United States ;,:
Environmental Protection
Agency ; ;
Office of Research and
Development .--. • ':
Washington DC 20460
EPA/630/R-00/005
December 2000
Summary Report of the
Technical Workshop on
Issues Associated with
Considering Developmental
Changes in Behavior and
Anatomy when Assessing
Exposure to Children
RISK ASSESSMENT FORUM
-------
-------
EPA/630/R-00/005
December 2000
Summary Report of the Technical Workshop on Issues
Associated with Considering Developmental Changes in
Behavior and Anatomy when Assessing Exposure to
Children
U.S. Environmental Protection Agency
Washington, DC
July 26-27,2000
Risk Assessment Forum
U.S. Environmental Protection Agency
Washington, DC 20460
Printed on Recycled Paper
-------
-..'.:. -.'.- , •- NOTICE
This document has been reviewed in accordance with U.S. Environmental Protection Agency
(EPA) policy and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
This report was prepared by Eastern Research Group, Inc., an EPA contractor (Contract No. 68-
C9-8148, Work Assignment Nos. 00-01 and 01-03) as a general record of discussion held during
the Technical Workshop on Issues Associated with Considering Developmental Changes in
Behavior and Anatomy When Assessing Exposure to Children (July 26-27,2000). As requested
by EPA, this report captures the main points and highlights of the meeting. It is not a complete
record of all details discussed, nor does it embellish, interpret, or enlarge upon matters that were
incomplete or unclear. Statements represent the individual v,iews.,of .each workshop,.participant,
none of the statements represent analyses by or positions of the FLisk AssessmentForum or me
EPA. ' :•:,. "''• '•••:••'. .-'•'•: • ;.= " '.•> •>•-..'• ;G^.^'/'> 4':' r ^vm^vV'vR
•i-f
11
-------
CONTENTS
FOREWORD •.;..;,-,.. ...V,<.; .%!.,-,;. .,.,'.'.%.;V.:.-v
EXECUTIVE SUMMARY , •'.,...,..;,,.,,. ,;,<...... -,y ...:•':..•..... vi
1. : . INTRODUCTION ,.,.,..., i.,.,,.,. ,•;.. ...:.,,..:..-] .•, ..';:.',..•:• /. ••.? 1-1
1.1 Workshop Purpose ......,...'......... 1-1
1.2 Workshop Participants 1-1
1.3 Charge to the Panel ,,........... :.\ ....,,, 1-1
1.4 Agenda 1-2
1.5 Workshop Summary ......................... ...;..,,........ 1-2
2. SUMMARY OF OPENING REMARKS . ............. .... ....... 2-1
2.1 Welcome '. 2-1
2.2 Background on EPA's Risk Assessment Forum ,.... -. 2-1
2.3 Children's Exposure Assessment at EPA: Current Practices and Future Needs 2-2
, .,2,4 : Methods of Exposure Assessment for Children ....;............... • •.--,• • • • 2r5
2.5 Age Bins in Existing Data Sets That Are Relevant to Children's Exposure
Assessment 2-12
,-• 2.6 Charge to the Experts;.;,,..;......... ^.::.. ^.. .;.s,.v.»j;,..-...-:; .,,i.,..., 2-18
:,.-, ,v ., ,:.v.r-t •>,:',•"( i •. •••r-'1" ••'•'>•. .i-.».-.. %•:'* <'••''.•-. \"'. .*-.':? ;•-«•• r '';
3. CHANGE IN CfflLDREN?S EXPOSURE DUE TO BEHAyiORAL
DEVELOPMENT ........ 3-1
3.1 The Defensibility of Representing Behavioral Difference in Terms of
Age Bins 3-2
3.2 Behavioral Development and Oral Exposure 3-6
3.3 Behavioral Development and Dermal Exposure 3-8
3.4 Behavioral Development and Inhalation Exposure 3-10
3.5 General Conclusions for Incorporating Behavioral Development into
Children's Risk Assessment 3-11
3.6 Research Recommendations 3-15
4. CHANGE IN CHILDREN'S EXPOSURE DUE TO ANATOMICAL
DEVELOPMENT 4-1
4.1 Individual Anatomical Characteristics, Organs, and Systems 4-2
4.2 General Issues 4-24
4.3 Research Recommendations 4-28
m
-------
S. FINAL PLENARY DISCUSSION 5-1
APPENDIX A: LIST OF PANELISTS , A-l
APPENDIX B: LIST OF OBSERVERS B-l
APPENDIX C: CHARGE TO THE EXPERTS C-l
APPENDIX D: AGENDA D-l
APPENDIX E: PRESENTER OVERHEADS , E-l
Overheads from Dr. Firestone's Presentation E-3
Overheads from Dr. Hubal's Presentation E-15
Overheads from Dr. Thompson's Presentation E-33
Overheads from Dr. Walker's Presentation E-57
APPENDIX F: BEHAVIOR GROUP OVERHEADS F-l
APPENDIX G: CHILDREN'S EXPOSURE ASSESSMENT (ISSUE PAPER PREPARED
BY ELAINE HUBAL) G-l
APPENDIX H: CHANGES IN CHILDREN'S EXPOSURE AS A FUNCTION OF AGE
AND THE RELEVANCE OF AGE DEFINITIONS FOR EXPOSURE AND RISK
ASSESSMENT (ISSUE PAPER PREPARED BY KIMBERLY THOMPSON) ..... H-l
IV
-------
FOREWORD
This report presents information and materials from a peer involvement workshop organized by
EPA's risk Assessment Forum. The meeting was held in Washington, DC on July 26 and 27,
2000. The meeting discussions focused on how to consider age related changes in behavior and
physical development when assessing childhood exposures to environmental contaminants.
These discussions are part of EPA's ongoing efforts to improve the assessment of risks to
children.
The 1993 National Academy of Sciences (NAS) report "Pesticides in the Diets of Infants and
Children" highlights important differences between children and adults with respect to risks
posed by pesticides. Some of the principles in the NAS report provided the foundation for the
Food Quality Protection Act of 1996 (FQPA) and the President's Executive Order 13045,
Protection of Children from Environmental Health Risks and Safety Risk. FQPA requires the
consideration of aggregate exposure to children when establishing pesticide tolerances (legal
limits for residues in food). Executive Order 13045 broadens consideration of impacts on
children by stating that "each Federal agency: shall ensure that its policies, programs, activities,
and standards address disproportionate risks to children that result from environmental health
risks or safety risks." Many of the comments the EPA received on the Proposed Guidelines for
Carcinogen Risk Assessment relate to the implementation of Executive Order 13045. In
response to these comments and regulatory initiatives, EPA has been investigating ways to
improve Agency risk assessments for children.
An Agency workgroup convened under the auspices of the Risk Assessment Forum has been
exploring children's exposure assessment issues. This workgroup has concluded that a major
issue facing Agency assessors is how to consider age related changes in behavior and physiology
when preparing exposure assessments for children. Children's behavior changes over time in
ways that can have an important impact on exposure. Further, children's physiology changes
over time in ways that can impact both their exposures and their susceptibility to certain health
effects. There are two aspects to these physiological changes. First, there are anatomical
changes resulting from physical growth. Second, there are changes in pharmacokinetics and
pharmacodynamics which affect the absorption, distribution, excretion and effects of
environmental contaminants. The Agency is examining the pharmacokinetic/pharmacodynamic
changes in children through other efforts and future meetings on this topic are anticipated. The
July 26 and 27, 2000 workshop focused on incorporating age related changes in behavior and
anatomy into Agency exposure assessments.
BILL WOOD
-------
EXECUTIVE SUMMARY
On July 26 to 27,2000, the U.S. Environmental Protection Agency (EPA) sponsored a workshop
to discuss issues associated with considering developmental changes in behavior and anatomy
when assessing exposure to children. The workshop panel comprised 22 experts in toxicology,
exposure assessment, risk assessment, and pediatrics from universities, state and federal
government, industry, and medical centers.
An opening plenary session provided background on the workshop's purpose as well as EPA's
activities and the availability of data regarding exposure factors for children's exposure
assessment. Panelists then divided into two discussion groups—one focusing on behavioral
changes during childhood and their impact on exposure to environmental contaminants, and the
other focusing on anatomical and physiological changes during childhood and then- impact on
exposure to environmental contaminants. Each discussion group met for about 8 hours to define
and characterize the important behavioral and anatomical facets of child development related to
exposure, to discuss how best to incorporate scientific and medical knowledge about childhood
development into the practice of exposure assessment, and to suggest what research should be
conducted to fill critical data gaps and enhance child-related exposure assessment. The
conclusions and recommendations developed by each group were presented in a final plenary
session. Key conclusions and recommendations included the following:
* Both groups noted the limitations inherent in using age bins to characterize
developmental change during childhood. They emphasized that while development is a
series of discrete events, these events occur along a number of continua. There is
considerable variability about when a change begins and ends, and some behavioral
patterns, once initiated, may never end. Race, ethnicity, culture, and socioeconomic
factors, as \vell as genetics, may contribute to the variability. For these reasons,
•developmental change for both behavior and anatomy should ideally be characterized as
distributions.
• Bins may be useful as a guide to the development of exposure scenarios, but EPA should
always keep in mind that bins are only a crude approximation of an underlying
distribution. Age bins, if used uncritically during exposure assessment, could lead to
significant error. Exposure assessors need to have an understanding of the biological
phenomena underlying age bins.
• Both groups offered preliminary ideas about possible bins for developmental change
related to exposure. However, they emphasized that these were based on very limited
discussion, working from general knowledge, and were provided only as a starting point
for further work. Even after further work, research would be needed to refine and
validate those bins over time.
VI
-------
Plata for both behavioral and anatomical exposure factors, are, limited in terms, of both
quality and coverage,; The adequacy of current data .sets is highly variable, and some of
the data may not be useful because they were gathered using outdated methods or because
lifestyle changes since the study was conducted make the results less relevant to today's
conditions., More up-to-date; data would be useful for refining distributions of critical •
developmental periods. ; ..,-••. •. ,i.••••:..
Despite the limitations of current data sets, more extensive use, of available datai.on child
development relative to exposure will likely greatly reduce the enormous errors currently
made when such data are ignored. An age bin approach would therefore be an
improvement over the status quo, however, in the longer term, the panel would prefer a
distributional approach.
To identify and fill data gaps, EPA should first have developmental specialists in the
areas of behavior and anatomy/physiology conduct an in-depth review of the literature to
determine what data are available and to evaluate the data in terms of methodology,
reliability, sample size, relationship to current exposure conditions, and variability. The
experts felt that a considerable amount of useful information already exists, albeit
somewhat dispersed, in the literature. A short-term goal could be to assemble this
information to examine what we know about distributions underlying the bins that might
be utilized pending development of models that can incorporate distribution.
A long-term project would be the development of integrated data sets (combining
information about children's behavior and anatomy, their estimated exposure, their
biomarkers for particular chemicals, and their health) to be used to evaluate the relative
importance of different kinds of exposure in order to identify exposure pathways that
appear to be associated with the most significant risks. Future research can then be
focused on developing data for exposure factors that appear to have the greatest
significance for risk. The long-term goal should be to develop good statistical data on
distributions for behavioral and anatomical change that can account for the variability
inherent in childhood development; the statistical data would be correlated with
biomarkers and clinically important endpoints.
Any physiological data for children will be inextricably linked to toxicokinetic and
toxicodynamic issues that must be taken into account when considering age bins.
Although the indirect exposure assessment approach can be valuable when direct data are
not available, some panelists felt that direct assessments are not necessarily too expensive
or too difficult to be conducted. This was thought to be feasible at least for the more
prevalent toxic chemicals and for the more prevalent exposures. Such studies should
incorporate information about both exposure and biomarkers of exposure.
VII
-------
Prenatal development was not discussed since it was outside the scope of the workshop,
but both groups strongly recommended that EPA look closely at maternal-fetal exposure,
since in utero development is such a critical and sensitive period.
Panelists felt that the interdisciplinary nature of the panel contributed significantly to the
quality of the discussion and recommended that the agency continue to involve a broad
range of specialists, including pediatric and obstetrical subspecialists, public health
specialists, exposure and risk assessors, and toxicologists in further discussion about
children's exposure.
viu
-------
^INTRODUCTION
1.1 Workshop Purpose
The Technical Workshop on Issues Associated with Considering Developmental Changes in
Behavior and Anatomy When Assessing Exposure to Children was held on July 26 and 27,2000,
in Washington, D.C. The workshop was sponsored by the U.S. Environmental Protection
Agency's (EPA's) Risk Assessment Forum, which has been exploring children's exposure
assessment issues. The purpose of the workshop was to discuss issues associated with
considering developmental changes in behavior and anatomy when assessing children's exposure
to environmental contaminants.
1.2 Workshop Participants
Panelists at the workshop consisted of 22 experts, including pediatricians, lexicologists, risk
assessors, and public health professionals from industry, universities, consulting, and state and
federal government agencies. Over 50 observers attended the workshop. Panelists and observers
. are listed in Appendices A and B, respectively.
1.3 Charge to the Panel
The complete charge to panelists is provided in Appendix C. Both behavioral and anatomical
changes over time can affect children's exposure; anatomical changes can also affect children's
susceptibility to certain health effects. Panelists were asked to focus their discussions on
defining and characterizing the important facets of behavioral and anatomical development
during childhood and on how best to estimate children's exposure given the limitations in
existing exposure information. They were asked to focus on broad issues (rather than specific
methodologies) and not to address pharmacokinetic issues, since these are being evaluated in a
1-1
-------
separate effort. They were also asked to consider whether existing exposure information is
adequate and what research should be conducted to enhance children's exposure assessment.
1.4 Agenda
The workshop agenda is provided in Appendix D. The workshop began with welcoming
remarks; a presentation about EPA's Risk Assessment Forum; and a presentation on the current
practices and future needs of EPA's Office of Children's Health Protection with respect to the
conduct of children's exposure assessments. These were followed by two technical
presentations. The first, given by a member of EPA's National Exposure Research Laboratory,
described the algorithms and parameters currently used to conduct indirect exposure assessments.
The second, given by a member of the Harvard Center for Risk Analysis, discussed how well the
different parameters mentioned in the previous presentation are currently documented in the
research literature.
The formal charge to the experts was then presented and the participants divided into two
discussion groups corresponding to the two sections of the charge. One discussion group was
charged to consider developmental changes hi children's behavior-related exposure factors and
the other was charged to consider developmental changes in children's anatomical exposure
factors. (Discussion group chairs and members are listed in Section 2.6.) The discussion groups
met for 3 hours and then reconvened in a brief plenary session to summarize their progress before
adjourning for the evening. The discussion groups resumed their work the next day and, after 6
hours of further deliberation, presented their findings to each other at the final plenary session.
Open discussion among the full panel of experts continued after the presentations.
1.5 Workshop Summary
This report summarizes the workshop presentations and discussions and is organized as follows:
1-2
-------
Section 2 of this report summarizes the presentations. Overheads used by the
chairperson, EPA presenters, and a commenting observer are provided in Appendix E.
The background papers that the presenters refer to are provided in Appendices G and H.
Section 3 and 4 report the conclusions that the two discussion groups (concerned with
behavior-related and anatomy-based exposure factors, respectively) presented at the
conclusion of the conference. Overheads used by the behavior-related discussion group
in presenting their results can be found in Appendix F.
Section 5 summarizes the final plenary discussion on issues related to assessing
children's exposure.
1-3
-------
-------
2. SUMMARY OF OPENING REMARKS
2.1 Welcome
Jan Connery of Eastern Research Group, Inc. (ERG), opened the workshop by welcoming
participants and observers. She introduced Dr. Kimberly Thompson, the workshop chair, and
asked other workshop participants to introduce themselves to the group. Dr. Thompson added
her own welcome to the participants and began the introductions. After the introductions, Ms.
Connery reviewed the workshop agenda and introduced the first speaker.
2.2 Background on EPA's Risk Assessment Forum
Bill Wood, Executive Director of EPA's Risk Assessment Forum (RAF), provided background
on the RAF and its broad goals in sponsoring the workshop. The RAF is a standing committee
within EPA that is responsible for providing agency-wide guidance in the area of risk
assessment. This workshop is one part of a larger consensus-building process that the RAF is
undertaking to improve its understanding of children's risks from environmental contaminants.
The RAF is currently revising EPA's cancer risk assessment guidelines. It is engaged in an
ongoing discussion with EPA's Science Advisory Board (SAB) about improvements that could
be made to these guidelines so as to better address children's cancer risks. The results of these
discussions should be released soon. In parallel with RAF's efforts to incorporate the issue of
children's risks into existing cancer risk assessment guidelines, there are several other agency-
wide programs to better address children's risks. The RAF, in this and previous workshops, is
trying to capture the kinds of expertise that are rapidly emerging (both within and outside EPA)
in connection with children's risk assessment.
At present, there is some consistency and some variation in how EPA's different departments
consider children's risks. The variations are rooted in the different kinds of data, decisions, and
2-1
-------
pieces of legislation with which the different departments work. The RAF would like to take a
broader perspective as it develops agency-wide guidance pertaining to these risks, and it
recognizes that such a perspective ought to be carefully developed in a consensus-building
process. Dr. Wood concluded by thanking the RAF Technical Panel for their assistance in this
regard. The Technical Panel is an advisory group composed of senior scientists from across the
Agency.
2.3 Children's Exposure Assessment at EPA: Current Practices and Future Needs
2.3.1 Presentation
Next, Dr. Michael Firestone, Science Director at EPA's Office for Children's Health Protection,
delivered a presentation on EPA's current policies for assessing children's environmental risks
and how the agency would like to further develop these policies. Dr. Firestone remarked that it
did not take him long after joining the Office for Children's Health Prevention to recognize the
key precept around which the Office is organized: "Children are not little adults." It is an
overarching goal of the Office to help others, within and outside EPA, to understand the specific
ways hi which children must be considered differently than adults. He provided several
examples of how children differ physiologically and behaviorally from adults:
• Children eat and drink more for their size than adults do.
• Children play arid act differently than adults do: very young children have more contact
with ground surfaces than adults and engage in a great deal of hand-to-mouth activity.
• Children's bodies are undergoing development.
• Children may be less able to metabolize and excrete certain toxic substances.
While it has been an important first step for EPA to distinguish children from adults, further
distinctions are necessary. The need for a more refined approach to considering children is
apparent from the fact that some of the rapid changes in human development take place within
2-2
-------
the first few years of life—it is the purpose of the workshop to gather advice from the
participants about how to take these developments into account and better.assess children's
exposure. Dr. Firestone referenced ;an earlier comment by Dr. Lynn Goldman, professor of.
Public Health at Johns Hopkins University: EPA should not consider children a "sub-population
of concern" but rather a "life-stage of concern." .
The ultimate goal of exposure assessment is to develop a day-to-day model of. human life that can
predict the chemical exposures an individual is likely to face at any point in his or her life. While
this is a laudable goal, it is not likely to be realized in the foreseeable future, so risk assessors
need to develop simpler models. One way to simplify exposure models is to classify individuals
into age bins, though some may be concerned that this procedure leads to over-simplification.
Different programs within EPA have been attempting to develop default approaches (including
the use of age bins) to address children's exposure when data are sparse. The different default
approaches ought to be replaced by a standardized approach that is based on science and that
provides justifications based on evidence: EPA has convened the workshop to gain insight and ,,
input into factors it should consider when developing such a standardized approach, as'well as to
identify what further scientific research may be necessary to accomplish these goals.. ,.
Recent EPA actions to improve assessment of children's exposure include:
The EPA Rule-Writer's Guide released in 1998. The guide is designed to help program
offices incorporate children's risks into their assessments. Under the auspices of EPA's
Science Policy Council, the Office for Children's Health Prevention is currently ,
reviewing the usefulness of this guide to the different program offices.
The Child-Specific Exposure Factors Handbook, which is currently undergoing peer
review. In 1997, EPA issued the latest version of the Exposure Factors Handbook, which
contains exposure factors useful for probabilistic risk modeling. EPA's Office of
Research and Development (ORD) has been developing a child-specific version of the
exposure factors handbook this year. This new document is presently in draft form.
2-3
-------
Efforts, such as those described above, to estimate children's exposure are hampered by the
uneven coverage of existing data sets. For example, the NHANES data on the biological
monitoring of pesticide exposure include no data for children under the age of 6.
Recent EPA risk assessments have tried to address the following age groups: fetuses, infants,
toddlers, children, and adolescents. There has been some variation, however, in the particular
age ranges attached to some of these qualitative categories.
In conclusion, Dr. Firestone expressed particular hope that the participants would provide
guidance in the following areas:
• Defining age bins more effectively by carefully identifying the particular characteristics
that distinguish them.
* Deciding how finely EPA should subdivide the overall life stage of childhood into age
bins.
» Describing how additional factors such as sex, culture, and geography might modify the
significance of standard age bins.
• Identifying the most pressing gaps in the base of scientific knowledge that would justify
age bins.
2.3.2 Questions and Comments
Dr. William Weil asked Dr. Firestone if he could more closely define the percentage of children
that he hoped to describe in each age group. For example, was a particular age bin meant to
accurately describe the average child or a range of children? If a range, was a bin meant to
describe 90 percent of the children within that age group, 95 percent, or 99 percent? Dr.
Firestone replied that the EPA definitely wanted to study distributions as far out as they could be
measured. The particular cutoff points used in risk assessments would depend on the particular
risks being considered. Dr. Weil continued to express some confusion about how EPA intended
to use age bins to summarize the widely varying development of complex organ systems. Dr.
2-4
-------
Robert Johnson suggested that it might be more appropriate to base exposure assessments
directly on the relevant behavioral and physiological properties of the child rather than by
generalizing from standard age categories. Dr. Firestone agreed that such a sophisticated
approach might be a good long-term goal for EPA, but cautioned that it was not a realistic short-
term goal. Dr. Melanie Marty suggested that while models based on age bins might often be
adequate, users of age bins should be alert to the complexities that underlie them. There may be
cases in which a specific factor (such as mouthing behavior) is a more significant indicator of
exposure than age. Dr. Firestone agreed with this caveat.
2.4 Methods of Exposure Assessment for Children
2.4.1 Presentation
Next, Elaine Hubal of EPA's National Exposure Research Laboratory (NERL) delivered a
presentation on some of the techniques presently used to assess exposure in children. Dr. Hubal
began by remarking that much of her work is oriented toward defining the particular kinds of
data that, if available, would be most helpful for use in exposure assessments. The definition of
human exposure is the contact (at some visible, external boundary for some period of time) of an
individual with a pollutant. It is important to distinguish exposure from dose, even though the
two concepts are related, as individuals do not necessarily absorb into their bodies all the
chemicals they are exposed to. Exposures can be measured either directly or indirectly:
Direct exposure assessment involves actually measuring the chemicals that an individual
is exposed to, using tools like personal air monitors or techniques like duplicate diet
sampling. Biomonitoring tests are useful as indicators of direct exposure, but it is often
difficult to develop quantitative exposure estimates from the results of these
measurements.
Indirect assessments estimate exposure from data about chemical concentrations in an
exposure medium (e.g., soil, toys, the floor). Concentration data are combined with
information about how an individual interacts with the exposure medium, and a series of
2-5
-------
exposure factors, to arrive at an estimate of personal exposure to the chemical in the
medium.
NERL is particularly interested in improving knowledge about the indirect exposure factors
involved in the transfer of chemicals from contaminated exposure media to children, whether by
inhalation, dermal contact, or ingestion. In general terms, these factors are the:
• Concentration of chemical in exposure medium.
• Contact rates of the individual with the medium.
• Contaminant transfer efficiency from the medium to the portal of entry.
" Contaminant uptake rates.
• Human activity patterns.
Dr. Hubal went on to discuss some of the characteristics of children that influence exposure.
With respect to physiological effects, she distinguished between those that affect a child's
susceptibility to toxic chemicals (e.g., growth in an organ system creating a window of
vulnerability) and those that affect a child's exposure to those chemicals (e.g., changes in food
consumption, respiration, and surface area to body weight ratio). There are many kinds of
specific developmental changes that are of interest to exposure assessors. When a child acquires
the ability to crawl, walk, run, or manipulate objects, his or her potential exposure changes
significantly. These different developmental capabilities affect the different environments to
which a child has access. Changes in how and what a child eats affect his or her exposure to
foodborne environmental contaminants. Other factors, such as gender, socioeconomic status,
race, and ethnicity are also extremely important because they can affect the location, quality, and
intensity of many other behaviors.
Dr, Hubal began summarizing the equations used to estimate children's exposure from sets of
exposure factors. They are included in the document titled Children's Exposure Assessment: A
Review of Factors Influencing Children's Exposure and the Data Available to Characterize and
2-6
-------
Assess That Exposure, which can be found in Appendix G and which has been published in the
June 2000 issue of Environmental Health Perspectives (volume 8, number 6, page 475). With
respect to these equations, she indicated that the exposure pathway of ingestion can be broken
down further into a dietary pathway (eating) and a non-dietary pathway (placing fingers and
Objects in one's mouth). Dietary ingestion pathways can be broken down further to include both
the contaminants present in the food itself and contaminants that get onto the food as it is
consumed.
The characterization of a child's activity patterns requires several kinds of information. The first
kind of information describes a child's microenvironment: it provides a specific and detailed
description of the place a child occupies during an activity (e.g., indoors in a kitchen, outdoors on
a lawn). The second kind of information is macroactivity: it is a general description of what a
child is doing (e.g., watching television, eating dinner, taking a shower). The third kind of
information is described as microactivity: the specific physical acts that are characteristic of a
macroactivity (e.g., the number of times a child touches the floor per hour while watching
television).
Children's inhalation exposure is relatively well characterized: it depends on atmospheric
pollutant concentration in the particular microenvironment where a child is located, that child's
rate of inhalation, and the length of time spent in the .microenvironment. There are four studies
that provide macroactivity data for children over a single day. These can be accessed through the
Consolidated Human Activity Database (CHAD) and used to estimate inhalation exposure.
There are some problems with these data: they are not longitudinal, they do not provide detailed
enough microenvironment information to make it possible to estimate other kinds of exposure
pathways (such as dermal exposure), and the macroactivity categories were developed for adults .
rather than for children.
Dermal exposure can be estimated with one of two alternative equations:
2-7
-------
Equation 1: Macroactivity Approach. To estimate dermal exposure using the
macroactivity approach, microenvironments are defined by location and surface type
(e.g., indoors at home on carpet). The dermal exposure associated with a given
macroactivity (e.g., actively playing in the yard) is measured and used to develop an
activity- and microenvironment-specific transfer coefficient. Exposure can then be
estimated individually for each of the microenvironments where a child spends time and
each macroactivity that the child conducts within that microenvironment. Exposure over
the 24-hour period is the sum of all of the microenvironment/macroactivity (me/ma)
exposures. For each microenvironmental/macroactivity (me/ma), dermal exposure over
the 24-hour period (Edme/ma) is defined as:
-'dme/ma"
xTC, xED
•'der
Where:
Csurf = total contaminant loading on surface (|ig/cm2)
TCder = dermal transfer coefficient for the me/ma (cnr/hr)
ED = exposure duration that represents the time spent in the me/ma (hr/day)
Equation 2: Microactivity Approach. To assess dermal exposure using the
microactivity approach, exposure is estimated individually for each of the microactivities
or events (e.g., each time a child touches a given object) from which dermal contact or
non-dietary ingestion occurs. Exposure over the 24-hour period is then the sum of all of
the individual exposures. For each microactivity, dermal exposure over the 24-hour
period (Eder/mi) can be defined as:
Eder/mi = CsurfxTExSAxEF
Where:
Eder/mi = dermal exposure for a given microactivity over a 24-hour period (jag/day)
Csurf= total contaminant loading on surface (jj-g/cm2)
TE = transfer efficiency, fraction transferred from surface to skin (unitless)
SA = area of surface that is contacted (cnr/event)
EF = frequency of contact event over a 24-hour period (events/day)
The first equation is simpler, but it has traditionally been used in agricultural rather than home
environments, so it needs to be tested in the residential environment with children. It uses a
single, lumped transfer coefficient to relate the contaminant loading on a surface to an
individual's rate of exposure for each hour spent in a given microenvironment/macroactivity
combination. Dermal transfer coefficients must be developed empirically. To do so will require
2-8
-------
studying groups of children — it is essential that these groups be selected according to appropriate
age bins in order to minimize the variability of the measurements.
The second and more complex equation is based on detailed microactivity data. Instead of a
single, lumped transfer coefficient, it uses data about how often a particular surface is.touched
per hour, the surface area that is contacted with each touch, and the transfer efficiency of the
contaminant from the surface to the skin. This methodology requires extremely detailed
information that is often slow, costly, and challenging to collect.
Non-dietary ingestion exposure can be estimated through an equation similar to the second,
microactivity-based dermal transfer equation:
For each microactivity resulting in non-dietary ingestion, exposure over the 24-hour
period (Ending/mi) can be defined as:
Ending/mi = C
x TExm
SAX x EF
Where:
Ending/mi = non-dietary ingestion exposure for a given microactivity over a 24-hour period
(ug/day)
x = hand or object that is mouthed
Cx = total contaminant loading on hand or object (jag/cm2)
TExm = transfer efficiency, fraction transferred from object or hand to mouth (unitless)
SAX = area of object or hand that is mouthed (cm2/event)
EF = frequency of mouthing event over a 24-hour period (events/day)
Determining indirect dietary ingestion involves measuring the detailed patterns of a food item's
contact with contaminated surfaces before it is consumed.
Dr. Hubal emphasized that the actual values and distributions of exposure factors are developed
in the context of specific exposure scenarios. An exposure scenario defines a particular source
for a chemical (e.g., the use of a chemical hi the home), a particular population that is potentially
exposed, and the timeframes, microenvironments, and macroactivities that are associated with
2-9
-------
the exposure pathway for that chemical. Dr. Hubal concluded by emphasizing that proper
selection of children's age categories will be crucial to the effective development of field studies
to characterize exposure factors for different scenarios.
2.4.2 Questions and Comments • . -
Dr. William Weil inquired whether air sampling measurements were conducted at the height of ,
adults or children. Dr. Hubal replied that although some studies have found significant
differences hi chemical concentrations between adult and child heights, many have not. She still
agreed with Dr. Weil that it was an important distinction to bear in mind, along with other
activity pattern data. Dr. Weil asked whether or not children's air exposure factor measurements
were conducted at the appropriate heights, noting that some compounds on ground surfaces have
limited volatility that can create layers of airborne chemicals confined to 6 inches above the
surface. Children who crawl are therefore likely to have very different inhalation exposures than
children who can walk. Dr. Hubal replied that, as yet, few exposure assessment studies have
been conducted that specifically consider children. The few assessments geared toward children
do sample air at the appropriate heights.
Dr. Richard Fenske asked for more information about how exposure scenarios are built and about
how many of them EPA puts together. They appear to be the crucial place where all the different
exposure factors come together. Dr. Hubal said that she did not believe exposure scenarios were,
at present, being developed systematically. She hoped that more information about age
differences among children would help exposure assessors to make better decisions about when
they can lump children together into a single scenario and when they need to develop separate
scenarios for two different age ranges. Dr. Fenske explained that at present, exposure assessors
first identify a scenario and then develop the particular data necessary to define that particular
scenario. He wanted to know whether EPA was hoping to develop a set of scenario-independent
life stages with associated exposure factors. Dr. Hubal believed that was EPA's goal. She
2-10
-------
commented further that general, scenario-independent data on children's activity patterns would
be an important resource for the proper development of specific scenarios of children's exposure.
Dr. Marty expressed some skepticism that toxicokinetic issues could be separated from exposure
issues. Any physiological data that are available about children are inextricably linked to
toxicokinetic and toxicodynamic issues. The participants, she continued, will not be able to
avoid these issues when they consider age bins. There is a concern that if one ignores
toxicological factors, one will pay inadequate attention to a particular exposure factor that is very
important to a particular chemical exposure. Dr. Hubal replied that EPA does not mean to ignore
toxicological factors—it is addressing them in another meeting and it is simply trying to focus
the present workshop on issues of exposure. Dr. Firestone suggested that assessments should be
developed as matrices of different developmental stages and different toxins of concern. A
matrix will be filled in where exposures are particularly high and toxicological risk is particularly
great. Dr. Firestone was concerned about allowing different EPA programs to define age bins in
an idiosyncratic fashion. Dr. Bruce Lanphear admitted that indirect exposure assessment was
valuable in cases where no direct data are available, but he expressed concern about the attitude
that direct assessment is too difficult and expensive to be useful. Dr. Lanphear indicated that he
did not believe direct assessments were too difficult and expensive to be conducted, at least for
prevalent toxic chemicals and prevalent exposures. These studies should incorporate information
about both exposure and biomarkers of exposure. Ultimately, he observed, direct assessment is
needed to validate the results of the "mechanistic models" of indirect assessment. Dr. Hubal
agreed with these comments but cautioned that it is not always possible to directly measure how
much of a chemical an individual absorbed and how that individual absorbed it.
Dr. Gary Ginsberg asked for a sense of how well studied and standardized the process of
developing transfer coefficients had become. Dr. Hubal replied that these methods are currently
being developed at NERL and exposure assessors need some way of focusing and coordinating
their efforts as they develop these methods. Dr. Ginsberg asked whether there were any
surrogate methods that did not need to be developed by working with children. Dr. Hubal replied
2-11
-------
that there are some surrogate methods but it is currently not well understood how the results of
those methods can be used in children's assessment.
Dr. Fenske expressed continuing skepticism, apparently shared by several other panelists, about
the possibility of developing scenario-independent life stages for children.
2.5 Age Bins in Existing Data Sets That Are Relevant to Children's Exposure
Assessment
2.5.1 Presentation
Next, Dr. Kimberly Thompson presented an issue paper that she developed for the workshop.
The paper summarized the use of age categories in the existing collections of children's exposure
data. Dr. Thompson expressed the hope that the workshop would produce guidance both about
how to make the best use of existing data sets (for the short term) and about how to develop new
data sets (for the long run).
Children's development, whether at the most obvious level of physical growth or in terms of
•social, behavioral, and psychological changes, affects the kinds of chemical exposures they are
likely to experience. The issue of development is further complicated by a significant degree of
variability among children's developmental pathways and exposure risks. The pediatrician's
growth chart is a familiar illustration of childhood development and variation. Pediatricians also
use similar standardized charts of behavioral milestones to assess behavioral development. The
milestones described on these charts, while significant from a medical perspective, are often not
closely suited to the characterization of childhood exposure. As an example, Dr. Thompson said
that although language and social skills are highly emphasized in pediatric medical charts, these
factors are only of secondary significance to exposure assessment. Conversely, the period of
teething is extremely important to exposure assessors but is not prominently defined in pediatric
charts. Dr. Thompson expressed the hope that the workshop might provide an opportunity for
2-12
-------
pediatricians and risk assessors to learn more about each other's work and might help them to
work together, in the future, in a more coordinated fashion.
There are both qualitative and quantitative differences between how chemical exposures affect
children and adults:
• Qualitative differences between adults and children exist when the effect of a chemical
.dose on a child is completely different from the effect the same dose would have on an
adult. Children's organ systems are undergoing development, creating windows of
vulnerability to particular chemicals.
• Quantitative differences between adults and children exist when the effect of a chemical
dose on a child is similar to the effect that the same dose would have on an adult, but is
present to a greater or lesser extent.
In the absence of careful studies of the effects of a chemical on children, it is very difficult to
extrapolate from adult risk profiles to child risk profiles. Comparing the effects of a particular
chemical in children and adults, the chemical may act completely differently or in the same way;
to lesser extent or to a greater extent. It is very important that exposure assessors know where
the critical windows of vulnerability in childhood development are, so as to develop exposure
data that specifically address those time periods.
The existing exposure data do use a fairly wide range of different age categories, but these
choices have not been irrational. Age categories are typically defined to reflect the particular
factor being studied (such as change in diet, for example). The use of existing data often
becomes problematic when one wishes to make extrapolations from them—for example, when
one is attempting to extrapolate from short-term exposure studies to long-term exposure studies
and vice versa. It is also unclear how one should deal with the spatial and temporal limitations of
existing studies. For instance, can a study of a small group of children be taken as representative
of all the children in the United States? Can a national study provide useful information about
exposure in a specific socioeconomic/cultural group? Exposure assessors need to be very careful
to avoid representing children in unrealistic ways. It is an open question whether it would be
2-13
-------
most efficient to answer these questions by developing new data or by learning how to better use
existing data. .
Dr. Thompson's report (Changes in Children's Exposure As a Function of Age in the Relevance
of Age Definitions for Exposure and Risk Assessment) is included in Appendix H. It describes
the data that feed into the exposure equations that Dr. Hubal described in her previous
presentation. Dr. Thompson reiterated that there are three major routes of exposure: inhalation,
dermal, and oral. The concept of the microenvironment, mentioned by Dr. Hubal earlier, was
originally developed with reference to the measurement of chemical concentrations in the air.
The importation of the concept of the microenvironnient to the assessment of dermal and oral
exposure has generated a great deal of uncertainty, since it is unclear how to characterize
microenvironments hi these new terms.
One of the major objectives of Dr. Thompson's paper is to describe the availability of the data
necessary to use the standard set of exposure assessment equations (described in Dr. Hubal's
paper) for children's exposure assessment. Another major objective is to describe the age
categories that are used hi the existing data relevant to assessing children's exposure. Other
objectives are to characterize the extent to which the data are accessible and the extent to which
the existing data age categories can be modified. A primary source of Dr. Thompson's research
was the draft copy of the Child-Specific Exposure Factors Handbook (mentioned earlier by Dr.
Firestone and provided to the participants at the workshop).
Exposure assessors are required by the Food Quality Protection Act to gauge aggregate exposure
to particular chemicals—that is to say, total exposure from all sources, pathways, and routes.
This is extremely difficult to accomplish with the present data sets because the age categories do
not coincide in the different studies describing different exposure pathways.
Dr. Thompson evaluated the existing exposure factor data for children (e.g., body weight, food
intake, soil ingestion), closely following pages 15 through 46 of her report. Summarizing her
2-14
-------
survey of the studies that describe the different exposure factors, Dr. Thompson concluded that
assessors lack the kind of information they need to extrapolate exposure data from the particular
populations that have been studied to broader or different populations. A great deal more is
known about easily measurable anatomical factors, such as body weight, than is known about
behavioral factors such as the mouthing of objects.
Direct, indirect, and biomarker studies can provide three independent techniques for filling in the
gaps in exposure assessors? knowledge. All three techniques should be used and can validate
each other. Dr. Thompson said that each of these techniques involves the taking of
measurements and the use of models. In the indirect assessment approach, one must make
measurements in order to build one's model. In the direct approach, one needs models in order
to interpret one's data. It will not do to call for the use of either measurements or models alone
because the two kinds of tools are closely interconnected.
Available exposure data describing children's surface areas, their fish consumption rates, and the
duration of their mouthing behaviors are currently not used in any of the exposure assessment
equations described by Dr. Hubal. Also, these equations call for pieces of data that are currently
not available in the scientific literature. These missing data primarily have to do with the precise
description of how children might contaminate their food by touching it with their contaminated
hands or dropping it on contaminated surfaces before consuming it.
Breast milk consumption is the only exposure factor for which data have been reported by
individual days of age. Water consumption, food consumption, and inhalation rates are the only
factors for which data are available in age ranges of months. On the yearly scale of age ranges,
there is a great deal of variability among the different age ranges that investigators have chosen
for their studies.
Dr. Thompson mentioned that many aggregate exposure models are being developed to meet the
demands of the Food Quality Protection Act and she gave an example from the Lifeline™ model
2-15
-------
to demonstrate the issues of how the model developers evaluated the natural breakpoints in the
data to develop appropriate age bins.
Dr. Thompson went on to outline some goals for improvement in exposure assessment data.
There are gaps in the data on breast feeding, the handling of food and how it relates to non-
dietary exposure, fish intake rates, soil intake, and soil adherence rates. Assessors also need
more information about the connections between different macroactivities, microactivities, and
microenvironments. They also need to learn more about the activities of school-age children in
the summertime and after school.
2.5.2 Questions and Comments
Dr. John Kissel made a comment about the use of the term "data" in reference to the information
hi the Exposure Factors Handbook (he had not looked yet at the Child-Specific Exposure Factors
HandbooK). He found that many of the numbers in this handbook are estimates based on prior
data rather than actual measurements. He also thought there was more need for critical
evaluation of the different degrees of certainty about the different numbers in the handbook. For
example, he thought that dermal soil adherence could be measured much more accurately than
soil ingestion numbers because of the different methodologies used in those two measurements.
He also thought that the consequences (in terms of assessing actual dosage) of underestimating
dermal soil adherence factors were far less grave than the consequences of underestimating soil
ingestion factors. For both reasons, he considers dermal soil adherence factors to be better
measures than soil ingestion factors. He was perplexed, therefore, that Dr. Thompson ranked a
set of soil adherence factors generated by him as being of low quality, while a set of soil
ingestion factors were ranked as being of medium quality. He is concerned that exposure factors
are being evaluated simply on the basis of the quantity of numbers that exist to describe them
rather than on the quality of those numbers. This misevaluation, he continued, might lead to the
misappropriation of resources intended to improve data quality. Dr. Thompson remarked that the
assessments of quality were not hers: she took them directly from the Child-Specific Exposure
2-16
-------
Factors Handbook. She made her own evaluations under the heading of "Extent of
generalization" and she ranked both the dermal soil adherence factors and the soil ingestion
factors as "low" under that heading. Dr. Thompson invited constructive criticism from the
participants about how EPA might better evaluate data quality. Dr. Kissel remarked that EPA
should make it a major research priority to learn more about soil ingestion in children with pica
because it represents a far greater and less well understood exposure pathway than dermal
absorption through soil.
Another panelist suggested developing growth charts for different organ systems. Dr. Weil asked
if Dr. Thompson had used the children's growth charts being developed by the Centers for
Disease Control and Prevention (CDC). Dr. Weil commented that CDC left out the NHANES III
data for children over age 6 because they have become too fat.
Dr. Walker, a workshop observer from EPA's National Center for Environmental Assessment,
made comments and presented some empirical evidence showing that children have multiple
critical growth periods, as they develop from birth to maturity (his overheads are provided in
Appendix E). He defined a critical growth period as an age when the child reaches a peak
growth velocity hi weight or height (growth spurt).
Dr. Walker said that he was hired by EPA to develop age-specific radiation dosimetry models,
but early on discovered this could not be done without devising a more adequate description of
organ masses as a function of age. He went on to say that many years were spent developing
these models, using cross-sectional data that were obtained from the literature, but it was not
until recently, after acquiring large sets of longitudinal height data for analysis, that he was able
to develop a model that was representative of bone growth in children.
Dr. Walker showed overhead slides illustrating how well his newly published WWHLA growth
model (recently published in Growth, Development, and Aging 2000, 64, 33-49) fitted
longitudinal height data of children from two major growth studies. He said that, although it was
2-17
-------
earlier believed that children experienced only two spurts (infantile and pubertal), his height
displacement and velocity curves showed six, from birth to maturity. These spurts occurred at
different times in each child. Dr. Walker named them according to the age when they reached
their peak height velocity (PHV); neonatal, infantile, early-childhood, middle-childhood, late-
childhood, arid pubertal. His analysis revealed that the mean ages at PHV for these spurts were
different in males and females (see his overheads in Appendix E). He indicated that, although
the ages at PHV for the different spurts varied between the children and depended on gender and
whether a child was a slow, average, or fast grower, they represented developmental milestones'
for height growth in children and should be considered in risk assessment. He felt that his model
could be a powerful tool for better characterizing height growth in children and identifying those
critical periods of rapid growth that may make children highly susceptible to xenobiotics from
the environment. Dr. Walker also presented graphs showing that the kidneys and liver undergo
multiple postnatal growth spurts; the periods of rapid growth in height correspond very closely to
periods where peak concentration levels in the bones were found for such bone seekers as
radium-226 and strontium-90.
2.6 Charge to the Experts
Kimberly Thompson reviewed the charge for the workshop. She urged panelists to think about
the questions in the charge from a "value of information" approach. That is, they should consider
how limited resources may be most efficiently allocated to meet research needs for the future.
Dr. Thompson suggested that panelists think about the different ways in which the problem of
children's exposure could be subdivided. Some reasonable ways to subdivide exposure are:
" By the class of chemicals involved in the exposure.
» By the organ affected in the exposure.
• By behavioral characteristics that lead to the exposure.
2-18
-------
• By age categories .(using the following categories as a starting point: newborn, infant,
toddler, young child, and adolescent).
Dr. Thompson advised the participants to think about the ways in which exposures, grouped in
these different ways, do or do not fall into similar groups.
Dr. Thompson reviewed two parallel sets of questions (see Appendix C) for the two discussion
groups (one concerned with anatomical development and the other concerned with behavioral
development). She asked the participants to identify the kinds of problems that might be
associated with the use of age bins. She also asked the participants to focus, as much as possible,
on anatomical and behavioral developments that affect a child's exposure to chemicals. The,
participants then broke up into two discussion groups (persons named in bold italics were
discussion leaders): -
Anatomy -Group
Thomas Armstrong
ExxonMobil Biomedical Sciences, Inc.
