453-R-93-038
September 1993
Descriptive Guide to Risk Assessment
Methodologies for Air Toxics
Prepared for:
United States Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
-------
ACKNOWLEDGEMENTS
This report was prepared under contract to the U.S. Environmental Protection Agency's
Office of Air Quality, Planning and Standards (OAQPS). This work was funded under EPA
Contract Number 68-D2-0065. Views expressed are those of the authors and do not represent
official policy.
This report was based on a previous draft version prepared by another contractor.
Special thanks goes to Kelly Rimer of OAQPS who provided project management and technical
review, and to the various EPA staff scientists and engineers who provided review of earlier
drafts including Dr. Dan Guth (ECAO), Charlie Ris (OHEA), Dave Guinuup (OAQPS), Nadine
Shear (OAQPS), Michael Dusetzina (OAQPS), and Beth Hassett-Sipple (OAQPS).
-------
TABLE OF CONTENTS
1.0 INTRODUCTION
1.1 Purpose 1-1
1.2 Scope 1-1
1.3 Background 1-2
1.4 How to Use This Guide 1-5
2.0 HAZARD pENTIFICATION - CARCINOGENS AND NONCARCINOGENS
2.1 Introduction 2-1
2.2 Types of Studies 2-1
2.2.1 Epidemiologic Studies 2-2
2.2.2 In vivo Toxicological Studies 2-5
2.2.3 In vitro Toxicologic Tests 2-7
2.2.4 Physical/Chemical Properties 2-8
2.2.5 Structure/Activity Relationships 2-8
2.2.6 Pharmacokinetic Properties 2-9
2.3 Weighing of Evidence 2-13
2.3.1 Weight-of-Evidence Classification for Carcinogens 2-14
2.3.2 Weight-of-Evidence for Noncarcinogenic^Effects 2-17
2.4 Methods and Data Sources for Hazard Identification 2-20
2.4.1 Carcinogens 2-20
2.4.2 Noncarcinogens 2-23
2.5 REFERENCES 2-31
3.0 DOSE-RESPONSE ASSESSMENT - CARCINOGENS AND NONCARCINOGENS
3.1 Introduction 3-1
3.2 Selection of Data for Dose-Response Assessment 3-3
3.2.1 Epidemiologic Data 3-3
3.2.2 Toxicological Data 3-4
3.3 Carcinogens 3-6
3.3.1 Overview 3-6
3.3.2 Process of Carcinogenesis 3-7
3.3.3 Mathematical Dose-Response Extrapolation Models 3-9
3.3.4 Unit Risk Estimates for Inhalation Exposure 3-15
3.3.5 Other Ways of Expressing Dose-Response
Relationships 3-16
3.3.6 Methods and Data Sources for Dose-Response
Assessment 3-17
3.4 Noncarcinogens 3-20
3.4.1 Overview 3-20
3.4.2 Derivation of RfCs and RfDs 3-21
3.4.3 Alternative Dose-Response Methodology 3-23
3.4.3.1 Structure/Activity Relationships 3-23
3.4.3.2 Dose-Response Modeling 3-25
3.4.3.3 Benchmark Dose 3-31
3.4.3.4 Decision Analysis Approach 3-32
3.4.3.5 Occupational Exposure Limits 3-35
3.4.4 Methods and Data Sources for Dose-Response
Assessment 3-37
3.5 References 3-40
11
-------
4.0 EXPOSURE ASSESSMENT
4.1 Introduction 4-1
4.2 Emission Characterization 4-4
4.2.1 Overview 4-4
4.2.2 Technical Background and Methods 4-4
4.2.2.1 Estimating and Measuring Toxic Emissions 4-4
4.2.2.2 Complex Chemical Mixtures 4-6
4.2.2.3 Chemical Reactions and Removal Mechanisms 4-7
4.2.2.4 Summary of Methods to Characterize
Emissions 4-8
4.2.3 Selection Criteria for Emissions Characterization 4-14
4.3 Fate and Transport Analysis 4-17
4.3.1 Overview 4-17
4.3.2 Technical Background and Methods 4-17
4.3.2.1 Atmospheric Transport and Dispersion 4-17
4.3.2.2 Pollutant Transformation and Deposition 4-19
4.3.2.3 Transport in Water and Soil Media 4-20
4.3.2.4 Uptake of Pollutants in the Food Chain 4-23
4.3.2.5 Summary of Methods to Estimate Fate and
Transport >- 4-24
4.3.3 Selection Criteria for Fate and Transport Estimation 4-29
4.4 Population Characterization 4-32
4.4.1 Overview 4-32
4.4.2 Technical Background and Methods 4-32
4.4.2.1 General Population Characteristics 4-33
4.4.2.2 Population Mobility 4-34
4.4.2.3 Summary of Methods to Characterize
Populations 4-43
4.4.3 Selection Criteria for Population Characterization 4-46
4.5 Exposure Calculations 4-47
4.5.1 Overview 4-47
4.5.2 Technical Background and Methods 4-47
4.5.2.1 Exposure Estimation -. . 4-47
4.5.2.2 Dose Estimation 4-49
4.5.2.3 Exposure Due to Ingestion Pathway 4-49
4.5.2.4 Summary of Methods to Estimate Exposure 4-50
4.5.3 Selection Criteria for Exposure Estimation 4-57
4.6 Monitoring Techniques 4-57
4.6.1 Overview 4-57
4.6.2 Technical Background and Methods 4-59
4.6.2.1 Ambient Air Monitoring 4-59
4.6.2.2 Personal Exposure Monitoring 4-61
4.6.3 Select Applications of Ambient Air
and Personal Exposure Monitoring 4-63
4.6.3.1 The NMOC Monitoring Program 4-63
4.6.3.2 The Urban Air Toxics Monitoring Program 4-63
4.6.3.3 The Integrated Air Cancer Project 4-65
4.6.3.4 The Total Exposure Assessment
Methodology (TEAM) Study 4-65
ill
-------
4.7 References 4-66
4.7.1 References for Section 4.2,
Emission Characterization 4-66
4.7.2 References for Section 4.3,
Fate and Transport Analysis 4-68
4.7.3 References for Section 4.4,
Population Characterization 4-70
4.7.4 References for Section 4.5, Exposure Calculations 4.71
4.7.5 References for Section 4.6, Monitoring Techniques 4-72
5.0 RISK CHARACTERIZATION
5.1 Introduction 5-1
5.2 RAC Guidance on Risk Assessment 5-2
5.2.1 Full Characterization of Risk 5-2
5.2.2 Consistency and Comparability 5-3
5.2.2.1 Risk Descriptors 5-3
5,2.3 Professional Judgement 5-4
5.3 Quantification of Cancer Risks 5-5
5.3.1 Overview 5-5
5.3.2 Technical Background and Methods . . „ 5-5
5.3.2.1 Cancer Unit Risk Estimates 5-5
5.3.2.2 Expression of Cancer Risks 5-6
5.3.2.3 Consistency in Risk Calculations 5-8
5.3.2.4 Multiple Chemical Mixtures 5-11
5.3.2.5 Senstivie Subpopulations 5-12
5.3.2.6 Multipathway Risks 5-12
5.3.2.7 Summary of Methods 5-13
5.3.3 Selection of Methods 5-19
5.4 Quantification of Noncancer Risks 5-21
5.4.1 Overview 5-21
5.4.2 Technical Background and Issues 5-21
5.4.2.1 Comparison of Exposure to RfC or RfD 5-21
5.4.2.2 Hazard Index for Chemical Mixtures 5-23
5.4.2.3 Multipathway Hazard Index 5-24
5.4.2.4 Dose-Response Modeling 5-25
5.4.2.5 Decision Analysis Approach 5-26
5.4.2.6 Summary of Methods 5-26
5.4.3 Selection of Methods 5-28
5.5 Characterization of Uncertainty 5-31
5.6 References 5-40
6.0 EMERGING ISSUES
6.1 Ecological Risk Assessment 6-1
6.1.1 Frameowrk for Ecological Risk Assessment 6-3
6.1.1.1 Problem Formulation 6-3
6.1.1.2 Analysis 6-6
6.1.1.3 Risk Characterization 6-8
6.1.2 Defining Ecological Risk Assessment Methods 6-10
6.1.2.1 Quaiitiative Methods 6-11
6.1.2.2 Quantitative Methods 6-11
6.1.3 Major Issues 6-13
IV
-------
6.2 Chemical Mixtures 6-15
6.2.1 Overall Approach 6-15
6.2.2 Existing Methods 6-16
6.2.3 Interactions 6-15
6.2.4 Other Mathematical Models 6-22
6.2.5 Uncertainties 6-22
6.3 References 6-24
7.0 USES OF RISK ASSESSMENT METHODOLOGIES
7.1 Introduction 7-1
7.2 Case I 7-3
7.2.1 Study Objectives 7-3
7.2.2 Scope 7-3
7.2.3 Selection and Use of Risk Assessment Methods 7-4
7.2.4 Other Considerations 7-12
7.3 Case H 7-17
7.3.1 Study Objectives 7-17
7.3.2 Scope 7-17
7.3.3 Selection and Use of Risk Assessment Methods 7-18
7.3.4 Other Considerations r 7-26
7.4 Case m 7-31
7.4.1 Study Objectives 7-31
7.4.2 Scope 7-31
7.4.3 Selection and Use of Risk Assessment Methods 7-32
7.4.4 Other Considerations 7-42
-------
LIST OF TABLES
Table Page
2-1 Illustrative Categorization of Carcinogenic Evidence
Based on Animal and Human Data (EPA, 1986") 2-18
2-2 Carcinogen Hazard Identification - Methods and Data Sources 2-21
2-3 Four Types of Response Levels (Ranked in Order of Increasing
Severity of Toxic Effect) Considered for Systemic Toxicants 2-28
2-4 Summary of and Selection Criteria Methods for Noncarcinogen
Hazard Identification 2-29
3-1 Carcinogen Dose-Response - Summary of and Selection for Methods 3-18
3-2 Guidelines for the Use of Uncertainty Factors in Deriving
Reference Concentrations (RfC) v. 3-24
3-3 TLV Adjustment Factors Used by States Air Toxics Programs 3-36
3-4 Noncarcinogen Dose-Response - Summary of Methods and Data Sources 3-38
3-5 Noncarcinogen Dose-Response - Summary of Methods 3-39
4-1 Office of Air Quality Planning and Standards Technology
Transfer Network 4-3
4-2 Summary of Methods for Characterizing Emissions 4-15
4-3 Selection Criteria for Source Sampling Method for
Characterizing Emissions 4-16
4-4. Summary of Available Exposure Models 4-28
4-5 Selection Criteria for Exposure Models 4-30
4-6 Geographic Units of U.S. Census Bureau Data 4-35
4-7 Summary of U.S. Census Data . . . .... 4-37
4-8 Summary of the Human Exposure Model-II 4-54
4-9 Selection Criteria for the Human Exposure Model-II 4-59
4-10 Summary of Selection Considerations for
Ambient Air Monitoring Capabilities 4-63
4-11 Summary of Selection Considerations for
Personal Exposure Monitoring -. . . . 4-64
VII
-------
LIST OF TABLES
(continued)
Table Page
5-1 Example Risk Distribution 5-9
5-2 Example Calculation of Cancer Risk 5-10
5-3 Summary of Cancer Risk Characterization Methods 5-14
5-4 Summary of HEM-H Results 5-17
5-5 Summary of Selection Criteria for Cancer Risk
Characterization Methods 5-20
5-6 Summary of Noncancer Risk Characterization Methods 5-27
5-7 Summary of Selection Criteria for Noncancer Risk
Characterization Method ^ 5-29
5-8 Sources of Uncertainty in Risk Assessment 5-32
5-9 Some Statistics Useful for Quantifying Uncertainty 5-37
6-1 Some Potential Endpoints for an Ecological Risk Assessment 6-7
6-2 Classification Scheme for the Quality of the
Risk Assesment of the Mixture 6-18
7-1 Pollutant: Arsenic [Unit Risk Estimate = 4.3 x 10'3] (ug/m3)'1] 7-13
7-2 Pollutant: Benzene [Unit Risk Estimate = 4.3 x 10'31 (ug/m3)'1] 7-14
7-3 Sample Output of Source Category Ranking for Single Pollutant 7-15
7-4 Sample Output of Source Category Ranking Based on All Pollutants . . 7-15
7-5 Summary of HEM-H Results for Noncarcmogens ""-28
7-6 Summary of HEM-II Results for Carcinogens 7-29
7-7 Source Categories Affected by the Proposed Regulation 7-43
7-8 Annual Exposure Scenarios for Proposed Waste-to-Energy Facility 7-43
7-9 Estimated Maximum Increased Cancer Risks from the
Waste-to-Energy Facility 7-44
7-10 Estimated Maximum Increased Noncancer Risks from the ,
Waste-to-Energy Facility : 7-45
via
-------
1.0 INTRODUCTION
1.1 PURPOSE
This report provides basic information and a general discussion of the steps and
technical issues involved in conducting a risk assessment. It also identifies and briefly
describes available risk assessment methods. This information is intended to assist State and
local agencies make informed decisions on whether to conduct a risk assessment and on what
methods are appropriate for a particular situation. Because method selection depends on
many factors such as the goals of the risk assessment, the available information, and time and
resource constraints, this report does not attempt to prescribe any single methodology that
should be used. Rather, it discusses key features, basic assumptions and uncertainties, inputs
and outputs, potential uses, and the relative level of expertise, time, and resources needed to
•N*
use each methodology. Based on this information, the reader can decide which methods may
be appropriate and obtain more detailed information from other referenced documents, data
bases, and agencies.
1.2 SCOPE
This document, developed by the Air Risk Information Support Center (Air RISC),
describes methods of estimating risks from toxic pollutants released into the air. The focus is
on routine releases from stationary sources, rather than releases from mobile sources or short-
term accidental releases from stationary sources. Many of the risk methodologies discussed
in this document focus on inhalation exposures, the primary exposure pathway for air
pollutants. However, pollutants released into the air can also enter other media through
deposition, so some multipathway risk assessment methodologies are also discussed.
Risk assessments may deal with only a single compound or with multiple compound
mixtures, depending on the emissions source characteristics and the risk assessment goals.
Methods for both of these scenarios are included. Most of the methodologies discussed in
this report are currently available for use in assessing potential risks to human health from
both carcinogens and noncarcmogens.
Risk assessment is a rapidly evolving field, and new methods are continually being
developed. Therefore, Section 6 deals briefly with some emerging issues, primarily
estimation of ecological (non-human health) risks.
1-1
-------
This document is intended to provide a broad understanding of risk assessment and
commonly available methods, so no single method is discussed in great detail. Basic
principles, key assumptions, and sources of uncertainty are described. This information will
help managers understand risk assessment, decide whether to conduct risk assessments, and
select appropriate methods. Other sources of information can then be consulted for the
detailed information needed to actually conduct a risk assessment.
1.3 BACKGROUND
Risk assessment is a multidisciplinary evaluation of factual information as a basis for
estimating the health effects that individuals or populations may experience as a result of
exposure to hazardous substances. Specific risk assessment steps are guided by
methodologies and may be qualitative or quantitative.
In 1983, the National Academy of Sciences (NAS) "published a study of how risk
assessment is conducted in the Federal government. This study covered risk assessment
methods for all agencies: the Food and Drug Administration, the EPA, and any other
agencies that deal with human health risks. The study concluded that there was a need for
uniformity and better definition of risk assessment within the Federal government. The NAS
then established a framework to guide future risk assessment by Federal agencies. As defined
by the NAS, risk assessment consists of four steps:
(1) hazard assessment
(2) dose-response assessment
(3) exposure assessment, and
(4) risk characterization.
Risk assessment should not be confused with risk management. Risk management is
the process of weighing policy alternatives and selecting an appropriate action to mitigate or
avoid possible dangers. Risk management integrates the results of risk assessment with other
information such as engineering data and social, economic, and political factors. Risk
assessment, rather than risk management, is described in this guide.
Hazard identification is the review of relevant biological and chemical information to
determine whether or not a chemical may cause health effects.
1-2
-------
• H
Z
UJ
UJ
o
i
^
(0
s
H
z
LU
+ f*
(f)
tf\
W*
Hi
uu
0)
en
£
RESEARCH
0?
X
WTIFICATION
jenl cause
» effect?)
S y2
a « «
o5-§
l£ «
S-8 ^
8 i S
§11
? .e o
1*1
l^'i
S » m
11-8
l|m
>-
/
>
ASSESSMENT
** A rtrrt iirv^v 41 i
!;
O !
a- ',
X -V
LU ;
•a
— c
CO O
§1
•SJ2
* S
1 Field measurements,
exposures, charact
1
i s
g .2
B-
S> *
d or anticipated
>nt conditions?)
tixperiencei
under differ*
••*
1
'J)
c
1 of populatio
CD
E
-------
Dose-response assessment defines the relationship between the exposure or dose of an
agent and the magnitude of the health response. This often includes a quantitative estimate
which measures the possible impact of a health effect for a range of doses.
Exposure assessment produces an estimate of the extent of exposure to which the
populations of interest are likely to be subject.
Risk characterization integrates the hazard identification, dose-response assessment,
and exposure assessment, in order to describe the nature, and often the magnitude, of health
risk. The risk characterization includes presentation of uncertainties and provides a
framework to help judge the significance of the hazard and risk estimate.
The relationship of the four steps of risk assessment and risk management as presented
by the NAS is shown in Figure 1-1. This report describes these four steps in sequence and
presents the distinct types of methodologies and models available for use by USEPA for each
step. In actually conducting a risk assessment, however, the first three steps may be
conducted in parallel, and consistency among them is important. For example, the results of
preliminary health hazard identification (step 1) and information on the types and quantities of
pollutants emitted (part of step 3) may be considered together in determining which
compounds require a detailed dose-response assessment (step 2) and exposure assessment
(step 3). And, in order to integrate dose-response assessment (step 2) and exposure
assessment (step 3) in the risk characterization (step 4), the two assessments should be
compatible in terms of the compounds and pathways considered, units of measure, and other
factors discussed under risk characterization.
Risk assessments are used in a variety of ways by Federal, State, and local agencies in
making decisions about regulating emissions of toxic air pollutants. For example, an
assessment may be done in order to prioritized sources or pollutants in establishing an overall
regulatory program. Assessments of risks, before and after air pollution controls are required,
may be used to evaluate the risk reduction benefits of specific regulatory alternatives for a
particular source category or pollutant. Detailed, site-specific risk assessments may be used
in making decisions on granting a permit to a new source. Different methods and levels of
detail would be appropriate in these different cases. This report includes information on the
1-3
-------
level of detail and types of studies for which various methods may be appropriate.
Illustrative case studies are given.
1.4 HOW TO USE THIS GUIDE
The report is organized into seven sections, including this introduction (Section 1).
Section 2 describes hazard identification for cancer and noncancer health effects. Section 3
discusses dose-response assessment for both carcinogens and noncarcinogens.
Section 4 describes exposure assessment methods. It includes sections on emission
characterization, fate and transport analysis, population characterization, exposure calculation,
and monitoring techniques.
Section 5 presents a discussion of the risk characterization step. It includes sections
on cancer and noncancer risk characterization and on characterization and presentation of
•\»
uncertainties.
Thus, Sections 2 through 5 complete the basic presentation of the principles and
available methods for the four steps of risk assessment. Each section begins with an
overview of the technical background and methods, to acquaint the reader with the scientific
and technical issues to be addressed in that step of a risk assessment. This is followed by a
summary of available methods and models. Tables present the inputs and outputs and
calculational approach of each method or model, along with some information on how the
model handles the basic technical issues and its key assumptions. The tables provide a quick
summary reference but, the accompanying text should be read to develop a fuller
understanding of the issues and methodologies.
Section 6 includes information on some emerging Issues which are relevant to
risk assessment for air emissions. Finally, Section 7 presents three illustrative case studies on
the uses of risk assessment. These case studies draw on the information and methodologies
presented in the proceeding sections. They provide examples of the potential goals and uses
of risk assessment studies, and describe how and why various methodologies would be
selected and used consistent with the objectives and constraints of the study.
-------
2.0 HAZARD IDENTIFICATION - CARCINOGENS AND NONCARCINOGENS
2.1 INTRODUCTION
Hazard identification is the initial, qualitative step of a risk assessment. In the hazard
identification process, pertinent available data for a compound of interest is reviewed to
determine if the compound is capable of eliciting adverse effects (i.e., cancer or noncancer
endpoints). This process is initiated to derive or locate existing hazard designations for the
carcinogenic or noncarcinogenic potential of a particular substance. For common chemical
carcinogens, several organizations including the EPA have developed weight-of-evidence
classification schemes. The formal classification of compounds with noncancer health effects is
somewhat different.
Data gathered in the initial hazard identification step are evaluated based on both their
scientific and statistical quality (EPA, 1986). Useful information for hazard identification
includes results of epidemiological studies, controlled human experiments, toxicologic studies,
in vitro and in vivo tests, and physical and chemical properties of the agent. Because these types
of studies and the resulting data are the basis for hazard identification for both carcinogens and
noncarcinogens, they are presented first in Section 2.2. Hazard classification schemes are
discussed in Section 2.3. Section 2.4 covers hazard identification methods and data sources.
2.2 TYPES OF STUDIES
Various types of information are available for determining the carcinogenic and
noncarcinogenic potential of a compound. Useful data for hazard identification comes from
epidemiologic studies, controlled human experiments, in vivo and in vitro toxicologic tests, and
analysis of physical/chemical properties, structure-activity relationships, and pharmacokinetic
properties (EPA 1990a; EPA, 1986). Epidemiologic and lexicological studies provide the most
important information and are essential in determining the hazardous potential of a compound.
The other types of information serve mainly as supporting data to the toxicoiogic and
epidemioiogic studies.
2.2.1 Epidemiologic Studies
Epidemiology is the study of the distribution and occurrence of disease in the human
population. In hazard assessment, epidemiology focuses on diseases resulting from various
factors or agents from outside the human organism. For instance, investigators have examined
the incidence of lung cancer in coke oven workers and the possibility of cancer risk posed by
2-1
-------
human exposure to electromagnetic fields. In noncarcinogenic studies, investigators may examine
a wide range of effects, for example, the incidence of hypertension in middle-aged males to
decrements in IQ in children following exposure to lead. Positive results from well-conducted,
well-designed analytic epidemiologic studies are the most conclusive type of hazard identification
data, because if an effect is observed in a human population that cannot reasonably be attributed
to other causes, the public health reality of environmental exposures is quite high. However,
negative results from epidemiologic studies do not necessarily mean the compound is not
hazardous.
There are two types of epidemiologic studies, descriptive studies and analytic studies. A
descriptive study characterizes the distribution and occurrence of disease in an entire population
(OSTP, 1985). However, no attempt is made to quantify individual exposures or to determine
if the disease occurred in exposed or nonexposed individuals. Therefore, these studies are useful
for characterizing human disease in general, but they cannot ascertain whether the exposure to
a particular compound actually caused the disease. For example,'a descriptive study might
determine that the population of Town A had 100 new cases of a particular cancer during the 10
years following the opening of a factory that released a suspected carcinogen to the atmosphere,
but it could not show that exposure to the compound caused the cancer cases. Descriptive studies
can suggest cause-and-effect relationships, but analytic studies are required to determine whether
a cause-and-effect relationship exists. In conducting analytic studies, epidemiologists collect data
on exposure and disease occurrence at the individual level. The study population is separated
into exposed and unexposed individuals to determine if an increased health risk is associated with
exposure to the compound of interest, fa the case of Town A, an analytical study would compare
cancer incidence or other relevant data on factory workers with similar data on other groups, or
similar data on residents who live near the factory as opposed to residents who live far from the
factory. Therefore, by using the analytic methodology, it is sometimes possible to both identify
and quantify health risks to humans (,OSTP, 1985).
Analytic studies are subdivided into two types, case-control studies and cohort studies.
These are graphically depicted in Figure 2-1. In case-control studies, individuals are initiallv
identified as either having the disease (cases) or not having the disease (controls). Then, past
exposure history for the individuals in both groups is ascertained through interview's.
questionnaires, examination of medical records, occupational logs, and other sources. The
2-2
-------
OK 1 *• *• OK
5K
•o
_3
1o
•^ 0)
O CO
i: co .
C 03 f
8(0 v-
, Q
03
CO
CO
O
03
3
CO _
o C
Q.
X
) f
i
1
i
»• c
Q
>»
T3
^*
co
•c
o
f~
>»
T3
^^
co
^^
0
-C -
^ a^
X
• C
) <
v>
>»
T3
_2
"S5
•««
o
o>
'E
c
"5>
03
.Q
«•#
<0
T3
03
O
< ^O
i: -o 03
o> 3 -
•g « di
•s •?; c
-C O •—
5 o) o
C = 22
o c -a
05 .£ C
'co _ O5 ^
O3 ^ 03 co
n^ /*^
_Q 02 o c
A'E "5 > ®
o: 5 £ c a
V)
4>
'•5
**
(0
"n
o
'cb
JD
o
1
-------
frequency of exposure of the cases and the controls to a particular toxicant can then be compared.
If the cases have a statistically significant greater frequency of exposure than the controls, an
associative relationship can be drawn between the exposure and the disease.
Cohort studies start with groups of individuals known to be exposed and unexposed and
follow them over a period of time to determine the disease rate that develops in each group.
With cohort studies, disease rates in exposed and unexposed groups are compared to try to
determine if the particular exposure is responsible for an increased incidence of disease. Cohort
studies can be done either prospectively or retrospectively to follow disease rates in a given
population. The prospective cohort study identifies exposed and unexposed groups and then
ascertains disease rates as time passes. Because this type of cohort study is time consuming and
expensive, it is not as commonly performed as the retrospective study, which examines past
exposure and disease rates. Information for retrospective studies is collected from medical
records, occupational records, physical examinations, questionnaires, tumor registries, and death
certificates. Cohort studies have been especially useful in examining increased cancer risk in
heavily exposed populations such as people who work in hazardous occupational settings.
Assessment of the overall quality of the epidemiologic study is an important part of the
hazard identification process. EPA and the Office of Science and Technology Policy (OSTP)
identify criteria for evaluating epidemiologic data (EPA, 199la; OSTP, 1985). These criteria
include the proper selection and characterization of exposed and control groups, the adequacy of
duration and quality of follow-up, the proper identification and characterization of confounding
factors and bias, the appropriate consideration of latency effects, the vaiid ascertainment of the
causes of disease and death, the ability to detect specific effects, use of reliable and valid
estimates of exposure, and a sufficiently large study group. A well-designed, well-conducted
epidemiologic investigation will address all of these factors together.
Epidemioiogic studies, regardless of how well they are designed and conducted, have
certain strengths and weaknesses. Epidemioiogic investigations have the advantage of directly
characterizing the experience of human populations and their response to environmental
exposures and other exogenous factors (OSTP. 1985). Information from the other types of
studies requires extrapolation from animal species to humans, which may be problematic because
of differences in anatomy and physiology between species. Even human cell culture studies
require extrapolation from in vitro to in vivo conditions. The advantage of epidemiologic studies
2-4
-------
is that they supply evidence of an exposure and its subsequent health effect as it typically occurs
in the human environment. Epidemiologic studies have been critical in establishing the
carcinogenic potential of agents such as tobacco smoke, asbestos, and benzene.
Although epidemiologic studies have the advantage of directly evaluating the human
experience, they have certain limitations. Exposure is often not well quantified in epidemiologic
studies, especially in retrospective studies. Specifically, concentrations are not well quantified
and exposures are typically defined categorically (e.g., low, medium, high). Therefore, a strict
concentration-response relationship often cannot be ascertained. Another problem is that disease
incidence is commonly established from human exposures at high or at intermediate levels
(OSTP, 1985). At lower exposure levels it becomes more difficult to define a cause-and-effect
relationship because they may cause lower disease incidence, which then may be open to various
alternative explanations such as chance, errors, and bias. Unless a very large (and costly)
epidemiologic study is used, only relatively large increases in disease incidences can be detected
(EPA, 1986).
The long latency period associated with some environmentally caused diseases also
presents a problem in epidemiologic studies. This long period of time between exposure and
manifestation of the disease makes it difficult to establish causal relationships and impossible to
evaluate toxic effects of chemicals recently introduced into the environment. Another problem
associated with epidemiologic investigations is the inability of the study design to completely
control for unknown risk factors. Disease rates can be influenced by a variety of unidentified
risk factors such as genetic predisposition, health and nutrition status, and exposures to other
unidentified toxicants. Because of these limitations, definitive epidemioiogic data exists only for
a small percentage of chemicals.
2.2.2 In vivo Toxicological Studies
In toxicological studies, the toxic potential of a chemical is evaluated in various test
species. These in vivo studies may be conducted in humans as clinical studies or controlled
experiments or in experimental laboratory animals. The use of animal bioassays in hazard
identification is based on'the assumption that, for some compounds and for some metabolic
pathways, the animal test species are similar enough to humans to allow comparison between the
two organisms. Although there has been controversy over this assumption in the past, especially
m regard to carcinogens, the National Research Council (NRC) states that "virtually every form
2-5
-------
of human cancer has an experimental counterpart, and every form of multicellular organism is
subject to cancer, including insects, fish, and plants" (NRC, 1977). Therefore, animal data are
commonly extrapolated to humans in order to assess cancer as well as noncancer risks. However,
a number of biological differences may need to be taken into account when extrapolating from
test animals to humans.
Long-term in vivo animal cancer bioassays are typically conducted at three different dose
levels, at and/or below the maximum tolerated dose (MTD), using a route of exposure the same
or similar to that seen in human populations. The MTD is the dose that is just high enough to
result in signs of minimal toxicity without significantly curtailing the animal's life span due to
effects other than carcinogenicity when administered for the duration of the chronic study (OSTP,
1985). This MTD exposure protocol is required in experimental animals in order to cause a
X-.
measurable carcinogenic response across a small group of test animals if the substance is a
carcinogen. At the conclusion, and also at intervals during the long-term study, test animals are
.sacrificed and examined microscopically for cancerous lesions. Animals that die during the
course of the study are also examined. These studies provide data on tumor location, presence
of benign and malignant tumors including unusual tumors, tumor type, time required to cause
tumor, and noncancer toxicological effects. A number of organizations, including the EPA,
National Cancer Institute/National Toxicology Program, and the International Agency for
Research on Cancer (1ARC), have developed guidelines for conducting long-term in vivo
bioassays for carcinogens. The OSTP (1985) summarizes them as follows:
• At least two species of test animals (most commonly rats and mice of both sexes)
should be examined at two, or preferably three dose levels.
• Dosage levels should be near or at the MTD and at fractions of the MTD (e.g.,
1/2, 1/3, etc.).
• The exposure route should be similar to that encountered in the human population.
• Each dose group and concurrent control group should contain at least 50 animals
of each sex.
• Duration of the study is conventionally 100 weeks for hamsters, 120 weeks for
mice, and 130 weeks for rats.
• Detailed pathologic examination of tissues should be conducted.
2-6
-------
• Controls should be subject to appropriate statistical evaluation.
The lifespan of the typical laboratory rodent is currently assumed to be 2 years.
Therefore, a chronic study covering 2 years is assumed to cover the lifespan of the animal and
thus may represent the lifespan of other species, including humans. The actual lifespan of a test
animal is species specific. However, given the predominance of using rodents, long-term
(chronic) tests are those lasting about 2 years. Shorter-term in vivo tests, such as subchronic (90+
days), short-term (14 to 90 days), and acute (single dose) tests, are also available. These studies
can be used to determine the type of adverse effects to be expected, to identify target organs, and
to provide data on how living organisms process compounds within the body (pharmacokinetics)
and how a compound exhibits its toxic activity (mechanism of action). Short-term studies also
serve as screening procedures for designing long-term toxicity studies.
2.2.3 In vitro Toxicological Tests
In vitro tests are conducted in cell cultures rather than in whole living organisms. These
tests are used to support evidence of carcinogenicity and noncancer adverse effects, and may also
provide data on the mechanism of action for a compound. (EPA, 1986). In vitro tests may
require less than 1 day or up to 8 months to complete, and are typically conducted as a screening
procedure for further study in in vivo animal bioassays. One type of in vitro test is the
genotoxicity or mutagenicity test used to assess the ability of a compound to alter
deoxyribonucieic acid (DNA), the genetic material in living organisms. Since many cheimcai
carcinogens exhibit mutagenic activity, in vitro tests can be used as predictors of potential
carcinogens.
Many in vitro tests are available to examine DNA adduct formation, gene mutations, DNA
repair, initiation and promotion activity, and cellular transformations. One of tne most widely
used in vitro tests is the Ames assay, which evaluates the mutagenic potential of a chemical or
mixture of chemicals in the bacterial strain Salmonella typhimurium. With another in vitro test,
the ability of a compound to induce chromosomal aberrations can be measured by the number
of sister chromatic exchanges (SCEs). SCEs are indicative of genetic damage; however their
biological significance is still uncertain. These SCEs are believed to result from the induction
of recombination by damage to the DNA (OSTP, 1985). An example of an in vitro bioassay
for noncarcinogens is the culturing of hepatocytes (liver cells) in the presence of a chemical agent
to assess the effects at the cellular level. In vitro tests can be combined with tests in live
2-7
-------
animals; for example, a potential carcinogen can be added to cultured mammalian cells and the
transformed cells subsequently injected into an appropriate recipient animal to evaluate the ability
of the transformed cells to induce tumors in a recipient host.
Caution should be exercised when the only data indicating potential toxicity are from in
vitro tests. The results of these tests alone cannot definitively support carcinogenic or
noncarcinogenic toxicity. In addition, the EPA states that the "lack of positive results in short-
term (in vitro) tests for genetic toxicity does not provide a basis for discounting positive results
in long-term animal studies" (EPA, 1986). The results of in vitro tests can, however, contribute
to the overall weight-of-evidence for a substance.
2.2.4 Physical/Chemical Properties
As part of the hazard identification process, information is gathered on the
N«
physical/chemical properties of the compound of interest. This type of information includes
properties such as vapor pressure, solubility, stability, pH sensitivity, and chemical reactivity.
These properties affect distribution of the chemical in air, water, and soil; its persistence in the
environment; its uptake, distribution, and metabolism in the body; and its excretion. These
factors in turn influence the way the compound affects human health, and provide useful fi
information for designers of toxicological studies.
2.2.5 Structure/Activity Relationships
Structure/activity relationships (SARs) can be used in a supportive way to suggest the
possibility that an adverse effect may exist. The process of evaluating SARs involves comparing
the molecular structures of known carcinogenic or noncarcinogenic chemicals to the chemical of
interest to determine the likelihood of carcinogenic activity or of a particular toxic endpoint. For
example, the EPA used SARs to crudely gauge the carcinogenic potential in a comparison of
chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans (EPA, 1987). Carcinogens are
known to be found within certain chemical classes, and the magnitude of carcinogenicity vanes
widely within a particular chemical class (OTA, 1981). Examining SARs during the hazard
identification process can support a concern about the compounds's carcinogenic potential, but
can not by itself support a determination of carcinogenicity. For noncarcinogens, SARs can point
to possible target organs, mechanisms of action, and adverse effects. However, there are
problems with the reliability of SARs. A SAR is an estimate based on comparison of two or
2-8
-------
more chemicals. The actual validity of the comparison cannot be known until confirmed by other
types of data.
2.2.6 Pharmacokinetic Properties
Pharmacokinetics is the study of the absorption, distribution, metabolism, and excretion
of compounds in biological systems (EPA, 1989a). Pharmacokinetics is defined as how living
organisms process compounds within the body. Pharmacokinetics may influence how a toxicant
enter the internal body systems, how it is distributed, biotransformed or eliminated and where it
may exhibit toxic activity. Absorption, distribution, and elimination are the major components
of pharmacokinetics and are summarized in Figure 2-2.
Absorption is the process by which a foreign agent crosses body membranes at routes of
exposure and enters the internal body systems. Each route of exposure—dermal, ingestion and
inhalation—represents a route of absorption through the skin, gastrointestinal tract, and lungs,
respectively. For absorption through the skin to occur, the agent must penetrate several layers
of skin, each of which has its own microenvironment and properties. The rate of transport or
absorption through the skin depends upon the permeability of the toxicant in these layers.
Absorption through the GI tract is most influenced by an agent's solubility and dissolution rate.
In general, the more lipophilic the compound the greater the absorption by the GI tract.
Absorption through the lungs is determined through the properties of the toxicant. If the agent
is a gas, it is absorbed through the alveoli. The transfer or absorption from the lungs to the
blood depends upon the blood solubility of the toxicant. If it is an aerosol or paniculate, the
absorption depends upon the site of deposition in the respiratory tract and the solubility of the
particle which is influenced by the size of the particle or the presence of active transport
processes. Clearance mechanisms in the respiratory tract (e.g., mucociliary action) can result in
the agent being eliminated or swallowed. If swallowed, absorption would then occur through the
GI tract.
Once a toxicant enters the systemic circulation, it is available for transport or distribution
throughout the body and potentially is available to all target organs in the body. The final site
of distribution for a particular toxicant depends upon the blood flow, the ability of that toxicant
to pass across cellular membranes of different types of tissues, and the toxicant's affinity for
various types of tissues. Toxicants may be stored or accumulated in various tissues in the body
that may or may not be the target organ for toxic effects. The liver and kidney generally
2-9
-------
(0
1H
-------
concentrate more toxicants than any other organ systems. Other places of storage include fat,
primarily for lipophilic compounds; and bone, primarily for elemental toxicants such as lead,
strontium and fluorine.
