. Hyattsville, MD. Vital & Health Statistics 1(15).
NCHS, 1982. A Statistical Methodology for Analyzing Data from a Complex Survey: The First
National Health and Nutrition Examination Survey. Hyattsville, MD. Vital & Health Statistics
2(92).
NCHS, 1985. Plan and Operation of the Hispanic Health and Nutrition Examination Survey
1982-1984. Hyattsville, MD. Vital & Health Statistics 1(19).
NCHS, 1987. Plan and Operation of 'the NHANES1Epidemiologic Followup Study 1982-84
Hyattsville, MD. Vital & Health Statistics 1(22).
NCHS, 1990. Plan and Operation of 'the NHANES I Epidemiologic Followup Study, 1986
Hyattsville, MD. Vital & Health Statistics 1(25).
NCHS, 1992a. Plan and Operation of the NHANES I Epidemiologic Followup Study, 1987.
Hyattsville, MD. Vital & Health Statistics 1(27).
NCHS, 1992b. Sample Design: Third National Health and Nutrition Examination Survey.
Hyattsville, MD. Vital and Health Statistics 2(113).
R-l
-------
NCHS, 1994a. National Health and Nutrition Examinatic n Survey HI: Accounting for Item
NonrespanseBias.Kyatt.svilie,MD. |
NCHS, 1994b. Plan and Operation of the Third National Health and Nutrition Examination
Survey, 1988-94. Hyattsville, MD. Vital Health Statistics 1(32).
NCHS, 1996a. Analytic and Reporting Guidelines: Third National Health and Nutrition
Examination Survey, 1988-94. Hyattsville, MD. j
NCHS, 1996b. Analytic and Reporting Guidelines: Third. National Health and Nutrition
Examination Survey, 1988-94. Appendix B. Hyattsville, MD.
NCHS, 1996c. National Health and Nutrition Examination Survey III, Weighting and
Estimation Methodology, Executive Summary. Hyattsville, MD
.
NCHS 1996d. The National Health and Nutrition Examination Surveys, A Selective _
Bibliography, 1980-96. October. This information was|Kstributed on floppy discs to NHANES
Consortium members by NCHS.
'.t, !! .
NCHS, 1997. Plan and Operation of the NHANES I Epidemiologic Followup Study, 1992.
Hyattsville, MD. Vital & Health Statistics 1(35).
NCHS, 1999a. Plan and Operation of the NHANES II Mortality Study, 1992. Hyattsville, MD.
Vital & Health Statistics 1(38).
i , i
NCHS, 1999b. NHANES-A Selective Bibliography 199\-1999. Hyattsville, MD.
.
NCHS, 2001. Laboratory Procedures Manual. April. I^yattsville, MD.
.
NCHS, 2001a. MECInterviewers Procedures Manual. Revised January. Hyattsville, MD.
.
NCHS, 2001b. MEC Interviewers Procedures Manual, Part W. Hyattsville, MD.
.
U S EPA (U.S. Environmental Protection Agency), 198|6. Air Quality Criteria for Lead.
Research Triangle Park, North Carolina:, Office of Health and Environmental Assessment; EPA
report no. EPA-600/8-83/028aF-dF. j.
;: !\
:r ; t
ฅ ' 11
R-2
-------
U.S DHHS (U.S. Department of Health and Human Services), 1988. The nature and 'extent of
lead poisoning ^children in the United States: a report to Congress. Public Health Service
Agency for Toxic Substances and Disease Registry. Atlanta.
R-3
-------
, 1-i i:.ป:4 Ml I
' :?' Hi hi j
.r i p! 'i
., ii i i
-------
-------
-------
-------
United States
Environmental Protection
Agency
National Center for
Environmental Assessment
Washington, DC 20460
Official Business
Penalty for Private Use
$300
Please make all necessary changes on i:he below label,
detach or copy, and return to the address in the upper
left-hand comer. :
If you do not wish to receive these reports CHECK HEREd ;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
EPA/600/R-02/044
March 2003
------- |