REPORT OF THE
REVIEW PANEL ON TOTAL HUMAN EXPOSURE
REVIEW OF THE AGENCY'S ONGOING RESEARCH
IN UNDERSTANDING TOTAL HUMAN EXPOSURE TO
INDOOR AND AMBIENT AIR POLLUTION-'
U.S. Environmental Protection Agency
Science Advisory Board
Washington, D.C. 20460
May 1985
-------
NOTICE
This report has been written as part of the activities of the
Environmental Protection Agency's Congressionally established Science
Advisory Board, a public group providing extramural advice on scientific
issues. The Board is structured to provide a balanced, independent,
expert assessment of scientific issues it reviews, and hence, the
contents of this report do not necessarily represent the views and
policies of the Environmental Protection Agency nor of other agencies
in the Executive Branch of the Federal Government.
-------
TABLE OF CONTENTS
PAGE
I. EXECUTIVE SUMMARY 1
II. INTRODUCTION 3
III. REPORT ON THE PROGRAM REVIEW ON TOTAL HUMAN EXPOSURE 4
A. Background and Limitations of This Review 4
B. Overview of Exposure Assessment Needs
Within the EPA 5
C. Review of Current EPA Programs 7
1. Program Management 7
2. Research Activities 8
3. Additional Issues Not Currently Being
Addressed 9
D. Reviews of Specific EPA Programs 12
1. Instrumentation 12
2. Field Studies 12
3. Modeling 15
4. Dosage Research 16
IV. APPENDICES
A* Charge to the Review Panel A-l
B. Roster of the Review Panel B-l
ii
-------
Key to Acronyms
EMTS
EPA
MEM
NAAQS
NEM
N02
OAQPS
ORD
PEM
SAB
SALTRP
SHAPE
TEAM
voc
Exposure Methods Test Site
Environmental Protection Agency
Microenvironmental Monitoring
National Ambient Air Quality Standards
NAAQS Exposure Model
Nitrogen Dioxide
Office of Air Quality Planning and Standards
Office of Research and Development
Personal Exposure Monitoring
Science Advisory Board
Strategic and Long-Term Research Planning (Subcommittee)
Simulated Human Air Pollution Exposure
Total Exposure Assessment Methodology
Volatile Organic Compound
iii
-------
I. EXECUTIVE SUMMARY
This is the report of the Science Advisory Board's (SAB) Total Human
Exposure Review Panel. The Panel met on February 25-26, 1985 at EPA's
Environmental Research Center, Research Triangle Park, NC, in order to
review the Agency's research program on total human exposure to indoor
and ambient air pollution.
The Review Panel focused on five major topic areas described in the
program document prepared by the Office of Research and Development:
1) development of measurement methods; 2) statistical protocols, data-base
development, and exposure models; 3) microenvironmental field studies;
4) total human exposure field studies; and 5) dosage research investigations,
This program document was well organized, informative, and provided the
Panel with an excellent means to focus their questions and discussion.
The Review Panel was impressed with the quality of the scientific
effort being expended by Agency staff scientists. It was clear to the
Panel that certain dedicated professionals have recognized the importance
of this issue and have carried out research of high scientific quality
to improve our understanding of total human exposure. While the Panel
recognizes this laudable effort, it also notes that the level of support
for this effort is less than it should be. The criticisms noted in this
report should be taken as the Panel's effort to help the Agency improve
the total human exposure program.
Based on a review of the Agency's program document and the presenta-
tions made at the meeting, the Panel has three major recommendations.
These are that the Agency:
Identify the issue of total human exposure as an emerging
scientific issue of critical importance to the Agency's
responsibilities;
Coordinate future research activities related to total
human exposure at senior levels of the Agency; and
Base future data collection activities on carefully developed
experimental design and probability sampling protocols.
The Panel has organized its overall findings into three areas.
These are: 1) program management; 2) research activities; and 3) areas
of research not currently being performed.
The overall objectives, mandate, organizational structure, and
level of support of the Total Human Exposure Program were not clearly
stated during the presentations. There also appeared to be no strategic
planning, coordination or organized direction of this program from senior
levels of the Agency. The technical work is generally excellent but the
significance, cost effectiveness, and ultimate regulatory impact is so
far-reaching that the EPA management should pursue the program with far
more emphasis and central leadership than it currently receives.
-------
The Panel recommends that the exposure assessment activities within
the Agency have a.central management and administration such that activities
in the required technical areas (development of instrumentation, field
studies, modeling, and health effects) are thoughtfully integrated. This
project integration can only be accomplished effectively by coordination
among the EPA Research Committees and project officers at the highest
levels within the Agency. This focus and concept of a strategy must be
formalized into a commitment by senior EPA management to understand
total human exposure to pollutants and its relationship to sources, doses,
toxicity, and the protection of public health.