Toxicants are removed from the body through the processes of elimination. Elimination
can be defined as the sum of all processes of biotransformation and excretion. Excretion can
occur by various routes: in urine, from the kidney; exhaled breath from the lungs; in feces, from
the gastrointestinal tract; or from skin, via sweat or saliva. Excretion may also occur in females
through milk. The kidney excretes more toxicants than any other route.
Biotransformation plays an important role in influencing an agents toxic property and
availability; it represents the sum of all biological or chemical processes to which a foreign
chemical is subject while in the body. Figure 2-3 presents a schematic of the disposition and
V-
toxic effects of a chemical including the role of biotransformation. Biotransformation reactions
can be divided into two different categories. Phase 1 reactions include hydrolysis, reduction, and
oxidation reactions; and Phase 2 metabolism reactions are conjugation and synthesis.
Biotransformation is primarily responsible for converting of lipophilic the compounds to ones
which are more hydrophilic. This conversion reduces the compound's ability to partition into
membranes which reduces the potential for distribution and reabsorption and as a result, promotes
excretion of the toxicant. The liver is predominantly responsible for biotransformation.
Biotransformation can also occur in other organ systems such as the lung, kidney, skin and
gonads although transformation rate and capacity are much lower in these organ systems.
Biotransformation is very important in consideration of toxic activity because it can be
divided into two essential categories: detoxification and bioactivation. In detoxification, the
products of biotransformation are less toxic or reactive than the parent compound. In
bioactivation, the products of biotransformation are more toxic than the parent compound.
Therefore, increased metabolism can either increase or decrease the potential for toxic response,
depending upon whether detoxification or bioactivation is taking place. The human body consists
of a complex network of metabolic pathways, many of which are competing. Each pathway
may result in unique metabolites. Dose level affects the pathway of biotransformation.
For example, enzymes that have a high affinity but low capacity are first responsible for
biotransformation for a particular toxicant. However, as this enzyme system becomes saturated
2-11
-------
b
CD
O
C
C
0)
CO
0)
'_
Q.
0
CC
o
•<— >
en
E
0)
.c
o
CO
CO
CM
C5
u_
(O
13
o
I
o
2-12
-------
and dose increases, other pathways with low affinity but higher capacity are activated. Therefore,
as dose increases, the percentage of toxicants transformed by the initial pathway decreases and
a greater percentage of biotransformation occurs via the secondary or tertiary pathways. Actual
toxicity would depend upon whether the products of these metabolic pathways are being
detoxified or being bioactivated.
Therefore, review of this type of data for suspected toxicants can result in a clearer
understanding of: 1) whether the agent is direct-acting or requires metabolic conversion to
become active, 2) metabolic pathways involved in both activation and detoxification of the agent,
3) macromolecular interactions and fate, and 4) similarities and differences between species
(EPA, 1986). Pharmacokinetic data can supply useful general information about the compound
and may contribute to other steps of the risk assessment, such as dose-response assessment. The
>»
EPA has developed guidelines on data gathering for pharmacokinetics (EPA, 1991).
Pharmacokinetic data should be evaluated in the context of these guidelines.
2.3 WEIGHING OF EVIDENCE
After all the data on a chemical have been collected, interpreted, and reviewed, the
weighing of evidence about the chemical's toxic potential is necessary. Weight-of-evidence is
defined as "the extent to which the available biochemical data support the hypothesis that a
substance causes an effect in humans" (EPA, 1989a). Items to consider when determining the
overall weight-of-evidence are the quality and adequacy of the data and the reproducibility of
toxic response in various test systems. Weight-of-evidence is determined primarily from two
sources of information: epidemiologic investigations and in vivo animal studies (EPA, 1986).
These results are supplemented with data from the structure/activity relationships,
pharmacokinetic properties, physical/chemical properties, in vitro bioassays, and other available
toxicologic information. The EPA (1986) identifies three major steps in characterizing the
weight-of-evidence for human toxicity: 1) characterization of evidence from human and animal
studies individually, 2) combination of these two types or data to indicate the overall weight-of-
evidence for human toxicity, and 3) evaluation of ail supporting information to determine if the
overall weight-of-evidence should be modified. Currently, EPA is reviewing weighing of
evidence.
2-13
-------
2.3.1 Weight-of-Evidence Classification for Carcinogens
The EPA uses qualitative descriptors for classifying the overall weight-of-evidence for
carcinogenicity which has some similarity to the IARC approach for classifying the weight-of-
evidence for human and animal data. These classification schemes are generally broad in nature
and, therefore, should include a written summary of the strengths and weaknesses of the evidence
in addition to the categorization (EPA, 1986). This overall summary is required because the
classification schemes place a finite description on the weight-of-evidence while, in reality, there
is a stratification within a category of weight-of-evidence. For. example, within the B2
classification there may be some strong B2s, those with 10 animal studies all showing the same
outcome as well as others with only 1 or 2 studies. Therefore, given the variation within a
classification a narrative description is also useful.
Under the IARC (1988) classification scheme, evidence of carcinogenicity in experimental
animals is placed in one of the following groups:
1. Sufficient evidence of carcinogenicity. Causal relationship has been established
between agent or mixture and an increased incidence of malignant neoplasms or
of an appropriate combination of benign or malignant neoplasms: (a) in multiple
species of strains; or (b) in multiple independent experiments (carried out at
different times or in different laboratories or under different protocols); or (c) in
a single exceptional study in one species when malignant neoplasms occur or type
of tumor, or age at onset of tumor. Additional evidence may be provided by data
on dose-response effects.
2. Limited evidence of carcinogenicity. The data suggest a carcinogenic effect but
are limited for making a definitive evaluation because, e.g., (a) the evidence of
carcinogenicity is restricted to a single experiment; or (b) there are unresolved
questions regarding the adequacy of the design, conduct or interpretation of the
study; or (c) the agent or mixture increases the incidence oniy of benign
neoplasms or lesions of uncertain neoplastic potential or of certain neoplasms
which may occur spontaneously in high incidences in certain strains.
3. Inadequate evidence of carcinogenicity. The studies cannot be interpreted as
showing either the presence or aosence of a carcinogenic effect because of major
qualitative <^r quantitative limitations.
4. Evidence suggesting lack of carcinogenicity. Adequate studies involving at least
two species are available which show that within the limits of the tests used, the
agent or mixture is not carcinogenic. A conclusion of evidence suggesting lack
of carcinogenicity is inevitably limited to the species, tumor sites and levels of
exposure studied.
2-14
-------
EPA is weighing criteria for classifying animal data that have a few small differences
from IARC.
1. An increased incidence of combined benign and malignant tumors will be
considered to provide sufficient evidence of carcinogenicity if the other criteria
defining the "sufficient" classification of evidence are met (e.g., replicate studies,
malignancy). Benign and malignant tumors will be combined when scientifically
defensible.
2. An increased incidence of neoplasms that occur with high spontaneous background
incidence (e.g., mouse liver tumors and rat pituitary tumors in certain strains)
generally constitutes "sufficient" evidence of carcinogenicity, but may be changed
to "limited" when warranted by the specific information available on the agent.
3. A "no data available" classification has been added.
^^ N-
For classifying evidence of carcinogenicity from studies in humans, IARC uses the
following four groups:
1. Sufficient evidence of carcinogenicity. A causal relationship has been established
between exposure to the agent, mixture or exposure circumstance and human
cancer. That is, a positive relationship has been observed between the exposure
and cancer in studies in which chance, bias and confounding could be ruled out
with reasonable confidence.
2. Limited evidence of carcinogenicity. A positive association has been observed
between exposure to the agent, mixture or exposure circumstance and cancer for
which a causal interpretation is considered to be credible, but chance, bias or
confounding could not be ruled out with reasonable confidence
3. Inadequate evidence of carcinogenicity. The available studies are of insufficient
quality, consistency or statistical power to permit a conclusion regarding the
presence or absence of a causal association.
4. Evidence suggesting lack of carcinogenicity. There are several adequate studies
covering the mil range of levels of exposure that human beings are known to
encounter, which are mutually consistent in not showing a positive association
between exposure to the agent, mixture or exposure circumstance and any studied
cancer at any observed level of exposure. A conclusion of 'evidence suggesting
lack of carcinogenicity' is inevitably limited to the cancer sites conditions and
levels of exposure and length of observation covered by the available studies. In
addition, the possibility of a very small risk at the levels of exposure studied can
never be excluded.
In some instances, the above categories may be used to classify the degree of evidence
for carcinogenicity for specific organs or tissues.
2-15
-------
EPA made the following modifications to the IARC approach classifying evidence from
human studies (EPA, 1986):
1. The observation of a statistically significant association between an agent and life-
threatening benign tumors in humans has been included in the evaluation of risks
to humans.
2. A "no data available" classification is another explanation for inadequate evidence.
The EPA classification system for the characterization of the overall weight of evidence
(animal, human, and other supportive data) of carcinogenicity of a compound includes five
groups (EPA, 1986).
Group A - Human Carcinogens: Sufficient evidence from epidemiologic studies to
support a causal association between exposure to the agents and cancer.
x-
GroupJB - Probable Human Carcinogens: Limited evidence of human carcinogenicity
based on epidemiologic studies or sufficient evidence of carcinogenicity based on animal
studies. This group is divided into two subgroups. Group B1 is reserved for agents for
which there is limited evidence of carcinogenicity from epidemiologic studies. It is
reasonable, for practical purposes, to regard an agent for which there is "sufficient"
evidence of carcinogenicity in animals as if it presented a carcinogenic risk to humans.
Therefore, agents for which there is "sufficient"-evidence from animal studies and for
which there is "inadequate evidence" or "no data" from epidemiologic studies would
usually be categorized under Group B2.
Group C - Possible Human Carcinogens: Limited evidence of carcinogenicity in animals
in the absence of human data. This group includes a wide variety of evidence such as
(a) a malignant tumor response in a single weil-conducted experiment that does not meet
conditions for sufficient evidence, (b) tumor responses of marginal statistical significance
in studies having inadequate design or reporting, (c) benign, but not malignant tumors
with an agent showing no response in a variety of short-term tests for mutagenicity, and
(d) responses of marginal statistical significance in a tissue known to have a high or
variable background rate.
Group D - Not Classifiable as to Human Carcinogemciry: Inadequate human and animai
evidence of carcinogenicity, or no data are available.
Group E - Evidence of Noncarcinogenicitv for Humans: No evidence for carcinogenicity
in at least two adequate animal tests in different species or in both adequate animal tests
in different species or in both adequate epidemiologic and animal studies. The
designation of an agent as being in Group E is based on the available evidence and would
not be interpreted as a definitive conclusion that the agent will not be a carcinogen under
anv circumstances.
2-16
-------
Table 2-1 illustrates how evidence based on animal and human studies is combined to yield a
tentative assignment to one of the five categories (EPA, 1986).
In addition to these EPA and IARC classification schemes weighing many types of
evidence, the National Toxicology Program (NTP) publishes reports of animal cancer bioassays.
The NTP provides a strength of evidence classification relative to the animal tested. NTP
indicates whether there is "clear" evidence, "some" evidence, "equivocal" evidence, "no
evidence", and "inadequate" study. Periodically, NTP also publishes its annual Report on
Carcinogens in which a weighing of evidence approach is used to classify evidence into two
categories as follows:
1. Known to be carcinogens: There is "sufficient evidence of carcinogenicity" from
studies in humans "which indicates a causal relationship between the agent and the
human cancer."
2. Reasonably anticipated to be carcinogens:
A. There is "limited evidence of carcinogenicity" from studies in humans,
"which indicates that causal interpretation is credible, but that alternate
explanation, such as chance, bias or confounding, could not be adequately
excluded," or .
B. There is "sufficient evidence of carcinogenicity" from studies in
experimental animals" which indicates that there is an increased incidence
of malignant tumors: (a) in multiple species or strains, or (b) in multiple
experiments (preferably with different routes of administration or using
different dose levels),or (c) to an unusual degree with regard to incidence,
site or type of tumor, or age at onset. Additional evidence may be
provided by data concerning dose response effects, as well as information
on mutagenicity or chemical structure."
2.3.2 Weighing Evidence for Noncarcinogenic Effects
Weighing evidence for noncarcinogemc compounds are simpler compared to those for
carcinogens though just as rigorous in terms of scientific consideration. The process of
developing weight-of-evidence criteria for systemic toxicants is time consuming because of the
multitude of toxic endpoints to be considered. EPA identifies the following factors in animal
studies for assessing the noncarcinogenic weight-of-evidence for a compound (EPA, 1987):
2-17
-------
TABLE 2-1. ILLUSTRATIVE CATEGORIZATION OF CARCINOGENIC EVIDENCE
BASED
ON ANIMAL AND HUMAN DATA (EPA, 1986)1
Animal Evidence
Human Evidence Sufficient Limited Inadequate No Data No Evidence
Sufficient
Limited
Inadequate
No Data
No Evidence
A
Bl
B2
B2
B2
A
Bl
C
C
C
A
Bl
D
D
D
A
Bl
D
D
D
A
Bl
D
"E
E
The above assignments are presented for illustrative purposes. There may be nuances in the
classification of both animal and human data indicating that different categorizations that
those given in the table should be assigned. Furthermore, these assignments are tentative and
may be modified by ancillary evidence. In this regard, all relevant information should be
evaluated to determine- if the overall weight of evidence needs to be- modified. Relevant
factors to be included along with tumor data from human and animal studies include
structure-activity relationships, short-term test findings, results of appropriate physiological,
biochemical, and lexicological observations, and comparative metabolism and
pharmacokinetic studies. The nature of these findings may cause an adjustment of the overall
categorization of the weight of evidence.
2-18
-------
• Clear evidence of a dose-response relationship;
• Similar effects across sex, strain, species, exposure routes, or in multiple
experiments;
• Biologically plausible relationship between metabolism data, the postulated
mechanism of action, and the effect of concern;
• Similar toxicity exhibited by structurally related compounds;
• Some correlation between the observed chemical toxicity in animals and human
evidence.
Consideration and evaluation of the above items is useful in developing the weight-of-evidence
classification of a particular compound. In fact, the greater the weight-of-evidence, the greater
•\-
the confidence in the conclusion derived from the data (EPA, 1989c).
Although weight-of-evidence schemes are not available for the majority of noncancer
health effects, the EPA have developed schemes for developmental toxicants. (EPA, 199la)
EPA proposed the following weight-of-evidence scheme for developmental toxicants. The
weight-of-evidence scheme has two major categories: sufficient evidence and insufficient
evidence. The sufficient evidence category can be used to describe both positive and negative
evidence of potential developmental hazard within the context of dose, duration, timing, and route
of exposure. This category includes both human and experimental animal evidence and is further
divided into:
Sufficient Human Evidence
This category includes data from epidemiologic studies (e.g., case control and
cohort) that provide convincing evidence for the scientific community to judge
that a causal relationship is or is not supported. A case series in conjunction with
strong supporting evidence may also be used. Supporting animal data may or may
not be available.
Sufficient Experimental Animal Evidence/Limited Human Data
This category includes data from experimental animal studies and/or limited
human data tnat provide convincing evidence for the scientific community to
judge if the potential for developmental toxicity exists. The minimum evidence
necessary to judge that a potential hazard exists generally would be data
demonstrating an adverse developmental effect in a single, appropriate, well-
conducted study m a single experimental animal species. The minimum evidence
needed to judge that a potential hazard does not exist would include data from
2-19
-------
appropriate, well-conducted laboratory animal studies in several species (at least
two) which evaluated a variety of the potential manifestations of developmental
toxicity, and showed no developmental effects at doses that were minimally toxic
to the adult.
Insufficient Evidence
This category includes situations for which there is less than the minimum
sufficient evidence necessary for assessing the potential for developmental toxicity,
such as when no data are available on developmental toxicity, as well as for data
bases from studies in animals or humans that have a limited study design (e.g.,
small numbers, inappropriate dose selection/exposure information, other
uncontrolled factors), or data from a single species reported to have no adverse
developmental effects, or data bases limited to information on structure/activity
relationships, short-term tests, pharmacokinetics, or metabolic precursors.
The noncancer weight of evidence schemes described above have been developed for
reproductive and developmental toxicity. Weight of evidence schemes for other endpoints will
be introduced when guidelines are developed addressing those endpoints. Therefore, specifics
of weight of evidence may vary among endpoints but the same general approach is
anticipated.
2.4 METHODS AND DATA SOURCES FOR HAZARD IDENTIFICATION
2.4.1 Carcinogens.
If a weight-of-evidence determination for a particular environmental contaminant has not
been performed, a review of the pertinent literature can be conducted to determine whetner a
chemical may be carcinogenic and where in the weight-of-evidence range the contaminant is
likely to belong. This determination could be arrived at by following the guidelines set forth by
the EPA, LARC, or NTP (described in the previous section;. However, expert judgement would
be required to fully assess the quality of the available human or animal studies and the
sufficiency of the evidence. Table 2-2 indicates where to find EPA, IARC, and NTP reviews and
other data sources that can be used in determining a weight-of-evidence classification for a
particular compound. The table also summarizes the time and expertise required to use each
source. The following paragraphs briefly describe the methods indicated in the table.
The EPA, IARC, and NTP have conducted carcinogenic hazard identifications for a
number of environmental contaminants. Listings of the chemicals reviewed and classified to date
can be obtained from published reports or by contacting these organizations. These organizations
should be used as the initial source of information in hazard identification for a particular
2-20
-------
CD
CO
LU
O
CC
o
CO
I
CO
Q
o
-S
O
i
§
cc
a
cr
a
o
z
o
cc
*V
C\J
CO
UJ
m
1
i
J
CO
lE
o o
i «
nj a.
_ a.
o 03
« -i
as
oj JH
•s -S
= C UJ
fO 'O —
S c
.£ '0
u
2
2
CO
1= u
-f J
"8
f
"i
o
CO
en
<
cc
a.
o >-
CC Q
a. z>
a. t—
< a>
Q
UJ
CC
o
LU
cc
LU
co
>—
cr
LU
a.
x
I
ce
.1
a
Ul
cc
LU
cc
co
LU
a
cc
O
CO
ui
cc
I
O"
cc
1
o
y^S
P£
< =
3 a
UJ UJ
CC GC
•5 p
2-21
-------
compound because their weight-of-evidence classifications have been developed by panels of
experts in the field of chemical carcinogenesis.
The preferred source of hazard identification information is the EPA's Integrated Risk
Information System (IRIS). This data base is available both on-line through the National Library
of Medicine, and a printed version is distributed by the National Technical Information Service
(NTIS). The weight-of-evidence determinations in IRIS are developed through an intraagency
work group, including scientists from throughout the EPA. An EPA weight-of-evidence
classification is established for each chemical reviewed by the work group. The basis for the
classification is included in the IRIS data base. The IRIS database can be accessed through on-
line computer databases and is also available in CD-ROM and PC-based versions. Each is
updated monthly to incorporate recently reviewed materials and can be used by any EPA, State,
or local agency staff.
In addition to the EPA reviews of potential chemical carcinogens, IARC and NTP also
have work groups that make carcinogenicity determinations. Like the EPA work group, the
IARC and NTP work groups are also composed of experts from a number of scientific disciplines
including epidemiologists, lexicologists, physicians, and statisticians. The NTP classifications
are published annually; IARC updates its assessments periodically as needed. The lARC's
carcinogenicity assessments, published as monographs, are very extensive reviews of the available
data for a particular chemical and are useful in conducting a hazard identification. The NTP
annual reports are also useful, though less extensive than the IARC monographs.
Additional information on carcinogenic potential is available through Gold's Tumor data
base, which contains information on the carcinogenic potential of some 492 rodent carcinogens.
Although it does not include a weight-of-evidence scheme, the data base can supply pertinent
information for the determination of carcinogenic potential. In addition, primary journal articles
on the compound of interest, including both toxicologic and epidemiologic studies, can be
obtained by searching the general scientific literature.
Table 2-2 presents an overview of the time and expertise required to use the available data
sources. Both EPA and LARC comprehensively assess carcinogenic potential, and are the best
sources of hazard identification information for both a screening and a refined carcinogenic risk
assessment. The time required to conduct a hazard identification through IRIS or IARC is
relatively small.
2-22
-------
If neither EPA nor IARC have evaluated the compound, the NTP reports should be
consulted. Two types of NTP reports can aid in hazard identification. The first is the Annual
Report on Carcinogens which contain the NTP's own weight-of-evidence classifications for
chemicals. The NTP also conducts long-term animal bioassays on suspected carcinogens and
generates reports on the findings. If the compound of interest is found in the NTP's Annual
Report, time required for hazard identification will again be relatively short. If the individual
NTP animal bioassays need to be reviewed, more time and expertise will be required.
The final methods available for conducting a hazard assessment are reviewing Gold's
Tumor data base and reviewing the general scientific literature. Both require time consuming,
extensive review of the data and critical analysis to determine carcinogenic potential.
2.4.2 Noncarcinogens.
>-
In the noncancer hazard identification process, an evaluation of the appropriateness,
nature, quality, and relevance of scientific data is necessary (EPA, 1990a). Other considerations
include the characteristics, magnitude, and relevance of experimental routes of exposure, and the
nature and significance to human health of the observed effects. The formal classification of
compounds with noncancer effects is simpler but no less rigorous than the classification of
carcinogenic effects. To date, EPA has developed guidelines for three different noncancer health
effects: male reproductive effects (EPA, 1988a), female reproductive effects (EPA, 1988b), and
suspect developmental effects (EPA, 199la). Since a single noncancer hazard identification
scheme does not exist, a discussion of general issues to be considered must be discussed prior
to a discussion of specific hazard identification methods.
Types of Noncancer Effects. Agents thai cause noncancer toxic endpoints are often
referred to as "systemic toxicants" due to their effects on the function of various organ systems
and to distinguish them from carcinogens (Barnes and Dourson, 1988). Systemic is defined as
pertaining to or affecting the whole body or acting in a part of the body other than the site of
entry (EPA, 1989b). However, adverse health effects may also occur at the site of entry. These
"portal of entry" effects are also included in the category of noncarcinogenic effects. A
compound that causes systemic or portal of entry effects may also be carcinogenic. For example,
carbon tetrachloride has been found to be carcinogenic, but it also can cause the systemic effect
of liver damage.
2-23
-------
Most noncancer risk assessment methodologies are based on the "threshold" concept of
mechanisms of action. The threshold concept assumes that below a certain level of exposure no
adverse response will occur. On the other hand, the "nonthreshold" concept assumes that any
level of exposure has a probability of evoking an adverse response. The nonthreshold concept
is used for most carcinogenic compounds and for a few noncarcinogenic agents, (i.e. criteria
pollutants. For noncarcinogens, EPA generally assumes that a threshold will exist and that
certain compensating and adaptive mechanisms must be overcome before a compound causes a
response (EPA, 199 Ib). Most techniques used for evaluating noncarcinogenic risk are based on
the threshold concept. Because individual sensitivities differ, individuals within a population may
have different thresholds. Therefore, a range of thresholds may exist for a population, making
a general population threshold difficult to define.
Noncancer health effects may be associated 'with respiratory, cardiovascular,
immunological, neurologic, renal, hepatic, reproductive, developmental, or myelitic systems.
Numerous potential toxic endpoints are associated with each of these systems. For example,
under female reproductive effects, toxic endpoints in animals include infertility, adverse
pregnancy outcomes, and adverse influences on offspring survival (EPA, 1988b). Other examples
of toxic endpoints are bronchitis (respiratory), myocardial infarction (cardiovascular), and cleft
palate (developmental).
Noncancer toxic endpoints can be generally categorized as either reversible or irreversible
effects. A reversible effect is defined as a stress that diminishes or disappears due to adaptive
or compensatory responses when the toxic exposure stops. An irreversible effect, on the other
hand, is persistent and may worsen or progress even after the exposure is removed (EPA, 1990aj.
The reversibility or irreversibility of an effect depends upon the concentration, the duration of
exposure, and the capacity of compensatory processes. Both reversible and irreversible effects
must be considered in the toxicologic evaluation of data because reversible effects may eventually
lead to irreversible effects.
When evaluating scientific data for noacarcinogenic risk assessment, it is important to
determine whether an effect is adverse or nonadverse. An adverse effect is defined as either a
biochemical change, functional impairment, or pathological lesion, that either singly or in
combination, adversely affects the performance of an organism, or affects the organism's ability
to respond to another environmental challenge (EPA, 199Ib). Exposures to toxicants can result
2-24
-------
in a number of different effects of varying severity. Figure 2-4 depicts the spectrum of response.
The vertical scale illustrates disability in terms of the overall organism, while the horizontal scale
represents the impairment of biological systems. Initially, exposed individuals may not show any
signs of toxicity because the body has the ability to detoxify or compensate for exposures to
pollutants, or because multiple cells perform the same function. However, at a certain level the
body can no longer accommodate or compensate for the exposure to pollutants. At this level,
physiological changes can be observed. These first physiological changes (e.g., changes in
enzyme levels or lung function) may not affect the overall health or integrity of the organism,
so their significance would be unknown or uncertain. However, a large change in these
parameters may be considered significantly adverse to health. As dose or exposure increases, the
body's protective mechanisms continue to break down, and clinical or pathological changes begin
>,-
to be apparent. These changes include damage to tissues, decreases in function, changes in
physiological functions, and severe irritation. As dose further increases, a percentage of
individuals begin to exhibit the obvious clinical effects known as morbidity which includes
obvious illness, requirement for medicine, or need for hospitalization. Where effects are often
subtle, their significance may be contentious. The evaluation of adverse response(s) for a
compound requires professional judgement and may require experts trained in a particular area.
Adversity is evaluated both on the toxicologic (or biological) and the statistical significance of
the particular response.
Identification of Critical Effect. Critical effect is defined as the adverse effect that occurs
at the lowest dose in trie identified animal studies. Data from animal studies are usually required
for systemic toxicants due to the lack of availability of adequate human data. A compound can
elicit more than a single toxic effect, so the "critical effect" must be identified in the hazard
identification process. The critical effect is identified after assessing the quality of the various
studies, determining the biological and statistical significance of the effects, and deiineatmg the
reversible and irreversible endpomts (EPA, 1990). If data are available m multiple species, the-
critical effect is determined from the most sensitive species, unless there is evidence that a
specific species has a response more similar to humans. Determination of the critical effect is
important because data concerning. the effect are necessary in the noncancer dose-response
assessment. The threshold for human exposure is then estimated based on the critical effect
found in animal studies.
2-25
-------
Death
;= Disease
JO
cc
CO
2
~ Disturbed
£ Function
Health
Adverse Effect
Homeostasis
Compensation Breakdown
and Adaptation
Organ System Impairment
Failure
Source: SPA, 1991b.
Figure 2-4. Continuum of severity of noncancer effects.
2-26
-------
While hazard identification is generally described as a qualitative procedure, the critical
effect and threshold concepts introduce a quantitative component. As a result, it may sometimes
be difficult to distinguish between the qualitative aspects of hazard identification and the
quantitative dose-response for noncarcinogens. This difficulty is supported by the fact the EPA
merged discussion of hazard identification and dose-response in their developmental toxicity risk
assessment guidelines (EPA, 199la). For carcinogens, in many cases, any dose level of exposure
is considered to be hazardous. But for noncarcinogens, the amount of a compound to which a
person is exposed or the length of exposure time must be considered to determine whether the
compound is a hazard. Therefore, dose response levels must be briefly introduced at this point.
To evaluate data for the dose-response assessment, EPA has developed a hierarchy of response
levels that can be identified from the experimental data. EPA describes four response levels for
systemic toxicants including the no-observed-effect-level (NOEL), the no-observed-adverse-effect-
level (NOAEL), the lowest-observed-adverse-effect-level (LOAEL), and the frank effect level
(PEL) (EPA, 1989c). Definitions of these response levels are presented in Table 2-3. NOAELs
and LOAELs are dependent on the study design and the dose used in the available studies. In
noncarcinogen hazard identification, the LOAEL is the level at which the critical effect occurs,
and the NOAEL is assumed to represent exposures at or below which no adverse effect would
result.
When establishing the critical effect, it is also necessary to consider the sensitivities of
certain segments of the human population such as the elderly, children, individuals predisposed
with a disease, and individuals with genetic susceptibility.
Methods and Data Sources. Table 2-4 indicates the EPA and other data sources available
for determining a weight-of-evidence classification for a noncarcinogenic compound and gives
an overview of the time and expertise required to use each source. To date, EPA has established
a protective exposure level commonly called a reference dose (RfD) for many compounds by the
oral route of exposure. An equivalent exposure level for the inhalation route is referred to as an
mnalation reference concentration (RfC). The RfC and RfD are accompanied by an EPA
statement regarding the overall confidence level associated with the study, the data base, and the
RfC/RfD itself. The initial source of information for hazard identification of hazardous air
pollutants should be the on-line IRIS data base, which contains EPA-reviewed data and a weight-
of-evidence determination for the compound, if one exists. If the compound has been evaluated
2-27
-------
TABLE 2-3. FOUR TYPES OF RESPONSE LEVELS
(RANKED IN ORDER OF INCREASING SEVERITY OF TOXIC EFFECT)
CONSIDERED FOR SYSTEMIC TOXICANTS
NOEL: No-observed-effect-level. That exposure level at which there are no
statistically or biologically significant increases in frequency or severity of
effects between the exposed population and its appropriate control.
x-
NOAEL: No-observed-adverse-effect-level. That exposure level at which there are no
statistically or biologically significant increases in frequency or severity of
adverse effects * between the exposed population and its appropriate control.
Effects are produced at this level, but they are not considered to be adverse.
LOAEL: Lowest-observed-adverse-effect-level. The lowest exposure level in a study or
group of studies that produces statistically or biologically significant increases
in frequency or severity of adverse effects between the exposed population and
its appropriate control.
PEL: Frank effect level. That exposure level which produces frankly apparent and
unmistakable adverse effects, such as irreversible functional impairment or
mortality, at a statistically or biologically significant increase in frequency or
severity between an exposed population and its appropriate control.
Adverse effects are defined as any effects resulting in functional impairment and/or
pathological lesions that may affect the performance of the whole organism's, or
that reduce an organism's ability to respond to an additional challenge.
2-23
-------
J3
"55
8
0.
Q
OC
-3
f
J
CD
.
(0
-s
X
0
CO
§
CO
X
8=
LU
I
o
cc
UJ
Q
OC
o
u_
s
o
8"
CD
o
cc
I
co
o
cc
o
o
o
UJ
CO
LL_
o
cc
CO
S
«
CO
CC
o
co
J2
CO
o
1 *
'
.a
c
JS '-
J2 *
2 E
fU O
£
5
(35
C
i §
I co
^
i
CO
O
"2
CO
£
|
!>
-------
by the Agency, IRIS will indicate the critical effect and provide a summary of the key data for
the compound. (IRIS also lists the NOAEL and LOAEL for the compound if either has been
identified.)
If a hazard identification has not been performed, a review of the available literature can
be conducted to determine the toxicity associated with a particular compound. Two sources of
information exist for gathering information on a compound: 1) data bases and reports containing
references to the primary literature, and 2) the primary literature itself. The advantage to
searching data bases and reports is that, in most instances, the references have been peer
reviewed. Reports, such as the Agency for Toxic Substances and Disease Registry toxicity
profiles and the EPA Health Assessment Documents, are particularly useful since they contain
actual summaries of the primary literature.
If a weight-of-evidence has not been established for a noncarcinogenic toxicant, it is
necessary to develop one from the available information (reports and primary literature). This
requires expert judgement for the identification of the critical effect and the NOAEL or LOAEL.
2-30
-------
2.5 REFERENCES
Barnes, Donald G., and Michael Dourson, U. S. Environmental Protection Agency. Reference
Dose (RfD): Description and Use in Health Risk Assessments. Washington, D.C. 1988.
California Department of Health Services. Draft Guidelines for Identification and Hazard
Assessment of Agents Causing Developmental and/or Reproductive Toxicity. Draft. December
29, 1989.
International Agency for Research on Cancer. IARC Monographs on the Evaluation of the
Carcinogenic Risk of Chemicals to Humans. Volumes 1 and 2, Supplement 4. World Health
Organization. Lyon, France, October 1982.
International Agency for Research on Cancer. LARC Monographs on the Evaluation of the
Carcinogenic Risk of Chemicals to Humans. Volume 34. Lyon, France. June 1984.
•\«-
National Research Council, Safe Drinking Water Committee. Drinking Water and Health.
Volume 1. National Academy Press. Washington, D.C. 1977.
Office of Science and Technology Policy (OSTP), Executive Office of the President. Chemical
Carcinogens; A Review of the Science and Its Associated Principles. Washington, D.C. 50 FR
10372. March 14, 1985.
Office of Technology Assessment (OTA). Assessment of Technologies for Determining Cancer
Risks from the Environment. Congress of the United States. Washington, D.C. June 1981.
U. S. Environmental Protection Agency. Guidelines for Carcinogen Risk Assessment.
Carcinogen Assessment Group, Office of Health and Environmental Assessment. Washington,
D.C. 51FR33992. September 24, 1986.
U.S. Environmental Protection Agency. Reference dose (RfD): description and use in health risk
assessment. Integrated Risk Information System (IRIS). Appendix A: online. Cincinnati, OH:
Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office.
1987.
U. S. Environmental Protection Agency. Proposed Guidelines Assessing Female Reproductive
Risk; Notice. 53 FR 24834. June 30, 1988a.
U. 3. Environmental Protection Agency. Proposed Guidelines Assessing Male Reproductive Risk
and Request for Comments. 53 FR 24850. June 30, 1988b.
U.S. Environmental Protection Agency. Occupational Exposure Limit Data in Relation to
Inhalation Reference Doses. Office of Health and Environmental Assessment, Cincinnati, Ohio.
1988c.
2-31
-------
U.S. Environmental Protection Agency. Proposed Amendments to the Guidelines for the Health
Assessment of Suspect Developmental Toxicants; Request for Comments; Notice. 54 FR 9386.
March 6, 1989a.
U. S. Environmental Protection Agency, Air Risk Information Support Center. Glossary of
Terms Related to Health, Exposure, and Risk Assessment. Research Triangle Park, North
Carolina. March 1989b.
U. S. Environmental Protection Agency. EPA 600/8-88/066F. Interim Methods for Development
of Inhalation Reference Doses. Office of Health and Environmental Assessment. April 1989c.
U.S. Environmental Protection Agency. General Quantitative Risk Assessment Guidelines for
Noncancer Health Effects. Second External Review Draft ECAO-CIN-538. Research and
Development. February 1991b.
U.S. Environmental Protection Agency. Metabolism and Pharmacokinetic Test Guideline.
Office of Toxic Substances and Office of Pesticide Programs, 56FR 32537-32545. 1991.
U.S. Environmental Protection Agency. Guidelines for Developmental Toxicity Risk Assessment,
Office of Health and Environmental Federal Register, Volume 56, No. 234, 63798-63826, 199 la.
2-32
-------
3.0 DOSE-RESPONSE ASSESSMENT - CARCINOGENS AND NONCARCINOGENS
3.1 INTRODUCTION
If the hazard identification process indicates that a chemical is likely to cause an adverse
health effect, then a assessment of the relationship between dose and response is conducted. The
NAS (1983) defined dose-response assessment as "the process of characterizing the relation
between dose of an agent administered or received and the incidence of adverse health effects
in exposed populations and estimating the incidence of the effect as a function of human
exposure to the agent. It takes account of intensity of exposure, age pattern of exposure, and
possibly other variables that may affect response, such as sex, lifestyle, and other modifying
factors."
Critical to a dose-response assessment is the basic assumption about whether thresholds
X-
exist for particular compounds and particular health effects of concern. In general, the EPA
assumes that carcinogenesis is a nonthreshold phenomenon whereby any exposure, no matter how
low, contributes to an increased lifetime probability of developing cancer. However, evidence
in the scientific literature exists which describes mechanisms of carcinogenesis that support the
concept of threshold. By contrast, pollutants causing noncancer health effects are typically
defined as having a threshold exposure concentration or dose below which adverse health effects
are not expected to occur. There is, however, some discussion in the scientific community of
the non-threshold nature of certain noncarcinogens. In general, the EPA treats carcinogens as
having thresholds and noncarcinogens as possessing thresholds. The threshold concept influences
the way in which dose response modeling or dose-response assessment is done. Thresholds are
assumed to exist because of homeostatic, compensatory, and saturating biochemical.
physiological, or metabolic processes in the exposed organism.
Figure 3-1 depicts the general dose-response curve, or the direct relationship between dose
and the incidence or probaoility of response in a population. It also shows the differences
between nonthreshoid and threshold toxicants. The dose response curve initiates at the origin or
some background value for nonthreshoid toxicants; that is, it is assumed that there is no threshold
and that any exposure equates with an increase in risk. Thus, as dose increases, so does rhe
probability of effect. Conversely, the threshold dose-response curve initiates at the threshold
3-1
-------
Non-threshold
Threshold
CD
V)
i
CO
Dose
Figure 3-1. Dose-response curve.
