United States      Science Advisory     EPA-SAB-IHEC-ADV-99-004
      Environmental      Board           February 1 999
&EPA AN SAB ADVISORY: THE
      NATIONAL HUMAN EXPOSURE
      ASSESSMENT SURVEY (NHEX/
      PILOT STUDIES
      PREPARED BY THE INTEGRATED HUM
      EXPOSURE COMMITTEE (IHEC) OF THE
      SCIENCE ADVISORY BOARD (SAB)

-------
                                   February 9, 1999

EPA-SAB-fflEC-ADV-99-004

Honorable Carol M. Browner
Administrator
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460

       Subject:      Advisory on the National Human Exposure Assessment Survey
                    (NHEXAS) Pilot Studies

Dear Ms. Browner:

       The Integrated Human Exposure Committee (IHEC) of the Science Advisory Board met
on September 28-29, 1998 to conduct an advisory on the National Human Exposure Assessment
Survey (NHEXAS) and to be briefed on the National Health and Human Nutrition Examination
Survey (NHANES) and its value to improving understanding of human exposure. The NHEXAS
pilot studies are funded and managed by the Environmental Protection Agency, Office of
Research and Development (ORD), National Center for Environmental Assessment (NCEA),
National Exposure Research Laboratory (NERL)  and the National Human and Ecosystem Effects
Laboratory  (NHEERL). NHANES is funded and managed primarily by the Centers for Disease
Control and Prevention (CDC). Since most of the NHEXAS field studies have recently been
completed, the NHEXAS results have not been fully analyzed or published.  Therefore, this
advisory report is based primarily on the preliminary NHEXAS data that were presented during
the meeting.

       At its November Strategic Planning Retreat, the Executive Committee of the Science
Advisory Board concluded that new directions taken by EPA demand new directions  for the SAB
if the Board is going to continue to make a positive difference in the way that science is developed
and used in environmental decision-making.  In particular, the Executive Committee decided that
the SAB needs to spend much more of its energy  providing strategic, forward-looking advice.
Consequently, the IHEC identified NHEXAS as the topic for its  strategic activity for Fiscal Year
1998 because it was designed to provide critical information about multipathway, multimedia
population exposure distributions to chemical classes.  These data should be very useful in
determining the most effective strategies to reduce public risk to hazardous environmental
chemicals.  The focus of this advisory is to provide the Agency with forward-looking  advice on
the analysis and future uses of NHEXAS based on the preliminary data.  Due to the wide range of
possibilities for analyses of the NHEXAS databases, the Committee plans to have a follow-up
advisory, towards the end of 1999, on the Agency's plans for the future analyses of the NHEXAS
data over the next 5 years.

-------
       In the three field studies, exposures in three study regions were measured using data on
pollutant concentrations in environmental and biological media and estimates of the frequency and
durations of exposure-related human activities.  These studies included: a) measurements of
pollutant concentrations in air, water, soil, dust, food, blood, and urine as well as on surfaces and
human skin; b) measurements of direct exposure using personal exposure monitors; and c)
estimations of human activity patterns using a series of questionnaires and diaries.  These data are
being used to estimate human exposures among the sampled populations and to test a series of
hypotheses related to these exposures. The target chemicals include: volatile organic compounds
(VOCs, except for the Maryland study), pesticides, metals and polyaromatic hydrocarbons
(PAHs).

       The Committee was charged to respond to the following Charge questions:

       a)    What are the strengths and weaknesses of multimedia, multipathway measurements
             of exposure as represented by the NHEXAS program, insofar as they can be
             defined at this point?

       b)    Are the ongoing and planned analyses appropriate and likely to further the goals of
             NHEXAS? At the level of each consortia?  At the level of NHEXAS?

       c)    What actions would be likely to increase the utility of the information from
             NHEXAS? In the  near-term? In the longer term?

       d)    What follow-up studies would be most useful in the near term, considering that
             key NHEXAS analyses will not be completed for a year?  What is the appropriate
             balance between large population surveys and more targeted follow-up studies?

       e)    What additional IHEC advice is offered for  strengthening the immediate and long-
             term utility of NHEXAS and studies like it?

       The IHEC commends the Agency for conducting the NHEXAS pilot study. The IHEC
found NHEXAS to be an excellent project that will provide vital information to the EPA for
addressing the many multimedia pollutant issues that the Agency is confronted with on a daily
basis. When fully analyzed, the results of the NHEXAS pilot study should provide a basis for the
design and implementation of an effective surveillance program for multimedia pollutants.
Therefore, the NHEXAS pilot study is highly relevant to the Government Performance and
Results Act (GPRA) and SAB's "Integrated Environmental Decision-making Framework" which
emphasize performance evaluation as a key to effective environmental health protection.
NHEXAS is also highly relevant to EPA's Cumulative Exposure Project (CEP) because it
provides measurement data against which the CEP models  can be evaluated.

       The Committee was particularly concerned about the limited resources  allocated to the
analysis of the NHEXAS data and the lack of a strategic plan for follow-up studies. If NHEXAS

-------
were to stop at this time, the considerable efforts and expenditures incurred during the last five
years would be of limited utility to the Agency.  Therefore, the Committee strongly encourages
the EPA to pursue the completion of the study in an expeditious manner.  It is important that the
costs of this program be presented within a frame of future savings as a result of improved public
health resulting from better targeted, more effective, and less costly regulatory efforts. In
addition, the Committee highly recommends that findings that are of considerable importance to
the Agency in the current risk management efforts be identified and given high priority for analysis
and publication.  However, the data should be presented in terms of the associated uncertainty
and limitations to add to the transparency of the information.

       The Committee also cites a weakness of the NHEXAS pilot studies and provides
recommendations for increasing the utility of the NHEXAS data. The weakness cited by the
IHEC was the criteria for the selection of the target chemicals which were measured analytes.
The Committee's recommendations to the EPA for increasing the utility of the NHEXAS data
include: a) the development of a strategic plan for analyzing the data; b) publicizing the NHEXAS
framework (by informing the public through various media such as an EPA publication that is
available in hard copy and on the Internet); c) analyzing the flexibility of NHEXAS to study
special populations such as minorities and sensitive populations; d) linking the exposure data from
NHEXAS with biological markers from NHANES where possible and e) improving the
communication between the NHEXAS investigators and state and local health officials.

       The Committee appreciates the opportunity to provide advice to the Agency on the
preliminary NHEXAS data. The Committee looks forward to receiving a written  response from
the Assistant Administrator for Research and Development (ORD), and to conducting a follow-up
advisory on the Agency's plans for the future analysis of the NHEXAS data.

                                  Sincerely,
                                         /signed/
                                  Dr. Joan M. Daisey, Chair
                                  Science Advisory Board
                                         /signed/
                                  Dr. Henry A. Anderson, Chair
                                  Integrated Human Exposure Committee
                                  Science Advisory Board

-------
                                       NOTICE
       This report has been written as a part of the activities of the Science Advisory Board, a
public advisory group providing extramural scientific information and advice to the Administrator
and other officials of the Environmental Protection Agency.  The Board is structured to provide a
balanced, expert assessment of scientific matters related to problems facing the Agency.  This
report has not been reviewed for approval by the Agency; hence, the comments of this report do
not necessarily represent the views and policies of the Environmental Protection Agency or of
other Federal agencies.  Any mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

-------
                                     ABSTRACT
       On September 29-30, 1998, the EPA Science Advisory Board (SAB) Integrated Human
Exposure Committee (IHEC) reviewed the preliminary data on the National Human Exposure
Assessment Survey (NHEXAS) pilot studies in Durham, North Carolina.  The NHEXAS pilot
studies were designed to provide critical information about multipathway, multimedia population
exposure distribution for selected chemical classes.