The Agency's ability to perform credible quantitative risk assessments
is severely constrained by the limitations in current capabilities in
exposure assessment. Although much of the progress in exposure assessment
technology that has taken place in recent years was supported by the
EPA, the level of support from the Office of Research and Development
(ORD), as well as from the program offices has been limited in terms of absolute
dollars invested.
The Panel recommends that future advances in the Agency's capabilities
for total human exposure assessments should be built on its past accomplish-
ments. In particular, the Panel refers to the pioneering and successful
major field studies represented by the Total Exposure Assessment Methodology
(TEAM) Study and the Denver, CO and Washington, DC carbon monoxide exposure
field studies. The results of these studies can help guide the Agency's
approach to environmental monitoring, assessment, and regulation of air
pollutants. In addition, work needs to be done on other pollutants
besides carbon monoxide.
In reviewing specific research programs, the Panel had several
observations: 1) Instrumentation - The Panel recognizes the importance
of instrument and method development in meeting the needs of exposure
assessment; the current EPA work is very good, however, this effort is
not sufficiently comprehensive at the existing level to meet the needs of
the field studies. Moreover, the reliance on passive and active sorbent
devices should be complemented by other sampling schemes such as sensor
arrays, fiber optic devices, solid-state transducers, and electrochemical
sensors. 2) Field Studies - The EPA staff are to be commended for their
scientific contributions to development of the indirect and direct approaches
to exposure assessment, and for their early recognition of the critical
role of survey sampling methodology in development of valued exposure
assessments. In addition, the Panel applauds EPA's efforts in conducting
the TEAM Study and recommends that comprehensive exposure studies of this
type be continued and expanded to other pollutants and additional locations.
3) Modeling - The Panel endorses the plan to validate the SHAPE (Simulated
Human Air Pollution Exposure) model with actual field data, and the
effort to validate and modify the National Ambient Air Quality Standard
(NAAQS) Exposure Model (NEM) to accept empirical activity data as input.
Such validation would enhance the use of those models in future studies.
4) Dosage Research - Although the Panel's review of this program was
brief, it was apparent that the work being done in this area was reasonable
and productive. However, it was clear that dosage research was not being
conducted in connection with exposure research. The Panel views this as
a weakness since a coordinated health component is important to the
exposure assessment activity.
-------
The Panel has developed a set of issues that the Agency should con-
sider in defining its research agenda for future human exposure assessments,
These include:
1) further efforts in understanding human exposure to nitrogen
dioxide and size-fractionated particulate matter;
2) exposure studies should be undertaken for other criteria
pollutants;
3) a coordinated effort should be undertaken to develop the
tools and methods required for these exposure assessment
activities;
4) residential and commercial applications of pesticides and
termiticides represent an important potential source of
inhaled or ingested exposure; and
5) exposure studies and health assessments should be more
carefully coordinated.
II. INTRODUCTION
At the request of the Agency's Office of Research and Development,
the Science Advisory Board has agreed to conduct a series of one-time
reviews on a number of ongoing research programs within the Agency. The
reviews will be coordinated under the auspices of the SAB's Subcommittee
on Strategic and Long-Term Research Planning (SALTRP). The purpose of
these reviews is to conduct a peer review of an ongoing research program
by a review panel of recogized experts in order to communicate to the
Agency the progress, or lack of progress being made in meeting research
needs pertinent to the development of Agency regulations and policy.
Each review will be conducted by a different review panel and will
take place at an appropriate laboratory or other location where the
panel will receive briefings and prepare its report. Approximately
two reviews will take place each calender quarter. Once a review is
completed, the chairperson of each review panel will brief the Deputy
Administrator on the findings and recommendations of that panel.
The Review Panel on Total Human Exposure was charged with the task of
advising the Agency on how well the research program for Total Human
Exposure is progressing toward answering the needs of the Agency, including
both whether or not the right research is being conducted to address
issues defined by the Research Committee process, and whether the research
is being conducted properly.
This report presents the results of the first program review in
this series: The Review of the Total Human Exposure Program.
-------
III. REPORT OF THE PROGRAM REVIEW ON TOTAL HUMAN EXPOSURE
A. Background and Limitations of This Review
This is the report of the Science Advisory Board's Total Human
Exposure Review Panel which met on February 25-26, 1985 in Research
Triangle Park, North Carolina. On the first day the Panel received
briefings, allowing EPA project managers and scientists to present the
highlights of the Total Human Exposure Program. On the second day the
Panel prepared the present report.