3-2
-------
dose. In other words, the amount of risk would be zero below the threshold dose. Above the
threshold, the curve increases directly with dose.
Threshold and nonthreshold concepts serve only as general guides, however, because the
threshold for any substance varies from individual to individual, especially for sensitive
subpopulations. Given a wide range of variability of individual thresholds, a population threshold
which would predict all or most individuals within a population is never clearly defined. Also,
recent research suggests that cellular repair mechanisms exist that can reverse the damage caused
by a carcinogen, and it is likely that these mechanisms operate most effectively after low doses
or in the absence of repeated doses.
Data selection is essentially the same for cancer and noncancer dose-response assessment.
Therefore, this topic is discussed first, in Section 3.2. Other aspects of dose-response assessment
specific to carcinogens are covered in Section 3.3, while Section 3.4 covers aspects of dose-
response assessment specific to noncarcinogens.
3.2 SELECTION OF DATA FOR DOSE-RESPONSE ASSESSMENT
A wide range of data can be used for dose-response assessment. The types of data used
include epidemiological studies for human exposures, controlled human experimentation (although
this is available only for a limited number of pollutants), in vivo animal bioassays, and possibly
short-term genotoxic studies and other in vitro studies for noncancer effects.
However, the two primary data sources for dose-response assessment are epidemioiogic
studies conducted in human populations and toxicological studies using in vivo animal bioassays.
These two types of studies were described in Section 2.2. This section focuses on the key
considerations for selecting and using epidemioiogic and toxicoiogicai data in relation r.o
carcinogen and noncarcinogen dose-response assessment.
3.2.1 Epidemioiogic Data.
The EPA states that adequate epidemioiogic data is preferred over animal data when
conducting a dose-response assessment (EPA, 1986). The main reason for preferring
epidemioiogic data is that extrapolation from animal to humans is not required. This eliminates
the uncertainty associated with interspecies extrapolation.
One of the main drawbacks of epidemioiogic studies is that exposure levels are sometimes
difficult to quantify. This problem is particularly serious when attempting to ascertain past
3-3
-------
exposures for cases and cohorts. Records of the exact amount of an agent in the air at a
particular time in the past are unlikely to exist. In fact, accurate records are not commonly
available for establishing the level of exposure for a group of individuals. Even if the overall
cohort's exposure can be characterized, the variation in exposure at the individual level can be
very large due to other exposures such as smoking, alcohol consumption, pesticide exposure
while gardening, diet, etc. For a more detailed description of the strengths and weaknesses
associated with epidemiologic investigations, see Section 2.2.1.
3.2.2. lexicological Data.
Epidemiologic studies are not available for most environmental agents and specific
exposure data may be lacking. Therefore, toxicological data from animal studies or bioassays
are commonly used for cancer and noncancer dose assessment. The criteria for evaluating the
Xs
adequacy of long-term in vivo animal bioassays are presented in Section 2.2.2. Animal studies
have two primary advantages over epidemiologic studies: 1) dose, environment, and extraneous
exposures are strictly controlled, and 2) adverse affects are directly measured through
pathological examination and necropsy.
In addition to the criteria listed in Section 2.2.2 for assessing the adequacy of animal data,
there are other important criteria to consider when selecting animal data for a dose-response
assessment. If epidemiologic studies are not available, data from a species that responds in a
manner similar to humans should be used (EPA, 1986). It is also important to consider the route
of administration in comparison to the human route of exposure u.e., inhalation, dermal contact,
ingestion). Ideally, the routes snouid be identical. However, an extrapolation from the animal
route of administration to a different human route of exposure may be considered. In order to
perform route-to-route extrapolation, pharmacokinetic data for the agent are desirable. The target
organ(s) and mechanism(s) of action must be considered to determine if route-to-route
extrapolation is appropriate. For an agent causing adverse effects at the point of contaa (e.g.,
skin, lung; route-to-route extrapolation would be questionable. But for carcinogens and
noncarcmogens with a systemic mode of action, route-to-route extrapoianon may be biologically
plausible. Pharmacokinetic data can be used in physiologically-based pharmacokinetic models
to convert the dose to a different route.
3-4
-------
In cancer dose assessment, tumor incidence should be combined when a significantly
elevated tumor incidence for multiple sites or tumor types is detected. Malignant and benign
tumors should also be combined unless the benign tumors are not considered to have potential
to progress to the associated malignancies of the same histogenic origin (EPA, 1986). The
rationale is that, given time, the benign tumor may progress to a malignant tumor.
The high background incidences of cancer at particular sites in certain species of animals
must be considered in performing a dose assessment. For example, the B6C3F1 mouse has high
background liver tumor incidence, the Swiss-Webster female mouse has a high background
incidence of mammary carcinoma, and the Fischer 344 male rat has a physiological predisposition
to kidney tumors. If the cancer incidence from the bioassay is statistically higher than
background incidence, then the data should be considered sufficient evidence for carcinogenicity
•v
(EPA, 1986). These data needs to be evaluated on a case-by-case basis in order to assure
scientific validity.
Exceedances of the maximum tolerable dose (MTD) can result in problems in interpreting
the scientific validity of animal bioassays. At doses above the MTD, normal enzymatic and other
biological processes become overwhelmed, and responses not typically encountered below the
MTD may result. Doses above the MTD may also alter the normal metabolic activation of the
carcinogen in a way that reduces the carcinogenic potential (OSTP, 1985). Positive and negative
studies that exceed the MTD should be carefully reviewed to ensure that the responses are not
due to factors chat do not operate at exposure levels below the MTD (EPA, 1986).
In noncancer dose-response assessment, criteria for evaluating the adequacy of animal
studies will vary depending upon the health effect of concern (i.e., reproductive, hepatotoxic. etc.)
and the thoroughness of endpomts analyzed. However, general criteria include chemical
characterization of the test compoundfs), the number of animals in the experimental groups and
whether both sexes are used, the number of experimental groups, the spacing and choice ot
dosing levels to determine an adequate dose-response relationship, the types of observations and
methods of analysis, the nature of pathologic changes, the consideration of pharmacokinetics, and
the route and duration of exposure to determine the appropriateness and relevance of the study
for human exposure (EPA, 1990a). For establishing the adequacy of animal data for specific
-------
health endpoints, the various EPA guidelines should be consulted (EPA, 1986a; EPA, 1988a;
EPA, 1988b; EPA 199la).
3.3 CARCINOGENS
3.3.1 Overview
In a dose-response assessment, a predicted human response is quantitatively determined
for any given level of exposure to a carcinogen (OSTP, 1985). The two basic reasons for
conducting a cancer dose-response assessment are 1) to extrapolate from high to low doses, and
2) to extrapolate from animal to human responses. Both epidemiologic and toxicologic studies
typically require doses higher than those normally encountered in the environment. Therefore,
in order to determine response at lower doses, an extrapolation from high to low dose must be
performed. Many models are available for dose-response estimation and high to low dose
extrapolation. These are described in detail in Section 3.3.3". The dose-response assessment must
also extrapolate from animals to humans if only animal data is available. This interspecies
extrapolation is carried out by applying a scaling factor to the experimental data (OSTP, 1985)
or through the use of physiologically based pharmacokinetic (PBPK) data.
The end result of the dose-response assessment is the quantification .of a carcinogenic
response associated with various dose levels. This end result is expressed at the individual level
as the Unit Risk Estimate (URE). The URE is described in detail in Sections 3.3.4 and 3.3.5.
For most chemicals, the dose-response assessment can proceed if the EPA weight-of-
evidence classification is Au B1; 8,, and C (EPA, 1989a). Direct evidence of carcinogemcity in
humans is not required; well-conducted animal studies that demonstrate carcmogenicity can
provide a plausible basis for conducting a dose-response assessment. In fact, a large portion of
the carcinogenic compounds were classified as carcinogens based on animal data.
Since dose-response models have varied strengths and weaknesses, some understanding
of the tneones of carcmogenesis is important in applying the models. Therefore, a. aescnption
of the carcinogenic theories (concentrating on the multistage process), the threshoid/nonthreshoid
concept, and sensitive subpopulations is presented. A description of the process of carcmogenesis
is presented in Section 3.3.2. Mathematical dose-response extrapolation models are oresented
in Section 3.3.3 and means for expressing dose-response relationships are contained in
Sections 3.3.4 and 3.3.5. Finally, methods and data sources for dose-response assessment are
3-6
-------
summarized and described in terms of criteria that could be used to select a method
(Section 3.3.6).
3.3.2 Process of Carcinogenesis.
During the carcinogenesis process, controlled, normally dividing cells are converted to
uncontrolled, rapidly dividing cells. The rapid proliferation results from a presently unknown
alteration at the cellular level, caused by the carcinogenic agent. Once the cells multiply and
form a group of uncontrolled dividing cells, they are called a tumor. A significant aspect of the
carcinogenic process is that the time between exposure to an agent and the manifestation of the
actual cancer is usually decades. The magnitude of carcinogenic exposure may be typically low
and can occur over an extended period of time.
The causes of the cellular alteration and the subsequent uncontrolled cell proliferation in
the cancer process are poorly understood. Presently, most health professionals assume that a
large number of carcinogens act directly on the genetic material, called the deoxyribonucleic acid
(DNA). This type of agent is called genotoxic. A genotoxic agent is defined as a chemical with
the ability to damage DNA or chromosomes (EPA, 1991). Similar to a genotoxic agent, a
mutagenic agent is capable of causing a permanent change in the structure of the DNA.
Although change caused by a mutagenic agent is not necessarily harmful, a mutational event is
considered one of the initial steps in the overall multistage process of carcinogenesis. A number
of carcinogens have also been determined to be mutagens in short-term assays such as the Ames
microbiological assay.
It was initially believed that a single exposure to a carcinogen, regardless of magnitude,
could start the carcinogenic process. This "nonthreshoid" concept of carcinogenesis stared that
there is no "zero risk" dose in terms of carcinogen exposure. However, recent advances in
molecular biology indicate cellular mechanisms exist that are capable of repairing the early
damage of a carcinogen. The determination of whether a carcinogen exhibits a threshold or not
should be based on the latest available biological evidence and be a case by case decision.
However, as a pracucal matter, humans are exposed to many carcinogens and the ability to repair
damage will vary among individuals within a population. Therefore, unless there is ciear
evidence to the contrary, cancer dose-response assessments usually assume no threshold.
j-/
-------
Although a number a theories exist to explain the process of carcinogenesis, the multistage
process is the most widely accepted. The multistage process consists of three distinct stages:
initiation, promotion, and progression (Weinstein, 1985). One reason the multistage process is
so well accepted is that it has been experimentally demonstrated for a number of carcinogens.
It has been shown to adequately describe carcinogenesis in the cells of some animal tissues,
including the skin, lung, liver and bladder (NRC, 1986). Also, some specific agents that are
specifically initiators and/or promoters have been identified.
During the first stage, initiation, the DNA in a cell undergoes a heritable change that is
manifested as damage or modification to the DNA (NRC, 1986). With some chemical initiating
agents, a single exposure may be sufficient to cause this mutational change (Weinstein, 1985).
The second stage, promotion, is characterized by the replication and/or proliferation of the newly
altered cell. Unlike initiators, agents that are promoters areTnot believed to interact with DNA,
but rather to somehow affect the normal regulatory activities of the cell. The final stage is
progression. Although this stage is not yet well defined, it is known that in this stage cells form
tumors and possess the ability to metastasize (transfer cancerous ceils to other parts of the body;
(OSTP, 1985; EPA, 1991). The ability of cells to metastasize distinguishes a malignant tumor
from a benign tumor. Benign cells are similar to malignant ones except they do not possess the
ability to metastasize, or spread and invade other tissues.
Certain groups of individuals within the population are inherently more sensitive to
carcinogen exposure than others. Factors that influence susceptibility include age, race, sex, and
genetic predisposition among others. An example of a sensitive subpopuiation is children. This
subpopuiation can be more sensitive to certain chemicals and more susceptible to cancer for a
variety of reasons, including:
Children have faster breathing rates than adults and, thus, inhale larger quantities
of a pollutant, relative to their body weights;
• Organs in children are still growing and developing and are, therefore, more prone
to disruption by an environmental agent;
Children spend substantially more time outdoors than adults and therefore are
exposed to ambient pollutants for longer duration; and
3-3
-------
• Young organisms appear to be inherently suspectable to many carcinogens.
Young experimental animals have been shown repeatedly to acquire more tumors
in a shorter time with a smaller dose than adult animals.
In most circumstances, there are not enough scientific data available to perform separate
quantitative dose-response assessments for these sensitive subpopulations. However, if
information is available, it should be considered in a dose-response assessment.
3.3.3 Mathematical Dose-Response Extrapolation Models.
Experimental animal studies use higher dose levels than would be normally encountered
in the environment. Extrapolation from these high to low environmental doses is typically
required. Dose-response models available for use in extrapolating from high to low doses are
described in this section.
No single dose-response model is appropriate in all situations. A dose-response model
is usually selected on an agent-specific basis. However, two categories of dose-response models
are generally used in carcinogen risk assessment: mechanistic models and tolerance-distribution
models.
Mechanistic models describe some mechanism by which carcinogenesis is believed to
occur. All of the mechanistic models assume that a tumor originates from a single cell that has
been altered by either the agent or one of its metabolites (OSTP, 1985). Examples of
mechanistic models are the one-hit, multi-hit, and multistage models.
The one-hit modei assumes that a single hit at a critical site can result in malignant
transformations. This model is highly conservative (i.e., reduces the chance of underestimating
the risk) because it does not account for cellular or DNA repair mechanisms. The one-hit mode!
is expressed by the equation
Pfd) = I - exp - (a + bd), a, b > 0
where:
P(d) = the probability of response at dose (d),
a = background incidence, and
b = a measure of potency of a test agent.
3-9
-------
Because the one-hit model has only one parameter other than background, it does not always fit
experimental observations (OSTP, 1985).
The multi-hit mechanistic model, an adaptation of the one-hit model, assumes that more
than one chemical exposure or biological event is required to elicit a carcinogenic response. The
following relationship describes this model:
bd
f vfc-i
P(d) =
where:
P(d) is the probability of response at dose d,
T (k) = (K-l)! for K > 1,
b = a measure of potency of a test agent,
k = the number of chemical "hits," and
x = the expected number of "hits."
The multistage model is the most frequently used of the low dose extrapolation models.
It is the model most frequently used by the EPA in conducting does-response assessments. This
model assumes that a number of mathematical stages may be useful in representing data n the
observed range and achieve a better fit to the data. -Like the one-hit model, the multistage model
is approximately linear in the low-dose region as EPA does and, therefore, is thought to be
relatively conservative. The version of the linearized multistage model most commonly employed
by EPA was developed by Crump et al. (1977) and is expressed as follows:
P(d) = i - exp Hq0 + q,d + q2d2 + ... + qkd*)], K > 1
where:
P(d) = the probability of cancer at dose d,
k = the number of stages, or k may also be assumed to be equal to the
number of dose levels minus one,
3-10
-------
qk = coefficients to be Fit to the data, and
dk = the applied dose raised to the kth power.
The second type of dose-response model, the tolerance-distribution model, is a form of
empirical model. Tolerance distribution models assume that for each individual in a population
there is a tolerance level below which that person will not respond to the exposure (OSTP, 1985).
These models assume a variability among individual tolerance levels that can be described in
terms of a probability distribution. This concept of individual tolerance levels differs from the
"threshold" concept used in most noncancer risk assessment, which posits a general level of
exposure that is "safe" for most of the population. Tolerance distribution models are actually
based on the "nonthreshold" concept of carcinogenesis, because they refer to an infinite number
x«-
of individual tolerance levels or thresholds distributed along a curve. The low-dose extrapolation
techniques based on the tolerance distribution theory include the probit (long-probit), logit
(log-logistic), and the Weibull model.
The probit model assumes a log-normal distribution of sensitivities in the human
population. It also assumes that the slope of the dose-response curve is dependent upon the •
background cancer rate and that the carcinogen is only supplementing an ongoing carcinogenic
process. The logit model is expressed in the following equation:
P(d) = <|> (a -r b log d)
where:
Pfd) = the probability of response at dose d.
q> = the standard cumulative normal distribution function evaluate at
(a + b log d),
a = the intercept (background incidence,), and
b = the slope of the log probit distribution.
Both the logit model and the probit model are symmetrical about the 50 percent response
level. However, the logit model approaches the extreme ends of the curve more slowly than the
probit model (Brown. 1982). The logit model was derived from the chemical kinetic theory and
3-11
-------
generates risk estimates that are lower than the probit model but the same as the Weibull model.
The logit model takes the form:
P(d) = [1 + exp - (a + b log,0 (d))]'1, a, b > 0
where:
P(d) = the probability of response at dose d,
a = the intercept (background incidence), and
b = the measure of potency of a test agent.
The Weibull model falls into the category of the tolerance distribution models, but it can
also be applied as a multi-hit mechanistic model when addressing multiple target cells. (As a
x»
tolerance-distribution model, the Weibull model assumes that cancer begins in a single cell).
This model attempts to account for the fact that some exposed individuals will die before they
have a chance to develop cancer. The Weibull model is expressed as:
P(d) = 1 - exp - (bdk), b, k > 0
where:
P(d) = the probability of response at does d,
b = the measures of potency of a test agent, and k is the number of
stages or events, and
k = the number of stages of events.
The ultimate selection of a dose-response model is either driven by scientific reasons or
by policy. Historically, EPA has found that the scientific reasoning is generally lacking and so
a policy choice of the linearized multistage model has been made. Recent advances in EPA
research suggests that a new generation of biologically based models may be on the horizon.
Given that ail currently recognized models and defaults for not knowing the true slope of the low
dose-response curve all current models are curve-fitting models. Curve fitting in the observed.
data range may not be influential in predicting the accuracy of defining a curve in the low dose
region. In Figure 3-2, response (risk) is plotted versus the dose of aflatoxin (a naturally
3-12
-------
P(d)
OH MS
10-*
10"
d
10"
Model
OH-One Hit
MS-Multistage
W-Welbult
MH-Multi-Hit
FIGURE 3. -2. Log-Leg Plot of Risk, P(d), vs. Dose, d, of Aflatoxin for
Four Doss-Response Models
3-13
-------
occurring mold) for five different dose-response models. As this figure shows, predicted risk
varies considerably depending on the dose-response model chosen. The selection of the
dose-response model should be considered and explained.
If animal data are used in the dose-response assessment, scaling factors are commonly
used to calculate a human equivalent dose. These scaling factors are applied to animal data to
account for differences between humans and animals regarding body size, life span, route,
metabolism, and duration of exposure (EPA, 1986). A scaling factor is also applied to some
epidemiologic studies, where the duration of exposure is less than a lifetime.
When making the interspecies comparisons, standardized dosage scales such as mg/kg
body weight/day, ppm in the diet or water, mg/m2 body surface area/day, and mg/kg body
weight/day are commonly used (EPA, 1986). The EPA prefers extrapolation on the basis of
>-
surface area (though other approaches may be defensible) because particular pharmacologic
effects commonly correlate to surface area. Because the body surface area is proportional to the
animal's weight to the two-thirds power1, and weight is more easily determined than surface
area, equivalent dose can be calculated as follows:
60 mg/0.20 kg*3 = x/70 kg273
where:
x = equivalent human dose (mg)
0.20 = weight of rat (kg)
70 = weight of average human (kg)
solving for x gives:
60 mg/0.34 = x/17
x = 3000 mg
Similarly, an equivalent average lifetime exposure for animals can be calculated if the available
data is from a less than lifetime exposure. An example is given beiow.
'The EPA is currently considering using, the 3/4 power to replace the
current (body weight) 2/3 power conversion.
3-14
-------
Example: Rats were exposed, via inhalation, to 300 ug/m3 of a carcinogenic agent for 6
hours/day, 5 days/week for 65 weeks.
Therefore, the equivalent lifetime exposure for rats (expected lifetime of 110
weeks) is:
300 ug/m3 x 6/24 hours x 5/7 days x 65/110 weeks = 32 ug/m3
3.3.4 Unit Risk Estimates for Inhalation Exposure.
A unit risk estimate (URE) represents an estimate of the increased cancer risk from a
lifetime (70-year) exposure to a concentration of one unit of exposure. The URE for inhalation
is expressed as risk per ug/m3 for air contaminants. The URE is a plausible upper-bound
estimate of the risk, upper bound meaning that the true risk, which we cannot define, is not likely
to be higher but may be lower and may be close to zero in some cases. An intermediate form
of the URE, which is sometimes used, is in units of risk per mg/kg/day, and is called a slope
factor or q,*. The slope factor is mathematically the slope of the linear extrapolation line in the
linearized multistage model. The qt* represents the 95% upper confidence limit on that slope.
The following equation is used to convert a slope factor to a URE for air contaminants (although
this conversion is not recommended without toxicologic justification):
URE = Slope Factor x 1/70 kg x 20 mVday x 10°
The URE is focused on an adult who is a 70 kg individual with a breathing rate of 20 nr/day
and is exposed to the carcinogen over a 70-year lifespan. The multiplication by iO"J in the URE
equation is required to convert from mg to ug. The URE is a linear proportional relationship,
whereas if a nonlinear low-dose-response extrapolation model were used, the unit risk would have
to be expressed as a more complex equation.
If the URE is derived from animal data, it usually represents the upper 95th percent
confidence limit of the unit risk. In other words, there is only a 5 percent chance that rhe actual
unit risk could be greater than the URE derived based on the animal data and model used. Using
the upper 95th percent confidence limit, greatly reduces the probability of underestimating the
unit risk.
3-1
-------
If epidemiologic data are available, either an upper bound URE estimate is estimated or
in some cases a maximum likelihood estimate (MLE) of the URE is derived. The MLE is
defined as a statistical best estimate of the value of a parameter from a given data set (EPA,
1991). The difference between the upper bound estimate and the MLE is that the upper bound
is more conservative, in the face of uncertainty whereas the MLE is better if there is a significant
level of confidence in deriving point estimates of risk. An MLE approach is better to use with
large numbers of data points.
A great deal of uncertainty is associated with the URE because many assumptions have
been made in the process of deriving it and the uncertainties are within the perspective of "upper
bound" estimation. The uncertainty arises from several areas in a dose-response assessment
including selection of the extrapolation model, intra- and interspecies variability, high to low dose
extrapolation, route to route extrapolation, and the development of equivalent doses. When
presenting a URE, it is important to also present the uncertainties associated with both the URE
value and the weight-of-evidence for the carcinogen. In presenting the uncertainties'along with
the URE one is insuring that this information will be passed to the risk manager so that undue
confidence will not be placed on the point estimates of for a particular carcinogen.
3.3.5 Other Ways of Expressing Dose-Response Relationships.
In addition to the UREs for the inhalation route, UREs for both the oral and drinking
water exposure pathways can be derived. The calculation of oral and drinking water UREs is
similar to calculation for the inhalation pathway in that the slope factor is converted to a unit
risk. UREs for oral and drinking water exposure are calculated as follows:
Water URE = Slope Factor x 1/70 kg x 2 L/day x 10'3
Oral URE = Slope Factor x 1/70 kg x 10"3
The units for the water and oral UREs are ug/L and ,ug/day, respectively.
For chemical mixtures, dose or response additivity is a feasible way to generate a URE.
The additivity concept assumes that the carcinogenic response of the individual chemicals in the
mixture is additive in the low dose region. The 3PA is developing two other alternative
approaches for quantifying dose-response relationships for mixtures: a) The comparative
potency approach, and b) the toxic equivalency factor (TEF) approach. Both of these approaches
use in vitro and short-term in vivo data on chemical mixtures as a surrogate to estimate a dose-
3-16
-------
response relationship. All three of these approaches for chemical mixtures assume that the
undivided components exert their effects through a similar mode of action.
The comparative potency approach is based on the assumption that the ratio of two
potencies is constant, whether they are based on comparisons between human studies, in vivo
assays or in vitro assays (EPA, 1988). If a relationship is established between in vitro and in
vivo potency, then the results of in vitro tests on other chemically related complex mixtures can
be converted to potency. For example, the human potency factor of a poorly-studied mixtrue can
then be estimated from its in vivo (or in vitro) potency multiplied by the potency ratios of a well-
studied similar mixture. The carcinogenic potency of combustion emissions has been estimated
by the comparative potency approach (EPA, 1988).
The TEF method can be used for complex mixtures that contain some components that
X'
are not well understood but are similar in mechanism of action. The TEF approach has been
useful in dose response assessments for chlorinated dioxin and dibenzofurans where mixtures
consist of structurally related congeners (EPA, 1988). This approach uses a set of derived TEFs
(based on acute and in vitro data verified with any available chronic or subchronic data) to
convert the Concentration of any mixture component into an equivalent concentration of another
well-studied reference component by multiplying the each TEF by its compound's concentration
in the mixture. The equivalent concentrations for each component of the mixture are then added
and the sum multiplied by the potency of the reference compound to derive a risk estimate (EPA,
1988).
3.3.6 Methods and Data Sources for Dose-Resoonse Assessment.
Table 3-1 summarizes the various methods available for conducting a dose-response
assessment for carcinogenic compounds. The first method is a review of the IRIS data base to
determine if a dose-response assessment has been performed by EPA. .\fost of the UREs
contained in the IRIS data base have been generated with the multistage model. All of the UREs
and/or slope factors in IRIS have been verified by the Carcinogen Risk Assessment Verification
Endeavor (CRAVE) workgroup. The CRAVE workgroup is comprised of EPA epidemiologists,
toxicologists, medical doctors, biostatisticians, and other health professionals. These experts
verify that Agency criteria have been followed. Since the CRAVE workgroup is continually
3-17
-------
CO
a
cc
£
o
UJ
CO
O
a.
CO
LU
CC
LU
CO
C3
O
a
cc
o
UJ
—I
<
,i
Review primary literature ai
develop an URE or slope fc
S2
%
•S
1
"
CD
.2
1
5
lr
2
s
s
Locate other soi
slope factors
i ll
cr:
3.
CO
£
-s
S
0)
j
URE or slope factor derivec
the available literature
° §
CO g
1 1
11
II
UJ O
59-
OB
2
URE or slope fa
-S
•-•-.
lit
ai _
ill
... —
? j -
lit
01
CO |=
iil
!H ^5
lit
i-
"i ^ as
11-1
2 2 i
,„
^
of
"O -^3 """
1 3 3
"""* irt ~^
^•jc 1 :i to 8 „,
ctP
LI_ TO ~— -
^3 ' ttl
c S.CT
ra 55 ~
.c 2 >• 'S a. S
2 - 03 » "3
0
5
1
c
•5 «o
IS ,*
• Requires thorough knowlc
the available literature and
dose-response models
• Subjective evaluation of It
literature is needed to setec
critical study to be used
• Peer-review process is
recommended
_ S
^ CO
flo
•o
O "c5 Q_ .£ en
e- _ m ^ en
^
Pin n
i .= 1 1 1 1 ? i
• .£ a. • en co • k-
1 £i
S ® ™^ ^
s 1 f ^ |
* S •§ -1 2
CO ^< CD Q3 CO
•^ (L Q ^^
I ts •= 1 g
c <= 2 « -|
c >» 3
03 *» t/J
S T3 0 OJ
-" 'U to c
*- 2 °5
i > .»• c
tO "O CD 'o
= « T3 O
= 2 -5 1
cp ,1 -a —
-g (D O T3
•o
i
^8
C
ci
T3
J
Screening to Re
-o
3J
i
o
5
ToxicologisI, epidemiologist
other health professional
S
Jrt
.££
1 1
's-'a
to 5.
«5~ £
f 1
II
^-
Ol
i
Scientist or toxic
33
"i
I
c_
"2
'"l3
c
01
o*
r~
IB
f " 1 §
__ ju_ yj c _^ c "32
. LU O . ca
C/I
03 /j
"«
"8
to
o
Q
3
= 3
Jt
dS
1
5
5
o
1
o
S
(ij X U
_i r2 1 '-U OC
1
O)
CD
X
1
|
-a
£
g.
-u
IX
cu
U
^2
CC
3-13
-------
reviewing and updating the carcinogen information in IRIS, the on-line data base should represent
the most up-to-date information on EPA-verified UREs and slope factors. If information is
available for a compound in the IRIS data base, this method will require the least amount of time
because the URE or slope factor has already been derived and verified by the EPA. Minimal
expertise is needed for using the IRIS data base since the critical study is identified and all the
information is summarized.
If a dose-response assessment for the compound of interest is not found in the IRIS data
base, another option would be to search for other sources containing developed UREs or slope
factors. Possible sources of URE and slope factor values include state agencies and published
reports. When using other sources, it is necessary to evaluate the extensiveness of the review
of available information that resulted in the URE or slope factor. An additional review may be
>•
required verify that the proper study was chosen as a basis for dose-response assessment.
Locating other sources of UREs or slope factors is more time consuming than searching IRIS and
could require additional expertise to review the data.
If no UREs or slope factors can be found in existing sources, two other methods axe
available. One method is to extrapolate the URE or slope factor from another route of exposure.
This is a complicated process requiring knowledge and application of physiologically-based
pharmacokinetic models. The other method is to review the primary literature and derive an
URE or slope factor from the available data. This requires knowledge of the dose-response
modeling techniques and involves subjective evaluation of the literature to identify the critical
study.
These two methods are the most time and resource intensive and require use of complex
techniques. Route-to-route extrapolation must be done by a knowledgeable health professional
because physiologically-based pharmacokinetic modeling has to be conducted. Review of the
primary literature also requires the expertise of a health professional. This method may be
slightly more time and resource intensive than route-to-route extrapolation since the critical study
has to be identified and a dose-response model selected.
Table 3-1 also presents information on level of study, expertise required, resources
required, and relative time required for each method discussed in the previous section.
3-19
-------
3.4 NONCARCINOGENS
3.4.1 Overview
If the hazard identification process determines that the compound exhibits noncancer
health effects, then the dose-response assessment is conducted. As with carcinogens, proof of
systemic toxicity in humans is not essential for performing the dose-response assessment, since
well-conducted animal studies provide adequate evidence. The methods used in the noncancer
dose-response assessment generally assume that a threshold exists below which adverse effect
are unlikely to result.
Four different approaches have been developed for performing a noncancer dose-response
assessment: 1) the structure-activity relationships (SAR) approach, 2) the chronic inhalation
reference concentration (RfC) method, 3) dose-response modeling, and 4) the decision analysis
approach (EPA, 199la). These noncancer dose-response approaches are listed in order of
increasing data and/or resource requirements, with the SAR approach requiring the least data and
the decision analysis approach the most resources. The approach most commonly used by the
EPA in noncancer air dose-response assessment is the inhalation RfC method. Through the RfC
methodology, a benchmark concentration is calculated below which adverse effects are not
expected to occur.
Most approaches developed for noncancer dose-response assessments are based on the
premise that a threshold exists for the adverse effects. This noncancer threshold concept is in
direct contrast to the view of carcinogens, which are assumed to be capable of eliciting a
response at any concentration. Although it is generally agreed that thresholds exist, the actual
threshold may vary significantly from individual to individual based on susceptibilities. In fact.
certain individuals may exhibit adverse effects even at very minute exposure levels. Given the
wide variation in individual response, a single threshold may not be representative of the
population as a whole. Therefore, any identified threshold level must be regarded as only a
general guide, and individual susceptibility must be considered when selecting the data to be used
in the noncancer dose-response assessment.
Concentration and duration of exposure are other important considerations when selecting
data for the noncancer dose-response assessment. High concentrations over a short period of time,
low concentrations over a long period of time, or high repeated exposures over a short time span
3-20
-------
can cause different types of adverse effects. Most risk assessments for routine air toxic emissions
from stationary sources focus on the effects of chronic exposures (low concentrations over a long
period of time). When selecting data for a dose-response assessment, the duration of exposure
in the study must be considered. If it does not reflect a chronic exposure, the equivalent chronic
or lifetime exposure must be calculated.
The normal output of a noncancer dose-response assessment for hazardous air pollutants
is the RfC or RfD. The following section explains the derivation of the inhalation RfC and the
oral RfD, while Section 3.4.3 briefly describes other dose-response assessment methodologies.
Finally, methods and data sources for dose-response assessment are summarized and described
in terms of criteria that could be used to select a method (Section 3.4.4).
3.4.2 Derivation of RfCs and RfDs.
~^^^^^^—"""• ^^^^^^^—P— ^^^^^^^^-^^^— x^
RfCs and RfDs are benchmark concentration values that represent estimates of the "safe"
daily exposure of the human population to a specific agent. Reference concentration (RfC)
applies to exposures through the inhalation route, and therefore, is commonly referred to as the
inhalation RfC. An exposure level derived for the oral route of exposure is typically referred to
as the oral reference dose (RfD). The RfC or RfD represents an estimate, with an uncertainty
of one order of magnitude or more, of the lifetime dose that is likely to be without significant
risk to the population including sensitive subgroups (Barnes and Dourson, 1988). These
estimates are based on the no-observed-adverse-effect-level (NOAEL) for a compound determined
from epidermoiogic or toxicologic studies.
As discussed in Section 2.4.2, the critical effect is a specific adverse effect that occurs
at the lowest dose in adequately designed and conducted animal studies. The lowest dose level-
at which the critical adverse effect is observed is called the lowest-observed-adverse-effect-level
(LOAEL). The EPA reviews many animal studies that have demonstrated the critical effect at
a variety of LOAELs. The EPA generally looks for a study that represents the highest level
dose tested in which the critical effect does not occur. This level is selected as the NOAEL.
.(See Figure 3.1.)
Selection of the NOAEL is important because it is the first component in the estimation
of the RfC. The following steps are useful in determining the appropriate study, species, and
NOAEL for RfC estimation (EPA, 1990a):
3-21
-------
• Choose the most appropriate NOAEL of the critical effect from a well-conducted
human study, or
• Choose the most appropriate NOAEL of the critical effect from a well-conducted
study on a species that is thought to resemble the human response to this
particular chemical (for example, by comparative pharmacokinetics), or
• When the above condition is not met, choose the most sensitive study, species,
and NOAEL as judged by an interspecies comparison of the NOAEL and LOAEL.
If adequate human data are not available, a human equivalent concentration (HEC) is
estimated for the dose-response assessment. The NOAEL from an animal study is converted to
a NOAEL[HEq by adjusting for exposure duration and respiratory tract differences between
animals and humans. A detailed description of the NOAEL to NOAEL[HEq conversion is
included the "Interim Methods for Development of Inhalation Reference Concentrations" (EPA,
X*
1990).
The second step in the quantitative estimation of the inhalation RfC is the application of
an uncertainty factor (UF) and a modifying factor (MF) to the NOAEL[HEC1. These three values
are incorporated into the following equation to derive a RfC:
RfC = NOAEL[HEC, / (UF x MF)
(1)
where:
NOAEL[HEC1 = NOAEL, adjusted for dosimetric differences between animal species and
humans, expressed as human equivalent concentration,
UF = an uncertainty factor suited to the characteristics of the data' and
MF = a modifying factor based on professional judgement of the entire data base
(e.g., sample size).
Uncertainty factors are applied to adjust for uncertainties in extrapolating from the type of study
serving as the basis for the RfC to the situation of interest for the risk assessment (EPA, 1989bj.
A modifying factor between one and ten is also added to reflect professional judgement of the
-------
scientific uncertainty associated with data available for the specific compound. Table 3-2 lists
UFs and MFs typically used and the rationale for using a certain factor.
The oral RfD (expressed as mg/kg/day) is developed using the same general equation as
the inhalation RfC, using a NOAEL based on the ingestion pathway.
3.4.3 Alternative Dose-Response Methodology.
In addition to the inhalation RfC and oral RfD methodology, several alternative
approaches are available for quantifying noncancer health risks. These include structure-activity
relationships, dose-response modeling, the decision analysis technique, and occupational exposure
limits (OELs) .
3.4.3.1 Structure/Activity Relationships. The advantage of the SAR approach is that it
can be used when very little or no dose-response assessment data exist for an agent. Instead,
data on structurally related compounds is used in calculating noncancer risks. There are four
major components to the SAR approach (EPA, 1990a):
Evaluation and interpretation of available and pertinent data on the chemical under
study or its metabolites;
Evaluation of test data available on "analogous" substances and potential
metabolites;
Generation of "quantitative structure-activity-relationships" to estimate physical
and chemical properties which aid in the selection of analogues; and
Interpretation and integration of available information by the incorporation of
knowledge and judgements by scientific assessors.
The final step is crucial because it determines whether the analogues are reasonably close to the
target compound and whether the data are reliable enougn to generate risk estimates. If the
analogues are close and tne daia reliable, a dose-response assessment is conducted using rfte
analogue data. Of all the noncancer methods, the use of SAR has the lowest confidence level
and greatest uncertainty and requires extensive expertise and interpretation. Therefore, it should
be used only in the absence of other data.
-------
O
S_
CO
tr
LLI
o
O
O
LU
o
S
cc
LU
LL.
LU
CC
oc
01
a
CO
CC
i
>-
1
CC
LU
O
Z>
u_
O
LU
CO
UJ
cr
O
u.
CO
I
1
i
o
= -1
w 55
c
.3
i!