Sophie Balk
Montefiore Medical Center
Jim Bruckner
College of Pharmacy, University of Georgia
Michael Dinovi
U.S. Food and Drug Administration
Gary Ginsberg
Connecticut Department of Public Health
Robert Johnson
Agency for Toxic Substances and Disease Registry
John Kissel
Department of Environmental Health, University of
Washington
Melanie Marty
California Environmental Protection Agency
Behavibr Group
Deborah Bennett
Lawrence Berkeley National Laboratory
Richard Fenske
Department of Environmental Health, University of
Washington
Lynn Goldman
Johns Hopkins School of Hygiene and Public Health
Celestine Kiss
U.S. Consumer Product Safety Commission
Bruce Lanphear
Department of Pediatrics, University of Cincinnati and
Children's Hospital Medical Center
James Leckie
Department of Civil and Environmental Engineering,
Stanford University
Mary Kay O'Rourke
College of Public Health, University of Arizona
P. Barry Ryan
Rollins School of Public Health, Emory University
2-19
-------
Anatomy Group
George Rodgers
Division of Pediatric Critical Care, University of
Louisville
Margo Schwab
Johns Hopkins School of Hygiene and Public Health
William Weil
Department of Pediatrics/Human Development,
Michigan State University
Behavior Group
Katherine Shea
Consultant
Robin Whyatt
Joseph L. Mailman School of Public Health, Columbia
University
'i
The two groups met separately during the afternoon of the first day of the workshop and the
morning of the second day. They convened briefly in a plenary session at the end of the first day
to report their progress, and again during the afternoon of the second day to report their
conclusions and recommendations to all participants. The discussions and conclusions of the
behavior and anatomy groups are summarized in Sections 3 and 4, respectively. Section 5
summarizes the brief plenary discussion that took place at the end of the workshop.
2-20
-------
3. CHANGE IN CHILDREN'S EXPOSURE DUE TO
BEHAVIORAL DEVELOPMENT
The behavior group consisted of 10 specialists in fields relating to children's exposure, including
pediatrics, risk assessment, and exposure assessment. A list of participants is provided in Section
2.6. The group was charged to consider the changes in behavioral patterns during childhood that
can impact children's exposure to environmental contaminants (see Appendix C for the complete
charge). The group was asked to consider questions such as:
• Does it makes sense to think about childhood behavioral development as a series of
discrete events which lend themselves to characterization using age group categories or
"bins?"
• What are the most important developmental milestones in children's behavior?
• For those behaviors that are likely to have an importance impact on exposure, is there
existing exposure information that is representative of the behavior?
• For those behaviors that are represented in existing exposure information, compare the
age groups identified for the developmental milestone in question 2 with the age groups
in the existing exposure information.
• For those behaviors where the age groups reported hi the exposure information are not
aligned with the age groups defined by the developmental milestone, what is the best
approach to representing the appropriate age groups in an exposure assessment?
The group began by carefully evaluating the use of age bins in describing children's behavior.
They then discussed specific issues related to characterizing behavioral changes relevant to
exposure, organizing their discussions by the three major exposure pathways (oral, dermal, and
inhalation). They concluded by integrating their discussions of the different exposure pathways
into a preliminary breakdown by age representing changes in constellations of behavior that
could signal new categories of exposure and by compiling a list of research recommendations.
The group's chair presented the group's conclusions at the workshop's final plenary session.
3-1
-------
This section summarizes the discussions and conclusions of the behavior group. It is divided
into five sections dealing with:
• The defensibility of age bins (Section 3.1).
• Specific behavioral factors related to oral, dermal, and inhalation exposure pathways
(Sections 3.2 through 3.4).
• Conclusions developed by synthesizing the discussions about the three exposure
pathways (Section 3.5).
• Recommendations for future research (Section 3.6).
3.1 The Defensibility of Representing Behavioral Difference in Terms of Age Bins
3.1.1 Principal Discussion Points
The group considered whether or not it made sense to think about children's behavioral
development as a series of discrete events that can be organized into age bins. They arrived at
the unanimous conclusion that it did not make sense to think about childhood behavioral
development in that fashion. Although behavioral development follows a recognizable
progression, it takes place differently between individuals and unevenly within individuals.
More specifically:
Children begin developing behaviors at widely different times, and that development
often occurs hi brief spurts that are hard to predict precisely.
Different domains of behavioral development (e.g., language skills, motor skills, social
skills) may not develop in a synchronized fashion, complicating attempts to peg an age
range of children to a particular overall stage of behavioral development.
It is extremely difficult, sometimes even impossible, to identify where in a child's
development of a particular behavior pattern ends. Many behavior patterns typically
associated with small children (such as mouthing non-food items) simply become less
prominent with time and persist into adulthood. It appears that, in many cases, new
behaviors do not completely replace older behaviors.
3-2
-------
• Modifying factors having to do with culture and socioeconomic status may substantially
limit one's capacity to generalize about the specific behaviors children exhibit within
particular age ranges.
Although it noted these major flaws with the practice of age binning, the group recognized the
practical utility of age bins to EPA in its day-to-day decision-making. The use of age bins, while
problematic, is a great improvement over not taking behavioral development into account at all.
Some pediatricians within the group pointed out that it is common practice for them to use age
bins as a starting point for evaluating their patients—an age bin provides a set of starting
expectations for a particular child, which can help them evaluate that child's particular
characteristics. There was very strong concern within the group, however, that if EPA developed
age bins in an exposure assessment context, these bins might take on a "life of their own" and
come to be accorded an unjustified degree of authority and precision. One panelist pointed out
that when pediatricians use age bins, they use them at a screening assessment level. Such use
depends upon a familiarity with the underlying continuous processes of children's development.
Age bins, if used uncritically by individuals unfamiliar with the behavioral development that
those bins crudely represent, could lead to significant errors of exposure assessment. EPA
exposure assessors, therefore, need to become familiar with the continuous distributions that
underlie any age bins they use.
The group suggested that it would make sense to focus on particular behavioral milestones that
indicate the beginning of the capacity to engage in particular domains of behaviors (e.g.,
crawling, mouthing) that potentially lead to particular kinds of exposure. EPA should always
bear in mind that the underlying expression of these domains of behavior are distributions rather
than neat categories. Some panelists considered it entirely plausible, in the absence of specific
evidence to the contrary, that the frequency distributions of different classes of behavior might be
spread out broadly, rather than clustered, across the range of children's ages. The usefulness of
the concept of general behavioral age bins depends on the degree to which the relevant behaviors
are spread out or clustered together.
3-3
-------
In their discussion of the merits of age bins, members of the group agreed that age bins could be
useful as a guide to the development of particular exposure assessment scenarios. They could
serve as reminders to the developers of exposure assessment factors that these factors need to be
developed for different broad age ranges among children.
The group members urged EPA to remember that the details they provided at the Workshop
about specific age categories were based solely upon their general knowledge. These details
require much more careful research and expert guidance before EPA can rely on them as
scientifically validated facts.
3.1.2 Other Discussion Points
Concern about EPA's direction in basing children's exposure assessment on indirect exposure
factors was expressed by the members of the group. A substantial number of the panelists
thought that EPA was overreaching the inherent limitations of indirect exposure assessment and
thought that it ought to be focusing more of its resources on biomonitoring efforts. There was
some suspicion that EPA wanted to use age categories to develop a broadly generalizable model
of child development rather than as a guide to better understanding the underlying elements of
childhood development. One panelist broadly referred to the equations hi the Hubal et al. paper
as "modeling efforts" and indicated that she was worried when she heard about them, given the
poor and limited data that are being used as inputs to them. She expressed strong skepticism
about the results of these equations, contrasting them to "empirical data." The chair reassured
the panel that EPA was interested in the underlying elements of childhood development, not
modeling efforts. Another panelist indicated that he disagreed with EPA's basic approach to
exposure assessment but considered "mechanistic modeling" to be adequate for addressing
urgent, present day issues. He said that his cooperation with EPA was based on the assumption
that age-bin/modeling efforts would be phased out in the long run. Some members of the group,
however, felt that the use of mechanistic modeling was justifiable (or even essential) in the
context of studying new chemicals for which exposure pathways are not yet well understood.
3-4
-------
Some members of the group were frustrated that EPA did not place more emphasis on
epidemiological approaches to exposure assessment (as opposed to focusing primarily on what
were seen as "mechanistic modeling" efforts). They felt that indirect exposure assessments
might be most valuable if they were conducted simultaneously with direct exposure assessments
and epidemiological studies so that the different kinds of data could be cross-referenced to each
other.
Despite all these caveats, the members of the group were generally willing to suggest which
domains of behavioral development were relevant to exposure assessment and to make rough
approximations of when they tend to first appear. They emphasized that their conclusions should
be taken as a starting point for further research and not as a definitive evaluation.
One panelist indicated that EPA could add a biological dimension to its Exposure Factor
Handbook by including information about the distribution of behaviors and by including
information about biomonitoring. Another panelist agreed. The use of distribution curves is an
important tool that EPA should use if it wishes to approach exposure factors in a biological
fashion.
A panelist suggested creating a field of study similar to industrial hygiene but focusing on
children's play rather than adult work.
Finally, the group emphasized that it was extremely important for EPA to look closely at fetal
exposure, although it was not directly relevant to the present discussion of children's behaviors.
Transplacental exposure should really be considered a fourth exposure pathway in addition to the
oral, dermal, and inhalation pathways.
3-5
-------
3.2 Behavioral Development and Oral Exposure
3.2.1 Principal Discussion Points
The group recognized four major domains of behavioral development that affect oral exposure to
chemicals:
• Gross motor development. For example, the onset of mouthing, sitting, crawling,
walking, rolling, climbing, bike riding.
i
• Fine motor development. For example, the onset of behaviors involving grasping objects,
placing hands in mouth, placing objects hi mouth, using utensils, and opening jars.
• Cognitive development. For example, the onset of understanding object permanency,
understanding the meaning of the word "no," and understanding the concepts of death,
danger, and poison.
• Social development. For example, the onset of willingness to follow directions, interest
hi risk taking, and ability to drive an automobile.
Changes hi these domains of development could plausibly affect oral exposure to chemicals
through the following six pathways:
In connection with this discussion of exposure pathways, the group expressed particular hope
that EPA would locate existing data on when parents start feeding their children with bottles and
with solid food as well as when they stop breast- and bottle-feeding their children. These
statistics are likely to be highly variable between different ethnic and social subgroups and may
merit closer investigation. EPA should attempt to find out at what ages children receive
"exclusive" breastfeeding (i.e., breast milk is the only food source), "full" breastfeeding (i.e.,
breast milk is the only milk source), and "any" breastfeeding (i.e., some breast milk is
consumed).
3-6
-------
3.2.2 Other Discussion Points
One panelist expressed some confusion about which behavioral characteristics were relevant to
oral exposure. Many of them, such as being able to sit or walk, are relevant in a "one-step-
removed" sense—they affect whether or not a child could get into a situation in which he or she
might-have an oral exposure.
Another panelist mentioned it was important to distinguish whether children were ingesting non-
food items or soil in particular.
Exposure Pathway
Breast Milk/Nursing
Bottle Feeding
Food
Water
Mouth-Hand Contact
Mouth-Object Contact
Examples of Relevant Age-Related Developments
(Ages, particularly when stated as single numbers, are spontaneous approximations
from the workshop and require further validation)
Nursing takes place roughly from 0 to 18 months of age, though this varies by culture.
Bottle feeding takes place roughly from 0 to 12 or 24 months.
Head control (2 months), sitting (6 months), finger feeding (8 to 9 months), use of
utensils (10 to 12 months), and the final shift to adult patterns of eating. Solid food,
served in a bottle as a.slurry, is often consumed as early as 1 month of age, but 4 to 6
months is the typical age range for beginning solid foods by themselves.
Use of cups (6 to 9 months).
Prevalence of hand-to-mouth behaviors, such as thumb-sucking. Gross motor skills
determine access to areas where the hand can become contaminated. Succession of
gross motor milestones: rolling (4 months), creeping (6 months), crawling (8 months),
walking (12 months), and climbing (1 8 months).
The ability to interact with objects is a major factor here. The ability to grasp an object
to one's mouth begins roughly at 3 to 5 months. A pincer grasp and moderate strength
are achieved by 9 months. Children become aware that objects exist even when
covered around 6 months but generally do not understand the meaning of the word
"no" until 12 months.
There was some discussion of whether or not it was important to distinguish between mouthing
and chewing of objects. There was some expression of non-specific concern about determining
how to differentiate between different kinds of oral behaviors. There was disagreement about
3-7
-------
whether or not socioeconomic status and education were effect modifiers. Panelists considered
them to be confounders.
Several panelists were interested in including transplacental food consumption as an oral
exposure, but the group came to the eventual conclusion that EPA should consider transplacental
exposure as a sort of fourth exposure pathway rather than trying to fit it into oral exposure.
Some members of the group felt that non-food mouthing behavior was comparable to pica
behavior. The chair pointed out that these were essentially two different behaviors.
One panelist suggested that the swallowing of coughed-up dusts might be considered a
behavioral factor contributing to non-dietary oral exposure, but there was some disagreement
within the group about whether this was an appropriate classification. The degree of exposure
varies depending on whether one spits or swallows coughed-up mucus. Some panelists noted
that playing hi water or dusty yards might contribute to non-dietary oral exposure.
Several panelists noted that then- discussion of oral exposure behaviors was primarily focusing on
very young children. The chair thought this was a result of the fact that behaviors never really
die out—it is simpler to point out the beginning of behaviors in early childhood than to estimate
how those behaviors scale back later hi childhood. An implication of this observation is that
exposure assessment for older children should also incorporate nonfood mouthing behaviors.
3.3 Behavioral Development and Dermal Exposure
3.3.1 Principal Discussion Points
The group recognized substantial similarity between behavioral development's effects on oral
exposure and its effects on dermal exposure. Thus, the previously discussed domains of gross
motor development, fine motor development, cognitive development, and social development
3-8
-------
were also found to be relevant to dermal exposure. However, the group noted that these
behavioral domains may have unique relationships to dermal exposure. Dermal exposure can
result from behaviors that cause any part of the body (not just the mouth) to come in contact with
contaminated surfaces.
Changes in these domains of development could plausibly affect dermal exposure to chemicals
through the following six pathways:
• Showering
• Bathing
• Recreational water use (e.g., swimming)
• Surface contacts
— Floor
— Object
— Ground (outdoor)
• Skin-to-skin transfer
— With hands
— With other skin
• Intentional applications to skin
The group also noted that changes in a child's clothing/diaper use throughout development would
affect the amount of his or her skin available for dermal exposure, as well as the permeability of
the skin.
3-9
-------
3.3.2 Other Discussion Points
One participant mentioned that there is significant concern about chloroform absorption into the
skin from water. Another member of the group wondered about the effects of different soaps on
the permeability of the skin.
Panelists were unsure of whether or not they should distinguish between pools, lakes, rivers, and
drainage ditches in terms of their dermal exposure profiles.
One panelist objected to distinguishing between soil and floor exposure in the case of dermal
contact because this distinction had not been made in the case of oral exposure. The panelist
wanted oral and dermal exposures to be classified in a parallel fashion.
A group member commented that children undergo a very abrupt transition from taking no
showers at all to taking regular showers.
3.4 Behavioral Development and Inhalation Exposure
3.4.1 Principal Discussion Points
The group attempted to develop a list of behavioral milestones that affect inhalation exposure,
but ended by concluding that behavioral effects on inhalation exposure do not change in discrete
jumps marked by milestones; rather, they change in a continuous fashion because exposure by
inhalation is driven substantially by activity level and exposure scenarios. The relevant domains
of behavioral change were:
Gross motor development. Gross motor development is important for characterizing the
atmosphere in which children breathe (i.e., their "personal clouds"). At different life
stages, children spend their time breathing in different microenvironments depending on
their modes of locomotion (e.g., rolling, crawling, and walking). Also, certain gross
3-10
-------
motor skills, such as the ability to run and play sports, tend to increase overall activity
levels as they develop.
• Activity level. The group concluded that activity level could not be considered simply as
a series of states through which a child develops, and agreed that children's broad level of
development certainly affects their average activity level. For example, infants are
generally relatively passive, but can exert themselves very intensely when crying.
Activity level affects inhalation rate and is thus an important factor in assessing inhalation
exposure.
• Breathing behavior. The group considered the transition from mouth breathing to nasal
breathing. They concluded that it is a significant factor in inhalation exposure, but is
complex and does not lend itself to discrete cut-offs.
3.4.2 Other Discussion Points
Some members of the group suggested that one should differentiate between activities that tend
to stir up dust and those that do not. This distinction determines whether one is exposed only to
atmospheric gases or also to aerosol particles.
The group gave some initial consideration to using sleeping, awake but quiet, and active as
behavioral stages, but the chair suggested that this was inappropriate and that activity level varies
continuously in children. There was prolonged confusion about how to incorporate these terms
(sleeping, quiet, and active) into a classification scheme parallel to that which was used for the
other exposure pathways.
3.5 General Conclusions for Incorporating Behavioral Development into Children's
Risk Assessment
3.5.1 Principal Discussion Points and Presentations
Dr. Katherine Shea, the group's chair, proposed a set of provisional behavioral age bins as a
starting point for further research. The bins are as follows (see also the corresponding overheads
in Appendix F):
3-11
-------
Age Bin
0 to 2 months
3 to 5 months
6 to 1 1 months
12 to 23 months
2 to 5 years
6 to 10 years
11 to 15 years
16 to 20 years
Characteristics Relevant to Oral and Dermal
Exposure Pathways
Breast and bottle feeding. Hand-to-mouth activities.
Rapid growth makes children particularly vulnerable
to chemicals.
Solid food is introduced. Contact with surfaces
increases. Object-to-mouth activities increase.
Food consumption expands. Children's floor mobility
increases. Children are increasingly likely to mouth
non-food items.
Children consume a full range of foods. They
participate in increased play activities, are extremely
curious, and exercise poor judgment. Breast and
bottle feeding cease.
Children begin wearing adult-style clothing. Hand-to-
mouth activities begin to approximate adult patterns.
There is decreased oral contact with hands and non-
food items, as well as decreased dermal contact with
surfaces.
Smoking may begin. There is an increased rate of
food consumption.
High rate of food consumption continues.
Characteristics Relevant to
Inhalation Exposure Pathways
Children spend a great deal of their
time asleep.
Children may breathe close to floor
level when placed in play pens or
infant seats on the floor.
Development of personal dust
clouds.
Children walk upright, run, and
climb. They occupy a wider
variety of breathing zones and
engage in more vigorous activities.
Occupancy of outdoor spaces
increases.
Children spend time in school
environments and begin playing
sports.
Increased independence. Work
outside of home begins.
Independent driving begins.
Expanded work opportunities.
The group briefly discussed Question 3 of their charge (Appendix A), which asked whether or
not information was presently available on the behaviors that affect exposure. Panelists could
only answer this question in general terms, because they did not have the resources to provide
specific examples of behavioral data. There is a definite need for more data on behavioral
exposure factors, as many of the existing data are of limited quality or coverage. Some members
of the group suggested that exposure factors be collected in the context of specific chemical
compounds, and there was general agreement that more work was needed on behavioral
development in specific sub-populations of children (e.g., developmentally impaired, physically
disabled). It is also very important, many felt, to better integrate indirect measurements of
exposure factors with physiological and mass-balance measurements so as to better validate
those exposure factors.
3-12
-------
Question 4 of their group's charge asks whether existing exposure factor data have been collected
in a manner that fits with the proper age bins for children. The group did not have time to review
and evaluate the age distributions chosen by the different studies Dr. Thompson listed in her
presentation. They did note a general lack of rich data sets describing the exposure of very young
children. They suggested, furthermore, that EPA re-evaluate the raw data of past exposure
studies when it appears that those studies did not organize their published results according to the
best set of age categories. Such plans, of course, would have to take into account some important
concerns about the cost of re-analyzing data and-the continued protection of the privacy of the
individuals from whom the raw data were collected.
Question 5 of the charge asks for guidance in assessing exposure to a particular age bin when
existing data do not provide uniform coverage of that group. The group had no simple answers
to this question and could only encourage EPA to develop more data, focusing on being as child-
protective as possible and emphasizing direct, empirical data collection. With respect to the
problem of aggregate exposure, EPA should note that oral and dermal exposure factors are
dependent on similar behavioral parameters and that inhalation exposure follows a fairly
independent set of parameters.
3.5.2 Other Discussion Points
The group generally agreed with the chair's proposal for different age categories, but one panelist
suggested that they should not try to arrive at any further specificity, given the lack of data and
specialized expertise to ground their discussion. Another panelist stated that EPA tends to .
overestimate the universality of its activity data—it tends to generalize from one set of data to
many different contexts, which may not be justified. A third panelist responded that age
categories would be useful not only for collecting activity data but also for the more direct,
epidemiological studies that other panelists were calling for. There is some value to mechanistic
modeling, argued another panelist, when it is unclear how exposure to a new chemical is
partitioned across different exposure pathways. The emerging consensus in the group was that it
3-13
-------
was reasonable for EPA to conduct indirect exposure assessment. The problem that panelists
perceived, however, was that EPA does not seem to give sufficient priority either to the direct
exposure assessment approach or to the validation of indirect exposure models.
The group noted that these categories do not align perfectly with the categories of cognitive
development articulated by Piaget. It is unclear, however, whether cognitive development (e.g.,
the ability to understand the concept of death) is as relevant to exposure assessment as it is to risk
management. For the purposes of scenario development, at least, the child's cognitive
development is a crucial factor for risk assessors to understand. Some members of the group also
briefly noted that the categories do not align perfectly with the Erickson's categories of
emotional development.
5.5.3 Observer Comments
Dr. Harvey Richmond of EPA's Office of Air Quality Planning and Standards delivered some
comments pertaining to the group's discussions. He argued that each behavioral age bin should
include information about the amount of time children in that category spend outdoors. From his
perspective, one of the most important factors in inhalation exposure is the amount of time that
one spends outdoors. Dr. Richmond also noted the group's enthusiasm for conducting
epidemiological studies that include biomarkers. He cautioned that, while such an approach is
valuable, it is only technologically feasible for a limited number of chemicals. Among air
pollutants, he ventured that it would only work for 10 percent of the chemicals EPA is concerned
about (e.g., certain metals and volatile organic compounds). He also noted that biomarker data
for a chemical is of no use to exposure modeling unless the paths of exposure for that chemical
are already very well understood.
3-14
-------
3.6 Research Recommendations
The behavior group highlighted three major data gaps that it believes EPA should immediately
address to improve its use of age bins. The group suggested the following measures:
• Carefully evaluating the distribution of behaviors across different age ranges, taking into
consideration the variability introduced by effect modifiers and specific sub-populations,
such as particular ethnic and socioeconomic groups, developmentally impaired children,
and disabled children.
• Using integrated data sets to evaluate the relative importance of different kinds of
exposure. It is important to focus time and resources on exposure pathways that are more
likely associated with significant risks.
• Collecting integrated data sets that combine information about children's behaviors, their
estimated exposure, their biomarkers for particular chemicals, and their health.
Other short-term data gaps mentioned by different members of the group include the lack of
information about:
The frequency, distribution, and duration of children's dermal and oral contact events.
Prenatal exposure. EPA should, perhaps, hold another conference on issues of prenatal
exposure and its modifiers. The accumulation of environmental contaminants in amniotic
fluid is a particular area of concern.
The relationship between adolescent developmental milestones and adolescent exposure.
The distributional properties around developmental milestones. ,
The exposure of children to household dusts and residues. Both oral and dermal
exposures need to be better characterized. Such studies could be conducted either by
direct exposure measurement or by measurement of the number of contact events children
have with household dusts and residues.
Hand-to-mouth and object-to-mouth transfer of contaminants. These exposure factors are
currently very poorly understood.
3-15
-------
" The particular biomarkers which are most useful for measuring children's exposures to
prevalent chemicals. Additional pharmacokinetic data on these chemicals may aid in the
proper interpretation of biomarker data.
" The extent to which EPA's definitions of soil ingestion (as a measure of exposure) are
validated by biomarkers.
*-!•'
• The relative significance of different exposure pathways for children of different ages.
» Probability-based biomarker surveys of prevalent exposures. A good model of this sort of
study is the one recently conducted on lead exposures by the National Center for Health
Statistics (with the support of the U.S. Department of Housing and Urban Development).
« The amount of time children spend near the floor or soil and how this affects their
personal dust clouds.
Individual members of the behavior group also mentioned several data gaps that EPA should
address in the long term. These include the lack of information about:
The relationship between cross-sectional studies and longitudinal studies. It is unclear to
what extent cross-sectional studies can stand in for longitudinal ones. It is common
research practice at present to simultaneously study the exposure of different children at
different ages, but this research strategy may leave out important information about how
the exposure of particular cohorts of children changes over time.
How exposure varies at different times of the year and among different geographical
locations. For example, children may have different dermal transfer coefficients in the
spring and summer, depending on how sweaty their palms are.
How long-term trends in child-rearing practices affect children's exposure.
Whether or not the recommended interventions for environmental hazards are indeed
effective. Randomized, controlled studies are needed to address this question.
3-16
-------
4. CHANGE IN CHILDREN'S EXPOSURE DUE TO
ANATOMICAL DEVELOPMENT
The anatomy group consisted of 11 specialists in fields relating to children's exposure, including
pediatrics, risk assessment, and exposure assessment. A list of participants is provided in Section
2.6. The group was charged to consider the changes in anatomy during childhood that can
impact children's exposure to environmental contaminants and their susceptibility to health
effects from that exposure. The group was asked to consider questions such as:
• Can anatomical development during childhood be considered as a series of discrete bins,
particularly when existing information is not adequate to construct a continuous exposure
function?
• What are the most important developmental milestones for anatomical changes related to
physical growth in children?
• For those anatomical characteristics that are likely to have an important impact on
exposure, is there existing exposure information that is representative of the
characteristics?
The group began by discussing the charge. Some panelists questioned whether "binning" was an
appropriate approach. There was initial concern that such an approach could result in a
unpractically large number of bins, and that the type of information provided by such an
approach may have an insignificant impact compared to other uncertainties in risk assessment.
Other panelists felt that binning could have merit. The group eventually agreed to pursue the
possibility of binning since this was an integral part of their charge. They began by discussing
the development of individual anatomical and physiological characteristics (such as body weight
and skin surface area), organs, and systems (including body fat, skin, skeleton, liver, immune
system, reproductive and endocrine systems, lung and respiratory system, gastrointestinal tract,
renal system, cardiac system, muscle, and sensory organs). They generally limited their
discussions to development after birth; fetal development and in utero exposure were not
covered. They then discussed general issues related to characterizing developmental changes
relevant to exposure, and concluded their discussions by compiling a list of research
4-1
-------
recommendations. The group's chair presented the group's conclusions at the workshop's final
plenary session.
This section summarizes the discussions and conclusions of the anatomy group. It is divided into
three sections:
» Individual anatomical characteristics, organs, and systems (Section 4.1).
» General issues (Section 4.2).
• Research recommendations (Section 4.3).
In each section, the group's conclusions, as summarized by the chair in the final plenary session,
are provided first, followed by a summary of the discussion that led to those conclusions.
4.1 Individual Anatomical Characteristics, Organs, and Systems
4.1.1 Weight
Conclusions
For weight, the folio whig preliminary bins were suggested based on changes in the rate of weight
gain: 0 to 6 months, 6 to 12 months, and 1 to 2 years. From 2 to 8 years the rate of weight gain
is relatively stable. It increases again at 8 years and at 10 years. After this, it peaks and then
decreases at 13 to 15 years depending on gender. Skin surface area as a function of weight is a
rapidly decreasing relationship until about 1 year, when it levels off. It remains stable until about
age 14, when it flattens out and becomes virtually horizontal.
4-2
-------
Discussion
The chair presented data1 on weight gain in children. The data showed rapid but different rates
of growth for three periods in early childhood: 0 to 6 months, 6 months to 1 year, and 1 to 2
years. By 2 years, weight gain continues at a relatively stable rate until about age 8, when it
increases again. Weight gain also increases around age 10 or 11, then decreases at age 13 to 15,
depending on gender. .
4.1.2 Body Fat
Conclusions i • .
The group considered body fat in terms of retention and mobilization of xenobiotic products.
Excess fat may act as a safety factor to the extent that it serves as a sink for lipophilic chemicals.
However, this potential "protection" becomes a liability during periods of weight reduction (e.g.,
in early adolescent boys and during dieting—which is common to adolescent girls—when
lipophilic substances stored in the fat become mobilized).
The proportion of fat in the body increases during the first 18 months and then remains stable
until about age 14. At age 14 in boys, it begins to decrease slightly and then increases again. At
14 in girls, it begins a more rapid increase, which continues up to adulthood., ,
'The chair presented data at many points during the discussions. These data were primarily drawn from five
sources:
• Pesticides in the Diets of Infants and Children, National Research Council, 1993.
• Research Needs on Age-Related Differences in Susceptibility to Chemical Toxicants, ILSI Risk Science
Institute, November, 1966.
• Bruckner, J.V., and Weil, W.B. 1999. Biological Factors Which May Influence an Older Child's or
Adolescent's Response to Toxic Chemicals. Regulatory Toxicology and Pharmacology, 29: 158-164.
• Analysis of the Michigan Department of Environmental Quality's Administered Environmental Standards
to Protect Children's Health, Michigan Environmental Science Board, February, 2000.
• Department of Labor Conference on Environmental and Occupational Risks of Adolescents.
4-3
-------
Discussion
The chair kicked off the discussion by showing body fat data for boys and girls by age. For boys
at age 20 years, there is more than a four-fold difference in body fat between the 10th percentile
(in which boys have 5 kilograms of body fat) and the 90th percentile (in which boys have more
than 20 kilograms of fat). For girls at that age, body fat ranges from 10 to 30 kilograms between
the 10th and 90th percentiles. The chair said there is a characteristic bump in fat mass for boys
just prior to adolescence, followed by a drop in body fat with the first stage of adolescence, then
an increase. This bump does not usually appear in girls.
The chair also showed data from the early National Health and Nutritional Evaluation Study
(NHANES); the Tecumseh, Michigan, study; and the Ten State Nutrition Survey of fat fold
thickness for triceps. These data show the same bump in fat mass for boys. In boys, extremity
fat is constant throughout childhood. For girls, it is fairly stable until early adolescence, when it
rises and then levels off at three times preadolescent levels. The chair was not sure whether this
would have any impact on exposure. However, he said, since fat is a major storage area for
lipophilic materials, it plays a significant role in what happens to chemicals once they enter the
body.
Another panelist said that work by the Connecticut Department of Health found that body fat at
birth is low relative to older ages, but by about 1 year of age it flattens out as a percentage of
body mass. This would affect various pharmacokinetic parameters, such as partitioning and how
much dose is retained in the fat compartment. The panelist said the content of adipose tissue
changes with maturation. It has more water content at birth than at older ages, so there is a
window at least in the first year hi which body composition is different enough to affect the
retained dose.
The chair responded that this composition difference is due in part to the fact that there are a
large number of fat cells at birth but their size is relatively small. Fat cells contain a certain
4-4
-------
amount of protein and water, which stay relatively constant. However, the protein and water
content decrease proportionally as the fat cell grows by adding more fat. This increase in fat
content probably occurs somewhere between the first and second years, while the number of fat
cells stays relatively constant between 2 years and adolescence. He also mentioned that the
increased body water of infants is primarily extracellular water rather than intracellular water.
Intraeellular water varies little and if anything may be a little low, because (in theory) there may
not be as many solute particles in infants' cells as in older children's. The loss of extracellular
water after birth is a constant proportion of mass in species from mice to giraffes.
A panelist pointed out that while fat could increase the retained dose of a chemical, it might also
act as a protective sink that lowers the chemical activity of toxic chemicals in the body. If so,
leaner people would be at greater risk, since a single exposure to a chemical could overwhelm
protective measures. Preadolescent males could also be at greater risk from chemical
mobilization that occurs with fat tissue loss, since this is a time when they typically lose fat
tissue. Also, data suggest that about 60 to 80 percent of girls diet, so they too could be
vulnerable to this type of risk, particularly during the first period of dieting. Another panelist
agreed that fat would not be protective when mobilized, such as during weight loss and breast
feeding. The greater the exposure and amount of chemical stored, the higher would be the risk
during periods of mobilization. Dr. James Walker, an EPA observer, said the some radioactive
chemicals such as xenon are stored in fat, so the amount of fat a child has could be an exposure
concern. A panelist pointed out that fat transport should also be considered. If there are
developmental changes in how fat is transported and deposited in the fat cells, this could affect
exposure.
4-5
-------
4.1.3 Skin
Conclusions
In the premature infant, the skin as a portal of entry appears to have differential permeability with
age, but members of the group did not know how long this differential permeability remained
after birth. The likelihood of skin being abraded can vary with age, since children are more
likely to abrade their skin as their mobility increases. Occluded skin may also be susceptible to
erosion. Skin under occlusive diapers in particular might have more permeability than
nonoccluded skin, especially when diaper rash is present. In later childhood, extensive eczema
or acne could alter skin permeability. For the ratio of skin surface area per unit of body weight,
there is a rapidly decreasing relationship until 1 year; then the ratio decreases steadily but more
slowly until about age 14, when it begins to flatten out to become virtually stable.
Discussion
The chair cited data that showed a rapid decline in skin surface area to weight ratio during the
first year of age, with a relatively slower but stable decrease in the ratio thereafter until about age
14. He also mentioned data showing increased skin permeability for about the first month of life.
This increased permeability was a problem when hexachlorophene was used to scrub newborns.
The chair said the stratum corneum begins to develop in the first month of life, and that shortly
after the first year of life, skin permeability does not appear to vary with age.
Other panelists thought that there might be somewhat different data in the literature, but did not
cite specific studies. One panelist speculated that there might be data on neonate absorption in
the pharmaceutical literature and suggested that EPA look into this. An observer mentioned
work by Fitzpatrick, which indicates that full-term infants have a complete stratum corneum, but
a thinner dermis. The stratum corneum is not necessarily complete in premature infants. A
panelist pointed out that dermal thickness can be critical to absorption of chemicals, including
4-6
-------
both the rate of absorption and total absorption (since not all chemicals absorbed into the skin
enter the circulatory system). Another panelist said that research has shown scrotal skin (which
is in contact with occlusive diapers) to be the most permeable of all.
Panelists discussed skin conditions that may affect absorption. For example, diapers are an
occlusive dressing, so they hydrate the skin and prevent the evaporation or volatilization of
chemicals, which could increase absorption. This could be an issue for babies and toddlers up
through toilet training. Any break in the stratum corneum can compromise the skin barrier.
Children with eczema can have a significant proportion of skin that, is invasively open^ and thus
could be a sensitive subpopulation, one panelist suggested. She also expressed concern about ,
adolescents with acne. This might not be as great a concern, another panelist responded, because
measurements of soil loadings show that the face stays relatively clean compared to hands or
knees. An observer mentioned that children with atopic dermatitis have a compromised barrier,
so bioavailability is much greater, and adults with psoriasis have a compromised barrier, which
can make as much as a 20-fold difference. A panelist pointed out that skin shedding can be
protective for highly lipophilic chemicals (greater than log 5 Kbw), which may be sloughed off
before they ;are absorbed in the circulatory system. Another panelist wondered, whether sunburns
in childhood and adolescence might have any impact on chemical exposure. ,
The chair mentioned that the faces and arms of young children have a greater proportion of body
surface area than adults. This is true for the arms, for example, up to age 6 or 7. Exposure
calculations often use adult proportions for young children, which can be off by 2 percent or so.
Another panelist thought this difference was too small to have regulatory significance, given all
the other uncertainties.
The panelists briefly discussed whether activities that increase blood flow to the skin (such as
crying in infants) could impact absorption. This might be the case for lipophilic chemicals for
which the rate at which they are transported away by the blood is slower than the rate at which
they penetrate the skin.
4-7
-------
4.1.4 Skeleton
Conclusions
Periods of bone growth can increase susceptibility to deposition of substances such as heavy
metals and some antibiotics. Rapid bone growth occurs during the first year, after which the rate
of growth is similar to that for the rest of the body until adolescence, when there is a period of
rapid skeletal growth from approximately age 10 to 16 in girls and 12 to 18 in boys. Epiphysial
closure2 starts around 11 years in girls and 13 years in boys and takes approximately 6 to 7 years,
depending on the site of closure. Interference with this process could shorten the length of the
bone or prevent closure and result in increased bone length, which can cause long-term joint
dysfunction in adult years.
Discussion
The chair presented graphs showing skeletal mass and calcium accretion. During the first year,
there is rapid growth in total skeletal (bony and cartilaginous) mass. This tapers off until
adolescence, when there is another significant increase in skeletal weight from about age 10 to 16
in females and 12 to 18 hi males. Calcium accretion peaks at about 14 years in boys; in girls the
peak is smaller and occurs earlier. During periods of rapid increase in skeletal mass, children
may be more vulnerable to exposure from heavy metals. Another period of vulnerability would
be the period of epiphysial closure and cessation of bone growth, which occurs during much of
adolescence. This is a time when the joint spaces may be susceptible to exposures that could
affect the cartilage. This susceptibility is present from about 11 to 18 years in girls and 12 to 19
years hi boys.
2Epiphyses are secondary bone-forming centers separated from a parent bone in early life by cartilage. When an
epiphysis closes, it becomes part of the larger (parent) bone.
4-8
-------
Dr. Walker, an EPA observer, said he had data showing that the uptake of some heavy metals,
such as, radium and strontium, is high during the neonatal and pubertal growth periods. The
chair responded that the skeletal growth rate is high during the first months of life, but (because
there is so much cartilage growth during the earliest months) this is largely linear growth rather
than an increase in bone mineral content. Another panelist pointed out that weight triples during
the first year. Dr. Walker, said that his milestones, developed using the empirical growth model,
correspond almost exactly with the ones that the group decided on for the skeleton. His model
discerned a growth spurt at birth, between 8 and 12 months, between 3 and 4 years, between 6
and 7 years, at about 8 years hi females and 10 years in males, and at 12 years in females and 14
years in males.
4.1.5 Liver
Conclusions
With regard to environmental agents, hepatic enzyme activity is relatively deficient during the
first year of life, reaches a peak level that exceeds adult efficiency sometime during the second
year of life, and then decreases to the adult range by 2 to 3 years of age. Data on developmental
changes in the liver need to be evaluated.
Discussion
In previous work, two panelists had examined liver enzyme development and found that, at about
1 year of age, enzyme activity exceeds adult activity per gram of liver by about 150 percent.
Enzyme activity returns to normal during the second year, after which it is relatively stable.
However, liver function during the first 6 months after birth is deficient in some respects
compared to adults. Data are available to define these early deficiencies more specifically. A
panelist said that much of the liver function data are fairly old and questioned how relevant they
would be given more recent exposures to compounds that could be liver enzyme inducers.
4-9
-------
4.1.6 Immune System
Conclusions ,
The group emphasized that the immune system was highly complex and that understanding of it
is rapidly evolving. During the first few months of life, infants have greater susceptibility to
certain types of infection, but by about 1 year of age most immune functions have matured.
Discussion
The group noted that the immune system is extremely complex. They pointed out that in utero
exposure could have a significant impact on immune system development but that discussion
was outside the scope of the group's charge. Panelists shared various facts about the immune
system development in children, including the fact that breast-fed babies are less likely to
develop type 1 diabetes and that infants up to about 2 months of age are more susceptible to a
host of infections they are not susceptible to later, while breast-fed babies may be less susceptible
to others due to1 maternal antibodies. Thus, the first 6 to 12 months after birth appear to be very ,
different from the rest of childhood in terms of the potential impacts of exposure. The panelists
pointed out that a number of objective measures, like white cell counts and immunoglobulin
levels, are age-dependent. These could be used as markers for what is normal at particular ages,
but little is known about the significance of deviations from normal levels.
The chair mentioned that though the thymus is large hi infants and small in adults, thymic
function appears to continue into adult life, so the gland's larger size during infancy may not be
significant. One panelist wondered whether data were available on alveolar macrophage
clearance of particles and how that varies with age, since slower clearance during childhood
could affect dose. Another panelist pointed out that EPA and the March of Dimes held a
Workshop to Identify Critical Windows of Exposure for Children's Health in September 1999 in
Richmond, Virginia. The workshop included a work group on the immune and respiratory
4-10
-------
systems, and a report of this work group was provided in Volume 108, Supplement 3 (June 2000)
to Environmental Health Perspectives, pages 483 to 490. This publication includes a series of
articles discussing the timeline for immune system development in humans, mice, and rats. The
authors suggest birth to 1 year as a significant bin for maturation of immune system competence,
and 1 to 18 years as a bin for establishment of the immune system memory (i.e., induction of
response to disease). He suggested that the authors might be useful sources for further
information. The chair suggested that bins for immunological development should be •
established by specialists after thorough examination of the literature in this complex area.
4.1.7 Reproductive and Endocrine Systems
Conclusions ,
Hyperplasia occurs dramatically beginning at about age 8 in girls and age 10 in boys, and
extending for about 8 years in each case. This could be a vulnerable time for potential
mutational changes that could have long-term health impacts. The group did not have data on
when the function of the thyroid, parathyroid, pancreatic islet cells, and renal cortical and
medulary cells become mature. This needs to be examined.
Discussion •
Panelists agreed that in utero exposure was an important concern for development of ovarian and
testicular tissues (as well as other organs), but agreed that in utero exposure was outside the
charge of the group. They discussed what ages would be appropriate to bracket bins for
reproductive development during adolescence. The chair noted that onset of puberty varies by
gender, there is substantial variability within each gender, and onset appears to be happening at
earlier ages. A bin for reproductive organ development that started at age 8 for girls would likely
capture about 95 percent of girls, he said. The bin for boys would need to extend to age 19 to
capture 95 percent of boys. Other panelists suggested 8 to 18 for girls and 10 to 19 for boys; 8 to
4-11
-------
15 years for girls, since few girls have menarche past age 15; and 8 to 16 years for girls, since
most textbooks cite 8 years as the lower limit for breast development (the beginning of puberty in
most girls) and the upper limit for onset is about 14 years, with an additional 2 years until
menarche. The chair pointed out that ovulation usually does not start until about 2 years after
menarche, so a bin should extend for 2 years after menarche to allow for potential effects on
developing oocytes. An observer expressed concern that too broad a bin could wash out the
potential for making associations. Some panelists thought that there were a lot of data in this
area, which would allow graphing of distributions; this would make it possible to create bins that
capture different percentiles of boys and girls.
The chair asked the group to consider the endocrine system as well, since research has shown
that many thyrotoxic drugs can interfere with thyroid development and function if given to
animals early in development. Also, many environmental agents have been shown to interfere
with thyroid function in animals. The chair asked panelists whether they had any information on
the nonreproductive components of the endocrine system, such as the pancreas, thyroid,
parathyroid, and adrenals.
One panelist mentioned that perchlorate, a groundwater contaminant in many areas of the
western United States, is a powerful thyroid toxin that affects iodine uptake in the thyroid.
Panelists did not have any data on whether thyroid sensitivity to environmental agents was
different hi children than in adults. Another panelist thought that even if there are windows of
developmental vulnerability for the endocrine system, this would not necessarily affect the
amount of internal or external dose per exposure event.
Dr. Walker, an EPA observer, said there are chemicals in the environment that have been shown
to affect the thyroid, and since thyroid hormones affect growth, the uptake and distribution of
these chemicals could be affected. The chair agreed with this and asked the group whether there
were periods when those organ systems are more likely to be differentially affected by exposure
than at other times. He said that studies with animals show the induction of permanent thyroid
4-12
-------
dysfunction by using an endocrine disruptor to insult thyroid anlage (the fetal tissue that will
ultimately form the thyroid gland) early in life, but the same compound will not cause
dysfunction if exposure occurs later in life. A panelist said that the State of California has been
concerned about both prenatal and perinatal exposure to perchlorate, which is a powerful thyroid
toxin. She also said that animal studies show that brain maturation is grossly affected by effects
on the thyroid. The chair pointed out that children born without a thyroid will suffer permanent
brain damage if the thyroid function is not replaced within a short time of birth. Another panelist
added that it is well known that thyroid deficiency in childhood can have permanent devastating
effects on brain development. This suggests there is differential susceptibility to thyroid
dysfunction in children compared to adults.
4.1.8 Lung and Respiratory Tract
Conclusions
Changes in the ventilatory rate, alveolar surface area, and oxygen requirements all alter the rate at
which volatile and particulate material enter the system through the respiratory tract.
Developmental changes in the upper airways (the nasal-pharyngeal pathway and the sinuses)
need further evaluation. Development of some sinuses does not begin until after birth and occurs
primarily during the first 6 years of life. Gender differences in ventilatory function develop
around adolescence.
' Discussion
The group discussed how alveolar surface area, ventilation rate, and behavior affected children's
exposure via the lung. One panelist said that, for the first 2 years of life, alveolar surface area
and ventilatory rate are significantly different, so that the minute volume exchange of oxygen and
other gases is significantly higher. He said that exposure to particulate matter in the lung is
approximately four times higher in the infant than in the older child. Another panelist said that a
4-13
-------
study by the California EPA at their Davis facility has shown that lung surface area to body
weight ratio did not change much after 12 years of age, but that infants had a much larger surface
area to body weight ratio and much higher oxygen requirements. She said that ventilation rates
are a key factor in the first 1 to 3 months of life and are quite different than, for example,
ventilation at 1 year of age. Thus, a consideration of ventilation rates would suggest different
bins than the numbers of alveoli or the surface area of alveoli. Only about IS.percent of alveoli
are present at birth; the rest are acquired during the first 3 to 4 years of life. Also, although
infants are not very mobile, their breathing function can be highly active when they cry and
scream. A third panelist pointed out that infants take about 40 to 60 breaths per minute,
compared to 12 breaths per minute in adults.
The chair noted that ventilatory rate decreases during the first 3 months, and then continues to
decrease at a slower rate. He said that alveolar surface area increases until about 3 years of age,
when it changes proportional to body size. He also pointed out that there are significant gender
differences during adolescence with respect to vital capacity, FEV,, (forced expiratory volume in
one minute) and so on. Boys generally have higher ventilatory function than girls of the same
age and weight, though part of that may be related to lean body mass rather than weight itself.