The Panel found the material presented to be both interesting and
useful. This review was a first, both for the Panel members and EPA
staff. There were, inevitably, a number of informational areas, mostly
of a non-scientific nature, that were not presented and which would have
been useful, in fact necessary, for the Panel to make a clear assessment
of the overall merits of the program. Some of these areas were:
The overall objectives of the program were not clearly
delineated, and some of the scientific presentations did not
discuss the hypothesis being tested.
There were no cost or effort estimates of the entire program or
its specific components. This made reviewer's curious about the
emphasis placed on each program component and its cost-effectiveness,
No attempt was made to elucidate the mandate for the program,
its organizational structure within the EPA, or the chain of
command of the staff involved.
In many cases staff did not discuss their future plans for
continuation of their projects, or if they did not expect to
continue, why not.
There was no discussion of other EPA projects that, at least
to some degree, could be termed Total Human Exposure Assessments,
or why there was not more integration with these projects.
The charge to the Panel with respect to the report was not
adequately expressed, either formally or informally.
On a positive note, the Panel was pleased with the content of the
scientific presentations, the manner of presentations, and the fact that
the scientific format was geared to the Panel's needs. The research
summaries delivered to the Panel prior to the meeting were well organized
and informative. Clearly the staff involved in the total assessment program
is of very high caliber. The Panel was pleased with the flexibility
displayed by EPA staff when requested to substantially alter the afternoon's
proceedings, at some inconvenience to individual staff.
-------
The criticisms discussed above should not reflect on the EPA staff
or the Panel members so much as on the newness of this particular review
process. The Panel recommends that, for future reviews, the Panel chair-
person be responsible for consulting with Agency staff prior to the meet-
ing in order to set the agenda and format, taking into account the items
discussed above. Whenever possible, the formats of the presentations
of the individual scientists should follow, within reason, a standard
format, including hypothesis, methods, cost, results, technical problems
and future plans.
The Panel trusts that the criticisms made above will not detract
from its general view of the merits of the program or of the excellent
scientific base developed thus far.
B. Overview of Exposure Assessment Needs Within EPA
Based on its review of the Agency's research program on Total Human
Exposure, the following represents the Review Panel's overall conclusions
and overview of EPA's programmatic needs.
The EPA's ability to perform credible quantitative risk assessments
is severely constrained by the limitations in current capabilities in
exposure assessment. Recent research has demonstrated that individual
exposure to criteria air pollutants is strongly influenced by indoor
environments. The contribution of gas cooking stoves to personal exposure
to nitrogen dioxide is one dramatic example, but indoor sources and sinks
modify exposure to all air pollutants. Recent work by EPA scientists has
shown that indoor sources are equally important for volatile organic
compounds (VOC). Better understanding of total human exposure to air
pollutants is critical to the long-term effectiveness of EPA's regulatory
strategy, to the choice of margins of safety to ambient standards, to
the identification of high-risk populations, and to future epidemiologic
research.
In another setting, a recent study by the National Academy of Sciences
on Toxicity Testing Needs found that the toxicity data base was inadequate
for toxicity evaluation for 50% of pesticides and 88% of chemicals in
commerce. The Academy recommended that priorities for further testing should
be guided by the potential for human exposure, for which the data base
was even more inadequate than for toxicity. For toxicity testing, there
may be inadequate data for most chemicals, but there is at least general
agreement on how the tests should be performed. The science of exposure
assessment is less well developed, especially for general population
exposures.
Much of the progress in exposure assessment technology that has
taken place in recent years was supported by EPA. However the level of
support from the Agency's Office of Research and Development (ORD) and
the program offices has been limited, and in its aggregate the level of
effort has been too meager to permit advances in capability commensurate
with the needs for increased numbers of ever more sophisticated risk
-------
assessments. While each program office has specific needs in exposure
assessment technology, they also share some generic needs with each
other and the ORD. Therefore, it is in their best interests to support a
research program focused on improvements in exposure assessment technology,
Such a program should be carried out by the ORD, in cooperation with
appropriate program office personnel, for several reasons. First, the
ORD has personnel with much valuable insight and experience on these
issues and is familiar with the needs of the program offices. Second,
none of the individual program offices is likely to invest in research
which is not closely tied to immediate program needs.
Future advances in the EPA's capabilities for total human exposure
assessments can, and should, be built on its past accomplishments. Among
these are the pioneering and successful major studies represented by the
TEAM (Total Exposure Assessment Methodology) Study and the Denver and
Washington, D.C. carbon monoxide exposure field studies.
The results of these studies can help guide the Agency's approach
to environmental monitoring, assessment, and regulation of air pollutants.
The issues these studies raise include:
What should be the Agency's regulatory stance toward indoor exposure?