       IHEC found NHEXAS to be an excellent project that has significant promise for
improving public health in a cost-effective manner.  The Committee also found NHEXAS to be
outstanding in both design and implementation. When completed, NHEXAS should greatly
improve understanding of human exposure to selected pollutants.  This, in turn, would be helpful
in determining the most effective strategies to reduce the public's risk to hazardous environmental
chemicals. (The uncertainty and limitations associated with the data should be presented  along
with the data in order to add to the transparency of the information.) Therefore, the  Committee
strongly encourages the EPA to pursue the completion of the study results in an expeditious
manner.

       To increase the utility of NHEXAS, the Committee recommends that the EPA:  a) develop
a strategic plan for analyzing the data; b) publicize the NHEXAS framework by informing the
public through various media such as an EPA publication that is available in hardcopy and on the
Internet; c) evaluate the flexibility of NHEXAS to study special populations such as minorities
and sensitive populations; d) link the exposure data from NHEXAS with biological markers from
NHANES where possible; and e) improve the communication between the NHEXAS
investigators and state and local health officials.

       The Committee was particularly concerned about the limited resources allocated to the
analysis of the NHEXAS data and the lack of a strategic plan for follow-up studies. If  NHEXAS
were to stop at this time, the considerable efforts and expenditures  incurred during the last five
years would be of limited utility  to the Agency. It is important, therefore, that the costs of this
program be presented within a frame of future  savings as a result of improved public health
resulting from better targeted, more effective, and less costly regulatory efforts.  The Committee
was  also concerned about the selection of the chemicals that were measured in NHEXAS.
Several recommendations are provided for the planned analyses of the data, actions for the
increased utility of the data, and follow-up  studies in both the near  term and in the future.

Keywords:   National Human Exposure Assessment Survey (NHEXAS); National  Health and
             Human Nutrition  Examination Survey (NHANES) strategic plan; biological
             markers; multimedia; multi-chemical; multi-pathway

-------
                           U.S. Environmental Protection Agency
                                  Science Advisory Board
                          Integrated Human Exposure Committee
               National Human Exposure Assessment Survey Advisory Panel

Chair
Dr. Henry A. Anderson, Chief Medical Officer, Wisconsin Bureau of Public Health, Madison, WI

Past Chair
Dr. Joan M. Daisey, Department Head, Indoor Environment Department, Lawrence Berkeley National
       Laboratory, Berkeley, CA

Members
Dr. Paul Bailey, Toxicology Consultant, Mobil Business Resources Corporation, Mobil Business
       Resource Corp., Product Stewardship & Toxicology, Paulsboro, NJ

Dr. Robert A. Harley, Associate Professor, Department of Civil and Environmental Engineering,
       University of California Berkeley, Berkeley, CA

Dr. Kai-Shen Liu, Epidemiologist, California Department of Health Services, Berkeley, CA

Dr. Thomas E. McKone, Staff Scientist/Adjunct Professor, School of Public Health, University of
       California, Berkeley, CA

Dr. Maria Morandi, Assistant Professor, University of Texas Health Science Center at Houston, School
       of Public Health, Houston,  TX

Dr. Jerome O. Nriagu, Professor, The University of Michigan, School of Public Health, Department of
       Environmental and Industrial Health, Ann Arbor, MI

Dr. Barbara Petersen, President, Novigen Sciences, Inc., Washington, DC

Dr. Charles Weschler, Senior Scientist, Bell Communication Research, Red Bank, NJ

Consultants
Dr. Lovell Jones, Director, Experimental Gynecology-Endocrinology, Depart, of Gynecologic Oncology;
       and Professor of Gynecologic Oncology; M.D. Anderson Cancer Center, Houston, TX

Ms. Leyla E. McCurdy, Director, Indoor Air Programs, American Lung Association, Washington, DC

Science Advisory Board Staff
Ms. Roslyn Edson, Designated Federal Officer, U.S. Environmental Protection Agency, Science Advisory
       Board (1400), 401 M Street, SW, Washington, DC 20460

Ms. Wanda Fields, Management Assistant, U.S. Environmental Protection Agency, Science Advisory
       Board (1400), 401 M Street, S.W., Washington, DC 20460
                                             in

-------
                            TABLE OF CONTENTS
1.  EXECUTIVE SUMMARY  	1

2. INTRODUCTION	3
      2.1 Introduction 	3
      2.2 Charge to the Committee	4

3. RESPONSE TO THE CHARGE	6
      3.1 Charge Question 1: Strengths and Weaknesses  	6
             3.1.1 Strengths  	6
             3.1.2 Weaknesses  	6
      3.2 Charge Question 2: Ongoing and Planned Analyses	7
             3.2.1 Analyses at the Level of Each Consortia	8
             3.2.2 Analysis at the Level of NHEXAS  	8
      3.3 Charge Question 3: Increasing the Utility of NHEXAS  	11
             3.3.1 In the Near Term 	12
             3.3.2 In the Longer Term	14
      3.4 Charge Question 4: Follow-up Studies 	15
             3.4.1 Follow-up Studies in the Near Term (over the next 12 months)	15
             3.4.2 Balance between Large Population  Surveys and More Targeted
                   Follow-up Studies	15
      3.5 Charge Question 5: Additional Advice	17
             3.5.1 Advice for Strengthening the Utility of NHEXAS in the Near Term .... 17
             3.5.2 Advice for Strengthening the Utility of NHEXAS in the Long-Term   ... 17

4.  CONCLUSIONS  	20
      4.1 Strengths and Weaknesses	20
      4.2 Ongoing and Planned Analysis	20
      4.3 Increasing the Utility of NHEXAS  	21
      4.4 Follow-up  Studies	21
      4.5 Balance between Large Population Surveys and more Targeted
             Follow-up Studies  	21
      4.6 Additional  Advice	21

REFERENCES	R-l

APPENDIX A -  ACRONYMS AND ABBREVIATIONS	A-l
                                         IV

-------
                           1. EXECUTIVE SUMMARY
       On September 28-29, 1998, the Integrated Human Exposure Committee (IHEC) met in
Durham, North Carolina to conduct an advisory on the National Human Exposure Assessment
Survey (NHEXAS). At this meeting, the IHEC was also briefed on the National Health and
Human Nutrition Examination Survey (NHANES) and its value to improving understanding of
human exposure by a representative of the Centers for Disease Control and Prevention.  Given the
early stage of the NHEXAS data analysis, the purpose of the first advisory was to provide advice
to the Agency on the strengths and weaknesses of the NHEXAS pilot study data and on the
ongoing and future analyses and uses of the data.

       The Committee selected NHEXAS as the topic for its strategic activity for Fiscal Year
1998 following the SAB Executive Committee's decision to spend much more of its energy
providing strategic, forward-looking advice to the EPA.  NHEXAS was chosen because it was
designed to provide critical information about multipathway, mulltimedia population exposure
distributions for chemical classes and will be very useful in determining the most effective
strategies to reduce the public's risk to hazardous environmental chemicals.

       The NHEXAS data has not been fully analyzed, and therefore, has not yet been published.
Therefore, the background material did not consist of published studies on the NHEXAS data
analysis.  Instead, the Committee was provided with articles that described the history of
NHEXAS; the strategies, goals, objectives, and the hypotheses to be tested; and the study design,
including the study area, population and sampling methods, and data evaluation and analytical
methods. During the meeting, the NHEXAS Principal Investigators gave presentations on the
preliminary data. The Committee's advice is based primarily on those data.

       The IHEC found NHEXAS to be an excellent project that has significant promise for
improving public health in a cost-effective manner. When completed, NHEXAS should provide a
greatly improved understanding of human exposure to selected pollutants.  This, in turn, will be
helpful in determining the most effective strategies to reduce the public's risk to hazardous
environmental chemicals.  The Committee also found NHEXAS to be outstanding in both design
and implementation. Therefore, the Committee strongly encourages the EPA to pursue the
completion of the study in an expeditious manner.