Panelists discussed the significance of nose breathing. Most children are nose breathers, unless
they get a cold (which children do on average six times per year). Style of breathing likely would
not impact exposure to water-soluble compounds, since these can be absorbed in the upper
airways; however, it could impact less soluble compounds, which penetrate the lower airways. A
panelist wondered whether differences in the nose structure with age might be relevant. Another
speculated that there might be a critical period in infancy when there is a high ventilation rate and
lower surface area of the upper respiratory tract that could make infants more susceptible to
exposure of less soluble air pollutants in the lung.
The chair said a lot of data are available on ventilation rate in very young children, including data
on the development of alveolar surface area, on minute volume rates, and therefore on air contact
4-14
-------
with the surface of alveoli. He thought those data were referenced in Chapter 2 of Pesticides in
the Diets of Infants and Children. A panelist thought, data were missing regarding measurements
for specific activities in children under 3 years of age. Another panelist said most data are for
sick children and thought that there were insufficient studies of heathy children. The quality of
the available data must be considered as well, another panelist pointed out, including
considerations of how the data were collected, how accurate they are, whether they are up to date,
and whether there is some reason why they might not be applicable (e.g., they pertain only to sick
children). Many of the techniques in older studies (for example, looking at the rate of blood flow
from specific tissues) are far from state-of-the-art now, said another panelist, and thereis
disagreement about their value. The chair said that many older morphologic studies (e.g.,
measuring the surface area of alveoli) were well done and the data are available in old biology
handbooks.
A panelist cautioned that future research should be directed toward data that would be truly
useful for exposure assessment, and should avoid areas that will not have a significant impact.
For example, an analysis of breathing rate distributions shows only a two-fold difference between
the 5th and 95th percentile, which is insignificant in the overall uncertainty of risk assessment.
The chair pointed outthat it would be difficult for the current panelists to cite specific data gaps,
since they had not recently reviewed the literature, and he suggested that the quality,
comprehensiveness, and significance of the data for exposure assessment would need to be
reviewed carefully for data gaps. He said most of the papers in the Environmental Health
Perspectives review (referenced earlier) recommend more research.
The group briefly discussed the upper respiratory tract. Panelists felt that the nose and upper
airways could have significance since they serve as a chemical entry point, a potential target for
effects, and a potential detoxifying organ; however, the panelists were not aware of data on
whether and how the nose and upper airways might differ between children and adults. A
panelist pointed out that large numbers of children have allergies or adenoidal hypertrophy; when
affected by these, they breathe through their mouths, bypassing the ciliary clearance mechanism
4-15
-------
in the nasopharynx. Another panelist noted that certain sinuses (frontal and sphenoid sinuses) are
not formed at birth, so development of some sinus function takes place after birth. Panelists
listed upper airway changes, including sinus development, as an area that should be examined.
4.1.9 Gastrointestinal Tract
Conclusions
Up to 3 months or so after birth, the infant stomach is more alkaline and has a greater capacity to
absorb intact proteins and large peptides. However, the group did not know whether this
difference had any significance for exposure.
Discussion
The group discussed development of the gastrointestinal tract during childhood. They noted that,
during the first 1 to 3 months of life, the stomach is more alkaline and has a greater capacity to
absorb intact proteins and large peptides than later. This could impact developing sensitivities to
certain proteins, since they get into the system intact at a time when immune tolerance could
occur. However, this difference lasts for 1 to 3 months, and very little data are available
regarding these gastrointestinal differences in babies after the first month of age. One panelist
questioned whether pH differences were relevant during the first month of life, when there is
little variability in diet. Another panelist responded that there are data suggesting that pH is
related to development of methemoglobinemia in infants fed water with high nitrate content. A
third panelist pointed out that the infant stomach may also be exposed to particles cleared from
the deep lung.
The group was not aware of changes in motility of gut during childhood and had no information
on how gut surface area changed. They noted that gut colonization changes with birth and again
with the transition from breast milk to other foods, but probably not much after that.
4-16
-------
Colonization of breastfed and bottlefed infants is different (e.g., breastfed infants are colonized
with lactobacillus rather than E. coli). Beyond the first year of life, even breastfed babies are
exposed to so many other foods that breast milk's impact is much less significant (unless it
contains significant toxicants).
A panelist said data suggest that children's excretions have a high fat content and wondered
whether that was because children have a high fat diet or because they are less efficient at
absorbing fat He said that decreased fat absorption has been suggested as a mitigating factor in
exposure to lipophiles in breast milk.
One panelist questioned the extent to which children with deficient diets or nutrition should be
noted as a subpopulation. A panelist said there likely would not be any gender-related ,
differences in growth rate of the gastrointestinal tract until age 8 or 9, though there could be
individual differences in growth spurts.
4.1.10 Circulatory System
Conclusions
There is a protein-binding deficiency during the first 6 months of life. Fetal hemoglobin, which
is present during the first few months of life, has different binding capacities than adult
hemoglobin. Around age 2 years or younger, children switch from liver and spleen as a source of
hematopoiesis. Extracellular fluid increases during the first 6 months or so of life and then
rapidly decreases until about age 2, when it becomes more consistently related to body size.
Children with G6PD (glucose-6-phosphate dehydrogenase) deficiency may be a susceptible
group. There may well be other susceptible groups of children that will need to be considered as
we learn more about the prevalence of genetic differences.
4-17
-------
Discussion
Panelists discussed features of the circulatory system that might distinguish children from adults.
For example, with regard to blood chemistry during the first 6 months, there are many more
protein-binding displacers relative to protein-binding sites, which means that there is excess
bilirubin. Also, there are fewer basic and acidic binding sites, so there is more drug
displacement. This suggests that the effective dose in the central compartment for chemicals that
are 'significant protein binders will be different in younger children compared to older age groups.
A panelist estimated that protein binding probably was an issue for the first 6 months of life.
Another panelist said that infants have more fetal hemoglobin, which makes them susceptible to
methemoglobinemia. A panelist pointed out that the hematopoietic capabilities of the marrow
change during the first year, but he did not know if this had any toxicological significance. Also,
extramedulary hematopoiesis (i.e., the liver and spleen producing hematopoeitic tissue rather
than just bone marrow) is very significant in the first year of life. After the first year, only bone
marrow produces new blood cells. A panelist said that total body water in infants was higher,
which affects volume, distribution, and clearance. Another panelist mentioned that G6PD
deficiency, a genetic condition that can result in anemia, could characterize a susceptible
subpopulation. Another panelist pointed out that the genetic polymorphism issue overlays the
entire discussion.
4.1.11 Renal System
Conclusions
Before 6 months, renal function (in terms of parameters such as glomerular filtration rate and
tubular maximum) is less than would be predicted by surface area, but by about 6 months of age,
those functions have approached adult values per unit of surface area. So after about 6 months,
renal function can be scaled by surface area.
4-18
-------
Discussion
The chair mentioned a paper he had written showing that kidney function is practically normal at
birth, and essentially reaches adult values per unit of surface area by age 6 months. This
suggested there should be a bin for the first 6 months after birth, since there is still recognizable
immaturity in renal function during that time. After 6 months a separate-bin for the kidney likely
is not necessary. Another panelist said that, while it is generally accepted that metabolic rate (P-
450 metabolism) is substantially greater after the first year and throughout childhood, he has seen
recent information suggesting it is not. These data (which have been criticized) do however
indicate that there are differences in the rate of clearance of drugs but no differences in the rate of
metabolism compared to adult levels. This opens to question the commonly accepted notion that
rates of metabolism and renal clearance are higher during childhood and then gradually diminish
to adult levels during adolescence. A third panelist discussed allometric scaling. He pointed out
that many of the P-450 and conjugation systems mature somewhere between 6 months and 1
year, at which point they can be allometrically scaled to surface area. For example, physicians
can use this scaling to predict blood flow to the liver based on surface area when prescribing
drugs. The fact that the liver system is immature during the first 6 months and the fact that
allometric scaling cannot be done for that period together argue for a bin covering the first 6
months.
4.1.12 Cardiac System
Conclusions
The group had little information about the changes in circulatory dynamics in the heart. They
recommended that EPA examine whether there are any significant changes in cardiac output and
regional blood flow during childhood. ,
4-19
-------
Discussion
The chair kicked off the discussion by saying that the heart generally appears to be fairly mature
at birth, and that he was not aware of any significant changes in shape or function after birth. A
panelist questioned whether blood flow to tissues or organs could change as a function of age.
Panelists'were not aware of significant changes to the cardiac system, but thought it would be
useful for EPA to examine whether there are any significant changes in cardiac output and
regional blood flow during childhood.
4.1.13 Central Nervous System (CNS)
Conclusions
The central nervous system changes rapidly during early development and likely is the system
that needs the most bins. Very little information is available about the duration and variability of
periods of change. Neuronal migration occurs during the first 6 months of life, followed by rapid
myelination, much of which is completed by 2 years of age. At about age 10 years, certain
synapses change and atrophy. This is probably reasonably complete by age 20. The blood-brain
barrier also changes during childhood, though the group did not know when this occurs, and this
could be significant.
Discussion
The chair said that the Environmental Health Perspectives supplement (referenced earlier)
contained a paper by Deborah Rice and Stan Barone, Jr. (pages 511 through 533), with a figure
on the development of the nervous system, which showed that the various areas of the brain
continue developing until age 19. The paper also had data showing that the order in which
nervous system components develop is not necessarily the same in rats as in humans, so there is
difficulty extrapolating from rats to children. Historically scientists have thought that most brain
4-20
-------
development takes place by 1 year of age, and brain size by age 2 is about 90 percent of adult
size. However, research on synaptogenesis indicates that children lose a lot of synapses in early
or mid adolescence in order to develop an adult brain configuration. This and other research
suggests that brain development is not static during childhood, so it is difficult to know what
might be critical periods. A panelist said that for some toxins like lead, the first 2 to 5 years are
critical because this is the period when brain growth is rapid.
The chair said the fact that neonatal deficiency in thyroid function or PKU (phenylketonuria) can
rapidly lead to permanent brain damage suggests that the first 6 or so months of life can be a very
sensitive time. Also, the maturation of the blood-brain barrier would affect the internal dose to
the central nervous system. The window for that maturation is approximately the first year of
life. The chair mentioned concern about the effect of thimerasol, a mercuric component of
vaccines. The effects of vaccination on the central nervous system may have increased as the age
of vaccination has decreased. Though the mechanism for these effects is not known, their
existence suggests that the nervous system is more vulnerable in younger children. He also
mentioned that there are increasing neuropsychiatric disorders for birth to adolescence at varying
times; however, it can be difficult to identify peaks in these disorders with age because the way
in which these diseases are defined and diagnosed at different ages has changed over time.
4.1.14 Muscle
Conclusions
Muscle goes through the same periods of hyperplasia and hypertrophy that other tissues do, but
little is otherwise known about the development of muscular tissue.
4-21
-------
Discussion
Panelists discussed data that might suggest critical periods for muscle development in children.
The chair showed data on creatinine coefficient (a surrogate for muscle mass/development) by
age-and gender. These data showed that the muscle mass of boys and girls began to differ around
12 to 14 years of age. Dr. Walker, an EPA observer, said that lots of data are available on lean
body mass as a function of age. These data show growth spurts as children develop and are
important for assessing the effects of chemicals like cesium, which target muscles in the body.
He also said the size of the muscle compartment is important, because it determines how much of
the chemical is stored. A panelist agreed, saying that while muscle had low perfusion and
metabolism compared to other organs, it is a significant large reservoir for. storage of chemicals
in the body. The size of the muscle compartment is a sensitive parameter in PBPK modeling. It
would therefore be an important variable for neonates because they have a relatively large muscle
compartment.
The chair showed data for lean body mass in boys and girls. He pointed out that changes in body
mass during childhood reflect changes in surface area and weight^ and that there are gender
differences. Dr. Walker pointed out that there are racial/ethnic differences in these parameters.
For example, African-Americans have higher bone densities than other groups, and Caucasians
have higher lung capacities than African-Americans. He thought that genetic or racial
differences should be considered. The chair asked whether data were available for children. A
panelist responded that there was a difference of about 10 to 15 percent in lean body mass in
children, though other panelists felt this was too small to be significant.
A panelist discussed the role of creatinine in dose assessment. Dose assessors routinely use
creatinine'as a corrector for hydration state when evaluating dose for various compounds that are
excreted in urine. The assumption has been that daily creatinine output is constant in children,
but the panelist's (and others') research suggests this is not so. He felt that creatinine output and
variability in children should be better studied, given creatinine's role in dose estimation. The
4-22
-------
chair expressed surprise at this finding, since many studies on urinary excretion assume that
creatinine clearance is a constant unless the subject is dehydrated.
Dr. Walker, an EPA observer, showed a graph illustrating how cancer incidence rates for
rhabdomyosarcomas vary with age. The curves showed that the rates peaked during the neonatal
and pubertal growth periods. He said this corresponded very closely to what is normally found in
height and muscle velocity curves and suggests an association between muscle growth rates and
soft tissue cancers in children. The chair pointed out that this raises the issues of hyperplasia and
hypertrophy in all tissues. Cells are more susceptible to mutation when they are multiplying (i.e.,
during hyperplastic periods) than" when they are simply enlarging (i.e., during hypertrophic
periods). Hyperplastic multiplication tends to occur early in development and continue for some
time; how long it continues varies with the tissue. After that, growth is largely hypertrophic until
adolescence, when hyperplastic growth occurs again. Identifying periods of hyperplastic growth
specific to the various organs and tissues (generally sometime during infancy and early
adolescence) could provide important information about periods of susceptibility. A panelist
agreed this was important and said the State of California was trying to identify hyperplastic time
periods in order to "weight" age at exposure for risk assessments.
Another panelist said that the first year or two of life, when the exposure measures change
rapidly, would be much more important in terms of exposure assessment than the later years of
childhood, when exposure measurement would not vary by age. A different panelist responded
that occupational exposures would start to occur in later adolescence; these exposures should be
considered in a risk assessment. The chair pointed out that muscle mass increases before
strength in adolescence, so adolescents are vulnerable to occupational injury because they are not
as strong as they look. • , :.
4-23
-------
4.1.15 Sensory Organs
Conclusions
The group reported no conclusions for this area.
Discussion
Panelists briefly discussed the eyes and ears. A panelist said that some toxins do enter through
the eyes and ears, but he did not know whether children were more susceptible than adults.
Other panelists pointed out that environmental tobacco smoke is associated with recurrent otitis
media, and recurrent ear infections in early childhood can cause learning disabilities and lifelong
difficulty. Another panelist mentioned data suggesting that young children are more susceptible
to drug toxicity in the middle ear.
4.2 General Issues
Conclusions
• The critical periods for exposure and impact on organs and systems need to be evaluated.
The scientific data are not readily available as to exactly what those periods are, when
they occur, and how sharply defined they are. Research should be done to gather
available data and evaluate their methodology, reliability, sample size, relationship to
current exposure conditions, and variability. Where no data are available, where the
methods are antiquated, or where the data do not cover the age groups, research should be
conducted to gather the data.
• Regarding bins for physiological development, the following general bins could be used
as a very preliminary starting point: 0 to 6 months, 6 to 12 months, and 1 to 3 years for
boys and girls; 3 to 8 years and 8 to 16 years for girls; and 3 to 9 years and 9 to 18 years
for boys. The 0 to 6 month bin will likely need to be further subdivided; possible
appropriate subdivisions might be 0 to 1 month, 1 to 3 months, and 3 to 6 months. The
bins should be correlated with behavior. Bins do not make sense unless there are real
data to back them up, so research should be conducted as mentioned above to gather and
4-24
-------
evaluate existing data to confirm or refine these bins, and data should be developed where
they do not exist.
• Almost all systems start with cell growth in number (hyperplasia) and then go to cell
growth in size (hypertrophy). Periods of hyperplasia have me greatest likelihood of
mutational changes. So many of the early bins reflect the period of cellular hyperplasia.
This becomes particularly obvious historically in studies of body fat, but is also true of
other systems.
• Both premature and low birth weight babies are sensitive subpopulations that need to be
considered separately. With very premature babies (under 1,500 grams), there is even
more concern, since their health status continues to differ from term babies even when
they reach what would have been term. Other susceptible populations should be
considered. The group did not discuss the fetus, but they did emphasize its importance
because so much critical development that can affect the impact of environmental agents
takes place during the fetal period.
• Variability is an important issue that must be considered. Variability is greater during
periods of rapid change and may also be related to gender, age, ethnicity, and genetic
polymorphism.
• The agency should continue to involve pediatricians in further discussion about children's
exposure.
Discussion
Several panelists noted that there are critical periods for children's health when developing target
organs and systems might be particularly vulnerable to toxic effects from exposure to
environmental chemicals. These periods would need to be overlaid with behavioral data to show
whether significant exposure could occur. Ages of particular concern would be those at which
exposure correlates with a critical period for developmental physiology. A panelist wondered
whether these periods were more relevant to hazard identification than exposure assessment, and
suggested the group might want to confine its discussion to physiological changes that affect
portals of entry for exposure to environmental chemicals. Others disagreed, saying that data on
vulnerable periods would help assessors understand where they needed to refine their exposure
estimates.
4-25
-------
Members of the group emphasized that the bins they had indicated for individual organs and
systems were highly preliminary. Data that panelists are aware of suggest there are critical
periods for effects in children, but this needs to be evaluated. They recommended that research
be conducted to systematically review the literature, gather data relevant to defining critical
periods hi target organs, and evaluate these data for reliability, applicability, and relevance to
current exposure and lifestyle factors. Such a search would likely reveal many data gaps in
which additional research would be needed to identify and define critical periods.
The group discussed whether overarching bins or breakpoints could be developed for childhood
based on similarities in the bins for the individual organ systems. A panelist suggested the
following general bins: 0 to 6 months; 6 to 12 months; 1 year to 2 or 3 years; 2 or 3 years to
early adolescence (8 or 9 years depending on gender); and 8 or 9 years to the end of adolescent
growth. Another panelist thought that 0 to 1 month might be a separate bin because so much
change happens during that first month. A third panelist said there would be substantial
variability during the first year, so that at least two bins if not more would be appropriate.
Another panelist expressed concern that too many bins might create a nightmare for the
regulatory community, who could be asked to provide data for each bin.
Panelists discussed the issue of variability. They noted that each bin, rather than being discrete
or fixed, represents a distribution. The larger the variability within a bin, the greater the concern
about using a median or central tendency value for that bin. Even a small bin, for example for
the first month after birth, could have great variability if dramatic developmental changes are
occurring, such as developments in lung function. Since children are a much more variable
population than adults, robust data sets are needed to generate bins small enough to adequately
characterize, and minimize the variability in, the groups the bins represent. Variability is greater
during periods of rapid change (e.g.* growth spurts) and may also relate to gender, age, ethnicity,
and genetic polymorphism. Sufficient data are needed to capture the range of variability and
ensure that each bin is small enough to minimize variability. Good statistical data on distribution
within the bin are important, because they make probabilistic analysis possible; this in turn
4-26
-------
makes it possible for an analysis to reflect the foil spectrum of the parameter, rather than a single-
point estimate that may bias the results. Panelists generally agreed that, in cases of high
variability, distributional data would be very valuable.
Panelists discussed premature and low birth weight babies. A number of panelists felt it
important to consider exposure of premature infants. For example, said one panelist, premature
infants are highly exposed to phthalates. Several panelists suggested that premature babies
represent a special subpopulation, particularly very small premature infants (e.g., under 1,500
grams). Premature infants during the first month or two of life are quite different from full-term
infants in terms of hepatic clearance and other systems. The chair suggested that the bin for .
premature infants could go up to the expected time of birth, at which point they are similar to
foil-term infants unless they are sick. ,
Several other panelists said that low birth weight babies (less than 2,500 grams) should also be
considered a special population. Since mortality and birth weight are a straight line, all low birth
weight children, even at term, are susceptible.
The chair suggested that weight in relation to gestational age should be considered. For example,
a premature infant with a low birth weight (e.g., 1,800 grams) could be more vulnerable than an
infant of the same weight born after 3 7 weeks of gestation. He, suggested that infants with a
weight too low for gestational age be considered a separate group from infants with a weight
appropriate for gestational age. .
Another panelist pointed out that the reason why a child had low birth weight was also important.
Low birth weight due to a congenital infection would likely be a greater concern than low birth
weight due to maternal reasons such as placental insufficiency, hypertension, or smoking, which
probably would not affect growth once the baby was born. Babies and children of any age with
chronic illness should also be considered a special subpopulation, the chair said. Around 10
4-27
-------
percent of children will have some kind of chronic illness (e.g., asthma, cerebral palsy) at some
point in their life that seriously interferes with then: daily activities.
Panelists mentioned that concurrent exposures to other chemicals could potentially be a problem
at any time during childhood. Concurrent exposure to recreational drugs could be a particular
problem in adolescence.
A panelist suggested that, for longer-term research, a longitudinal study to develop a data base to
model growth would be useful. Some other panelists questioned the value of longitudinal
studies. They said that these studies are very expensive and that the results from following
individuals over time are not different enough from cross-sectional data for different ages to
make them worth doing. Another panelist pointed out that the factors that influence exposure
and development (e.g., activity patterns, types of food consumed) change over time. For this
reason, contemporary cross-sectional data would be more valuable than longitudinal data because
they would reflect current conditions.
Dr. Walker, an EPA observer, said that cross-sectional data do not allow identification of discrete
critical growth periods, whereas longitudinal data do. He also said that a database on
developmental variables in childhood would be very helpful for risk assessment. The agency, he
said, needs to better characterize anatomical and physiological parameters, determine whether
sufficient data are available to define critical periods, and develop a scientific basis for defining
critical periods. Ideally, sufficient information ultimately will be available to replace each bin
with a continuous function/model from infancy to maturity.
4.3 Research Recommendations
At the end of the discussion, panelists were asked to individually compile a list of the research
needs for children's exposure they felt were most important. Panelists listed the following needs.
4-28
-------
(Note that since these needs were suggested by individual panelists, they do not necessarily
represent the views of more than one panelist.)
4.3.1 Short-Term Research Needs
EPA should critically evaluate the existence and quality of data bases for organ system
functional development with respect to bins and should particularly identify areas where
there are no data. Variability within the bins should be characterized. A database for
system development through childhood, from birth to maturity, should be developed.
Research is needed to improve our understanding of exposures in the 0 to 6 month bin
(particularly for inhalation) and the 6 to 12 month bin (particularly for nondietary
consumption) because so much change takes place during this time. Also, adolescence
appears to be an important physiological window for which further research may be
needed.
The development of the gastrointestinal tract and its impact on bioavailability for several
classes of compounds should be studied. There is significant uncertainty about degree of
nondietary ingestion and how age and development impact absorption.
A prospective study (perhaps of siblings) would be useful to better understand sex
differences in lipophilic absorption and mobilization and differences in plasma levels for
lipophilic substances. This would be particularly important to a better understanding of
the impact of absorption during adolescent growth spurts.
Research is needed to evaluate common modes of action for pesticides and to develop
and validate biomarkers of exposure that may be useful for a broad spectrum of
pesticides.
Data on food and water intake should be examined to determine their adequacy in
covering intake for each age group. More people should be surveyed and the surveys
should cover more than 2 days. Development of an algorithm to determine the ideal
duration of sampling (i.e., the minimal number of days needed to provide data relevant to
a long-term average) would be useful. Optimal duration may vary with age. For
example, what food is eaten in early life varies much less than what food is eaten at older
ages. The length of the study may need to increase as food intake and variability increase.
Research is needed to improve the general exposure assessment (measuring and
modeling) techniques to improve the power of epidemiologic data for the dose-response
component of risk assessment.
4-29
-------
Research is needed to improve understanding of differences in exposure in critical time
periods for different socioeconomic and ethnic groups.
Research is needed to improve our understanding of, and ability to measure, what specific
contaminants are present in various environmental media to which children are exposed,
including dust, soil, medicines, and vaccines.
In the area of inhalation exposure, research is needed into upper respiratory function
during the first 2 years of life, including scrubbing capability, transport, and ciliary
function.
Research is needed to clarify and generate new data on the lung, including the major
factors that influence absorption in the lung (e.g., surface area, cardiac output), respiratory
rate, and flow rates of water-soluble and lipid-soluble compounds for the deep lung and
upper respiratory tract. Since almost all studies of respiratory situations have been done
with people at rest, respiratory function during exercise appears worth looking at. In
general, the state of the art for measuring absorption in lung of children for different
classes of chemicals needs to be improved.
Research is needed into pulmonary clearance in children, particularly both ciliated and
macrophage clearance of particles from the deep lung in very young children.
Research is needed into the upper respiratory barrier, including the sinuses, oral pharynx,
nasal physiology and scrubbing activity, and the distribution of chemical flow pathways
in the nose (the relationship of flow change to surface area).
More information is needed on liver clearance functions. Data related to liver clearance
mechanisms and enzyme levels (except, possibly, for recent P-450 data) should be
reexamined.
More information is needed on cardiac output.
More information is needed on skin permeability in children with or without occlusion.
For example: At what age does the rate of skin absorption in children match adult
values?
Information on dermal transfer efficiency and absorption kinetics of chemicals such as
pesticides is grossly inadequate. Better information is needed on transfer efficiency and
absorption kinetics to skin.
Research is needed into lipophilic and nonlipophilic substances in breast milk and what
kind of transfer rates exist between milk and the neonatal gut, given the gut pH and
developmental changes. These transfer rates likely vary with time as the gut starts
4-30
-------
maturing. .Current breast rnilk models focus on lipophilic toxics. There is a pH
difference between plasma and breast milk, so breast milk is a sink for some chemicals,
depending on the pharmacokinetics. Research should examine factors other than lipid
solubility that determine what enters breast milk. Also, there is a need for better
measurements of breast milk contaminants (for example, better data are needed on
metals). Breast milk data primarily cover white upper middle class women; better data
may be needed for other socioeconomic groups.
• On the behavioral side, research is needed into transfer rates and efficiencies between
surfaces and the skin, such as hand- or toy-to-mouth transfer rates. In general, better data
are needed on the transfer efficiency for mouthing.
• Research is needed to better define rates of consumption of game fish in children of
different cultures. There are significant cultural differences in how much fish people eat.
• Information on soil ingestion rates in children is not adequate. Both incidence and
prevalence data are needed. Also, better understanding is needed of what fraction of soil
ingestion might be house dust, which could have higher concentrations of some
contaminants.
• Better information is needed on consumption and content of folk and herbal medicines.
Data suggest that about 60 percent of the U.S. population uses these medicines. Some of
the medicines contain toxic components, such as lead.
• Data are needed on the incidence and prevalence of children accompanying parents to
farm fields during periods of farm activities.
4.3.2 Long-Term Research Needs
Panelists debated whether a longitudinal correlative study of children starting at birth and
following them for about 20 years would be useful. Some felt such a study would be
useful, while others felt it would too expensive and yield results that might be out of date
by the time the study was completed. However, somewhat longer-term exposure
information may be needed than the short-term/cross-sectional data currently available.
In particular, longitudinal data through first year or two of life could be useful, and
differences due to socioeconomic status or ethnicity should be noted.
Long-term research should include research on critical periods and possible exposure
differences in fetal development. • '
4-31
-------
Ultimately, sufficient data should be gathered to develop a continuous multivariate model
that can replace bins. Over the long term, EPA should work to develop this type of
model.3
3In a post-meeting comment, one panelist noted that he strongly dissented with this
recommendation and felt that it would not be a worthwhile line of research. Given the variation
in the physical-chemical behavior of agents of interest and the multiplicity of exposure scenarios
and toxicological endpoints EPA must deal with, he felt that a single model would not be
feasible.
4-32
-------
5. FINAL PLENARY DISCUSSION
Kimberly Thompson delivered some summary remarks to synthesize the findings of the two
discussion groups and introduce the plenary session of the workshop. It is clear, she gathered,
that there is a great deal of work that remains to be done in learning how to assess children's
exposure. The discussions of the two groups have highlighted the basic fact that it is inadequate
to lump children into a single category when conducting risk and exposure assessment. They
have also pointed out that there are existing data sources that could tell us a great deal about
children's exposure factors if they are only properly analyzed or re-analyzed. There ought to be a
comprehensive review of the data that have not been adequately analyzed.
The behavior group, which addressed the question of whether or not age bins were really
accurate representations of childhood development, indicated that bins should be replaced by
some other analytical framework in the long run. For the short term, while one must use age
bins, it is encouraging to note that there is substantial overlap between the provisional binning
systems developed by the anatomy and behavior groups. There are, however, some
inconsistencies to be worked out in the age ranges between 2 and 15 years. One panelist
reiterated the point that users of bins should understand the underlying distributions of behavior
that those bins are approximating. Another panelist pointed out that there was no reason to use a
fixed set of bins for all children's exposure assessment. One could use a different set of exposure
bins for the different organ systems targeted by a chemical, so as to make sure that the bins were
firmly centered around the relevant windows of vulnerability.
One panelist asked whether the present exposure factor data were sufficiently copious that they
could be re-analyzed and broken down into more detailed age bins. Dr. Shea, chair of the
behavior group, stated that the answer to that question is unknown, but estimated that the
adequacy of present data sets would be highly variable. In some exposure scenarios, children
could be rebinned, but in others one would create sample sizes that would be too small (e.g., less
than 10 children) if one broke age categories down more finely. Another panelist seconded the
5-1
-------
point that exposure factors should be evaluated in a chemical-specific or exposure-specific
fashion.
Several panelists inquired about the next steps that EPA would take in connection with the issues
raised at the workshop. William Wood suggested that EPA might produce a supplement to the
1992 Exposure Assessment Guidelines that would incorporate some of the conclusions from .the
workshop. He also indicated that EPA was organizing a workshop on children's
pharmacokinetics. One panelist suggested that EPA use the ToxProfile document as a guide to
organize data well.
Pediatricians among the panelists were interested in continuing and strengthening the partnership
between the risk assessment community and. the pediatric community that they saw developing at
the workshop. One panelist suggested that EPA reach out to pediatric specialists who might be
able to provide more specific answers to its questions about behavioral and physiological child
development. The EPA should also start to actually use age-specific exposure factors in its risk
assessments.
Dr. Thompson asked the panelists and observers to suggest important lessons for EPA that
emerged from the workshop. Individual panelists and observers suggested that EPA:
• Draw on the knowledge of experts in developmental biology, physiology, pharmacology,
and toxicology.
• Identify three or four epidemiologists to serve on an ongoing basis as risk assessment
advisors.
• Carefully verify the general conclusions of the panelists by going back to the specific
exposure data. EPA should realize that the panelists did not have access to this data
while forming their conclusions—they were working solely from their general
knowledge.
5-2
-------
• Use a group like the International Life Sciences Institute or the National Academy of
Sciences to pull together the necessary experts on a long-term basis to address children's
exposure issues.
• Work in the context of specific exposure scenarios.
• Gather new exposure factor data to flesh out the new age bins that are created.
• Evaluate the variability within particular age bins and study how different exposure
pathways might require different age bins.
• Carefully distinguish behavioral data from other kinds of data and avoid exaggerating
their precision. There is a particular temptation to treat quantified behavioral data as if
they were engineering data, simply because they are expressed in numbers.
» Be careful to reflect ethnic variability.
Some panelists indicated they were not adequately notified that they would be asked to come up
with short-term research goals at the workshop. Therefore, EPA should not take the suggestions
offered at the workshop to be comprehensive—panelists might have come up with more
suggestions if they had had more notice.
Dr. Thompson thanked the panelists for their participation and hard work in connection with the
workshop. Jan Connery thanked the panelists on behalf of ERG and adjourned the meeting.
5-3
-------
-------
APPENDIX A
LIST OF PANELISTS
-------
-------
&EPA
United States
Environmental Protection Agency
Office of Research and Development
Technical Workshop on Issues Associated with
Considering Developmental Changes in Behavior and
Anatomy When Assessing Exposure to Children
Holiday Inn on the Hill
Washington, DC
July 26-27, 2000
List of Participants
Thomas W. Armstrong
Senior Staff Industrial Hygienist
Exposure Sciences Section
ExxonMobil Biomedical Sciences, Inc.
1545 Route 22, E - Room LF294
Annandale, NJ 08801-0971
908-730-1114
Fax:908-730-1192
E-mail: twarmst® erenj.com ,
Sophie J. Balk
Attending Pediatrician
Montefiore Medical Center
65 Hunter Avenue
New Rochelle, NY 10801
718-405-8090
Fax:718-405-8091
E-mail: sbalk@montefiore.org
Deborah H. Bennett
Post-doctoral Researcher
Lawrence Berkeley National Laboratory
Energy Technologies Division
1 Cyclotron Road (90-3085)
Berkeley, CA 94720
510-486-6945
E-mail: dhbennett@lbl.gov
James V. Bruckner
Professor of Pharmacology & Toxicology
College of Pharmacy
University of Georgia
D.W. Brooks Drive
Athens, GA 30602-2352
706-542-5405
Fax: 706-542-3398
E-mail: bruckner@rx.uga.edu
Michael Dinovi
Chemist
Center for Food Safety & Applied Nutrition
U.S. Food & Drug Administration
200 C Street, SW (HFS-246)
Washington, DC 20204
202-418-3003
Fax:,202-418-3030
E-mail: mdinovi@cfsan.fda.gov
Richard Fenske
Professor
Department of Environmental Health
University of Washington
Health Sciences Building - Room F-233
Seattle, WA 98195
206-543-0916 ., .
Fax:206-616-2687
E-mail: rfenske@u.washington.edu
Gary Ginsberg
Toxicologist
Department of Environmental Epidemiology
and Occupational Health (EEOH)
Toxic Hazards Assessment
Connecticut Department of Public Health
410 Capitol Avenue (MS 11CHA)
Hartford, CT 06134
860-509-7750
Fax: 860-509-7785
E-mail: gary.ginsberg @ po.state.ct.us
i Printed on Recycled Paper
A-3
-------
Lynn Goldman
Adjunct Professor
School of Hygiene and Public Health
Johns Hopkins University
624 North Broadway - Room 441A
Baltimore, MD 21205-1996
410-614-9301
Fax:410-614-8964
E-mail: lgoldman@jhsph.edu
Robert Johnson
Medical Officer
Exposure Investigation Section
Agency for Toxic Substances and
Disease Registry
1600 Clifton Road (E-32)
Atlanta, GA 30333
404-639-5177
Fax: 404-639-0655
E-mail: rdj2@cdc.gov
Celestine Kiss
Engineering Psychologist
Division of Human Factors
U.S. Consumer Product Safety Commission
4330 East West Highway - Room 717B
Bethesda,MD 20814-4408
301-504-0468, Ext. 1284
Fax:301-504-0407
E-mail: ckiss@cpsc.gov
John C. Kissel
Associate Professor
Department of Environmental Health
University of Washington
1705 Northeast Pacific Street
Health Sciences E179A
Seattle, WA 98105
206-543-5111
Fax:206-543-8123
E-mail: jkissel@u.washington.edu
Bruce P. Lanphear
Associate Professor
Department of Pediatrics
The University of Cincinnati and .
Children's Hospital Medical Center
3333 Burnet Avenue
CH-1 South - Room 1123
Cincinnati, OH 45229-3039
513-636-3778
Fax:513-636-4402
E-mail: bruce.lanphear@chmcc.org
James O. Leckie
Professor
Department of Civil & Environmental
Engineering
Terman Engineering Center (M25)
Stanford University
Stanford, CA 94305-4020
650-723-2524
Fax:650-725-3164
E-mail: leckie@ce.stanford.edu
Melanie A. Marty
Supervising Toxicologist & Section Chief
Air Toxicology & Epidemiology Section
Office of Environmental Health
Hazard Assessment
California Environmental Protection Agency
1515 Clay Street -16th Floor
Oakland, CA 94612
510-622-3154
Fax:510-622-3210
E-mail: mmarty@oehha.ca.gov
Mary Kay O'Rourke
Research Associate Professor
College of Public Health
The University of Arizona
1435 North Fremont Avenue
Tucson, AZ 85721-0468
520-626-6835
Fax: 520-882-5014
E-mail: maryk@hrp.arizona.edu
George C. Rodgers
Professor of Pediatrics and
Pharmacology & Toxicology
Division of Pediatric Critical Care
University of Louisville
571 South Floyd Street - Suite 332
Louisville, KY 40202
502-852-3720
Fax: 502-852-8626
E-mail: gcrodgers@pol.net
P. Barry Ryan
Professor
Department of Environmental &
Occupational Health
Rollins School of Public Health
Emory University
1518 Clifton Road, NE
Atlanta, GA 30322
404-727-3826
Fax:404-727-8744
E-mail: bryan@sph.emory.edu
A-4
-------
Margo Schwab
Assistant Director
Risk Sciences & Public Policy Institute
Johns Hopkins School of Public Health
615 North Wolfe Street - Room W6033
Baltimore, MD 21205
410-614-4962
Fax:410-955-0863
E-mail: mschwab@jhsph.edu
Katherine M. Shea - Discussion Leader
Consultant
1 Buttons Road
Chapel Hill, NC 27514
919-933-2699
E-mail: tkmjshea@mindspring.com
Kimberly Thompson - Workshop Chair
Harvard Center for Risk Analysis
718 Huntington Avenue
Boston, MA 02115
617-432-4285
Fax:617-432-0190
E-mail: kimt@hsph.harvard.edu
William B. Weil - Discussion Leader
Professor Emeritus
Department of Pediatrics/Human Development
College of Human Medicine
Michigan State University
B140 Life Sciences Building
East Lansing, Ml 48824-1317
517-351-5615
Fax:517-353-4584
E-mail: weilw@pilot.msu.edu
Robin M. Whyatt
Assistant Professor of Clinical Public Health
Division of Environmental Health Sciences
Joseph L. Mailman School of Public Health
Columbia University
60 Haven Avenue - B-1
New York, NY 10032
212-304-7273
Fax:212-541-1943
E-mail: rmw5@columbia.edu
A-5
-------
-------
APPENDIX B
LIST OF OBSERVERS
-------
-------
&EPA
United States
Environmental Protection Agency
Office of Research and Development
Technical Workshop on Issues Associated with
Considering Developmental Changes in Behavior and
Anatomy When Assessing Exposure to Children
Holiday Inn on the Hill
Washington, DC
July 26-27, 2000
List of Observers
Lucy Ament
Assistant Editor,
Pesticide & Toxic Chemical News
CRC Press, LLC
1725 K Street, NW - Suite 506
Washington, 20006
202-887-6320 Ext.: 111
Fax: 202-887-6335
E-mail: lament@crcpress.com
Silas Anamelechi
Research Assistant
Engineering Department
Howard University
Washington, DC 20059
202-806-9250
Fax: 202-806-4430
E-mail: yrking@hotmail.com
Katharine Anitole
Biologist
Existing Chemicals
Assessment Branch
Office of Prevention, Pesticides &
Toxic Substances
U.S. Environmental
Protection Agency
401 M Street, SW (7403)
East Tower - Room 611G
Washington, DC 20460
202-260-3993
Fax:202-260-1279
E-mail: anitole.katherine@epa.gov
Vincent Arena
Assistant Professor of Biostatistics
Department of Biostatistics
University of Pittsburgh
318 Parran Hall
Pittsburgh, PA 15261
412-624-3023
Fax:412-624-2183
E-mail: arena+@ pitt.edu
Ayaad Assaad
Toxicologist
Registration Action Branch
Health Effects Division
U.S. Environmental Protection
Agency
1921 Jefferson Davis Highway
(7509C)
Arlington, VA 22202
703-305-0314
Fax:703-305-5147
E-mail: assaad.ayaad@epa.gov
Robert Beiiies
Toxicologist
National Center for
Environmental Assessment
U.S. Environmental Protection
Agency
Ariel Rios Building (8623D)
1200 Pennsylvania Avenue
Washington, DC 20460
202-564-3273
Fax: 202-565-0078
E-mail: beliles.robert@epa.gov
Charlotte Bertrand
Environmental Scientist
U.S. Environmental
Protection Agency
Ariel Rios Building (5307W)
1200 Pennsylvania Avenue
Washington, DC 20460
703-308-9053
Fax: 703-308-0511
E-mail:
bertrand.charlotte@epa.gov
Erin Birgfeld
Environmental Protection
Specialist
U.S. Environmental
Protection Agency
Ariel Rios Building (6205-J)
1200 Pennsylvania Avenue
Washington, DC 20460
202-564-9079
E-mail: birgfeld.erin®
epa.gov
Elizabeth Boa
Senior Manager,
Policy Economics & Risk Analysis
American Chemistry Council
1300 Wilson Boulevard
Arlington, VA 22209
703-741-5234
Fax:703-741-6040
E-mail: elizabeth_boa@
americanchem istry.com
B-3
-------
Christine Chaisson
CF Chaisson Scientific Advisors
4610 Quarter Charge Drive
Annandale, VA 22003
703-978-6496
Fax: 703-978-6962
E-mail: chaissoninc@erols.com
David Chen
Health Scientist
Office of Children's Health Protection
U.S. Environmental
Protection Agency
1200 Pennsylvania
Avenue, NW (1107)
Washington, DC 20460
202-260-7778
Fax:202-260-4103
E-mail: chen.david@epa.gov
H. Gregg Claycamp
Associate Professor
Department of Environmental &
Occupational Health
University of Pittsburgh
260 Kappa Drive
Pittsburgh, PA 15238
412-967-6524
Fax:412-624-1020
E-mail: hgc2@cis.pitt.edu
Cheryl Cleveland
Research Scientist/Risk Leader
Global Exposure & Risk Assessment
Dow Agro Sciences
9330 Zionsville Road
Indianapolis, IN 46268
317-337-3532
E-mail: cbcleveland@dowagro.com
Jeffrey Dawson
Chemist
Office of Pesticide Programs
Health Effects Division
U.S. Environmental Protection
Agency
401 M Street, SW (7509C)
Washington, DC 20460
703-305-7329
E-mail: dawson.jeff@epa.gov
Emma Demastrie
Intern
Synthetic Organic Chemical
Manufacturers Association
1850 M Street, SW - Suite 700
Washington, DC 20036
202-721-4186
E-mail:
emma.demastrie@socma.com
Angelina Duggan
Director of Science Policy
American Crop Protection Association
1156 15th Street - Suite 400
Washington, DC 20005
202-872-3885
Fax: 202-463-0474
E-mail: angelina@acpa.org
Carol Eisenmann
Research Associate
The Cosmetic, Toiletry, &
Fragrance Association
1101 17th Street, NW - Suite 300
Washington, DC 20036
202-331-1770
Fax:202-331-1969
E-mail: eisenmannc@ctfa.org
Penelope Fenner-Crisp
Senior Science Advisor
Office of Pesticide Programs
U.S. Environmental Protection
Agency
1200 Pennsylvania Avenue, NW
(7501C)
Washington, DC 20460
703-605-0654
Fax: 703-308-4776
E-mail: fenner-
crisp.penelope@epa.gov
Michael Firestone
Science Director
Office of Children's
Health Protection
U.S. Environmental
Protection Agency
401 M Street, SW (1107)
Washington, DC 20450
202-260-2899
E-mail: f irestone.michael @ epa.gov
Elaine Francis
National Program Director
Endocrine Disrupters
Research Program
Office of Research & Development
National Center for
Environmental Assessment
U.S. Environmental
Protection Agency
Ariel Rios Building (8701R)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-6789
Fax: 202-565-2444
E-mail: francis.elaine@epa.gov
Yosef Gebrekristios
Atmospheric Chemistry
Department
Howard University
733 Harvard Street, NW
Washington, DC 20001
202-939-0374
E-mail: yosfu@hotmail.com
Lee Hofmann
Environmental Health Scientist
U.S. Environmental
Protection Agency
1200 Pennsylvania Avenue, NW
(5202G)
Washington, DC 20460
703-603-8874
Fax:703-603-9133
E-mail: hofmann.lee@epa.gov
Karen Hopfl-Harris
Associate Director of Policy
Environment & Health Division
Physicians for Social
Responsibility
1101 14th Street, NW - Suite 700
Washington, DC 20005
202-898-0150
Fax:202-898-0172
E-mail: khopfl@psr.org
Elaine Cohen Hubal
Chemical Engineer
U.S. Environmental
Protection Agency (MD-56)
Research Triangle Park, NC
27711
919-541-4077
Fax:919-541-0905
E-mail: hubal.elaine@epa.gov
Abby Jacobs
Pharmacology/Toxicology
Team Leader
Division of Dermatologic &
Dental Drug Products
U.S. Food & Drug Administration
5600 Fishers Lane (HFD-540)
Rockville, MD 20857
301-827-2020
Fax:301-827-2075
E-mail: jacobs@cder.fda.gov
B-4
-------
Patrick Kennedy
Exposure Analysis Branch
Office of Prevention, Pesticides
& Toxic Substances
U.S. Environmental
Protection Agency
1200 Pennsylvania Avenue, NW
(7406)
Washington, DC 20460
202-260-3916
Fax: 202-260-0981
E-mail: kennedy.patrick@epa.gov
Carole Kimmel
Senior Scientist
National Center for
Environmental Assessment
U.S. Environmental
Protection Agency
Ariel Rios Building (8623D)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-3307
Fax: 202-565-0050
E-mail: kimmel.carole@epa.gov
Steve Knott
National Center for
Environmental Assessment
Office of Research & Development
U.S. Environmental
Protection Agency
Ariel Rios Building (8601 D)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-3359
Fax:202-565-0062
E-mail: knott.steven@epa.gov
Krishna Kumar
Professor of Biopharmaceutics &
Pharmacokinetics
School of Pharmacy
Howard University
2300 4th Street, NW
Washington, DC 20059
202-806-6540
Fax: 202-806-7805
E-mail: kkumar@howard.edu
Jim Laurenson
Project Manager
Risk & Enviromental
Assessment Practice
ICF Consulting
9300 Lee Highway
Fairfax, VA 22031-1207
703-934-3648
Fax: 703-934-9740
E-mail: jlaurenson@icfconsulting.com
Timothy Leighton
Environmental Health Scientist
Health Effects Division
Office of Pollution Prevention &
Toxic Substances
U.S. Environmental Protection
Agency
Ariel Rios Building (7509C)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
703-305-7435
E-mail: leighton.timothy@epa.gov
Benjamin Lim
Chemist
Program Assessment &
Outreach Branch
National Program Chemical Division
Office of Pollution Prevention & Toxic
Substances
U.S. Environmental Protection
Agency
401 M Street, SW (7404)
Washington, DC 20460
202-260-1509
Fax: 202-260-3453
E-mail: lim.benjamin@epa.gov
Amal Mahfouz
Senior Toxicologist
Science & Technology Branch
Office of Water
U.S. Environmental
Protection Agency
1200 Pennsylvania Avenue, NW
(4304)
Washington, DC 20460
202-260-9568
Fax:202-260-1036
E-mail: mahfouz.amal@epa.gov
Susan Makris
Toxicologist
Office of Pesticide Programs
U.S. Environmental
Protection Agency
1200 Pennsylvania Avenue, NW
(7509C)
Ariel Rios Building
Washington, DC 20460
703-305-5222
Fax: 703-605-0670
E-mail: makris.susan@epa..gov
Elizabeth Margosches
Statistician
Existing Chemicals
Assessment Branch
Office of Prevention, Pesticides &
Toxic Substances
U.S. Environmental
Protection Agency
1200 Pennsylvania Avenue, NW
(7403)
Ariel Rios Building
Washington, DC 20460
202-260-1511
E-mail: margosches.elizabeth®
epa.gov
Alec McBride
Senior Analyst
Office of Solid Waste
U.S. Environmental
Protection Agency
1200 Pennsylvania Avenue
(5307W)
Ariel Rios Building
Washington, DC 204600
703-308-0466
Fax:703-308-0511
E-mail: mcbride.alexander®
epa.gov
Linda Meredith
Risk Assessor/Environmental
Scientist
Engineering & Environmental
Management Group
Science Applications
International Corporation
11251 Roger Bacon Drive
(MSR-3-1)
Reston, VA 20190
703-318-4741
Fax:703-709-1042
E-mail: linda.a.meredith@
cpmx.saic.com
B-5
-------
Marsha Morgan
Environmental Health Scientist
Human Exposure Analysis Branch
U.S. Environmental
Protection Agency
79 West Alexander Drive (MC56)
Research Triangle Park, NC 27711
919-541-2598
Fax:919-541-0905
E-mail: morgan.marsha@epa.gov
Siroos Mostaghimi
Environmental Engineer
Antimicrobials Division
Office of Pollution Prevention
& Toxic Substances
U.S. Environmental
Protection Agency
401 M Street, SW (751OC)
Washington, DC 20460
703-308-8337
E-mail: mostaghimi.siroos®
epa.gov
Jacqueline Moya
Environmental Engineer
Exposure Analysis & Risk
Characterization Group
National Center for
Environmental Assessment
U.S. Environmental
Protection Agency
1200 Pennsylvania Avenue, NW
(8623D)
Washington, DC 20460
202-564-3245
Fax: 202-565-Q079
E-mail: moya.jacqueline@epa.gov
Barbara Neal
Senior Toxicologist
BBL Sciences
1801 Robert Fulton Drive - Suite 400
Reston.VA 20190
703-375-8575
E-mail: bhn@bbl-inc.com
Stephen Olin
Deputy Director
Risk Science Institute
International Life Sciences Institute
1126 16th Street, NW
Washington, DC 20036
202-659-3306
Fax:202-659-3617
E-mail: solin@ilsi.org
Dennis Pagano
Environmental Health Scientist
Risk & Exposure
Assessment Group
Emission Standards Division
U.S. Environmental Protection
Agency (MD-13)
Research Triangle Park, NC 27711
919-541-0502
Fax:919-541-0840
E-mail: pagano.dennis@epa.gov
Jerome Paulson
Children's Environmental
Health Network
110 Maryland Avenue, NE - Suite 511
Washington, DC 20002
202-543-4033
Fax: 202-543-8797
E-mail: jpaulson@cehn.org
Andrea Pfahles-Hutchens
Epidemiologist
Existing Chemicals
Assessment Branch
Risk Assessment Division
U.S. Environmental Protection
Agency
Ariel Rios Building (7403)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-260-0288
Fax: 202-260-1279
E-mail: pfahles-
hutchens.andrea@epa.gov
Lorence Pope
Environmental Engineer
Office of Air Quality,
Planning & Standards
U.S. Environmental
Protection Agency (MD-15)
Research Triangle Park, NC 27711
919-541-0682
Fax: 919-541-0237
E-mail: pope.lorence@epa.gov
Harvey Richmond
Environmental Protection
Specialist
Air Quality Strategies &
Standards Division
Office of Air Quality
Planning & Standards
U.S. Environmental
Protection Agency (MD-15)
Research Triangle Park, NC
27711
919-541-5271
Fax:919-541-0237
E-mail: richmond.harvey@epa.gov
James Rowe
Science Administrator
Cross Program
Office of Science Policy
U.S. Environmental
Protection Agency
Ariel Rios Building (8103R)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-6488
Fax: 202-565-2925
E-mail: rowe.james@epa.gov
June Samuel
Performance Improvement
Department
Synthetic Organic Chemical
Manufacturer's Association
1850 M Street
Washington, DC 20036
202-721-4163
E-mail: june.samuel@socma.com
Erica Schmitt
Staff Assistant
Synthetic Organic Chemical
Manufacturers Association
1850 M Street, SW - Suite 700
Washington, DC 20036
202-721-4100
E-mail: erica.schmitt@socma.com
Debbie Smegal
Toxicologist
Health Effects Division
Office of Pollution Prevention
& Toxic Substances
U.S. Environmental
Protection Agency
401 M Street, SW (7905C)
Washington, DC 20460
202-305-7457
E-mail: smegal.debbie@epa.gov
B-6
-------
Bob Sonawane
Chief, Effects Identification &
Characterization Group
National Center for
Environmental Assessment
U.S. Environmental
Protection Agency
Ariel Rios Building (8623D)
1200 Pennsylvania Avenue
Washington, DC 20460
202-564-3292
Fax: 202-565-0078
E-mail: sonawane.bob@epa.gov
Greg Susan ke
Biologist
Office of Pollution Prevention
& Toxic Substances
U.S. Environmental
Protection Agency
401 M Street, SW (7404)
Washington, DC 20460
202-260-3547
Fax: 202-260-0001
E-mail: susanke.greg@epa.gov
Nancy Sussman
Assistant Professor
Department of Environmental Health
University of Pittsburgh
260 Kappa Drive
Pittsburgh, PA 15217
412-967-6545
Fax:412-624-1020
E-mail: nbsl @ pitt.edu
Sandra Tirey
Assistant Vice President,
Regulatory & Technical Affairs
American Chemistry Council
1300 Wilson Boulevard
Arlington, VA 22209
703-741-5202
Fax:703-741-6056
E-mail: sandra_tirey@
americanchem istry.com
Nicolle Tulve
Physical Scientist
Human Exposure Analysis Branch
U.S. Environmental Protection
Agency
79 TW Alexander Drive (MD-56)
Research Triangle Park, NC 27711
919-541-1077
Fax:919-541-0905
E-mail: tulve.nicolle@epa.gov
Vanessa Vu
Associate Director
National Center for
Environmental Assessment
U.S. Environmental Protection
Agency
Ariel Rios Building (8601-D)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-3282
Fax: 202-565-0066
E-mail: vu.vanessa@epa.gov
James Walker
National Center for
Environmental Assessment
Office of Research & Development
U.S. Environmental Protection
Agency
Ariel Rios Building (8623D)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-3316
Fax: 202-565-0078
E-mail: walker.james@epa.gov
Isabel Walls
Senior Scientist
Risk Science Institute
International Life Sciences Institute
1126 16th Street, NW
Washington, DC 20036
202-659-3306
Fax:202-659-3617
E-mail: iwalls@ilsi.org
Karen Werner
Reporter
Bureau of National Affairs
1231 25th Street, NW
Washington, DC 20037
202-452-4130
Fax:202-452-4150
Amina Wilkins
Senior Environmental Scientist
National Center for
Environmental Assessment
U.S. Environmental Protection
Agency
Ariel Rios Building (8623D)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-3256
Fax: 202-565-0076
E-mail: wilkins.amina@epa.gov
Bill Wood
Executive Director
Risk Assessment Forum
Office of Research & Development
U.S. Environmental
Protection Agency
Ariel Rios Building (8601 D)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
202-564-3358
Fax: 202-565-0062
E-mail: wood.bill@epa.gov
B-7
-------
-------
APPENDIX C
CHARGE TO THE EXPERTS
-------
-------
Technical Workshop on Issues Associated with Considering Developmental
Changes in Behavior and Anatomy When Assessing Exposure to Children
U.S. Environmental Protection Agency
Washington, D.C.