Even if the Agency were to take the position that it has no
regulatory role in indoor environments, it must recognize in
its approach to assessment and regulation that indoor exposures
are often greater than outdoor exposures.
The emerging information about the indoor environment raises
doubts about the adequacy and, for the same pollutants, even
the relevance, of outdoor monitoring for protection of public
health.
To assess the importance of sources and the amount of a dose,
exposures must be characterized.
What advances in measurement technology are needed for future
studies?
On a more operational level, other questions arise:
Are current and planned expenditures for total human exposure
assessment consistent with the Agency's information needs for
assessment and regulation?
What level of precision of exposure information is appropriate
in different research and regulatory settings?
How should monitoring resources be allocated between studies
of exposure distributions and monitoring activities in support
of epidemiologic research?
-------
To address these issues, the Panel recommends that the Agency:
Identify the issue of total human exposure as an emerging
scientific issue of critical importance to the Agency's
responsibilities;
Coordinate future research activities related to total human
exposure at senior levels of the Agency; and
Base future data collection activities on carefully developed
experimental design and probability sampling protocols.
C. Review of Current EPA Programs
1. Program Management
The presentations to the Review Panel by EPA management and staff
scientists reflected the view that total exposure assessment is required
in order to perform credible risk assessments and, ultimately, to perform
the regulatory functions of EPA. The Review Panel agrees with this
concept. Moreover, the individual presentations made it clear that
several dedicated professionals within the Agency have recognized the
importance of this issue and have carried out research of high scientific
quality to improve our understanding of total human exposure.. Important
findings have been reported that can affect both exposure assessment and
the EPA's regulatory policies of the future. It was also clear, however,
that there is no strategic planning, coordination, or organized direction
of this total human exposure assessment activity from senior levels of
the Agency. The technical work is generally excellent but the significance,
cost-effectiveness, and ultimate regulatory impact is so far-reaching
that, in the Panel's opinion, EPA management should pursue the program
with far more emphasis and central leadership than it currently receives.
The Panel recommends that the exposure assessment activities within
EPA have a central management and administration such that activities in
the required technical areas (development of instrumentation, field
studies, modeling, and health effects) are thoughtfully integrated.
Further, the exposure assessment activities should be linked to the
overall risk assessment model presented, i.e., to sources and to the
dose/toxicity assessment activities.
This project integration can only be accomplished effectively by
coordination among the EPA Research Committees and project officers at the
highest levels within the Agency. This focus and concept of a strategy
must be formalized into a committment by senior EPA management to under-
stand total human exposure to pollutants and its relationship to
sources, doses, toxicity, and the protection of the public health.
-------
2. Research Activities
Of the many components that relate to an overall risk assessment for
any particular chemical, many, but certainly not all, were presented to
the Review Panel. The overall focus of the research activities presented
was on exposure assessment, but there was also some discussion of dosage
assessment, Dosage assessment activities, however, do not appear to be
coordinated with any of the exposure assessment activities.
Agency scientists have made significant progress in developing the
conceptual approach and methodology for determination of total human
exposure to pollutants. Basically, two procedures have been suggested
and actually implemented on a limited scale. The first is the concept
of personal exposure monitoring (PEM). This requires that a small
measuring device actually be worn by subjects as they go about their
normal activities. The second concept is that of microenvironmental
monitoring (MEM), which requires the measurement of a pollutant's
concentration in a variety of locations inhabited by people during a
normal day's activity and a reconstruction of a complete exposure history
by combining these measurements with an activity log. The two methods
should yield comparable results if each is performed properly. As yet,
comparable results have not been achieved. For example, exposure estimates
differ by about forty percent for the Washington, D.C., carbon monoxide
study.
The EPA has a substantial program in the development of measurement
methods for assessing total human exposure, although most of this activity
is clearly funded by the base research and development program. Most of
the research presented to the Panel was concerned with the measurement
of volatile and semi-volatile hydrocarbons, either by means of a personal
exposure monitor or by means of a larger sampler suitable for micro-
environmental monitoring. Although there is a major expenditure in
developing a sensor for nitrogen dioxide and some effort in developing a
sampler to measure volatile organic compounds that are associated with
particles of diameter less than lOjum, there is very little activity in
developing instrumentation for the measurement of other pollutants.
Several field studies have been conducted in the past by the EPA.
Two of these were related to exposures to carbon monoxide and have been
essentially completed and reported. Another study, the Total Exposure
Assessment Methodology (TEAM) study was presented in detail. This study
is examining exposure to airborne volatile organic pollutants in several
locations over time periods long enough to establish temporal patterns.
The results, as they define sources of exposure to organic chemicals,
are most provocative and indicate that "traditional" sources of exposure
may be minor compared to other more "personal" sources. This study,
more than any other, provides compelling evidence of the necessity of
evaluating total human exposure and of not basing exposure estimates on
pollutant concentrations measured only by fixed station monitors.