> .o
o ~
ing
p
tl
x "2
a) a3
g.s
i
U3
03
if
" OJ
II
II
'i i
0. o
03 o
O o
CO
CD
3
w
g
^
03
cS
U
03
-Z
_O
CO
O — — 03 2
§
e
03 C= -5
C ca G
2 £
-= 3 en
c « S
CO 03 O-
03 '-5 2
'73 3 ">?
(13
if
« s
1 s?
—
ental animals o
extrapolating fr
s
0)
0
in
03
.
" 13
-2 -§
05 '*z c c
g-i is
*£
01 .en
2.2
o c
ll
•= ^
c —
CO w.
2. ~
o. c
.2 c
U. 03
ra
C\l
CT3
LU
•2
03
|
•o
5
c
S
CO
co
ra
^r
o
CO
en
S
LU
§
.2
J3
a.
S
03
o
c
T3
O
3-24
-------
3.4.3.2 Dose-Response Modeling. Dose-response models can be used in noncancer risk
assessments to characterize the response of target organs along a concentration gradient. These
noncancer dose-response models are similar to cancer models in that risks (or incidence) are
estimated at various exposure levels. Noncancer dose-response modeling is complicated by the
concern over multiple organ systems, each of which may have multiple health endpoints.
Furthermore, for each toxic endpoint there would be multiple levels of severity or intensity of
effect. Therefore, noncancer dose-response modeling may focus on a single endpoint. The
selection of the effect modeled (usually jhe critical effect) is of extreme importance. EPA's
noncancer risk assessment guidelines (EPA, 199la) describes dose-response modeling in great
detail, some of the major elements are summarized below.
The Agency has identified the following main steps in using a mathematical model in
noncancer dose-response descriptions (EPA, 199la):
• description of the risk assessment goal, (e.g., prediction of incidence v. defining
a protective exposure level),
• evaluation of the quality and suitability of the toxicity data in terms of meeting
that goal,
• selection of the appropriate form(s) of mathematical model(s),
• estimation of model parameters (assuming adequate data quality),
• evaluation of the overall quality of the dose-response model in describing the
toxicity data, including a discussion of key assumptions and uncertainties.
Specific guidance on data acceptability cannot t>e given due to the enormous variability
and complexity of toxicity studies. As a general rule, the dose-response model is more plausible
if the toxicity data mimic rhe conditions for which the risk assessment is performed, (e.g., same
exposure route, duration, species, age group, health, reproductive status). In addition, the model
is usually more accurate if it is derived from well-conducted experiments that study responses
at multiple doses and investigate a variety of endpoints.
Selection of Appropriate Mathematical iModei. The selection of a dose-response moaei
can be based on any of several characteristics. Two common criteria are the type of model to
3-25
-------
be developed (empirical or mechanistic) and the type of available data, i.e., whether the response
is reported as a prevalence rate (quantal), a measured value (continuous), or a description of the
severity of the effect (ordered categories).
Models that are intended to be mechanistic involve substantial biological interpretation
of covariates and model parameters. For example, these models may incorporate information
about chemical uptake into the body, pharmacokinetics, mechanisms of toxic action, and
threshold. Examples of mechanistic models include the "hit" models (U.S. EPA, 1984) used for
cancer and radiation-induced teratogenesis (e.g., mental retardation), and pharmacokinetic models
applied to a variety of chemicals. The better mechanistic models are those whose parameters are
independently estimated in separate experiments. In contrast, empirical models are generally
curve-fitting tools , whose parameters have limited mechanistic or biological interpretation. If
•\«
the assessment goal is to estimate risks or doses within or near the experimental range, then
empirical models can usually serve quite well. In such cases, model selection is not critical since
models that adequately fit the data are usually in close agreement for doses near the experimental
range. Empirical models can also be selected for ease of use, (i.e., simplicity of model structure),
such as the simple exponential or polynomial models, with parameters added until adequate fit
is obtained. However, when the goal requires significant extrapolation (e.g., to doses
substantially lower than the data or across species or exposure routes) then empirical models are
less satisfactory. For example, a dose-response relationship for a test animal is not plausible
when applied to humans unless many key biological factors are incorporated into the model.
Some of these factors include appropriateness of the observed effect to human toxicity,
circulating or tissue concentrations of the chemical, time-variability of concentrations of chemical
over the full duration of exposure, and the homogeneity or heterogeneity in the exposed
population.
The model should be selected so it is appropriate for the type of data that is modeled.
Models can be developed to use data related either to dose or exposure. Many of the same
features are shared by both types. Therefore, models using dose will be described beiow.
Several general models have been used or proposed for dose-response assessment, which
collectively can handle nearly all types of data on general toxicity. These are summarized below
according to data type:
3-26
-------
DATA MODEL
1. Dose-response (prevalence) Threshold and nonthreshold models,
tolerance distributions.
2. Dose-intensity (continuous Polynomial and exponential-polynomial
measurement) functions.
3. Dose-effect or dose-severity (ordered Generalized linear models,
categories for severity of effects in dose Cummlative log odds model.
groups or populations)
Dose-response data. Dose-response models may be used to characterize quantal data, i.e.,
where only the presence or absence of an effect is of interest. These models require data on
frequencies of occurrence of a given effect and directly describe the relationship between dose
and risk (probability) of the effect. The threshold models'for dose-response data replace dose
(d) by the dose in excess of threshold (d-d0), but are otherwise nearly identical to the
nonthreshold models (e.g., multistage, one-hit, Weibull) used in cancer dose-response assessment.
Most of the cautions related to the use of the nonthreshold models also apply to their threshold
counterparts. There is no strong biological support for selecting any specific threshold model.
For example, support for the threshold multistage model is limited to descriptions of some
general toxicity in terms of discrete stages; however, experiments have not yet been conducted
to validate the model. As a result, threshold models should be treated as empirical curve-fitting
tools.
Furthermore, if some averaging of several models is performed (e.g., when no model fits
•
significantly better than any other; care must be taken to avoid including several models of the
same class into the average. For example, the one-hit, multi-hit and multistage are similar
mathematical functions; if only two parameters can be estimated, then these three models are
nearly identical. Averaging them with other potential models (e.g., iog prooit) would give
unjustified emphasis to «:he multistage/ hit model. Finally, without direct independent
measurement, the threshold estimate appears only as another parameter estimate, so its
interpretation as a biological threshold is not necessarily accurate (Cox, 1987: Gaylor. 1983).
Since the application of threshold dose-response models has been limited, any use of these
models for significant extrapolation must be justified.
3-27
-------
Tolerance distribution models require the assumption that for every member of a
population there is a critical (threshold) dose below which the individual will not respond. Such
a model then describes the probability distribution of the tolerances (thresholds) across the
exposed population. The well known log-probit and logistic models have been successfully used
in several diverse applications, although neither has a strong theoretical basis for toxic responses.
Distribution models are considered to be empirical, curve-fitting models. Because of their long
history of use, moderate extrapolation beyond the toxicity data set is usually considered plausible.
Logistic models have been used for some air pollutant assessments and are preferred over probit
models in situations involving heterogeneous human populations, since logistic models assume
greater variability in the threshold dose distribution. These dose-response models usually
describe the prevalence of only one type of effect (e.g., mortality, reproductive effects). The
X--
models should then be limited to cases where only a few effects need be considered, since a
separate modeling effort is required for each toxic endpoint. Since chemical toxicity usually
involves several kinds of effects, multivariate dose-response models, called response surfaces
(NRC, 1988), are preferred. Unfortunately, the application of response surfaces to health risk
assessment is extremely difficult. Consequently, the use of dose-response surfaces should be
justified in each case.
Dose-Intensity Data. Models describing continuous measurements use the assumption that
the measured deviation from the normal value increases (or decreases) with increasing dose, e.g.,
a decrease in body weight with increasing dose (Crump, 1984; Dourson et ai., 1985). These
models are purely empirical, curve-fitting descriptions, and should therefore only be used within
or near the range of the experimental data. One difficulty in using such models is that the
measured effect must be interpreted in light of the variation in healthy or control populations; its
numerical value alone is not an absolute indicator of hazard. For example, if the forced
expiratory volume (FEV) decreases by 10% at the low dose, and by 20% at the high dose, the
20% decrease does not necessarily indicate twice the health hazard of the 10% decrease. The
application of such a model must compare the modeled value for the measured parameter with
values for healthy individuals. The parameter range for healthy individuals that is used to
evaluate the model results must be referenced. Models appropriate for dose-effect (i.e.,
continuous) data have been proposed (Crump, 1984). There has been, however, little practical
3-28
-------
application of these models. Consequently, any use of dose-effect models must include detailed
guidance on their interpretation.
Dose-Severity Data. A different approach to characterizing dose-effect data is to recast
the toxic effects in terms of risk. First, one characterizes the probability distribution of normal
measurements. Then, the probability of finding the specific changed- value in a normal animal
(or person) is used in a decision rule to assign the response either to the "adverse" or "normal"
groups. For example, one might determine that 95% of body weight gains with age are within
10% of the average weight gain. If this range is the decision rule for being normal, then any
deviation of more than 10% from the average weight gain is considered an adverse toxic
response. The data then become quanta! and are amenable to the dose-prevalence models in the
first category.
The models for toxicity data grouped into severity categories have been proposed as a
simple method for dealing with multiple response curves or multiple types of effects without
using higher dimensional response surfaces (Hertzberg and Miller, 1985; Hertzberg, 1988). The
toxicity data are first judged in terms of overall toxic severity for each dose. Linear regression
is then applied to the dose-severity data. Schemes have been proposed for assigning severity
categories to general toxicity (DeRosa et al., 1985), but no standard approach has yet been
adopted by the Agency. Since only the overall severity is being related to dose, this categorical
regression approach is best applied to cases where the regulatory goal is similarly concerned only
with the general severity of the toxic response, and not with specific organs or types of lesions.
This approach is consistent with the goals of the RfD for preventing any adverse effects
regardless of target organ and has been investigated to address short-term inhalation exposure
(Guth et al., 1991). The results of categorical regression models can be presented as probabilistic
risk of a specific seventy of effect at a given dose.
Model Parameters. The implementation of mathematical models to describe toxicity data
involves the estimation of model parameters and evaluation of the overall acceptability of the
model. When specific computer programs are dictated for the model application, the programs
must be carefully checked for accuracy and numerical stability.
Some programs and models require specific forms for the exposure values, for example,
that dose be represented as a daily intake rate. In such cases, assumptions may be required to
3-29
-------
convert the reported exposure units into the requisite units. The internal dose, as a steady-state
tissue concentration or blood level, should be estimated when possible using chemical-specific
dosimetry or using generic dosimetric scaling factors. In most such cases, no additional dose
adjustment will be used. If information suggests that other factors influence toxic sensitivity,
additional scaling or dose adjustment may be calculated, but must be justified in each case.
When chemical-specific information on dosimetry is not available, then standard
conversions should be used and referenced for estimating average daily intake for oral exposures
(in units of mg/kg/day) and average daily air concentration for inhalation exposures (in units of
mg/m3). Interspecies scaling of dose will involve a default 10-fold reduction, as is performed
by the species uncertainty factor in the Reference Concentration procedure. Other scaling
approaches, or more appropriate uncertainty factors, are encouraged but must be justified on a
x*
case-by-case basis.
Duration will be scaled according to fraction of lifespan for exposures of 90 days or
longer. Procedures for scaling duration with shorter-term data, such as using the actual time
exposed or scaling dose and duration by their product (Haber's principle), should be justified in
each case.
Quality of the Dose-Response Model. The evaluation of how well the model or models
describe the toxicity data should include measures of goodness-of-fit, estimates of variation (i.e.,
standard error) for each model parameter, and their respective levels of statistical significance.
When substantial extrapolation is required, the validity of biological assumptions in the model
and the impact of alternative assumptions should be discussed. When the use of alternative
assumptions can be quantified, the results of the dose-response modeling should be presented as
a range, in addition to the preferred values. The mathematical assumptions should also be
discussed, and checked whenever possible. For example, if a model requires the data to be
normally distributed, then the data should be -checked for goodness-of-fit by the normal
distribution function. When several dose-response models are used, perhaps for different toxic
endpoints, all should be presented. If the models are applied to animal data, then there snouid
also be a discussion of which endpoints are applicable to human toxic responses. Particular care
should be exercised when interpreting results of dose-response (prevalence) models. Unless
substantial additional information is incorporated (e.g., on mechanism of action, pharmacokmetics
3-30
-------
and species differences in tolerance distributions), response rates in animals, for general toxicity
or for specific organs, are unlikely to be reliable indicators of human response rates.
3.4.3.3 The Benchmark Dose. Several limitations in the use of the RfD/RfC have been
described. The use of the NOAEL focuses only on the dose that is the NOAEL and does not
incorporate information on the slope of the dose-response curve or the variability in the data.
Since data variability is not taken into account, the NOAEL from a small study will likely be
higher than the NOAEL from a similar but larger study in the same species, whereas the opposite
may be true when variability is considered. Additionally, the NOAEL must be one of the
experimental doses, and the number and spacing of doses in a study can influence the dose that
is chosen for the NOAEL. The NOAEL is defined as a dose that does not produce an observable
change in adverse responses from control levels and is dependent on the power of the study.
Theoretically, the risk associated with it may fall anywhere between zero and an incidence just
below that detectable from control levels (usually in the range of 7-10% for quantal data).
Because of the limitations associated with the use of the NOAEL, the Agency has decided
to begin using an additional approach to quantitative dose-response evaluation, known as the
benchmark dose (Crump, 1984) which is based on a model-derived estimate of a particular
incidence level, such as 1, 5 or 10% incidence. More specifically, the benchmark dose is the
lower confidence limit on the effective dose that produces a certain increase in incidence above
control levels. The benchmark dose is derived by modeling the data in the observed range,
calculating the upper confidence limit on the dose-response curve, and selecting the point on the
upper confidence curve corresponding to, for example, a 10% increase in incidence of an effect.
The dose corresponding to the model estimate for a 10% increase in incidence is the ED,0, wmle
the benchmark dose is the dose that corresponds to the upper confidence limit on the 10%
incidence, or the LED,0 Using the benchmark dose approach, an LED10 will be calculated for
each agent for which there is an adequate database. In some cases, the data may be adequate to
also estimate the ED05 or ED01 incidence levels which may be closer to a true no-effect dose.
A level between the ED01 and the ED10 is usually the lowest level of risk that can be estimated
adequately for binomial end points from standard developmental toxicity studies.
As indicated earlier, the advantages of the benchmark dose approach are that it takes into
account the slope of the dose-response curve, the size of the study groups, and the variability in
3-31
-------
the data. However, there is little practical experience with the application of this approach in
calculating the RfD/RfC.
As for the choice of the appropriate model to use in deriving a benchmark dose, various
mathematical approaches have been proposed for modeling developmental toxicity data (e.g.,
Crum, 1984; Kimmel and Gaylor, 1988; Rai and Van Ryzin, 1985; Faustman et al., 1989; Kodell
et al., in press). Such models may be used to calculate the benchmark dose, and choice of the
model may not be critical since estimation is within the observed dose range. Since the model
is only used to fit the observed data, the assumptions about the existence or nonexistence of a
threshold for a particular model are not pertinent. Thus, any model which fits the empirical data
well is likely to provide a reasonable estimate of the benchmark dose, although if there is some
biological reason to incorporate particular factors in the model (e.g., intralitter correlation for
developmental toxicity data), these should be included to account as much as possible for
variability in the data. The Agency is currently supporting studies to evaluate the application of
several models to data sets for calculating the benchmark dose.
In addition to identification of the NOAEL/LOAEL or benchmark dose, the dose-response
evaluation defines the range of doses that are effective in producing toxicity for a given agent,
the route of exposure, timing and duration of exposure, species specificity of effects, and any
pharmacokinetic or other considerations that might influence the comparison with human
exposure scenarios. This information should always accompany the characterization of the
adequacy of heaith-reiated data.
3.4.3.4 Decision Analytic Approach. The decision analytic approach to dose-response
assessment shares many of the features of the mathematical modeling of the dose-response data
approach described m the previous section. The major distinguishing characteristic of tne
decision analytic approach is the emphasis on explicitly characterizing and representing major
uncertainties using probability as the language to convey the degree of uncertainty. These
techniques have been applied to dose-response assessment for some noncancer health effects
associated with exposures to lead and ozone in the ambient air (Whitfieid and Wallsten, 1989,
Whitfield et al., 1993). The decision analytic methods have been used to support regulation of
criteria air pollutants. The decision analytic approach may be applicable to both exposure- and
3-32
-------
dose-response relationships. However, methods for dose-response relationships will be discussed
to introduce the appropriate methods.
Uncertainties about the shape of dose-response relationships exist, both because of such
things as measurement error, sample size, and sampling protocol in particular studies and because
of the paucity, incompleteness, and indirect nature of much of the available noncancer health
effects data. Where the data base is relatively strong and complete, it is possible to represent
uncertainty due to sample size considerations using Bayesian statistical techniques (e.g., Winkler,
1972). Generally, this is useful only where there is relatively good dose-response data on a given
health effect with doses in or near the range of regulatory interest.
It is possible to use the data directly to estimate probabilistic dose-response relationships
and to estimate the uncertainty due to sample size. Prerequisites for this approach include:
1. that the data be for humans (either a high quality epidemiologic or a controlled
human exposure study) and
2. that the population from which the sample was drawn be sufficiently close to the
population of interest.
The Bayesian approach to representing uncertainty due to sample size is to begin with a
prior distribution to represent the uncertainty. The prior distribution is then updated using
experimental data and standard Bayesian statistics to obtain a posterior distribution, which is
taken to be the desired final form of the distribution.
For many noncancer health effects the available data are too indirect, conflicting, or
incomplete to draw inferences directly about dose-response relationships of concern. Whiie
classical statistical techniques ore not available to quantify these types of uncertainties.
approaches have been developed in the field of decision analysis for eliciting judgmental
probabilities from scientific experts.
The use of the decision analytic approach to dose-response assessment is appropriate when
the objective is to obtain a distribution of risk estimates for some defined health endpomt(s)
associated with given levels and conditions of human exposure. Like mathematical modeling,
this approach allows risk estimates to be derived for different exposures or doses, as well as for
different durations of exposure, including acute, subchronic, and chronic.
3-33
-------
Decision analytic approaches are more data- and resource-intensive than the RfD/RfC
approach described earlier. In addition to the constraints described in the section on
mathematical modeling, the decision analytic approach to dose-response assessment requires
expertise in eliciting judgments from experts and the availability and cooperation of relevant
scientific experts for successful implementation.
An important issue in implementing a decision analytic assessment is the selection of
health experts to provide probabilistic dose-response relationships. The Agency develops explicit
selection criteria on a case-by-case basis. ^Experts are chosen that are recognized, competent
scientists who have done research in the area of interest and have published in the peer-reviewed
literature. In addition, the group of experts selected should represent the range of credible and
respected scientific viewpoints encompassing the range of potentially different interpretations of
the scientific data base. In many cases, the set of well qualified experts may be small enough
to include most of the relevant experts, assuming that all of the experts are willing to participate.
If the set of potential experts is large, or if resource limitations preclude including all experts,
selection of experts should be done with great care and follow predetermined criteria developed
for the specific assessment.
Probabilistic judgments from experts concerning dose-response relationships are obtained
during an interview session, which involves one or two analysts and the expert. Decision
analysts refer to the elicitation process as "probability encoding." Probability encoding typically
includes five phases, starting with motivating, in which the purpose of the assessment is
established and possible motivational biases are explored. In the second phase, structuring, the
analyst(s) clearly defines the unknown quantity or relationship for which judgments will be
elicited, making explicit any assumptions that exist. In the third phase, conditioning, the
analyst(s) and expert discuss both the scientific literature relevant to relationships of interest and
various factors that behavioral psychologists have found to affect the way people form and
express probabilistic judgments. In the fourth phase, encoding, the analyst elicits probability
judgments using one or more techniques that have been developed by decision analysts. These
techniques include: "fixed-probability methods" in which the expert is asked for values of the
quantity that bound specified probability intervals and "fixed-value methods" in which the expert
is asked for the probability that a given quantity lies within a specified range of values. In the
3-34
-------
final phase, verification, the analyst(s) checks stability and coherence of the probability
distributions obtained using alternative elicitation methods, to see if they satisfy the laws of
probability.
3.4.3.5 Occupational Exposure Limits. Occupational exposure limits (OELs) are
determined, based on the available epidemiologic and toxicologic data, in order to minimize
adverse health effects in workers occupationally exposed to chemical agents. OELs may be
established by one of several occupational regulatory or advisory agencies. The most commonly
used OEL is the Threshold Limit Value (TLV) established by the American Conference of
Governmental Industrial Hygienists (ACGIH). The TLVs are assumed to be protective of
otherwise health workers exposed 8 hours a day, 40 hours per week for their entire working life.
Recently, state agencies have promulgated air toxic regulations that are modifications of existing
X-
TLVs. Typically the TLV is divided by some combination of safety or scaling factors (ranging
from 10 to 420) to extrapolate to continuous exposures. These scaling factors may include
factors intended to protect for sensitive subpopulations and to adjust for exposure duration and
severity of effect. Table 3-3 summarizes the approach used by many states. This application of
the TLVs is controversial because TLVs are intended for applications to occupational settings
only. Several major issues may preclude the use of TLVs. TLVs have been established to
protect health workers and do not address sensitive subpopulations (e.g., children, elderly, those
with pre-existing illnesses or conditions). TLVs are also based on workweek exposures which
allows for 16 hours per day and 2 days per week of recuperation. There is typically little data
on the effects of continuous exposure related to TLVs since the recuperative process would be
eliminated. Furthermore, TLVs are not necessarily established in relation to some effect
threshold. Some TLVs have been established above effect thresholds if that effect (e.g.,
irritation) is believed to be transient (exposure-related) and to not significantly affect worker
performance or overall health.
The EPA Risk Assessment Forum has evaluated the use of OELs m generating mnaiation
RfCs (EPA, 1988d). After considering ail the issues described above, the EPA technical panel
recommended that OELs should not be used in developing inhalation RfCi, even with
mathematical manipulation. However, the scientific data base supporting an OEL can provide
useful information for both the hazard identification and dose-response assessment.
3-35
-------
Table 3-3
TLV Adjustment Factors Used by States Air Toxics Programs
State
Method/Factors
Alabama
Arkansas
Connecticut
Illinois
Minnesota
Mississippi
Montana
Nevada
New Hampshire
South Carolina
Vermont
Virginia
Washington^
Wisconsin
Wyoming
TLV/40, 1-hour avg
New sources
TLV/100, 24-hour avg
TLV/50, 8-hour avg
TLV/300, 24-h, for noncarcinogens
TLV/100,. 24-hour
TLV/30, 24-h
TLV/42 for annual average
TLV/10, 8-h
TLV/100
TLV/420
TLV/420, 24-hour and annual
TLV/60
TLV/300
TLV/42, annual
TLV/42, annual
TLV/50, 24-h
TLV/300, 1-h
a Compiled by Radian Corporation for North Carolina, Department of Environmental
Management, 1985
b Personal communication with 0. Oevoli, EPA Region X, 1990
3-36
-------
3.4.4 Methods and Data Sources for Dose-Response Assessment.
The EPA has established an interagency RfD/RfC Work Group to determine inhalation
RfCs and oral RfDs. This work group has developed a systematic approach to summarizing its
evaluations, conclusions, and reservations regarding RfCs and RfDs. As part of this effort, the
work group generates a statement of confidence (high, medium, or low) in the stability of the
RfC or RfD. An RfC or RfD with a high confidence rating is unlikely to change because it is
based on multiple, high quality epidemiologic or animal studies. A low confidence rating
indicates that subsequent information could lead to a change in the current RfC or RfD value.
The IRIS data base is the most desirable source of RfCs and RfDs and should be consulted first
when performing a dose-response assessment for noncarcinogens (see Table 3-4). Information in
the IRIS data base is EPA reviewed and supplies RfCs and/or RfDs for many noncarcinogenic
compounds.
If the compound of interest is not reviewed in IRIS, state or other agencies can be
contacted to see if an RfC or RfD has been developed. RfCs or RfDs developed by agencies or
groups other than EPA should be reviewed for proper selection of the critical effect and adequacy
of data.
After an RfC or RfD has been identified for a compound, a more quantitative method
may also be desired. In that case, dose-response modeling can be conducted if adequate data are
available.
After all sources of existing RfCs and/or RfDs are exhausted, the other approaches for
conducting a dose-response assessment (i.e., dose-response modeling, SARs. etc.) should be
explored. The selection of the appropriate methodology is dependent upon the data available.
Table 3-5 also shows the levels of study, expertise, resources, and relative time required
for the various methods. The relative time and level of expertise required vanes from low for
the IRIS data base search to high for developing an RfC or RfD from the existing primary
literature. Resources required also increase with the increasing complexity of the methodology.
••> -»--7
J-J /
-------
Dther Characteristics of Methods
f5
^
T3
O
~ZH
2
i
o.
fa>
^5
'
i i
\^- _ _
§ |
led by an intra-agenc
equired
jmber of chemicals, i
osure concentrations
1 RIDs have been generated by EPA and verif
1 Subjective evaluation of the literature is not r
> Inhalation RICs available only for a limited ni
1 Present point estimate for comparison to exp
zz~* -• _
~3> TD
2 E
S a
Ji ^
_2 "s
S 0
o
ac
S
CO
cc
f
-g
D
C/3
1
1
s J
fs j "s
C 3
^ « «
5. o fe
|1 2
cx ^3
I"!™
3wed according to EF
cal effect and critical
ty factors and modily
1 May not have been developed and peer-revii
1 Should review to determine if the proper critii
1 Should review whether appropriate uncertain
o
cc
^o
0
a:
a>
3
o
a>
^
3
8
£
CD
•-a
o
a.
03
UJ
eo_
"to
a
"I
1
~a>
5~~
M
®
leralure and deterrnin
1 factors to use
1 Have to subjectively evaluate the available hi
*JOAEL and/or LOAEL for the compound
1 Need to determine uncertainty and modifying
1 Peer-review process is recommended
UJ
o S
__J -™
o -~
uj 2
< —
0 =
Z. 0
1 "S
•=j '=
13 , ~
!D O
~2 S
11
UJ O
o £
a ™
<0 o"
5 "o3
"^ •*
<3 ^o
|.l
s 13
3. 'c
5CU
u =
5 °
CC -o
CO
•&
1
CO
S5
09
O
S
s
m
CO
RfC-type value
ave been developed
' Results in probability of effect rather than an
1 Dose-response factors for some chamicals;h
nodels used
' Peer-review process is recommended
OJ
to
—
2
qj'
CJ
03
1
O
-Q
I i
0. en
o>
S
^
CD
O
y5
2
y5
O
a
3-38
-------
CO
Q
O
UJ
2
u_
O
O
LU
LU
LU
co
O
Q_
CO
LU
CO
o
CC
o
UJ
ca
03
.i -o
"- 2
S ^
1 =
Q>
cr
-a
03
'5
oT
CC
CO
1
CC
-E3
03
1
cr
CD
CO
Q.
X
UJ
>,
^
CO
"o
"s
03
1
r-
03
S
o
o!
8
co .
t—
J§ 2
7S ^**
03 UJ
^•s
c 3
O o
-a
If
•5 .*
o
O3 ~O
C J=
£ CC
co1
o
ac
5
CO
ac
"o
CO
1
T3
CD
CO
CD
S
§,
ca
CO
o
.M
§
S
§
o
o
03
1
CL
O
£•
CO
.-9
.'OT
"o
1
5
^5
1
co
0
O) ~O
1 "3
o £E
3
CO
O
cc
"o
CO
O3
^
O
(O
CD
S
03
b
o
_j
^
.2^
X
CO
1
8-
o
ca
CD
CO
1
2
Q)
•^
CD
a.
o
2?
2
Zj
o
'w
o1"™
o §
E "35
03 «
'H^
^^ ^
V5 "T^
"05 15
O fU
li
t— 0
-^
O3
CC
c/3
-H
"3
0
2
o
1
CO
O
CC
"to
o
Q.
"x
UJ
j.
.5^
X
S
1
-s
s
03
1
•^
CD
I
8
Zj
o
Is
g5"™
o g
s «
03 CO
•a ^g
'5. g
TX •£
'en ca
0 03
O •*—
X ^
o £
1— O
-3
03
C
"35
CC
LU 0
0 £
"H S
.3 "®
2 ?
03 -a
= _o
g .2
•— "oD
5- "c
ll
II
„
ir
o
S
~*
O3
§ a
°_ s
.I--8
:f! 2_
fl
11
CJ U
ra 2,
CD
"co
O cu
p •*•
"x _«
o £
V- 0
T3
03
C
CC
s
CD
-3
O
03
O
Q.
CO
03
03
CO |
Q
o>
1
o
ca
CO
o
o>
O3
03
^:
o
,0
O)
E
03
CO
§
g-
03
03
CD
C
O
-
CO
CO
CO
03
O3
CO
O
03
_CO
co
J
"o
03
'-5
03
O
3-39
-------
3.5 REFERENCES
Brown, Charles C. High Dose to Low Dose Extrapolation in Animals. Presented at the
American Chemical Society Symposium on Assessing Health Risks from Chemicals. Kansas
City, Kansas. September 1982.
National Research Council, Safe Drinking Water Committee. Drinking Water and Health.
Volume 6. National Academy Press. Washington, D.C. 1986.
Office of Science and Technology Policy (OSTP), Executive Office of the President. Chemical
Carcinogens; A Review of the Science and Its Associated Principles. Washington, D.C. 50 FR
10372. March 14, 1985.
U.S. Environmental Protection Agency, Air Risk Information Support Center. Glossary of Terms
Related to Health, Exposure, and Risk Assessment. Research Triangle Park, North Carolina.
March 1989a.
x»
U.S. Environmental Protection Agency. Guidelines for Carcinogen Risk Assessment. Carcinogen
Assessment Group, Office of Health and Environmental Assessment. Washington, D.C. 51 FR
33992. September 24, 1986.
U. S. Environmental Protection Agency. Guidelines for the Health Assessment of Suspect
Developmental Toxicants. Office of Health and Environmental Assessment. 51 FR 34028.
1986a.
U.S. Environmental Protection Agency. Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office. Technical Support Document on Risk
Assessment of Chemical Mixtures. November, 1988.
U. S. Environmental Protection Agency. Proposed Guidelines Assessing Female Reproductive
Risk; Notice. 53 FR 24834. June 30, 1988a.
U. S. Environmental Protection Agency. Proposed Guidelines Assessing Male Reproductive Risk
and Request for Comments. 53 FR 24850. June 30, 1988b.
U.S. Environmental Protection Agency. Occupational Exposure Limit Data in Relation to
Inhalation Reference Doses. Office of Health and Environmental Assessment, Cincinnati, Ohio.
1988d.
U.S. Environmental Protection Agency. Risk Assessment Guidance for Superfund - Human
Health Evaluation Manual, Part A. EPA/540/1-89/002. Office of Solid Waste and Emergency
Response. Washington, DC. July 1989a.
3-40
-------
U. S. Environmental Protection Agency. EPA 600/8-90/066A. Interim Methods for Development
of Inhalation Reference Concentrations. Office of Health and Environmental Assessment. August
1990.
U.S. Environmental Protection Agency. Guidelines for Developmental Toxicity Risk Assessment;
Office of Health and Environmental Assessment, 54 FR 6398-63826, 1991.
U.S. Environmental Protection Agency. General Quantitative Risk Assessment Guidelines for
Noncancer Health Effects. External Review Draft, ECAO-CIN-538. Research and Development.
February 199 la.
Weinstein, I.B. The Relevance of Tumor Promotion and Multistage Carcinogenesis to Risk
Assessment. (In) Banbury Report 19: Risk Quantitation and Regulatory Policy. D.G. Hoel,
R.A. Merrill, and F.P Perera, eds. Cold Spring Harbor Laboratory. Cold Spring Harbor,
New York. 1985.
3-41
-------
-------
4.0 EXPOSURE ASSESSMENT
4.1 INTRODUCTION
Guidelines for Estimating Exposures (51 FR 34042-34054) was one of the original EPA
risk assessment guidelines issued in 1986. These guidelines discussed the general contents of
an exposure assessment, data availability and uncertainty analysis, uncertainty evaluation, and a
clarification of terminology. When the 1986 guidelines were issued, the Science Advisory Board
recommended that the EPA develop supplementary guidelines for conducting exposure studies.
Supplementary guidance was developed by an Agency work group composed of scientists from
throughout the Agency. A draft was developed and peer reviewed by environmental, industry,
academia, and other governmental agencies. In 1988, this draft was then proposed for public
comment as the Guidelines for Exposure-Related Measurements (53 FR 48830-48853).
Comments received during the public comment period were in favor of combining the proposed
guidelines with the 1986 guidelines. As a result, in 1992 the guidelines have been reformatted,
revised, and reissued as final Guidelines for Exposure Assessment (51 FR 22888-22938) which
became effective May 29, 1992. These new guidelines supersede and replace both the original
1986 guidelines and the proposed 1988 guidelines.
The new 1992 guidelines establish a broad framework for exposure assessments by
describing the general concepts of exposure assessment including definitions and associated units.
They also provide guidance on the planning and conducting an exposure assessment, presenting
the results of the exposure assessment, and characterizing uncertainty. Although most of the
guidelines focus on human exposures, much also pertains to assessing wildlife exposures as well.
These guidelines are organized in a similar manner to the 1986 guidelines but with significantly
more detail on general concepts, planning, data development, calculating exposure, uncertainty
evaluation, and presentation of results. These guidelines can be referred to for specific details.
In addition to these guidelines, recent EPA policy regarding risk characterization also has
implications for exposure assessments. The Deputy Administrator of the EPA introduced new
guidance on risk characterization in March 1992 (see Section 5). This new guidance emphasizes
a change in the descriptors of risk, including individual risk, to include trie central tendency and
high end portions of the risk distribution, important subgroups of the population, and population
nsk. Exposure assessments must be taking these risk descriptors into account.
4-1
-------
Exposure assessment has four major components: emissions characterization,
environmental fate and transport, characterization of the study population, and exposure
calculation. This section discusses each of these components in terms of models and
methodologies applicable to exposure resulting from atmospheric pollutant emissions. Although
toxic pollutants may be emitted into both indoor and outdoor air from a wide variety of sources
and activities, this report focuses on emissions from stationary outdoor pollutant sources.
In the emissions characterization component of exposure assessment, the, emission rate
of the pollutant and the parameters of the source are defined. The emission rate specifies the
mass of pollutant released to the atmosphere over time. Source parameters, notably the flow rate
of the stack gas volume, the stack gas exit temperature, and the stack height, define how the
pollutant is released to the environment. These factors as well as determining the magnitude of
•\-
potential exposure significantly affect the initial dispersion of the pollutant in the atmosphere.
The fate and transport, with the emission rate, components describe how the pollutant is
° ultimately transformed and dispersed over the region of interest. Transport and possible
transformation of an airborne pollutant are influenced by the pollutant's physical and chemical
properties and by meteorological and environmental conditions.
The population characterization component defines the study population in terms of
geographic distribution and other characteristics of interest. Factors such as age, sex,and activity
level affect the amount of pollutant actually inhaled by individuals, while mobility affects the
concentration levels to which an individual is exposed over time.
In the fourth component, exposure calculation, the pollutant concentration and study
population are spatially integrated to estimate maximum and cumulative exposure.
The EPA Office of Air Quality Planning and Standards (OAQPS), has set up a technology
transfer network (TTN), which provides information on a variety of topics related to exposure
assessment. Information necessary to access the TTN and topics available are outlined in Table
4-1 (U.S. EPA, 1991).
4-2
-------
a
2!
UJ
03
>>
o «
•a 8
< |
logy,
ech
a
Reasonably A
1*
tl
I 8,
u-i
3 =
*-a
.2 a
SS •§
'§ I
u E
u <2
3 .£
2 «
o jj
31
S -3
so -S a
III
CJ r- .3
3
o
I
G
V5 ^J
J <
aa a
HJ
I d
.i 'g
00 c
.s §
ll
•i-a
g.5
o a
•35 .2
"3 ">
§'§
_ U
11
w 2
S 3
I3
ll
go
2 1
2 "=»
31
c/) 'J
ll
l§
i
-J
<« i
s
I
M
•
- 3
u C
o
3
O CJ
•3
1
ts.
in
u
i'3
Ii
12
"> 2
M
a o
13
. a
» ^
g'S
•s 8
a
s ta
E s e
"
-LJ '_U _=
,_2
"5
. U
ll
S
I
S&
M
5 u
8 -o
•3§
§1
2 3
.2 O
•
a. 2 =
cu a m
2 2 U
3
00
C
(3 •§
Is
00
QQ
&
it
u
Ig
_• >o
S j,
« TT
8-
u
X
-55
« CS OO
>* ^
jj X m
3 -3 —
o
00 U
i
< o 3
Q S >
4-3
-------
4.2 EMISSIONS CHARACTERIZATION
4.2.1 Overview
Characterization of pollutant emissions is accomplished in the early stages of a risk
assessment to identify and quantify the amount of each specific chemical released to the
environment. Once the quantity of emissions has been estimated, potential exposure of the study
population can be assessed.
Pollutants may be released in indoor and outdoor environments from a wide variety of
sources and activities. This document focuses on stationary sources of emissions released into
an outdoor atmosphere. For use in human exposure models, chemical emissions are typically
defined in terms of the mass released to the atmosphere over time. Emission rates may be
expressed on an annual basis to assess chronic exposure, or on a short-term basis, to estimate
v»
more acute exposures.