July 26-27, 2000
Charge to Experts/Discussion Issues
This workshop is being held to discuss issues associated with how to consider important
developmental changes when assessing the exposure of children to environmental contaminants.
The workshop discussions will focus on broad technical issues rather than any one specific
methodology. These issues were raised by Agency scientists who have been working to improve
exposure and risk assessment methodologies for children in response to the President's
Executive Order (Executive Order 13045) and such legislative mandates as the Food Quality
Protection Act of 1996. The focus of the workshop discussions will be on defining and
characterizing the important facets of child development and how best to estimate childhood
exposure given the limitations in existing exposure information.
Background
The 1993 National Academy of Sciences (NAS) report "Pesticides in the Diets of Infants
and Children" highlights important differences between children and adults with respect to risks
posed by pesticides. Some of the principles in the NAS report provided the foundation for the
Food Quality Protection Act of 1996 (FQPA) and the President's Executive Order 13045,
Protection of Children from Environmental Health Risks and Safety Risk. FQPA requires the
consideration of aggregate exposure to children when establishing pesticide tolerances (legal
limits for residues in food). Executive Order 13045 broadens consideration of impacts on
children by stating that "each Federal agency: shall ensure that its policies, programs, activities,
and standards address disproportionate risks to children that result from environmental health
risks or safety risks." Many of the comments the EPA received on the Proposed Guidelines for
Carcinogen Risk Assessment relate to the implementation of Executive Order 13045. In
response to these comments and regulatory initiatives, EPA has been investigating ways to
improve Agency risk assessments for children.
An Agency workgroup convened under the auspices of the Risk Assessment Forum has
been exploring children's exposure assessment issues. This workgroup has concluded that a
major issue facing Agency assessors is how to consider age related changes in behavior and
physiology when preparing exposure assessments for children. Children's behavior changes over
time in ways that can have an important impact on exposure. Further, children's physiology
changes over time in ways that can impact both their exposures and their susceptibility to certain
health effects. There are two aspects to these physiological changes. First, there are anatomical
C-3
-------
changes resulting from physical growth. Second, there are changes in pharmacokinetics and
pharmacodynamics which affect the absorption, distribution, excretion and effects of
environmental contaminants. The Agency is examining the pharmacokinetic/pharmacodynamic
changes in children through a separate effort. The present workshop discussions will focus on
how to consider age related changes in behavior and anatomy.
Discussion Issues
The broad technical issues identified for the workshop discussions can be organized into
two categories: issues associated with behavioral changes in children and issues associated with
anatomical changes and physical growth. Although organized in this manner to facilitate
discussions, it is understood that these two categories are considerably intertwined. The issues
identified under each category will be the focal point for discussions during this workshop. These
issues/questions are intended to help structure and guide, not limit the workshop discussions.
In addressing these issues/questions, workshop participants are asked to consider several
overarching questions:
• what is the ideal approach to preparing childhood exposure assessments that reflect
changes in children's behavior and anatomy over time;
• is the existing exposure information adequate to implement the ideal approach; if not,
what additional information is needed;
• what short term studies could be conducted to supply the necessary information or
provide additional guidance; and
what longer term research may be needed to achieve the ideal approach to preparing
childhood exposure assessments.
Behavioral Changes During Child Development and
Their Impact on Exposure to Environmental Contaminants
Childhood behavior changes over time in ways that can have an important impact on
exposure to environmental contaminants. These changes are linked to physical and mental
growth and can influence where children spend their time, what physical activities they engage
in, and what foods they eat. Rephrased in terms of exposure factors, these changes can influence
time spent in microenvironments, the frequency and duration of micro and macro level activities,
and the intake rate for water and selected foods and beverages. Recognizing the importance of
these changes in behavior, exposure assessors have invariably estimated exposure for such
subgroups as infants, toddlers, children, and adolescents. The ages ascribed to these groups vary
and are often based on the exposure pathways and routes of concern, expert judgment, and/or the
C-4
-------
availability of exposure information. The goal of the present discussion is to examine how
childhood behavior changes over time, identifying those aspects of behavior that are most
important for consideration in exposure assessment. The following questions will serve as a
guide for this discussion.
1. Does it make sense to think about childhood behavioral development as a series of
discrete events which lend themselves to characterization using age group
categories or "bins?" Alternatively, should exposure assessors be thinking in terms
of a continuum of behavioral development that contributes to an exposure
function over all ages? If so, how would one pursue this later approach? When
existing information is not adequate to construct an exposure function that reflects
continuous behavioral development, a consistent, default approach using age
group "bins" may be needed. In such cases, what "bins" serve as a reasonable
surrogate for the continuous function? How would one characterize the
uncertainties that arise from the use of such "bins?" . ,,
2. What are the most important developmental milestones in children's behavior?
For each milestone, what is the range of ages during which the behaviors are ,
typically observed? How much variability is there among children with respect to
the age of onset and the age of abandonment (if applicable) for these behaviors?
Are the observed changes in behavior associated with these milestones likely to
affect children's exposure to environmental contaminants? If so, how?
3. For those behaviors that are likely to have an important impact on exposure, is
there existing exposure information that is representative of the behavior?
Comment on the existing information including some indication of accessibility
and quality. If such information is not available, is there exposure information
that could serve as a reasonable surrogate? Comment on this information
including some indication of accessibility and quality. , .
4. For those behaviors that are represented in existing exposure information,
compare the age groups identified for the developmental milestone in question 2
with the age groups in the existing exposure information. Were the age groups
reported in the exposure information based on consideration of child
developmental milestones, are they an artifact of study/survey design and/or
responses, or are they based on the expert judgment of the study investigator?
5. For those behaviors where the age groups reported in the exposure information are
not aligned with the age groups defined by the developmental milestone, what is
the best approach to representing the appropriate age groups in an exposure
assessment? The issue of alignment is compounded when attempting to aggregate
exposure across multiple routes (e.g., dermal, inhalation, and ingestion). For
example, exposure information may be available to characterize children's
C-5
-------
inhalation exposure at a particular stage of development while ,such information
may be lacking to characterize exposure by the dermal and ingestion routes.
Under these circumstances, what is the best approach to characterizing childhood
aggregate exposure?
Anatomical Changes and Physical Growth
During Child Development and Their Impact
on Exposure To Environmental Contaminants
As stated in the background, children's physiology changes over time in ways that can
impact both their exposures to environmental contaminants and their susceptibility to certain
health effects. These physiological changes include anatomical changes resulting from physical
growth. The focus of the present discussion will be on the anatomical changes that relate directly
to commonly used exposure factors information (e.g., body weight, skin surface area, skin
permeability, gut absorption, and inhalation rate). The following questions will help to guide this
discussion.
1. Does it make sense to think about childhood anatomical development as a series
of discrete events which lend themselves to characterization using age group
categories or "bins?" Alternatively, should exposure assessors be thinking in terms
of a continuum of anatomical development that contributes to an exposure
function over all ages? If so, how would one pursue this later approach? When
existing information is not adequate to construct an exposure function that reflects
continuous anatomical development, a consistent, default approach using age
group "bins" may be needed. In such cases, what "bins" serve as a reasonable
surrogate for the continuous function? How would one characterize the
uncertainties that arise from the use of such "bins?"
2. What are the most important developmental milestones for anatomical changes
related to physical growth in children? For each milestone, what is the range of
ages during which the characteristics are typically observed? How much
variability is there among children with respect to the age of onset for the
characteristics? Are the observed characteristics associated with these milestones
likely to affect children's exposure to environmental contaminants? If so, how?
3. For those anatomical characteristics that are likely to have an important impact on
exposure, is there existing exposure information that is representative of the
characteristics? Comment on the existing information including some indication
C-6
-------
of accessibility and quality. If such information is not available, is there exposure
information that could serve as a reasonable surrogate? Comment on this
information including some indication of accessibility and quality.
4. For those characteristics that are represented in existing exposure information,
compare the age groups identified for the developmental milestone in question 2
with the age groups in the existing exposure information. Were the age groups
reported in the exposure information based on consideration of child
developmental milestones, are they an artifact of study/survey design and/or
responses, or are they based on the expert judgment of the study investigator?
5. For those anatomical characteristics where the age groups reported in the exposure
information are not aligned with the age groups defined by the developmental
milestone, what is the best approach to representing the appropriate age groups in
an exposure assessment? The issue of alignment is compounded when attempting
to aggregate exposure across multiple routes (e.g., dermal, inhalation, and
ingestion). For example, exposure information may be available to characterize
children's inhalation exposure at a particular stage of development while such
information may be lacking to characterize exposure by the dermal and ingestion
routes. Under these circumstances, what is the best approach to characterizing
childhood aggregate exposure?
C-7
-------
-------
APPENDIX D
AGENDA
-------
-------
4MEPA
United States
Environmental Protection Agency
Office of Science Policy
Technical Workshop on Issues Associated
with Considering Developmental Changes
in Behavior and Anatomy When Assessing
Exposure to Children
Holiday Inn on The Hill
Washington, DC
July 26-27, 2000
Agenda
Workshop Chair: Kimberly Thompson, Harvard
Center for Risk Analysis
WEDNESDAY, JULY 26, 2000
8:OOAM Registration
8:30AM Welcome & Introductions Jan Connery
Eastern Research Group, Inc.,
Lexington, MA
8:45AM Background William Wood
Risk Assessment Forum (RAF),
U.S. Environmental Protection Agency (U.S. EPA),
Washington, DC
9:1 SAM EPA Perspective on Childhood Exposure Assessment:
Current Practices and Future Needs Michael Firestone
Office of Children's Health Protection,
U.S. EPA,
Washington, DC
10:OOAM BREAK
10:15AM Exposure Assessments for Children, an Overview Elaine Hubal
National Exposure Research Laboratory (NERL),
U.S. EPA,
Research Triangle Park, NC
) Printed on Recycled Paper
(over)
D-3
-------
WEDNESDAY, JULY 26, 2000 (continued)
10:45AM Changes in Children's Exposure as a Function
of Age and the Relevance of Age Definitions
for Exposure and Risk Assessment Kimberly Thompson
Workshop Chair,
Harvard Center for Risk Analysis,
Boston, MA
11:45AM Observer Comments
12:15PM LUNCH
1:15PM Charge to Experts Kimberly Thompson
1:30PM Discussion Sessions
• Behavior-Related Exposure Factors - Katherine Shea, Discussion Leader
• Physiologically-Based Exposure Factors - William Weil, Discussion Leader
4:30PM Discussion Session Reports and Wrap-Up
5:OOPM ADJOURN
THURSDAY, JULY 27, 2000
8:OOAM Discussion Sessions
B Behavior-Related Exposure Factors - Katherine Shea, Discussion Leader
• Physiologically-Based Exposure Factors - William Weil, Discussion Leader
NOON LUNCH
DISCUSSION GROUP PRESENTATIONS
1 :OOPM Behavior-Related Exposure Factors
2:OOPM Physiologically-Based Exposure Factors
S:OOPM Observer Comments
3:30PM BREAK
3:45PM Open Discussion/Next Steps/Wrap-Up
4:30PM ADJOURN
D-4
-------
APPENDIX E
OVERHEADS USED AT WORKSHOP
-------
-------
Michael Firestone
Office of Children's Health Protection
U.S. Environmental Protection Agency
E-3
-------
-------
EPA's Perspective on
Childhood Exposure
Assessment - Current
Practices and Future
Needs
Michael Firestone, Ph.D., Science
Director
Uttif ed States
Environmental
Our Mantra - Children are not little
adults
> They eat and drink more for their size
> They play and act differently than adults
> Their bodies are still developing
> Children may be less able to metabolize
and excrete certain toxic substances
Yes but...
• Newborns are not tiny toddlers
• Infants are not small adolescen
E-5
-------
Administrator Browner's Seven
Step National Agenda to Protect
Children's Health from
Environmental Threats
1. Protective standards
2. Expand research on children's
risks
3. New policies on childhood
exposures
4. Expand Community Right-To-Know
5. Provide basic information to
parents and care givers
6. Expand education efforts 3
7. Pro vide fund ina
GOAL of this
Presentation
I. Discuss EPA's exposure
assessment needs with respect to
considering the impact of
developmental changes
II. Summarize current agency
practices
III. Present some of EPA's ongoing
activities and future needs
IV. Outline next steps
E-6
-------
I. EPA's Exposure Assessment
Needs with respect to Considering
the Impact of Developmental
Changes
• Suggest and define specific
early lifestages which EPA
should consistently utilize
when assessing exposure
• Provide the scientific
rationale for these lifestages
• Identify related reseach
needs
II. Current Approaches Tor
Childhood Exposure
Assessments
• Assessments are conducted
by EPA's program offices (Air
and Radiation; Prevention,
Pesticides & Toxic
Substances; Water; Solid
Waste/Superfund and 10
regions
• Executive Order 13045
http://www.epa.gOV/children/w
hatwe/executiv.htm
E-7
-------
II. Current Approaches for
Childhood Exposure
Assessments
• EPA's Rule Writer's Guide to
Executive Order 13045
http://www.epa.goV/children/w
hatwe/rrguide.pdf
• EPA Exposure Factors
Handbook
http://www.epa.gov/ncea/pdfs/
efh/front.pdf
• Draft Child-Specific Exposure
Factors Handbook ?
II. Current Approaches for Childhood
Exposure Assessments
Age Group Selection:
• Varies somewhat from program to
program and case to case
Often depends on data availability
e.g., Exposure Factors
dbook, Continuing Survey of
d Intake by Individuals, etc.
E-8
-------
II. Current Approaches for
Childhood Exposure
Age
• Often based on professional judgment
about where children spend time and
what activities they engage in
e.g., periods of increased mouthing of
hands and objects
• May consider specific health concerns
e.g., age related differences in iron
•deficiency, cognitive impacts of methyl
mercury due to fetal exposure 9
II. Current Approaches for Childhood
Exposure Assessments
Age Groups that are Sometimes
Addressed in Agency Assessments
Include:
« Fetus
» Infants
• Toddlers
• Children
• Adolescents
10
E-9
-------
II. Current Approaches for Childhood
Exposure Assessments
Examples of How Exposure Data are
Used to Represent the Different Age
Groups:
• median values for 3 year olds used
to represent ages 1 to 6 years
• time weighted averages of data for 1
2 years and 3 to 5 years used to
represent 1 to 5 years
Ilia. EPA's Ongoing Activities
• Child Specific Exposure Factors
Handbook
•New Food and Water Consumption
Data for Children through the Continuing
Survey of Food Intake by
Individuals
• Probabilistic Approaches^ Aggregate
Exposure and CumwLWTfeisk
Assessment
12
E-10
-------
Elaine Hubal
National Exposure Research Laboratory
U.S. Environmental Protection Agency
E-15
-------
-------
Exposure Assessments for Children: an Overview
Elaine A Cohen Hubal
National Exposure Research Laboratory, U.S. EPA,
Research Triangle Park, NC27711
Risk Assessment Forum Workshop
Issues Associated with Selecting Age Groups for Assessing Exposure to
Children
July 26-27, Washington D.C. :
Definition of Human Exposure
The contact at visible external
boundaries of an individual with a
pollutant for a specific duration of time.
E-17
-------
Exposure Assessment
Exposure assessments (half of a risk assessment)
are developed to characterize "real-life" situations
* Identify potentially exposed populations
* Identify potential exposure pathways
* Quantify the magnitude, frequency, and duration of
chemical exposure
Direct Assessment
Measure receptor contact with chemical concentration
in the exposure media over an identified period of time
Personal monitoring techniques are used to directly
measure exposure to an individual during monitored
time intervals (personal air, duplicate diet)
Biomarkers are an indicator of absorbed dose that
resulted from direct exposure.
E-18
-------
Indirect Assessment
To estimate exposure, use
- available information on concentrations of chemicals in
exposure media,
- information about when, where, and how individuals might
contact the exposure media,
- algorithms and a series of exposure factors (i.e., pollutant
transfer, pollutant uptake)
Because of difficulty performing direct exposure
assessments, indirect assessments are often used to
perform the risk assessments required to make regulatory
decisions.
indoor
reside Will
residential
mobile
sources
Indoor
ar
witer
house dust
food
surftces
cbfh.es
E-19
-------
Exposure Pathways
In general terms, a pathway is defined as the course that a
chemical takes from i.ts source to the receptor's portal of
entry.
To specifically evaluate potential for exposure, pathways
are defined here by the exposure medium and the route of
exposure.
The pathway crosses the environmental medium with the
human activity that leads to exposure
Examples:
Indoor air -^ Inhalation
Turf —» Dermal contact
Exposure Factors
Indirect exposure assessments require data on the
following exposure factors:
* Contaminant concentrations in the exposure media in the
environment where the individual spends time
* Contact rates of the i ndividual with the exposure media
* Contaminant transfer efficiency from the contaminated
medium to the portal of entry
* Contaminant uptake rates through portal of entry
* Human activities
E-20
-------
Characteristics of Children that Influence""
Physiological characteristics
Behavioral characteristics
- Development (motor capacity, mouthing)
- Physical Activities
- Diet and eating habits
Other characteristics
- Gender
- Socioeconomic Status
- Race/ethnicity
E-21
-------
Baby Blues
Characteristics of Children that Influence
Fsmosure
Physiological characteristics
Behavioral characteristics
- Development (motor capacity, mouthing)
- Physical Activities
- Diet and eating habits
Other characteristics
- Gender
- Socioeconomic Status
- Racefethnicity
E-22
-------
Exposure Algorithms
* For each route, the algorithm mathematically
expresses exposure as a function of
- chemical concentration in the exposure medium
- contact rate
- rate of transfer from the exposure medium to the portal of
entry
- exposure duration
* Aggregrate assessments include all three exposure
routes: inhalation, dermal contact, and ingestion
* Ingestion can be divided into two subroutes, dietary
and non-dietary ingestion.
Children's Activity Pattern Data
* Microenvironment
The location the child occupies
* Macroactivity
General activities such as watching TV, eating dinner,
taking a shower
* Microactivity
Detailed actions that occur within a general activity, such as
hand-to-surface and hand-to-mouth behavior
E-23
-------
Inhalation Exposure
For each microenvironmentfmacroactivity (me/ma),
inhalation exposure over the 24-hr period is defined as
-inhale me/ma ~
me* '"ma*
Caiune = air concentration measured in the rnicroenvironment
(mg/rrP)
IR^ = child's respiration rate for the macroactivity (rrP/h)
ED = time spent in that me/ma over the 24-hour period
(h/24h)
Exposure over the 24-hr period is the sum of all of the me/ma
exposures.
Inhalation: Data Requirements
* Definition of important me/ma for inhalation exposure
* Air concentration in each rnicroenvironment
* Inhalation rate for each me/ma
Estimated for each macroactivity based on child's age and
weight
* Amount of time child spends in each me/ma over 24-
hr s
Questionnaires designed to collect this data
E-24
-------
Macroactivity Data
* Macroactivity information for an individual contains at least one
complete day of sequential locationtectivity data for each discrete
major behavior. There are 9 studies that recorded such data, but
only 4 include data on children.
* Data from all 9 studies contained in CHAD; a relational database
using a common set of codes for activities, locations, intensity
levels, and questionnaire information.
* Limitations of existing macroactivity data:
- Location information not sufficient to assess dermal exposure
- Activity codes are much too broadly defined and ignore many
child-oriented behaviors
Dermal Exposure - Macroactivity
For each me/ma, dermal exposure over the 24-hour period is
defined as
dermaLme/ma
- ^surface x
Vfer
ED
= transferable surface residue loading
measured in the microenvironment (jig/cm2)
= dermal transfer coefficient for the me/ma
= time spent in the me/ma over a 24-hr period
(h/24h)
E-25
-------
Dermal: Data Requirements
Definition of important me/ma for dermal exposure
Transferable surface loading in each microenvironment
Time child spends in each me/ma over 24-hrs
Questionnaires designed to collect this data
Transfer coefficient for each me/ma
Data need to be generated experimentally
Mac reactivity Data
* Macroactivity information for an individual contains at least one
complete day of sequential location/activity data for each discrete
major behavior. There are 9 studies that recorded such data, but
only 4 include data on children.
* Data from all 9 studies contained in CHAD; a relational database
using a common set of codes for activities, locations, intensity
levels, and questionnaire information.
* Limitations of existing macroactivity data:
- Location information not sufficient to assess dermal exposure
- Activity codes are much too broadly defined and ignore many
child-oriented behaviors
E-26
-------
&EPA
Dermal Exposure - Microactivity
Approach
For each microactivity, dermal exposure over the 24-
hour period is defined as
F — C v TC
*~demial_me/ma "" w surf ace
der
FD
t-l-F
^surface = transferable surface residue loading
measured in the microenvironment
(jig/cm2)
TE = transfer efficiency, fraction
transferred from surface to skin (unitless)
SA = area of surface that is contacted
(crn2/event)
EF = event frequency over a 24-hr period
(events/24h)
Dermal: Data Requirements
Data on important microactivities that lead to contact with
objects/surfaces
Residue loadings for the objects/surfaces contacted
Fraction of residue tra nsferred from surfa ce to skin during
contact event
Surface area of objects/surfaces contacted
Number of contact events over 24-hours
E-27
-------
Microactivity Data
Approaches to gathering data
- Real-time hand recording
- Videotaping
Comparing results among studies is difficult due to
differences in
- Ages of children
- Reported summary statistics
- Categories of body parts and objects contacted
Limitations
- Few data sets, small sample sizes
— Require knowledge on important contact parameters
Non-dietary Ingestion Exposure
For each microactivity resulting in n on-dietary ingestion,
exposure over the 24-hour period is defined as
TEm
SA,
EF
X TExm X SAX X EF
hand or object that is mouthed
contaminant loading on hand or object
dig/cm2)
transfer efficiency, fraction transferred from x to mouth
(unitless)
area of x that is contacted by the mouth (cmP/event)
mouthing event frequency over a 24-hr period (events/24h)
E-28
-------
ssg
Non-dietary Ingestion Data Requirements
Information required to asses non-dietary
exposure from surface-to-mouth activities
* Data on important microactivities that lead to
object/surf ace-to-mouth ingestion
* Residue loadings for the objects/surfaces mouthed
7
* Fraction of residue transferred from surface to mouth
during mouthing event
* Surface area of objects/surfaces contacted by mouth
* Number of mouthing events over 24-hours
Non-dietary Ingestion Data Requirements
Information required to asses non-dietary exposure from
hand-to-mouth activities
* Data on important microactivities that lead to hand-to-mouth
ingestion
» Residue loadings on the hands
+ Fraction of residue transferred from hand to mouth during
mouthing event
* Surface area of hand contacted by mouth
* Number of mouthing events for each me/ma over 24-hours
E-29
-------
Dietary Ingestion
Exposure is estimated by summing contributions from:
- Chemical residue on the food prior to handling in the
residence
- Pesticide transferred to the food during contact with
contaminated surfaces ^'
- Pesticide transferred from surface to hand to food during
handling and eating
y, • •«
Algorithms and data requirements similar to those for
non-dietary ingestion with addition of information on:
- Concentrations of contaminant in foods coming into house
- How food is handled
Exposure Scenarios
* For any given pathway there are a set of
associated exposure scenarios
• Exposure scenarios combine
— Source (application method, residential use of a consumer
product)
- Population (age group, geographic location, SES)
- Timeframe (acute, short term, chronic)
— Microenvironment (indoors and outdoors at home, indoors
and outdoors at daycare/school, indoor and outdoor other, in transit)
— Macroactivity (active play, quiet play, sleeping, eating)
E-30
-------
Exposure Pathways vs Exposure
Systematically identify potential exposure
pathways to frame exposure assessments
Identify exposure scenarios to specify values of
exposure factors and to estimate distribution of
exposure by any given pathway
To identify exposure scenarios, need identify
appropriate age/developmental benchmarks for
categorizing children
Research Needs
To improve the database available to assess children's
exposures, three areas of research are required.
- Identification of appropriate age/developmental benchmarks
for categorizing children in exposure assessments
- Development and improvement of methods for monitoring
children's exposures and activities
- Collection of physical activity data for children (especially
young children) required to assess exposure by all routes
E-31
-------
Summary
«• Activity patterns provide information about when, where,
and how individuals might contact exposure media.
* Contact rates, transfer efficiencies, and uptake rates are all
a function of activity patterns:
* To guide field studies and select scenarios for exposure
assessment, it is critical to develop relevant
age/developmental based milestones for children
E-32
-------
Kimberly Thompson
Harvard Center for Risk Analysis
Note: the charts and illustrations shown on pages
E-36, E-37, E-40 through E-45, E-48, E-49, andE-50
also appear in Appendix H of this document.
E-33
-------
-------
Changes in Children's Exposure
as a Function of Age and
Relevance of Age Definitions for
Exposure and Risk Assessment
EPA Technical Workshop
Kimberly M. Thompson, Sc.D.
Harvard School of Public Health
® 2000 Knribety M. Tlwnpam
Key Issues
Childhood - distinct phase of life
Growth - transition from birth to adulthood
(physical, social, behavioral, psychological)
Some things common to all (e.g., teething)
Some specific to children with certain
characteristics (e.g., kids with fair skin)
Some specific to child-activity patters (e.g., kids
that swim)
© 2000 TSnixa^W. Ttampson
E-35
-------
Physical changes (birth - 3 years)
•
Boys 2
~ '>'—
T.
?:.
"
'
J l> r K-.
L— JL— ™
|](3{ttl4:l
^
-
>
f
fj
i//
'/,-•:
f
-•
-. 2
jS
fj/^^
y
j
M
7S
/ /
f
• £ i!
;
X
/
/
,*
;
5 31 2-S :
s$
r^
f
'jS"
\r
^
^X"
s*
^
7 :
*•
f
t
S
f
**
%-
) i
i u :
I
«J
Girls "
/ r
o 31
- Ii it-
£5 '*'
"
• xl
f
u
• u
. K:
^"'
±
: « ;
'
^
if
X
/
*t
t
3 tip
y
/
/
/
i
^
^
/
/
r
i
*
s
s
$
/
f
. IS 13 i
1
j^
/x"
>
fjf
'/
f
*~
- * * :
12
>
^
r^*
^
(**"
^
n<
4 27 :j
^
-""'
X
X
x"^
^
>
^ ^
i_
!*• '
;X
1 *
Jft€ *
•''^ ^ .
VT 5
:!
/ *
i mi
U •«:
© 2000 Kinibedj-M, Thompson
Physical changes (3-18 years)
j, s r a a m 11 i? n a ii in IT u,
i * 4 r. r a a m 11 i> ii t4 ii ir. if
«-
«
e?
»-
« -
X'
X
11.
7
J
/
Boys
g»
•KV.-
i^.
rf«^-
ia
113
3>
- ?J
4 S e > t « O 1! Gt 13 w IS I: 1> 1*
S «•
^
sjfc
T3FW
•20}
• 1X1
• in
• m
iw
- ix
• m
• s>
• 4>
3}
2 4 > « 7 t; ^ ii u ii 13 14 iQ |j i; •
E-36
-------
Physical changes
ran?
£m&(fetoi)
Newborn £yr. 6yr.
Z5yr.
© 2000 KnribedferftL Thompson
Developmental milestones
(birth - 6 years)
Charts available
Categories (personal-social, fine motor-
adaptive, language, gross motor)
May have different significance for
physicians and exposure/risk assessors
Continuous changes, growth spurts,
measurements at discrete time points
Qualitative and quantitative differences
© 2000 KMde%M. Thomson
E-37
-------
Working with what we have
• Existing exposure data
• use a wide array of age categories
•may not be representative of the children of
interest or concern
• As a result
• analysts must pull together data from different
databases to create modeled children, but avoid
"hypothetical" children that could not exist
• significant data gaps may exist
® 2000 KinJjedj-M. Thomson
Exposure Equations
• 7 Equations given, based on Hubal et aL (2000)
* Routes: Ingestion (dietary, non-dietary),
inhalation, and dermal
• Plus equation to go from exposure to dose
(Pharmacokinetics and pharmacodynamics are
not covered in this meeting)
® 2000 Knriberij-M. Thompson
E-38
-------
Objectives
Characterize
• the availability of data for use in exposure equations
and the age categories used
• the extent to which the data are accessible and the age
categories could be modified
• the current quantification of variability among
children of similar ages and uncertainty in the data
Use
• primarily the EPA's Child-Specific Exposure Factors
Handbook (CSEFH)
• a few other studies from the peer-reviewed literature
© 2000 Knribedtr M. Thomson
Tables
Summarize
• exposure factor
• original source
• age category used, and when available the numb er
of subjects in each
• general assessment of (1) data quality based on
the criteria and judgments (EPA's CSEFH)
• extent of generalization (as judged by author)
© 2000Khribe4rM. Tfampson
E-39
-------
Figures
• Summarize
exposure factor (Jinked to the table) for 3 different
age ranges (a=birth to 1 month by days, b=birth
to 3 years by month., and c=birth to 21 years by
year)
an "x" under a specific age indicates that the
study reported measurements for children at that
age
a bar between endpoints of a range indicates that
the study reported measurements b etween that
range
© 2000 Kimbeifc-M. Tfanpson
. -ftm', u » > *
Body weight
-i K ii is -a •* -a t i; is. \-j
of Araife.b v
',yfi\ 3d'.:i by &ays
na*r x x
Full it 4b Suiriiria-v v! A-/ai!ablt? Suds' VVtiii;!:. I'D'.Vj Dali by Murltii
n'.vrn xxxxxxxxxxxxxxxxxl-
ir.vis-i X A x x
f 13jre 4« Sannary st Available Bocy WclgSI (H\Vi licla fey
® 2000 KbvibeifrM. Thoinpson
E-40
-------
Skin surface area
«:• ;DlviC> •_ 1 i S -4 .-> = v j i> 1-1 -1 1S •* W '6. 1i •• IS •-• i-J il 0< ii U ii. S- J'' a- "
Ski-:
SA.iy.Vi-:
Fstra Si 5jrm£.rvot Avai at-li SL'ace .J.fi£ SA: Da'.a.tvMor:li
-irnrft .5.1 Sa^mary af Av.nsn^lr Surncc; Ar;Q ;$*; I>n1r by Y-io
® 2000 KinAeifyM. Tlionpson
Food intake
\Dl\tii i 1 i S 4 i i -' -. & M -1 12 •:> 1J -S. 1i •'. 18 -J i>l il 22 JS i^ is £i iV aj is s;
fx-*<>t Av
Eft Sumroa-y r4 Avatl^hl^ Ir.qetMr-n
E-41
-------
Hz. :
Water consumption
• ; s 4 s ft .• s •; i. -i i< •& ii is -ft if •!• « a: n sx in- ;•= 45 m Sf a a s.
?c. sjrr i>a y d Avai
; ni o' '.'.'a:*: Daii by Daya
F"gvKr7t Sinnray of Ass! at'.- lrj»'Jci Ri'1t{IS)&rft'ai'-'t 3alt-1> Months
7; = jinriia-f-1-' A><:il«ble lujcsirtyi fials :!RJ c' V^aie- Data ay Yeai
® 2000 KmJbedyM. Thnnpstin
Breast milk ingestion
• S : *• j t } S I Ti II l< 13 M 1i I: •» 7i i» i: « 2i iS !« JS te S? CS 2S St
, n ._* i:>
. •.-.— ! I.
Sum-ia-j IB Available nsBs:icT Rslc CIR'i ;f 8-*si1 'Jilk Caia by Ds.'3
FICMC Kb Sunircrjf af AM abi: Irjcrfor Para flR; ot Breast U; k 3s1s by Warrthi
s Sunn-cry D' Avf.i abis ingssdar Ra:e (13; ot Breas: viiK Date by Y&y*
(£> ZUUU HdntteOyS/L llvnnpson
E-42
-------
Fish consumption
•V* iT'vvI ;ti5-ssT^r'"ii'ys i-,is it IT -s i; jw j ci ^f ;« is a* B: » .w «
- * in-
nsiT .•" Agitable [- ^s.o'i Rvfe : P.I yl F =lt 3:ris jr i-ji-i DaU; bj
ir^ iifc fiuflimT *> cj AvnllnMn - "•: w rrt Rftt-: II R; r* Rsf Ga t*s. tp* oft Dns: ay Vo
t i 5 * •
= .11.-
.r^fi.
i - « in
>•*«:•
JH—.*
=,-;jrc s: Gunmur,-y .".•oilablg ->g»Kbi Rate I R;
© 2000 Kin*ed(j-M. Thmpson
Soil ingestion
C 0 T 3 3 10 | |J -3 -i IE 16 IT « |» 4; 21
2: s= 2; «c-
i t. il!
«. is;
•=-. •-;
F pic*? "On Sjirtrary ol *..T Inbl ^ ; t(je-slici ?{a-n ; fi) ol Snil IrgnnTinr Kiln ty Ut-ji
;T. n: I
i i. 11;
«. .<;
! 1. "I;
A <•
iQurn in*: Sumnary 3l 4'^:i!ab c; ngcMicn 3nr3 110) nf Snil lr:jn,?rrRr £2!.? by Yearn
© 2000 Khribeity-M. Thrnvs
E-43
-------
Other non-dietary ingestion
i s . i £
is •< •» 1* I;
Tgu'c " 1a Surrna-y c" Sva- stlc O'.lic" Ko:i-D s;ary Ir-gua.iar, Da;a 3tf I'a/s
•"•:c:;vi:rh:: -i t • i * - T t z i- ; % i r. • c. 1* '* *> :n ; » M fi ss.1-»? »s <•= s- si >s .I
I-
Sun-ns-y «• Av«. afcl* O:h6'
Olhnr
ll.v"
R . jSi
Inhalation
a s * » s 5 * » « -i -t i* w i£ ie i; 1= -i so £•
gure *2a QLnmzfx' c- A/£:!ab'± !n*>a!diofi Rc1c Data by Days
R.-Jn
.'5 cf Awt lab* Inhatel
•"-•:••
!«.: i
in^.yi
i11-- '
© 2000 KinJbedbrKL Thtm^pson.
E-44
-------
Dermal contact
V'.l- i
C3L.-I
Fty f
-niM-y nl AJKIX|:|M
j ^ - 8 - s : 1:111: i: H is : r s Jj>iisi>^^:<:?:!;= 2;3
=;ri.r? 13s S'.nn-r.-v y -V.T.ljb : zcrna Ccnbc: Dsa 6? M3"-»
: c ? c s ii s 2 4 ;: i: s?
ca-1
C:L. i
© 2000 Kimbei^M. Thmnps
Time/activity patterns
T-Mrl?l
T /.. i«t
F Qiirs lia 5nmi(i:'« nt ficcl
v':": by Mcntffl
z 5 i r
E-45
-------
Synthesis and observations
Lack of representative/available data for the individual,
population, temporal and/or spatial scale of interest
Know relatively more about easily observable anatomical
exposure factors (weight) than behavioral not-easfly
observable factors (non-dietary consumption)
© 3000 KinfcedyM. Thomson
Exposure factors reviewed and
equations that use them
Factor Quality
BW H
SA H
SAffiW
IRWBW H/L
IRwt* H
rRi~~»~n» M
IRai. H/L
M&2m£& H/L
IR*«a M
TR~. M
TKT. T.
T«.r™ M
Extent of "X" Denotes Used inEquation Number
generalization 123
H
M
M
H
L
M
M
L
M/L
M/T.
M X
M XX
4567
X
X
X
X
X
X
X
HNKgh, M=mediuni, L=low, H/L=high for arerageflow for long-term and upper-
perc entiles
MSL- mediiimforaveraffeflowforbne-teim anduwier-TKrcentiles
E-46
-------
]
Exposure factors in equations that
did not occur in the review
Factor
EFdorm»l
TEdemal
O-n^ownal
WT
TEs/r
EFs/F
5-As/F
TErnr
EFn/F
SAH/F
"X" Denotes Use din Ea
1
234
X
X
X
X
X
X
X
X
X
X
X
uation Number
5
6 7
© 2000 KinfceiferM. Ttanpson
•• — .