       The Committee was particularly concerned about the limited resources allocated to the
analysis of the NHEXAS data and the lack of a strategic plan for follow-up studies. If NHEXAS
were to stop at this time, the considerable efforts and expenditures incurred during the last five
years would be of limited utility to the Agency. It is important, therefore, that the costs of this
program be presented within a frame of future savings as a result of improved public health
resulting from better targeted, more effective, and less costly regulatory efforts.

-------
       To increase the utility of NHEXAS, the IHEC recommends that the EPA: a) develop a
strategic plan for analyzing the data; b) publicize the NHEXAS framework by informing the
public through various media such as an EPA publication that is available in hardcopy and on the
Internet; c) evaluate the flexibility of NHEXAS to study special populations such as minorities
and sensitive populations; d) link the exposure data from NHEXAS with biological markers from
NHANES, where possible; and e) improve the communication between the NHEXAS researchers
and the state and local health officials.

       The Committee also recommended several additional actions for the Agency to increase
the utility of the NHEXAS data. Those  actions include:

       a)      analyzing the questionnaire data (including analysis of the responses on the
              questionnaires and analysis of the questionnaire itself to determine what worked
              and what did not work),
       b)      analyzing the exposure data,
       c)      identifying the most cost effective approaches to exposure assessment based on the
              NHEXAS data,
       d)      conducting a prototypical analysis of exposure and assessments of intervention
              strategies,
       e)      archiving the information in a manner such that the data can be retrieved years
              from now for further analysis,
       f)      integrating the databases  from the three pilot studies, where possible, and
       g)      developing a feasibility study for a national human exposure survey.

       For longer term activities, the Committee recommends that the Agency: a) place emphasis
on developing physically-based quantitative models that can be used to evaluate various possible
risk management options; b) consider using advanced measurement technologies such as
advanced sensors and sampling technologies, and robotics for multiple analyses, and advanced
software for analysis of chromatographic outputs;  and c) applying the NHEXAS pilot data,  after
the analyses have been completed, to on-going programs within the Agency such as validation of
multimedia transport models.

       The Committee looks forward to the Agency's response and to the follow-up advisory on
the Agency's plans for future analyses of the data.

-------
                                2. INTRODUCTION

2.1 Introduction

       In order to assess the risks posed by chemical pollutants in the environment, the EPA must
be able to estimate the number of people exposed to these chemicals as well as the intensity of
exposure. In the past, most studies have focused on exposures to one chemical at a time by one
route of exposure. While such studies are important, they do not reflect actual patterns
(distributions) of human exposure to chemicals in the environment because people are exposed to
multiple chemicals from various routes and media.  The ultimate goal of the National Human
Exposure Assessment Survey (NHEXAS) is to document status and trends of national
distributions of human exposure to potentially high risk chemicals to improve the accuracy of
exposure (and risk) assessments, and to evaluate whether exposure (risk) is decreasing or
increasing over time with the application of risk management steps by taking into account
exposure to multiple chemicals from various routes and media.  In doing so, NHEXAS will help
to identify those pathways and routes that are responsible for the greatest exposure and will
therefore provide direction for decisions on the most effective strategies to reduce  risks.

       The five major objectives for NHEXAS are:

       a)     to evaluate the feasibility of NHEXAS concepts, methods, and approaches for the
              conduct of future population-based studies;
       b)     to evaluate the utility of the NHEXAS data for improved risk assessment and
              management decisions;
       c)     to test the hypothesis that the distribution of exposure given by modeling and
              extant data does not differ from the measurement-based distribution of exposure;
       d)     to define the distribution of total human exposures for a relatively large geographic
              area by:
              (1)     identifying the upper 10th percentile of distribution,
              (2)     providing baseline data for later comparisons,
              (3)     evaluating pathways,
              (4)     identifying predictors, and
              (5)     comparing exposure data with biomarkers of exposure; and
       e)     to stimulate exposure research and forge strong working relationships between
              government and non-government scientists.

       Phase I of NHEXAS consists of pilot studies designed to evaluate the feasibility of
evaluating total human exposure to multiple chemicals on a community and regional  scale. The
pilot studies focus on the exposure of people to environmental pollutants during their daily lives.
Hundreds of volunteer participants were randomly selected from several areas of the country to
obtain a population-based probability sample. The levels of target chemicals to which participants
were exposed were measured in the air they breathe, in the food and beverages they consume, in
the water they drink, and in the soil and dust around their homes.  In addition, chemicals or their

-------
metabolites were measured in biological samples (including blood and urine) provided by the
participants. Finally, participants completed questionnaires which will be used to help identify
possible sources of exposure to chemicals and to characterize major activity patterns and
conditions of the home environment.

       The EPA Office of Research and Development entered into three Cooperative Agreements
with three different Consortia to conduct these field studies.  The Consortia include: 1) a
consortium from the University of Arizona, Battelle Memorial Institute, and the Illinois Institute
of Technology; 2) a consortium from the Research Triangle Institute (RTI) and the Environmental
and Occupational Health Sciences Institute (EOHSI); and 3) a consortium from Harvard
University, Johns Hopkins University, Emory University, Southwest Research Institute, and
Westat. In addition, the EPA Office of Research and Development has entered into Inter-agency
Agreements with: 1) the Centers for Disease Control and Prevention (CDC) to analyze blood and
urine samples; 2) the Food and Drug Administration (FDA) to analyze food and beverage
samples; 3) the National Institute for Standards and Technology (NIST) to provide quality
assurance support; and 4) the General  Services Administration (GSA) to provide data
management support.

       Sample collection began in mid-1995 and was completed for  all of the projects in late
1997. Almost all of the planned sample analyses were completed late in 1998. Statistical analyses
of the data were ongoing in 1997 and 1998 and funded for additional analyses in 1999. Initial
publications by the Principal Investigators are expected beginning in early 1999. Analyses of the
combined databases from the three consortia can provide considerable additional information, but
will require a strategic plan for this data analysis due to the wide range of possibilities for
analyses. The Integrated Human Exposure Committee plans to conduct a follow-up advisory, late
in 1999, on the Agency's  plans for a strategy for the data analysis. In 2000, the actual NHEXAS
databases for the pilot studies should become available to the public in hardcopy and on the
Internet.

2.2 Charge to the Committee

       On September 29-30, the IHEC met in Durham, North Carolina to conduct an advisory on
the National Human Exposure Assessment Survey and to be briefed on the National Health and
Human Nutrition Examination Survey. The IHEC was charged to respond to the following five
questions:

       a)      What are the strengths and weaknesses of multimedia, multipathway measurements
              of exposure as represented by the NHEXAS program, insofar as they can be
              defined at this point?

       b)      Are the ongoing and  planned analyses appropriate and likely to further the goals of
              NHEXAS? At the level of each consortia?  At the level of NHEXAS?

-------
c)     What actions would be likely to increase the utility of the information from
       NHEXAS? In the near-term?  In the longer term?

d)     What follow-up studies would be most useful in the near term, considering that
       key NHEXAS analyses will not be completed for a year? What is the appropriate
       balance between large population surveys and more targeted follow-up studies?

e)     What additional IHEC  advice is offered for strengthening the immediate and long-
       term  utility of NHEXAS and studies like it?

-------
                        3. RESPONSE TO THE CHARGE
3.1 Charge Question 1: Strengths and Weaknesses

       What are the strengths and weaknesses of multimedia, multipathway measurements of
       exposure as represented by the NHEXASprogram, insofar as they can be defined at this
       point?

       Overall, the Committee found that the NHEXAS pilot program is strong scientifically and
addresses many of the issues of concern to the Agency. Another strength of NHEXAS is that it
can be used to identify serious human health risks. The criteria for the selection of the target
chemicals that were measured was identified as a weakness of the NHEXAS pilot program.