-------
Finally, some material on dosage evaluation was presented to the
Panel. Most of the time was devoted to a study in China that seeks to
explain the very high incidence of lung cancer in a small region. Other
material of a diffuse and general nature was presented, but it was fairly
clear that there is no dosage-research study that is connected with the
exposure-research study. The Review Panel views this as a weakness
since a coordinated health component is important to this exposure assess-
ment activity. The present health-component would need significant
redirection in order to complement the exposure studies reviewed here.
3. Additional Issues Not Currently Being Addressed
As discussed above, the current "Total Human Exposure to Air Pollutants"
program within the EPA has been limited in scope. Human exposures have been
examined for only a few pollutants. These projects were primarily directed
toward deriving a frequency distribution of exposures in a representative
population. The relevancy to other useful applications of exposure assess-
ments was not explicitly considered, although further data analysis will
greatly enhance their usefulness. Further, the biological and health related
endpoints of the EPA's exposure assessment work are not well characterized,
are limited primarily to mutagenicity, and are not an integral part of the
field and laboratory studies.
This section sets forth a set of issues that the Agency should con-
sider in defining its research agenda for future human exposure assessments.
This research should help to quantify population exposure, identify the
portion of the population at greatest risk, and identify determinants of
exposure (including specific sources, locations, and activity patterns).
When linked to health studies, human exposure studies can assist in the
quantification of potency factors for a variety of hazardous substances
and air contaminants.
a. Nitrogen Dioxide (NO?) - Studies have demonstrated the import-
ance of unvented combustion sources (mostly in homes) to inte-
grated and peak N02 indoor concentrations. Limited personal
monitoring has identified the importance of these indoor sources
as contributors to total human exposure, but these studies did
not utilize probability sampling in selecting participants or
monitoring plans. Health endpoints were not examined concurrent-
ly with exposures. Actual peak exposures are not well-document-
ed and will not be until instrumentation is improved. The EPA
has a direct interest in understanding human exposures to N02
in order to provide improved estimates of health effects and to
properly restructure air quality standards.
b. Particles - Human exposures to particulate matter is a vexing
problem. Indoor concentrations are an important contributor
to the integrated exposures to respirable size particles.
-------
However, the size'distribution, chemical composition, and,
quite likely, general toxicity of indoor particulate matter
will differ from that of outdoor particles. The Agency needs
to understand human exposures to size-fractionated particulate
matter from the perspective of source contributions and by
fractional components. Fractional components will have different
toxicological effects. The relevant issues include:
1) Characterizing acute integrated exposures to acid aerosols;
The covariance of outdoor activities and acid events
should be examined. The penetration of acid aerosols
into indoor environments, the neutralization of acid
aerosols indoors and the generation of acid gases and particles
indoors could be additional area of inquiry. Panel members
disagreed about the importance and cost-effectiveness of
research on these issues.
2) Characterizing human exposure to vehicle exhaust; Receptor
modeling of urban aerosols has identified vehicle exhaust
as a ubiquitous contributor to ambient particulate matter.
Limited in-vehicle studies have demonstrated several-fold
higher concentrations than measured at :fixed sites. Al-
though in-transit time may be only 5 to 10% of an individual's
daily activities, this may represent an important exposure
opportunity for specific compounds and metals.
3) Characterizing human exposure to fossil and biomass fuel
related primary emissions; It will be important to quantify
the amount and location of human exposure to combustion
derived particulate matter. The emphasis should be on
condensible organic fractions and a few metals (e.g. arsenic,
vanadium, selenium).
4) Characterizing human exposures to cigarette smoke; Passive
smoke exposure occurs predominantly indoors. Respiratory
health effects, including cancer, are either documented or
suspected to be associated with exposure to passive smoke.
Discerning possible health effects associated with other
components of respirable particulate matter will require
careful separation of the possible passive tobacco smoke
effects. Therefore, an exposure assessment program should
utilize appropriate markers for tobacco smoke in character-
izing exposures to segments of our population. This might
be an> area where microenvironment sampling is an appropriate
approach.
5) Characterizing asbestos fiber exposures; While friable
asbestos insulation material has been noted in schools,
offices and homes, an understanding of actual airborne
concentrations of asbestos fibers is missing. To undertake
appropriate risk assessment and to determine remedial
actions, better information on either environmental
10
-------
concentrations of, or personal exposures to asbestos fibers
is needed. These studies should be longitudinal because of
the episodic nature of fiber release. They should include
residential locations.