Synthesis and observations
Lack of representative/available data for the individual,
population, temporal and/or spatial scale of interest
Know relatively more about easily observable anatomical
exposure factors (weight) than behavioral not-easily
observable factors (non-dietary consumption)
THxnpson
E-47
-------
Data available from birth to 30 days
E-I: (Days'i P • Z 3 4 f S 76 j> 10 11 '2 "13 14 15 13 i~ 1? 1S- '3 Z\ V. ?3 S4 ?.5 K 87 ?f S3 50
HW X
'"UVMT-IH XXXXXXXXXXXX
Rgj.-R 15. nets Availebla fn' ARHB 0 1a 30 Hfiyg for Expnaure Fantors Usuri n EqiiHtinia 1. ?, 5, 5. sn.n 5
X —
© 2000 KirAedyM. Xlmnpson
Data available from birth to 3 years
A£X IMcnllo: C12S4E87SS 'Of 1Z 13 U :5 16 17 IS "3 20 21 2Z 23 St E52S272S23 SO 31 32 3254 35 S3
IBtrsacteOt XXXXXXXXXXXX
Figure 16. Data Available lor Ages D to 36 Months for Exposure Faetnrg Used in Equations 1, 8. 3, 5 and 6
£t specific £3?
© 3000 KinJie^M. Thompson.
E-48
-------
Data available from birth to 21 years
- s s : « • a •:• 11 ia is •< -s is 1- 1; 10 20 21 a: j; M
X XX X X X > XX.)! X X XX X K X (-
III,,,.
ir* "7. Data Avai-shk foi Ages £ :i> 84 v«ars f
-------
Time spent at home according to NHAPS
data (Source: The LifeLine™ Project)
B g g H 8
0p p a D
8 B 5 6
Em (Mlnuto cat Hums)
o P n H B 5
Bfsii^yEio*
© 2000 KinAedj-M. Thompson
Children
© 2000 Kiimfcedy M. IKxnnpson
E-50
-------
Discussion issues
What is the ideal approach to preparing childhood
exposure assessments that reflect changes in children's
behavior and anatomy over time?
Is the existing exposure information adequate to
implement the ideal approach, if not, what additional
information is needed?
What short term studies could be conducted to supply
the necessary information or provide additional
guidance?
What longer term research may be needed to achieve the
ideal approach to preparing childhood exposure
assessments?
© 2000 KinfcedyM. TJboompson
Behavioral question 1
Does it make sense to think about childhood behavioral
development as a series of discrete events which lend
themselves to characterization using age group "bins?"
Alternatively., should exposure assessors be thinking in
terms of a continuum of behavioral development that
contributes to an exposure function over all ages? If so,
how would one pursue this later approach? When
existing information is not adequate to construct an
exposure function that reflects continuous behavioral
development, a consistent, default approach using age
group "bins" may be needed. In such cases, what "bins"
serve as a reasonable surrogate for the continuous
function? How would one characterize the uncertainties
that arise from the use of such "bins?"
© 2000 KnribeityM. Thomson
E-51
-------
Behavioral question 2
What are the most important developmental milestones
in children's behavior? For each milestone, what is the
range of ages during which the behaviors are typically
observed? How much variability is there among
children with respect to the age of onset and the age of
abandonment (if applicable) for these behaviors? Are
the observed changes in behavior associated with these
milestones likely to affect children's exposure to
environmental contaminants? If so, how?
© 2000 Kimfeedj-M.
Behavioral question 3
For those behaviors that are likely to have an important
impact on exposure, is there existing exposure
information that is representative of the behavior?
Comment on the existing information including some
indication of accessibility and quality. If such
information is not available, is there exposure
information that could serve as a reasonable surrogate?
Comment on this information including some indication
of accessibility and quality.
© 2000 KmfcedyM. Tlnnjps
E-52
-------
Behavioral question 4
For those behaviors that are represented in
existing exposure information, compare the age
groups identified for the developmental milestone
in question 2 with the age groups in the existing
exposure information. Were the age groups
reported in the exposure information based on
consideration of child developmental milestones,
are they an artifact of study/survey design and/or
responses, or are they based on the expert
judgment of the study investigator?
© 2000 Kbnbedy M. Thompson
Behavioral question 5
For those behaviors where the age groups reported in the
exposure information are not aligned with the age groups
defined by the developmental milestone, what is the best
approach to representing the appropriate age groups in an
exposure assessment? The issue of alignment is
compounded when attempting to aggregate exposure across
multiple routes (e.g., dermal, inhalation, and ingest ion).
For example, exposure information may be available to
characterize children's inhalation exposure at a particular
stage of development while such information may be
lacking to characterize exposure by the dermal and
ingestion routes. Under these circumstances, what is the
best approach to characterizing childhood aggregate
exposure?
E-53
-------
Anatomical question 1
Does it make sense to think about childhood anatomical
development as a series of discrete events which lend
themselves to characterization using age group "bins?"
Alternatively, should exposure assessors be thinking in
terms of a continuum of anatomical development that
contributes to an exposure function over all ages? If so,
how would one pursue this later approach? When
existing information is not adequate to construct an
exposure function that reflects continuous anatomical
development, a consistent, default approach using age
group "bins" may be needed. In such cases, what "bins"
serve as a reasonable surrogate for the continuous
function? How would one characterize the uncertainties
that arise from the use of such "bins?"
® 2000 KMbedyM. Thompson
Anatomical question 2
What are the most important developmental milestones
for anatomical changes related to physical growth in
children? For each milestone, what is the range of ages
during which the characteristics are typically observed?
How much variability is there among children with
respect to the age of onset for the characteristics ? Are
the observed characteristics associated with these
milestones likely to affect children's exposure to
environmental contaminants? If so, how?
© 2000 Km*edyM. Ttampson
E-54
-------
Anatomical question 3
For those anatomical characteristics that are likely to
have an important impact on exposure, is there existing
exposure information that is representative of the
characteristics? Comment on the existing information
including some indication of accessibility and quality. If
such information is not available, is there exposure
information that could serve as a reasonable surrogate?
Comment on this information in eluding some indication
of accessibility and quality.
© 2000 Kmjbedtr M. Thompson
Anatomical question 4
For those characteristics that are represented in
existing exposure information, compare the age
groups identified for the developmental milestone
in question 2 with the age groups in the existing
exposure information. Were the age groups
reported in the exposure information based on
consideration of child developmental milestones,
are they an artifact of study/survey design and/or
responses, or are they based on the expert
judgment of the study investigator?
© 2000 KfaritafrHL Ttan^son
E-55
-------
Anatomical question 5
For those behaviors where the age groups reported in the
exposure information are not aligned with the age groups
defined by the developmental milestone, what is the best
approach to representing the appropriate age groups in an
exposure assessment? The issue of alignment is
compounded when attempting to aggregate exposure
across multiple routes (e.g., dermal, inhalation, and
ingestion). For example, exposure information may be
available to characterize children's inhalation exposure at
a particular stage of development while such information
may be lacking to characterize exposure by the dermal and
ingestion routes. Under these circumstances, what is the
best approach to characterizing childhood aggregate
exposure?
© 2000 KxnfceifrM. Tfampson
E-56
-------
James Walker
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
E-57
-------
-------
i.
s 0>
PM
CS
^ ^™^
^
"«
u
.= u
a, *c
s -s
u
^ « "^3
o S ,-S
O CL 2
— u
? s
£ •-
Q
H
05
^^
Q
<
c
2
W
c 13
S
S o
o ^>
•-P3 W
O -rj
^ e
O °i-
5 1
2 I
G 05
O
O
O
^ Cis
bD O g
-jH ^—H £2
o
^^^j ^^^j
^^^^ s ^^^^
^^ ^^^^«
s g
•> u
w ^
5-H
a
|
6
S Dd w
g f-^ , ,
^~( M^^ prt
a3
5=1
O
1
03
SH
Q
05
05
O
05
05
E-59
-------
uc
ro
o
•§
s g
x -g
«£« ,«J CD
O 5
ti 'fcl *9
05 _£ 05
TS
T3
CD
N J> ^
•TH >N *—«
OS ^
CD
la
d,
^
O
d cT
«> -S
8 o ^
^ too too
8 ^ 2
^=! ^ CD
^.i d S^
J±H S 8
O r"! 'K-
00 O ^
E-60
-------
CD
a
CO
1
Q o
fe
o
^ w>
*g
co
CD
0
CD
11
£ s
v*H *^^
3 13
J3 O
^•£
• t-H
0,
O
o
oo
e
-S3 ^
o ^ too
^ 52 T!
•^ CD
F^ ^
* *
G T—4 "rl
jg S
T3 «J
Q
i s.g
*1 'c
'S ? S
O «i
^ *-i CD
>> feO'TS
CD
OJ
CD
6
.a
T3
I
.&
.3
a,
o
-S -
^ 3
/-N CO
CO <-!
CD ^
*3 ?
S 2
& d
O *-g
•Si
c3
CO
JJ e!
•* O
GO
d>
S
«
CD
csi :=!
4=1 0> /CD
° 7^ S
3 s ^
-| -si
ni ? C^
cL P^ j ^
'« b -S
•ff-.b
CD
^H
H
Bl
j»»rv ^rH
,
E-61
-------
•8
§
o.
•8
CD
CD a
• TH ^3
G C5
> Q
rr, CD
|
£< CD
§1
a g
Si *£H
CD
> 5-<
J? CD
-2 §-
TH >
•S-8
S3 CD
CD ^
too &
a n
s
CD -^
WM
ge
C5 ^ cj
^ rt fe
H O 55
e
S
•£P
CD
45
I
CD
§
CD
05
1
S .
&"3
O r£
CD
>
•8-s
^
CD
g O
TH '•"
§•
cc 03
CD rj
toO-S
c3 «»<
CD ^
I i
H too
E-62
-------
•a s
&
I g
go
•Sk-1 S
S*r
•^^
«H -»-l
C 4=!
^ o
Switz
I
•N
o
a>
^ S
S i!
® Ti
Tl rt'
en
1
4=1
«2
00
H CO
E-63
-------
a
o
0
o
PH
£• a
I'o
&••£
•S o-
li* CD
in
CD
<^H
O -g
*s §
OS O
=H=
.•8.
;=! CD
CO *d
r. (D
Ji Tg ^
'&-« S
S3 * ^ ft
& *T"4 *"55
K-o .g
CD ^ g
^**l • ^H !~^
§ .-g N
co Q p
T3
•s
£•8
155
B !>'g 8
& g RqH
,v d
CD 05
S 0°
CD ^
§ H a
1
1
CD
Of}
O
:^
CD
oo
CD
OO
0
05
1
s a
&o *
05 jsg
Q CD
CD ^J
CD
E-64
-------
;=3 o>
CD
§
1
50
at other hum
8 15
a ^
CD
O
•1 6b 5-e
ss _2 05
• '•H I«~H y^
•J3
ni
1
S3 O
-8 *
•^ ^3
13 Q
T3 S
o ii
r-i w
5--a
J3 §
'S go
^ o
^ CD
O CD
CS
I
OJ
.sill
ss T3 *-> e5
s* .co bo S
E-65
-------
GO
us
Conc
^ Tfr
*d ;g
S co
3 -a s CD
E-66
-------
03
_0
"cS" ^
^
H <•-'
•••-** nj^J
P •£ 55
«g
<« -a •-
-------
&
CO
CD
CN
O
CO
CO
CN
O
*
CD
in
8
O
CN
S ' 8
E-68
-------
JJ2 >> co
f—H ^j ^
O
oo T3
CO
-------
S
O
o
CO !
s o
f1
.1
• rs
O
g
•a
ffi
o
C3
CO
8
o
.V
•
-f.
CD
CO
CN
O
00
CO
- CN
CD
4P
E-70
-------
MO
-------
tfl
^ 9£ ^
T3 r: ^>
O H %
^^X ^^ Pr*-
0 ^ ^
|*V| ^ r"| *i~H
*-3 1> co
C3 f> I
I § *
1> Q CD
§, toO^
J^-aa.
^ so
^5*. f^ ^^
^ ^ ^3
o a £
,-H 5 fe
^ tH O
2 ^ oo-.
M=* T3 CM '
*m s o
°l
^M
ra
o>
u
1
u.
bi
s
Q
^
W
ll
g 5
£
Q
2
&2
II
•a .§
••al
•a s
O PL,
jncjgg 0^
DO C^ CD C^ r^ ^"
o o o o o o
•r> o> — • >o co t-
fl en oo oo CD
"<-. CN CN ^ CN C3
Q 00 CN 0 o ^.
o o c-a »o — — •
CN M OO *O O> ^
CO CN OO — O O>
^^ CN *O »-* r^ VD
-< <>J "O 00 CJ Ov
0 O O O ~-> O
o o>
-------
o
N.
-a
•fH
w
<+H. !>-
o %
S ^
^-t
GO
° i
^- G
S P
ii
^H • ^H
-=!
^ O
^
~ l l
Q
-I
(015)
in
ano«Nanw*N
E-73
-------
o
SI
o
O
VI
-------
CD
ft O
o
O
00
0<§
5 .i=H
-a .a
-
Q PH
O
N
. f
o
z
\
o
co
o
in
r
o
o
ro'
(f. / UJ 0) f.\\ 3 0| 8 /\
_ o
CN
_ CD
_ CO
0)
E-75
-------
1
W
l
tx)
o a.
M v^ CN 1-> C«N o»
<£> fS tx f v% -< «
OO —•
il
E-76
-------
.3
50
s
I
8
-3
•K
-8
^
I 1 T-
CD
§
lt> ^
°. «.
O 0
o a
. E-77
-------
*H
£
.5
J
ffi
^
0
b
a
•i-H
r
tt)
U7
0
r>
d
E-78
-------
i
-------
tn
ttf
V4
O
«w ,£3
0) -H
•^) -f^J CQ
JP 0> g-H
1-1 >-« 0>
<8) ^
J> 18
A, rH CJ
'C!^ 09
0) 5-1
S
•8 4
,---**
W U flj 0
CD CJ v^4
•4-4 O EJ ^
iW O O<
-H ?T.
0} JJ
m
fi
C0
E-80
-------
I
TJ
.5!
£ ^
&
-S
3
OQ
*s
"O
o
§
0
S
-------
-------
APPENDIX F
BEHAVIOR GROUP OVERHEADS
-------
-------
i
Q.
_0
1
•o
2
I
,8
"to
S
at
I
a.
a.
(0
I
c
8
re
Q.
C
_g
4*
JS
"5
f^
£
1
O
o
£
*£Z
Ul
*J
O)
*
Q.
O
I
s
>
O i"
i
in
m
CO
o
CD
Q.
>."§
CO 0
CD O
food expands
increased floor contact/mobility
contact nonfood items
CO
CD
CO
Fine motor
Increased mobility
Creeps/crawls/crui
in
O
T-
£ ••-
all foods
increased play/exploration of non
items
cessation of breast and bottle fee
T3
O
O
x:
"O >,
1 1
CO 3
111 <
CO
Extreme mobility
Walks, runs, climb
Poor judgment
Curiosity
S
«o
CM
CN
|
3
O
CD i~
CO 3
CO CO
§ 1
.£ CD
£
out of diapers/into adult clothing
increasing adult patterns hand to
j_
CO,
T3
I!
Ic '*£
O £
"co
1
CD
1
Q.
CO
CD
More outdoor gam
CO
1
in
if
O "O
w ^
To
£
.~
2
CD
Q.
O
CD
CD
O
O
O
2
CO
5.
o
CD
C
S
i--o
i!
^ "o
O CO
r"* 03
.2 ^
.1 j=
•- .2
CD
o
c
8 ^
CD C
T5 ®
< £
To
c
o
'ffl
CD
Q.
o
To
o
£
*—
Rapid growth
Increased risk takii
i
in
£1
o
S.
CO
8
1
•5
CO
.CD
.5?
Q 5
2
CO
5.
S
I
F-3
-------
Routes and pathways of exposure that are significant for age groups
Routes
11-15 years
16-20 years
ADULT
In this figure, shading has been used to indicate a qualitative tailing off of frequencies in distribution.
F-4
-------
APPENDIX G
CHILDREN'S EXPOSURE ASSESSMENT (PAPER DISTRIBUTED AT
WORKSHOP BY ELAINE HUBAL)
Note: this paper is published in
Environmental Health Perspectives, volume 8, number 6 (June 2000), page 475.
-------
-------
CHILDREN'S EXPOSURE ASSESSMENT: A REVIEW OF FACTORS INFLUENCING
CHILDREN'S EXPOSURE, AND THE DATA AVAILABLE TO CHARACTERIZE AND
ASSESS THAT EXPOSURE
Elaine A. Cohen Hubal*
Linda S. Sheldon
Janet M. Burke
Thomas R. McCurdy ...
Maurice R. Berry
Marc L. Rigas,
Valerie G. Zartarian
National Exposure Research Laboratory, U.S. EPA, RTF, NC
Natalie C.G. Freeman
•" ' ' •"'•,• ' '* '-','"
Environmental and Occupational Health Sciences Institute (EOHSIX Rutgers University,
Piscataway, NJ
* Author to whom correspondence should be addressed
U.S. Environmental Protection Agency
National Exposure Research Laboratory
HEAB (MD-56)
Research Triangle Park, NC 27711
Telephone: (919) 541-4077
FAX: (919) 541-0905
Internet: hubal.elaine@epamail.epa.gov
Send express mail to:
U.S. Environmental Protection Agency
Annex Building
79 TW Alexander Drive
Durham, NC 27709
G-3
-------
Running Title: CHILDREN'S EXPOSURE ASSESSMENT
Key Words:
Children
Environmental Exposure
Exposure Assessment
Activity Patterns
Susceptible Populations
Aggregate Exposure
EPA Disclamer:
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's
peer and administrative review policies and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
G-4
-------
Abstract:
This paper reviews the factors influencing exposure of children to environmental contaminants
and the data available to characterize and assess that exposure. Children's activity pattern data
requirements are demonstrated in the context of the algorithms used to estimate exposure by
inhalation, dermal contact, and ingestion. Currently, data on children's exposures and activities
are insufficient to adequately assess multimedia exposures to environmental contaminants. As a
result, regulators use a series of default assumptions and exposure factors when conducting
exposure assessments. Data to reduce uncertainty in the assumptions and exposure estimates are
needed to ensure chemicals are regulated appropriately to protect children's health. To improve
the database, advancement in the following general areas of research is required:
• Identification of appropriate age/developmental benchmarks for categorizing children in
exposure assessment.
• Development and improvement of methods for monitoring children's exposures and activities.
• Collection of activity pattern data for children (especially young children) required to assess
exposure by all routes.
• Collection of data on concentrations of environmental contaminants, biomarkers, and transfer
coefficients that can be used as inputs to aggregate exposure models.
G-5
-------
I. Introduction
Children's exposures to environmental contaminants are expected to be different and, in many
cases, much higher than adults (1-7). Differences in exposure are due in part to differences in
physiological function and surface-to-volume ratio. However, differences in the behavior of
children, particularly the way in which children interact with their environment, may also have a
profound effect on the magnitude of exposures to contaminants.
The U.S. Environmental Protection Agency has pledged to increase its efforts to provide a safe
and healthy environment for children by ensuring that all EPA regulations, standards, policies,
and risk assessments take into account special childhood vulnerabilities to environmental
contaminants. The Food Quality Protection Act of 1996 (FQPA) requires that exposure
assessments be used in the pesticide tolerance setting process. Exposure assessments for FQPA
must consider the potential susceptibility of infants and children to pesticide exposures from all
sources including those from food, water, dust, soil, and air. To meet these regulatory
requirements, existing information on children's exposure to environmental contaminants needs
to be used to develop and improve exposure assessment methods and models for children. In
addition, research on exposure that will answer questions about age-related differences and will
lead to better exposure assessments for children needs to be designed and conducted.
This paper reviews the factors influencing exposure of children and the data available to
characterize and assess exposure, with a focus on children's activity patterns. Activity pattern
data requirements are demonstrated in the context of algorithms used to estimate exposure by
inhalation, dermal contact, and ingestion. Finally, data gaps and areas for future research to
improve exposure assessment for children are identified.
II. General Principles for Studying Exposure of Children
Exposure is defined as the contact (at visible external boundaries) of an individual with a
pollutant for specific durations of time. Exposure assessments are developed to characterize
"real-life" situations, whereby: a) potentially exposed populations are identified, b) potential
pathways of exposure are identified, and c) the magnitude, frequency, duration and time-pattern
of contact with a chemical (potential doses) are quantified. Exposure assessments are conducted
using either a direct or an indirect approach. A direct assessment measures a person's contact
with a chemical concentration in a media over an identified period of time using personal
monitoring techniques. Due to high study costs, direct exposure assessments are not often
conducted and few methods exist for making them. For a few environmental contaminants,
biomarkers can serve as a useful measure of direct exposure aggregated over time for all sources
and pathways. However, few studies using biomarkers have collected all of the information
required to accurately estimate exposure. An indirect assessment uses available information on
concentrations of chemicals in the various media, along with information about when, where, and
how individuals might contact the chemical. The indirect approach uses models and a series of
exposure factors (e.g., pollutant transfer, pollutant uptake) to estimate exposure. The specific
information and factors needed to conduct an indirect assessment for a given contaminant
depends on the significant routes and pathways for exposure to that contaminant.
Because of difficulties associated with performing direct exposure assessments, indirect exposure
assessments are used typically to perform formal risk assessments needed to make regulatory
decisions. Indirect exposure assessments require data on the following exposure factors:
G-6
-------
• Contaminant concentrations in the exposure media in the environment where the individual
spends time,
• Contact rates of the individual with the exposure media,
• Contaminant transfer efficiency from the contaminated medium to the portal of entry,
• Contaminant uptake rates,
• Activity patterns.
It is difficult to develop and verify exposure factors such as contaminant uptake rates and transfer
rates for young children. Children cannot intentionally be exposed to contaminants; thus,
controlled laboratory studies with children cannot be conducted. Using adult surrogates for these
studies introduces bias, because adults do not behave like young children and therefore cannot
mimic their contact activities. It also is difficult to collect personal air, blood, urine, and
duplicate-diet samples from a child. In addition, it is difficult to accurately record a child's
activity patterns. Direct observation (which may include videotaping) is considered the most
accurate way to record a child's activities, especially as they relate to dermal absorption and
ingestion. However, this methodology is very labor intensive and costly. Finally, children
engage in a wider range of contact activities than adults, so a much wider distribution of
activities must be considered. Developing realistic estimates of children's exposures to
environmental contaminants requires the understanding and quantification of children's activity
patterns.
It is important to understand that physiological characteristics and behavioral patterns will result
not only in different exposures for children and adults, but also for children of different
developmental stages. Thus, exposure assessments are required for children in each age group,
with age group being defined by developmental stage. Classification of children by age group
should be based on estimates of when developmental changes commonly occur. For example,
walking typically develops between 12 and 14 months. However, there are children who are
early walkers (8-11 months) and late walkers (after 15 months). This variability in development
produces challenges for exposure assessment. If an age-dependent model of exposure is based
on a prototypical child at that age, it may have little bearing on exposure patterns of specific
individuals who are delayed or advanced in development.
III. Characteristics of Children That Influence Exposure
Both physiological and behavioral characteristics influence children's exposures to
environmental contaminants. Physiology and behavior is a function of age, gender,
race/ethnicity, and socio-economic status. All of these characteristics pose challenges for
categorizing children and collecting data on their exposures; these challenges are briefly
reviewed here.
Physiological characteristics
Physiological characteristics influence exposure by affecting a child's rate of contact with
exposure media or by altering the exposure-uptake relationship that governs internal dose
resulting from an exposure. Children have a much larger surface area relative to body weight
than do adults. The surface-area-to-body-weight ratio for newborn infants is more than two times
greater than that for adults. This ratio decreases by about a third within the first year of life and
remains constant until about 17 years of age, when it decreases to the adult value (8). In addition
G-7
-------
to providing more area for dermal absorption, the larger relative surface area of children means
that body heat will be lost more rapidly to the environment, requiring a higher rate of metabolism
to maintain body temperature. In addition, extra metabolic energy is needed by children to fuel
growth and development. The higher basal metabolic rate and energy requirements in children
mean that both oxygen and food requirements are greater per kilogram body weight for a child
than for an adult. The higher breathing rate and food consumption rate required to meet these
physiological needs for children will result in higher relative exposures to environmental
contaminants in air and food.
The absorbed dose—the amount of chemical that crosses a receptor's external boundaries—of an
environmental contaminant probably is the relevant measure of exposure for assessment of health
risk. Age-dependent barrier properties of the skin, respiratory-tract lining, and gastrointestinal-
tract lining influences absorbed dose. The permeability of the skin, highest at birth, decreases in
the first year such that the skin of a 1-year old child is similar to that of an adult (5). In addition,
a layer of subcutaneous fat develops in infants about 2-3 months old and continues to exist
through the early toddler period (9). This layer of fat may act as a sink for lipophilic chemicals
absorbed through the skin. Changes in the permeability of lung epithelial cells during childhood
have not been reported. However, the gas-exchange sacs, or alveoli, continue to develop until
adolescence, increasing the surface area for absorption so that the same exposure might lead to a
higher absorbed dose as a child ages. Finally, in the neonate, the stomach produces gastric acid
at about 50% of the adult level (10). As a result, stomach pH exceeds 2 until several months
after birth, when it drops by more than 15% to adult levels. Gastric pH affects absorption by
altering the ionization state of chemicals. Absorption and permeability in the gut are also
regulated by the body to provide nutritional needs which vary with age. For example, children
can absorb more calcium than adults from the gastrointestinal contents to satisfy growth needs.
The absorption of similar positive ions (such as lead) can also be enhanced inadvertently by the
same mechanism used to actively absorb calcium.
Behavioral development
Children's behavior and the way that children interact with their environment may have a
profound effect on the magnitude of their exposures to contaminants.
Motor Capacity. A child's motor capacities determine how that child interacts with her
environment. The manner in which infants and toddlers move is significantly different from the
manner in which adults move and can significantly impact their exposure to contaminants in the
air and on residential surfaces. Motor capacity increases as a child develops. As a result,
children spend less time playing on the floor and touching other potentially contaminated
surfaces as they gain mobility and extend the boundaries of their interactions.
Measurements or descriptions of the changes in motor capacity that occur as a child develops is
described in the developmental psychology and pediatrics literature (11). Much of this literature,
however, focuses on changes in motor capacity that can be used to identify developmental
disabilities and whether children have arrived at various developmental milestones (12). None of
it directly addresses how a child's behavior might contribute to exposure to environmental
chemicals. Using developmental milestones as an indication of children's interactions with the
environment is problematic, as there is significant variability between when a child first achieves
G-8
-------
a milestone and when the child performs the activities on a regular basis. In addition, activities
such as crawling are not included because not all children crawl, and there is tremendous
variation in how and when children first move around. Despite these drawbacks, developmental
milestones can serve as useful guidelines for classifying children in exposure studies.
Manual dexterity includes the ability to pick up, hold, and manipulate objects held in the hand. A
child's hands are the means for placing food in the mouth and are the immediate source of
non-dietary exposure through hand-to-mouth and object-to-mouth behavior. Because the hand is
used to act on the environment and probably has more contact with water, soil, and dust than any
other part of the body, hands have been used as the equivalent of dermal surfaces in several
studies (13-15).
Extensive research has been done to document the changes in manual coordination of very young
children as they mature (26-20). Children show wide variability in manipulative performance. A
young child has not developed a stable manner of handling objects, and the performance is
variable in both style and effectiveness (19). Quantifying significant intra- and inter-child
differences for exposure assessment in moving about and handling objects remains a challenge.
Mouthing Behaviors. Characterizing and quantifying children's mouthing behaviors is also
important for assessing the potential for contacting and transferring contaminants from objects
and surfaces in the environment. Sucking and mouthing hands and objects are natural behaviors
in childhood development. Infants are born with a sucking reflex, providing them with both
nutrition and a sense of comfort or security. If infants do not receive unrestricted breast feeding,
they will suck on a pacifier, thumb (or other finger), or other object like a blanket or stuffed
animal. As infants develop, they begin to explore their world through mouthing (21). During
this stage of development, children put almost everything that they contact into their mouths for a
few seconds. Young children may also begin to use the mouth as a third hand, placing some
objects in the mouth in order to manage them.
Teething is another important stimulus for mouthing activities. Biting and chewing on fingers
and objects to relieve the discomfort of teething may be extensive. Teething usually begins
between 4 and 7 months of age, but may start several months earlier or later. As with all
childhood behaviors, mouthing activities vary significantly from child to child and, therefore, the
impact on exposure will also be highly variable.
Physical activities
Exposure to contaminants is a function of the specific physical activities in which a child is
engaged (e.g., playing games, watching television), the location of these activities (e.g., outdoors,
at school, in the living room), and the child's activity level while so engaged. Different activities
lead to exposures by different pathways. Locations where a child spends time determine the
exposure media that may be contacted, and affect the activity level that determines contact rate
with those media. Differences in duration and frequency of periods spent in particular locations
result in different exposures and risks to children that vary with age and development stage.
Additional variability among children of similar developmental stages is associated with seasonal
and geographic differences in activity patterns and use of indoor and outdoor space. These
concepts are discussed further in subsequent sections on physical activity data.
G-9
-------
Diet and eating habits
Children's diets differ significantly from those of adults. The diet of newborns is limited
exclusively to breast milk or formula, both of which may expose infants to significant
concentrations of environmental contaminants (22-23). Infants and young children eat more fruit
and milk products in proportion to their body size and have a less varied diet than adults. In
addition, there may be tremendous variability in diet among young children of similar ages and
for a single child at different periods in time. Some infants and toddlers go through phases where
only a few preferred foods are eaten for weeks and months at a time. Such a limited diet may
potentially increase dietary exposure of young children to environmental contaminants such as
pesticide residues in fruit (3, 6).
In addition to the exposures associated with the foods that children eat, the manner in which
children handle food as they eat may also impact their exposure to environmental contaminants.
Small children are less likely than adults to consume food in a structured environment. Small
children may sit on the floor or lawn to eat and often pick up and eat foods that have fallen on the
floor. Infants and young children also eat most of their food with their hands. Increased
exposure occurs when children handle and eat foods that have come in contact with the floor or
other contaminated residential surfaces (24-25).
Gender
Gender has been identified as a factor influencing activity level and the types of behaviors and
activities in which children participate (26-28). As early as preschool (ages 3-5), gender
differences exist in the types of games played, frequency of play, and activity level. Locations in
which children spend time also vary with gender. Clear differences in frequency and type of
outdoor activities have been found between boys and girls ages 7 to 15 (28-29). Boys are more
likely than girls to play outdoors, and the character of their activity is different from girls. Boys
are more likely to be involved in physically vigorous activities such as soccer, hockey, and
bicycling, while girls are reported to sit and go for walks. Thus, in exposure assessment for
school-aged children, gender differences in activity level and activity type must be addressed.
There are insufficient data to indicate whether there are gender differences in activity levels of
infants and toddlers. It is useful for exposure modeling to know when the differences emerge, as
well as the degree to which they influence exposure.
Socioeconomic status and race/ethnicity
Children's exposure to environmental contaminants is likely to vary based on the socioeconomic
status (SES) of the child. Though evidence exists to suggest that low-income groups tend to be
more exposed to many environmental pollutants than the general population, data are currently
insufficient to characterize the relationship between SES, ethnicity/race, age, and exposure (30).
Exposure factors related to children that may be affected by SES and race include the following:
• Proximity to source (e.g., distance from toxic release inventory sites)
• Location (e.g., urban, suburban, rural)
• Housing stock (e.g., age, condition, type)
• Activity patterns (e.g., hygiene, housekeeping, activity level, child care)
• Diet and drinking water supply
While there are substantial data on the influence of housing stock, location and socioeconomic
G-10
-------
status on environmental exposure and adverse health outcomes, there are few data on the
relationship of these influences to children's activities and potential contact with the physical
environment. One study of Swedish children from two housing projects found that proximity to
parks and play areas and the floor on which children live in an apartment house influence where
young children play and the amount of time urban children play outside (29). However, there is
little to suggest that housing stock and location have any influence on children's behavior, and
there are no comparable data evaluating children's activities in the United States.
Comparisons of play activities across social classes have been studied for preschool children (31-
33). Some of the studies were conducted within the home and others at day care centers. When
the location was the same (i.e., daycare), no differences in behaviors were observed in children of
different social classes. However, within the home, class (as an indicator of poverty, social
stimulation, and poor parental education) influenced what the children had to play with and the
type of play in which the children engaged (34-36). For subjects tracked from age 15 to age 25 at
5-year intervals, social class and education level were related to the type and level of activities in
which the children participated (37) . Children identified as low social class were less active, and
children who eventually went to college were more active.
Maternal influences on children's activity patterns have been evaluated using the HOME survey
(34-35). Particularly for infants and toddlers, the mother is a major factor hi determining what
the child does, what the child eats, and where the child is located.
Though a disproportionate percentage of ethnic and racial minorities belongs to economically
disadvantaged populations, there are few studies that specifically address the relationship
between race or ethnicity and behaviors that might influence exposure to environmental
contaminants. Most of the studies that address this issue consider lead exposure. One such study
found that black urban children are more likely than white urban children to ingest paint lead
from window sills, while white children ingest soil and suck fingers more than black children
(38). These behaviors contributed to the children's exposure to lead in multivariate analyses.
However, a study of 3-4 year old children in daycare programs found no differences in the
behaviors of black, white, and Mexican American children within the context of the daycare
setting (39). This does not mean that differences that are culturally or economically driven might
not exist when the children are at home or away from the daycare setting.
IV. Children's Exposure Monitoring Data
A variety of methods have been used to collect information about children's exposure.
Telephone surveys and questionnaires can be used to capture global events, particularly those
that relate to air pollutant exposure. Diaries go into more detail than surveys and collect
information related to temporal variations in activities and behaviors that may contribute to
exposure through multiple routes. Observations, personal monitoring, and biological monitoring
are valuable tools for collecting precise and detailed information. Because monitoring methods
are often labor intensive and costly to implement, these are typically used with smaller groups of
subjects. Methods for collecting personal and biomonitoring data for children are discussed hi
this section, while physical activity data are addressed separately in the next section.
G-ll
-------
Personal monitoring
To assess dietary exposure, prototypical diets have been used to characterize children. However,
these do not characterize specific subpopulations such as ethnic groups or inner-city poor. In
addition, the available FDA data sets are out of date and do not reflect the dramatic shift to fast
food diets that has occurred hi the United States. Existing dietary contaminant models assume
that all contaminants can be accounted for prior to the food entering the home or institution.
Data presented by Wilson et al. (40) and Sheldon et al. (41) suggest that there are sources of food
contamination within the institution and home that need to be addressed. These include the
influence of residential and institutional pesticide treatment on food pesticide levels and the
influence of hygiene habits on other food contaminants such as lead. To obtain more specific
information on dietary exposures, data are obtained by collecting duplicate-diet samples. These
samples include a duplicate portion of all food and beverages prepared and consumed in the
home. Results of duplicate-diet analysis are used in combination with food diaries and
supplemental questionnaires to assess exposures by dietary ingestion. More refined protocols to
assess dietary exposures of young children caused by contact of foods with contaminated
surfaces during eating are currently under development and testing (42-44).
For inhalation exposure, a variety of motion detectors and personal monitoring backpacks have
been developed to quantify activity levels and sample air within the individual's breathing zone
(45). While motion detectors have been used with some children, most of these studies were
designed to evaluate the technique and have not proceeded to thoroughly characterize the level of
activity in a large population of children. Breathing zone air monitors have been used with the
few children who participated in NHEXAS-Region V (46). Monitoring backpacks that can be
worn successfully by children of all ages have not been developed. As a result, personal
monitoring is seldom done on infants and preschool children.
Current techniques for measuring dermal exposure are limited in utility. Measures of skin
contamination do not reflect changes hi dermal loading that occur subsequent to sampling and do
not indicate the amount of contamination actually absorbed through the skin (47-48). In addition,
dermal measurement methods developed for occupational use (where the environment and
physical activities are homogenous) may not be useful for measuring children's residential
exposures.
Finally, some of the most significant exposures to environmental contaminants experienced by
children may be related to non-dietary ingestion of contaminant residues, dust, and soil during
mouthing of hands subsequent to dermal contact with contaminated surfaces and objects.
Reliable methods to monitor non-dietary ingestion of environmental contaminants have not been
developed (8). However, non-dietary ingestion of soil and dust has been monitored in fecal
samples using tracer elements (49-53). These studies require collection of dietary data and
concentrations of contaminants hi residential soil and dust to link the tracers to ingested soil and
then to estimate ingestion of contaminants.
Biological monitoring
Biomarkers can serve as a useful measure of direct exposure aggregated over all sources and
pathways; measuring integrated exposure from all routes. However, to use biomarkers for this
purpose, several important criteria must be met. Biomarkers that can accurately quantify the
G-12
-------
concentration of an environmental contaminant or its metabolite(s) in easily accessible biological
media (blood, urine, breath) must be available. It has to be specific to the contaminant of
interest, so that its presence can be linked to that contaminant. The pharmacokinetics of
absorption, metabolism, and excretion must be known. Finally, the time between exposure and
biomarker sample collection must be known. Although there are a number of biomarkers that
meet these criteria, very few studies using biomarkers have collected all the information required
to accurately estimate exposure. In addition, significant challenges are associated with collecting
biomarker data from children (54).
Biomarker data have been collected for children to evaluate environmental exposures to lead
(55), benzene (56), arsenic (57), chromium (58-60), and pesticides (61-62). Most recently, the
Minnesota NHEXAS children's pesticide exposure study collected urine samples from children
on three alternate days and analyzed them for metabolites of chlorpyrifos, malathion, atrazine and
diazinon. Thus far only the chlorpyrifos values are available (61). The children's median levels
of the chlorpyrifos biomarker, TCPY, over the three measurements was 8.6 ppb, as compared to
2.2 for the population-based NHANES III adult population. About 60% of the homes in the
NHEXAS study were identified as using or storing pesticides in the home within a year, and were
considered to be "user" homes (though the data do not show whether or not pesticides were
applied during monitoring). Levels for children in these homes were significantly higher than
levels for children from homes classified as low-users. However, some of the highest monitored
values were found in the low-user children, suggesting that sources of exposure could not be
identified based only on categorization of household pesticide use. Similar results were found in
a study that attempted to determine whether children who lived near a pesticide manufacturing
plant were exposed to poly chlorinated biphenyls (63). There was no difference between the
proposed exposed children compared to controls; all children had measurable levels of the
metabolite, and no additional sources of exposure were reported. In a study by Loewenherz etaL
(62), children up to 6 years of age who lived with pesticide applicators hi an agricultural region
of Washington State were monitored for increased risk of pesticide exposure. Results of this
study did indicate that applicator children experienced higher pesticide exposures than did
reference children in the same community and that proximity to spraying is an important
contributor to these exposures.
V. Children's Activity Pattern Data
As noted previously, a child's exposure is greatly affected by where the child is and what the
child is doing. In exposure modeling, the location a child occupies is known as a
microenvironment. A microenvironment is a physical, three-dimensional space having a well-
characterized, relatively homogenous pollutant concentration level over a specified time period
(64). A child's activity in a microenvironment (e.g., indoors at home) can be described by what
the child is doing in a general sense, such as watching TV, eating, playing games, and crawling
around on the floor. This type of information has been used since the early 1980's to assess
inhalation exposures (65). However, in recent years it has become obvious that general activity
descriptions do not provide enough information on the specific contacts with exposure media that
occur within a microenvironment of interest to estimate dermal and non-dietary ingestion
exposures.
In response to this need for more detailed information, a distinction is now made between
G-13
-------
"macro-" and "micro-" activity information. The general activities described above are
macroactivities. Microactivities are detailed actions that occur within a general activity, such as
hand-to-surface and hand-to-mouth behavior. The physical activity data, both macro- and micro-
activity, available to assess exposure are reviewed in the following sections. Activity pattern
data requirements are demonstrated in the context of algorithms used to estimate exposure by
inhalation, dermal contact, and ingestion. These algorithms for combining the environmental
monitoring data with the exposure factors to estimate an exposure or a dose should be used to
guide the type of data collected to assess children's exposures.
Activity data required and available to assess inhalation exposures
For inhalation, exposure is estimated for each of the microenvironments where a child spends
time and each macroactivity that would result in a different inhalation rate while engaging in that
activity. Exposure over the 24-hour period is then the sum of all of the
microenvkonmental/macroactivity (me/ma) exposures.
For each individual me/ma, inhalation exposure over the 24-hour period (Eime/ma) is defined as
ma = ^me/ma.x Qime XIR^ (I)
where
= the time spent hi that me/ma over the 24 hour period (h/24h)
= ^e ^ concentration measured hi the microenvironment (mg/m3)
= the child's respiration rate representing his activity level for that macroactivity
(m3/h)
In order to apply the above model data are required on the amount of time the child spends hi
each me/ma over a 24-hour period (macroactivity data) and on the child's inhalation rate for each
me/ma. Inhalation rates are typically estimated based on age and weight of the child and on the
macroactivity.
Macroactivity data are obtained using a variety of survey techniques, such as time-budget diaries
or recall (yesterday) telephone surveys (66). For a review of a number of these macroactivity
studies, see Ott (67) and McCurdy (68). Macroactivity information relevant to inhalation
exposure assessment for an individual contains at least one complete day of sequential
location/activity data for every discrete major behavior that is undertaken (and disclosed!) by a
respondent This is known as a "person-day" of information. There are nine studies that
recorded person-day macroactivity data on a flexible-time basis, but not all include data on
children. The data from all of these studies are contained in the EPA National Exposure
Research Laboratory's Consolidated Human Activity Database (CHAD)1 (69). Data from four of
these studies are also available hi EPA's THERdbASE software system on the Internet (70).
For children and adolescents younger than 18, CHAD contains about 4,300 person-days of
information. An explicit breakdown of these data for children <12 years old appears as Table 1.
'CHAD is a relational database using a common set of codes for activities, locations, intensity levels, and
questionnaire information (80). Thus, it allows a user to easily combine information from the nine studies in order
to increase the sample size of human activity data.
G-14
-------
For these children, data are available from only three studies: a) the 1990 California "children
and youth" recall survey (71); b) the 1983 Cincinnati diary study sponsored by the Electric Power
Research Institute (72); and c) the "air" and "water" versions of the 1992-1994 National Human
Activity Pattern Survey (NHAPS) recall survey (73). Altogether, there are 3009 person-days of
macroactivity data in CHAD available from 2640 children <12 years old. Another survey of
children's activities (being incorporated into CHAD) was just released by the University of
Michigan's Institute for Social Research (74).
The person-days of activity data can be used in exposure assessments in a number of ways. Each
person-day of data can be used separately to represent individuals in a modeling exercise or they
can be organized into "cohorts" (such as female babies <6 months old) and used as a "pool" from
which a random sampling routine selects one individual to represent the cohort for a day (75-77).
Macroactivity data can also be aggregated over the total population or a cohort of the population
to obtain "average" or other statistical measures of activity for some specified time period. This
approach is most commonly used in exposure assessment, but it removes the inherent
correlations among activity, location, and time—and the pattern of exposures experienced—that
truly determine dose received from an environmental contaminant. In addition, misleading
results can occur if the assessor is not careful about how the data are prepared to represent a '
group.
Specific examples of the type of macroactivity data available for children are presented in Tables
2 and 3. The number of hours per day children spend in various microenvironments is
summarized in Table 2. Nearly all children in CHAD spent some part of their diary day indoors
at home, and the amount of time spent in this microenvironment ranged from 15 to 20 hours per
day on average (63 - 83% of day) for habitues. Children less than 2 years old spend the most
time indoors at home while older children spend the least amount of tune. Variability within
each age category was substantial, but also fairly consistent across all the age categories (standard
deviations of approx. 4 hours). This high variability remained when comparing hours spent
indoors at home between weekdays and weekends or between seasons, indicating that inter-child
variability in daily activities within each year of age is significant compared to trends due to the
day of the week or season when the diary was collected.
Approximately half of the children in CHAD reported spending time outdoors at home, except
for children in the youngest age categories (< 2 years old). Less than one-third of children under
2 years old reported being hi this microenvironment. Children less than 2 years old also spend
the least amount of time outdoors at home on average, while 4-7 year olds spend more time in
this microenvironment than older children.
Children also spend a significant amount of time in non-residential microenvironments,
including indoors at school, stores, and restaurants; outdoors at parks and playgrounds; and in
vehicles. Approximately 40% of children were hi school during their CHAD diary day for each
age category of school-aged children (^ 5 years old). On average, children spend approx. 6 hours
per day in school. Time spent indoors at school was fairly consistent for children s 7 years of
age, with lower standard deviations (1.0-1.5 hours) than for younger children. A small number
of children less than 5 years old (2 -16%) also reported being in school for as much as 6 hours
per day on average. This highlights the lack of appropriate microenvironment categories for
G-15
-------
young children in the CHAD activity pattern studies. Only the California study included
'childcare facility' as a separate microenvironment category. In the other studies, the 'school'
category may have been used for pre-school or other non-residential childcare facilities, or the
non-specific 'other indoor' category may have been used also.
The number of children in CHAD that reported spending time outdoors at a park or playground
also varied significantly with age. Only 10% of the children in the youngest age categories (< 2
years old) reported being in this microenvironment, while approximately 40% of the older
children (10-11 years old) spent time outdoors at a park or playground. For those children that
reported being outdoors in mis microenvironment., the amount of time spent at a park or
playground did not have a trend across age categories. In addition, age differences were least
evident in the percent of children that reported being in vehicles, as well as the amount of time
spent in vehicles for those children.
The number of hours children spend doing various macroactivities while indoors at home are
summarized hi Table 3; age differences in children's macroactivities are also evident, as shown
in Table 3. On average, the number of hours children spend, both eating and sleeping decreases
gradually with age of the child, so that children less than 2 years old spend the most time doing
these macroactivities. Although showering/bathing times are fairly consistent across ages, the
other macroactivities displayed hi Table 3 do show age differences in the number of hours
children spend playing games, watching television and doing other 'passive' activities while
indoors at home. This table also illustrates another area where macroactivity data in the CHAD
studies are inadequate for characterizing children's activities and exposures. Categories such
'playing games' do not provide any information on the activity level of the child while playing,
which can significantly affect inhalation exposure for example. In addition, appropriate
macroactivity categories for infants were not used in the CHAD studies, so a large percentage of
children less than 1 year old (62%) have a substantial amount of time (3* hours on average) for
the non-specific 'other passive activity' category.