3.1.1  Strengths

       Scientifically, NHEXAS is outstanding in both study design and implementation and
provides an excellent model for future EPA  studies. Key scientific strengths include: a) a
hypothesis-driven design; b) the integration of models with measurements to test and refine
models and to understand causality;  c) some advances in measurement methods; and d) a strong
quality assurance plan. These are essential features of the "scientific method" and provide great
power for addressing environmental health problems.  The framework of NHEXAS makes it
possible to track many of the  exposure pathways back to sources of exposure and provides a
sound scientific basis for exposure and risk reduction. In addition, the NHEXAS pilot study has
also generated a rich database.

       Another important strength of the emerging NHEXAS data is the ability to quantify
multiple pathways of exposure from different source media to a single or to multiple exposure
media. No other data set includes temporally and spatially concurrent measurements that makes it
possible to simultaneously trace chemicals from  air and soil to house dust or from air and soil to
indoor air. NHEXAS data can help  to identify serious human health risks, to decide what
interventions would be helpful in reducing these  risks, and to document the effectiveness of
interventions in actually reducing exposures.

3.1.2  Weaknesses

       If EPA decides to proceed with further phases of the NHEXAS program, better criteria
for the selection of target analytes (chemicals of concern that are measured in food, air, water,
etc.) need to be developed. In future phases, there should be more two-way interaction between
risk assessment experts and NHEXAS program participants so that the list of target chemicals to
be measured evolves dynamically as our understanding of human exposures and health risks
advances.

-------
3.2 Charge Question 2: Ongoing and Planned Analyses

       Are the ongoing and planned analyses appropriate and likely to further the goals of
       NHEXAS?  At the level of each consortia?  At the level ofNHEXAS?

       The ongoing and planned analyses were outlined well at the meeting with the exception of
setting limitations on the degree of extrapolation of the data.  One strength of the data analyses is
that all food/beverages, water, and biologicals were analyzed by FDA, EPA contract laboratories,
and CDC, respectively, with the exception of some food samples that were analyzed by RTI.
However, there are some significant differences in the methods for collection for some of the
samples and in the protocols for the use of those samples. Therefore, the integration and meta-
analysis of the food^everages, water and biologicals data from all three Consortia should be done
where possible. In addition, each consortium had certain samples analyzed in its own
laboratories. There were inter-laboratory analyses of performance evaluation samples
(coordinated by NIST) and of split samples (or extracts).  Since there are  some differences in
sample collection and analysis, the IHEC realizes that there are some limitations on the extent of
data integration and "meta-analyses" that can be done.

       The NHEXAS pilot surveys that have been conducted to date have emphasized data
collection. During the meeting, a representative of the Agency mentioned that a minimal amount
of resources have been allocated for further analysis of the NHEXAS data. The EPA should
ensure that sufficient resources are available to fully analyze the data from this large study.

       The ability to analyze the data at the level ofNHEXAS (combined three studies) is
contingent upon the development of a strategic plan to assure that the data from the consortia are
comparable to allow for meta-analysis. IHEC recommends that the Agency complete the quality
assurance of the data before entering the data into a master database. The Committee strongly
recommends that the data be analyzed using the original protocol and that the hypotheses be
tested first using  the constraints of the protocol, e.g. there were only sufficient samples to
calculate the upper 90th percentile.  Criteria should be established before the analyses are
conducted for each of the parameters, including such things as: a) the number of observations
required to calculate the mean, 50th percentile, 75th percentile and 90th percentile; b) the
treatment of censored and missing data; c) non-response calculations; and d) application of
weighting factors.

       The Committee highly recommends that findings that are of considerable importance to
the Agency in current risk management efforts be identified and given high priority for analysis
and publication.  The Committee also provides recommendations for analyzing the questionnaire
data, analyzing the exposure data, conducting hypothesis testing, developing models to integrate
exposure from different media and estimating long-term exposures from short-term
measurements.

-------
3.2.1  Analyses at the Level of Each Consortia

       With limited resources, all three consortia explored their data to different depths and
varying dimensions, some even beyond the original goal of the Phase I field work to plan, design,
test, and demonstrate critical components of NHEXAS.

       The Committee was concerned about the difference in participation rates amongst the
three consortia and the potential for a difference in representativeness of the population sampled
in the different consortia.  To assess the feasibility of the population-based probability sampling
design, all three consortia presented some data on representativeness, response rates, and dropout
rates.  The representativeness of the probability  samples of Region V and the State of Arizona
does not appear to be problematic since the demographics of samples from these two regions are
not very different from census data. A simple observed-against-expected chi-square test can be
used to test whether or not deviations from proportions of census data reach a statistically
significant level.

       The response and completion rates are comparable to other large surveys, but are still
lower than the expected 75%. Therefore, demographics of nonparticipants and dropouts should
be compared with those who participated and remained in the study to see if there is any bias
throughout the various stages of the project. If detailed analysis can be carried out to identify
factors related to non-response and dropout, investigators may gain some insight on how to
increase the response rate for the national survey. NHANES, a national sample of over 30,000
people had a response rate of approximately 80%. The NHEXAS pilot studies should be able to
achieve response rates close to those of NHANES.

       Unlike the other two consortia, higher-than-expected completion rates (>95%) and high
retention rates (>95%) have been observed in the Maryland study. The level of incentive was
based on the extent of participation, with higher incentives offered for more difficult procedures.
The IHEC recommends that the Agency explore the use of incentives for a national survey. In
doing so, the Agency should conduct an analysis of the cost-effectiveness of incentives.

3.2.2  Analysis at the Level of NHEXAS

       a)      Master database and strategic plan for data analysis - The IHEC strongly
              recommends that the Agency develop a strategic plan to analyze the data from the
              three consortia in a systematic way so that the core information can be delivered to
              the Agency in a timely manner.  This strategic plan would ensure that all datasets
              have been checked and found to  meet quality  control/quality assurance standards
              before they are entered into a master database. The master database should be
              user-friendly and easily installed on a personal computer. Data from the three
              consortia should be comparable in format, so that cross-checking and meta-
              analysis can be easily carried out. The database should be designed so that it can
              easily incorporate other relevant  datasets, such as time-activity studies and

-------
       NHANES data, for an integrated data analysis.  The master database should
       provide a user-friendly environment for future data management, statistical analysis
       and modeling work of all datasets stored in the database.  The Committee is
       concerned about the Agency's ability to retrieve electronically stored data in a few
       years from now given the rapid pace at which computer technology is advancing
       and changing.  Therefore, the IHEC strongly recommends that, in addition to any
       electronic versions of the NHEXAS data, the EPA maintain a complete paper copy
       of all data, including measurement results and questionnaires.

       Key elements of the strategic plan for data analysis should include:

       (1)     completion of quality assurance of the data,
       (2)     statistical analyses to develop some feel for the data, e.g., means,  shapes of
              data distributions, ranges, etc.,
       (3)     testing of study design hypotheses plus any related hypotheses,
       (4)     model validation and refinement,
       (5)     evaluation and comparison of tools, e.g., questionnaires and measurements,
       (6)     critical evaluation of the potential value of meta-analysis across the three
              sub-components of NHEXAS and development of a plan for any meta-
              analysis, and finally
       (7)     the identification of findings of considerable importance and help  to the
              Agency in some current risk management efforts (the early analysis of the
              NHEXAS data suggest that there may be findings of this nature).

b)     Evaluation of survey instruments and Quality Assurance/Quality Control  data - In
       order to assess total exposures, questionnaires on baseline information and time-
       activity were administered, diet-diaries were obtained, direct observations by field
       personnel were noted, and environmental samples from air, soil, dust, water, food,
       beverage, urine, and blood were taken.  Classes of chemicals analyzed include
       metals, volatile organic compounds (VOCs) , polyaromatic hydrocarbons (PAHs),
       and pesticides. Some of the monitoring and measurement techniques are well
       established, some require validation (e.g., using market basket surveys and food
       diaries to estimate dietary exposures), and others still need further development
       (e.g., skin wipes for dermal  exposure to pesticides). Due to the  diversity
       (questionnaires, diaries, environmental measurements and biological samples) of
       the survey tools, one of the most important objectives of the Phase I field work is
       to evaluate the survey instruments.  None of the three consortia presented data on
       the evaluation of the survey tools.  However, this was not expected at this stage of
       the research. In the follow-up advisory, the Committee is interested in the analysis
       of the questionnaire responses.  The Committee is also interested in the analysis of
       the questionnaire itself to determine what worked and what did not work.
       Additional advice on the questionnaire is provided under the next subheading,
       Questionnaire data.