Section 4.2.2 describes general factors to consider in emissions characterization and
summary of various methods and techniques for estimating pollutant emission rates. Section
4.2.3 describes the methods presented in terms of selection criteria for actual application.
4.2.2 Technical Background and Methods
4.2.2.1 Estimating and Measuring Toxic Emissions. The focus and level of detail
involved in characterizing emissions depends on the scope and depth of the overall risk
assessment. For example, the focus may be geographical, geared towards pollutants released
from a variety of sources within a defined area; it may be industrial, based on characterizing
emissions from a particular type of industry, or may be oriented towards a specific chemical.
Once the sources or chemicals of interest are identified, preliminary estimates of the chemical
emission rates can be made and used to screen for potential risk. Detailed emission estimates,
based on more precise information, may then be used in refined risk assessments.
The "Toxic Air Pollutant/Source Crosswalk - A Screening Tool for Locating Possible
Sources Emitting Toxic Air Pollutants, Second Edition" is a useful resource for identifying che
types of toxic compounds that may potentially be emitted from a source category (EPA, 1989a).
Tables within the crosswalk are cross-referenced by pollutant. Standard Industrial Classification
(SIC) codes and Source Classification Codes (SCC). Data from the SARA Title HI (TRIS) and
NATICH data bases, further described in Section 4.2.2.4, could also be used to .identify cnemicals
that may be emitted from particular source categories.
4-4
-------
Source Sampling. Emissions estimates can be obtained by directly measuring the source
of interest. Although direct measurement is likely to provide the most accurate data for a
specific emission source, data collection via sampling may be impractical because of technical,
resource, and/or time constraints. If emissions testing is conducted, it should be done under
representative conditions and should be well documented. The EPA has established various
reference or state-of-the-art methods for direct measurement of air toxic emissions. Methods
exist for determining total emission rates for groups of similar compounds, as well as for
identifying compound-specific emission rates. The EPA has developed compound-specific
methods to determine emissions for five major classes of air toxics:
• volatile organic compounds (VOCs),
• semi-volatile organic compounds,
• toxic metals,
• aldehydes/ketones, and
• acid gases and halogens.
Emission Factors. Emission factors indicate the quantity of a pollutant typically released
to the atmosphere with a particular source operation. They are usually representative of an
industry or emission source type as a whole. Emission factors are a commonly used alternate
method for defining emissions when specific emissions measurements are not feasible or
available. Emission factors are usually expressed by units of mass of pollutant emitted per unit
of mass, volume, heat input, distance, or duration of a process that emits the pollutant. For
screening purposes, emission factors may not be appropriate, unless the entire source category
is being modeled. Emission factors are not conservative and should only be used with caution.
Actual emissions from a source may be higher or lower than the emission factors indicate
because of site-specific process design, control equipment, operation and maintenance practices,
or other factors. Before using an emission factor, available documentation on how the emission
factor was derived should be studied to determine whether it is appropriate for the source under
study. Specific data sources containing emission factors are described in Section 4.2.2.4.
Chemical speciation factors are used to estimate ratio (or percentage) of specific chemicals
that make up the total VOC or particuiate matter (PM) emissions. Speciation factors are
available for some types of sources and pollutants. As described in Section 4.2.2.4, if total VOC
or PM emissions have been measured or estimated, speciation factors may be used to estimate
emissions of individual chemical species. This method introduces uncertainty, since species
4-5
-------
composition may vary from source to source. The user should review information on the
speciation factor derivation and applicability to determine whether this method is appropriate for
the source in question.
Other Methods. Methods other than source testing or emission factors could be used to
estimate emissions. For example, if there is no published emission factor for a given source type,
but reliable emissions test data are available from a similar source operation, this information
might be used to derive an emission factor. Similarities or differences in the processes,
capacities, and pollution control systems would need to be considered. If chemical use but not
emissions are known, a chemical mass-balance approach could be used to estimate emissions.
The mass-balance approach requires that all pathways of toxic materials entering and leaving a
facility be identified. As with any method, the basis for assumptions and calculations should be
clearly documented. Another method, the TSCREEN computer model, was developed for the
Technical Evaluation Section of the OAQPS specifically for determining the types of toxics
released from Superfund sites.
Source Release Parameters. In addition to estimating the quantity of emissions, release
characteristics of the source must be defined. Knowledge of the emission rate and release
characteristics enables the pollutant fate and transport to be estimated, as discussed in Section
4.3. Modeling of emissions released from a stack requires knowledge of the stack height, inner
stack diameter, gas exit velocity or flow rate, and gas exit temperature. For facility area sources
(e.g., storage pile fugitives or emissions from ponds), the dimensions of the area source should
be identified. While point source emission rates are expressed in terms of mass per unit :ime,
area source emission rates are more typically modeled in terms of mass per unit time per unit
area. Another important consideration in specifying the source emission rates is whether the rates
should reflect short-term or annual operating conditions. This will depend on whether the focus
of the assessment is on acute or chrome exposure.
4.2.2.2 Complex Chemical Mixtures. Emissions and subsequent exposures are commoniy
calculated separately for each toxic chemical in a mixture. However, if there are health effects
data for a mixture as a whole (e.g., coke oven emissions), then the emissions of the entire
mixture could be estimated and used in a risk assessment. In this case, the emission rate of the
mixture (e.g., the rate of total VOCs emitted), can be used to model the chemical mixture.
Mixtures can also be modeled from the emission rates of each of the individual compounds. The
4-6
-------
toxicity of the mixture is then estimated by comparing it to a similar mixture or evaluating each
compound individually, evaluating for interactions, and adding their effects contributed by each
of its components.
For estimating risks associated with exposure to complex chemical mixtures of chlorinated
dibenzo-p-dioxins and dibenzofurans, an interim approach using toxicity equivalence factors
(TEFs) has been established (Bellin and Barnes, 1986). This approach is based on estimates of
the concentrations of congeners present in the mixture. Using available lexicological data and
reasoning on the basis of structure-activity relations, the significance of exposure to each of the
components is estimated and expressed as an equivalent amount of a reference compound. Risk
assessment using this approach requires emissions estimates for the specific congeners involved.
4.2.2.3 Chemical Reactions and Removal Mechanisms. Because changes may occur in
the pollutants after they are emitted, the chemical properties of the pollutants should be assessed
for potential transformations that might significantly affect exposure. In the case of aldehydes,
for example, the combined effects of direct emission into the atmosphere (from manufacturing,
power production, etc.) and of generation and removal within the atmosphere can result in highly
variable ambient concentrations (NRC, 1981). Aldehydes are generated in the atmosphere
through interaction with various reactive chemical species. Formaldehyde vapors may be released
from a variety of non-targeted sources, such as building materials and insulation and chemically
treated cloth. Aldehydes are removed through photodecomposition and reaction with the
hydroxyl free radial present in a sun-irradiated atmosphere (NRC, 1981). Consideration of the
potential for chemical interaction helps to determine to what degree the released pollutant may
be depicted. It also raises the question of whether compounds more toxic than their precursors
may be created.
Pollutant emissions are subject to various removal processes, particularly dry deposition
and scavenging by rain and clouds. Dry deposition involves the transport of pollutants to the
earth's surface, followed by the physical and chemical interactions between the surface and the
pollutant. To simplify the computations involved, air quality models express the rate of dry
deposition in terms of a deposition velocity, which is defined as the ratio of the ground-ward flux
of the species to the concentration of the species at some reference height.
Wet deposition (precipitation scavenging) is a function of the intensity and size of the
raindrops and the solubility and reactivity of the species. Chemicals with high water soiubiiity,
4-7
-------
such as formaldehyde, will tend to transfer readily into rain and surface waters. Fog and cloud
droplets can absorb gases, capture particles, and enhance chemical reactions. Precipitation
scavenging is not included in most urban-scale models since it is more significant on the regional
scale. Fog chemistry can be important on the urban scale; however, few attempts have been
made to model the relationship between pollutant emissions and fog chemical dynamics in an
urban area (Russell, 1988).
4.2.2.4 Summary of Methods to Characterize Emissions. The Clearinghouse for
Inventories and Emissions Factors (CHIEF), available to the OAQPS TTN (see Table 4-1), is
EPA's central clearinghouse for the latest information pertaining to inventories and emissions
factors. CHIEF provides access to the tools necessary to estimate emissions and perform air
emission inventories (U.S. EPA, 1991).
X-
Applicable methods applicable for sampling and analyzing the 191 compounds and classes
of compounds specified in the 1990 proposed Clean Air Act Amendments are described in the
EPA document entitled "Measurement Protocol for Air Toxics" (Radian, 1990). This protocol
presents a series of air toxic screening methods, emphasizing those that apply to a large number
of compounds from the list of 191. Listed chemicals for which no known sampling and analysis
methods exist are identified. Criteria for the selecting methods appropriate for specific
applications are presented, as well as an estimate of the cost of each method based upon a sample
survey of the sampling/analytical services marketplace.
Information about emission test procedures may also be obtained through the EPA
Emission Measurement Technical Information Center (EMTTC) in Research Triangle Park. North
Carolina. EMTIC's purpose is to promote consistent and accurate test method applications in
national, state, and local emission prevention and control programs. The Technical Support
Division, Emission Measurement Branch of the EPA OAQPS manages an EMTIC electronic
Bulletin Board System fBBS). BBS provides technical guidance on stationary source emission
testing issues for people doing emissions testing in support of emissions standards, emissions
factors, and state implementation plans. Access to the EMTIC BBS is through the OAQPS TTN
(see Table 4-1) (U.S. EPA. 1991).
Emissions Test Reports. The most accurate information on pollutant emissions generally
comes from the results of specific emissions tests. These tests are typically performed co
demonstrate compliance with regulatory standards or permits, to document the performance of
4-8
-------
an air pollution control device, or to provide desired technical data. Reports summarizing the
results of a compliance test are generally available to the public; reports of tests conducted for
other purposes are generally located on the property of the facility.
The amount of information available on the tested emissions source depends on the
purpose of the test. At a minimum, some idea of the capacity at which the source was operating
during the test will be given (e.g., heat input rate, production rate). Also, the exhaust gas
temperatures and flow rates and stack diameter will generally be provided, but the height of the
stack may not be reported- Emissions are expressed in terms of concentrations (the time period
of which depends on the sampling method) and mass emission rates.
The National Air Toxics Information Clearinghouse (NATICH) provides selected source
testing data for unusual test cases or cases of particular interest to other agencies. For each
source, test information is provided on: the agency, source testing contact, test ID number,
facility category, location of measurement, SIC code, test data, sampling technique and analytical
method used, Source Classification Code (SCC) if available, pollutant names. CAS numbers, and
emission rates.
NATICH also provides information on acceptable ambient concentration levels, air toxics
emission research activities, ambient monitoring activities, permitting and source testing activities,
and emission inventory and risk assessment activities (EPA, 1990a-d). NATICH is managed
through the Pollutant Assessment Branch of OAQPS. The NATICH data base system is
maintained on the EPA National Computer Center (NCC) IBM mainframe. State and local
agencies can arrange access to NATICH information through their EPA Regional Office. The
public may access the information the National Technical Information Service. The OAQPS
TTN will soon offer a NATICH bulletin board to improve the exchange of information among
federal, state, and local agencies concerned with the control of toxic air pollutants (U.S. EPA,
1991).
Emission Factor Data Sources. Emission factors are published in the EPA document
"Toxic Air Pollutant Emission Factors for Selected Air Toxic Compounds and Sources - Second
Edition' and can also be found in the computerized versions of this document known as "The
Crosswalk Air Toxic Emissions Factor (XATEF) Data Base Management System" (EPA,
1990e,f). This report identifies emission factors by pollutant name, CAS number, process
description, SIC codes, emission source descriptions, and SCC. Also included are brief
4-9
-------
descriptions of factor derivations, notes on any control measures associated with the emission
factor, and references. Emission factors are sorted by pollutant and source type. Over 5,000
factors covering more than 300 pollutants are presented in this report.
The EPA has developed a document series addressing air toxic emissions from specific
.sources (e.g., EPA, 1989d-g). Reports in this series identify source categories for which
emissions of the subject toxic air pollutant have been characterized. Each report characterizes
all toxic releases from a specific source category. These reports include: general process
descriptions of the emitting processes, identification of applicable control equipment and pollutant
emission points, controlled and uncontrolled emission factors; physical and chemical property
data for the subject toxic, and general descriptions of the emitting processes. Documents have
been prepared for acrylonitrile, carbontetrachloride, chloroform, ethylene dichloride,
>»
formaldehyde, nickel, manganese, chromium, phosgene, epichlorohydrine, polychlorinated
biphenyls, polycyclic organic matter, benzene, organic liquid storage tanks, coal and oil
combustion sources, municipal waste combustors, perchloroethylene, trichloroethylene, 1,3-
butadiene, styrene, and sewage sludge incinerators.
The Aerometric Information Retrieval System (AIRS) Facility Subsystem (AFS) contains
emission factors for the six criteria pollutants: PM10, sulfur oxides, nitrogen oxides, volatile
organic compounds, carbon monoxide, and lead (U.S. EPA, 1990g). It contains a consolidation,
by SCC, of all currently available emission factors and provides newly developed SCO; and
emission factors. Most of these emission factors are taken from the "Compilation of Air
Pollutant Emission Factors, Fourm Edition, AP-42." They are intended for use as default values
in the absence of better estimates of emissions. In certain cases, the factors may be derived from
information not yet incorporated into AP-42, or may be based simply on the similarity of one
process to another with known emissions information. A personal computer (PC) diskette version
of this information is also available from the National Air Data Branch of OAQPS. An AIRS
bulletin board is available on the OAQPS TTN for information exchange among state and local
agencies that use AIRS documents and information (U.S. EPA, 1991).
Emission factors and species profiles for many consumer/commercial products are
presented in the "Compilation and Speciation of National Emission Factors for
Consumer/Commercial Solvent Use" (U.S. EPA, 1989b). This publication focuses on area
4-10
-------
sources and presents emissions factors as national averages with geographical allocation for
identifying regional variations-
Methods for estimating fugitive emissions for chemical process units have been
documented in the EPA "Protocols for Generating Unit-Specific Emission Estimates for
Equipment Leaks VOC and VHAP" and in the closely related Chemical Manufacturers
Association's "Guidance for Estimating Fugitive Emissions" (U.S. EPA, 1988; CMA, 1987).
These documents discuss the estimation of fugitive emissions from components such as values,
pumps, connectors, and open-ended lines; and promote implementation of leak detection and
repair programs. Five methods are presented in order of increasing complexity, cost, and
accuracy:
>•.
• Synthetic Organic Chemical Manufacturing Industry (SOCMI) average emission
factors;
• Leak/No Leak emission factors;
• SOCMI stratified emission factors;
EPA correlation equations; and
• Unit-specific correlation equations.
With the exception of the SOCMI emission factors, all of the other fugitive emission estimation
methods require that a screening survey be made of the components in the process area. This
screening involves relatively quick measurements, using a portable detector, to indicate which
sources are emitting and roughly in what magnitude.
Procedures for estimating air toxics emissions from area sources are presented in a draft
document available through the Pollutant Characterization Section of EPA. The document
includes emissions factors for the most prevalent urban ares source categories, such as mobile
sources, home heating, and gasoline service stations.
National emission estimates and factors for mobile sources are presented in rhe EPA
document "Air Toxics Emissions from Motor Vehicles" (U.S. EPA, 1986). Emission factors for
13 air toxics are included in the report, which is periodically updated.
Speciation Factors. Chemical speciation factors are used in conjunction with emission
factors or total emissions estimates to provide emission rates of individual chemical species
within the effluent of a given source type. The "Air Emissions Species Manual - Volume I,
Volatile Organic Compound Species Profiles and Volume II, Paniculate Matter fPM) Species
Profiles - Second Edition" contain chemical speciation profiles for a variety of VOC and PM
4-11
-------
-f
i
emission sources (U.S. EPA, 1989c). Speciation data are organized by source category and by
SCC. Information in this two-volume set is also available in a PC-based computerized version
known as the "Speciation Data System" (U.S. EPA, 1990h).
To determine emissions for an individual chemical from a particular source type, it is
necessary to identify the source category corresponding to the source type of interest. Then the
appropriate species profile can be identified from the "Air Emissions Species Manual." To
produce estimates of the individual chemical emission rates, obtain the weight percentages of the
chemicals of interest from the profile and then multiply them by the emission factor, or total
VOC or PM emissions, for a particular source category. Emission factors can be obtained from
the National Emission Data System (NEDS) or, for particulate matter, from the "Criteria Pollutant
Emission Factors for the 1985 NAPAP Emissions Inventory" (U.S. EPA, 1987a). Using the
reported distributions associated with the weight percentages from the Emission Air Species
Manual, a range of the chemical emissions can be determined.
SARA 313 Emissions Data. Section 313 of the Emergency Planning and Community
Right-To-Know Act is also known as Title HI of the Superfund Amendments and Reauthorization
Act (SARA). Section 313 of SARA requires EPA to establish a national inventory of toxic
chemical emissions from facilities that have 10 or more full-time employees, that are in SIC
codes 20 through 39, and that manufacture, process, or otherwise use a listed toxic chemical in
excess of specified threshold quantities. SARA Title HI reports provide information on the
amount of chemicals released to the atmosphere from point or fugitive sources in units of pounds
per year. These estimates may be based on a variety of methods (U.S. EPA, 1987b), The EPA
maintains a data base of the SARA Title EQ reports on its computerized Toxic Release Inventory
System (TRIS). TRIS data can be retrieved and organized in a number of ways, including by.
state, county, pollutant, or by SIC code (U.S. EPA, 1987c).
Source Permit Data. Air toxic pollutant emissions on a state or local ievel are usually
controlled through source permits available from various agencies. To obtain a permit, the
operator must provide information about emissions rates and source parameters. The operations
of the source is modeled based on the information, and the emissions predicted by the model
must be in compliance with state or local air toxics standards. The permit identifies the specific
emissions limits demonstrated by the model, and typically provides information on the stack
4-12
-------
parameters (i.e., stack height, inner stack diameter, flow rate and exit temperature) used in
modeling compliance with the standards.
NATICH provides a registry of selected permitted sources of air toxics for cases that are
unique or of particular interest to other agencies. For each permit case, information is provided
on the agency, permitting contact, permit ID number, facility category, SIC code, year permit
issued, last year amended, control equipment, SCC if available, emission sources, pollutant
names, CAS numbers, and emission limits.
National Emissions-Data System (NEDS). State environmental control agencies collect
emission and source parameter data on new sources, modified sources, and source shutdowns
within their jurisdiction. For point sources, NEDS has now been completely replaced by the
AIRS Facility Subsystem, administered by the National Air Data Branch of the OAQPS (U.S.
EPA, 1990i). Historical NEDS data, back to the year 1985, will eventually be loaded into the
AIRS Facility Subsystem.
Sources with emissions greater than 100 tons per year have been reported in NEDS., but
the actual range of data available in the data base varies. Additional data are often available for
states that develop inventories with much lower ton per year reporting criteria (i.e., some states
send in the same inventory collected for their own state programs). The source parameter data
requested on the NEDS form is not provided in all cases.
NEDS emission data are reported in terms of total VOC and PM emissions per facility
per year. Speciation profiles from other sources can then be used to estimate emissions of a
particular chemical species. At best. NEDS data are only estimates and accuracy vanes
depending on how each estimate is derived. In the future, agencies will be able to enter their
emissions data by PC directly into AIRS. Because AIRS will perform some automatic data
quality checks, the quality of the data is expected to improve. For example, AIRS will check the
geographical location of sources for a state to ensure that source coordinates ore within the state
boundary.
PIPQUIC. The "Program Integration Project - Queries Using Interactive Commands,"
otherwise known as PIPQUIC, is a computerized data system which provides for the statistical
analysis and graphical display of urban air toxics emissions, exposure and risk for a given study
area. The system is accessible through the EPA National Computer Center. It provides
emissions inventory data for a number of established study areas: Philadelphia, Santa Clara.
4-13
-------
Baltimore, Kanawha Valley in West Virginia, Rhode Island, Chicago, Los Angeles, New
York/New Jersey, 35 selected counties nationwide, and selected state pilot programs. Air toxics
emissions inventories for other areas can be generated through a work assignment with the
PIPQUIC support contractor. Information about PIPQUIC is available from the Non-Criteria
Pollutant Assessment Branch of OAQPS.
Control Technology Center. The EPA Office of Air and Radiation (OAR) and the Office
of Research and Development (ORD) developed the Air Toxics Control Technology Center
(CTC) to assist state and local agencies in the implementation of air toxics control programs by
providing technical guidance and support on air pollution control technology. The CTC also
provides a mechanism to transfer available engineering information of broad interest through
workshops and seminars, publications, and computer software. Through CTC, telephone access
to EPA expertise as an initial, quick response to individual problems, as well as in-depth
engineering assistance. Finally, CTC encourages feedback to EPA on the technical support needs
of state and local agencies. The CTC has produced over 20 technical reports and several PC-
based software programs in response to needs identified by the air pollution control community.
The CTC bulletin board offers engineering assistance to state and local air pollution control
agencies through the OAQPS TTN (see Table 4-1) (U.S. EPA, 1991).
Table 4-2 summarizes the major methods that can be used to characterize emissions.
4.2.3 Selection Criteria for Emissions Characterization
The method to be used in characterizing emissions is determined by the scope of the risk
assessment, the available resources, and existing emissions information. The first step is to
identify what information currently exists about the sources and pollutants of interest. The most
accurate information can be obtained from a carefully planned and conducted emissions test;
however, it may not be feasible to test all the sources for which estimates are desired. It may
be possible to obtain existing emission test reports or to use data collected from a source (or
sources) with similar process operations. If adequate test data are not available, emissions
estimates may be derived from reliable and representative emission factors or mass-balance data.
In the absence of specific data, some assumptions and analogies will have to be made in order
to arrive at an emissions estimate. Assumptions and analogies made due to lack of data should
be made conservatively. Table 4-3 presents the variables for selection of methods described in
Section 4.2.2.4.
4-14
-------
I
2
LU
V)
FACTORS
SS
ds
oximately 20
lished
E
Knowle
produc
technol
charact
pr
pol
Ol
^r
01
a
a.
RCE
ir
s.
."=
al
S
S.I
O
cn
11
ii
§,§
^. o
'J
-------
oc
LU
UJ
a-
•2 i e
CD
M
QC
UJ
I
o
cr
O
LU
CO
CO
ac
V)
2.
s
•§
a.
&
£
•S
-g
1
5
o
o
o
UJ
C3
CJ
cc
•s
I
1
•3>
'-
o
I
Q
J'x
2
* §
I i ?"s
° " :s .2
If II
.a S S. «
I?i2
IIJl
^— J^- *~*
'x (O
5 S
i £
-
s
a.
LU
•5
.§
I
S •
5-8
•= UJ
II
O_ LU
4-16
-------
4.3 FATE AND TRANSPORT ANALYSIS
4.3.1 Overview
Once the pollutants of interest and their rates and sources of emission are defined, the
process of conducting a risk assessment continues with estimation of the pollutant fate and
transport. In this step, pollutant emissions are translated into concentrations to which the study
population is ultimately exposed.
The transport and dispersion of pollutants, in the context of established air quality models,
is addressed in. Section 4.3.2.1. Assessing the pollutant fate once it is released to the
environment is discussed in Section 4.3.2.2. These two sections emphasize fate and transport in
the atmosphere, while transport within soil and water media and uptake within the food chain are
discussed in Sections 4.3.2.3 and 4.3.2.4, respectively. Section 4.3.2.5 presents summaries of
some available fate and transport models and Section 4.3.3 presents information to aid in model
selection.
4.3.2 Technical Background and Methods
4.3.2.1 Atmospheric Transport and Dispersion. After air pollutants are released to the
atmosphere, their transport and dispersion are governed by fundamental meteorological principles,
as well as source-related characteristics. Initially, the diffusion of pollution is largely determined
by the source release characteristics, particularly the effective height of release. This effective
height is a combination of the physical release height and any additional rise which may be due
to buoyancy or momentum effects (in the case of stationary point sources).
Buoyant rise is driven by the temperature difference between the stack gas and the
ambient air and the gas volume flow rate. The amount of buoyant rise is also affected by the
thermal stability of the atmosphere into which the plume is released. Momentum rise is directly
proportional to the stack exit velocity and stack diameter and is significant when little
temperature difference exists between the stack gas and ambient air.
Wind and turbulence are important meteorological factors affecting pollutant dispersion.
Pollutants are naturally transported with the wind and are diluted with increasing wind speed.
(An increase in wind speed can also lead to suppression of plume heig.n:, augmenting plume
impact at ground level.)
Dispersion by circular motions (eddies) of varying sizes in the atmosphere is the principal
means of turbulent mixing. Turbulence may be both mechanically and buoyantly produced.
4-17
-------
Mechanical turbulence is exhibited by wind flow over objects, with magnitude dependent on the
type of terrain or the number, size, and spacing of the objects. Buoyant turbulence results from
strong surface heating during unstable atmospheric conditions. This strong solar radiation
produces rising thermals, vertical currents consisting of updrafts of heated air, and corresponding
downdrafts of air cooled by the entrainment of surrounding air. This circular flow pattern
frequently results in the formation of convective cells, and the dispersal of the plume over large
vertical distances. Rising thermals are lifted upward in the narrow convective updrafts.
Surrounding these updrafts are general regions of subsiding air; hence, during unstable conditions,
a plume becomes mixed within a relatively large vertical distance.
A variety of deterministic mathematical models have been developed to describe the
transport and fate of pollutants released to the atmosphere. A widely used formulation is the
Gaussian plume model, a statistical treatment of the atmospheric dispersion equation or the
species continuity equation. According to this model, the rate of change in pollutant
concentration is a function of: (1) the movement of the pollutant by the mean wind, (2) the
turbulent transport or mixing by variations in the wind, and (3) the amount of pollutant emitted
or removed by wet or dry .deposition.
For continuously emitting sources, pollutant concentrations in the Gaussian model are
assumed to be directly proportional to the pollutant emission rate and are diluted at a rate
inversely proportional to the wind speed at the height of release. Concentrations within the
plume are assumed to exhibit a normal or Gaussian distribution in the horizontal and vertical
directions and are, thus, a function of the receptor height and crosswind distance from plume
centeriine. Horizontal and vertical standard deviations of plume concentration (av and crz) depend
on the downwind distance from the release point, the mean wind speed, and the atmospheric
stability. Because the Gaussian equation satisfies principles of continuity, it is difficult to apply
it to situations involving pollutant sources and sinks along the plume path (e.g., chemical
transformation and deposition) (Hanna. 1985). Differences among Gaussian plume models
generally stem from variations in methods of determining o"y, o~z, effective plume height, and
wind speed at stack height (Hanna, 1985). Within the Gaussian model, contributions from
multiple sources are taken into account by summing the calculated ambient concentrations due
to each source at the specified locations.
4-18
-------
Emissions assumed to be uniformly and widely distributed (such as benzene from urban
gasoline stations) are often regarded as coming from area sources. Another way to solve the
dispersion equation is the box model, which is useful for continuously emitting area sources. For
these sources, emission rate is expressed in units of mass/(area-time) and concentrations can be
approximated with a box model solution, in which the mass flux into the box (the pollutant
emission rate times the along-wind width of the box) equals the mass flux out of the box (defined
by the wind speed, times the height of the box, times the pollutant concentration). The box
model solution can be derived from the Gaussian plume model equation, assuming a "narrow
plume" hypothesis, for which dependence on the crosswind distance (y) is eliminated (Hanna et
al., 1982). Wind speed throughout the box is generally assumed to be uniform and the source
emission rate constant. For large along-wind widths (grid distances), the box height may be
defined by the mixed-layer height, while for shorter grid distances it is often expressed in terms
of the vertical dispersion parameter (Gz). Removal processes, such as wet and dry deposition,
may be incorporated into the box model equation.
4.3.2.2 Pollutant Transformation and Deposition. The concentration of a chemical
released to the environment is determined not only by physical dispersion but also by physical,
chemical, and biological interactions between the chemical and other substances in the
environment. Prediction of chemical fate involves knowledge of aqueous solubility, vapor
pressure, air-water partition coefficient (Henry's Law constant), molecular diffusivity, phase
portion coefficient, melting point and absorbtivity (U.S. EPA, 1988a).
For risk assessment purposes, a pollutant compound may be considered chemically
reactive or unreactive, depending on whether reactions noticeably affect its concentration within
the time period modeled (Russell, 1988). A pollutant's degree of chemical reactivity may be an
important consideration; atmospheric chemical reactions may lessen the pollutant concentration,
through its transformation to other products, or conversely its concentration may be increased
through its formation from other compounds. In air pollution modeling, estimation of chemical
reactions is made feasible through a condensed treatment of the atmospheric chemistr , in what
is generally referred to as a chemical mechanism. This strikes a balance within the model
between computational feasibility and chemical detail. The chemical mechanisms were developed
to simplify hydrocarbon chemistry in photochemical modeling (i.e., the formation of O, and NO-,
4-19
-------
in photochemical smog), but the basis of these mechanisms may be extended to account for the
formation of other pollutants.
Pollutant emissions are also subject to other removal processes, particularly dry deposition
and scavenging by rain and clouds. These removal processes may significantly affect the fate
of hazardous pollutants released to the atmosphere.
Dry deposition involves pollutant transport to the earth's surface and subsequent physical
and chemical interactions between the surface and the pollutant. To simplify the computations
involved, air quality models express the rate of dry deposition m terms of a deposition velocity.
This is defined as the ratio of the ground-ward flux of the pollutant to the concentration of the
pollutant at some reference height.
Precipitation scavenging or wet deposition, is a function of the intensity and size of the
N-
raindrops and the solubility and reactivity of the chemical. For nonreactive gases, solubility is
the most important physical property to consider. Reliable values of solubility for various
ambient temperatures are needed for deposition models (Dana et al., 1984). Measurements of
solubility for ethylene oxide, nitrobenzene, and methyl chloroform are reported in "Hazardous
Air Pollutants: Wet Removal Rates and Mechanisms" (Dana et al., 1984).
4.3.2.3 Transport in Water and Soil Media. Toxic pollutants originally released to the
atmosphere may deposit on water or soil surfaces and affect individuals through these media.
The amount of pollutant deposited can be estimated through air dispersion modeling. Soil and
water surfaces may also be sources of atmospheric emissions through volatilization of deposited
pollutants.
In freshwater bodies, pollutant transport is more advective (moving with the mean flow)
than dispersive (related to eddy activity), but advective factors like wind and temperature are also
important (U.S. EPA, 1988a). In estuarine and coastal systems, the dispersive component is more
significant than the advective, with temperature and wind being important factors in defining the
rate of transport.
The sediment bed underlying the water body acts as both a soui ce and sink of dissolved
and paniculate pollutants. Many organic chemicals and heavy metals partition to the organic and
clay fraction of the sediment bed. Accumulation depends on the characteristics of the solids and
on the turbulence and shear at the interface between solids and water (U.S. EPA, i988a).
4-20
-------
Estuaries, lakes, and reservoirs tend to deposit pollutants in sediment, whereas rivers or streams
tend to reentrain previously deposited material.
In surface water models, deposited pollutants generally enter a water system from runoff,
precipitation, and groundwater discharge and exit by evaporation or downstream flow. The
Exposure Analysis Modeling System II (EXAMS II) is a model that simulates the transport and
fate of organic chemicals in surface water environments. The model combines the chemical
loadings, transport, and transformations into a set of differential equations, using the law of
conservation of mass as an accounting principle (U.S. EPA, 1990a). The total mass of chemicals
entering and exiting each section of a body of water is computed as the algebraic sum of external
loadings, transport processes that distribute the chemicals and export them across its external
boundaries, and transformation processes (such as photolysis, hydrolysis, biolysis and oxidation)
that convert the chemicals into daughter products (U.S. EPA, 1990a). Output produced by the
model includes the expected environmental concentrations (long-term, 24-hour, and 96-hour), the
chemical distribution in the aquatic system, and the relative dominance of each transport and
transformation process.
Soil compartment models commonly stratify the soil column into two or more layers. The
upper soil layer typically contains the most decomposed plant matter. The lower soil layer is
often defined as the unsaturated zone between the upper soil layer and the water table. In reality,
the depths of these layers can vary dramatically in various locations. Some models assume no
unsaturated zone as a worst-case scenario; in this case, the pollutant goes from the upper layer
directly into the water. The soil layers are characterized by parameters, such as depth, bulk
density (dry soil mass per unit volume), porosity, water content, and organic carbon fraction.
Pollutants introduced into the upper layer are removed by chemical transformation, volatilization,
runoff, uptake by plants, and downward leaching. Chemical transport to a lower soil layer is
estimated by the product of the recharge rate (liters/year) and the pollutant concentration
(mg/liter) in soil water (McKone and Layton, 1986).
The SESOIL model is one example of a soil compartment model. It estimates the
seasonal rate of vertical chemical transport and transformation m the soil column, based on mass
and concentration distributions among the soil, water, and air phases in the unsaturated soil zone
(Clark, 1987). The hydrologic cycle is simulated by means of probability density functions of
independent climatic input parameters. These result m probability distributions of surface runoff.
4-21
-------
evapotranspiration, and groundwater runoff (Clark, 1987). In the SESOIL model, the sediment
cycle is also simulated by means of stochastic techniques, which use random variables to estimate
chemical concentrations and mass quantities in the soil, soil moisture, and soil gas in various
zones of the soil column on a monthly or annual basis.
Although vital for estimating integrated exposure, models that simulate pollutant transport
and fate in soil and water media often cannot be applied to specific situations because the
required data is not available. As with any model, the credibility of the output depends entirely
on the certainty of the input. Models generally require input of various rate constants and
parameters that may be difficult to determine, particularly when attempting to simulate chemical
processes such as volatilization, photolysis, hydrolysis, oxidation, and bio transformation.
Multimedia models have been developed to integrate the potential human exposure to
pollutants that occur in more than one environmental media. An example of the multimedia
concept is provided by the GEOTOX model (McKone and Layton, 1986). This model was
developed to demonstrate that multimedia exposure could be approximated mathematically
(McKone, 1990). GEOTOX simulates the movement of pollutants in a generic landscape
comprised of the following compartments: the atmosphere, as paniculate matter or as gaseous
contaminants; the terrestrial biosphere; an upper and lower soil layer; groundwater; surface water;
and .a sediment layer located at the bottom of surface water bodies. In this model, the
partitioning, reaction, and interphase-transport characteristics of a chemical are based on a
combination of physical, chemical, and landscape properties. The model is designed as a
screening tool, intended U^ provide information on a chemical's characteristic behavior and the
resulting exposure to individuals living their full life in and receiving all of their air, water, and
food from the reference landscape. A box model is used to represent the atmosphere. The
chemical concentration within the box is assumed to be uniform and the height of the box is set
equal to the mixing height. Chemicals enter the atmosphere directly from a source or by
diffusion from soil and/or surface waters and leave through dry deposition, wet deposition.
advective loss due to wind transport, venting out the box top, and chemical decompositicr.
These three models (EXAMS H, SESOIL, and GEOTOX) and other soil and water or
multimedia modes are useful in approximating the transport and fate of pollutants in media other
than air. Once toxic chemicals deposit in soil and water, they can affect human health through
the food chain and through reentrainment into the atmosphere by vaporization.
4-22
-------
4.3. 2 A Uptake of Pollutants in the Food Chain. Calculation of the concentrations of
contaminants in food is a complex process requiring the integration of physical, chemical, and
biological factors. For example, plants accumulate pollutants via root uptake, direct deposition
onto plant parts, and air-to-plant transfer of vapor-phase pollutants. Many factors affect the
relative importance of each of these, including:
• plant type (leafy vegetables, exposed produce, such as fruits, protected produce
such as root crops, grains, and forage);
• pollutant type (organics or metals); and,
• duration of plant exposure (usually defined as the growing season at the affected
site).
Each of these factors carries its own set of constraints. For example, to calculate the amount of
chemical deposited on exposed vegetation, the interception fraction must be known. The
interception fraction is used to weight the annual deposition to account for the geometry of the
vegetation. Since leafy vegetables have broader, flatter surfaces than a string bean, for example,
they will intercept a larger proportion of the pollutant. The interception fraction can be
calculated as a function of plant spacing and plant geometry (U.S. EPA, 1990d); default values
have been calculated for some plant types (Baes et al., 1989).
Pollutants also enter the food chain through animals. For example, cattle accumulate
pollutants by ingesting contaminated food and soil. To estimate the amount of pollutants m beef,
the amount of pollutant in the forage, grain, and soil consumed by cattle and the biotransfer
factor for each type of animal tissue must be calculated. The biotransfer factor is defined as the
concentration of pollutant in animal tissue (or milk) divided by the daily intake of pollutant
(Travis and Arms, 1988). The concentration of pollutant in the jth animal tissue group A,, can
be calculated as follows (U.S. EPA, 1990d):
n
AJ = [I (QPij P,p + (QSj Sc)J Ba,
where:
Qpy = quantity of ith plant group eaten by the jth animal each day
Pu = total concentration of pollutant in the ith plant group eaten by the jth animal
QSj = quantity of soil eaten by the jth animal each day
Sc = soil concentration factor
a^ = biotransfer factor for the jth animal tissue group
4-23
-------
Fish from polluted water are also consumed by humans. The daily intake from fish
ingestion is calculated as the product of water concentration of pollutant, bioconcentration factor
(BCF), and fish ingestion rate (U.S. EPA, 1990d). The BCF is the ratio of the contaminant
concentration in an aquatic organism to the contaminant concentration in the water body.