CHAD contains about 140 activity codes and 110 location codes, but data generally are not
available for all activities or locations for any single respondent. In fact, not all of the codes were
used for most of the studies. And even though many codes are used in macroactivity studies,
many of the activity codes do not adequately capture the richness of what children actually do.
They are much too broadly defined and ignore many child-oriented behaviors. Thus, there is a
need for more and better-focused research into children's activities.
Aggregate human activity data are available from additional sources other than those cited above.
Summary and distribution uiformation regarding the time that children spend in various
microenvironments and their activities can be found hi EPA's Exposure Factors Handbook (8)
and the American Industrial Health Council's Exposure Factors Sourcebook (78). These are
comprehensive source documents. More limited information about American children's
activities has been published in Berry, et al. (79), Harlos, et al. (80), Roth Associates (81-82),
Schwab et al. (83-84, 28), and Silvers et al. (85-86). The last study is a 1990-1991 survey of
1000 households with children 5-12 years old in six states. Results of that survey closely match
those of the California study mentioned earlier in this section.
G-16
-------
Activity data required and available to assess exposure by dermal contact and non-dietary
ingestion
Two main approaches are currently used to assess dermal and non-dietary ingestion exposure.
These assessment approaches provide different ways of integrating exposure over time and space.
In the macroactivity approach, exposure is estimated individually for each of the
microenvironments where a child spends time and each macroactivity that the child conducts
within that microenvironment. To do this, exposure is modeled using empirically-derived
transfer coefficients to aggregate the mass transfer associated with a series of contacts with a
contaminated medium. In the microactivity approach, exposure is explicitly modeled as a series
of discrete transfers resulting from each contact with a contaminated medium. It is important to
understand that the temporal and spatial scales of activity patterns, exposure media
concentrations, and transfer efficiencies to be measured will depend on the assessment approach
that is used.
To estimate dermal exposure using the macroactivity approach, microenvironments are defined
by location and surface type (e.g., indoors at home on carpet). The dermal exposure associated
with a given macroactivity (e.g., actively playing in the yard) is measured and used to develop an
activity- arid microenvironment-specific transfer coefficient. Exposure can then be estimated
individually for each of the microenvironments where a child spends time and each macroactivity
that the child conducts within that microenvironment. Exposure over the 24-hour period is the
sum of all of the microenvironment/macroactivity (me/ma) exposures. For each me/ma, dermal
exposure over the 24-hour period (Edme/ma) is defined as
(2)
Where
Csurf = total contaminant loading on surface (mg/cm2)
TCder = dermal transfer coefficient for the me/ma (cm2/hr)
ED = exposure duration that represents the time spent in the me/ma (hr/day)
In order to apply the macroactivity approach to assess dermal and nondietary ingestion exposure,
data are required on the amount of time the child spends in each me/ma over a 24-hour period.
While the CHAD activity pattern studies can provide data on time spent in various me/ma, the
types of surfaces associated with each me/ma are not included in the database. Alternatively,
CHAD does include information on time spent in different rooms within a home, which may be
useful in the macroactivity approach to modeling dermal and non-dietary exposures. According
to data in CHAD, children spend the majority of their time indoors at home in the bedroom (an
average of 65-75%) and living room (15-25%). These rooms likely contain textured surfaces
such as carpet and upholstery, compared to the kitchen and bathroom which are likely to have
hard smooth surfaces (linoleum, tile). Since surface types are required to estimate dermal
exposures, this additional data should be collected in future activity pattern studies.
To assess dermal exposure and non-dietary ingestion using the microactivity approach, exposure
is estimated individually for each of the microactivites or events (e.g., each time a child touches a
given object) from which dermal contact or non-dietary ingestion occurs. Exposure over the 24-
hour period is then the sum of all of the individual exposures. For each microactivity, dermal
G-17
-------
exposure over the 24-hour period (Eder/mi) can be defined as
£*„„,,• = CsurfxTExSAxEF (3)
Where
Edcr/mi = dermal exposure for a given rnicroactivity over a 24-hour period (mg/day)
Csurf=total contaminant loading on surface (mg/cm2)
TE = transfer efficiency, fraction transferred from surface to skin (unitless)
SA = area of surface that is contacted (cmVevent)
EF ~ frequency of contact event over a 24-hour period (events/day)
For each microactivity resulting in non-dietary ingestion, exposure over the 24-hour period
(Ending/mi) can be defined as
^xxTExm
x SA* x EF
(4)
•p .
•*-*nding/mi
Where
Ending/mi = non-dietary ingestion exposure for a given microactivity over a 24-hour period
(mg/day)
x = hand or object that is mouthed
Cx = total contaminant loading on hand or object (mg/cm2)
TEjo,, = transfer efficiency, fraction transferred from object or hand to mouth (unitless)
SAx = area of object or hand that is mouthed (cmVevent)
EF = frequency of mouthing event over a 24-hour period (events/day)
To use the microactivity approach, a greater level of detail (i.e., "microactivity data") is needed
to characterize people's dermal contact with chemical residues in their environments and
quantify subsequent dermal absorption and non-dietary ingestion. Microactivities required to
estimate dermal and non-dietary ingestion exposure include frequency and duration of contact
between skin surfaces (including the mouth) and objects and parameters describing the nature of
contact, such as pressure, motion type, and exposed surface area.
Literature about children's activities from the fields of child development and psychology tends
to focus on social development and peer interactions of infants, toddlers and kindergarten
children. The literature seldom reports how children act on, or move about in, their physical
space (87-88). A review of the child behavior and psychology literature can be found in U.S.
EPA, 1998 (89). Frequency and duration of handling and mouthing events were documented in
several of the reviewed studies. However, in these studies caretakers introduced objects to
children sitting on their laps. Handling and mouthing behaviors will differ for a child in his own
environment under normal conditions.
Because of the age dependencies and labor-intensive nature of gathering microactivity data, few
data sets relevant to exposure assessments currently exist. Two general approaches to gathering
such data have been used: a) real-tune hand recording, in which trained observers watch an
individual and write down the information of interest on a score sheet; and b) videotaping, in
which trained videographers tape an individual and then subsequently extract the data of interest
by hand or by computerized software.
A recent study used the first approach to quantify duration of mouthing in awake infants ages 3-
G-18
-------
36 months in the Netherlands (21). Five parents were asked to observe 8 children (10 times, 15
minutes per day on two days) and measure mouthing time with a stopwatch. No differences were
present between the two observed days, across different periods of the day, or between boys and
girls; however, total mouthing time did differ among age groups. The mean daily extrapolated
mouthing times (minutes) for ages 3 to 6 months, 6 to 12 months, 12 to 18 months, and 18 to 36
months were 36.9 (s.d. 19.1), 44 (sd 44.7), 16.4 (sd 18.2), and 9.3 (sd 9.8), respectively. The
youngest children mouthed mainly their fingers, while children 6-12 months mouthed toys not
meant for mouthing. The older age groups mouthed mostly non-toys and their fingers. On ,
average, children sucked or bit on objects two-thirds of the time and licked objects the other one-
third of the time. The children aged 12 to 18 months sucked or bit the most, and the percentage
of licking was largest in the youngest age group. Note this study reported difficulties in parent
training and compliance that may have influenced the reliability of the reported data.
Several studies have used the videotaping approach to quantify children's microactivity data.
The U.S. EPA's National Human Exposure Assessment Survey (NHEXAS) included videotaping
19 children 3 to 12 years old in Minnesota with a hand-held camera. Observers then replayed the
videotapes and recorded the frequency of object-to-mouth contact, hand-to-mouth contact, and
hand contact with the following object categories: clothing, dirt, smooth surface, textured
surface, and hand-held object (90). Reed (91) videotaped 30 children, between 18 months and 5
years old, in New Jersey (20 in a daycare facility and 10 in their homes) for a total of 168 hours
and then recorded hand and mouthing behaviors in the same way as Freeman (90). As in
NHEXAS, observers recorded the frequencies of hand-to-object contacts over 5-minute intervals.
Objects recorded included clothing, dirt, another hand, mouth, object, other items, smooth
surfaces, and textured surfaces. Zartarian et al. (13, 92) reported results for the left hand, right
hand, and mouth from a videotape study of 4 agricultural children (whose ages were 2 to 4 years)
in California (31 hours of videotape). In this study a computer software application (93) was
used rather than a scorecard to obtain the sequence of a wide array of objects contacted and the
duration of each contact. Table 4 summarizes the type of microactivity data collected in these
studies.
Comparing results among these studies is difficult because ages of children, reported summary
statistics, and categories of body parts and objects contacted were different among the studies.
Despite these differences and the small sample sizes, some interesting observations can be
drawn. The children studied exhibited short average duration of mouthing and surface contacts
(on the order of seconds) and high contact frequencies. Average contact frequencies across the
studies for the same object categories were reasonably similar, but the variability for, a particular
object category was high hi each study. Object categories contacted the most frequently by hands
were smooth surfaces (e.g., wood furniture), bedding, clothes, plastic toys, and paper. The only
variable which was statistically different across age groups in the NHEXAS (ages 3-4, 5-6, 7-8,
and 10-12) was object-to-mouth contacts, which were greater for the 3-year olds (6+/-7 per hr)
than the other groups. For age-matched boys and girls, girls exhibited higher object-to-mouth
contacts. However, this may be related to the fact that boys spent substantially more time
outdoors in active play (90). In the New Jersey study, contacts with another hand (either the
child's own hand or another person's hand) were found higher for children 1 to 3 years old
(25/hr) than for children 4 to 6 years old (13.5/hr); hand-to-mouth contacts were significantly
higher in the spring (10.4/hr) than the winter (4.6/hr); no variables were significantly different by
gender; and some variables (contact with dirt, objects-to-mouth, other items, and textured
G-19
-------
surfaces) were statistically significant between daycare and residential children (91). Some
micro-activities appeared to be setting-dependent (e.g., contact with dirt, grass, and toys) while
others (e.g., contact with clothes, body parts, and mouths) did not. In general, non-dietary object-
to-mouth contacts were less frequent than hand-to-mouth contacts. All of these results, however,
may reflect the types of behaviors quantified, the small sample size, and the setting and
conditions under which the observations were made.
In summary, the current database on children's microactivities is sparse. More data for different
ages and body parts over a wide range of scenarios are needed to reduce uncertainty in modeled
estimates of dermal and non-dietary ingestion exposure and dose and to identify important
objects for measuring pollutant concentrations. However, before these data can be collected, the
important activities and contact parameters (e.g., surface type, contact duration, skin condition)
need to be identified to determine the type of microactivity data that should be collected. Then, a
standard protocol for collecting and reporting relevant children's microactivity data could be
developed.
Activity data required and available to assess dietary exposure
Young children do not consume foods in a structured manner. While eating, their foods contact
surfaces (hands, floors, eating surfaces, etc.) that may be contaminated. Thus, dietary exposures
of young children are difficult to accurately assess or measure. A young child's dietary exposure
to environmental contaminants is characterized by the sum of three major terms (42):
1 . the original contaminant residue on foods before they are handled by the child;
2. surface-to-food contamination as the foods come into contact with contaminated surfaces
before being consumed by the child; and
3. surface-to-hand-to-food contamination as the child touches contaminated surfaces and
then handles and eats the foods.
. To assess dietary ingestion, exposure is estimated individually for each item of food consumed
by the child. Total dietary exposure is then the sum of exposures for all food items consumed
over a 24-hour period. For each food item, dietary exposure (Ediet) can be defined as the sum of
the three terms listed above. The intake of a contaminant associated with one food item, i
specific eating activities resulting hi that food item's contact with contaminated surfaces, and j
specific activities resulting hi the food item's contact' with the child's hands before it is eaten can
be described as follows.
Ediet = Cfood WT + L [Csurf TE-^p SAS/F EF S/F ] + £. [Chand
Term 1
EF
(5)
Term 2
TermS
where:
Ediet = Total dietary exposure to the environmental contaminant for one food eaten
(mg/food item)
Cfood = Contaminant concentration of food item after preparation for consumption (ug/g food)
WT = Total amount of the individual food consumed (g food/food item)
G-20
-------
SAS/F
FF
c>r
SfF
Csurf ^ Contaminant loading on a contacted surface (ug/cm2)
Surface to food contaminant transfer efficiency (where transfer efficiency is a
function of duration of contact, surface type, moisture etc.) (unitless)
Area of contaminated surface that is contacted by the food item (cnr/event)
Frequency of surface to food contact events that occur during consumption of
the food item (events/food item)
Contaminant loading on child's hand (ug/cm2)
Hand to food contaminant transfer efficiency (unitless)
Area of the contaminated hand that is contacted by the food (cm2/event)
Frequency of hand to food contact events that occur during consumption of the
food item (events/food item)
FF =
•E/r
H/F
In measurable quantities, Term 1 summed for all foods consumed over the day may be obtained
by duplicate-diet sampling procedures which provides total, daily dietary intake of contaminants
that are present on the foods themselves, plus those that have been introduced during preparation.
Terms 2 and 3 are much more difficult to quantify even for the simplest eating scenario, and
require measurements of specific factors (e.g., surface concentrations, contact areas, transfer.
efficiencies) in the eating environment of the child and analysis of eating activities.
Recent studies on dietary exposure of children to lead (24, 94) and to pesticides (42-44) have
begun to explore potential pathways of dietary contamination caused by the child's eating
activities, and ways to measure them. These studies are focused on young children (1-3 years
old). In the study by Barlion (24), children's dietary exposure to lead was evaluated by collecting
a 24-hour duplicate of all foods plus sentinel foods (i.e., individual food items used to represent
foods contaminated during handling) from 48 children 2 to 3 years of age. Sentinel foods were
contacted with the child's hands and other surfaces to represent ways the child might handle the
foods while eating. Additional information collected included lead concentrations from hand
wipes, floor wipes, and venous blood; and questionnaire responses on activities related to
exposure. Results showed that children's dietary exposure to lead may potentially increase by a
factor from 4-20 when foods are handled by a child in a contaminated environment.
Akland et al. (43) video taped the eating activities of young children to determine the frequency
and duration of activities that may lead to contamination, including hand-to-surface, hand-to-
food, and food-to-surface contacts. The frequency and duration of hand and food contacts with
different surfaces, types and amounts of foods consumed, and other location factors were
recorded for 10 children 1-3 years of age, eating both at home and in day-care facilities.
Summary results from analysis show that there is a wide range of time and contact frequency
between children. A specific food item contacting the child's hands during an eating event
depended on the type of food eaten, and age. Bread, cereal and banana were the food items most
commonly handled while being eaten by these children. Food is in contact with a plate or eating
G-21
-------
utensil for the longest period of time (on the average about 10 minutes), with food and hand
contact, and food and surface contact, each about 2 minutes. Food items come in contact with
plate, hands and mouth about the same number of times on average during an eating event.
Field testing is being conducted to collect additional activity pattern data and to measure other
input parameters required for the dietary exposure model (equation 5) under realistic conditions
to improve dietary exposure assessments for young children. The field testing will also provide
indirect confirmation of the dietary exposure model through comparisons of dietary exposures
estimated by the model with measurements of handled foods and child biomarkers (42).
VI. Total Exposure Studies
An important component of current exposure and risk characterization is the consideration of
aggregate exposures. When assessing exposure and health risk to children, exposure information
should be aggregated from all potential exposure media including the following:
• the air that children breath,
• the foods that children eat,
• groundwater or surface water that is consumed as drinking water or used for bathing,
• other contaminated media contacted under nonoccupational circumstances (i.e., dermal or
non-dietary contact with contaminated residential surfaces).
In Table 5, several examples are presented to demonstrate the type of data required to assess
aggregate exposure to a variety of environmental contaminants. The first two examples,
depicting exposure to methylmercury and lead, might be considered simple systems, each with
one chemical and typically only one route of exposure. The final two examples, depicting
exposure to chloroform and pesticides, require consideration of multiple exposure media and
routes. As shown in Table 5, some of the most useful studies for assessing exposure collect a
combination of personal and biological monitoring data, environmental concentration data, and
activity pattern data. These types of studies are required to assess aggregate exposure by the
indirect approach. Some examples of studies for which a combination of children's exposure
data were (or are currently being) collected are presented in Table 6.
VII. Conclusions
Currently, data on children's exposures and activities are insufficient to adequately assess
exposures to environmental contaminants. As a result, regulators use a series of default
assumptions and exposure factors when conducting exposure assessments. The more uncertain
the assumptions and exposure factors used, the more conservative they must be to protect
children's health. Data to reduce uncertainty hi the assumptions and exposure estimates are
needed to ensure chemicals are regulated appropriately. To improve the database available to
assess children's exposures, three areas of research are required.
1. Identification of appropriate age/developmental benchmarks for categorizing children in
exposure assessments
As discussed previously, the physiological characteristics and behavioral patterns of children not
only result in differences in exposures between children and adults, but also result in differences
in exposures among children of different developmental stages. Classification of children by
age group should be based on estimates of when developmental changes most commonly occur.
G-22
-------
Both physiological and behavioral development need to be considered in developing appropriate
age classifications. Protocols for addressing variability in development need to be established to
insure that exposure patterns of specific individuals who are delayed or advanced in development
can be adequately characterized. In addition, methods need to be developed for addressing
developmental characteristics, such as teething, that will likely span age classifications, yet may
have a very significant influence on a child's exposure.
2. Development and improvement of methods for monitoring children's exposures and
activities
Significant challenges are associated with developing and verifying exposure factors for young
children, such as contaminant contact rates and transfer rates. Novel methods must be developed
and validated in the manner of Sheldon et al. (41), Noland et al. (95), Kissel et al. (96), and
Gurunathan et al. (14) to elicit information from or about young children who are non-verbal or
lack a well-developed sense of time about their activities and exposures. New and improved
methods are needed to monitor personal exposures, measure biomarkers, and survey activities, in,
these young children. Methods that can be used with infants should also be developed.
3. Collection of physical activity data for children (especially young children) required to
assess exposure by all routes
The data availabile for conducting exposure assessments for children are highly variable,
depending on the route of exposure considered. The data that are available for assessing
inhalation exposures is the most complete. However, even for inhalation, limited data are
available for very young children. For all routes of exposure, sufficient population-based data
are needed to better characterize children's exposures and behaviors as a function of age, gender,
setting (residence, school, daycare), socioeconomic status, race/ethnicity, location (urban,
suburban, rural), region, and season. These data gaps are particularly significant for young
children less than 4 years old.
In addition, the following route-specific data are required to improve assessment of children's
exposures.
Dietary ingestion. Improved information on the foods children eat and the residues on them is
needed. Those foods most frequently consumed by infants and children need to be identified;
and distributions of amounts consumed need to be quantified more specifically. Because of the
changing nature of children's diets, food consumption surveys should include adequate sample
sizes of children aged 0 to 6 months, 6 to 12 months, 12 to 24 months, 24 to 36 months, 3 to 5
years, 5 to 10 years, and 11 to 18 years. The residues associated with a child's diet (prior to food
preparation and handling by the child) need to be better characterized. Methods to assess
exposures caused by contamination of foods during consumption by the child need to be
evaluated. Activities specifically related to the way children consume foods need to be
categorized. Current information is not specific enough to determine the relative magnitude of
the child handling component to the total dietary intake of a contaminant.
Inhalation. There is a need for more and better-focused research into children's activities. The
current database, seemingly extensive, is deficient from an exposure modeling perspective
because many of the activity codes do not adequately capture the richness of what children
actually do. They are too broadly defined and ignore many child-oriented behaviors, limiting the
G-23
-------
utility of these data for assessing the frequency and duration of children's contact with
contaminated air, children's activity levels and, consequently, inhalation rates.
Dermal contact and non-dietary ingestion. Currently, there are no methods available to directly
assess dermal and non-dietary ingestion exposures. Therefore, it is particularly important that
studies be performed to identify the most important exposure factors for assessing dermal
exposures. Characteristics of surfaces and objects contacted by children are important in
assessing children's dermal and non-dietary ingestion exposures. Consequently, the definition
used to identify microenvironments hi which children spend tune must be modified to include the
surface type. In addition, more survey and observational studies across all ages of children are
required to characterize both macro- and micro-activities that contribute to dermal exposure in
these microenvironments, as well as contact and transfer necessary for non-dietary ingestion and
contamination of food.
The research needed to better characterize and quantify children's exposures to environmental
contaminants is best conducted by carefully considering the data needed to assess aggregate
exposure. The algorithms for combining the environmental monitoring data with the exposure
factors to estimate an exposure or a dose should be used to guide the type of data collected. In
this way, future research efforts will most efficiently provide the knowledge base needed to
improve exposure assessments for children.
G-24
-------
VIII. References
1. Rogan WJ. Sources and routes of childhood chemical exposures. J Pediatr 97:861-865 (1980).
2. Rogan WJ. Environmental poisoning of children - lessons from the past. Environ Health Perspect 103(suppl
6): 19-24 (1995).
3. National Academy of Sciences. Pesticides in the Diets of Infants and Children. Washington, DCrNAS, 1993.
4. Schneider D. American Childhood: Risks and Realities. New Brunswick, NJ:Rutgers UP, 1995.
5. Bearer, CF. How are children different from adults? Environ Health Perspect 103:7-12 (1995).
6. Goldman LR. Children - unique and vulnerable environmental risks facing children and recommendations for
response. Environ Health Perspect 103(suppl 6):13-18 (1995).
7. WargoJ. Our Children's Toxic. New Haven, CT:Yale UP, 1996. "
8. U.S. EPA. Exposure Factors Handbook Volume I - m. EPA/600/P-95/002Fa,b,c. Washington, DC: Office of
Research and Development, 1997. http://www.epa.gov/ordntrnt/ORDAVebPubs/exposure/
9. Thompson H. Physical Growth. In: Manual of Child Psychology (Carmichael L, ed). New York:John Wiley and
Sons, 1946;255-294.
10. Blackburn ST, Loper DL. The gastrointestinal and hepatic systems. In: Maternal, Fetal, and Neonatal
Physiology: A Clinical Perspective. Philadelphia:Saunders, 1992;379-438.
11. Gesell A. Child development. In: Manual of Child Psychology (Carmichael L, ed). New York: Wiley and Sons,
1946.
12. Shelov, SP, Hannemann, RE. (1993) Caring for Your Baby and Young Child: Birth to Age 5. The American
Academy of Pediatrics. Bantam Books, New York, NY.
13. Zartarian VG, Ferguson AC, Leckie JO. Quantified dermal activity data for a four child pilot field study. J
Expo Anal Environ Epidemiol 7:543-552 (1997).
14. Gurunathan S, Robson M, Freeman NCG, Buckley B, Roy A, Meyer L, Bukowski J, Lioy PJ. Accumulation of
chlorpyrifos on residential surfaces and on/in toys accessible to children. Environ Health Perspect 106:9-16 (1998).
15. Fenske RA, Black KG, Elkner KP, J_£e C, Methner M, Soto R. Potential exposure and health risks of infants
following indoor residential pesticide application. Am J Public Health 80:689-693 (1990).
16. Connolly RS, Elliot JM. Evolution and ontogeny of hand function. In: Ethological Studies of Child Behavior
(Blurton-Jones N, ed). London:Cambridge, UP, 1972.
17. Humphrey R, Jewell K, Rosenberger RC. Development of in-hand manipulation and relationship with
activities. Am J Occup Ther 49:763-771 (1995).
18. Thelen E. Motor development. AmPsychol 50:79-95 (1995).
19. Pehoski C, Henderson A, Tickle-Degnen L. In-hand manipulation in young children: rotation of an object in the
fingers. Am J Occup Ther 51:544-552 (1997).
20. Pehoski C, Henderson A, Tickle-Degnen L. In-hand manipulation in young children: translation movements.
Am J Occup Ther 51:719-728 (1997).
G-25
-------
21. Groot ME, Lekkerkerk MC, Steenbekkers UP A. Mouthing behaviour of young children: an observational study.
Wageningen, The NetherlandsrAgricultural University Wageningen, Household and Consumer Studies, 1998.
22. Mukerjee D. Assessment of risk from multimedia exposures of children to environmental chemicals. J Air
Waste ManagAssoc 48:483-501 (1998).
23. Chance GW, Harmsen E. Children are different: environmental contaminants and children's health. Can J
Public Health 89:89-815 (1998).
24. Barlion, LJ. "Dietary exposure of children living in lead-laden environments" Ph.D. Dissertation, University of
Cincinnati, 1999.
25. Childs BH, and Pellizzari, ED. "Child Dietary Lead Study: Final Report" Cooperative Agreement: CR 822070-
01-D, US EPA, (1999).
26. Gill DC. Gender differences in competition and sport participation. Int J Sports Psy 19:145-159 (1988).
27. Garcia CA. Gender differences in young children's interactions when learning fundamental motor skills. Res Q
Exerc Sport 65:213-225 (1994).
28. Schwab M, McDermott A, Spengler JD. Using longitudinal data to understand children's activity patterns in an
exposure context: data from the Kanawha County health study. Environ Intern 18:173-189 (1992).
29. Bjorklid-Chu P. A survey of children's outdoor activities in 2 modem housing areas in Sweden. In: Biology of
Play (Tizard, Harvey D, Heinneman W, eds). London:Heinneman, 1997;146-159.
30. Sexton K, Adgate JL. Looking at environmental justice from an environmental health perspective. J Expo Anal
Environ Epidemiol 9(l):3-8 (1999).
31. von Zuben MV, Crist P, Mayberry W. A pilot study of differences in play behavior between children of low
and middle socioeconomic status. Am J Occup Ther 45:113-118 (1991).
32. Vandell DL, Powers CP. Day care quality and children's free play activities. Am J Orthopsychiatry 53:493-500
(1983).
33. Rubin K, Maioni T, Homung M. Free play behaviors in middle- and lower-class preschoolers. Child Dev
47:414-419(1976).
34. Bradley RH, Caldwell, BM. 174 children: a study of the relationship between home environment and cognitive
development during the first 5 years. In: Home Environment and Early Cognitive Development (Gottfried AW, ed).
New York:Academic Press, 1984.
35. Bradley RH, Caldwell BM, Rock SL. Home environment and school performance: a ten-yr follow-up
examination of three models of environmental action. Child Dev 59:852-867 (1988).
36. DuRant RH, Baronowski T, Puhk J, Rhodes T, Davis H, Greaves KA, Thompson WO. Evaluation of the
children's activity rating scale (CARS) in young children. Med Sci Sports Exerc 25:1415-1421 (1993).
37. Engstrom, L-M. Physical activity of children and youth. Acta Paediatr Scand Suppl 282:101-105 (1980).
38. Lanphear BP, Weitzman M, Eberly S. Racial differences in urban children's environmental exposures to lead.
Am J Public Health 86:1460-1463 (1996).
39. Baranowski T, Thompson WO, DuRant RH, Baranowski J, and Puhl J. Observations on physical activity in
physical locations: age, gender, ethnicity, and month effects. Res Q Exerc Sport 64:127-133 (1993).
G-26
-------
40. Wilson NK, Chuang JC, Nishioka M, Lyu, C. Measurements of persistent organic chemicals in several day care
centers. 7th Annual meeting of International Society Of Exposure Analysis, 11/2-5, Research Triangle Park, NC,
1997.
41. Sheldon L, Melnyk L, Berry M, Freeman NCG. Determination of children's dietary exposure to lead. 7th
Annual meeting of International Society of exposure analysis, 11/2-5, Research Triangle Park, NC, 1997.
42. Berry MR, Adcox C, Melnyk LJ, Akland GG, Clayton CA, Hu YA, Aragon ED, Roberds JM, Pellizzari ED.
Measuring dietary exposure of young children. Presented at the Analytical Challenges for Assessing Environmental
Exposure to Children Symposium, ACS National Meeting. Aug 22-26, New Orleans, LA; 1999.
43. Akland GG, Pellizzari ED, Hu YA, Clayton CA, Long K, Roberds JM, Berry MR, Leckie J. The three
interacting factors associated with children's dietary exposures: environmental concentrations, food contamination,
and children's behaviors. Presented at the Analytical Challenges for Assessing Environmental Exposure to Children
Symposium, ACS National Meeting. Aug 22-26, New Orleans, LA; 1999.
44. Adcox C, Berry MR, Akland GG, Roberds JM, Pellizzari ED. Transfer of pesticides from surfaces to foods for
the estimation of dietary exposure of children. Presented at the Analytical Challenges for Assessing Environmental
Exposure to Children Symposium, ACS National Meeting. Aug 22-26, New Orleans, LA; 1999.
45. Pellizzari E, Lioy P, Quackenboss J, Whitmore R, Clayton A, Freeman N, Waldman J, Thomas K, Rodes C,
Wilcosky T. Population-based exposure measurements in EPA Region 5: A phase I field study in support of the
national human exposure assessment survey. J Expo Anal Environ Epidemiol 5:327-358 (1995).
46. Pellizzari ED, Perritt RL, Clayton CA. National human exposure assessment survey (NHEXAS): exploratory
survey of exposure among population subgroups in EPA Region V. J Expo Anal Environ Epidemiol 9(l):49-55
(1999).
47. Fenske RA. Dermal exposure assessment techniques. Ann Occup Hyg 37(6): 687-706 (1993).
48. McArthur B. Dermal Measurement and Wipe Sampling Methods: A Review. Appl Occup Environ Hyg
7(9):559-606 (1992).
49. Binder S, Sokal D, and Maghan, D. Estimating soil ingestipn: the use of tracer elements in estimating the
amount of soil ingested by young children. Arch Environ Health 41(6):341-345 (1986).
50. Calabrese EJ, Barnes R, Stanek EJ ffl, Pastides H, Gilbert CE, Veneman P, Wang X, Lasztity A, Kostecki PT.
How much soil to young children ingest: an epidemiological study. Regul Toxicol Pharmacol 10:123-137 (1989).
51. van Wijnen JH, Clausing P, Brunekreff B. Estimating soil ingestion by children. Environ Res 51:147-162
(1990).
52. Davis S, Waller P, Buschbom R, Ballou J, White P. Quantitative estimates of soil ingestion in normal children
between the ages of 2 and 7 years: population-based estimates using aluminum, silicon, and titanium as soil tracer
elements. Arch Environ Health 45:112-122 (1990).
53. Calabrese EJ, Stanek EJ HI, Pekow P, Barnes RM. Soil ingestion estimates for children residing on a superfund
site. Ecotoxicol Environ Saf 36:258-268 (1997).
54. Weaver VM, Buckley TJ, Groopman JD. Approaches to environmental exposure assessment in children.
Environ Health Perspect 106(3):827-832 (1998).
55. Brody DJ, Pirkle JL, Kramer RA, Flegal KM, Matte TD, Gunter EW, Paschal, DC. Blood lead levels in the US
population. J Am Med Assoc 272:277-283 (1994).
G-27
-------
56. Weaver VM, Davoli CT, Heller PJ, Fitzwilliam A, Peters HL, Sunyer J, Murphy SE, Goldstein GW, Groopman
JD. Benzene exposure, assessed by urinary trans, trans Muconic Acid in urban children with elevated blood lead
levels. Environ Health Perspect 104:318-323 (1996).
57. Hwang Y-H, Bomschein RL, Grote J, Menrath W, Roda S. Urinary arsenic excretion as a biomarker of arsenic
exposure in children. Arch Environ Health 52:139-147 (1997).
58. Freeman NCG, Wainman T, Stern AH, Shupack S, Lioy PJ. The effect of remediation of chromium waste sites
on chromium levels in urine of children living in the surrounding neighborhood. J Air Waste Manag Assoc
45:604-614(1995).
59. Lioy PJ, Freeman NCG, Wainman T, Stem AH, Boesch R, Howell T, Shupack SI. Microenvironmental
analysis of residential exposure to chromium laden wastes in and around New Jersey homes. Risk Anal 12:287-299
(1992).
60. Stem AH, Freeman NCG, Pleban P, Boesch R, Wainman T, Howell T, Shupack SI, Johnson BB, Lioy PJ.
Residential exposure to chromium - urine biological monitoring in conjunction with environmental exposure
monitoring. Environ Res 58:147-162 (1992).
61. Quackenboss J, Pellizzari ED, Clayton A, Lioy PJ, Shubat P, Sexton K. Measurement and analysis of children's
exposures to pesticides and PAHs. The 7th annual meeting of the International Society of Exposure Analysis, Nov
2-5, Research Triangle Park, NC; 1997.
62. Loewenherz C, Fenske RA, Simcox NJ, Bellamy G, Kalman D. Biological monitoring of organophosphorus
pesticide exposure among children of agricultural workers in central Washington State. Environmental Health
Perspectives 105:1344-1353 (1997).
63. Wolff M, Schecter A. Accidental exposure of children to polychlorinated biphenyls. Arch Environ Contam
Toxicol 20:449-453 (1991).
64. U.S. EPA. Guidelines for exposure assessment. Federal Register 57 (29 May 1992):22888-22938. Microfiche.
65. McCurdy TR. Estimating human exposure to selected motor vehicle pollutants using the NEM series of models:
lessons to be learned. J Expo Anal Environ Epidemiol 5:533-550 (1995).
66. Robinson JP. Time-diary research and human exposure assessment: some methodological considerations.
Atmos Environ 22:2085-2092 (1988).
67. Ott WR. Human activity patterns: a review of the literature for estimating time spent indoors, outdoors, and in
transit. In: Proceedings of the Research Planning Conference on Human Activity Patterns (Starks TH, ed). Las
Vegas:U.S. Environmental Protection Agency, 1989. (EPA-450/4-89-004).
68. McCurdy TR. Human exposure to ambient ozone. In: Tropospheric Ozone: Human Health and Agricultural
Impacts (McKee DJ, ed). Boca Raton:Lewis Publishers, 1994;85-128.
69. GlenG, LakkadiY, Tippett JA, del Valle-Torres M. Development of NERL/CHAD: The National Exposure
Research Laboratory Consolidated Human Activity Database. Research Triangle Park, NC: ManTech
Environmental Technology, 1997.
70. U.S. EPA. THERdbASE software system on the Internet, 1999. http://www.epa.gov/nerl/heasd/therdbase.htm
71. Wiley JA, Robinson JP, Cheng Y-T, Piazza T, Stork L, Pladsen K. Study of Children's Activity Patterns.
Berkeley, CA:Survey Research Center, University of California, 1991.
72. Johnson T. Human Activity Patterns in Cincinnati, Ohio. Palo Alto, CA: Electric Power Research Institute,
1989.
G-28
-------
73. Klepeis N, Tsang A, Behar JV. Analysis of the National Human Activity Pattern Survey (NHAPS)
Respondents from a Standpoint of Exposure Assessment. Las Vegas, NV:National Exposure Research Laboratory,
U.S. Environmental Protection Agency, 1995.
74. University of Michigan's Institute for Social Research (1999), http://www.isr.umich.edu/src/child-
development/data.html. . .
75. Johnson T, Gapel J, McCoy M. Estimation of Ozone Exposures Experienced by Urban Residents Using a
Probabilistic Version of NEM and 1990 Population Data. Durham NC:IT Corporation, 1996.
76. Hayes SR, Rosenbaum AS, Wallsten TS, Whitfield RG, Winkler RL, Richmond H. A health risk assessment
for use in setting the U.S. primary ozone standard. Paper presented at the 3rd U.S.-Dutch International Symposium
on Atmospheric Ozone Research and Its Policy Implications; May, 1988.
77. Lurmann F, Colome SD, Hogo H. Modeling current and future human exposure to ozone in Southern
California. In: Tropospheric Ozone and the Environment n (Berglund RL, ed). Pittsburgh:Air & Waste
Management Association, 1992;725-745.
78. American Industrial Health Council. Exposure Factors Sourcebook. Washington, DC:AIHC, 1994.
79. Berry M, Lioy PJ, Gelperin K, Buckler G, Klotz J. Accumulated exposure to ozone and measurements of
health effects in children and counselors at two summer camps. Environ Res 54:135-150 (1991).
80. Harlos DP, Marbury M, Samet J, Spengler JD. Relating indoor NO2 levels to infant personal exposures.
Atmos Environ 21:369-376 (1987).
81. Roth Associates. A Study of Activity Patterns Among a Group of Los Angeles Asthmatics. Palo Alto,
CA:Electric Power Research Institute, 1988.
82. Roth Associates. A Survey of Daily Asthmatic Activity Patterns in Cincinnati. Palo Alto, CA:Electric Power
Research Institute, 1989.
83. Schwab M, Spengler JD, Ozkaynak H,Terblanche P. The time/activity component of the Kanawha County
health study. In: Total Exposure Assessment Methodology. Pittsburgh:Air & Waste Management Association,
1990;! 18-129.
84. Schwab M, Terblanche A, Spengler J. Self-reported exertion levels on time/activity diaries: application to
exposure assessment. J Expo Anal Environ Epidemiol 1:339-356 (1991).
85. Silvers A, Florence BT, Rourke DL, Lorimor RJ. How Children Spend Their Time: A Sample Survey for Use
in Exposure and Risk Assessment. Washington, DC: Resource Planning Corporation, 1992.
86. Silvers A, Florence BT, Rourke DL, Lorimor RJ. How Children Spend Their Time: A sample survey for use in
exposure and risk assessment. Risk Anal 14:931-944,1994.
87. Brownell, CA. Peer social skills in toddlers: competencies and constraints illustrated by same-age and
mixed-age interaction. Child Dev 61:838-848 (1990).
88. Putallaz M, Gottman JM. An interactional model of children's entry into peer groups. Child Dev 52:986-994
(1981).
89. U.S. EPA. The Role of Child Behavior and Activities in Determining Exposure to Xeniobiotics, Child Behavior
Patterns: An Analysis of the Data. EPA/600/X-98/005. Washington, DC:Office of Research and Development,
1998.
G-29
-------
90. Freeman, N., 1999, "Susceptibility related to differential exposure and/or doserstate of the science," presented
at "The Role of Human Exposure Assessment in the Prevention of Environmental Disease", National Institute of
Environmental Health Sciences, Rockville, MD, Sept 22-24,1999.
91. Reed KJ. Quantification of children's hand and mouthing activities through a videotaping methodology [PhD
Thesis]. Camden, NJ:Rutgers University and UMDNJ-Robert Wood Johnson Medical School, 1998.
92. Zartarian, V.G., Ferguson, A.C., Leckie, J.O. (1998). "Quantified mouthing activity data from a four-child
pilot field study." Journal of Exposure Analysis and Environmental Epidemiology. 7(4): 543-553.
93. Zartarian VG, Ong CG, Ferguson AC, Leckie JO. Quantifying videotaped activity patterns: video translation
software and training methodologies. J Expo Anal Environ Epidemiol 7(4):535-542 (1997).
94. Melnyk LJ, Berry MR, Sheldon JJL, Freeman NCG, Pellizzari ED. Dietary exposure of children to lead.
Presented at the Analytical Challenges for Assessing Environmental Exposure to Children Symposium, ACS
National Meeting. Aug 22-26, New Orleans, LA; 1999.
95. Noland M, Danner F, Dewalt K, McFadden M, Kotchen JM. The measurement of physical activity in young
children. Res Q Exerc Sport 61:146-153 (1990).
96. Kissel JC, Richter KY, Fenske RA. Field measurement of dermal soil loading attributable to various activities:
implications for exposure assessment. Risk Anal 16:115-125 (1996).
97. Wilson NK, Chuang, JC, Lyu, C. Total PAH exposures of nine preschool children. Presented at the 17*
International Symposium on Polycyclic Aromatic Compounds. Bordeaux, France, October 25-29, 1999.
98. Wilson NK, Morgan MK. Urinary biomarkers of exposure of several preschool children to pentachlorophenol,
chlorpyrifos, 2,4-dichlorophenoxyacetic acid, and polycyclic aromatic hydrocarbons. Presented at the EPA/NIEHS
Workshop on Applying Biomarker Research in Research Triangle Park, NC, August 30-31,1999.
99. Chuang, JC, Callahan, PJ, Lyu, CW, and Wison, NK, (1999) Polycyclic aromatic hydrocarbon exposures of
children in low-income families. Journal of Exposure Analysis and Environmental Epidemiology 9(2): 85-98.
100. Wilson NK, Chuang, JC, Lyu, C. Multimedia concentrations of PAH in several day care centers. Polycyclic
Aromatic Compounds (In press).
101. Lewis, RG, Fortmann, RC, and Camann, DE. 1994. Evaluation of methods for monitoring the potential
exposure of small children to pesticides in the residential environment. Archives of Environmental Contamination
and Toxicology 26:37-46.
102. Lu A, Fenske RA, Touchstone JA, Moat T, Keden G, Knutson D. Characterization of children's exposure to
organophosphorus pesticides in rural and urban communities. Presented at the Analytical Challenges for Assessing
Environmental Exposure to Children Symposium, ACS National Meeting. Aug 22-26, New Orleans, LA; 1999.
103. Sexton, K, Kleffman, DE, Callahan, MA, (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.
104. Melnyk LJ, Berry MR, Sheldon LS. (1997) Monitoring Dietary Exposure from Pesticide Application on
Farms in the Agricultural Health Pilot Study. Journal of Exposure Analysis and Environmental Epidemiology. 7,
(1): 61-80.
105. Streicher J and Mage,DT, (1997) "Pesticide Applicator Inhalation Exposure in the Agricultural Health Pilot
Study", Presented at the7th Annual ISEA Meeting in Research Triangle Park, NC, November 1997.
G-30
-------
106. Camann, DE, Akland, GG, Bond AE, and Mage, DT, (1997) "Carpet Dust and Pesticide Exposure of Farm
Children", Presented at the 7th Annual ISEA Meeting in Research Triangle Park, NC, November 1997.
G-31
-------
a
1
1
ildren's Aggregate E>
6
g
f£5
1
Table 6. Summary of A
References
r play Wilson ct al. (197)
and at Wilson and Morgan (98)
ling
l||
1^1
Exposure Data
Indoor air, outdoor air, food and beverages, indoor dust,
area soil, handwipes and urine samples were collected be
day care center and analyzed for persistent organic pollu
20 target PAHs and several pesticides.
Participants
9 preschoolers ages 2-5
pilot study
c
J
Study
Children's Exposure to Pei
Organic Pollutants
nples Chuang et al. (99)
k«
5
- 3
2-a
S %
|sf
S.&
Environmental and biological samples to account for all
Indoor air, outdoor air, housedust, surface wipes, handw
samples collected and analyzed for selected pesticides.
Children 1 to 5
Number unknown
2"s
Total OP pesticide Exposu
Among Children in Rural i
Urban Environments
ate Pellizzari et a\:(46);
Sexton et al. (103)
u w
11
» to
1 °r
Indoor air, outdoor air, housedust, soil, dislodgeable resi
diet, and urine samples collected and analyzed for VOCs
metals, and PAHs.
Children older than 8
NHEXAS
*Quackenboss et al. (61)
o
l-g
8 %
>- T3
ft) O
!3 ^
Indoor air, outdoor air, water, housedust, soil, dislodgeal
handwipe, duplicate diet, urine, and blood samples colle
analyzed for selected pesticides
100 children
3 to 12 years
3
Children's Pesticide Expos
Study
'ipe, Melnyk et al. (104);
Streicher et al. (105);
Camann et al. (106)
*
"M
5 •»
*l
3 .£•
T3 n
Indoor air, outdoor air, housedust, soil, dislodgeable resi
duplicate diet, blood, and urine samples collected and ar
selected pesticides
U
_z
Farm workers, spouses,
children. Six farms in 1
and IA, pilot study
Agricultural Health Pilot
Study
5 for * Recently funded
study. K. Sexton is the PI
'5
A\
Outdoor, in-home, in-school, personal, and human tissu<
VOCs, metals, ETS, PAHs, and pesticides.
° 8
800 children attending
elementary school in tv
low-income neighborh<
in south Minneapolis
53
•S,l
ti£
School-Based Study of Co
Environmental Exposures
Related Health Effects in (
* M. Lebowitz is the PI
t-
fi S2
•8.2
•3.5
e -=
cs .£
Indoor air, surfaces, housedust, hands, and other media i
pyrethroids and OPs. Blood sampled for cholinesterase
!"§
100-300 children, prim
low income Hispanic a
S
en
fli
Exposure of Children to Pi
in- Yuma County, Arizona
te at http://www.epa.gov/ncerqa
ranee webs
Dnal Center for Environmental Research and Quality Assu
'i
Cocopah
be found on US EPA's
O
CO
^
CO
O
* Abstracts describing thes
G-32
-------
r-
VO
C3 w5
o §
to T3
•S p
o\
CO
•*
VO
C B
DCO^->OCN
COfNCSCNcNcMtSfN
"-• o\
f>J co
CD
I
z
JJ
XI
1
&
i
k- OO
e» -i
"7* ts
S,
0
S
o — ro
G-33
-------
.2
o
"I ^x-N^^x-s^.^x-s^^ ^
> in \o \o co" oo & S\ f^ c? vo" ?? ^
_. \ovof-r~f-oot--f~oor--oot--
,_fa \_X Ni^S N_X NWS NwX VM/ VvX S_X S^/ NwX v3^ \_x
pc\ino\oqoqoo-^—5-Hc4 —< H -H o5 cs
.hi to
> 3 '^ ^-V X-X ^-v ^^ ^-N X-V /—V ,-V ^-V
S § &-^-N X-N •* o1 S\" ^ o" CT S" ^ S~
o -c Ci G 2^ ^ Ci 52- S3 S> S- S3 G. S
o® E-i "o t^ooc'ioocsioc^m'-jirjirjio
"5 A Z<2° cococoescocJcs—<-H-M-^—<
ao H § o +1 -H +1 +1 -H +1 -H -H +1 -H -H +1
S.S la's00 in T^ c>> t~; o\ •* oq ro'es c; c\ o\
ot; 2>5ra cncnvdio-^ininsdvdvdviin
'C o O
| 2 I
*a 1 «
C *o ^\
IS i ^^^^^^^^^^^^
_,o is oncor~-oooo<-HCSO«nooc\ro
nsg §g —<—<—J-^^csc>)(Ncscs—5o5
•go "3 o -H-H-H-H-H-H-H-H-H-H-H-H
•^ eS 9 "*~ •ovOvp'^cnp-;O\
rt D-j, O"« -^—3(Nc4c4c>i«so5oqc>cncopinvqespoj
^ e-^ c\o\r~odr-:\dvdviin. o -
H <«O o—
-------
i
2
o
"8
•*-»
1
en
•s ^^^^^^^^^ :
-S 6;o\r--r-:^^oo'o'>o'5:'f:^P1
•aSJp S-d'G-CiG'C'C'S'S'S/S,'*
SS° -*ovqoor--ooc\cf>oo>r!—•
^ceJI^ffi oocJocScJcJo—5-5-^ -J
Z
l&i LH Q '*^*» ^.^ ^.^.^ ^_^ ^.^^ ^^^
O_.e;s S^—'O^—'csin'c-r^-'f'T'cn'S'io'
ttJ"5s« S-S-S-SSSS'sSSSSSSSS-C^S'
">^ij« *~;lx?~*v£J1/"5rr!cr?1/"}t"":-Hl/">'-!