-------
       Although IHEC cited a strong quality assurance plan as one of the key scientific
       strengths of the NHEXAS design, the Committee was particularly concerned that
       Quality Assurance/Quality Control (QA/QC) data were not presented at the
       meeting, and was further concerned that limited resources have been allocated for
       data analysis.  The completeness, accuracy, and validity of questionnaires should
       be evaluated.  Issues on accuracy, precision, sensitivity, specificity, portability, and
       affordability of environmental measurements should be addressed. After the
       evaluation, questions with too many missing or incorrect responses should be
       deleted from further analysis and invalid chemical measurements should be deleted
       from further processing.

c)     Questionnaire data - So far, data analysis has been focused on descriptive and
       summary statistics, tables, and figures of chemicals measured in various media.
       The analysis of questionnaire data has been delayed.  At  the preliminary stage of
       data analysis, efforts should be spent on entering, checking, correcting and
       cleaning all measured variables. All questionnaire and environmental data are
       needed for the  estimation of integrated long-term exposure. Without the
       questionnaire data, such as time-activity and diet diary, it is not possible to
       integrate exposures across different media and from various microenvironments. It
       is important that the data format and reports of descriptive and summary statistics
       from the three consortia are uniform  so that preliminary results from different
       research groups can be easily compared.  Given that this was a pilot study which
       will provide information on how to improve future studies, the Committee strongly
       recommends that the Agency analyze the questionnaire itself to determine what
       worked and what did not work.

d)     Exposure data - Once  descriptive and summary statistics have been completed,
       concentration data should be transformed into exposure  data.  All three consortia
       generated some exposure distributions from  measured chemicals.  The Committee
       emphasizes the importance of presenting the data in terms of the associated
       uncertainty and limitations to add to the transparency of the information.

       When examining the shapes of exposure distributions that were presented at the
       meeting, it became clear that not all exposure distributions are log-normal.  The
       data presented  at the meeting indicated that some compound-specific distributions
       are close to exponential while others are log-normal, with a second mode towards
       the tail. Further work needs to be done to integrate total exposures from all media
       and to estimate long-term exposures from short-term measurements.  To predict
       more long-term exposures, it is important to know intra-individual/household,
       inter-individual/household, day of week, and seasonal variations.  The optimal
       timing and frequency of sampling are determined by the relative size of different
       variations.
                                     10

-------
       e)     Hypothesis testing - Long lists of hypotheses were formulated by the Arizona and
             Region V consortia (Pellizari, et al., 1995).  However, the results of the
             hypotheses testing were not presented at the meeting.  Some hypotheses may not
             be testable due to the lack of data or insufficient sample size (e.g. for sensitive
             populations and/or minority groups). However, the Committee recommends that
             the Agency  attempt to test hypotheses that can be addressed with the collected
             data.

       g)     Modeling - The data collected from the Phase I field work provide many
             possibilities for modeling work.  The interrelationship of measured variables can be
             explored. Models can be developed to integrate exposures from different media
             and to estimate long-term exposures from short-term measurements. Relative
             contribution of various sources to total exposures and body burdens can be
             assessed. Factors related to the high-end of exposures can be identified. Among
             all possible modeling efforts, the most important one is the exposure assessment
             modeling work.  Examples of modeling projects where the NHEXAS data can be
             applied include the Cumulative Exposure Project and the Total Risk Integrated
             Methodology (TRIM).  The Office of Policy, Planning, and Evaluation developed
             the Cumulative Exposure Project to provide a national distribution of cumulative
             exposures to environmental pollutants, providing comparisons of exposure across
             communities, exposure pathways, and demographic groups. The Total Risk
             Integrated Methodology is a multimedia, time-series simulation modeling system
             developed by the EPA Office of Air Quality Planning and Standards for the
             assessment of human and ecological risks resulting from hazardous  and criteria air
             pollutants.

                    Rigorous and thoughtful data analysis is an essential component of any
             research and requires adequate time and funding. It is not sufficient to simply
             make environmental measurements and stop. The analysis phase of research  is
             essential to fully understanding and utilizing the results of expensive field
             measurements.

3.3 Charge Question  3: Increasing the Utility of NHEXAS

       What actions would be likely to increase the utility of the information from NHEXAS?  In
       the near-term?  In the longer term?

       The IHEC found NHEXAS to be an excellent project that has significant promise for
improving public health in a cost-effective manner.  When completed it should provide a greatly
improved understanding of human exposure to selected pollutants. This, in turn, will be helpful in
determining the most effective strategies to reduce the public's risk. Therefore,  the Committee
strongly encourages the EPA to pursue the completion of the study in an expeditious manner.
                                           11

-------
3.3.1  In the Near Term

       The EPA should review the goals of this research and develop a strategic plan for the
analyses of the Phase I data.  Included should be a clear statement of the limitations of the Phase I
studies (with attention to distributions, tails, 95th percentiles, etc.). Detailed recommendations
for the analysis of the data are provided in the IHEC response to Charge question 2.

       The EPA should provide funding to complete the primary data analyses. The NHEXAS
information will be of only limited utility if such analyses are not completed. Furthermore, EPA
should facilitate cooperation among the three consortia during the data analysis phase in order to
achieve optimal coordination of the resulting information. Detailed recommendations for
analyzing the data in a systematic way are presented in the Committee's response to Charge
question 2.

       While the  experience is fresh in the minds of the various members of the consortia, the
EPA should encourage the investigators to: a) evaluate the different field study designs; b)
identify the most cost-effective approaches (to exposure assessment) that were used during the
study; and c) identify less expensive ways to accomplish exposure assessment.

       The EPA should conduct a cost-benefit analysis of the various components of the study.
Such an analysis could prove quite useful in discussions regarding the value of the program and
would also be of utility in seeking future funding.  It is important that the costs of this program be
presented within a frame of future savings as a result of improved public health resulting from
better targeted, more effective, and less costly regulatory efforts.

       a)     Prototypical analysis of exposure and assessments of intervention strategies - In
              the near-term, prototypical analyses of exposure and assessments of intervention
              strategies should be made for a variety of chemicals measured in the NHEXAS
              pilot program. It is recognized that NHEXAS  data collected to date are for
              demonstration purposes, so any assessments based on the data would be  limited
              and preliminary.  Nevertheless, a diverse set of analyses such as those listed below
              could be performed using NHEXAS data.  Examples of chemicals or products for
              inclusion in NHEXAS that should provide significant information to the Agency
              include the following:

              (1)    trace metals - For example, the mercury concentrations measured in the
                     NHEXAS pilot studies in various environmental media and the
                     contributions of each to overall  human exposure should be reported to
                     other EPA programs that are responsible for regulating mercury in the
                     environment.  Although late, input on this issue could still be helpful to
                     EPA's risk assessment and risk management for mercury.
                                            12

-------
       (2)     disinfection byproducts - Many water systems are switching from chlorine
              to chloramines to provide drinking water treatment and residual
              disinfection in the water distribution system.  This is motivated, in part,
              because there are concerns about the formation of disinfection byproducts
              (notably trihalomethanes such as chloroform) when chlorine is used as the
              disinfectant, and also because there are concerns about hazardous materials
              present during accidental atmospheric releases of chlorine gas. It may be
              possible, using NHEXAS data, to distinguish between water systems in the
              same area using chloramines instead of chlorine to disinfect drinking water.
              This would be a good prototypical example of assessing the effects of a
              source intervention on human exposure to chloroform. However, the
              Agency must keep in mind that the data from these NHEXAS pilot studies
              are not necessarily nationally representative.