A detailed description of food chain methodology can be found in Methodology for
Assessing Health Risks Associated with Indirect Exposure to Combustor Emissions (U.S. EPA,
1990d). Site-specific information to be used in this methodology may be available from local
officials. Every county has a Soil Conservation Service office which can supply soil data. The
agricultural extension agent may be able to supply information on locally grown crops. Default
values for many variables have been calculated (Travis and Arms, 1988; Baes et al., 1984).
4.3.2.5 Summary of Methods to Estimate Fate and Transport. The passage of the Clean
Air Act in 1970 and subsequent amendments in 1977 and 1990, promoted the development of
mathematical air quality models to simulate pollutant transport and estimate ambient
concentration. Some of the atmospheric dispersion models used in exposure assessments today
are models originally designed for regulatory purposes. The EPA OAQPS oversees the
distribution and application of models approved for regulatory use. The "Guideline on Air
Quality Models (Revised)," issued and updated by OAQPS, is a primary source of information
on the proper selection and application of these air quality models (commonly referred to as
guideline models) (U.S. EPA, 1986 and 1987a). The Technical Support Division, Source
Receptor Analysis Branch of OAQPS manages a Support Center for Regulatory Air Models
(SCRAM) bulletin board through the OAQPS TTN (see Table 4-2) (U.S. EPA, 1991). SCRAM
is the primary source of EPA information related to acquisition of computer code for regulatory
air models, changes to the models, and model-related news. Table 4-4 summarizes the features
of four models commonly used to simulate pollutant transport and estimate ambient concentration
of toxics. These four models will be discussed below.
ISC2. One of the most commonly used guideline models of atmospheric dispersion is the
Industrial Source Complex (ISC) modei (U.S. EPA, 1987b). Two versions of this model are
available, a long-term (ISCLT) and a short-term (ISCST). Both are steady-state Gaussian plume
models, used to predict pollutant concentrations from continuous point, area, or volume sources
associated with industrial source complexes located in flat or rolling terrain. In 1992, revised
ISC2 models (ISCST2 and ISCLT2) were developed as replacements for the previous versions
4-24
-------
of the model. The reprogramming did not modify the ISC algorithms; rather, the changes were
made to (1) improve the quality, reliability, and maintainability of the code, (2) improve user
interfaces by modifying the input and output file structures, and (3) provide better "end user"
documentation for the revised models (U.S. EPA, 1992b). For regulatory purposes, ISC2 has
superseded ISC; however, some human exposure models still contain the earlier version.
The ISC2 models account for the influence of multiple sources, settling and dry deposition
of particulates, downwash, plume rise as a function of downwind distance, and terrain below
stack top. A chemical decay coefficient may be specified to account for time-dependent pollutant
removal by physical or chemical processes. As with many dispersion models, ISC2 may be
executed in either a rural or urban mode, to take into account the observed differences in plume
dispersion between rural and urban areas.
The long-term version of ISC2 is based on annual (or seasonal) meteorological data in
the form of Stability Arrays. Generally referred to as STAR data, stability arrays display joint
frequencies of occurrence of wind speed, wind direction, and air stability by combining these
factors into a frequency distribution. In estimating long-term concentration or deposition, the
area surrounding a continuously emitting source is partitioned into sectors of equal angular width
corresponding to the sectors of annual frequency distributions of wind speed, wind direction and
stability. Annual emissions are divided among the sectors according to the frequencies of wind
blowing toward the sectors. Concentration fields predicted for all sources are expressed in terms
of a common receptor grid.
Preprocessed, hourly, sequential meteorological data are used by the short-term version
of ISC (both forms of meteorological data are available, for National Weather Service stations
throughout the United States, through the National Climatic Data Center in Asheville, NC).
ISCST calculates average concentration or total deposition in 1-, 2-. 3-, 4-, 6-, S-, 12-, 24-hour
and n-day time periods.
ISC models are appropriate for transport distances less than 50 kilometers from the model
grid origin. Either polar or cartesian coordinate receptor grids may be generated by the model,
with the option of predicting concentrations (deposition) at discrete, user-specified locations.
The EPA Human Exposure Model II (HEM-II) uses ISCLT to predict pollutant
concentrations from industrial facilities (U.S. EPA, 1990b). ISCLT model options programmed .
into HEM-II are: polar receptor grid, flat terrain, annual averaging, pollutant concentration, final
4-25
-------
plume rise, urban or rural mode, Briggs stack-tip downwash, and buoyancy-induced dispersion.
In situations involving intermediate (between stack base and stack top) or complex (above stack
top) terrain, the flat terrain assumption typically underpredicts concentrations. For these
situations, HEM-EL allows the output from other dispersion models (including ISCLT runs with
terrain option) to be used in the calculation of human exposure. Several of the models suitable
for modeling areas of complex terrain, such as Multiple Source Algorithm with Terrain
adjustments (MPTER) (Rao et al., 1982) and COMPLEX I (Turner, 1986) are available through
the SCRAM bulletin board. HEM-II also contains a meteorological data base of STAR data for
348 National Weather Service Stations throughout the United States.
TOXBOX. To estimate area source concentrations, HEM-n incorporates two box model
algorithms based on the work of Hanna et al., (1982). The more widely known of these the
ToxBox model is a steady-state box model that incorporates the removal processes of dry
deposition, wet deposition, and chemical decay. Concentration is defined by a mass continuity
equation, with the assumption of a Gaussian distribution of concentration in the vertical. An
annual average concentration for the box is obtained by solving the equation over the range of
atmospheric stability class, weighing each solution by the frequency of occurrence of the stability
class over all wind directions and wind speed categories.
GEMS. The Graphical Exposure Modeling System (GEMS), developed for the EPA
Office of Toxic Substances, uses ISCLT and an area box model algorithm to estimate annual
average concentrations (U.S. EPA, 1989, 1990c). The ISCLT application in GEMS is similar 10
that in HEM-n.
SCREEN. The EPA SCREEN model is a screening technique for estimating short-term
(essentially hourly average) pollutant concentrations due stationary point source emissions (U.S.
EPA,1988b). SCREEN uses a Gaussian plume model to estimate maximum ground level
concentrations based on worst case meteorological conditions (that combination of wind speed
and stability class resulting in the highest concentration for the source modeled). SCREEN is
capable of predicting concentrations in both simple (below stack height) and complex (above
stack height) terrain and is most appropriate for estimating impacts from -single point sources.
The EPA document entitled A Workbook of Screening Techniques for Assessing Impacts
of Toxic Air Poilumnts presents screening techniques for estimating short-term amoient
concentrations resulting from various types of toxic releases (U.S. EPA, 1988c). The workbook
4-26
-------
is structured around a series of release scenarios, categorized as paniculate, gaseous and liquid
phase. For each scenario, the workbook guides the user through estimation of emissions
(principally through the use of established emission factors), to prediction of atmospheric
dispersion and concentration. Impacts from continuous releases are reported as hourly averages.
TSCREEN, a model for screening toxic air pollution concentrations, is a computer program that
implements the procedures in the workbook (EPA, 1991).
Multi-Media Models. Additional models are available for predicting dispersion of
pollutants through soil and water. The Unified Transport Model .for Toxic,Materials (UTM-TOX)
is a multimedia model designed to predict the dispersion of pollutants through air, soil and water
(Clark, 1987; U.S. EPA, 1989). Atmospheric transport and dispersion are calculated with a
Gaussian plume model (in much the same manner as in ISCLT). Chemicals may be deposited
in wet or dry form on vegetation or soil and consequently transported through runoff and erosion,
percolation through the soil to a stream channel and in the stream channel to the outfall of a
watershed. Model output includes plots and tables summarizing average monthly and annual
concentrations in ambient air in the various wind sectors, in saturated and unsaturated soil layers,
in runoff and out of each stream segment. UTM-TOX is part of the GEMS.
The EPA Center for Exposure Assessment Modeling (CEAM) provides many exposure
assessment techniques for metals and organic chemicals released in aquatic and terrestrial
environments. CEAM currently distributes simulation models with the following applications:
• urban stormwater runoff;
• • leaching and runoff from soils;
• toxic and conventional pollution of streams;
• toxic and conventional pollution of lakes and estuaries;
• tidal hydrodynamics;
• geochemical equilibrium;
• aquatic food chain bioaccumuiation;
• unsaturated/saturated transport of pollutants in ground water;
• multimedia exposure from hazardous waste sites; and
• terrestrial exposure and food chain accumulation.
Model documentation and code may be readily obtained through the Center's electronic
Bulletin Board System (BBS). (To access the CEAM BBS, a user must call (404)-546-3402 and
follow the interactive prompts. Communication parameters are: 9600/2400/1200 baud, no parity,
8 data bits and 1 stop bit.)
Table 4-4 presents summaries of models that can be used to estimate fate and transport.
4-27
-------
c'-i
g.8
•§.2
•Ms
.a
.2
5*
<
*
>-
•sH-
ill.
s-I
s °»
S i
S.s
S''a -f;
2 -1 1 '5 3 i -T
- T .
"S
01
of
X
LU
_J
o_
o
a:
04
o
2 J
o V.
CO 15
o ^~
— c
t § .2
l^^2
oj
•1
m
u.
"S
•s
'I
5|
B >
2 £
Il
S §
11
.<=£
«
s
«
II
II
•8-5
'
CD
X
LU
I
i
s i-
II
O"
o
cc
J 1
•* •= o
A" -3 3
',- S -3
«• a i
-3s
i
a.
i
2 -S
i "3 -s <=
'IP
litl
^ ^ O. /n
s e »
- .§ o,
•fl.i
-
a.
2 '£
a. «
i z
a>
13
3.
8
-------
4.3.3 Selection Criteria for Fate and Transport Estimation
The best model or technique to use in estimating the pollutant concentrations to which
the study population may be exposed depends on a variety of pollutant, source and study related
factors. If the main intent of the study is to produce screening level estimates of potential
exposure, then a more conservative model based on worst-case dispersion assumptions should be
employed. A conservative model should assume that the chemical is not subject to removal
processes. The toxic pollutants of interest should be investigated to determine whether they
would be reactive or nonreactive over the time period of interest. If chemical reactions would
significantly reduce the pollutant's concentration, then it may be reasonable to consider a model
that at a minimum accounts for chemical decay. The fate of chemicals with high solubility
would be more realistically treated in models accounting for wet deposition. Similarly, it may
be desirable to account for the dry deposition of particulates, such as metals.
The number and types of sources must also be considered. Screening-level models such
as SCREEN are designed to predict concentrations due to a single source. Models such as the
Industrial Source Complex (ISC) model can accommodate multiple sources located within a
circular area of radius 50 kilometers. To assess the combined impact of sources on a regional
or nationwide scale would require use of an exposure modeling system such as HEM-II or
GEMS.
A study of chronic exposure would require a model designed to produce long-term
(annual) concentration estimates. Short-term concentrations for use in assessing acute pollutant
effects would require models capable of averaging over shorter time periods.
If indirect exposure pathways such as accumulation in the food chain and ingestion are
expected to be significant, food chain modeling may be considered. Food chain modeling is
generally complex, and techniques are less well established than ambient transport modeling
techniques.
Table 4-5 presents the models summarized in Section 4.3.2.5 in terms of variables to
consider for model selection. The EPA's Guideline on Air Quality Models (Revised) contains
more information about the strengths and weaknesses of the various models and should be
consulted before selecting a model.
4-29
-------
1
11
III
lii
s
&
UJ
I
UJ
i
o
o
01
(/)
up
« O
73 -5
•5 u
Q. £ -
-S.S •
•3 8-
5 c
-il
= Ti-
ll
-s £
I
S
c o
ill
^ c ^
III
8.2 "
e-2 H
8H
<8 . 3
>- OB- a
III
i-fl
"3 S S
-2-5 »
zs-I
Js|
g 5 a
e-o
8 c
C 33
||
i~^ CtJ
(O >
oc ^H.
ID O
U H
2 3
(13 03
Q cr
si
73
S
«
.= *
(O ,„
* S ^
f? I
<= § to
"2 ~ <»
I'l
II?
ill
a. a o
g o S
£ "3 5
-§
O
(0
1
I
O
?
en
1
f
3>
CO
O
oc
?
4-30
-------
LU
LU
CC
s
C9
cc
2
QC
LU
CC
O
o
LU
GO
tfi
2
S
cc
CO
CJ O
22 o
a
6
.c
3
£
51
lll-
O co *i,cc co Z co
I
a
4-31
-------
4.4 POPULATION CHARACTERIZATION
4.4.1 Overview
Through modeling and monitoring, the ambient concentrations of the pollutants can be
estimated geographically. Exposure and risk to.human populations via the inhalation route and
through secondary exposure routes (such as food and water contaminated by deposited pollutants)
involves combining pollutant concentration information with information on the geographic
distribution of people in the study area. Actual exposure (or dose) and susceptibility to pollutant
effects vary with age, occupation, intensity and amount of exercise, and with time spent in
various microenvironments. Depending on the level of detail involved in the assessment, it may
be important to consider various characteristics of the population, such as age, sex, and activity
patterns.
Section 4.4.2. discusses techniques available for characterizing study populations.
Mobility, as a unique population characteristic, is described in Section 4.4.2.2 and a summary
of various methods and data sources for characterizing a study population is given in Section
4.4.2.3, while Section 4.4.3 describes general factors to consider in selecting a method of
characterizing the study population.
4.4.2 Technical Background and Methods
The magnitude of inhalation exposure is principally defined by the atmospheric
concentration to which the individual is exposed; the exposure duration; and by individual's
inhalation rate and body weight. The atmospheric concentration will vary from place to place,
and will typically be higher in some places than others. The health effect attributed to amoient
concentration will depend on how much is actually inhaled, which in turn depends on the
individual's breathing rate. Commonly, risk assessments assume a default breathing rate •<[ 20
m3/day. However, inhalation rates are affected by the individual's age, sex, and activity. For
example, children have different inhalation rates than adults. In a detailed study, if there is
concern about effects on sensitive subpopulations, different respiratory rates couid be used for
different subpopuiations. The concepts of microenvironment and population cohort were
developed to assess the influence of these variables.
A cohort is a small, statistically homogeneous unit of population. In the NAAQS
Exposure Model (NEM), cohorts are developed on the basis of age and occupation (U.S. EPA,
1983). These cohorts are further classified into activity pattern subgroups. Activity patterns
4-32
-------
associate each cohort with a specific micro-environment and exercise level on an hourly basis.
A basic premise of the model is that each cohort is located either in a home or work setting
during each hour of the year.
A microenvironment is a place where the pollutant concentration is considered uniform.
Microenvironments potentially important for exposure assessment may be generally classified as
outdoors, in areas of high air pollution; outdoors in areas of low or medium air pollution; indoors
in an occupational setting; indoors in a residential or public setting and inside a transportation
vehicle (Sexton and Ryan, 1988). Accounting for indoor microenvironments may significantly
lower long-term population exposure estimates for pollutants initially released outdoors (Hayes,
1989).
4.4.2.1 General Population Characteristics. In order to estimate aggregate population
exposure, one must identify how people are distributed within the area of interest. Depending
on the scope of the assessment and the extent to which health effects data are known for the
pollutant(s) of interest, information about age, sex and activity patterns may be needed.
U.S'. Census Bureau Data. The U.S. Bureau of the Census is the major source of
demographic and geographic information. U.S. Census data, collected and revised every decade,
provide a complete population count of the entire United States population and more detailed
population and socio-economic characteristics for a subset of the entire population. Census data
are organized according to geographical area.
Data collected by the Census Bureau provides population counts down to the block level
(essentially city blocks, refer to Table 4-5 for the census definition). Questions asked of all
persons (sometimes referred to as short-form or 100-percent questions) provide basic population
and housing statistics, such as sex, race, marital status and housing type. More detailed
characteristics, such as occupation, migration (residence in 1985), presence of a disability, place
of work and journey to work are available for a subset of the entire U.S. population (data
obtained on the long-form or sample data; approximately 17 percent of all household units
nationwide, for the 1990 census). These characteristics are useful indicators of population
mobility. Census data may be obtained in printed reports or in computer tape files for ease in
assimilation (other dissemination media are microfiche and laser disks). Section 4.4.2.3 describes
U.S. Census data in further detail.
4-33
-------
Table 4-6 summarizes the various geographic units into which the 1990 census data are
categorized. These units are similar to those defined for the 1980 census, however, the Block
Group unit for 1990 is equivalent to and a substitute for the Enumeration District unit used in
previous censuses (U.S. Department of Commerce, 1989). Most useful for obtaining data down
to the level of towns and townships are the documents: Number of Inhabitants provides
population counts only, while General Population Characteristics provides population counts by
sex, by age, as well as household type, family type, and marital status. Population counts and
characteristics within standard metropolitan statistical areas (SMSAs) are available in a
publication called Census Tracts, Maps corresponding to the various geographic units are
available for use with the census data. A series of computerized summary tap files (STFs) are
also available. The STFs provide much greater geographic and subject detail than is possible in
x«
the published reports.
Table 4-7 summarizes the basic information provided in the 1990 census. Figure 4-1
provides contacts for Census Bureau information.
4.4.2.2 Population Mobility. Exposure over a given time period is a function of the
amount of time the population is estimated to be in various microenvironments and, therefore,
depends on the movement of people from one place to another. In a general sense, population
migration may be categorized as indoor-outdoor, within the study area, and out of the study area.
Population movement may be achieved through the assignment of population cohorts
(total population or specific subgroup) to various microenvironments based on a prescribed
activity pattern. Trip distribution models are another way of allocating people to various
locations (Young et ai,, 1985). A commonly used form of the trip distribution model is the
gravity model, in which trip interchange between areas is directly proportional to the relative
attraction of each of the areas, and inversely proportional to a function of the distance or travel
time between areas.
Land Use Data. Land use data are typically presented in map format and can be used to
identify where people are located. Population density can be determined by correlating people
with land use type (e.g., residential, commercial). Land use maps obtained from counry planning
commissions will generally delineate various residential, public, commercial and industrial areas.
Land use and land cover data, compiled by the U.S. Geological Survey, are available in digital
4-34
-------
TABLE 4-6. GEOGRAPHIC UNITS OF U.S. CENSUS BUREAU DATA
Statistical Area
Description
Region
Division
State
Metropolitan
Statistical Area (MSA)
County
Minor Civil
Division (MDC)
Place
Large, geographically contiguous group of states (with the
exception of Alaska and Hawaii). Four regions are defined:
Northeast, North Central, South and West
Subdivision of a region. Nine divisions are defined: New
England, Middle Atlantic, South Atlantic, East South Central,
West South Central, West North Central, Mountain and Pacific.
A primary governmental division of the United States.
Highly populated, economically integrated area defined by the
Office of Management and Budget as a Federal statistical
standard. An area qualifies as an MSA in one of two ways: it
contains a city of at least 50,000 with a total metropolitan
population of at least 100,000 (75,000 in New England). MSAs
are defined in terms of counties, except in New England, where
cities and towns (MCDs) are used.
The primary subdivision of a State.
The primary political and administrative subdivision
of a county (in 28 States). MCDs are identified by a variety of
legal definitions such as township, town borough, magisterial
district or gore.
May be one of two types. Incorporated: a governmental unit
incorporated under State law as a city, town, village or borough
and having legally prescribed limits, powers and functions.
Census designated: a statistical area comprising a densely settled
population that is not incorporated, buc resembles a incorporated
place.
(Continued)
4-35
-------
TABLE 4-6. (Continued)
Statistical Area
Description
Urbanized Area (UA)
Urban and Rural Area
Census Tract
Block Numbering Area
Block Group
Block
Encompass a central city and surrounding incorporated and
unincorporated areas which meet certain population size or
density criteria. Urbanized area differs from SMSA mainly by
excluding rural portions of counties comprising SMSA as well
as those places separated by rural territory from the densely
populated central city perimeter.
All persons living in an urbanized area and in places of 2500
inhabitants or more outside the urbanized area. Population not
classified as urban is divided into rural-farm and rural-nonfarm.
A small, relatively permanent division of a metropolitan
statistical area or selected nonmetropolitan counties. Census
tracts are designed to be relatively homogeneous for population
characteristics, economic status and living conditions, and to
contain between 2,500 and 8,000 inhabitants. Census tracts do
not cross county boundaries.
An area delineated cooperatively by the States and the Census
Bureau for grouping and numbering blocks in block-numbered
areas where census tracts have not been established.
Combination of census blocks comprising a subdivision of a
census tract or block numbering area. Block groups are equal to
and a substitute for the enumeration districts used for reporting
data in prior censuses. A block group consists of all blocks
whose numbers begin with the same digit in a given census tract.
A clearly identified piece of land, bounded by streets, roads,
railroad tracks, streams or other ground features. Blocks do not
cross census tract boundaries and are the smallest areas for
which census data are tabulated.
4-36
-------
<
£
•<
Q
3
<
uj
C£.
""^
03
C/3
CO
W
u
C/5
3
Q
O
d
UJ
2
<
f'T
••4
3
2
a-
1
c
• »
tX)
a-
M
• *•
4J
•s
c
O
'§
^s
u •
V3
3
u
8
09
U
i
s?
cs
M3
&
>»'
k
fi
11
3
8 •«
if
.2 "o
2 <«
3 ir
3§
a -a
"i'"3
•3 §
£ .2
.S
•S
2
mination
5!
vi
5
.2
«
^^
2
i
o
Sex, race, age, marital status, hispanic origin
U. S., States, MSAs counties, minor civil divisions, places, census tracts, block gr
3
•3
KJ3
&5
y*
^j
'-3
•c ^
a 3
u «a
^3 .-S
1 |
cj •<
a 2
•s ^
« -5
"3 w
1 f
U ~o
"r, 0
S U
so a
u
Q-oT
f.. 00
a o*
s *••
B <« S S
O — j
Social characteristics: Education, place of birth; ancestry; language spoken at
migration (residence in 1985); disability; fertility; veten
Economic Characteristics: Labor force; occupation.industry and class of
of work and journey to work; work experienc
income in 1989; year last worked
so
e
' ' 'fl
§
a
#
r-
••«
!
'./)
•a
•yj
<
*» &
8ft **•
1
2 "S
f'i
*J O
United States, regions, divisions, States, counties, places of 10,000 or more inhabi
of 10,000 or more inhabitants in selected States, MSAs and UAs. Sample data at
geographical subareas within States down to the block group level
u
3
m
<
.2
5
u
'3
I
SO
O
t»
w
^-\
a\
CO
O\
i
a
"o
s
1
5/5
D
BO
'l/»
0
•o
i
o
J
1
*s
V)
3
D
5 Js
a
o
•i!
eo •fl
Data products are available for the following geographic units: United States, Re
Stale, County, Minor civil division/census county division, place, census tract/bkx
area, block group and block
__
_o
a
0
5
Q>
1
f
oo
u
•l^
4-37
-------
U.S. Deportment of Conrmerce • Bureau of the Census •Washington. D.C 20200
. .<•*••*• -*--
-------
Telephone Contacts for Data Users
Demographic Programs
Decennial Census:
Content (General) - Al Paez (DPLD) 763-4251
Content and Tabulations (Program Design) -
Patricia Berman (DPLD) 763-7094
Count Questions -1990 Census -
Ed Kobilarcik (DPLD) 763-4894
Counts for Current Boundaries - Joel Miller (GEO) ... 763-5720
Count Information, Decennial Census -
Staff (POP) 763-5002/5020
Demographic Analysis - Gregg Robinson (POP) 763-5590
Litigation - Valerie Gregg (PPDO) '...•'. 763-7787
Post-Enumeration Surveys - Howard Hogan (STSD) . 763-1794
Reapportionment/Redistricting -
Marshall Turner/Cathy Talbert (DUSO) 763-5820/4070
Sampling Methods, Decennial Census -
Henry Woltman (STSD) 763-5987
Special Tabulations of Housing Data -
Bill Downs (HHES) 763-8553
Special Tabulations of Population Data -
Rosemarie Cowan (POP) 763-7947
Tabulations and Publications (General) -
Cheryl Landman/Gloria Porter (DPLD) 763-3938/4908
User-Defined Areas Program (Neighborhood
Statistics) - Adrienne Quasney (DPLD) 763-4282
Housing and Income:
American Housing Survey -
Edward Montfort (HHES) 763-8551
Components of Inventory Change Survey -
Jane Maynard (HHES) 763-8551
Income Statistics - Staff (HHES) 763-8576
Information, Decennial Census - Bill Downs (HHES) . 763-3553
Market Absorption/Residential Finance -
Anne Smoler/Peter Fronczek (HHES) 763-8552
New York City Housing and Vacancy Survey -
Margaret Harper (HHES) 763-8552
Vacancy Data - Wallace Fraser (HHES) 763-8165
Census Bureau Telephone Contacts for Data Users
April 1991
Editor: Mary G. Thomas
To contact any of the specialists by mail, use the name.
the division. Bureau of the Census, Washington. DC 20233.
To receive additional cooies of Census Bureau Teleohone
Contacts for Data Users, contact Customer Services at the
Census Bureau (301/763-4100). Address comments ex-
suggestions on this publication to Mary G. Thomas, Data
User Services Division, Washington Plaza, Bureau of the
Census, Washington, DC 20233 (301/763-1584).
International Statistics:
Africa. Asia. Latin America, North America, and
Oceania - Frank Hobos (CIR) 763-4221
China. Peopte's Republic - Judith Banister (CIR) ... 763-4012
Europe - Godfrey Baldwin (CIR) 763-4022
Health-Peter Way (CIR) 763-4086
International Data Base - Peter Johnson (CIR) 763-4811
Soviet Union - Barry KostinsJcy (CIR) 763-4022
Women in Development - Ellen Jamison (CIR) 763-4086
Population:
Age and Sex (States, Counties) - Staff (POP) 763-5072
Age and Sex (U.S.) - Staff (POP) 763-7950
Aging Population - AmokJ Goldstein (POP) 763-7883
Apportionment - Robert Speaker (POP) 763-7962
Child Care - Martin O'Connell/Amara Bachu (POP).. 763-5303
Citizenship - Slaff (POP) 763-7955
Commuting: Means of Transportation; Place of Work -
Phil Satopek/Celia Boertlein (POP) 763-3850
Consumer Expenditure Survey - Gail Hoff (DSD) ... 763-2063
Crime - Larry McGinn (DSD) 763-1735
Current Population Survey - Ronald Tucker (DSD) .. 763-2773
Disability - Jack McNeil (HHES) 763-8300
Education - Staff (POP) 763-1154
Employment, Unemployment -
Thomas Palumbo/Selwyn Jones (HHES) 763-8574
Estimates - Staff (POP) 763-7722
Families - Staff (POP) 763-7987
Farm Population - Don Dahmann (POP) 763-5158
Fertility/Births -
Martin O'Connell/Amara Bachu (POP) 763-5303
Foreign Bom - Staff (POP) 763-7955
Group Quarters Population - Oenise Smith (POP) . 763-7883
Health Surveys - Robert Mangold (DSD) 763-5508
Hispanic and Other Ethnic Population Statistics -
Staff (POP) 763-7955
Homeless Population - Cynthia Taeuber (POP) 763-7883
Household Estimates for States and Counties -
Staff (POP) 753-5221
Household Wealth - Enrique Lamas (HHES) 763-8578
Households and Families - Staff (POP) 763-7987
Immigration (Legal/Undocumented), Emigration -
Karen Woodrow (POP) 763-5590
Journey to Work -
Phil SalopeK/Gloria Swieczkowski (POP) 753-3350
Language - Staff (POP) 763-1154
Longitudinal Surveys - Ronald Dopkowski (DSD) . 763-2767
Mantai Status; Living Arrangements -
Arlene Saluter (POP) 763-7987
MetrocoWan Areas (MSA's) - Richard Forstall. (POP) 763-5158
Migration - Diana DeAre (POP) 763-3850
National Estimates and Projections - Staff (POP) ... 763-7950
Occupation and Industry Statistics -
John Priebe/Wilfred Masumura (HHES) 763:8574
Outlying Areas - Michael Levin (POP) 763-5134
Place of Birth - Kristin Hansen (POP) 763-3850
4-39
Figure 4-1
-------
Telephone Contacts for Data Users
Population Information - Staff (POP).... 763-5002/5020 (TTY)
Poverty Statistics - Staff (HHES) 763-8578
Prisoner Surveys - Larry McGinn (DSD) 763-1735
Race Statistics - Staff (POP) 763-2607/7572
Sampling Methods - Preston J. Waite (SMD) 763-2672
School District Data - Jane Ingotd (POP) 763-3476
Special Population Censuses -
Ronald Dopkowski (DSO) 763-2767
Special Surveys - Ronald Dopkowski (DSD) 763-2767
Stale Projections - Staff (POP) 763-1902
State and Outlying Areas Estimates - Staff (POP) ... 763-5072
Survey of Income and Program Participation (SIPP):
SIPP: General Information - Staff (DSD) 763-2764
SIPP: Statistical Methods - Raj Singh (SMD) 763-7944
SIPP: Products - Carmen Campbell (DUSD) 763-2005
Travel Surveys - John Cannon (DSD) 763-5468
Veterans Status -
Thomas Palumbo/Selwyn Jones (HHES) 763-8574
Voting and Registration - Jerry Jennings (POP) 763-4547
Women - Denise Smith (POP) 763-7883
Economic Programs
Agriculture:
Crop Statistics - Donald Jahnke (AGR) 763-8567
Data Requirements and Outreach -
Douglas Miller (AGR) 763-8561
Farm Economics - James Liefer (AGR) 763-8566
General Information - Tom Manning (AGR) 763-1113
Irrigation and Horticulture Statistics and Special
Surveys - John Blackledge (AGR) 763-8560
Livestock Statistics - Linda Hutton (AGR) 763-8569
Puerto Rico, Virgin Islands, Guam, and Northern
Mananas-Kern Hoover (AGR) 763-8564
Business:
Business Owners - Donna McCutcheon (ESDI 763-5517
County Business Patterns - Zigmund Decker (ESD) .. 763-5430
Minority - and Women-Owned Businesses -
Dorma McCutcheon (ESD) 763-5517
Foreign Trade:
Trade Data Services -
Staff/Haydn Mearkle (FTD) 763-5140/7754
Shipper's Export Declaration - Hal Blyweiss (FTD) ... 763-5310
Retail Trade:
Advance Monthly Sales. Annual Sales, Monthly
Inventories - Ronald Piencykoski (BUS) 763-5294
Census - Anne Russell (BUS) 783-7038
Monthly Retail Trade Report - Irving True (BUS) . . . 763-7128
Service Industries:
Census - Jack Moody (BUS)
Current Selected Services Reports -
Thomas Zabelsky (BUS)
Finance, Insurance, and Real Estate -
Sidney Marcus (BUS)
Utilities, Communication, and Transportation
Census - Dennis Shoemaker (BUS)
Wholesale Trade:
Census - John Trimble (BUS)
Current Sales and Inventories - Date Gordon (BUS)
Construction:
Building Permits (C40 Series) - Linda Hoyte (CSD) .
Census - Bill Visnansky (CSD)
New Residential Construction:
Characteristics, Price Index. Sales (C25/27 Series)
Steve Berman (CSD)
Housing Starts (C20 Series)/CompJetions
(C22 Series) - David Fondelier (CSD)
New Construction in Selected MSA's (C21 Series) -
Dale Jacobson (CSD)
Survey of Residential Improvements and Repairs
(C50 Series) - George Roff (CSD)
Value of New Construction Put in Place (C30 Series) -
AJIan Meyer (CSD)
Governments:
Criminal Justice Statistics - Diana Cull (GOVS) ...
Employment - Alan Stevens (GOVS)
Federal Expenditure Data - David Kelterman (GOVS) . .
Finance - Henry Wulf (GOVS)
Governmental Organization - Diana Cull (GOVS) . ..
Operations Support and Analysis -
William Fanning (GOVS)
Survey Operations - Genevieve Speight (GOVS) .
Taxation - Gerard Keffer (GOVS)
763-702
763-172
763-138
763-266
763-528
763-391!
763-724^
763-754<
763-784J
763-5731
763-7842
763-5705
763-5717
763-7789
763-5086
763-5276
763-7664
763-7783
763-7783
753-5356
; Manufacturing:
' Concentration, Exports from Manufacturing
I Establishments, and Production Index -
i Bruce Goldhirsch (IND)
i Durables (Census/Annual Survey) -
; Kenneth Hansen (INO)
Durables (Current Industrial Reports) -
; Malcolm Bemnardt (IND)
] Fueis/Electric Energy Consumed -
John McNamee (IND)
'• Industries - John P. Govoni (IND)
Monthly Shipments. Inventories. Orders -
Suth Runyan (IND) '
Nondurables (Census/Annual Survey) -
Michael Zampogna (IND)
Nondurables (Current industrial Reports) -
Thomas Flood (IND) "
763-1503
763-7304
753-2518
7S3-593S
763-7666
7S3-2502
763-2510
763-5911
Figure 4-1
4-40
-------
Telephone Contacts for Data Users
Research and Development, Capacity. Pollution
Abatement - Elinor Champion (INO) 763-5616
Special Topics:
County Business Patterns - Zigmund Decker (ESD) .. 763-5430
Economic Census Products - Paul Zeisset (ECS) ... 763-1792
Employment/Unemployment Statistics -
Thomas Palumbo (HHES) 763-8574
Enterprise Statistics - Johnny Monaco (ESD) 763-1758
Geographic Areas of the Economic Censuses -
Staff (GEO) 763-4667
Industry and Commodity Classification -
Alvin Venning (ESD) ..: 763-1935
Investment in Plant and Equipment -
John Gates (IND) 763-5596
Mineral Industries - John McNamee (IND) 763-5938
Quarterly Financial Report - Paul Zarrett (ESD) 763-2718
Accounting and Related Issues -
Ronald Lee (ESD) 763-4270
Classification - Frank Hartman (ESD) 763-4274
Puerto Rico, Virgin Islands, Guam:
Censuses of Agriculture, Construction. Manufactures,
Retail Trade, Services, and Wholesale Trade -
Odell Larson/Kent Hoover (AGR) 763-8226/8564
Transportation:
Truck Inventory and Use - William Bostic (BUS) 763-2735
Geographic Concepts and Products
Area Measurement -
Don Hirscnfeid (GEO) 763-5720
Boundaries of Legal Areas:
Annexations, Boundary Changes -
Nancy Goodman (GEO) 763-3827
State Boundary Certification -
Louise Stewart (GEO) 763-3827
Census Geographic Concepts - Staff (GEO) 763-5720
Census Tracts:
Address Locations - Ernie Swapshur (GEO) 763-5720
Boundaries, Codes, Delineation - Cathy Miller
(GEO) 763-3827
Centers of Population - Don Hirscnfeid (GEO) 763-5720
Congressional Districts:
Address Locations - Ernie Swapshur (GEO) 763-5692
Boundaries, Component Areas -
Robert Hamill (GEO) 763-5720
GBF/DIME System - Staff (DUSD) 763-1580
Maps:
1980 Census Mao Orders-Leila Baxter (DPO) . 812/288-3192
1990 Census Maps - Staff (OUSD) 763-4100
Cartographic Operations - Staff (GEO) 763-3973
Computer Mapping - Fred Broome (GEO) 763-3973
Metropolitan Areas (MSA's, PMSA's, and CMSA's) -
James Fitzsimmons/Ricrwd Forstall (POP) 763-5158
Outlying Areas - Staff (DFUD) 763-2903
Statistical Areas - Staff (GEO) 763-3827
TIGER System:
Applications - Larry Carbaugh (DUSD) 763-1580
Future Plans - Staff (GEO) 763-4664
Products - Staff (DUSD) 763-4100
Urban/Rural Residence - Staff (POP) 763-7962
Voting Districts - Cathy McCully (GEO) 763-3827
ZIP Codes:
Demographic Data - Staff (DUSD) 763-4100
Economic Data - Anne Russell (BUS) 763-7038
Geographic Relationships - Rose Quarato (GEO).. 763-4667
Statistical Standards and Methodology
Statisticat Research for Demographic Programs -
Lawrence Ernst (SRD) 763-7880
Statistical Research for Economic Programs -
Nash J. Monsour (SRD) 763-5702
DATA CENTER LEAD AGENCIES
(Data Centers are usually State government agencies or aca-
demic centers that disseminate data, often in customized
forms. State Data Center lead agencies in boldface. Business
and Industry Data Center lead agencies in italic. An asterisk
indicates both are combined in one agency.)
Alabama - Annette Waiters. University
of Alabama. Center for Business and
Economic Research
Alaska - Kathryn Uzik, Department of Lacor,
Alaska State Data Center
Arizona - Betty Jeffnes, Arizona Department
of Economic Security
Arkansas - Sarah Breshears. University of
Arkansas- Little Rock. State Data Center
California - Linca Gage, State Census Data
Center, Department of Finance
Colorado - Reid Reynolds, Coloraco
Department of Local Affairs
Connecticut - Theron Scnnure, Connecticut
Office of Policy ana Management .