£ Exposure studies should be undertaken for other criteria
pollutants - The emerging evidence on the importance of indoor
environments and personal activities to air pollution exposure
suggests that current approaches to air pollutant regulation
based on outdoor measurements may become less defensible in
the near future. The importance of total human exposure assessment
should be reviewed for each criteria pollutant to identify
those for which integrated exposure assessment is most urgently
needed.
d. A coordinated effort should be undertaken to develop the tools
and methods required for these exposure measurement activities.
e. Residential and commercial applications of pesticides and
termiticides represent an important potential source of inhaled
or ingested exposure - The Agency should undertake a carefully
designed study (i.e., such as TEAM) of human exposures to
halogenated compounds and to quantify the sources and routes
of exposure.
f. Exposure studies and health assessments should be more carefully
coordinated - These two components do not necessarily have
to be integrated into every study. However, the relevance
of the health related research to actual human exposures should
be explicit. For example, the mutagenicity work currently
associated with the indoor air pollution research program, and
identified as a component of the total human exposure program,
could be better integrated into measurement and risk assessment
activities. Because of the cost and time involved in conducting
multi-dose bioassys with pollution mixtures, carefully designed
sampling schemes should be employed. Environmental conditions
and sample composition must be well characterized. The products
of bioassays will be more useful if consistent with observational
data (clinical and/or epidemiological). For example, bioassay
analysis on suspended particulate matter containing tobacco
smoke obtained in homes and offices might help to elucidate
the relative hazards of different environments. A significant
effort has been made (>$20 million) by the Department of Energy
in recent years to provide a battery of chemical fractionation
and bioassay techniques that can be used in complex mixtures.
This useful published literature should be considered by the
Agency in its health effects component.
11
-------
D- Reviews of Specific EPA Programs
1. Instrumentation
The program described instruments developed primarily for measure-
ments of chemical exposure to selected air pollutants. This work has
been successful and useful to field activities. The prime focus has been
the development of passive and active sorbent collection devices that are
returned to the laboratory for analysis. Plans were suggested to further
develop such approaches. This does not seem a proper focus for continued
effort. The types of measurement needs for field exposure assessment are
broad. Requirements for real-time data, temporal and spatial resolution,
large dynamic range, and multidimensional measurements can never be
totally met by more or even better sorbent collection schemes. Such
schemes have limitations and, therefore, must be complemented by others.
The Panel recognizes the importance of instrumentation in meeting the
needs of exposure assessment; however, the current effort is not sufficiently
comprehensive at the existing level to meet these needs. A recent
workshop at Harvard University sponsored by the EPA and the Gas Research
Institute defined the need for field instrumentation to perform exposure
measurement. The latest technology in chemical monitoring (sensors
and detectors) should be used by EPA researchers to provide new concepts
and approaches for chemical pollutant measurement. Such modern technologies
as integrated sensing systems (sensor arrays), fiber optic devices,
solid-state transducers, and electrochemical sensors can be applied to
field sensing problems and should be a part of the EPA's future instrumen-
tation research program.
The development of instrumentation (and all field techniques) is related
to the prioritizing of compounds, situations, and parameters required in
exposure measurement studies. Then, appropriate strategies (nearterm and
long-range) should be proposed by instrument developers to satisfy these
most important needs using the best available technology. The present
schemes for sorbent measurement have important field uses and should be
supported but should not be the main focus of future efforts.
2. Field Studies
The EPA's measurement activities related to total human exposure to
environmental pollutants fall with the following areas:
Statistical protocols and experimental design
Microenvironmental field studies
Total human exposure field studies
12
-------
Dosage research investigations
The first three topics are discussed in (this section; dosage research
will be discussed in Section D.4 of this report.
a. Statistical Protocols and Experimental Design - The EPA staff
are to be commended for their scientific contributions to
development of the indirect and direct approaches to exposure
assessment, and for their early recognition of the critical
role of survey sampling methodology in development of valued
exposure assessments. This work has raised a number of out-
standing critical statistical issues associated with exposure
assessment, such as:
1) optimal designs for direct and indirect exposure
assessment studies;
2) statistical validity of indirect exposure assessment;
3) evaluation of benefits (in reducing uncertainty)
versus costs of various levels of effort of exposure
measurement, with regard to both epidemiological studies
and risk assessment.
Methodologic research to clarify these issues should be encouraged,
so as to determine the incremental value of intensified measurement
programs in exposure assessment, and to guide the design of
future studies.
b. Microenvironmental Field Studies - These studies characterize
the pollutant concentrations to which people are exposed when
they are in specific microenvironments. Most of the work
conducted so far is related to carbon monoxide exposures.