       (3)     volatile organic compounds - Reformulated gasoline (RFG) was required
              beginning in 1995 in Chicago, Phoenix and other cities nationwide.  In
              addition to requiring the use of oxygenated compounds such as methyl
              tert-butyl ether (MTBE) or ethanol in gasoline, a separate mandate of the
              Clean Air Act limited  benzene in gasoline to 1%  by volume or less.  It may
              be possible to document changes in benzene inhalation exposures by
              comparing areas in all three of the pilot study regions where RFG was used
              to other cities where RFG was not required,  and/or by comparing with
              historical benzene exposures measured in the Total Exposure Assessment
              Methodology (TEAM) study (EPA, 1987).

       (4)     pesticides - Similar analyses to those presented for mercury could be
              conducted to demonstrate how NHEXAS data could help assess the
              importance of various environmental media in contributing to exposure of
              one or more pesticides of concern.  The pesticide measurements may also
              be of value to the Office of Prevention, Pesticides and Toxic Substances in
              its effort to implement the Food Quality Protection Act.

       The EPA should archive the information from NHEXAS in a manner that the data
       can be readily searched several years from now.  This is  especially true for the
       analytical data. As an example, the Agency should consider the total ion
       chromatograms collected as part of the program. Although specific sets of VOCs
       were chosen as target chemicals to be measured, there is the potential of searching
       these chromatograms for other analytes. This would require that all the associated
       information be stored in a cross-referenced fashion that is readily retrievable.

b)     Database integration - The IHEC recommends the following to improve the quality
       and utility of the databases from the three pilot studies:
                                    13

-------
              (1)    the databases from Arizona and Region 5 should be integrated in some
                    fashion,

              (2)    the NHEXAS researchers should assess the implications of the Maryland
                    study for the Arizona and Region 5 study,

              (3)    the NHEXAS results should be integrated with information on criteria
                    pollutants from fixed site monitoring stations, and

              (4)    the databases from NHEXAS and NHANES should be cross referenced in
                    a fashion that fully takes advantage of the "complementarity" of the data.

       As recommended in the response to Charge question 2, the Committee recommends that
the EPA initiate the development of a strategic plan and feasibility study for a national survey.

       During the meeting, the Committee was informed of the interagency involvement
including local departments of health.  However, the Committee noted gaps in the interagency
communication. For example, there were cases in which high levels of lead were measured in
homes but this information was not provided to the local health department due to reasons of
confidentiality.  The Committee recommends improvement in the communication between the
NHEXAS investigators and state and local  public health officials to maximize the reduction of
health risks from exposure to environmental chemicals that can be achieved through the
NHEXAS program.  For example, the  confidentiality agreement might be modified to permit
notification of health officials and subjects when high exposures have been measured.

       The EPA should also develop a public education strategy. The extent to which the
community leaders were aware of the NHEXAS studies is unclear.  When the data are  released to
the public, information should also be communicated on the findings and their implications. The
data that are released should include specific recommendations for reducing exposures to
potentially harmful agents, if appropriate.

3.3.2  In the Longer Term

       The EPA should consider studies designed to bridge NHEXAS and NHANES.  Such
studies have the potential to significantly enhance the utility of information derived from both
programs.

       Implementing a national survey could significantly improve public health. However, it
should be designed in the most cost-effective manner. The EPA should use the experiences of the
Phase I consortia to evaluate the feasibility of a national survey. The EPA should focus on those
areas that most contribute to risk reduction. It may be desirable to coordinate such a survey with
NHANES.  The Committee recommends that if the EPA finds a national survey to be feasible
                                           14

-------
upon evaluation of the NHEXAS data and conducts such a survey, the Agency should also
consider sub-populations such as, children, minorities, and inner city residents.

       In light of a potential national survey, the EPA should reassess the target chemicals to be
measured. This includes specific volatile organic compounds as well as selected biologicals that
are emerging as substantial risks to susceptible populations such as , children, minorities and inner
city residents.

3.4 Charge Question 4: Follow-up Studies

       What follow-up studies would be most useful in  the near term, considering that key
       NHEXAS analyses will not be completed for a year?  What is the appropriate balance
       between large population surveys and more targeted follow-up studies?

3.4.1  Follow-up Studies in the Near Term (over the next 12 months)

       It is difficult to propose near-term studies with the limited analysis of the data obtained to
date. In addition, the time line for completion of the NHEXAS pilot data analyses (i.e., one year)
is probably optimistic, and it is highly unlikely that any new studies could be proposed, reviewed,
and performed during such short time frame. At this time, EPA will be best served by directing
resources to the analysis of the data already collected, and to a thorough evaluation of the
methodologies used during the pilot studies.  In this context, it is important to carefully examine
the different approaches taken by the three consortia in  order to determine what methods did or
did not work, and give careful thought to alternative methods that could be more cost effective.
Another important issue to address is the amount and nature of the new knowledge derived from
the pilot studies as it relates  to the methodologies implemented. New studies should be
considered only after the pilot data are analyzed. The Committee recommends that the Agency
present the results of the data analysis and methodology evaluation at the follow-up IHEC
advisory so that the Committee may provide advice and assistance in the development of future
studies.

3.4.2  Balance between Large Population Surveys and More  Targeted Follow-up Studies

       The Committee recognizes the Agency's struggle to identify the appropriate balance
between large population surveys and more targeted follow-up  studies. Given the goals of risk
management, to lower the central tendency of an exposure distribution and to reduce the
percentage of the population above the health-related benchmark, both large population surveys
and more targeted follow-up studies provide valuable information. Population probability-based
studies are the only means for collecting baseline information that can be used for current risk
assessment as well as trend analysis. On the other hand, more targeted special studies tend to
assess high end exposures more precisely and identify sources and factors associated with such
exposures. For populations  at or above such benchmarks, identification of major sources, media
and pathways are essential in order to reduce unacceptably high risk through mitigation strategies.
                                           15

-------
If the majority of the national population is exposed to pollutants at levels under health-related
benchmarks, source identification for such exposures is not a priority from a health standpoint.

       Due to the cost of a national survey, population sample size and the number of analytes
must be minimized. A well designed population study can be accomplished with a population-
based sample of less than one thousand individuals.  To minimize the cost and maximize the utility
of the data that are collected in population surveys designed to establish baseline exposure data,
the IHEC recommends that the Agency: a) focus on obtaining important parameters of the
population survey, such as, the central tendency (mean, median, and mode), the variation
(standard variation, 25th percentile, and 75th percentile), and the shape of the distribution; and b)
evaluate the six pages of hypotheses (Pellizari, et al., 1995) to decide which worked and which
did not work and apply this information to future studies.

       Determining the upper tail of the exposure distribution requires a much larger sample size
than establishing baseline exposure.  A stratified sampling scheme would be more efficient for this
purpose, but the survey design must stratify the population based on exposure-related
characteristics and over-sample those sub-populations at the high end of the distribution. A
national survey of this kind would require measurements of many environmental contaminants.
This type of survey would be more difficult than a population study because the exposure-related
characteristics of different chemicals can be quite variable and the characteristics of the sub-
populations at the extremes of the distributions can also be quite variable. This type of
information can only be acquired through more targeted studies.