Delaware - Judy McKinney- Cherry, Delaware
Development Office
District of Columbia - Gan Ahuia, Mayor's
Office of Planning
Florida - Sieve Kimcie. Florida State
Data Center . . . .
. 205/348-6191
.. 907/465-4500
. 602/542-5984
. 501/569-8530
. 916/322-4651
303/866-2156
. 203/566-3285
. 302/739-4271
. 202/727-6533
904/487-2814
4-41
Figure 4-1
-------
Telephone Contacts for Data Users
Florida - Sally Ramsey. Florida Department of
Commerce, Bureau of Economic Analysis ..
Georgia - Marty Sik, Georgia Office of
Planning and Budget
Guam - Peter Barcinas, Guam Department
of Commerce
Hawaii - Emogene Estores. Department of
Business and Economic Development ..
Idaho - Alan Porter. Idaho Department of
Commerce
Illinois - Sue Ebstsch, Illinois State Data Center
Coop., Illinois Bureau of the Budget
Indiana - Roberta Eads, Indiana State Library,
Indiana State Data Center
Indiana - Carol Rogers, Indiana Business
Research Center
Iowa - Beth Henning, State Library
Kansas - Marc Galbraith, State Library
•Kentucky - Ron Crouch, Kentucky State Data
Center.Urban Research Institute
Louisiana - Karen Paterson, Louisiana
Planning and Budget
Maine - Jean Martin, Maine Department
ofLaoor
•Maryland - Michel Lettre, Maryland Office
of Planning
904/487-
404/656-
671/646-
808/548-
208/334-
217/782-
317/232-
317/274-
515/281-
913/296-
502/588-
504/342-
207/289-
301/225-
2568
0911
•5841
•3082
•2470
•1381
3733
2205
4350
3296
7990
•7410
•2271
4450
•Massachusetts - Dr. Steve Coelen, University
of Massachusetts, Massachusetts Institute
for Social and Economic Research
Michigan - Eric Swanson, Mictiigan Information
Center, Department of Management and
Budget
Minnesota - Oavid Birkhoiz, Minnesota Slate
Planning Agency, State Demographic Unit ...
Minnesota - David Rademacher, Minnesota
State Demographer's Office
Mississippi - Pattie Byrd, University of
Mississippi, Center for Peculation Studies
Missouri - Marlys Davis, Missouri State Library .
• Montana - Patricia Roberts, Montana
Deoartment of Commerca
Nebraska - Jerome Oeicnert, University
of Neoraska-Omaha. Canter for Public
Affairs Research
Nevada - Betty McNeal, Nevada State
Library and Achwes
New Hampshire - Thomas J. Duffy, Office of
State Planning
' New Jersey - Connie O. Hughes, New Jersey
Department of Labor
New Mexico - John Beasiey, Economic
Development and Tourism Deoartment .
New Mexico - Juliana Boyle. University of
New Mexico. Bureau of Business and
Economic Research
413/545-3460
517/373-7910
612/297-2360
612/297-3255
601/232-7288
314/751-3615
406/444-2896
•102/595-2311
702/687-5160
503/271-2155
609/984-2593
505/827-0276
505/277-2216
New York - Robert Scardamalia. Department
of Economic Development
•North Carolina - Francine Stephenson, State
Data Center, North Carolina Office of State
Budget and Management
North Dakota - Dr. Richard Rathge, North
Dakota State University, Department of
Agricultural Economics
Ohio - Barry Bennett. Ohio Department of
Development
Oklahoma - Karen Seiland, Oklahoma
Departmentof Commerce, Oklahoma State
Data Center
Oregon - Maria Wilson- Figueroa, Portland
State University, Center for Population
Research and Census
•Pennsylvania - Michael Behney, Pennsylvania
State University at Hamsburg, Pennsylvania
State-Data Center
Puerto Rico - Lillian Torres Aguirre, Puerto Rico
Planning Board
Rhode Island - Paul Egan, Rhode Island
Department of Administration, Office of
Municipal Affairs
South Carolina - Mike MacFarlane, South
Carolina Budget and Control Board
South Dakota - DeVee Dykstra. University of
South Dakota, Business Research Bureau ...
Tennessee - Charles Brown, Tennessee State
Planning Office
Texas - Susan Tuiiy. Texas Department of
Commerce
Utah - Linda Smith, Office of Planning and
Budget
Vermont - Office of Policy Research and
Coordination
Virgin Islands - Or. Frank Mills. University of
the Virgin Islands
Virginia - Dan Jones. Virginia Employment
Commission
•Washington - Sharon Estee. Office of
Financial Management
West Virginia - Mary C. Hapless, Governor's
Office of Community and Industrial
Development
West Virginia - Linda Gulp. Center tor
Economic Research
Wisconsin - Robert Naylor. Demographic
Sen/ices Center. Department of Administration
Wisconsin - Michael Knight, University of
Wisconsin-Madison
Wyoming - Mary Byrnes. Department of
Administration and Fiscal Cootol
518/474-601
919,733-7CX
701/237-86;
614/466-21'
405/841 -51i
503/725-3&
717/948-63C
809/728-44;
401/277-64J
803/734-37?
605.677-525
615,741-167
512'472-505
SOV'SSS-IO;
802328-332
809,776-92C
8Q4~S6-83C
206 5S6-25C
304.348-^01
304/293-582
608266-192
608/262-309
307,777-750
4-42
Figure 4-i
-------
form, coded according to an established land use and land cover classification scheme (Anderson
et al., 1976). Digital data are available for 1:250,000-scale quadrangles nationwide.
4.4.2.3 Summary of Methods to Characterize Populations. This section describes some
commonly used exposure models that incorporate population estimation, microenvironments and
population cohorts. Other techniques for developing data sets for use in modeling are also
described.
Human Exposure Model-II. The Human Exposure Model-H (HEM-II) uses nationwide
1980 Block Group/Enumeration District (BGED) and 1990 Block Lavel data from the U.S.
Bureau of the Census. The population data base contains the population of each BGED or Block
and its latitude and longitude coordinates for mapping with the predicted concentration data. An
option is provided to allow the population to grow, from the base year to the specified study date,
based on user-defined or census projections of county-level growth rates. Populations in different
geographic areas are allowed to grow at unique rates.
In the default mode, HEM-n allows for differentiation of*pollutant concentrations within
an outdoor (ambient) and indoor microenvironment. Outdoor concentrations are not modified,
while indoor exposure levels are adjusted to 60 percent of outdoor exposure levels. These
pollution correlation coefficients may also be selected by the user and may be specified for
geographic areas of interest. Up to 10 microenvironments may be defined by the user.
Occupancy coefficients are used to weight the proportion of time spent in each
microenvironment.
NEM. The NEM was developed by OAQPS to evaluate the impact of alternative ambient
air quality standards in urbanized areas. There are two versions of the model: the exposure
district and the neighborhood-type versions. The exposure district version estimates the pollutant
concentrations expected to occur in selected exposure districts within a study area. This version
is more appropriate for assessing exposure due to sources of pollution that are widely distributed
(U.S. EPA, 1983).
Population within each exposure district is assigned to a single discrete point, or
population centroid. Concentrations are adjusted to account for five microenvironments: indoors
at work or school, indoors at home or other locations, inside a transportation vehicle, outdoors
near a roadway, and outdoors at other locations. The movement of population cohorts through
the exposure districts and microenvironments is simulated on an hourlv basis.
4-43
-------
The total population of each study area in the exposure district version of NEM is divided
into 12 age-occupation groups. Each age-occupation group is further subdivided into 56 groups
representing typical patterns of activity. For each hour of simulation, activity-pattern subgroup
members are assigned to one of the five microenvironment types and to one low, medium, or
high levels of exercise. Unique activity patterns are associated with weekdays, Saturdays, and
Sundays.
The other version of NEM divides the study area into zones, termed "neighborhood
types," which are classified according to the types and.intensities of emission sources within
them. This version is more appropriate for assessing exposure from specific point sources of
pollution. For example, the neighborhood version has been used to analyze carbon monoxide
(CO) exposure because source patterns of CO commonly vary with neighborhood type (e.g.,
-v»
commercial, industrial, residential).
Cohorts within the neighborhood version are defined as groups of individuals who all live
in the same neighborhood type, work in the same neighborhood type, are members of the same
age-occupation group, and are members of a subgroup with a specified daily activity pattern.
SHAPE. The Simulation of Human Air Pollution Exposure Model (SHAPE) model was
designed to estimate human exposure to carbon monoxide (CO). To attain exposure estimates,
the model simulates the activity patterns of urban dwellers, allocating each minute of the day to
one of 14 microenvironments (e.g., home, automobile, office, sidewalk, parking garage).
Individuals are exposed to varying concentrations of CO as a function of time spent in the
differing microenvironments. The duration associated with a specific activity and the transitions
from one activity to another are treated as random variables, sampled from probability
distributions derived from activity pattern studies (Thomas, et al., 1984). CO concentrations are
provided by a random number generator, based on empirical field studies. The background
concentration level is defined from the CO data collected from one or more fixed monitoring
stations located in the urban area of interest. Sequential rrucroenvironmentai exposures and urban
background data are combined to produce an "exposure profile" for an individual, from which
hourly and S-houriy moving average CO concentrations are calculated over a 24-hour period.
Other Techniques. During the 1980s, human activity pattern studies conducted in the
cities of Denver, Washington, and Cincinnati resulted in an activity data base appropriate for
NEM, SHAPE, and models of similar nature. Through an OAQPS work assignment, an Activity
4-44
-------
Pattern Simulator (APS) methodology and computer software was developed to utilize this data
base, create files representing specific population group activities in terms of probability, and then
use these files to produce activity patterns consisting of minute-by-minute exposure state
assignments (draft report: Johnson, 1987). As in NEM and SHAPE, the exposure state refers to
a combination of microenvironment and respiration rate. In this probabilistic, model, five
exposure states are used by the program to produce activity patterns:
Vehicle-affected — low or medium breathing rate
Vehicle-affected — fast breathing rate
Other indoors — unspecified breathing rate - - -
Other outdoors — low or medium breathing rate
Other outdoors — fast breathing rate.
In generating the activity patterns, it is assumed that the respondents of an activity diary
study, possessing a certain set of demographic characteristics, will yield similar activity patterns
by season and day of the week. Furthermore, the identity and duration of an exposure state being
entered, for a particular combination of season, day of the week and population group, are
determined only by the exposure state being vacated and the time of day. APS output includes
such general information as month, day (weekday, Saturday or Sunday), the initial random
number used to start the sequencer program and population group. The generated activity pattern
is presented in terms of: clock time an event begins, the corresponding minute (out of 1440
minutes in a day), the numerical code of the exposure state entered and its associated
microenvironment and breathing rate, and the duration of the event.
A method for determining urban population distributions for a contiguous subset of census
tracts within an SMSA was developed for the EPA Environmental Monitoring Systems
Laboratory of the EPA's ORD (Young, 1985). In this method, the study area is segmented into
1-km2 grid cells. Based on computerized land use and land cover data from the U.S. Geological
Survey and census tract data from the U.S. Bureau of the Census, the area associated with several
land use categories and the proportions of census tracts within each grid ceil are determined.
Population distributions in specific areas of the grid ceil are estimated during twelve 2-hour
periods of the day (expressed in terms of person-hours), based on time-of-day trip data from the
Nationwide Personal Transportation Study, Bureau of Labor Statistics work-shift data, and trip-
production and trip-attraction rate algorithms. Additional data required at the census tract level
are total resident population, total dwelling units, and total automobiles. In essence, the method
calculates the resident population within each grid cell, based on land use and census data, and
4-45
-------
subsequently predicts the number of trips attracted to and produced from (directed out of) the
grid cell as a function of the number of land use types, employees, automobiles, and dwelling
units within the grid cell.
Field Studies. Field studies are conducted to determine specific characteristics of a study
population. Common techniques used to define human behavior include questionnaires
(administered face-to-face or by mail) and time diaries. Careful planning and implementation
is required to obtain accurate and adequate responses with these techniques (U.S. EPA, 1989).
Typically, a time diary will track an individual's activities over a day or week on an
hourly basis. In addition to the type of activity, participants may also report where the activity
took place, with whom, and any associated secondary activities (Robinson, 1988).
Time diary data can be analyzed to obtain a composite of a wide range in daily behavior,
such as work, commuting and other travel, household "activities, recreation, and sleeping.
Furthermore, the times and locations at which activities occur provide useful information that can
be combined with data on the potential exposure associated with such locations.
National time-use studies have been conducted over the past few decades (Robinson,
1988). Comparison of these studies reveals how time spent in various activities changes,
indicating a need to periodically revise tune-activity estimates used in exposure modeling. Time
budget data derived from these studies has been used in developing population activity data
bases, such as that accessed by NEM.
Market Research. Specialized population data sets may be developed through market
research organizations. Through a market research organization, for example, latitude and
longitude coordinates have been associated with a U.S. Census Block Group/Enumeration District
file, allowing computerized mapping of the population data (U.S. EPA, 1985). Local activity
pattern studies may also be conducted through this type of organization.
4.4.3 Selection Criteria for Population Characterization
Determining how to characterize the study population will depend on the scope of the
assessment and the level of available information on the population of interest. The U.S. Bureau
of the Census is the most comprehensive, single source of population information and through
programs such as its User-Defined Areas Program, individuals may acquire population and
housing information for locally-specified geographic areas on a user-fee basis. Land area
4-46
-------
associated with all census geographic units, and latitude and longitude coordinates will be
provided in the 1990 census data, facilitating mapping of population data.
On a local scale, it may be possible to develop representative activity pattern data for use
in microenvironmental studies; however, for this data to be most useful it would also be
necessary to determine how the outdoor pollutant concentrations are modified within the
microenvironments.
4.5 EXPOSURE CALCULATIONS
4.5.1 Overview
In exposure calculation, the pollutant fate and transport elements are combined with the
population characterization elements to estimate human exposure. Section 4.5.2.1 discusses
techniques used to estimate exposure to ambient air concentrations. A brief discussion of the
concepts of exposure, intake, and dose are included in Section 4.5.2.2, along with a method for
dose estimation. A summary of various methods available for calculating population exposure
is given in Section 4.5.2.3. Section 4.5.3 outlines points to consider in determining what
exposure calculation method to select.
4.5.2 Technical Background and Methods
4.5.2.1 Exposure Estimation. Human exposure models are based on output from air
quality dispersion models. They generally match the predicted pollution concentrations with
population data on the BGED level. Exposure estimation models that produce concentrations
over a spatial receptor grid require some method of either interpolating concentrations to
locations of population or interpolating populations to locations of predicted concentration.
Interpolation is necessary because fate and transport models usually predict ambient
concentrations at discrete points within the study area (e.g., at polar grid locations defined by
distance and direction from a source). However, population data (e.g., BGED level census data)
are not compiled using the same grid system. The interpolation process is a source of uncertainty
in exposure estimation.
A related issue in exposure estimation is characterization of the levels of exposure. In
order to fully describe the risk, assessors need to present information on a range of exposures,
from the worst case (the hypothetical maximum possible exposure) to the median or geometric
mean exposure, in which one-half the population would be exposed to a level above and one-half
to a level below (U.S. EPA, 1992). These levels of exposure can be determined in different
4-47
-------
ways. To determine maximum exposure, it can be assumed that people are exposed to the highest
modeled ambient concentration, although in reality no people may reside at the location of
highest concentration. A maximum level of exposure would serve as a simplifying assumption
and would support the most conservative safety standards. A more detailed study may use
BGED census data or even more detailed, site-specific data on population distribution, to support
a more limited exposure level.
It is often assumed, especially in screening studies, that exposure to the modeled, outdoor
concentration occurs 24 hours per day at the place of residence. This does not^account for
contrasting indoor and outdoor concentrations or for population movement between areas of
differing ambient concentration. Some models adjust concentrations to reflect the amount of time
the population is assumed to be outdoors. Other models adjust monitored pollutant
X-
concentrations to reflect levels for various microenvironments.
Risk assessments for routine air emissions from stationary sources typically focus on
cancer and noncancer health effects resulting from chronic exposures. The average lifetime
exposure is the measure of interest in such studies. Various units can be used to express average
lifetime exposure. The units selected partially depend on the form of the dose-response model
output, since the exposure assessment and dose-response assessment must be combined in the risk
characterization step (Chapter 5). When the dose-response assessment is expressed in the form
of an URE (i.e., risk per ug/m3 or ppm ambient concentration), then the lifetime exposure should
be expressed in units of ambient air concentration (average ug/m3 or ppm). The general equation
for lifetime exposure would be (U.S. EPA, 1989a):
Lifetime Ambient Air x Exposure
Exposure = Concentration Duration
Lifetime (70 years)
Inhalation exposure can also be expressed in units of lifetime average mg of pollutant inhaled
per kg of body weight per day. These units would be compatible with the "slope factors"
discussed in the dose-response assessment section of Chapter 2. The general equation for
expressing inhalation exposures in these units is:
Lifetime
Average Inhalation x Ambient Air x Exposure
Inhalation = Rate Concentration Duration
Exposure Body Weight x Lifetime
4-48
-------
Inhalation rates vary depending upon exertion level, sex, and age. Ranges of measured values
are presented in the literature (U.S. EPA, 1985). Common default values (which are inherent in
the URE approach) are 70 kg average adult body weight and 20 nrVday inhalation rate.
4.5.2.2 Dose Estimation. The actual dose an individual receives is the amount of a
pollutant taken into the body. To estimate actual dose, another conservative and simplifying
assumption is often used. This assumption is that when an individual is exposed to a given
pollutant concentration (the amount present in the atmosphere), the individual is affected by all
pollutant available. In this case, the dose rate of the individual (in units of ug/day, in this
example) is equal to:
Concentration in- Rate of Air
Dose Rate = Atmosphere x Consumption
(e.g., ug/day) (e.g., ug/m3) (e.g., 20 nrVday)
Total dose is then calculated from the dose rate and the duration of exposure. A
distinction can be made between total dose and biologically effective dose because all of the
chemical taken into the body may not be absorbed or reach the critical organ. The biologically
effective dose is the concentration or amount of a chemical that reaches receptor site(s) in the
exposed individual to exert a toxicologic effect. However, in the absence of evidence to the
contrary, 100 percent absorption is conservatively assumed. Unless the exact relationship between
total dose and effective dose is known, it must be assumed that these doses are equal.
4.5.2.3 Exposure Due to Ingestion Pathway. The methodology for assessing exposure
due to the ingestion pathway is less well-established than that for inhalation. This is due to the
complexity of even the simplest model that adequately describes the ingestion pathway. Even
more critical is the difficulty of getting site-specific data to parameterize a moaei.
Information on human physiology, human benavior patterns, and environmental rate and
transport must be integrated to determine exposure by ingestion. The following discussion is
based on Methodology for Assessing Health Risks Associated with Indirect Exposure to
Combustor Emissions (U.S. EPA, 1990b). An alternative approach is described in McKone and
Daniels (1990) and McKone (1989).
Direct ingestion exposure may result from the ingestion of contaminated grains, fruits, and
vegetables. Pollutants may accumulate in the tissues of animals who eat contaminated vegetation
4-49
-------
or in fish from polluted surface waters. Human consumption of this contaminated meat, milk,
or fish represents other sources of exposure. This ingestion exposure can occur in the vicinity
of the pollutant source or some distance away should contaminated food be transported to other
locations or markets.
After contaminant concentrations in food have been calculated, the lifetime consumption
of each food type must be calculated. It is necessary to determine the proportion of the diet that
is locally grown on commercial farms or in backyard gardens as compared to imported foods,
as well as the consumption patterns and locations of locally grown foods.
In addition to foods, the ingestion of water and soil is included in a food chain analysis.
Soil ingestion is generally higher for very young children (ages 1 through 6) than for adults, and
includes inadvertent ingestion as well as abnormal soil consumption (pica). Default values and
•v*
methodologies are described in the Exposure Factors Handbook (U.S. EPA, 1989c).
The target population may be evaluated with respect to age, diet, and activities to
determine the exposure to pollutants through various ingestion pathways. Generally, at least two
populations are considered: the average adult and the average child. The definition of these
"average" individuals typically includes their body weight, life span (or duration of childhood)
and the length of time spent in the target area (U.S. EPA, 1990b). However, highly exposed and
highly susceptible subgroups such as the sick, the elderly, pregnant women, and nursing mothers
should also be considered.
4.5.2.4 Summary of Methods to Estimate Exposure. This section summarizes the various
methods used to estimate population exposure. For models that produce concentrations over a
spatial receptor grid, methods of interpolating concentrations to locations of population or
interpolating populations to locations of the predicted concentration are discussed. Models
without spatial grids may estimate exposure by passing individuals through microenvironments
with designated concentration levels.
Screening Level Assessments. For a quick approximation of the magnitude of potential
exposure to a given pollutant, an estimate of the maximum pollutant concentration may be used
without direct consideration of the study population. This approach, like the concept of the
maximum exposed individual, assumes that a hypothetical person resides at the point of
maximum predicted concentration. This maximum pollutant concentration is compared to an
4-50
-------
established exposure threshold for the pollutant to determine whether potential risk to human
health could occur.
EPA's A Tiered Modeling Approach for Assessing the Risks Due to Sources of Hazardous
Air Pollutants (U.S. EPA, 1992a), takes the assessor through a three-level screening process that
becomes increasingly more complicated and less conservative. Thus, if the level of concern is
not exceeded in the most conservative first tier method (a simple look-up table), the more
complicated modeling approaches of the higher tiers will not be necessary.
The EPA Guideline on Air Quality Models (Revised) recommends various screening level
techniques to produce conservative ambient air concentrations (U.S. EPA, 1986a, 1987). The
model selected should be appropriate for the type of terrain and predominant land use (urban or
rural) in the study area. A representative set of meteorological conditions for each source locale
should be used in the modeling analysis. A majority of assessments will require predictions of
annual average concentrations. Estimates of long-term pollutant impacts vary from year to year,
based on annual variations in meteorological conditions. A climatic period of 5 years should
be used in predicting the maximum pollutant concentration because 5 years generally represents
adequate variation in both annual and short-term impacts (U.S. EPA, 1986a, 1987).
HEM-II. The exposure methodology within HEM-II involves matching the annual
average concentrations predicted over the receptor grid with census BGED or Block populations
located within the grid (U.S. EPA, 1990).
Polar grid concentrations resulting from point sources are interpolated linearly between
direction radials and exponentially along them to obtain concentrations at the BGED or Block
centroids within the study area (log-linear interpolation). For a given pollutant, the highest
predicted BGED concentration is identified, and maximum individual exposure and risk are
determined for this concentration. The maximum exposure is simply this highest BGED
concentration. The total maximum exposure is the product of the maximum exposure and the
associated BGED population. The maximum individual risk is calculated as the product of the
maximum BGED concentration and the unit risk estimate for the given pollutant. The unit risk
estimate is the dose-response factor in cancer risk assessment. The unit risk factor represents an
upper bound estimate for the slope of the dose-response line for cancer.
For assessing exposure in densely populated study areas, BGED or Block populations may
be allocated to particular ceils of a master grid. The master grid is a rectangular array of grid
4-51
-------
cells that represents a geographic region. Each master grid cell becomes a pseudo-BGED, with
coordinates corresponding to the center of the grid cell and a population equal to the total of all
BGED populations within the grid cell. Source impact areas are expressed in terms of the master
grid, and each cell (pseudo-BGED) within the impact area is identified. Pollutant concentrations
predicted on the polar grid are then interpolated to the pseudo-BGED centroids, as mentioned
above.
Cumulative exposure and risk estimates are also produced in HEM-II. Population
exposures for given concentration intervals are calculated as the cumulative product of the BGED
or Block population exposed to a concentration within the interval and the concentration to
which that BGED or Block population is exposed. The cumulative totals of population and
exposure are determined at each interval to yield the total population exposure for the study.
To determine cumulative risk estimates, risk levels are also assigned. The risk (for
example, the number of probable cancers) is calculated by multiplying the pollutant unit risk
estimate by the exposure for each BGED or Block centroid. Risks for BGED or Block centroids
within an interval are summed to obtain the population risk for that interval. The cumulative risk
represents the risk to the entire population over 70 years (the assumed human life span).
Dividing the cumulative risk by 70 gives an estimate of the probable number of cancers occurring
each year in the study population.
Exposure and risk tables are presented for each pollutant and source modeled. Depending
on the number and spatial extent of the sources modeled, summaries are also made for source
groups and for the HEM-II study overall. A Source Group is comprised of sources whose impact
areas overlap (for point sources no more than an area of 50-kilometer radius). Maximum
exposure in the Source Group report is determined by summing the contribution of each source
at each BGED or Block location. Cumulative exposure now encompasses a larger grid domain.
Therefore, the source group cumulative exposure values are equal to the sum of the source-
specific cumulative exposures.
Exposure and risk estimates generated by HEM-II may be adjusted to reflect the influence
of microenvironments. This is done with occupancy coefficients, which define the percentage
of time spent by the population in the defined microenvironments (the total number of
coefficients summing to one) and by pollutant correlation coefficients, which define the
microenvironment concentration as a percentage of the predicted outdoor concentration.
4-52
-------
The exposure and risk estimates generated by HEM-II are based on certain simplifying
assumptions, because of data limitations and in order to facilitate the model's general use. The
primary assumptions are as follows:
• Exposure is assumed to occur at population-weighted centers of U.S. Census
Block Group/Enumeration Districts or Blocks, since actual residence location
information does not exist in the available data bases.
• Given the definition of the unit risk estimate, people are assumed to reside
at the population centroids for an entire year and pollutants are assumed
to be emitted at a constant rate during that year. In addition, the
conditions are assumed to persist for the averaging time of the URE
• Pollutant concentrations are predicted assuming a homogeneous, flat terrain
over the study area.
N-
Recent improvements to the HEM-II model have expanded the capability to quantify key
uncertainties associated with the model results. Monte Carlo analysis is useful in characterizing
uncertainties related to key input variables. HEM-II now has the capability to carry out Monte
Carlo analysis where input variables may be described and sampled as distributions rather than
point estimates. The resulting risk estimates can then be presented as a distribution. The revised
HEM-II model allows for six input variables to be described as distributions. These include the
unit risk factor, emission rate, microenvironment concentrations, time spent in
microenvironments, years at present residence, and variability in concentration predicted at
receptors. For each of these variables, HEM-II allows for one of several distributions to be
selected which best describes the distribution for that particular input variable. These include a
lognormai, normal, umfrom, empirical (using actual data), triangular, and Johnson series.
Characteristics of HEM-II are summarized in Table 4-8.
GEMS. Within the GEMS Atmospheric Modeling System (otherwise known as GAMS),
the polar grid is partitioned into sector segments, which represent a given wind direction and ring
distance interval (U.S. EPA, 1989b). Each segment encompasses an area between two
neighboring polar grid rings and two direction radials. Concentrations are predicted along the
primary direction radials, at each ring distance, and at three intermediate points between rings.
The sector segment concentration is determined by averaging the three intra-ring and the two-ring
concentrations corresponding to the segment. The total population for each sector segment is
estimated by summing the populations of each BGED centroid located within the boundaries of
4-53
-------
UJ
Q
O
UJ
DC
§
0_
X
UJ
ID
I
UJ
I
oc
CO
op
•4
LU
m
XPOSURE MODEL-II (HEM-II)
UJ
z
^5
i
0.
UJ
lETHOD:
2
ra
T:
D
2
t:
a.
c
ra
j—
0
.t
^.^
c
a
a>
1
§
3
1
CO
CD
2
i
CO
m
£
a
i
4)
Q.
CD
O
O
co
1
"o
a.
CO
^j
g
a>
i
•g
i
o
CO
1
•5)
'•5
.1
2!
CD
E
o
2
-**
8
3
Nationwide population count distributed by
Census Data
a
"3
o
J
JS
1
0.
CD
H
?
.0
1
C
.2
"S
STAR data for National Weather Service S
3
Q
15
o
'ra
leteorolo
2
1
CO
c
Default: Indoor/outdoor
User may specify up to 10 microenvironme
in
T3
2
CD
O
Q
UJ
a
03
O
.2
1
"H
8
Log linear Interpolation of receptor grid con
Identify maximum concentration for MEI
c
o
"5
u
•a
O
£
>n
O
O.
X
UJ
„
•^
«)
Latitude and longitude of sources
Source emission rate and release characte
Unit risk factor for carcinogenic pollutant
Q.
*^
2 2
| 8
i f
1 1
X 03
C "^
£^
2 8.
o "ft
*- w
.10 .2
1 1
« «
£ =6
For carcinogens: Cumulative exposu
For noncarcinogens: MEI concentration;
3
i"
o
o
'c
ID
1
o -
o
£
1
X
0)
1
CT
J
E
w
O)
C
'£.
CO
£
.52
*w
"ra
'^
Estimation of aggregate and maximum indh
compounds
"S
-^
c ^
ecomme
pplicatiol
cr <
4-54
-------
the sector segment. The annual population exposure for each segment is calculated by
multiplying the average concentration of the sector segment by the total population for the sector.
The cumulative population exposed as well as the cumulative population exposure are also
estimated.
SHED Model. One of the two basic models used in the original version of HEM was the
Systems Applications Human Exposure and Dosage (SHED) model (U.S. EPA, 1986b). This
model is frequently referred to as HEM or HEM1. This model is still available for use on the
EPA VAX cluster. SHED employs two schemes for matching pollutant concentrations and
population. The first scheme pertains to population centroids relatively distant from the source
and the second to populations near the source. The two schemes are required because close to
the source, a single concentration point on a polar grid represents a relatively small area
x»
compared to the area associated with a BGED. Between 0.2 and 3.5 km from the source, scheme
2 assigns, polar grid points to the nearest BGED centroid. The centroid population is then
apportioned among all polar grid points assigned to that centroid, based on the area of the polar
sector associated with the grid point (U.S. EPA, 1986b). Here a polar sector encompasses an
area delineated by two neighboring concentric rings and two direction radials that are evenly
spaced between the primary (modeled) direction radials. This method of apportioning produces
the same population density for all polar sectors assigned to a single centroid. Beyond 3.5 km,
scheme 1 is used and grid point concentrations are interpolated to the BGED centroids using a
log-linear interpolation scheme.
NEM. In the neighborhood-type version of NEM, hourly concentration levels within the
various microenvironments are determined based on monitored air quality data representative of
the study area (U.S. EPA, 1983). Activity patterns representative of the population in the study
area are used to define the rnicroenvironment inhabited by, and activity level (high, medium, or
low) associated with the cohort population of interest on an hourly basis. Exposure of the cohort
population is determined by matching cohort individuals with the concentration appropriate to
the microenvironment and neighborhood type in which they reside.
Personal Air Quality Model. Another example of a technique using activity pattern data
and the microenvironment concept is given by the Personal Air Quality Model (Hayes, 1989).
This model, applicable to any outdoor-generated air pollutant, uses outdoor hourly concentration
data (either monitored or modeled) for the pollutant of interest. The model calculates the hourly
4-55
-------
air concentration for five unique indoor and outdoor microenvironment types: office, home, inside
a vehicle, outdoors near a roadway, and other outdoors. These five microenvironments are
further subdivided into ten more specific microenvironments:
office with typical heating, ventilation and air conditioning;
office with energy-efficient heating, ventilation and air conditioning;
school;
home with closed windows;
• energy-efficient home with closed windows;
• home with open windows;
• home with air conditioning operating;
• inside a transportation vehicle;
• outdoors, near a roadway; and
• outdoors, other.
N-
The activity pattern data base established for NEM was used in the Personal Air Quality
Model to assign up to 56 population groups, differentiated by age/occupation and activity level,
to one of the 10 microenvironments during each hour of the day. Varying degrees of mobility
are accounted for by operating the model in one of four modes: (1) personal monitor mode--
hourly outdoor concentration data input to the model is assumed to be identical to that which
would be measured outside of the population group member's location during the day, (2) sphere-
of-influence—outdoor concentration data are assumed to be representative of a geographical area
large enough that population members do not leave its sphere of influence, (3) home/work mode--
different outdoor concentration data are specified for home and work periods, and (4) stationary-
source mode—home or work outdoor concentrations are assumed to be zero for time periods spent
outside the sphere of influence of the modeled stationary source. Model output includes hourly
personal concentrations to which population group members are exposed and ratios of personai-
to-outdoor air quality for different averaging periods of interest.
Food Chain Exposure. In performing a food chain exposure analysis as described in
Section 4.5.2.3, site-specific data should be used whenever possible. It is rarely feasible to gather
detailed site-specific data for all variables; however, some data are readily available from local
government offices. The agricultural extension agent may be able to supply information on crops
grown locally. The Fish and Wildlife Service (Federal or local) may be able to supply data on
fish consumption. Local health departments may also have useful information. Information
characterizing the population in terms of number of people, location, age, and activity could be
4-56
-------
gained through various population characterization methods described in Section 4.4. Even if
quantitative data are not available, local experts may have useful information that can be used
to construct plausible exposure scenarios. Default values for many variables can be found in the
Exposure Factors Handbook (U.S. EPA, 1989).
4.5.3 Selection Criteria for Exposure Estimation
As with all components of an exposure assessment, method selection depends on the
scope of the assessment. If maximum individual exposure estimates are desired (i.e., exposure
of a hypothetical individual to the peak concentration), then it would not be necessary to select
a method that incorporates detailed population characteristics; instead, established air quality
modeling or monitoring techniques could be used to obtain the exposure estimate. A study
focused partially or entirely on calculating aggregate exposure would require knowledge of the
population distribution and a method to integrate that distrifiution with the spatial distribution of
pollutant concentration. Cumulative or individual exposure estimates may be refined by
incorporating additional population characteristics, such as age, sex, and mobility. Levels of
exposure to outdoor-generated pollutants may be modified by taking into account the amount of
time people spend in various microenvironments. Modeling of ingestion pathway exposures in
addition to inhalation exposures adds further complexity to exposure analysis. Table 4-9
indicates characteristics of HEM-II that pertain to criteria for selection.
4.6 MONITORING TECHNIQUES
4.6.1 Overview
The exposure assessor is likely to encounter several types of monitoring and measurement
techniques. One type is outdoor fixed-location monitoring used to identify general levels of
concentrations and trends in concentrations. This method can evaluate the data from fixed
networks for relevance to an exposure assessment, and to document baseline values in the
environment. This method, referred to as ambient air monitoring, evaluates the overall quality
of outdoor air in a relatively large area. As the distance from the point of contact increases, the
certainty of the data decreases, and the assessor is responsible for showing the relevance of the
data (U.S. EPA, 1992) .
Microenvironmental measurement, another type of concentration measurement, defines
specific zones (and occasionally time periods) in which the concentration in the medium is
4-57
-------
LU
Q
O
LU
CC
§
a.
x
LU
I
LU
I
cr
O
LL
CC
UJ
o
z
o
o
LU
UJ
CO
cp
•*r
LU
m
LU
I
LU
Q
O
LU
CC
o
CL
X
LU
O
JS
CD
"5
•g
(0
1
CO
(O
.Q
|
E
o
O
CD
Q.
O
O
i
CD
0-
CD
OJ
-a
o
•£. co
1
>• ®
c 3 «
— O
§ 8 2 i
S 8 « 1
O co .£ co
CD
0)
•c .
CO CD
52
o S
_ CD
3 Q.
t|
S =
§S.
1 S
1
.
«. 0 :=.
.1 «
3
cr
£
CD
CD co
CO
E
'c
_CD
3
§
03
LU
CO
D.
CD
2
a
(0
0
u>
"8
2
3
in
Ex
(0
3
X
0)
CO
o
X
LU
LU
a
o
i
LU
iS
CC
CD
Q.
O
O
-------
thought to be relatively homogenous, and attempts to characterize the concentration in that zone.
Examples would be an automobile in heavy traffic, or a kitchen while cooking.
Personal monitoring is the concentration measurement that provides the closest link to
actual point of contact. Direct exposure measurements are collected by a person wearing the
sampling device. The collection medium is positioned in or as near has possible to the breathing
zone for the most accurate results. For a variety of reasons, personal monitoring is not always
feasible (U.S. EPA, 1992).
Biological monitoring can only be used directly for exposure assessment where the
relationship between the result and the absorbed dose is clear. Frequently, only inferences about
the exposure can be made. In this method, samples of biological fluids or tissues are examined
for a biomarker of the exposure that has or continues to occur. The concentration of the
x-
biomarker is used to estimate the dose received.
Because this document focuses on risk assessment from stationary sources, the emphasis
in this section is on ambient air monitoring techniques. Ambient air monitoring has the
advantage over more direct biological monitoring methods for exposure prediction and allowance
for preventive measures, because exposure can be more directly correlated with a specific source.
It is also more economically feasible than direct exposure measurement because the results may
be applied to a larger population.