Clearly, additional studies of the concentration distributions
of other pollutants in appropriate microenvironments are needed
at this stage. Even within a single microenvironment (e.g., a
parking garage), pollutant concentrations are highly variable
in space and time, and great care must be taken to obtain
suitable measurements so that these concentration distributions
can be accurately characterized.
We agree that in future personal monitoring studies, the following
should be applied:
1) use of closed format questionnaires;
2) use of automated instrument output; and
3) locating activities to specific census tracts.
These are crucial lessons learned from the Denver and Washington
carbon monoxide studies and will improve the validity of future
studies.
13
-------
The objectives and expected benefits of the Exposure Methods
Test Site (EMTS) project are not clear to the Panel. Although
there are advantages to having a single locale where micro-
environmental field studies can be conducted, it would be
dangerous to assume that the data obtained at such a locale
would be representative of exposures in other cities in other
parts of the country. This is because of the many demographic,
geographic, and climatological differences between various
locations. Hence, the Panel recommends that the degree to which
the data from an EMTS is expected to be generic and extrapolatable
be established before the project is implemented.
Total Human Exposure Field Studies - The Panel applauds the
EPA's efforts in conducting the TEAM study, and recommends that
comprehensive exposure studies of this type be continued and
expanded to include other pollutants and additional locations.
New developments in pollutant monitoring technology and biological
monitoring should be incorporated as they become available.
Some of the findings of the studies conducted to date
suggest that greater attention should be given to incorporating
source-receptor considerations into the survey design. In some
instances, it appears that much of the variability that shows
up in the exposure data could be explained by meteorological
factors, such as some receptors being downwind of the source(s),
while others are not. A more careful experimental design that
includes consideration of these factors, including measurement
of appropriate meteorological parameters, will likely lead to
more meaningful data in future studies.
The following are some technical comments on TEAM:
1) The intraperson temporal variation in VOC exposure is
crucial in risk assessment and should be given a high
priority in future studies.
2) Given the substantial measurement error, the estimated
exposure distributions can be substantially more hetero-
geneous than the true exposure distributions. For example,
the variance of the estimated exposures is the sum of the
variance of the true exposures and the variance of the
measurement errors, assuming that: a) measurement errors
are homoscedastic, and b) there is no correlation between
measurement error and true exposure. Empirical Bayes
methods are available for such adjustments.
3) We share the concern about the high refusal rate in the
sample enrollment. We would like to see more rigorous
efforts in the future to assess the impact of the refusal
on the generalizability of the sample. For example, a
subsample of the accessible part of the refusals can be
offered an incentive to participate, or be offered a less
intensive protocol for their participation; the data from
14
-------
the would-be refusals can then be compared with the "regular1
participants to assess the possible magnitudes of selection
bias.
3. Modeling
As an alternative to direct personal monitoring of human exposure,
the indirect modeling approach offers certain attractive features and
has the potential to be cost effective. The essence of the indirect
approach is to assess human exposure as it is related to human activities,
and to combine pollution measurements with activity data (which could be
collected separately) to estimate exposure. Simulation models such as
SHAPE and NEM are variants of this approach in which either part or all
of the input data are simulated or imputed.
Compared to the direct personal monitoring approach, the indirect
approach has the following advantages:
a. Existing pollution arid activity data can be reused in comparable
future studies, thereby reducing the cost of future studies;
b. The human subjects need not be burdened with the inconvenience of
carrying personal monitors during their activities;
c. In the absence of feasible personal monitors, it might still be
possible to estimate exposure using the indirect approach as a
stop-gap measure; and
d. In conjunction with source-transport models, the indirect
approach can be used to impute the would-be exposures under
alternative regulatory strategies.
The presentation at the review meeting was focused on one aspect of
the indirect approach, namely, the use of microenvironmental monitoring
to collect pollution data. The indirect approach can also be implemented
with pollution data from a personal monitoring study. For example,
pollution data from an earlier personal monitoring study can be combined
with the activity data from a comparable sample in a new study. Another
example is enhanced personal monitoring, in which extra activity data
are collected to make more efficient use of the pollution data. We
recommend the collection of activity data in future studies whenever
feasible.
The indirect approach remains to be validated empirically with
comparable personal monitoring data. We endorse the plan to validate
the SHAPE model with actual field data and the mentioned effort by the
Agency's Office of Air Quality Planning and Standards (OAQPS) to validate
and modify the NEM model to accept empirical activity data as input.
Such validation would enhance the use of those models in future studies.
15
-------
However, it is neither clear which microenvironments and chemicals are
of greatest importance nor how to define each microenvironment. Further,
identification of the microenvironmental scenarios which will be most effective
at modeling exposure must be considered. Subsequent to this prioritization
and analysis of microenvironments, cost effective research in this field
can be planned.