       The IHEC provides the following recommendations to reduce the cost and maximize the
utility of more targeted follow-up studies. The EPA should use the experience, lessons learned,
and wisdom gained from the pilot studies  to guide the Agency and external investigators on how
to optimize the measurement and analytical approaches. The Committee also recommends that
the Agency design surveys that can address multiple goals.  For example, if the exposure-related
determinants are similar for the chemicals/subpopulations, they could be combined into one
survey.  Examples of multiple theories that can be tested are: a) Are exposures to pesticides high
in agricultural  areas; and are Hispanics who live in those areas more likely to  be  exposed?; and b)
Are exposures to indoor air pollutants in urban areas of health concern; and are African
Americans more likely to be exposed to the indoor air pollutants and more likely to have high
asthma rates?

       In summary, it was difficult to answer the question because the NHEXAS data have not
yet been fully analyzed.  The Committee found utility  in both population surveys and more
targeted follow-up studies and provides recommendations for increasing the cost-effectiveness
and utility of both types of studies.
                                            16

-------
3.5 Charge Question 5: Additional Advice

       What additional IHEC advice is offered for strengthening the immediate and long-term
       utility ofNHEXAS and studies like it?

3.5.1  Advice for Strengthening the Utility ofNHEXAS in the Near Term

       a)     Model Development - There is a tendency in the exposure analysis community to
             adopt the epidemiological/statistical analysis methods utilizing "factors" such as
             presence or absence of a pollutant source, human activity, etc., rather than models
             based on mechanisms and physical causality.  Models based on the underlying
             physical causes are much more powerful because they are more explanatory and
             more quantitative, and because they have predictive power as well as the ability to
             integrate existing measurement data. For example, indoor air models have often
             been based in the past on the presence or absence of a source rather than a mass
             balance model with quantitative source emission rates. With the measurements
             made in the NHEXAS pilot study, emphasis should be on developing physically-
             based quantitative models that can be used to evaluate various possible risk
             management options.

       b)     Other EPA programs - It is important that NERL disseminate the  NHEXAS data
             to other divisions within the Agency, and that NERL personnel be aware of which
             projects in other divisions could use such data.

       c)     Assessing source-to-dose trends - The EPA should use the NHEXAS data, where
             possible, to assess source-to-dose trends for chemicals such as lead, benzene and
             PAHs. This would provide the Agency with information on how environmental
             measurements compare with known changes in sources and emissions.

3.5.2  Advice for Strengthening the Utility ofNHEXAS in the Long-Term

       a)     Instrumentation - More streamlined and less costly methods will be needed for a
             national survey, e.g., advanced sensors and sampling technologies, robotics for
             multiple analyses, and advanced software for analysis of chromatographic outputs.
             NERL should explore using some of the  advanced  measurement technologies that
             are being developed by EPA and agencies such as the Department of Defense
             (DOD) and the Department of Energy (DOE), that might be adapted for use in a
             national survey.

       b)     Other EPA programs - The data already collected during the pilot phase of
             NHEXAS can be useful for on-going programs within the Agency. Several of
             these include:
                                          17

-------
       (1)     Outdoor, indoor and personal air concentrations of multiple contaminants
              that can be utilized for validation of concentration and exposure estimates
              derived from models (for example, the Office of Planning, Policy, and
              Evaluation Cumulative Exposure Project);

       (2)     Concentrations of contaminants in other media that can also be used for
              validation of multimedia transport models;

       (3)     Information  on biological analytes that can be used for the purpose of
              comparison with exposure benchmarks (Even though the NHEXAS pilot
              data do not provide exposure distributions for a representative sample of
              the US population);

       (4)     The concentration, external exposure, and biological analyte data which
              can be used for the development/optimization of exposure factors;

       (5)     Information  on minority populations, especially Hispanics, that is seriously
              lacking including activity patterns, food preferences, and exposures (These
              data should be used with care, however, given its limited statistical
              power.); and

       (6)     Both in the short term and long term, the data can be useful in the
              prioritization of programmatic and regulatory efforts. (For example, the
              NHEXAS pilot data suggest that exposure pathways involving water intake
              may not be significant for several commonly used pesticides. This finding
              could be used for re-targeting regulatory efforts in this area to other
              contaminants for which water intake represents a significant pathway or
              limiting those efforts to sources of drinking water that may indeed result in
              significant pesticide intake, e.g., shallow water wells in rural areas).

c)     Data simulation - The EPA should work with  statisticians and modelers to better
       understand the capabilities and limitations of the NHEXAS data for making
       multimedia source-to-dose assessments.

d)     Iterative analysis - The EPA should consider using the NHEXAS data to develop
       preliminary concentration distributions for exposure concentrations of a chemical
       in multiple media. These distributions can be used to develop premises regarding
       exposures prior to the collection of additional  site-specific or local data. In the
       context of Bayesian statistics, the preliminary  distributions are referred to as
       "prior" distributions. The collection of additional data can be used to develop a
       likelihood function that expresses the likelihood, given the prior distributions, of
       observing the value  ranges and distribution of collected data. When the prior
       distributions;  a smaller, more  site-specific data set; and the likelihood functions are
                                     18

-------
       combined, the resulting "posterior" distribution includes both the broad range of
       data in the prior distribution and the calibration of that data using a small, but
       better-targeted sample set.  The resulting posterior distribution is statistically more
       robust and better tailored to the likely distribution of local exposures than either
       the NHEXAS data (the prior) or the small site-specific sample set used alone.

e)     National survey - The Committee recommends that the EPA use the data from the
       NHEXAS pilot studies to build the premises for a national exposure survey.
       Planning for Phase 2, a national surveillance plan, should require a plan that uses
       the experience of the pilot study so that the most appropriate multimedia
       measurements (including questionnaires) are used in Phase 2. Given the very
       limited information on actual multimedia exposures which now exists, the emphasis
       should be placed on determining median exposures rather than measuring highly
       exposed populations.

f)     Community-based environmental programs - The Committee recommends that the
       Agency continue to communicate and disseminate the NHEXAS data in the
       communities.  The Committee also recommends that the Agency determines how
       the NHEXAS data can be used to address community-based questions regarding
       public health and environmental trends.

g)     Quantification of sources and emissions - In the NHEXAS pilot studies, there are
       qualitative measures of sources, i.e. presence of smokers, cooking patterns,
       proximity to traffic and industrial emissions, etc.  However, the Agency should
       attempt to match  sources to some of the environmental measurements.  Data on
       sources were collected using the questionnaires.  However, during the
       presentations, minimal information was presented on the content of these
       questionnaires.  Therefore, it is unclear at this point,  until further analysis is
       completed and published, whether the information collected was adequate to make
       important links between sources of contaminants and their concentrations in
       environmental media to allow the Agency to determine which sources should be
       controlled to reduce exposure. As recommended in Section 3.3 of this report, the
       Committee recommends that the Agency analyze the responses on the
       questionnaires and that the Agency also analyze the  questionnaire itself to
       determine what worked and what did not work.
                                    19

-------
                                 4. CONCLUSIONS
4.1 Strengths and Weaknesses

       Overall, the IHEC found the NHEXAS pilot program to be strong scientifically.  Given
that NHEXAS provides vital information to the EPA for addressing the many multimedia
pollutant issues that the EPA is confronted with on a daily basis, the Committee also found
NHEXAS to be highly significant research which will be useful in identifying serious human health
risks.  Key scientific strengths include: a) a hypothesis-driven design; b) the integration of models
with measurements to test and refine models and to understand causality; c) some advances in
measurement methods; and d) a strong quality assurance plan.  Another strength of the emerging
NHEXAS data is the ability to quantify multiple pathways of exposure from different source
media to a single or to multiple exposure media. The criteria for the selection of the target
chemicals that were measured was identified as a weakness of the NHEXAS pilot program.