ZS>7!.° —<—/_x/_>/_sx_x^_v/_x^^.^_^/_^
" co e3\OOC\O\^-Oiir>OO lO^ O? OO~ f--"
W S G- S- G- 5G- |o^rtncnmtNcnt>i
S J2118 C*l°>l/^v4?vc;c?o>'^pr-;r^o\
O OnO T^cricac^o^fvi^fvj^j^^^
ffi
£
5* ^~ • •
^_l u t^^~,/-v^-«^-,^-N/—1X-V^«,^^/_SX_X
IT1 OrS •*\Oflt-riC>>'!l-C\O-H?)'cn1i/TV
£2 S * S^C.C-G-G'IC-S^C'SG'S-^.Sl.
H ^*2 Ttuivi'!i:viir>'*<^:'<:i:|'i-*'*
M COO OOOCJCSOCSOCJOOCJ
§ g- g, g.
y OSC^OO^OOO'oo'x-vG'.-s^vx-N
<. O\C3\-HON—HO\O\O>c3\vO^-Tj-
»5J _^ ^-' **—^ ^-^ ^^' '*—' ^—' v*-' c^ s—/ OS C7t O\
n,« \o—<Ti-o>~'x-''N—'
KZ citN-S-H'cscso^of^^^
JJ.^ -H-H-H—..X-J-HCS—1O\OO\
00 O
i?'^,^^^.^^^^^^^
St^-c^v^cnvi'^'m-H&vo S"
S' S, S- S, £, S- ®* °^ ®* °* °° °°
•Ja CNiOfOOJ-^-^—
<«2> o—<
-------
§
1
[ethod Used
S
**3
"3
S
!".
52
1
i
1
"o
.£
"5
™
s.
8
W
5
'?
CO
S3
,
XI
e
_g
'•£3
C3
deotape observ
'>
i
o
o
8
CA
£
'•£
1
a
I
E
CO
minute interval
>o
J^
X)
c
_o
CiJ
o
3
Q-i
0
u
deotape observ
isearchers with
•?
2:
B
frt
anslation softw
IM
o
"5
a
•g
O
1
Q
at
n,
uthing duration
1
o
*1*
1
i
X?
O
6
i
cs
.n
•—
c$
1
JS
3
0
,'.
U
u
i?
o
51
§
3
er
«g
^
•a
tj
.u
9
o
1
.c
•g
cc
O
U
j:
3
4
•L
B
S
1"
U
1
f
¥
P
i
T3
J=
i
c
B
S1
9
o
1
JS
1
c
0
1
T3
T3
CQ
1
U
3
cr
eg'
2 22
object-to-mouth c
frequency and du
•z!
2
o
I
•o
1
S
I
o
CO
^
CO
•§
C3
"i
i
M
cs
o
f>
CO
2
'co
1
S -S
4'55
•S o o
3 tN — i
u
"2
"I
U
1
1
'&
as
C-J
P
S2
C3
I
co
0
en
03
3
CO
(U
I
ce
c •*-*
g u
b %i
O O
G^rf O
eu S
at o
S
CD
a
PH
"S
w
"8
c
CS
Zartari
G-36
-------
ital Pollutants
c
CO
O
ex
&
CO
(D
1
IS
u
'o
8
'i
*• c
c s
S 0
!!
CO 0
.2
cu
IS
i
t
- Concentrations of methyl mercury in fish
- Fish consumption rates
1
Sa
5
.c
•a ^
S CO
||
1 1
- Resulting concentrations in mother's milk
- Consumption rates of mother's milk
CO
IS
- Concentrations of lead in dust, soil, paint c
c
0
1
&
53
.2
•3
1
CO
CU
'£
O
a
' C3
o-
'5
CO
to
O
•s
00
i|
CO S
u O
(11 *T-1
- Activity patterns (mouthing behavior, fingi
eating behavior, hand washing, out<
- Nutritional status
of exposure)
c
CO
CO
53
H
CO
C
1
1
•a
o
m
S
- Concentrations of chloroform in water
- Bathing, showering, and swimming activiti
•5
e
Inhalatio
Dermal c
&"
C3
f
CO
O
&.
X
J~
- Breath concentrations (direct assessment o
C
o
'i
.1
£>
Non-diet
cposure media
f exposure)
S o
- Pesticide use patterns
- Concentrations of pesticides in all relevant
- Activity patterns
- Biomarkers of exposure (direct assessment
c
_o
c 1
o *j eo
•S o c
CO 53 — *
& S £*
n s o to
•^•S^ .S
j- CO 2 T3
III!
CO
- U
- to •£
^ ^ o
s-l"-
|i|
TO j5 U
5t f-S1
• " fc2 ®
'S .3 g
r s °
O C* *4M
o — ^
PH 0, M
A S
•P* _M
T) ^
^ O
c2
o
5
S
G-37
-------
-------
APPENDIX H
CHANGES IN CHILDREN'S EXPOSURE AS A FUNCTION OF AGE AND
THE RELEVANCE OF AGE DEFINITIONS FOR EXPOSURE AND RISK
ASSESSMENT (PAPER DISTRIBUTED AT WORKSHOP BY KIMBERLY
THOMPSON)
-------
-------
DRAFT-DONOTCITEORQUOTE
Changes in Children's Exposure as a Function of Age and the Relevance of
Age Definitions for Exposure and Risk Assessment
Draft 7/19/00
Kimberly M. Thompson, Harvard Center for Risk Analysis
1. Introduction
During the past decade, improving the lives of children has emerged as a priority on the
national agenda. In the public and environmental health arena, this priority has been reflected in
changes to statutory requirements (e.g., the 1996 Food Quality Protection Act and the Safe
Drinking Water Act) and in President Clinton's Executive Order 13045. This Executive Order
requires federal agencies to ensure that their "policies, programs, activities, and standards address
disproportionate risks to children that result from environmental health risks or safety risks"
(EO13045,1997). The Food Quality Protection Act creates a demand for estimating aggregate
exposure (i.e., estimating exposure from multiple pathways for the same substance) and
cumulative risk (i.e., assessing the risk of all substances that act with same mechanism of toxicity
over all the multiple pathways in which they may act).
The focus on children's health raises many challenges for exposure and risk analysts.
Children are in a distinct phase of human life with unique characteristics that distinguish them
from adults. From birth to adulthood their physiology and behavior are constantly evolving,
making them a "moving target" for exposure and risk assessment. This raises a number of issues:
• How should the age-related changes in children's behavior and physiology be considered
when assessing children's exposure to environmental contaminants?
• What is the most appropriate way to categorize the available data into age groups when
assessing children's exposure?
• Given the rapid change in modern society, how representative are data from previous
studies for today's children?
• What it is the most appropriate way to estimate childhood exposure given the limitations
in currently available exposure information?
• To what extent is further research needed to provide the data necessary for estimating
children's exposure? What short-term studies or longer-term research are needed to
provide the missing data?
This issue paper has been prepared to stimulate discussion on these issues, with a
particular focus on age-related anatomical and behavioral changes in children (changes in
pharmacokinetics and pharmacodynamics are not covered). The paper synthesizes the most
current and relevant information regarding children's anatomical and behavioral changes and
discusses their value in exposure and risk assessment. The paper is organized as follows:
• Section 2 reviews some of the key issues regarding children's exposure and risk
assessment.
H-3
-------
DRAFT-DO NOT CITE OR QUOTE
• Section 3 presents a series of equations, developed Hubal et al. (2000) that provide a
useful approach for estimating exposure in children. These equations utilize a
number of exposure factors, including child-based exposure factors concerning
physiology and behavior, as well as environmental factors, such as the concentration
of contaminant to which a child may be exposed. A Child-Specific Exposure Factors
Handbook (CSEFH) currently being developed by the U.S. Environmental Protection
Agency (EPA) will provide additional information to analysts about the inputs in
these equations. (Note that the handbook will offer recommended values for exposure
factors based on existing data, but will not specify exposure factors as a function of
particular ages or age ranges.)
• Sections 4 through 8 of this paper synthesize and discuss the available data (as
provided in EPA's Child-Specific Exposure Factors Handbook) for each of the child-
based exposure factors utilized in Hubal et al.'s equations.
• Section 4 discusses on anatomical changes that occur during growth (i.e., body
weight and skin surface area).
• Sections 5,6,7, and 8 discuss behavioral factors related to ingestion (food intake,
drinking water consumption, breast milk, fish consumption, soil ingestion, and
other non-dietary exposure factors); inhalation; dermal exposure; and time-
activity patterns, respectively.
• Finally, Section 9 characterizes the challenges and constraints that analysts face when
using these data in exposure and risk assessments.
• References are listed in Section 10.
2. Key Issues for Children's Exposure
2.1 Unique Characteristics of Children
Children experience remarkable change from birth to adulthood. Two of the most
dramatic changes are rapid increases in weight and height. Figures la-Id summarize the increase
in height and weight for each gender up to age 3 and for ages 3-18 (note that these show
continuous functions fit to cross-sectional discrete data to show the continuity of growth). Figure
2 shows the changes in body proportions that occur from age 2 months to adulthood.
In addition to physical growth, children pass through numerous other physiological,
psychological, social, and behavioral phases. These phases have different duration. Figure 3
shows a typical chart for normal developmental milestones. Note that milestones for fine motor,
gross motor, language, and personal or social development are categorized separately. Also note
that these charts do not include anatomical changes such as teething that could impact children's
exposure and risk. Some phases, such as crawling and mouthing objects, are common to all (or
almost all) developing children. Other phases are common only to children with specific
characteristics (e.g., kids with fair skin), while others may depend on child-specific activity
patterns (e.g., children that swim, children that consume a lot of a particular food, teenage girls
that wear make up).
H-4
-------
DRAFT- DO NOT CITE OR QUOTE
Figure l(a): Growth chart for girls birth to 3 years.
(Source: http://www.ama-assn.org/insight/h_focus/nemours/baby/grow.htm)
B 3 6 S 12 15 IS 21 24 27 30 33 36
Vf
I
I I I I I
B 3 6 9 I2IS18212427303336
Age(mcntfis)
H-5
-------
DRAFT- DO NOT CITE OR QUOTE
Figure l(b): Growth chart for boys birth to 3 years.
(Source: http://www.ama-assn.org/insight/h_focus/nemours/baby/grow.htm)
12 IS IS 21 24 27 30 33 36
369
I I I I
12 15 IS 21 24 27 30 33 36
Ape (months)
H-6
-------
DRAFT- DO NOT CITE OR QUOTE
Figure l(c): Growth chart for girls 3 to 18 years.
(Source: http://www.ama-assn.org/insight/h_focus/nemours/baby/grow.htm)
4 56 73 S ID 11 12 13 14 15 16 1? 13
3 4 56 78 9 10 11 12 13 14 15 16 17 13
Age (years)
H-7
-------
DRAFT- DO NOT CITE OR QUOTE
Figure l(d): Growth chart for boys 3 to 18 years.
(Source: http://www.ama-assn.org/insight/h_focus/nemours/baby/grow.htm)
3 4 56 78 9 10 11 12 13 14 15 16 17 13
75
3 456 78 9 ID II 12 13 14 15 16 17 13
Age (yeans)
H-8
-------
DRAFT- DO NOT CITE OR QUOTE
Figure 2: Changes in body proportions with age from 2 months to 25 yrs (Nelson etal., 1998).
Enaffctol) 5 mo. Newborn 2yr. 6yr. IZ.yr. Z5yr.
H-9
-------
DRAFT- DO NOT CITE OR QUOTE
Figure 3: Developmental milestones as a function of age (Nelson et al., 1998).
Denver II
MONTHS
Examiner: _ Name:
Date: Birthdate:
ID No.:
9 12 15 18 24
YEARS
3.4 .5.6
s
fvctrc cteMdrm pawp
25
njcrectftiKffDawfs,
TEST BEHAVIOR
(Check boxes for 1st, 2nd, or 3rd test)
Typical 1 2 3
Yes
Interest In Surroundings
Alert
Somewhat Disinterested
Seriously Disinterested
Fsarfulness
None
Mild
Extreme
Attention Span
Appropriate
Somewhat Distractable
Very Distractable
12 15 18 24
MONTHS
3
YEARS
H-10
-------
DRAFT- DO NOT CITE OR QUOTE
Different developmental stages, milestones, and activities may have different significance
for physicians and exposure/risk assessors. For example, developmental milestones such as
talking and reading may be important to physicians but generally not to exposure/risk assessors.
Conversely everyday behaviors such as drinking a lot of water or playing outside may be
significant to exposure/risk assessors but not to physicians.
Many aspects of child development reflect continuous change, though they may not be
recorded as such. For example, physical growth is continuous even though measurements are
typically collected only at discrete points in time (e.g., at annual physical exams) and the growth
rate is not constant (e.g., growth spurts).
The developmental phase(s) or time periods that are relevant to a risk assessment depend
on the type of assessment. For lifetime cancer risks, childhood exposure is simply one
component of the entire lifetime. In contrast, when assessing acute hazards, exposure/risk
assessors may be most interested in the peak exposure for a young child over the course of an
hour or less. For some non-cancer health effects, the relevant exposure duration could be a day, a
week, a year, etc. For toxic effects that only occur if the child is exposed during a certain period
of development (e.g., during the formation of the limbs in uterd), only exposure during that
developmental window may be significant.
Adults and children can react differently to the same exposure. For example, radiation
treatment for cancer can permanently damage the child's developing central nervous system,
which can inhibit normal growth (Bearer, 1995). For exposures such as this that may cause
permanent or latent health impacts, analysts must develop approaches to characterize the impacts
of health effects at different times on the developmental trajectories of children (McCormick,
1999).
Compared to adults, children have higher daily requirements for food, water, and oxygen
per unit of body weight, and they have a higher ratio of surface area to volume. However, this
does not necessarily mean that they are more vulnerable to health impacts than adults. In fact,
their exposures and risks can be higher or lower than those experienced by adults (ILSI, 1992;
Bearer, 1995; ILSI, 1996). This is because they have different exposures, pharmacokinetics, and
pharmacodynamics than adults, and because the developmental changes during childhood can
affect the metabolism, absorption, and excretion of substances and make children more or less
vulnerable to health effects. Consequently, attention to toxicological information is critical when
characterizing risks to children. (As noted earlier, pharmacokinetics/pharmacodynamics in
children is not covered in this paper.)
2.2 Variability
Variability is a key challenge for children's exposure assessment. Children of the same
age can exhibit tremendous variability. This generally limits the extent to which fixed age ranges
can be used for assessing children's development, exposure, and risk. Nonetheless defining some
standard age ranges for children would be helpful, particularly in dealing with data gaps an
mismatches that arise in the consideration of aggregate exposure and cumulative risk. Ideally,
analysts would know everything they need to know for every child and would have good
H-ll
-------
DRAFT-DO NOT CITE OR QUOTE
estimates of the exposures that children really experience. Since this type of data is not available,
exposure/risk assessors typically use multiple data sources when assessing aggregate exposure
and cumulative risk. Because these data often have a wide array of age categories, they often do
not allow direct modeling of the aggregate exposures for children. An important challenge for
analysts is to model the child of interest as he/she develops, rather than piecing together data to
create "hypothetical" children that could not really exist (e.g., children that live 25 hours/day or
consume more food than-biologically possible). The ability to model children's exposure should
improve over time with the collection of better information.
23 Representativeness
Another challenge when assessing children's exposure is the extent to which the available
exposure data represent the population of interest (Thompson, 1999). Exposure data are
collected for a specific group of people, in a specific place, and at a specific time. They can be
used in a risk assessment only to the extent that they are sufficiently relevant to the population
being assessed hi the current tune and place. The rapid pace of social and behavioral change may
diminish the relevance of study data. For example:
• In the past decade, many fruits and vegetables that were available only seasonally now
are available virtually year round.
• Many people consume an ever-increasing percentage of food away from home.
• Diets for children, which have historically included a large amount of fresh produce
and tapwater, are. shifting to include larger amounts of processed food, bottled water,
and soft drinks.
3. Exposure Equations
Hubal et al. (2000) reviewed the factors that influence children's exposure, and discussed
the data available to characterize these factors. They defined three terms, which they used to
develop a series of equations for estimating exposure:
• A microenvironment (me) is the location a child occupies for a specified period of
time. Examples include outdoors-home lawn and indoors-home kitchen.
• A macroactivity (ma) is a highly aggregated description of what a child is doing
during a specified period of tune. Examples include playing games, watching
television, eating, running, sleeping, and crawling.
• A microactivity (mi) is a detailed description of an event that takes place during a
macroactivity. Examples include hand contact with a floor or an object and mouthing
a hand or an object.
Hubal et al. (2000) provided several equations for estimating exposure. (These are
discussed as Equations 1,2,3,4, and 6, below. Equations 5 and 7 were not developed by Hubal
et al. but they have been added because they reflect typical exposure relationships used by
analysts.)
H-12
-------
DRAFT- DO NOT CITE OR QUOTE
Equation 1: Inhalation Exposure
Inhalation exposure averaged over a day for a single microenvironment/macroactivity
(Eime/ma) (in mg/day) is defined as:
Ein
wh
IRma
"ime/ma
where
L me/ma ^ame
(1)
= the child's respiration rate representing his activity level for that macroactivity
(m3/hr)
= the time spent in that me/ma during the 24 hour.period (hr/day)
Came = the air concentration measured in the microenvironment (mg/m3)
Equation 2: Dermal Exposure (Series of Contacts with Contaminated Medium)
i
Dermal exposure can be estimated individually for each microenvironment and
macroactivity by using empirically derived transfer coefficients to aggregate the mass transfer
associated with a series of contacts with a contaminated medium (Hubal et al., 2000. Dermal
exposure averaged over a day for a single microenvironment/macroactivity (Edme/ma) (in mg/day)
is defined as: - ' • '
•'-'dme/ma
where
DTCder
T
x me/ma
Qiurf
- DTC
der me/ma
surf
(2)
: dermal transfer coefficient for the me/ma (cm2/hr)
: the time spent in that me/ma during the 24 hour period (hr/day)
•• total contaminant loading on surface (mg/cm2)
Equation 3: Dermal Exposure (Single Contact with Contaminated Medium)
Dermal exposure 'can also be modeled as a series of discrete transfers resulting from each
contact with a contaminated medium (Hubal et al., 2000). Dermal exposure averaged over a day
for each microactivity (Eder/mi) (in mg/day) can be defined as:
= TE • SA • EF • C,
'surf
(3)
where
TE
SA
EF
= transfer efficiency, fraction transferred from surface to skin (unitless)
= area of surface that is contacted (cm2/event)
= frequency of contact event over a 24-hour period (events/day)
= contaminant concentration on surface (mg/cm2)
For contaminants contacted in soil, exposure assessors may estimate the contaminant
concentration on the surface (mg/cm2) (Csurf) by using information about the dermal soil loading
on the surface (mg/cm2) (DSL) and a concentration of the contaminant in the soil (mg
contaminant/mg soil).
H-13
-------
DRAFT- DO NOT CITE OR QUOTE
Equation 4: Dietary Ingestion Exposure (Food Consumption—Complex)
Hubal et al. (2000) defined dietary ingestion exposure averaged over a day (E^,) (in
mg/food item) as the amount of exposure that results directly from the food plus the amount that
comes from the food contacting a contaminated surface i times and a child's contaminated hand j
times:
where
WT
Qood
food
S AS/F • EFS/F
Chand] (4)
SA
S/F =
Qiand =
amount of the individual food consumed (g/food item)
contaminant concentration on food item as prepared for consumption (mg/g)
transfer efficiency, fraction transferred from surface to food, may be a function of
duration of contact, moisture, surface type, etc. (unitless)
area of food item in contact with contaminated surface (cm2/event)
frequency of surface to food contact events that occur during consumption of food
item (events/food item)
contaminant loading on contacted surface (mg/cm2)
transfer efficiency, fraction transferred from hand to food (unitless)
area of food item in contact with contaminated hand (cm2/event)
frequency of hand to food contact events that occur during consumption of food
item (events/food item)
contaminant loading on child's hand (mg/cm2)
Converting this exposure to units of mg/d requires multiplying by the number of food items
consumed per day (N) (in food items/day).
Equation 5: Dietary Ingestion Exposure (Food Consumption—Simple)
Equation 4 provides a relatively sophisticated assessment of exposure from food
consumption. However, when some of the exposure factors required for Equation 4 are not
known, dietary ingestion exposure can be estimated by the following simpler traditional equation
(not from Hubal et al., 2000) in which dietary ingestion exposure averaged over a day (E^) (in
mg/day) is defined as:
where
food
(5)
= the amount of the specific food that the child consumes in a day (g/day)
= me concentration of the contaminant in the food (mg/g)
Equation 6: Non-Dietary Ingestion
Non-dietary ingestion exposure averaged over a day for each microactivity in which it
occurs (Ending/mi) (in mg/day) can be defined as:
= TE^-SA.-EF-C,
(6)
where
H-14
-------
DRAFT- DO NOT CITE OR QUOTE
= object that is mouthed (including hand)
= transfer efficiency, fraction transferred from object or hand to mouth (unitless)
= area of object or hand that is mouthed (cm2/event)
= frequency of mouthing event over a 24-hour period (events/day)
= total contaminant loading on hand or object (mg/cm2)
Estimating Total Exposure
SAX
EF
a
To estimate total exposure for an entire day or longer, exposures must be added and
averaged appropriately. For air pollutants, total exposure has traditionally meant adding the
exposures to the contaminants from the various microenvironments that the child experiences
over the course of a day. However, the appropriate dose-response relationship for the health
effect of concern will determine the appropriate dose metric, which determines the level of
aggregation and averaging required. For most risk analyses, estimating exposure typically
requires averaging over a longer time period than a day (e.g., a year or a lifetime). For this
reason, it is very important for exposure/risk assessors to recognize that short-term exposures
tend to be more variable than long-term ones. For example, the amount of daily exposure to a
contaminant on grapes will be zero (on days when no grapes are consumed) and non-zero on
another day (when grapes are consumed). Over the longer term, the average grape consumption
will be greater than zero, but less than highest daily consumption amount. Thus, over time, there
will be regression to the mean. This phenomenon must be properly accounted for in
exposure/risk assessment, but it is challenging because currently very few longitudinal data exist.
Equation 7: Estimation of Potential Dose
To estimate a potential dose (mg/kg/d) for risk assessment the results of the equations
above may need to be divided by body weight of the exposed individual (BW) or some function
of body weight:
Dose
where
E
BW
= E/BW
= exposure (mg/d)
= body weight (kg)
(7)
Note that some exposure factors (e.g., ingestion rate and skin surface area) can be
expressed as a function of body weight. When correlation exists between exposure factors this
correlation should be conserved. Equation 7 is used only in situations where body weight is not
already included in the exposure factor.
Discussion of Anatomical and Behavioral Exposure Factors
Sections 4 through 8 of this paper discuss the various anatomical and behavioral exposure
factors utilized in Equations 1 through 7. For each exposure factors, the sections:
• Describe the types of information needed in the context of exposure models
H-15
-------
DRAFT- DO NOT CITE OR QUOTE
• Assess the extent to which the data are accessible and the age categories can be
modified, and
• Discuss quantification of variability and uncertainty in the information.
Also for each exposure factor, a summary table is provided that:
• Lists the key available data sources. Note that EPA's Child-Specific Exposure
Factors Handbook (CSEFH) was the source for nearly all the data sources listed in
this paper. Each source is given "source number" in the left-hand column of the table
for identification purposes. For example, in Table 1, the first source, NHANES ffl, is
given the source number "BW(1)."
• Lists the age categories used by each source and, when available, the number of
subjects in each age group.
• Provides a general assessment of (1) the data quality based on the criteria and,
judgments given in the EPA's Child-Specific Exposure Factors Handbook, and (2)
the extent of generalization (as judged by the issue paper, author).
Following each table, a figure is provided for each exposure factor that graphically
displays the ages of the children for which data were collected by each source. Each figure is
divided into three age ranges:
• Figure (a) shows birth to 1 month (by days).
• Figure (b) shows birth to 3 years (by month).
• Figure (c) shows birth to 21 years (by year).
For each of these three age ranges, the figure provides information on the ages of children studied
by each source listed in the preceding table. Data are displayed as follows:
• The figures identify the studies using the source number provided in the left column of the
table. The reader should refer to the table for the specific source reference.
• An "x" under a specific age in the figure indicates that the study in question did report
measurements for children at that age.
• A bar is used between endpoints of a range to indicate that the study reported measurements
for children within that age range. Note when an age range is given, the available data
generally include data for the entire month or year given as the end of the age range. This
fact is reflected in the figures. For example, for body weight, one of the sources provides
data for the age range 7- 12 months. This range is shown on the corresponding figure as a bar
extending from 7 months to just before 13 months to indicate that the data cover the full
duration of twelfth month.
All figures (a-c) are shown for each factor, even when no data are available for a
particular age range. Section 9 of this paper synthesizes these figures to show the availability of
data for all of the factors and to reveal where additional information may be needed. Note that
the age categories used in the figures were selected as a convenient means of summarizing the
available data and were not intended as recommendations of appropriate age categories.
H-16
-------
DRAFT- DO NOT CITE OR QUOTE
4. Anatomical changes during growth , ,
4.1 Body weight (BW)
Body weight is critical to appropriately assessing dose (see Equation 7). Data from large
cohorts can be used to develop complete growth charts and to characterize the variability in body
weight around each age (see Figures la-Id for an example). Any age grouping is possible since
these data are continuous and they may be converted into discrete age bins. Table 1 summarizes
the age groupings provided by the data sources listed in EPA's Child-Specific Exposure Factors
Handbook. Figures 4a-c display these data by age categories.
The most extensive studies of body weights for children come from the National Center
for Health Statistics (NCHS) National Health and Nutrition Examination Survey (NHANES) n
andHI:
• NHANES n provided body weight data for children between 6 months and 19 years at
each age. Burmaster et al. (1997) reanalyzed NHANES n data and found that body
weight data distribute lognormally. (The fact that these data have been reanalyzed
suggests that they are likely to be accessible.)
• The data from NHANES EC were recently released and provide body weight data for
children and young adults between 2 months and 24 years of age. They are
represented in Table 1. These data are also publicly available. Since NHANES in
data are reported for each year of age between 1 and 17 years, combining the results
into different age ranges and quantifying variability among children of the same age
should be feasible.
The NHANES n and NHANES El could be compared to determine whether there are
significant differences that might indicate a time trend. For example, are children larger now
than children in the past? Recent advances in medical technology also allow'many more low • .
birth weight (less than 2500 g) and very low birth weight (less than 1500 g) infants to survive.
This might lead to greater variance in the weights of infants and children.
Remarkably, weight change of an individual child as a function of age and the correlation
of body weight with other exposure factors are less well studied. For example, do children born'
at the 90th weight percentile stay at the 90th percentile or even continue to be larger than the
median child? Anecdotal evidence of small babies growing up to be large adults and large babies
growing up to be small adults suggests that genetics and other factors play a role in changes of
body weight. Few longitudinal data exist concerning body weight as a function of age. This type
of data may not be very significant when analyzing chronic effects for an average child (e.g., the
median or mean). However, if the analysis focuses on a low percentile individual child (e.g., a
5th percentile child), then it may be important to factor in the tendency of regression to the mean
and to be cautious in constructing a 5th percentile time weighted average estimate of body weight
by using weights observed for 5th percentile individuals at different ages.
H-17
-------
DRAFT- DO NOT CITE OR QUOTE
Table 1: Key Body Weight Data Sources and Age Categories Used
Exposure
Variable
(Source
number)
BW(1)
BW(2)
Description
Body weight
Body weight
Data Sources
NHANESm,
2000 (NCHS as
reported in
CSEFH)
Hamill et al.,
1979
(NCHS as
reported in
CSEFH)
Age groups used for
reporting data
2 mo. (n=243)
3 mo. (n=190)
2-6 mo. (n=1020)
7-12 mo. (n=1072)
1 yr. (n=1258)
2 yr. (n=1513)
3 yr. (n=1309) -
4 yr. (n=1284)
5 yr. (n=1234)
6 yr. (n=750)
7 yr. (n=736)
Syr. (n=711)
9 yr. (n=770)
10yr.(n=751)
llyr. (n=754)
12yr.(n=431)
13 yr. (n=428)
14 yr. (n=415)
15 yr. (n=378)
16 yr. (n=427)
17 yr. (n=410)
18-24 yr. (n=2532)
(n=867)
Omo.
1 mo.
3 mo.
6 mo.
9 mo.
12 mo.
18 mo.
24 mo.
30 mo.
36 mo
Quality and
extent of
generalization
Quality = High
Extent of
generalization =
High
H-18
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
BW(1)
BW(2) X
Figure 4a Summary of Available Body Weight (BW) Data by Days
Age: (Months)
BW(1)
BW(2)
012345
X X
XX X •
6789 101112131415161718192021
X
XXX X
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
X X
XXX
Figure 4b Summary of Available Body Weight (BW) Data by Months
Age: (Years) o 1 2-3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
BW(1) XXXXXXXXXXXXXXXXX
BW(2) XXX X
Figure 4c Summary of Available Body Weight (BW) Data by Years
X = Reported at specific age
HH = Reported within age range
H-19
-------
DRAFT- DO NOT CITE OR QUOTE
4.2 Skin surface area (SA and SA/BW)
Skin surface area information is most often used in dermal assessments. (Dermal
assessments also incorporate a number of behavioral factors, which are discussed in Section 7.)
Direct surface area measurements are much less common than body weight measurements. For
example, body weight and height (which correlate with skin surface area) are frequently
measured by physicians, but body surface area is rarely measured.
Instead, skin surface area is generally calculated from body weight and using
relationships based on data collected 65 years ago by Boyd (1935). Table 2 summarizes the age
groupings for surface area provided by the data sources listed in EPA's Child-Specific Exposure
Factors Handbook. Figures 5a-c display these data by age categories. Note that insufficient data
exist for children under age 2 years. For very small infants (e.g., low birth weight infants),
extrapolations must be performed to estimate surface area because Boyd's relationship for
estimating surface area from height and weight data did not include these children.
Many of the ideas discussed in Section 4.1 for the body weight factor apply to surface
area as well. Assuming that the NHANES IE data are available, the estimates for SA could be
updated to reflect these new data. Since surface area correlates with body weight, the
uncertainties about body weight estimates also affect surface area estimates.
For most assessments of dermal exposure, analysts consider the extent to which different
parts of the child's body might be exposed. Although reasonably reliable estimates for total
surface area for children over 2 years old are available, estimates of surface area associated with
specific parts of the body are less available and less reliable. This can be important when
combined with information about children's behavior. For example, consider a child wearing
shorts who sits in sand to play, or a child who is crawling and pulls his or her legs and hands over
the floor. The fact that children do these behaviors at different ages (and sizes) may impact
estimates of exposure.
H-20
-------
DRAFT- DO NOT CITE OR QUOTE
Table 2: Surface Area Data Sources and Age Categories Used
Exposure
Variable (Source
number)
SA(1)
SA/BW (1)
Description
Surface area
Surface area/
body weight ratio
Data Sources
EPA, 1985,
Using data from
NHANESH- .
separately
reported
percentiles for
both boys an4
girls (from
CSEFH)
Phillips et al.,
1993 (reported in
CSEFH)
Age groups used for
reporting data
2-3 yr. (n=?)
3-4 yr.
4-5 yr.
5-6 yr.
6-7 yr.
7-8 yr. " " "
8-9 yr.
9-10 yr.
10-1 lyr.
ll-12yr.
12-1.3 yr.
13-14 yr.
14-15 yr,,
15-16 yr.
16-17 yr.
17-18 yr.
0-2 yrs.
2.1-17.9 yrs.
Quality and
extent of
generalization
Quality = High
Extent of
generalization =
Medium (based
on extrapolation
using relatively
old data,
inadequate
information for
children under
age 2)
H-21
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
SA(1)
SA/BW(1)
Figure 5a Summary of Available Surface Area (SA) Data by Days
Age: (Months) 0123456789 101112131415161718192021222324252627282930313233343536
SA(1)
SA/BW(1)
Figure 5b Summary of Available Surface Area (SA) Data by Months
Afl«: (Years) o 1 2 3 4 5 6 7 8 9 10 11 12 13 14 IS 16 17 18 19 20 21
Figure 5c Summary of Available Surface Area (SA) Data by Years
) = Reported within age range
H-22
-------
DRAFT- DO NOT CTTE OR QUOTE
5. Changes in ingestion and mouthing behavior
Assessing exposure from ingestion is probably the most difficult of all the exposure
routes because so many things are ingested or mouthed. Initially, children consume large
amounts of breast milk or formula and then gradually their diets become increasingly more
varied. National dietary studies including the U.S. Department of Agriculture (USD A)
Nationwide Food Consumption Survey (NFCS) and Continuing Survey of Food Intakes by
Individuals (CSFII) provide a large amount of information about dietary exposure. The data
collected in these large studies are generally reported in age range categories and they are
available for reanalysis. These studies are typically cross-sectional in nature and capture
variability in the population well, at least at the time of the survey, but they do not capture
longitudinal changes in dietary consumption patterns for individual children as they grow. This
lack of data makes the assessment of lifetime aggregate exposure challenging, particularly with
respect to understanding important sources of correlation. For example: Do children that eat a
relatively large amount of grapes as 9-month olds continue to consume a lot of grapes into their
teens, or is eating a lot of grapes more characteristic of a phase or of parental concepts about
what young children should eat? Do children who consume a lot of apples also eat a lot of
pears? These types of questions are difficult to answer with the available data.
Section 5 reviews six key exposure factors concerning ingestion in children:
• Section 5.1 reviews the age categories used in the one major survey that assesses food intake
amounts.
• Section 5.2 covers drinking water ingestion, for which data are available from
several large studies.
• Sections 5.3 and 5.4 review the age category information in the ingestion exposure
databases for breast milk and fish, respectively. These databases are based on much
smaller studies in local populations.
• Finally, several small studies provide some information about non-dietary ingestion
exposures of children. These studies generally do not include national data, but
instead report the results for a small convenience sample of children studied in a
specific local area. Typically, the categories used for children in the smaller studies
represent the age ranges of the children in the study. While the data may be
accessible by contacting the researchers that conducted the study, given the relatively
small sample sizes involved reassessing them to look at different age categories is less
likely to be useful than for some of the larger national studies. Nonetheless, Sections
5.5 and 5.6 review the age category information for soil ingestion and for mouthing
non-dietary objects, respectively.
5.1 Food intake
Table 3 summarizes the age groupings for food intake provided by the U.S. Department
of Agriculture's Continuing Survey of Food Intakes by Individuals (CSFII) study — the data
source listed in EPA's Child-Specific Exposure Factors Handbook. Figures 6a-c display these
data by age categories. The USDA's CSFII study provides data from a national survey of food
consumption. The results include data for total fruits, total vegetables, total grains, total meats,
H-23
-------
DRAFT- DO NOT CITE OR QUOTE
total fish, total dairy products, and total fats. USDA collects these national data using a stratified
sampling strategy that specifically collects food consumption information from children. The
data are reported as intake rates per unit of body weight (g of food/kg of body weight/d) (i.e.,
data were collected for each individual so the reported data preserve the correlation between food
consumption and body weight). Thus, no additional calculation is needed to account for body
weight when estimating dose. The food intake factor comes from Equation 5 and it relates to the
WT factor mentioned by Hubal et al. (2000) and given in Equation 4 (when the WT is multiplied
by the number of such food items consumed per day [N] and divided by body weight [BW]).
Unfortunately the USDA data do not provide information over long time periods, or
longitudinal data for individual children. For a "typical child," there are some long-term dietary
constraints that must apply (e.g., requirements for caloric intake, sufficient vitamins and
minerals). However, the extent to which individual children meet these requirements is
unknown. Correlation of diet with socioeconomic factors may also be an important issue in the
context of the exposure assessment.
Table 3: Food Intake Data Sources and Age Categories Used
Exposure
Variable (Source
number)
Food Ingestion
ORfood/BW) (1)
Description
Food intake (for
various foods)
Data Sources
EPA, 2000.
(Data from
USDA CSFH)
(Reported in the
CSEFH)
Age groups used for
reporting data
<1 yr, (n=359)
1-2 yr. (n=1356)
3-5 yr. (n=1435)
6-llyr.(n=1432)
12-19 yr. (n=1398)
Quality and
extent of
generalization
Quality=High in
average, low in
long-term, upper
percentiles
Extent of
generalization =
Medium (Lack of
longrterm focus)
H-24
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 Z 3 4 S 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Figure 6a Summary of Available Ingestion Rate (IR) of Food Data by Days
Age: (Months) 0123456789 101112131415161718192021222324252627282930313233343536
Figure 6b Summary of Available Ingestion Rate (IR) of Food Data by Months
Age: (Years) o 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
IRM/BW (i) 1 ;
Figure 60 Summary of Available Ingestion Rate (IR) of Food Data by Years
^ = Reported within age range
H-25
-------
DRAFT- DO NOT CITE OR QUOTE
5.2 Drinking water consumption (IR^e,. or IRwater/BW)
Several large studies on drinking water intake provide good estimates of the amount of
drinking water consumed. Table 4 summarizes the age groupings for drinking water consumption
provided by the data sources listed in EPA's Child-Specific Exposure Factors Handbook. Figures
7a-c display these data by age categories.
Not surprisingly, the amount of drinking water consumed may depend on the type of
physical activity being done by the individual and on the temperature and humidity (e.g., people
might consume more water in the summer). The existing studies provide information about both
the total tap water consumed and the total fluid intake. The data from these surveys are generally
available for reanalysis and have been analyzed to characterize the variability in the population.
The data appear to distribute lognormally (Roseberry and Burmaster, 1992).
Table 4: Drinking Water Data Sources and Age Categories Used
Exposure
Variable (Source
number)
IRwaterand
HWBW (1)
IR^and
HWBW (1)
Description
Drinking water
intake for total
fluid intake
Drinking water
intake for total
fluid intake
Data Sources
EPA, 2000
(Using data from
USDA's CSFID
and Ershow and
Cantor, 1989
(Reported in
CSEFH)
EPA, 2000
(Using data from
USDA'sCSFn)
and Ershow and
Cantor, 1989
(Reported in
CSEFH)
Age groups used for
reporting data
0<1 yr. (n=359)
1-10 yr. (n=3980)
ll-19yr. (n=1641)
<0.5(n=199)
0.5-0.9 (n=160)
1-3 (n=1834)
4-6 (n=1203)
7-10 (n=943)
ll-14(n=816)
15-19 (n=825)
Quality and
extent of
generalization
Quality=High
Extent of
generalization =
High
H-26
-------
DRAFT- DO NOT CTTE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 IB 17 18 19 20 21 22 23 24 25 25 27 28 29 30
Figure 7a Summary of Available Ingestion Rate (IR) of Water Data by Days
Age: (Months) 01 23456789 101112131415161718192021222324252627282930313233343536
IR.ai.rd)1
Figure 7b Summary of Available Ingestion Rate (IR) of Water Data by Months
Age:,(Years) o 12 34 5 6 78 9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 7c Summary of Available Ingestion Rate (IR) of Water Data by Year
"Reported for IRwater and IRwa,n/BW
|—I = Reported within age range
H-27
-------
DRAFT- DO NOT CITE OR QUOTE
5.3 Breast milk
Five studies provide estimates of breast milk intake that can be used to estimate infant
exposure to substances in the milk. Table 5 summarizes the age groupings for breast milk data
from these studies. Figures 8a-c display these data by age categories. These studies included
estimates of infants up until age 1. Most of the studies have focused on quantifying milk
ingestion for young infants (under 6 months). Note that no data are available for children that are
breast-fed beyond age 1.
Information about the percentages of infants that are breast-fed is relatively sparse. One
study (NAS, 1991) provides data for the percentage of newboms being breast-fed and of 5- to 6-
month-old infants being breast-fed. To estimate a population risk (or population exposure) for
this exposure pathway, additional information about the decline in the percentage of infants that
are breast-fed may be needed. In general, these data are relatively sparse and they may not reflect
current trends for breast-feeding. Shorter postpartum hospitalization for normal deliveries and
longer infant hospitalization for very premature infants may impact the amount of breast milk
consumed and the tendency to breast-feed.
H-28
-------
DRAFT- DO NOT CITE OR QUOTE
Table 5: Breast Milk Ingestion Data Sources and Age Categories Used
Exposure
Variable (Source
number)
™breastmilk (1)
ERbreastmilk (2)
IR-breastmilk (3)
IR-breastmilk (4)
IR-breastmilk (5)
Description
Breast milk
intake
Breast milk
intake
Breast milk
intake
Breast milk
intake
Breast milk
intake
Data Sources
Pao et al., 1980
(Reported in
CSEFH)
i
Dewey and
Lonnerdal, 1983
(Reported in
CSEFH)
Butte et al, 1984
(Reported in
CSEFH)
Neville et al.,
1988 (Reported
in CSEFH)
Dewey et al.,
1991a,b
(DARLING
Study) (Reported
in CSEFH)
Age groups used for
reporting data
Completely breast-
fed
lmo.(n=ll)
3mo;(n=2)
6mb. (n=l)
Partially
Breastfed
1 mo. (n=4)
3mo.(n=ll)
6 mo. (n=6)
9 mo. (n=3)
1 mo. (ri=16)
2 mo. (n=19)
3 mo. (n=16)
4 mo. (n=13)
5 mo. (n=l 1)
6 mo. (n=ll)
1 mo. (n=37)
2 mo. (n=40)
3 mo. (n=37) ,
4 mo. (n=41)
Intake per day
Each day for days 1
toll(n=7tol2)
For days 14, 21, 28,
35, 42, 49, 56 (n=10
to 13)
For days 90, 120,
150, ...360(n=9to
13)
3 mo. (n=73)
6 mo. (n=60)
9 mo. (n=50)
12 mo. (n=42)
Quality and
extent of
generalization
Quality=Medium
Extent of
generalization =
Low (based on
small sample size
and inability to
characterize
variability)
H-29
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
(5)
Figure 8a Summary of Available Ingestion Rate (IR) of Breast Milk Data by Days
Agi: (Months) 0123456789 1011 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
m»_«-* (1)
mM>M (2)
IR.^_(3>
IR.—0-. (4)
IR^.^. (5)
xx x x
X X X X
X X X X
x x x x
Figure 8b Summary of Available Ingestion Rate (IR) of Breast Milk Data by Months
Age: (Years) Q 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 8c Summary of Available Ingestion Rate (IR) of Breast Milk Data by Years
X = Reported at specific age
H-30
-------
DRAFT-DO NOT CITE OR QUOTE
5.4 Fish consumption (IRfis,,)
Amounts of fish consumed depend on the segment of the population under consideration.
For children in some segments of the population, data about the amount and types of fish eaten
are relatively sparse. In particular, people who catch fish, either for sport or for sustenance, are
generally likely to consume more fish than those people who do not. Does this tendency translate
to greater fish consumption for their children? For those fishing for sustenance, it probably does,
but for sport fishers it may not. In either case, few data are available to answer these questions.
Because different population segments consume different types and amounts of fish, the
data for fish consumption cover several different categories of fish consumed and types of
consumers. Table 6 summarizes the age groupings for fish consumption reported in the studies
covered by EPA's Child-Specific Exposure Factors Handbook. Figures 9a-c display these data by
age categories. These studies included estimates of infants up until age 1.
While general intake data are available from USDA's large CSFn database, most of the
fish consumption data come from relatively small studies. These data are difficult to extrapolate
to the larger population and make characterization of variability a challenge. The age categories
used in the studies differ, and very little information is available at all for fish consumption by
relatively young children. These data are not as readily accessible as the data from the national
surveys, but they have been reassessed to characterize variability in some cases. For example,
Ruffle et al. (1994) fit lognormal distributions to the daily fish consumption rates obtained in the
Tuna Research Institute Survey.