Although the simulation modeling approach to estimating total human
exposures is not without difficulties and weaknesses, it is less costly
and potentially more generalizable than direct measurements. The most
serious problem with the modeling approach is that, before any model can
be used with any confidence, its validity must be carefully and completely
evaluated. This requires large quantities of experimental data from
specially designed field studies. Failure to obtain a probability sample
for microenvironment monitoring is a shortcoming of the Washington micro-
environment study, and needs to be addressed in future studies.
Logically, the direct (measurement) and indirect (modeling) approaches
should be considered complementary, in that the former provides essential
information for the development, validation, and refinement of the latter.
Hence, the Panel recommends that the planning for human exposure field
studies be carefully designed specifically to provide for the needs of
exposure models.
4. Dosage Research
The Panel's review of the dosage research program was relatively
brief and occurred at the end of a long day. Hence it was, of necessity,
somewhat superficial. Nevertheless, we found that the choice of assays
to be included in this program review by the EPA staff was reasonable
and that productive work was being done. The people involved appeared
to have a good grasp of the scientific issues and potential of their
methodology, and to be capable of undertaking additional productive work
in this important area. The following are some specific comments:
a. The results of the analyses of volatile organic chemicals
(VOC's) in exhaled air in the TEAM study provide an excellent
example of the utility of such measurements in field studies.
They showed that most of the VOC's were traceable to indoor
rather than outdoor sources of air pollution, and identified
many previously unsuspected causes of human pollutant exposures.
b. The results of the measurements of alveolar carbon monoxide
in Denver and Washington were very useful in demonstrating
that ambient carbon monoxide concentrations had very little
effect on actual variation in individual overall exposures to
carbon monoxide.
16
-------
c. Current research on the use of short-term mutagenesis bioassays
offer promise of developing reliable and efficient assays for
source-specific pollutant mixtures from common indoor pollution
sources such as cigarette smoke, wbodstoves and kerosene heaters,
Research of this kind may prove to be useful to the EPA
for many other applications involving exposure to organic
vapors such as waste recovery and disposal sites.
d. Evidence for human exposure to pollutants can often be made
from analyses of biological materials such as exhaled air,
urine, blood, hair, etc. Such analyses are particularly
valuable when there are multiple routes of exposure, and/or
highly variable levels of exposure. Biological samples can
show evidence for cumulative exposures, and can indicate
whether adverse effects are occurring or are likely to occur.
17
-------
APPENDICES
-------
REVIEW PANEL ON TOTAL HUMAN EXPOSURE
CHARGE
The purpose of the total human exposure research program is to
determine the frequency distributions of exposures of the pop-
ulations to selected chemicals. The issues center around five
major topic areas:
1) Development of measurement methods,
2) Statistical protocols, data base development,
and exposure models,
3) Microenvironmental field studies,
4) Total human exposure field studies,
5) Dosage research investigations.
The Review Panel on Total Human Exposure shall advise the
Agency on how well the research program for Total Human Exposure
is progressing toward answering the needs of the Agency, including
both whether or not the right research is being conducted to address
the issues defined by the Research Committee process, and whether
the research is being conducted properly. The Panel's findings and
recommendations will be presented both in a report and in a briefing
to be presented to the Deputy Administrator.
A-l
-------
U.S. Environmental Protection Agency
Science Advisory Board
Total Human Exposure Review Panel
Roster
Co-Chairmen
Dr. James Ware, Department of Biostatistics, Harvard School of
Public Health, 677 Huntington Avenue Boston, MA 02115
Dr. Morton Lippmann, Institute of Environmental Medicine, Lanza
Laboratory, Long Meadow Road, New York University, Tuxedo,
NY 10987
Members
Dr. Lynn Anspaugh, Lawrence Livermore National Laboratory, P.O.
Box 5507; L-453, University of California, Livermore, CA 94550
Dr. Naihua Duan, Economic Department, Rand Corporation, 1700
Main Street, Santa Monica, CA 90406
Dr. Warren Johnson, Director, Atmospheric Science Center, Advanced
Development Division, SRI International, 333 Ravenswood Avenue,
Menlo Park, CA 94025
Dr. John Spengler, Harvard School of Public Health, Department
of Environmental Science and Physiology, 665 Huntington Avenue,
Boston, MA 02115
Dr. Joseph Stetter, Argonne National Laboratory, Building EES 362,
9700 Cass Avenue, Argonne, IL 60439
Dr. Jerry Wesolowski, California Department of Health,
2151 Berkeley Way, Berkeley, CA 94704
Executive Secretary
Mr. Robert Flaak, Science Advisory Board (A-101F), U.S. Environmental
Protection Agency, 401 M Street, S.W., Washington, DC 20460
B-l
------- |