4.2 Ongoing and Planned Analysis

       The ongoing and planned analyses  were outlined well at the meeting with the exception of
setting limitations on the degree of extrapolation of the data. Although the quality assurance plan
was identified as a scientific strength of the NHEXAS pilot program, the Committee was
particularly concerned about the minimal analysis of the quality assurance data to date.
Therefore, it was difficult to determine the quality of the data. However, this may be attributable
to the timeframe of the first NHEXAS advisory. The Committee was also concerned about the
difference in participation rates amongst the three consortia and the potential for a difference in
representativeness of the population sampled in the different consortia.

       Analysis of the data at the level of NHEXAS is contingent upon the development of a
strategic plan to assure that the data from the consortia are comparable to allow for meta-analysis.
IHEC recommends the completion of quality assurance of the data prior to entering the data into
a master database.  The Committee highly  recommends that findings that are of considerable
importance to the Agency in the current risk management efforts be identified and given high
priority for analysis and publication. The Committee also provides recommendations for
analyzing the questionnaire data, analyzing the exposure data, conducting hypothesis testing,
developing models to integrate exposure from different media, and estimating long-term
exposures from short-term measurements.  The Committee emphasizes the importance of
analyzing the questionnaire, itself, to determine what worked and what did not work since the
NHEXAS Phase I studies are pilot studies.
                                           20

-------
4.3  Increasing the Utility of NHEXAS

       In the near term, the IHEC recommends that the Agency focus on developing and
implementing guidelines and a strategic plan for completing the data analysis. The Committee
also recommends that the EPA: a) identify the most cost effective approaches to exposure
assessment based on the NHEXAS pilot study data; b) conduct a prototypical analysis of
exposure assessment of intervention strategies; c) archive the information in a manner such that
the data can be retrieved years from now; d) integrate the databases from the three pilot studies,
where possible;  and e) develop a strategic plan and feasibility study for a national human exposure
survey.

       In the longer term, the Committee recommends that the Agency  develop a public
education strategy, consider studies designed to bridge NHEXAS and NHANES, evaluate the
feasibility of a national survey, and reassess the target chemicals to be measured.

4.4 Follow-up  Studies

       The Committee found it difficult to propose near-term studies given that the analysis of
the data obtained to date is limited.  Therefore, the Committee recommends, as previously stated,
that the Agency direct its resources to the analysis of the data that has already been collected and
to an evaluation of the methodologies used during the pilot studies. Also, the six pages of
hypotheses (Pellizari, et al., 1995) need to be reviewed and analyzed to determine which worked
and which did not work.

4.5 Balance between Large Population Surveys and more Targeted Follow-up Studies

       In summary, it was difficult to answer the question because the analysis of the NHEXAS
has not been completed.  However, the Committee found utility in both  population surveys and
more targeted follow-up studies and provides recommendations for increasing the cost-
effectiveness and utility of both types of studies.  In addition, The Committee recommends
defining the high end of exposure in the more targeted studies based on  a health-related
benchmark.

4.6 Additional Advice

       The IHEC recommends that the EPA:

       a)    place emphasis on developing physically-based quantitative models that can be
             used to evaluate various possible risk management options;

       b)    consider using advanced measurement technologies such as advanced sensors and
             sampling technologies, and robotics for multiple analyses; and advanced software
             for analysis of chromatographic outputs; and

                                          21

-------
c)     disseminate the NHEXAS pilot data, after the analyses has been completed, to on-
       going programs within the Agency, such as programs that are trying to validate
       multimedia transport models.
                                    22

-------
                                  REFERENCES
*Buck, RJ, Hammerstrom, K.A., and Ryan, B.P., 1995, "Estimating long-term exposures from
       short-term measurements," Journal of Exposure Analysis and Environmental
       Epidemiology, 5(3): 359-373.

EPA, 1987, The Total Exposure Assessment Methodology (TEAM) Study: Summary and
       Analysis: Volume 1, USEPA Office of Research and Development, Washington, DC,
       EPA/600/6-87/002a, June 1987.

*EPA,  1998. NHEXAS: The National Human Exposure Assessment Survey Overview,
       September, 1998.

international Society of Exposure Analysis, 1997. 1997 NHEXAS Abstracts from the 7th Annual
       Meeting of the International Society of Exposure Analysis, November 2-5, 1997,
       Research Triangle Park, North Carolina.

international Society of Exposure Analysis and International Society for Environmental
       Epidemiology, 1998.  1998 NHEXAS Abstracts from the Tenth Conference of the
       International Society for Environmental Epidemiology and the Eight Conference of the
       International Society of Exposure Analysis, Boston, Massachusetts.

*Lebowitz, M., O'Rourke, M., Gordon, S., Moschandreas, D.J., Buckley, T., and Nishioka, M.,
       1995, "Population-based exposure measurements in Arizona: a Phase I field study in
       support of the National Human Exposure Assessment Survey," Journal of Exposure
       Analysis and Environmental Epidemiology, 5(3): 297-325.

*Pellizari, E., Lioy, P., Quackenboss, J., Whitmore, R., Clayton, A., Freeman, N., Waldman, J.,
       Thomas, K., Rodes, C., and Wilcosky, T., 1995, "Population-based exposure
       measurements in EPA Region 5: a Phase I field study in support of the National Human
       Exposure Assessment Survey," Journal of Exposure Analysis and Environmental
       Epidemiology, 5(3): 327-358.

* Sexton, K., Kleffman, D.E., and Callahan, M.A., 1995, "An Introduction to the National Human
       Exposure Assessment Survey (NHEXAS) and related Phase I field studies," Journal of
       Exposure Analysis and Environmental Epidemiology, 5(3): 229- 232.

*Sexton, K., Callahan, M.A., Bryan, E.F, et al., 1995, "Informed decisions about protecting and
       promoting public health: rationale for a national human exposure assessment survey,"
       Journal of Exposure Analysis and Environmental Epidemiology, 5(3): 233- 235.

       *      Material reviewed by the Committee prior to the meeting.

                                        R-l

-------
           APPENDIX A - ACRONYMS AND ABBREVIATIONS
CDC
CEP
DOD
DOE
EOSHI
FDA
GPRA
GSA
fflEC
MTBE
NAAQS
NCEA
NERL
NHANES
NHEERL
NHEXAS
NIST
ORD
PAHs
QA/QC
RFG
RTI
SAB
TEAM
TRIM
VOCs
Centers for Disease Control and Prevention
Cumulative Exposure Project
Department of Defense
Department of Energy
Environmental and Occupational Health Sciences Institute (EOHSI)
Food and Drug Administration
Government Performance and Results Act
General Service Administration
Integrated Human Exposure Committee
methyl-tert-butyl ether
National Ambient Air Quality Standards
National Center for Environmental Assessment
National Exposure Research Laboratory
National Health and Human Nutrition Examination Survey
National Human and Ecosystem Effects Laboratory
National Human Exposure Assessment Survey
National Institutes of Standards and Technology
Office of Research and Development
Polyaromatic Hydrocarbons
Quality Assurance/Quality Control
reformulated gasoline
Research Triangle Institute
Science Advisory Board
Total Exposure Assessment Methodology
Total Risk Integrated Methodology
Volatile Organic Compounds
                                       A-l

-------
                              DISTRIBUTION LIST
Administrator
Deputy Administrator
Assistant Administrators
Deputy Assistant Administrator for Science, ORD
Director, Office of Science Policy, ORD
EPA Regional Administrators
EPA Laboratory Directors
EPA Headquarters Library
EPA Regional Libraries
EPA Laboratory Libraries
Library of Congress
National Technical Information Service
Congressional Research Service

-------