H-31
-------
DRAFT- DO NOT CITE OR QUOTE
Table 6: Fish Intake Data Sources and Age Categories Used
Exposure
Variable (Source
number)
IRfchandlRfcfc/BW
(1)
»«, (2)
»*, (3)
IRfth and BE^EW
andNfch^W
Kfch(5)
N&hnxaj, (6)
Description
General intake
rates (freshwater
and estuarine,
marine, and total)
General intake
rates of fish
consumers
General intake
rates of fish
consumers
(freshwater
finfish, saltwater
finfish, and
shellfish)
Fish meals per
month for
anglers with
fishing licenses
Make rate for
Native American
fishers
General number
of fish eating
events (meals)
Data Sources
EPA, 1996 (Data
from USDA
CSFE)
(Reported in the
CSEFH)
Javitz, 1980
analysis of the
Tuna Research
Institute Survey
(Reported in the
CSEFH)
Rupp et al., 1980
of the Tuna
Research
Institute Survey
(Reported in the
CSEFH)
West et al., 1989
(Reported in
CSEFH)
Columbia River
Inter-Tribal Fish
Commission
(CRTTFC), 1994
(Reported in
CSEFH)
Tsang and
Klepeis, 1996
(Reported in
CSEFH)
Age groups used for
reporting data
14 or under
(n>1000)
15-44 yr. (n>1000)
0-9 yr.
10-19 yr.
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
IR«, OC
IR«» (2)
IR,« (3)
IR,,. <4>"
IR«, (5)
N*.™» (6)
Figure 9a Summary of Available Ingestion Rate (IR) of Fish Consumption Data by Days
Age: (Months) 0123456789 101112131415161718192021222324252627282930313233343536
. (2)
» (3)
, (4)"
« (5)
M> (6)
Figure 9b Summary of Available Ingestion Rate (IR) of Fish Consumption Data by Months
Age: (Years) o
NU^. (6)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 9c Summary of Available Ingestion Rate (IR) of Fish Consumption Data by Years
• Reported for IRfisti and IR|iSf/BW
" Reported for IR||Shl IR|lsh/BW, and N|fc|,meais
— Upper age limit not specified
(—i = Reported within age range
H-33
-------
DRAFT- DO NOT CITE OR QUOTE
5.5 Soil ingestion (IR^n)
The amount of soil ingested by children depends on whether or not the ingestion is
intentional:
• For incidental ingestion, several studies have attempted to measure soil intake
indirectly by measuring the amounts of trace elements in stool and urine samples, and
in some studies by subtracting the amounts of these elements in food (using duplicate
meals).
• Very limited exposure data are available for intentional ingestion of soil, known as
pica.
Table 7 summarizes the age groupings for soil ingestion data from the studies covered by
EPA's CSEFH. Figures lOa-c display these data by age categories. The methodology used to
estimate soil ingestion is indirect, relatively complicated, and prone to errors. In addition, the
studies reflect short-term, small local analyses that do not extrapolate easily to national, long-
term studies.
In general, soil ingestion tendencies probably vary considerably over days and
characterization of this variability is relatively limited. A significant amount of uncertainty exists
about the amount of soil ingested by children. In addition, the extent to which children ingest the
soil as a function of different microactivities is unknown. As discussed by Hubal et al. (2000)
and shown in Equation 4, one mechanism for soil ingestion is when children eat with dirt on their
hands that gets transferred to the food, or when food drops onto a dirty surface and children pick
it up and eat it. The existing soil ingestion data do not distinguish between different activities
that lead to soil ingestion, and more effort is needed to combine activity monitoring/modeling
with amounts of soil ingested.
Also, remarkably, the existing studies do not include children under age 1 year, even
though these children are likely to be in contact with the floor. More data for very young
children is needed. In addition, data for children over age 7 years are also missing. While this
behavior is likely to be reduced significantly by age 7 years, some soil ingestion may continue
beyond that age associated with outdoor play, etc.
H-34
-------
DRAFT- DO NOT CITE OR QUOTE
Table 7: Soil Ingestion Data Sources and Age Categories Used
Exposure
Variable (Source
number)
IRsoH (1)
IRsoi. (2)
IRsoH (3)
IRsoi,(4)
IRsoil(5)
IRsoi, (6)
Description
Soil intake -
non-intentional
Soil intake -
non-intentional
Soil intake -
non-intentional
Soil intake -
non-intentional
Soil intake -
non-intentional
Soil intake (pica)
Data Sources
Binder et al.,
1986 (Reported
in CSEFH)
Clausing et al.,
1987 (Reported
in CSEFH)
Calabrese et al.,
1989 (Reported
in CSEFH)
Davis et al., 1990
(Reported in
CSEFH)
Van Wijnen et
al., 1990
(Reported in
CSEFH)
Calabrese et al.,
1991 (Reported
in CSEFH)
Age groups used for
reporting data
1-3 yr. (n=65)
2-4 yr. (n=18)
1-4 yr. (n=64)
2-7 yr, (n=104)
1-5 yr. (n=292)
3.5 yr. (n=l)
Quality and
extent of
generalization
Quality=Medium
for average,
long-term central
estimates)
Extent of
generalization =
Low (All non-
national data,
short-term
studies, not all
ages of children
included)
H-35
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
IR-16)
Figure 10a Summary of Available Ingestion Rate (IR) of Soil Ingestion Data by Days
Age: (Months) 0123456789 101112131415161718192021222324252627282930313233343536
I-
Figure 10b Summary of Available Ingestion Rate (IR) of Soil Ingestion Data by Months
Age: (Years) o 1 234 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
^ (6)
Figure 10c Summary .of Available Ingestion Rate (IR) of Soil Ingestion Data by Years
X = Reported at specific age
f—| = Reported within age range
H-36
-------
DRAFT- DO NOT CITE OR QUOTE
5.6 Other non-dietary ingestion
Children may be exposed to environmental pollutants when they place non-food items
into their mouths as discussed for Equation 6. The studies regarding this behavior tend to be
divided into studies that estimate the duration of mouthing (Tmouth) and studies that estimate the
frequency of mouthing (EF). While these are related concepts, they are not the same, and slightly
different equations are needed to estimate exposure based on these different data.
Table 8 summarizes the age groupings for non^dietary ingestion data from the studies
included in EPA's CSEFH. Figures lla-c display these data by age categories In general the
link between duration of mouthing and micfoactiviiies or inacroactivities is relatively
unexplored. The studies included here represent very small non-national studies that are
typically of a short-duration. The mouthing duration data collected by Juberg et al. (2000)
suggests that longitudinal studies of mouthing duration are needed because regression to the
mean does occur and children's mouthing duration of objects does decrease over the first 3 years.
No studies provide information about mouthing behavior for children over age 6. While this
behavior is likely to be reduced significantly by age 6 years, some mouthing of objects may
continue beyond that age associated with outdoor play, cigarettes, etc.
Table 8: Non-dietary Ingestion Data Sources and Age Categories Used
Exposure
Variable (Source
number)
Tmouth (1)
Tm0uth (2)
Tmouth (3)
EF(1)
EF(2)
Description
Duration of
mouthing
Duration of
mouthing
Duration of
mouthing
Frequency
of mouth
contact
Frequency
of mouth
contact
Data Sources
Groot et al., 1998
(Reported in
CSEFH)
EPA analysis of
Davis et al., 1995
(Reported in
CSEFH)
Juberg et al., 2000
Reed etal., 1999
(Reported in
CSEFH)
Zartarian et al.,
1997
(Reported in
CSEFH)
Age groups used
for reporting data
3-6 mo. (n=5)
6-12 mo. (n=14)
12-18 mo. (n=12)
18-36 mo. (n=ll)
10-60 mo. (n=92)
0-18 mo. (n=275)
19-36 mo. (n=l 10)
2-6 yr. (n=30)
2.5-4.2 yr. (n=4)
Quality and extent of
generalization
QwSty=? (North
CSEFH)
Extent of generalization
= Medium for average
for young children, low
for long-term central
estimates and for all
extremes
H-37
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
EF{1)
EF(2)
Figure 11a Summary of Available Other Non-Dietary Ingestion Data by Days
Age: (Months) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
EF(1)
EF(2)
Figure 11b Summary of Available Other Non-Dietary Ingestion Data by Days
Ago: (Years) o 1 2 ' 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 11 c Summary of Available Other Non-Dietary Ingestion Data by Days
|—i = Repotted within age range
H-38
-------
DRAFT- DO NOT CITE OR QUOTE
6. Inhalation
A number of studies provide data on inhalation rates for children. Inhalation rates vary as
a function of activity (i.e., higher than average ventilation rate when exercising, lower when
sleeping), and estimates are available for several different types of activities for both healthy and
asthmatic children. Table 9 summarizes the age groupings for inhalation data from the studies
covered by EPA's handbook. Figures 12a-c display these data by age categories.
The estimates by Layton et al. (1993) rely on estimating inhalation rates based on food
energy intake and on basal metabolic rate. Since dietary data are available for large numbers of
people, the sample sizes possible with this approach can be very large. However, since a model
is required to go from the food intake to the inhalation rate estimate, error may be introduced into
the estimates associated with the model. The fact that Layton et al. (1993) found similar
estimates of inhalation rates using different data and different modeling approaches is reassuring.
Nonetheless, some uncertainties remain about the results. In addition, different age categories
were used in the different approaches due to differences in the age categories used in the input
data for the models.
One challenge in using the inhalation rate data is the need to characterize the daily
activities to obtain good estimates of the average daily inhalation rate. Exposure and risk
assessors typically want to know the inhalation rate over a longer time period than simply during
an activity, so some time/activity weighting is needed to meet the needs of risk analysts.
Remarkably, none of the studies report inhalation rate as a function of body weight or address
their correlation. Studies with longitudinal data on inhalation rates are missing and additional
studies are needed to better characterize inhalation rates of children as a function of age and to
estimate their average inhalation rates.
H-39
-------
DRAFT- DO NOT CITE OR QUOTE
Table 9: Inhalation Rate Data Sources and Age Categories Used
Exposure
Variable (Source
number)
Kma CD
K«<3>
»«
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
R~ (D
FU (2)
FU (3)
FU,(4>
R~ (5)
Figure 12a Summary of Available Inhalation Rate Data by Days
Age: (Months) 0123456789 101112131415161718192021222324252627282930313233343536
FU (2)
f- (3)
FU (4),
R» (5)
Figure 12b Summary of Available Inhalation Rate Data by Months
Age: (Years) o 1
1R» (D
IR™(1)
IR~ (2)
IR™ (3)
2 3 4 5 5 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 12c Summary of Available Inhalation Rate Data by Years
|—I = Reported within age range
H-41
-------
DRAFT- DO NOT CITE OR QUOTE
7. Dermal contact
In contrast to many other exposure factors for which daily rates are common and
generally available, the available data for dermal contact are primarily activity-based. Most of
the existing data focus on dermal exposure from contaminated soil. The EPA's CSEFH does not
recommend using a single value for soil adherence. Instead the Agency recommends that
analysts find the activity that is most similar to the one of interest and estimate the dermal soil
loading based on this.
Table 10 summarizes the age groupings for inhalation data from the studies covered by
EPA's CSEFH. Figures 13a-c display these data by age categories. Kissel et al. (1996a; 1996b)
and Holmes et al. (1996) provided data for the rate of soil adherence to the skin as a function of
different activities based on controlled experiments. The relatively small database leaves a high
degree of uncertainty in estimation of dermal exposure and provides only a limited ability to
characterize variability in the population. In addition, the observations made in the field studies
may not be fully representative of actual activities that occur. As a result of the design of these
studies, the age ranges reported reflect the age ranges of the participating subjects.
In some analyses of dermal exposure, knowing the surface area of the body or part of the
body may be necessary. The age categories for surface area data are provided in Section 4.2.
One factor that may impact the amount of skin in contact with contaminants is the
amount of clothing worn by the individual for various activities. Seasonal variations are likely to
impact both the activities and the clothing worn, and data that account for this correlation are not
currently available for children. Kissel et al. (1996) and Holmes et al. (1996) do note the types of
clothing worn by participants and the month of the data collection. Reanalysis of the existing
data might be possible by contacting the study authors, but it is probably of limited use given the
very small sample sizes.
Given the limitations in the number of microenvironments and activities for which data
are available, additional data are needed to better characterize dermal contact.
H-42
-------
DRAFT- DO NOT CITE OR QUOTE
Table 10: Soil Adherence Data Sources and Age Categories Used
Exposure
Variable (Source
number)
DSL (1)
DSL (1)
DSL (1)
DSL (1)
DSL (1)
DSL (1)
DSL (1)
Description
Soil Adherence
from 1.5 hours of
indoor
Tae Kwon Do
Soil Adherence
from 2 hours of
indoor play on
carpeted floor
Soil Adherence
from indoor and
outdoor exposure
during daycare
(four groups of
children in day
care 3.5, 4, 8,
and 8 hours,
respectively)
Soil Adherence
from 0.67 hours
of outdoor soccer
Soil Adherence
from 4 hours of
outdoor
gardening
Soil Adherence
from 11.5 hours
of archeological
work
Soil Adherence
from kids playing
in mud (2 times
for 0.17 and 0.33
hours,
respectively)
Data Sources
Kissel et al.
(1996b), Holmes
et al. (1996)
(Reported in
CSEFH)
Kissel et al.
(1996b), Holmes
et al. (1996)
(Reported in
CSEFH)
Kissel et al.
(1996b), Holmes
et al. (1996)
(Reported in
CSEFH)
Kissel et al.
(1996b), Holmes
etal.(1996)
(Reported in
CSEFH)
Kissel et al.
(1996b), Holmes
et al. (1996)
(Reported in
CSEFH)
Kissel et al.
(1996b), Holmes
et al. (1996)
(Reported in
CSEFH)
Kissel et al.
(1996b), Holmes
et al. (1996)
(Reported in
CSEFH)
Age groups used for
reporting data
8-42 yrs. (n=7)
6-13 yrs. (n=4)
3-13 yrs. (n=6)
1-6.5 yrs. (n=6)
1-6.5 yrs. (n=6)
1-4 yrs. (n=5)
1-4.5 yrs. (n=4)
13-15 (n=8)
16-35 (n=8)
16-35 (n=7)
9-14 (n=6)
Quality and extent
of generalization
Quality= Low
Extent of
generalization =
Low (Data are very
limited)
H-43
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
DSL (1)
DSHD
DSL(1)
DSL(1)
DSL(1)
DSL<1)
DSL(1)
DSL(1)
DSL(1)
DSL(1)
Figure 13a Summary of Available Dermal Contact Data by Days
Age: (Months) 0123456789 101112131415161718192021222324252627282930313233343536
OSL(1)
DSHD
DSHD
DSHD
DSHD
DSU (1)
DSHD
DSHD
DSL(1)
DSHD
Figure 13b Summary of Available Dermal Contact Data by Months
Age: (Years) o 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 13c Summary of Available Dermal Contact Data by Years
|—4 = Reported within age range
H-44
-------
DRAFT- DO NOT CITE OR QUOTE
8. Time/activity patterns (Tme/ma)
Several national and local studies provide data on how children of various ages spend
their time in various microenvironments and on various major activities. Table 11 summarizes
the age groupings for time/activity pattern data from the studies covered by EPA's CSEFH.
Figures 14a-c display these data by age categories.
These data have been reanalyzed to assess the variability in the population for some
factors (e.g., Funk et'aL, 1998 reanalyzed data collected by the California Air Resources Board,
as did Hubal et al., 2000). The fact that the data listed below are not independent should be taken
into consideration, but it does provide evidence of the availability of the existing data and the
different age categories that have and can be used.
One issue associated with time/activity is the need to meet the constraint of 24 hours/day
when combining time/activity data. This issue has not been thoroughly addressed in the context
of children's exposure estimates or in the characterization of variability in time/activity patterns.
In particular, if times spent tend to distribute lognormally, then the means will exceed the
medians and adding the mean values could lead to average time/activity estimates that exceed 24
hours/day. Analysts need to develop appropriate methods for dealing with this issue.
Similar to other factors discussed in this issue paper, no longitudinal studies exist and all
of the time/activity data requires extrapolation from short-term to long-term, which suggests the
need for additional study.
H-45
-------
DRAFT- DO NOT CITE OR QUOTE
Table 11: Time/Activity Pattern Data and Age Categories Used
Exposure
Variable
Tmefo]a(l)
1 me/ma W)
Tme/ma (3)
•*• me/ma v*v
Tnw/ma (5)
Tmefaa(6)
Description
Average time
spent for major
activities
Average time
spent for major
activities
Average time
spent indoors,
outdoors, and in-
vehicle and in
various activities
Average time
spent in various
micro-
environments
Average time
spent in various
major activities
Average time
spent at home
and away from
home and level
of activity with
respect to
inhalation
exposure
Data Sources
Timmer et al.,
1985 (Reported
in CSEFH)
Timmer et al.,
1985 (Reported
in CSEFH)
Robinson and
Thomas, 1991
compared data
from the
California Air
Resources Board
study and a
national study
(Reported in
CSEFH)
Robinson and
Thomas, 1991
(Reported in
CSEFH)
Wiley et al.,
1991 (Reported
in CSEFH)
Funk et al, 1998
(Reported in
CSEFH)
Age groups used for
reporting data
3-11 yrs. (n=229)
12-17 (n=160)
3-5 yrs.
6-8 yrs.
9-11 yrs.
12-14 yrs.
15-17 yrs.
<12 yrs.
12-17 yrs. (n=183)
18-24 yrs. (n=250)
0-2 yrs (n=313)
3-5 yrs.(n=302)
6-8 yrs (n=269)
9-11 yrs. (n=316)
6-8 yrs. (n=269)
9-llyrs. (n=316)
12-17 yrs. (n=183)
Quality and
extent of
generalization
Quality=
Medium
Extent of
generalization =
Medium (data
are limited for
infrequent
activities, but do
allow good
characterization
of major
activities)
H-46
-------
DRAFT- DO NOT CITE OR QUOTE
Tme/ma (/)
Tme/ma (8)
Tme/ma(")
Average time
spent in various
activities
Average time
spent indoors and
outdoors at home
and away from
home
Average time
spent showering,
in the bath, or in
the bathroom
Hubal et al.,
2000 (Reported
in CSEFH)
-
Davis et al., 1995
(Reported in
CSEFH)
Tsang and
Kleipis, 1996
(Reported in
CSEFH)
0 (n=199)
1 (n=238)
2 (n=264)
3 (n=242)
4 (n=232)
5 (n=227)
6 (n=199)
7 (n=213)
8 (n=226)
9 (n=195)
10 (n=199)
11 (n=206)
10-60 mo. (n=92)
1-4 yrs (n=40)
5-11 yrs.(n=139)
12-17 yrs (n=268)
H-47
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
T - (1)
T - (4)
Figure 14a Summary of Available Time/Activity Data by Days
Aga: (Months) o
T_~ (1)
T (2)
T (3)
T~».(4)
T (6)
Tw~(7) x
T^.(8)
TW~(9)
123456789 101112131415161718192021222324252627282930313233343536
X X X
Figure 14b Summary of Available Time/Activity Data by Months
Age
T
T _
TM.««
: (Years) o 1 2.3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 14c Summary of Available Time/Activity Data by Years
X = Reported at specific age
f—| = Reported within age range
H-48
-------
DRAFT- DO NOT CITE OR QUOTE
9. Discussion . ,
9.1 Synthesis and observations
As demonstrated in Sections 4 to 8, substantial exposure information is available, but
significant gaps in our knowledge still remain. The lack of data in key areas and the
representativeness of those data that are available pose key challenges to exposure and risk
assessors for children's health, as discussed below.
Representativeness of data
For all the exposure factors discussed, the representativeness of the available data to the
individual, population, temporal, and/or spatial scale of interest is an ongoing issue. Not
surprisingly, we know the most about children's observable anatomical characteristics, such as
body weight and height, and we know the least about less easily observable behavioral
characteristics such as where and how they spend their time or how much soil they ingest.
Nonetheless, even in the area of anatomical development where we have substantially more data,
issues of representativeness pose challenges for exposure and risk assessors. For example, we
have limited information about how well measurements collected for today's children will
represent children of future generations. Further, with advances in medical technologies,
significant numbers of low birth weight babies (less than 2,500 g) and very low birth weight
babies (between 1,000-2,500 g) now survive. Essentially no exposure assessment information
exists for these children.
Extrapolation from today's children to future generations also raises challenges for
exposure and risk assessors in the context of behavioral changes. For example, eating habits and,
practices have changed so dramatically that diary studies of eating habits from 10 years ago
might not mention foods that children eat today (e.g., new breakfast cereals, exotic foods). In
addition, with the increased globalization of trade, today's children can eat "seasonal" fruits and
vegetables nearly all year.
Lack of Data
With respect to using a microenvironment, macroactivity, and microactivity approach in
exposure models, the data are somewhat limited in some contexts to support these efforts. Table
12 summarizes the different factors discussed in Sections 4 through 8 and indicates which of the
exposure equations listed in Section 3 use that factor. Note that overall body surface area (SA),
the number of fish meals (Nfishmeals), and the duration of mouthing of objects (Tmouth) are not listed
in any of Equations 1 through 7 but they do appear in the EPA's Child-Specific Exposure Factors
Handbook. Using these factors requires modification of the exposure equations or modification
of the factor to be consistent with the equations.
Table 13 lists the exposure factors, which were not discussed in Sections 4 through 8, but
which are required in Equations 1 through 7. Note that equations 3 and 4 in particular require
many of these factors. These factors indicate some of the challenges that analysts will face in
using Equations 3 and 4 given the existing data. The presence of the transfer coefficients (DTC
H-49
-------
DRAFT- DO NOT CITE OR QUOTE
and TEs) in Table 13 is not surprising because, like concentrations, these values are often
contaminant-specific and medium-specific. They may also depend on the contact mechanics (for
which good information about children's activities are needed) and on skin characteristics (which
may vary with age). Comparing Tables 12 and 13 suggests the presence of data gaps in the
context of assessing dermal exposure. This comparison also clearly shows that very few data
exist related to non-dietary intake. In particular, how do children's handling of food impact their
exposures, what is the amount of a contaminant that can be ingested from food items retrieved
from the floor, and in what contexts do these exposures matter?
Table 12: Summary of Physiological and Behavioral Factors Discussed in Sections 4
through 8
Factor
BW
SA
SA/BW
IRfcxv/BW
JRwaler
JR^JBW
J-^brea-rtmilk
Elfish
EWBW
Nfi,h me,,),
3R^n
mouth
Ji*JL mou llii TIP
IR™
DSL
T™.,m,
Quality
H
H
H/L
H
M
H/L
H/L
M
M
L
M
Extent of
generalization
H
M
M
H
L
M
M
L
MIL
M/L
M
L
M
"X" Denotes Used in Equation Number
1
X
X
2
X
3
X
4
5
X
X
X
X
X
6
X
7
X
t .
H=high, M=medium, L=low, H/L=high for average/low for long-term and upper-percentiles
M/L= medium for average/low for long-term and upper-percentiles
H-50
-------
DRAFT- DO NOT CITE OR QUOTE
Table 13: Summary of Other Exposure Factors
Factor
DTCder
FF
'-''• dermal
•I ^dermal
^"dermal
WT
TEs/F
EFS/F
SAg/p
TEjvF
EPH/p
oA.H/p
"X" Denotes Used in Equation Number
1
2
X
3
X
X
X
4
X
X
X
X
X
X
X
5
6
7
Key of symbols:
BW = body weight (kg)
= area of surface that is contacted (cm2/event)
= area of food item in contact with contaminated surface (cm2/event)
= area of food item in contact with contaminated hand (cm2/event)
= area of object x or hand that is mouthed (cm2/event)
= the amount of the specific food that the child consumes in a day (g/day) (general
SA
SAS/F
SAjj/p
SAV
^food
category includes breast milk, drinking water, fish, etc.)
W
TE
TES/F
EF
EFS/F
= amount of the individual food consumed (g/food item)
= transfer efficiency, fraction transferred from surface to skin (unitless)
= transfer efficiency, fraction transferred from surface to food (unitless)
= transfer efficiency, fraction transferred from hand to food (unitless)
= transfer efficiency, fraction transferred from object x or hand to mouth (unitless)
= frequency of contact event over a 24-hour period (events/day)
= frequency of surface to food contact events that occur during consumption of food
item (events/food item)
EFj^F = frequency of surface to food contact events that occur during consumption of
food item (events/food item)
IRma = the child's respiration rate representing his activity level for that macroactivity
(nrVhr)
DSL = dermal soil loading on surface (mg/cm2)
DTCder = dermal transfer coefficient for the me/ma (cm2/hr)
Tme/ma = the time spent in that me/ma during the 24 hour period (hr/day)
Figures 15,16, and 17 show the data available by age category for the exposure factors in
Equations 1, 2, 3, 5, and 6. (Equation 4 is not included because so many factors are missing.):
H-51
-------
DRAFT- DO NOT CITE OR QUOTE
• Figure 15 shows data available for 0 to 30 days (by day). (Note that data are available for
only two exposure factors.)
• Figure 16 shows the data available for 0 to 36 months (by month). (Note that data are
available for only five exposure factors.)
• Figure 17 shows the data available for 0 to 24 years (by year). (Data are available for 11
exposure factors.)
These figures illustrate the age-related compatability of the available data for applying
Equations 1,2,3,5, and 6 (followed by Equation 7 as appropriate to estimate dose). They clearly
indicate that some important differences in age categories exist for the individual equations used
to estimate exposure and dose. For analysts attempting to estimate aggregate exposure (multiple
pathways for the same substance), the consistency of the age categories for the data used in
different equations might also be an issue.
Age: (Days) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
BW
"breastmflk
X XXXXXXXXXXX
X
X
Figure 15. Data Available for Ages 0 to 30 Days for Exposure Factors Used in Equations 1, 2, 3, 5, and 6
X = Reported at specific age
H-52
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Months)
BW
IRm.
"WBW
"^waler
IRbteaslmilk
0 1 2
X X
XXX
3 4 5 6 7 8 9 .10 il 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
XX X X X XX X
X X X X X X X X X
Figure 16. Data Available for Ages 0 to 36 Months for Exposure Factors Used in Equations 1, 2, 3, 5, and 6
X = Reported at specific age
|««j = Reported within age range
H-53
-------
DRAFT- DO NOT CITE OR QUOTE
Age: (Years) o 1 2 3 4 5 6 7 8 9 10 11 12- 13 14 IS 16 17 18 19 20 21 22 23 24
BW
IRma
Tnw/m»
DTCa.,
SA-
TE
EF
DSL"
xxxxxxxxxx xxx xx xxx
XXXXXXXXXX X X r"
IRwat«
IRb«»stnillk
"42
IRsol1
fmouth
EF
Figure 17. Data Available for Ages 0 to 24 Years for Exposure Factors Used in Equations 1, 2, 3, 5, and 6
• Total body surface area
" Limited data—depends on activity
X = Reported at specific age
|—| = Reported within age range
H-54
-------
DRAFT- DO NOT CITE OR QUOTE
In general, the lack of longitudinal data that would allow correlation of the exposure with
growth limits the ability to confidently model children's exposure. For example, when modeling
aggregate exposure as required by the Food Quality Protection Act, analysts face the challenges
of building long-term exposure profiles based on short-term data from a wide range of sources.
In this context, they must make assumptions about the interindividual variability (i.e., the extent
to which the observed differences in daily exposures represent differences between children) and
the intraindividual variability (i.e., the extent to which the observed differences in daily
exposures represent differences for each child). In reality, both sources of variation probably
exist. Unfortunately our ability to characterize them is limited by the lack of repeated samples in
longitudinal studies.
9.2 Challenges
While data are limited, the demand for exposure and risk analysis to inform risk
management decisions concerning children's health continues to increase. For most of the
exposure factors discussed in this issue paper, some information is available. Significant data
gaps remain in the following areas:
• Breast milk consumption by infants today and for children over age 1 year.
• Children's food handling practices and how this leads to exposure (e.g., by eating
with dirty hands or by eating food that has dropped onto a contaminated surface).
• Fish intake rates for young children and for children whose families include sport
fishers or whose families rely on self-caught fish for sustenance.
• Incidental and intentional soil intake by children.
• Soil adherence for dermal exposure.
• Relationships between various microactivities, macroactivities, and
microenvironments where children spend time.
• Correlation between exposure factors and growth (i.e., how children's exposure
behaviors change over time).
The demand for aggregate exposure assessment and cumulative risk assessment under the
Food Quality Protection Act (1996) creates a much greater need for information about correlation
between exposure factors and growth and places emphasis on the combined exposures that
children experience instead of on their exposure from a single pathway. Currently the greatest
challenge lies in combining the data from various independent studies in a way that appropriately
models the experiences of real American children.
Analysts building models to estimate aggregate exposure have already had the experience
of assessing and dealing with the inconsistencies of the age categories for children (Personal
communication with Paul Price and Chris Chaisson, 2000). For example, in building Lifeline™
(a model being developed to estimate aggregate exposure from pesticides), Price and Chaisson
evaluated all the existing data bases to explore where the natural breakpoints of age categories
occur in the data. In one example, for the time activity factor of time spent at home, they found
four age ranges that correspond to infancy (0-1 year), pre-school children (1-5 years), school-age
children (6-18 years), and post school-age children (>18 years) as the relevant categories. Figure
18 shows the distributions of data for these factors in these age categories for both genders.
H-55
-------
DRAFT- DO NOT CITE OR QUOTE
Figure 18: NHAPS Data on Time Spent at Home for Males and Females in Four
Age Groups (Source: The LifeLineTM Project)
0.9
0.8-
0.7"
0.6"
0.5'
0.4"
0.3"
0.2"
0.1 ••
Age group (years)
•-Males < 1
*- Males 1-5
—Males 6-18
-*-Males 19-24
•—Females <1
*~ Females 1-5
—Females 6-18
•~ Females 19-24
Bin (Minutes at Home)
H-56
-------
DRAFT- DO NOT CITE OR QUOTE
Notably, gender does not appear to be significant. Although these data provide good information
about the time spent at home by age category, they do not provide extensive information about
the activities pursued while at home. In addition, very little information exists about the details
of how time is spent in different microenvironments, although videography studies and other new
methods provide a means for collecting these data. In addition, how school-age children spend
their time in summer months and after school is relatively uncertain. The advantage that models
like the Lifeline™ model have over traditional exposure models is their ability to track the child
over the smallest relevant time unit (e.g., exposure over a day) and to build longer exposures
based on cumulating these units.. This is just one example, and the same issue would apply in
the development of other aggregated exposure models like Calendex™, the Cumulative and
Aggregate Exposure Risk Evaluation System (CARES), MENTOR, Stochastic Human Exposure
and Dose Simulation (SHEDS), and the Total Risk Integrated Methodology (TRIM).
A large, national longitudinal study of children's exposure would provide valuable data to
support exposure and risk analyses. However, currently enough information does exist to
support modeling efforts as long as the uncertainty in the analysis is appropriately considered.
The most significant challenges to modelers come from extrapolating the existing data to project
short-term data to longer-term averages required for the evaluation of chronic hazards and
dealing with the existing data gaps.
10. References
Adams, W.C. (1993) Measurement of breathing rate and volume in routinely performed daily
activities, Final Report. California Air Resources Board (CARB) Contract No. A033-205. June
1993. 185 pgs.
Basiotis, P.P.; Thomas, R.G.; Kelsay, J.L.; Mertz, W. (1989) Sources of variation in energy
intake by men and women as determined from one year's daily dietary records. Am. J. Clin. Nutr.
50:448-453.
Bearer, C.F., 1995. "How are children different from adults?" Environ. Health Perspect.
103(Suppl 6):7-12.
Binder, S.; Sokal, D.; Maughan, D. (1986) Estimating soil ingestion: the use of tracer elements in
estimating the amount of soil ingested by young children. Arch. Environ. Health. 41(6):341-345.
Bpyd, E. (1935) The growth of the surface area of the human body. Minneapolis, Minnesota:
University of Minnesota Press.
Brainard, J.B.; Burmaster, D.E. (1992) Bivariate distributions for height and weight, men and
women in the United States. Risk Anal. 12(2):267-275.
Burmaster, D.E.; Lloyd, K.J.; Crouch, E.A.C. (1994) Lognormal distributions of body weight as
a function of age for female and male children in the United States. Risk Anal.
H-57
-------
DRAFT- DO NOT CITE OR QUOTE
Butte, N.F.; Garza, C.; Smith, E.O.; Nichols, B.L. (1984) Human milk intake and growth in
exclusively breast-fed infants. Journal of Pediatrics. 104:187-195.
Calabrese, E.J.; Pastides, H.; Barnes, R.; Edwards, C.; Kostecki, P.T.; et al. (1989) How much
soil do young children ingest: an epidemiologic study. In: Petroleum Contaminated Soils, Lewis
Publishers, Chelsea, MI. pp. 363-397.
Canadian Ministry of National Health and Welfare (1981) Tapwater consumption in Canada.
Document number 82-EHD-80. Public Affairs Directorate, Department of National Health and
Welfare, Ottawa, Canada.
Cantor, K.P.; Hoover, R.; Hartge, P.; Mason, T.J.; Silverman, D.T.; et al. (1987) Bladder cancer,
drinking water source, and tapwater consumption: A case-control study. J. Natl. Cancer Inst.
79(6): 1269-1279.
Clausing, P.; Brunekreef, B.; Van Wijnen, J.H. (1987) A method for estimating soil ingestion by
children. Int. Arch. Occup. Environ. Health (W. Germany) 59(l):73-82.
Columbia River Inter-Tribal Fish Commission (CRITFC). (1994) A fish consumption survey of
the Umatilla, Nez Perce, Yakama and Warm Springs tribes of the Columbia River Basin.
Technical Report 94-3. Portland, OR: CRIFTC.
Costeff, H. (1966) A simple empirical formula for calculating approximate surface area in
children. Arch. Dis. Childh. 41:681-683.
Davis, S.; Waller, P.; Buschbon, R.; Ballou, J.; White, P. (1990) Quantitative estimates of soil
ingestion in normal children between the ages of 2 and 7 years: Population based estimates using
aluminum, silicon, and titanium as soil tracer elements. Arch. Environ. BQth. 45:112-122.
Dewey, K.G.; Lonnerdal, B. (1983) Milk and nutrient intake of breast-fed infants from 1 to 6
months:relation to growth and fatness. Journal of Pediatric Gastroenterology and Nutrition.
2:497-506.
Dewey, K.G.; Heinig, J.; Nommsen, L.A.; Lonnerdal, B. (1991a) Maternal versus infant factors
related to breast milk intake and residual volume: the DARLING study. Pediatrics. 87:829-837.
Dewey, K.G.; Heinig, J.; Nommsen, L.; Lonnerdal, B. (1991b) Adequacy of energy intake among
breast-fed infants in the DARLING study: relationships to growth, velocity, morbidity, and
activity levels. The Journal of Pediatrics. 119:538-547.
Dubois, D.; Dubois, E.F. (1916) A formula to estimate the approximate surface area if height and
weight be known. Arch, of Intern. Med. 17:863-871.
Ershow, A.G.; Brown, L.M.; Cantor, K.P. (1991) Intake of tapwater and total water by pregnant
and lactating women. American Journal of Public Health. 81:328-334.
H-58
-------
DRAFT- DO NOT CITE OR QUOTE
Ershow, A.G.; Cantor, K.P. (1989) Total water and tapwater intake in the United States:
population-based estimates of quantities and sources. Life Sciences Research Office, Federation
of American Societies for Experimental Biology.
Funk, L.; Sedman, R., Deals, J.A.J., Fountain, R. (1998) Quantifying the distribution of
inhalation exposure in human populations: Distributions of time spent by adults, adolescents, and
children at home, at work, and at school. Risk Analysis 18(l):47-56.
Gehan, E.; George, G.L. (1970) Estimation of human body surface area from height and weight.
Cancer Chemother. Rep. 54(4):225-235.
George, S.L.; Gehan, E.A.; Haycock, G.B.; Schwartz, GJ. (1979) Letters to the editor. J. Fed.
94(2):342.
Groot, M.E.; Lekkerkerk, M.C.; Steenbekkers, L.P.A. (1998). "Mouthing behaviour of young
children: An observational study." Wageningen: Agricultural University, Household and
Consumer Studies (ISBN 90-6754-548-1).
Hamill, P.V.V.; Drizd, T.A.; Johnson, C.L.; Reed, R.B.; Roche, A.F.; Moore, W.M. (1979)
Physical growth: National Center for Health Statistics Percentiles. American J. Clin. Nutr.
32:607-609.
Holmes, K.K.; Kissel, J.C.; Richter,K.Y. (1996) Investigation of the influence of oil on soil
adherence to skin. J. Soil Contam. 5(4):301-308.
International Life Sciences Institute (ILSI), 1992. Similarities and Differences Between Children
and Adults: Implications for Risk Assessment. Edited by P.S. Guzelian, CJ. Henry, and S.S.
Olin. Washington, DC: ILSI Press.
ILSI, 1996. "Research needs on age-related differences in susceptibility to chemical toxicants:
Report of an ILSI Risk Sciences Institute Working Group." Washington, DC: ILSI, November.
Javitz, H. (1980) Seafood consumption data analysis. SRI International. Final report prepared for
EPA Office of Water Regulations and Standards. EPA Contract 68-01-3887.
Juberg, D.R.; Alfano, K.; Coughlin, R.J.; Thompson, K.M. (2000). "An observational study of
object mouthing behavior by young children." Pediatrics. In Press.
Kissel, J.; Richter, K.; Duff, R.; Fenske, R. (1996a) Factors Affecting Soil Adherence to Skin in
Hand-Press Trials. Bull. Environ. Contamin. Toxicol. 56:722-728.
Kissel, J.; Richter, K.; Fenske, R. (1996b) Field measurements of dermal soil loading attributable
to various activities: Implications for exposure assessment. Risk Anal. 16(1): 116-125.
Layton, D.W. (1993) Metabolically consistent breathing rates for use in dose assessments. Health
Physics 64(l):23-36.
H-59
-------
DRAFT- DO NOT CITE OR QUOTE
Linn, W.S.; Shamoo, D.A.; Hackney, J.D. (1992) Documentation of activity patterns in "high-
risk" groups exposed to ozone in the Los Angeles area. In: Proceedings of the Second
EPA/AWMA Conference on Tropospheric Ozone, Atlanta, Nov. 1991. pp. 701-712. Air and
Waste Management Assoc., Pittsburgh, PA.
Linn, W.S.; Spier, C.E.; Hackney, J.D. (1993) Activity patterns in ozone-exposed construction
workers. J. Occ. Med. Tox. 2(1): 1-14.
Maxwell, N.I.; Burmaster, D.E. (1993) A simulation model to estimate a distribution of lipid
intake from breast milk during the first year of life. Journal of Exposure Analysis and
Environmental Epidemiology. 3:383-406.
McCormick, M.C. 1999. "Conceptualizing child health status: Observations from studies of very
premature infants" Perspectives in Biology and Medicine. 42(3):372-386.
Murray, D.M.; Burmaster, D.E. (1992) Estimated distributions for total surface area of men and
women in the United States. J. Expos. Anal. Environ. Epidemiol. 3(4):451-462.
National Academy of Sciences (NAS). (1991) Nutrition during lactation. Washington, DC.
National Academy Press.
National Academy of Sciences (NAS). (1974) Recommended dietary allowances, 8th ed.
Washington, DC: National Academy of Sciences-National Research Council.
National Center for Health Statistics (NCHS) (1987) Anthropometric reference data and
prevalence of overweight, United States, 1976-80. Data from the National Health and Nutrition
Examination Survey, Series 11, No. 238. Hyattsville, MD: U.S. Department of Health and
Human Services, Public Health Service, National Center for Health Statistics. DHHS Publication
No. (PHS) 87-1688.
Nelson, W.E.; Benrman, R.E.; Kliegman, R.M. (1998) Essentials of Pediatrics, 3rd Edition. New
York, NY: W.B. Saunders Co.
Neville, M.C.; Keller, R.; Seacat, J.; Lutes, V.; Neifert, M.; et al. (1988) Studies in human
lactation: milk volumes in lactating women during the onset of lactation and full lactation.
American Journal of Clinical Nutrition. 48:1375-1386.
Pao, E.M.; Hines, J.M.; Roche, A.F. (1980) Milk intakes and feeding patterns of breast-fed
infants. Journal of the American Dietetic Association. 77:540-545.
Pao, E.M.; Fleming, K.H.; Guenther, P.M.; Mickle, S.J. (1982) Foods commonly eaten by
individuals: amount per day and per eating occasion. U.S. Department of Agriculture. Home
Economics Report No. 44.
H-60
-------
DRAFT- DO NOT CITE OR QUOTE
Phillips, L.J.; Fares, R.J.; Schweer, L.G. (1993) Distributions of total skin surface area to body
weight ratios for use in dermal exposure assessments. J. Expos. Anal. Environ. Epidemiol.
3(3):331-338.
Price, P.; Su, S.; Gray, M. (1994) The effects of sampling bias on estimates of angler
consumption rates in creel surveys. Portland, ME: ChemRisk.
Reed, K.J.; Jimenez, M.; Freeman, N.C.G.; Lioy, PJ. (1999). "Quantification of children's hand
and mouthing activities through a videotaping methodology." J. Exposure Analysis and
Environmental Epidemiology 9:513-520.
Robinson, J.P; Thomas, J. (1991) Time spent in activities, locations, and microenvironments: a
California-National Comparison Project report. Las Vegas, NV: U.S. Environmental Protection
Agency, Environmental Monitoring Systems Laboratory.
Roseberry, A.M.; Burmaster, D.E. (1992) Lognormal distribution for water intake by children
and adults. Risk Analysis 12:99-104.
Ruffle,. B.; Burmaster, D.; Anderson, P.; Gordon, D. (1994) Lognormal distributions for fish
consumption by the general U.S. population. Risk Analysis 14(4):395-404.
Rupp, E.; Miler, F.L.; Baes, C.F. ffl. (1980) Some results of recent surveys of fish and shellfish
consumption by age and region of U.S. residents. Health Physics 39:165-175.
Spier, C.E.; Little, D.E.; Trim, S.C.; Johnson, T.R.; Linn, W.S.; Hackney, J.D. (1992) Activity
patterns in elementary and high school students exposed to oxidant pollution. J. Exp. Anal.
Environ. Epid. 2(3):277-293.
Thompson, K.M. (1999). "Devleping univariate distributions from data for risk analysis."
Human and Ecological Risk Assessment 5(4):755-783.
Timmer, S.G.; Eccles, J.; O'Brien, K. (1985) How children use time. In: Juster, F.T.; Stafford,
P.P.; eds. Time, goods, and well-being. Ann Arbor, MI: University of Michigan, Survey
Research Center, Institute for Social Research, pp. 353-380.
Tsang, A.M.; Klepeis, N.E. (1996) Results tables from a detailed analysis of the National Human
Activity Pattern Survey (NHAPS) response. Draft Report prepared for the U.S. Environmental
Protection Agency by Lockheed Martin, Contract No. 68-W6-001, Delivery Order No. 13.
USD A. (1996) Data tables: results from USDA's 1995 Continuing Survey of Food Intakes by
Individuals and 1995 Diet and Health Knowledge Survey. U.S. Department of Agriculture,
Agricultural Research Service, Riverdale, MD.
U.S. EPA. (1985) Development of statistical distributions or ranges of standard factors used in
exposure assessments. Washington, DC: Office of Health and Environmental Assessment; EPA
report No. EPA 600/8-85-010. Available from: NTIS, Springfield, VA; PB85-242667.
H-61
-------
DRAFT- DO NOT CITE OR QUOTE
U.S. EPA. (1995) Fish consumption estimates based on the 1991-92 Michigan sport anglers fish
consumption study. Final Report. Prepared by SAIC for the Office of Science and Technology.
U.S. EPA. (1996) Daily average per capita fish consumption estimates based on the combined
1989,1990, and 1999 continuing survey of food intakes by individuals (CSFII) 1989-91 data.
Volumes I and n. Preliminary Draft Report. Washington, DC: Office of Water.
U.S. EPA. (2000) Child-Specific Exposure Factors Handbook. Washington, DC: Office of
Research and Development.
Van Wijnen, J.H.; Clausing, P.; Brunekreff, B. (1990) Estimated soil ingestion by children.
Environ. Res. 51:147-162.
West, P.C.; Fly, M.J.; Marans, R.; Larkin, F. (1989) Michigan sport anglers fish consumption
survey. A report to the Michigan Toxic Substance Control Commission. Michigan Department of
Management and Budget Contract No. 87-20141.
West, P.C.; Fly, J.M.; Marans, R.; Larkin, F.; Rosenblatt, D. (1993) 1991-92 Michigan sport
anglers fish consumption study. Prepared by the University of Michigan, School of Natural
Resources for the Michigan Department of Natural Resources, Ann Arbor, MI. Technical Report
No. 6. May.
Wiley, J.A.; Robinson, J.P.; Cheng, Y.; Piazza, T.; Stork, L.; Plasden, K. (1991) Study of
children's activity patterns. California Environmental Protection Agency, Air Resources Board
Research Division. Sacramento, CA.
Zartarian, V.G.; Ferguson, A.C.; Leckie, J.O. (1998). "Quantified mouthing activity data from a
four-child pilot field study." J. Exposure Analysis and Environmental Epidemiology 8:543-553.
Acknowledgements and Disclaimer
This reported was prepared under contract to the ERG as part of its Peer Review contract with
the U.S. Environmental Protection Agency. The author thanks Jan Connery, Kate Schalk, Greg
Mark, and James Fiore of ERG for their assistance in preparing this manuscript, and the work
assignment managers, Bill Wood and Steve Knott, and the members of the EPA technical
workgroup for helpful suggestions. This document is a draft that has not been reviewed by the
EPA and should not be construed to represent the views of the EPA.
H-62
-------
-------
------- |