4.6.2 Technical Background and Methods
4.6.2.1 Ambient Air Monitoring. Ambient air monitoring results are used in conjunction
with data on sources, environmental processes, target organisms, and activity patterns to estimate
exposures. In addition to use in risk assessment models, the data are used to validate exposures
based on air dispersion modeling techniques. Several new methods are now available to measure
specific toxic air pollutants. One example is the analysis of nonmethane organic compounds
(NMOCs) collected by evacuated canisters to differentiate and quantify the species. A
comprehensive reference is the EPA's Compendium of Methods for the Determination of Toxic
Organic Compounds in Ambient Air. These documents provide standardized measurement
techniques that include liquid impingers, passivated steel canisters, and various adsorbent,
cryogenic, and foam trapping technology. Some of the techniques combine methodologies to
permit determination of multiple pollutants; these and others may also have real-time analysis
capabilities (U.S. EPA. 1990c).
4-59
-------
The pollutant to be evaluated determines the type of trapping medium to be used. A
sample of the air can be collected and sent to a laboratory for direct analysis, or the air can be
directly drawn through the analytical instrument in the field. Field analysis is generally more
desirable because it reduces the potential for error associated, for example, with desorbing the
compounds from the sampling medium. Some compounds are not easily desorbed, resulting in
under valuing; others undergo chemical reaction causing underestimation or interference with the
analysis of other compounds on the sorbent.
A pump is frequently used to draw ambient air through the trapping medium. Other
methods use an evacuated canister to provide the negative pressure required to draw the sample.
Both pumps and evacuated canisters have flow meters and flow controllers and are used over a
specified sampling time to determine the volume of air sampled.
>••
Remote optical sensing, also known as "long path spectroscopy," is an emerging technique
for measuring air toxic compounds. Long path monitoring is described in the document
Application of Long Path Monitors to Superfund Sites - Draft Report. An overview of the
current remote sensing technology is given in the Air/Superfund National Technical Guidance
Study Series Volume II (U.S. EPA, 1989a).
A program to collect ambient monitoring data for exposure assessments will involve
definition of a monitoring network. General principles for network design relate to the objectives
for monitoring, the siting of monitoring stations and the network density and spacing, all of
which are interrelated. General requirements and considerations for ambient monitoring of
criteria pollutants are documented in the EPA publication Ambient Monitoring Guidelines for
Prevention of Significant Deterioration (PSD) (U.S. EPA, 1987a).
Monitoring networks may be designed to characterize the ambient concentrations resulting
from the emissions of a particular source, or may be designed to characterize the overall
background concentration in an area. The purpose of the investigation determines the optimum
network. Practical considerations, such as site accessibility, availability of electric power, and
security must also be considered. Monitoring networks usually require meteorological information
to account for the influence of emission sources outside the network.
The objectives of the exposure assessment must be clearly defined before any network
design is initiated. Because ambient concentrations can vary widely within a study area, the
location of monitors is critical in properly characterizing exposures. Sampling to isolate a
4-60
-------
particular facility's contribution to the ambient composition of air toxics will require simultaneous
sampling of meteorological variables (notably wind direction) to determine when the air sampled
by the monitor is representative (i.e., coming from the direction of the facility).
The number of stations required to obtain representative characterization will depend on
the local meteorology and terrain and the size of the area to be characterized, as well as other
factors. Table 4-10 summarizes the capabilities and selection criteria for ambient air monitoring
techniques.
4.6.2.2 Personal Exposure Monitoring. Personal exposure to toxic pollutants may be
determined through direct measurement techniques. In direct measurement, chemical
concentrations contacting a person's body are measured using split samples of the air the person
breathes, the food and water the person consumes, and by using patch or other techniques to
X-
estimate dermal exposure. These concentrations are measured as a function of time to obtain an
individual exposure profile. A set of individual profiles can be statistically aggregated to make
inferences about the exposure profiles of the population as a whole, provided the individuals
sampled have a known relationship to the entire population (U. S. EPA, 1988).
Several methods are available for personal exposure monitoring, each developed to
determine exposures to a single chemical or to compounds with similar chemical properties. The
most common reference is the NIOSH Analytical Methods Manual that contains procedures for
occupational exposure monitoring (NIOSH, 1985). As in ambient air monitoring, the trapping
medium is selected for the contammant(s) of interest. Equipment such as pumps and canisters
that provides air flow across the medium is not limited to specific media or pollutants.
Some personal exposure sampling devices rely on passive diffusion and movement of the
person wearing the device for the contaminant to contact the medium. These are known as
passive dosimeters.
Area sampling devices are not strictly direct exposure measuring instruments because they
are not worn. However, because these characterize relatively small areas, they can be used to
provide personal exposure information. They are most frequently used where people cannot or
prefer not to wear the personal sampling devices.
Personal exposure through ingestion and dermal contact can also be determined. Split
samples of food and drinking water may be taken simultaneously as an individual is ingesting
them to characterize individual or population exposure through the ingestion route. An example
4-61
-------
CO
UJ
m
o
E
o
o
UJ
m
CC
2
CO
cr
LU
9
CO
Q
O
o
5
UJ
UJ
CO
Q
u.
O
tr
co
UJ
—i
m
Monitoring
JH
.a
_ra
'5
co
£
co
UJ
-n
o
"CB
CD «,
2*8 I
5 2
•
"
c g
•Bs
CD TJ
^ ®
8-g
l
2 c
co 3
0
C/)
CDC?-0
IIs
.M.§ 2
J3 (A CD
°,>^|
.£ c c5
e 2* °
Q £ >
E "- lc
l||
= «
TJ "5 J=
CD O ••
52
5 CD x
!•!
or
ue
s
for
, the
at
la's
•S.2 fe
Hi
CD o -a
o o ^s
§ "c5 ]°
^•§,^
*-§ £
•= 2 to
1 8 S
°- « 3
^ ^ 5,
.SCO
a.
_f
.5
m c -c
8-i I
_j .s c
"y «=
§.°
f o
.11
forni
CO
75
o
.2 co
M
CU O
CC <
"8
1
O
O
.2
03
o3
CC
4-63
-------
of the use of split samples is given in the Total Exposure Assessment Methodology (TEAM)
study (U.S. EPA, 1987c). Skin patch samples characterize the dermal exposure of an individual
or population. An example of the use of skin patch samples is given in the EPA document
Pesticide Assessment Guidelines for Applicator Exposure Monitoring-Subdivision U (U.S. EPA,
1987b).
Table 4-11 summarizes the methods and selection criteria for personal exposure
monitoring. The advantage of this type of monitoring is the evaluation of individual exposures;
a significant limitation is its inability to identify the source.
4.6.3 Select Applications of Ambient Air and Personal Exposure Monitoring
This section describes several programs that use ambient air monitoring techniques. An
example of a program that used personal exposure monitoring, drinking water monitoring and
•\.
breath analysis, is also provided.
4.6.3.1 The NMQC Monitoring Program. The nonmethane organic compound (NMOC)
monitoring program was initiated in 1984 to assist states with areas that are out of compliance
with the National Ambient Air Quality Standard for ozone. The measurements are used in
revising their ozone control strategies. Three-hour integrated ambient air samples are collected
in SUMMA® polished stainless steel canisters and analyzed by a cryogenic preconcentration,
direct flame ionization detection (PDFID) method.
4.6.3.2 The Urban Air Toxics Monitoring Program. A state assistance program was
established in 1987 through which States could evaluate their urban air toxics problems. Ambient
air samples were taken and analyzed for specific air toxic compounds, primarily aromatics and
halocarbons. Since 1987, the method has been refined to determine concentrations of
38 compounds (U.S. EPA, I990a). Under the Urban Air Toxics Monitoring Program (UATMP),
three types of samples are collected: 24-hour ambient air samples every 12 days through the year
for air toxic compound analysis, cartridges for carbonyis determination, and high-volume filter
samples for metals and benzol a)pyrene concentrations.
The objective of the program is to provide screening data that states can use to set
priorities for their air toxics programs (U.S. EPA, 1987a). Recently, use of the data has been
used to estimate excess cancer risks (U.S. EPA, 1990b).
4-63
-------
CO
UJ
o
UJ
CD
•z
cc
o
o
UJ
cc
CO
o
Q.
X
UJ
o
CO
cc
UJ
Q.
CC
2
CO
cc
UJ
Q
CO
o
o
z
o
i—
o
UJ
UJ
CO
LL
o
>
oc
CO
LU
m
O)
Monitorir
£
3
in
8.
X
UJ
_
2
0
(O
5
0.
Q
O
I
(_
UJ
2
CD
C
CD
M
S.
CD
75
•g£
frj?
nsidered representative of others' exposures, the person sar
It may also be located in the vicinity where the activity is tak
o
U _j
o 2
CD —
m'S
"3 03
CD "5
•S a-
= a.
.2
To
g
|
Q
**
1
_£,
2
CD
CD
O)
en
'or several air toxics; however, the number of devices worn i
0
J3
JO
to
CO
£
co
v>
•**
^j
2
CO
5
"o
Q.
•s
1
1
c
S
•c
i
CD
j?
•a
•i
lative person requires experienced judgment.
ented or purchased and requires a skilled technician
ive and should be performed by a laboratory, preferably cert
with the method.
i labor-intensive compared to area (fixed location) sampling.
^~ W ""^ ^^
il) 33 C £* "Jfl
S f | S 5 s
0. 3 CD ~ .2 0>
5|S18-
o> CD » J> ">. O.
•8 '!• a i a §
co UJ < 51< co
I
CD
O) CO
.C ^rf
n5
CD CD g>
c is 1"
2 CD ,5
~ w "5_
^ Si
posure assessment is determined in part by the cooperation
> applied to others if the person sampled had been selected
ie time period sampled and exposure during the activities ss
activities may produce significantly different results.
x s — —
CD -a _ c
f til
° n °-~°
& ° 2.£
U (Q (Q ^
> "b "b "2
CD g CD CD O
..tX..i.-5,.-=
—
c
CD E
1.2
O 3
03 0)
oc cc
O
"cO
^
Zl
^
'i
CD
'C
CD
>
"o
Q
c
J3
•c
CD
C
cf-S
.2 Q-
?2
l§
ca n
a.
•73
1
o
O
§
75
»_
2.
~
CD
^
1
•o
C
a
.2
o
§
1
OSH Manual of Analytical Methods. U. S. Department of Hi
sed 1985.
~7
^— > I
CD
vi CC
O
M
c'si
Q..S
P u
5 -i=
CO O
U)
CD
i.
^
"55
CC
4-64
-------
Both the Urban Air Toxics and NMOC programs have been used in approximately 25
States and the District of Columbia, involving more than 50 urban centers. These programs are
managed through the Technical Support Division of the EPA OAQPS.
Increasing attention is being given to the concept and application of continuous or semi-
continuous NMOC and speciated organic compound (air toxic) sampling. Continuous sampling
allows identification of short-term peak concentrations that would be obscured in longer-term
(i.e., 24 hour) integrated concentrations.
4.6.3.3 The Integrated Air Cancer Project. The Integrated Air Cancer Project was
conceived in 1984. The goals of the project are to: (1) identify major carcinogenic chemicals
emitted or arising from atmospheric transformation of chemicals emitted, (2) identify the sources
of these chemicals, and (3) improve the methodology and data base for assessing human exposure
and comparative risk to airborne carcinogens. The goals will be achieved through conducting
field studies in airsheds of increasing complexity. The early studies developed methods for
identifying and quantifying mutagens and carcinogens from residential wood combustion and
motor vehicles. Field studies in 1987 furthered these developments but also characterized human
exposures to complex organics in ambient and residential environments. Other accomplishments
included characterizing the mutagenicity of emissions that had undergone atmospheric
transformations and successfully combining bioassay and receptor modeling technologies to
apportion the mutagenicity and mass paniculate organic matter (U. S. EPA, 1989b).
4.6.3.4 The Total Exposure Assessment Methodology (TEAM) Study. The goals of the
TEAM study were to develop methods to determine total individual exposure (exposures through
air, food, and water) and the resulting body burden, and then to apply these methods to estimate
exposures and body burdens of urban populations in several U.S. cities. The study design
required personal exposure and ambient air monitoring, drinking water sampling, and breath
analysis. Individuals who were sampled maintained a diary of their activities and potential
sources of exposure. The TEAM study began in 1979 and concluded in 1985 (U.S. EPA, 1987c).
The outcome of the study showed chat the Tenax personal monitor is appropriate for this
type of sampling and that exhaled breath is a sensitive method to determine body burden.
Furthermore, indoor exposures appeared to significantly contribute to exposures because breath
samples correlated well with personal exposures but not with the ambient samples. Finally, a
number of sources of exposure were identified (Wallace, 1986).
4-65
-------
4.7 REFERENCES
4.7.1 References for Section 4.2, Emission Characterization
Bellin, J.S. and D.G. Barnes, 1986. Interim Procedures for Estimating Risks Associated with
Exposures to Mixtures of Chlorinated Dibenzo-p-Dioxins and -Dibenzofurans (CDDs and CDFs),
Part I. U.S. Environmental Protection Agency. Risk Assessment Forum.
Chemical Manufacturers Association (CMA), 1987. Guidance for Estimating Fugitive Emissions.
Washington, DC.
National Research Council, 1981. Assembly of Life Sciences. Committee on Aldehydes Board
on Toxicology and Environmental Health Hazards. Formaldehyde and Other Aldehydes.
Washington, DC.
Radian Corporation, Research Triangle Park, 1990. Measurement Protocol for Air Toxics, Draft
Report. Prepared for U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Pollutant Characterization Section.
Russell, A.G., 1988. Mathematical Modeling of the Effect of Emission Sources on Atmospheric
Pollutant Concentrations. In: Air Pollution, the Automobile, and Public Health, pp. 161-205,
Health Effects Institute, National Academy Press, Washington, DC.
U.S. Environmental Protection Agency, 1986. Air Toxics Emissions From Motor Vehicles.
Office of Mobile Sources, Emissions Control Division, Ann Arbor, MI. EPA AA-TSS-PA-86-5.
U.S. Environmental Protection Agency, 1987a. Criteria Pollutant Emission Factors for the 1985
NAPAP Emissions Inventory. Office of Research and Development, Air and Energy Engineering
Research Laboratory, Research Triangle Park, NC. EPA 600/7-87-015.
U.S. Environmental Protection Agency, 1987b. Estimating Releases and Waste Treatment
Efficiencies for the Toxic Chemical Release Inventory Form. Office of Pesticides and Toxic
Substances, Washington, DC. EPA 560/4-88-002.
U.S. Environmental Protection Agency, 1987c. Toxic Release Inventory System. Office of
Pesticides and Toxic Substances, Washington, DC.
U.S. Environmental Protection Agency, 1988. Protocols for Generating Unit-Specific Emission
Estimates for Equipment Leaks of VOC and VHAP. Office of Air Quality Planning and
Standards, Research Triangle Park, NC.
EPA 450/3-88-010.
4-66
-------
U.S. Environmental Protection Agency, 1989a. Toxic Air Pollutant/Source Crosswalk - A
Screening Tool -For Locating Possible Sources Emitting Toxic Air Pollutants, Second Edition.
Office of Air Quality Planning and Standards, Research Triangle Park, NC. EPA 450/2-89-017.
NTIS PB 90-170002.
U.S. Environmental Protection Agency, 1989b. Compilation and Speciation of National Emission
Factors for Consumer/Commercial Solvent Use. Office of Air Quality Planning and Standards,
Research Triangle Park, NC. EPA 450/2-89-008.
U.S. Environmental Protection Agency, 1989c. Air Emissions Species Manual - Volume I,
Volatile Organic Compound (VOC) Species Profiles and Volume II, Particulate Matter (PM)
Species Profiles - Second Edition. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA 450/2-90-OOla (VOC) and EPA 450/2-90-OOlb (PM).
U.S. Environmental Protection Agency, 1989d. Locating and Estimating Air Toxic Emissions
from Coal and Oil Combustion Sources. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA 450/2-89-001.
U.S. Environmental Protection Agency, 1989e. Locating and Estimating Air Toxic Emissions
from Sources of Chromium. Office of Air Quality Planning and Standards, Research Triangle
Park, NC. EPA 450/2-89-002.
U.S. Environmental Protection Agency, 1989f. Locating and Estimating Air Toxic Emissions
from Municipal Waste Combustors. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA 450/2-89-006.
U.S. Environmental Protection Agency, 1990a. NATICH Data Base Report on State, Local, and
EPA Air Toxics Activities. Office of Air Quality Planning and Standards, Pollutant Assessment
Branch, Research Triangle Park, NC. EPA 450/3-90-012.
U.S. Environmental Protection Agency, 1990b. NATICH: Bibliography of Selected Reports and
Federal Register Notices Related to Air Toxics. Office of Air Quality Planning and Standards,
Pollutant Assessment Branch. Research Triangle Park. NC. EPA 450/3-90-014.
U.S. Environmental Protection Agency, 1990c. NATICH: Ongoing Research and Regulatory
Development Projects. Office of Air Quality Planning and Standards, Pollutant Assessment
Branch, Research Triangle Park, NC. EPA 450/3-90-013.
U.S. Environmental Protection Agency, 1990d. NATICH: Bibliography of Selected Reports and
Federal Register Notices Related to Air Toxics, Index 1990. Office of Air Quality Planning and
Standards, Pollutant Assessment Branch, Research Triangle Park, NC. EPA 450/3-90-014a.
U.S. Environmental Protection Agency, 1990e. Toxic Air Pollutant Emission Factors for Selected
Air Toxic Compounds and Sources - Second Edition. Office of Air Quality Planning and
Standards, Noncriteria Pollutant Programs Branch. Research Triangle Park, NC. EPA 450/2-90-
011.
4-67
-------
U.S. Environmental Protection Agency, 1990f. Crosswalk Air Toxic Emission Factor (XATEF)
Data Base Management System. Office of Air Quality Planning and Standards, Noncriteria
Pollutant Programs Branch, Research Triangle Park, NC.
U.S. Environmental Protection Agency, 1990h. Speciation Data System. Office of Air Quality
Planning and Standards, Research Triangle Park, NC.
U.S. Environmental Protection Agency, 1990g. AIRS Facility Subsystem Source Classification
Codes and Emission Factor Listing for Criteria Air Pollutants. Office of Air Quality Planning
and Standards, Research Triangle Park, NC. EPA 450/4-90-003.
4.7.2 References for Section 4.3, Fate and Transport Analysis
Baes, C. F., R. D. Sharp, A. L. Sjoreen, and R. W. Shor. 1984. A Review and Analysis of
Parameters for Assessing Transport of Environmentally Released Radionuclides through
Agriculture. Prepared for the U. S. Department of Energy under Contract No. DE-AC05-
840R21400.
Clark, L.D., 1987. Evaluation of Available Multi-Media Exposure/Risk Models. Prepared for
the U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
Research Triangle Park, NC. by Alliance Technologies Corporation, Chapel Hill, NC.
Dana, M.T., R.N. Lee and J.M. Hales, 1984. Hazardous Air Pollutants: Wet Removal Rates and
Mechanisms (Final Report). Battelle Pacific Northwest Labs, Richland, WA. EPA Publication
No. 600/3-84-113.
Hanna, S.R., G.A. Briggs and R.P. Hosker, Jr., 1982. Urban Diffusion Models. In: Handbook
on Atmospheric Diffusion, pp. 57-66, U.S. Department of Energy, Technical Information Center,
DOE/TIC-11223.
Hanna, S.R., 1985. Air Quality Modeling over Short Distances. In: Handbook of Applied
Meteorology, (D.D. Houghton, ed.), pp. 712-743, John Wiley and Sons, New York. NY.
McKone, T.E. and D.W. Layton, 1986. Screening the Potential Risks of Toxic Substances Using
a Multimedia Compartment Model: Estimation of Human Exposure, Regulatory Toxicology and
Pharmacology, 6, 359-380.
McKone, T.E., 1990. Personal communication between S. Templeman, Radian Corporation and
Tom McKone, Risk Assessment and Environmental Policy Division of Lawrence Livermore
Laboratory, September 1990.
Russell, A.G., 1988. Mathematical Modeling of the Effect of Emission Sources on Atmospheric
Pollutant Concentrations. In: Air Pollution, the Automobile, and Public Health, pp. 161-205,
Health Effects Institute, National Academy Press, Washington, DC.
4-68
-------
Sehmel, G.A., R.N. Lee and T.W. Horst, 1984. Hazardous Air Pollutants: Dry-Deposition
Phenomena. Battelle Pacific Northwest Labs, Richland, WA. EPA Publication No. 600/3-84-
114.
Travis, C. C. and A. D. Arms. 1988. Bioconcentration of Organics in Beef, Milk, and
Vegetation. Environmental Science Technol. 22(3):271-274.
U.S. Environmental Protection Agency, 1982. PTPLU - A Single Source Gaussian Dispersion
Algorithm. Office of Research and Development, Environmental Sciences Research Laboratory,
Research Triangle Park, NC. EPA 600/8-82-014.
U.S. Environmental Protection Agency, 1986. Guideline on Air Quality Models (Revised).
Office of Air Quality Planning and Standards, Research Triangle Park, NC. EPA 450/2-78-027R.
U.S. Environmental Protection Agency, 1987a. Supplement A to the Guideline on Air Quality
Models (Revised). Office of Air Quality Planning and Standards, Research Triangle Park, NC.
EPA 450/2-78-027R.
U.S. Environmental Protection Agency, 1987b. Industrial Source Complex (ISC) Dispersion
Model User's Guide - Second Edition (Revised), Volume I. Office of Air Quality Planning and
Standards, Research Triangle Park, NC. EPA 450/4-88-002a.
U.S. Environmental Protection Agency, 1988a. Appendix A: Strategies for Sources, Transport
and Fate Research. Report of the Subcommittee on Sources, Transport and Fate, Research
Strategies Committee. Office of the Administrator, Science Advisory Board, Washington, DC.
SAB-EC-88-040A.
U.S. Environmental Protection Agency, 1988b. Screening Procedures for Estimating the Air
Quality Impacts of Stationary Sources. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA 450/4-80-010.
U.S. Environmental Protection Agency, 1988c. A Workbook of Screening Techniques for
Assessing Impacts of Toxic Air Pollutants. Office of Air Quality Planning and Standards,
Technical Support Division, Research Triangle Park, NC. EPA 450/4-88-009.
U.S. Environmental Protection Agency, 1989. Graphical Exposure Modeling System i'GEMS)
User's Guide. Prepared for Office of Pesticides and Toxic Substances, Exposure Evaluation
Division by General Sciences Corporation, Laurel, MD.
U.S. Environmental Protection Agency, 1990a. Exposure Analysis Modeling System: User's
Guide for EXAMS II Version 2.94. Office of Research and Development, Environmental
Research Laboratory, Athens, GA. EPA 600/3-89-084.
U.S. Environmental Protection Agency, 1990b. Human Exposure Model-II User's Guide. Office
of Air Quality Planning and Standards, Pollutant Assessment Branch, Research Triangle Park,
NC.
4-69
-------
U.S. Environmental Protection Agency, 1990c. PCGEMS User's Guide Release 1.0. Prepared
for Office of Pesticides and Toxic Substances, Exposure Evaluation Division under Contract No.
68024281 by General Sciences Corporation, Laurel, MD.
U.S. Environmental Protection Agency, 1990d. Methodology for Assessing Health Risks
Associated with Indirect Exposure to Combustor Emissions. Office of Health and Environment
Assessment, Cincinnati, OH. EPA 600/6-90-003.
4.7.3 References for Section 4.4, Population Characterization
Anderson, J.R., E.E. Hardy, J.T. Roach and R.E. Witmer, 1976. A Land Use and Land Cover
Classification System for Use with Remote Sensor Data. U.S. Geological Survey Professional
Paper 964. U.S. Government Printing Office, Washington, DC.
Hayes, S.R., 1989. Estimating the Effect of Being Indoors on Total Personal Exposure to
Outdoor Air Pollution. The Journal of the Air and Waste Management Association 39:1453-
1461.
Johnson, T., L. Wijnberg, and R. Mersch, 1987. Draft. A Probabilistic Model for Simulating
Human Activity Patterns. Prepared by PEI Associates, Inc., Durham, NC for the U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC.
Robinson, J.P., 1988. Time-Diary Research and Human Exposure Assessment: Some
Methodological Considerations. Atmospheric Environment 22:2085-2092.
Sexton, K. and P.B. Ryan, 1988. Assessment of Human Exposure to Air Pollution: Methods,
Measurements, and Models. In: Air Pollution, the Automobile, and Public Health, pp. 207-238.
Health Effects Institute, National Academy Press, Washington, DC.
Thomas, J., D. Mage, L. Wallace and W. Ott, 1984. A Sensitivity Analysis of the Enhanced
Simulation of Human Air Pollution Exposure (SHAPE) Model. Prepared by General Software
Corporation, Landover, MD, for the U.S. Environmental Protection Agency, Environmental
Monitoring System Laboratory, Research Triangle Park, NC.
U.S. Department of Commerce, 1989. 1990 Census of Population and Housing. Tabulation and
Publication Program. Bureau of the Census, Washington, DC.
U.S. Environmental Protection Agency, 1983. The NAAQS Exposure Model (NEM) Applied to
Carbon Monoxide. Prepared by T. Johnson and R.A. Paul, PEDCo Environmental, Inc., Durham,
NC for the
U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality
Planning and Standards, Research Triangle Park, NC.
U.S. Environmental Protection Agency, 1985. Methods for Assessing Exposure to Chemical
Substances. Volume 4: Methods for Enumerating and Characterizing Populations Exposed to
4-70
-------
Chemical Substances. Office of Pesticides and Toxic Substances, Exposure Evaluation Division,
Washington DC. EPA 560/5-85-004.
U.S. Environmental Protection Agency, 1989. Proceedings of the Research Planning Conference
on Human Activity Patterns.
T.H. Starks, ed. Exposure Assessment Division, Environmental Monitoring Systems Laboratory,
Office of Research and Development. EPA 600/4-89-004.
U.S. Environmental Protection Agency, 1990. Human Exposure Model-II User's Guide. Office
of Air Quality Planning and Standards, Pollutant Assessment Branch, Research Triangle Park,
NC.
Young, J.M., R.A. Dulaney, S.E. Wright, and J.F. Scholl, 1985. A Computer Method for
Estimating Time-Weighted Urban Population Distributions, Volume I: Description of the Method.
Prepared by Lockheed Engineering and Management Services Company, Inc., Las Vegas, Nevada
for the U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory,
Office of Research and Development.
4.7.4 References for Section 4.5, Exposure Calculations
Hayes, S.R., 1989. Estimating the Effect of Being Indoors on Total Personal Exposure to
Outdoor Air Pollution. The Journal of the Air and Waste Management Association 39:1453-
1461.
McKone, T. E. 1989. Multiple Pathway Exposure Factors (PEF's) Associated with Multimedia
Pollutants, pp. 283-299 in D. T. Allen, Y. Cohen, and I. R. Kaplan (eds). Intermedia Pollutant
Transport: Modeling and Field Measurements, Plenum Press, NY. 298 pps.
McKone, T. E. and J. I. Daniels. 1990. Estimating Human Exposure through Multiple Pathways
from Air, Water, and Soil. Lawrence Livermore National Laboratory, Livermore, CA. UCRL-
JC-103995.
U.S. Environmental Protection Agency, 1983. The NAAQS Exposure Model (NEM) Applied to
Carbon Monoxide. Prepared by T. Johnson and R.A. Paul. PEDCo Environmental, Inc.. Durham,
NC for the U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air
Quality Planning and Standards. Research Triangle Park, NC.
U.S. Environmental Protection Agency, 1985. Development of Statistical Distributions or Ranges
of Standard Factors Used in Exposure Assessments. Prepared by GCA Corporation, Chapel Hill,
NC for U.S. Environmental Protection Agency, Office of Health and Environmental Assessment.
Office of Research and Development, Washington, DC. EPA 600/8-85-010.
U.S. Environmental Protection Agency, 1986a. Guideline on Air Quality Models (Revised).
Office 01 Air Quality Planning and Standards, Research Triangle Park, NC. EPA 450/2-78-027R.
4-71
-------
U.S. Environmental Protection Agency, 1986b. User's Manual for the Human Exposure Model
(HEM). Office of Air Quality Planning and Standards, Research Triangle Park, NC. EPA 450/5-
86-001.
U.S. Environmental Protection Agency, 1987. Supplement A to the Guideline on Air Quality
Models (Revised). Office of Air Quality Planning and Standards, Research Triangle Park, NC.
EPA 450/2-78-027R.
U.S. Environmental Protection Agency, 1988. Screening Procedures for Estimating the Air
Quality Impacts of Stationary Sources. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA 450/4-80-010.
U.S. Environmental Protection Agency, 1989a. Exposure Factors Handbook. Office of Health
and Environmental Assessment, Washington, DC. EPA 600/8-89/043.
U.S. Environmental Protection Agency, 1989b. Graphical Exposure Modeling System (GEMS)
User's Guide. Prepared for Office of Pesticides and Toxic Substances, Exposure Evaluation
Division by General Sciences Corporation, Laurel, MD.
U. S. Environmental Protection Agency, 1989c. Exposure Factors Handbook. Office of Health
and Environmental Assessment, Exposure Assessment Group, Washington, D. C. EPA/600/8-
89/043. NTIS PB90-106774.
U.S. Environmental Protection Agency, 1990. Human Exposure Model-II User's Guide. Office
of Air Quality Planning and Standards, Pollutant Assessment Branch, Research Triangle Park,
NC.
U. S. Environmental Protection Agency, 1990b. Methodology for Assessing Health Risks
Associated with Indirect Exposure to Combustor Emissions. Office of Health and Environment
Assessment, Cincinnati, OH. EPA 600/6-90-003.
4.7.5 References for Section 4.6, Monitoring Techniques
U.S. EPA/AREAL, 1990. Project Summary. Second Supplement to Compendium of Methods
for Determination of Toxic Organic Compounds in Ambient Air. EPA/600/54-89/018.
March 1990.
NIOSH, 1985. NIOSH Manual of Analytical Methods. U.S. Department of Health, Education,
and Welfare. Cincinnati, Ohio. Revised 1985.
53 FR 48832. Proposed Guidelines for Exposure-Related Measurements. Federal Register
Volume 53, No. 232. December 2, 1988. pp. 48830-48853.
51 FR 34042. Guidelines for Estimating Exposures. Federal Register Volume 51, No. 185.
September 24, 1986, pp. 34042-34054.
4-72
-------
EPA/OAQPS, 1990. 1989 Nonmethane Organic Compound Monitoring Program and Three-Hour
Air Toxics Monitoring Program. EPA-450/4-90-011. May 1990.
EPA/OAQPS, 1987. Urban Air Toxics Monitoring Program. EPA-450/4-87-022.
September 1987.
EPA/NPPB, 1990. Memorandum. Tom Lahre to internal distribution. Calculation of Cancer
Risks from 1988 UATMP Data. September 21, 1990.
EPA/AEERL, 1989. Integrated Air Cancer Project Status Report. (Joint Effort with
EPA/AREAL and EPA/HERL.) February 1989.
Wallace, L. A. Personal Exposures, Indoor and Outdoor Air Concentrations, and Exhaled Breath
Concentrations of Selected Volatile Organic Compounds Measured for 600 Residents of New
Jersey, North Dakota, North Carolina, and California. Toxicological and Environmental
Chemistry. Volume 12. 1986. pp. 215-236.
X"
U.S. Environmental Protection Agency, 1987a. Ambient Monitoring Guidelines for Prevention
of Significant Deterioration (PSD). Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA Publication No. 450/4-87-007.
U.S. Environmental Protection Agency, 1987b. The Total Exposure Assessment Methodology
(TEAM) Study. Volume I: Summary and Analysis. Office of Acid Deposition, Environmental
Monitoring and Quality Assurance, Washington, DC. EPA Publication No. 600/6-87-002a.
U.S. Environmental Protection Agency, 1987c. Pesticide Assessment Guidelines for Applicator
Exposure Monitoring-Subdivision U. Exposure Assessment Branch, Hazard Evaluation Division,
Office of Pesticide Programs, Washington, DC. EPA Publication No. 540/9-87-.127.
U.S. Environmental Protection Agency, 1989. Air/Superfund National Technical Guidance Study
Series, Volume II. EPA Publication No. 450/1-89-002a.
4-73
-------
-------
5.0 RISK CHARACTERIZATION
5.1 INTRODUCTION
The National Academy of Sciences (NAS) defines risk characterization as a description
of the nature and magnitude of human or nonhuman risk and the attending uncertainties. Risk
characterization is the final step in risk assessment and is primarily used to integrate the
information from the other three components. That is, it integrates the information that has been
identified during the hazard identification dose-response relationship analysis, and exposure
assessments. Risk characterization represents the link to risk management and provides decision
makers with information and data for use in developing, evaluating, and selecting risk
management strategies.
Unlike the other components of risk assessment, risk characterization did not have its own
official guidelines among the original 1986 EPA guidelines. The general content of risk
characterization was defined by the NAS, and to a limited degree in each EPA Risk Assessment
Guideline. However, much was left to the professional judgment of those involved in risk
assessment preparation. As a result, a high degree of variability in how risk characterization has
been practiced has developed, resulting in problems related to public perception of the reliability
of EPA's scientific assessments (EPA, 1992). Although a great deal of careful analysis and
scientific judgment goes into the development of EPA risk assessments, significant information
is often omitted as the results of the risk assessment are passed along in the decision-making
process. Often, when risk information is presented to the ultimate decision maker and to the
public, the results have been boiled down to a numerical risk value. Such "short-hand"
approaches do not fully convey the information needed to consider the health hazard nor heip the
health oriented decision maker understand the nature of the hazard.
To address these concerns, the Risk Assessment Council (RAC) of the EPA, composed
of risk assessors and risk managers from within the Agency, evaluated EPA risk assessment
practices. The RAC. after careful evaluation, recommended a guidance on risk assessment
focusing on the risk assessment-risk management interface, risk characterization, and exposure
and risk descriptors. This guidance was introduced to EPA managers for adoption as official
policy by the EPA Deputy Administrator in March of 1992. The major elements of the guidance
are summarized below in Section 5.2 followed bv specific discussions of risk characterization for
5-1
-------
carcinogens (Section 5.3) and noncarcinogens (Section 5.4), and the characterization of
uncertainty (Section 5.5)
5.2 RAC GUIDANCE ON RISK ASSESSMENT
The RAC reached several conclusions in their evaluation of EPA risk assessment practices
including:
• a full and complete presentation of risk is needed including a statement of
confidence about data and methods used to develop the assessment;
• a basis for greater consistency and comparability should be developed; and
• professional judgment plays a necessary and important role in the overall
statement of risk.
>*
Further, the RAC recommended that Agency-wide guidance would be useful.
5.2.1 Full Characterization of Risk
As practiced at EPA, the risk assessment process depends on many different kinds of
scientific data (e.g., exposure, toxicity, epidemiologic), all of which are used to "characterize"
the expected risk to human health or the environment. Informed use of reliable scientific data
from many different sources is a central feature of the risk assessment process.
Two elements are required for full characterization of risk. First, the characterization
must address qualitative and quantitative features of the assessment. That is, along with
quantitative estimates of risk, full risk characterization must clearly identify all assumptions, their
rationale and the effect of reasonable alternative assumptions on the conclusions and estimates.
Second, it must identify any important uncertainties in the assessment as part of a discussion on
confidence in the assessment. This statement on the confidence of the assessment must identify
all major uncertainties and comment on their influence on the assessment. The uncertainty
statement is important for several reasons.
Information from different sources carries different kinds of uncertainty and
knowledge of these differences is important when uncertainties are combined for
characterizing risk.
• Decisions must be made about expending resources to acquire additional
information to reduce the uncertainties.
• A ciear and explicit statement of the implications and limitations of a risk
assessment requires a clear and explicit statement of related uncertainties.
5-2
-------
• Uncertainty analysis gives the decision-maker a better understanding of the
implications and limitations of the assessments.
5.2.2 Consistency and Comparability.
The EPA's Science Advisory Board's (SAB) report, "Reducing Risk: Setting Priorities
and Strategies for Environmental Protection, recommended that EPA target efforts to achieve the
greatest risk reduction. Therefore, a common measure of risk is needed. EPA's newly revised
Exposure Assessment Guidelines provide standard descriptors of exposure and risk. Use of these
terms in all Agency risk assessments will promote consistency and comparability. Use of several
descriptors, rather than a single descriptor, will result in a more complete picture of risk that
corresponds to the range of different exposure conditions encountered by various populations
exposed to most environmental chemicals. EPA risk assessments will be expected to address or
provide descriptions of (1) individual risk to include the central tendency and high end portions
of the risk distribution, (2) important subgroups of the populations such as highly exposed or
highly susceptible groups or individuals, if known, and (3) population risk. Assessors may also
use additional descriptors of risk as needed when these add to the clarity of the presentation.
With the exception of assessments where particular descriptors clearly do not apply, some form
of these three types of descriptors should be routinely developed and presented for EPA risk
assessments.
5.2.2.1 Risk Descriptors. This section will briefly define the risk descriptors introduced
above. Subsequent sections related to carcinogens and noncarcinogens will be discuss these in
more detail in the appropriate sections (Sections 5.3 and 5.4).
Individual Risk. Individual risk descriptors are intended to estimate the risk borne by
individuals within a specified population or subpopulation. These descriptors are used to answer
questions concerning the affected population, the risk levels of various groups within the
population, and the average or maximum risk for individuals within the populations of interest.
Population Risk. Population risk descriptors are intended to estimate the extent of harm
for the population as a whole. This typically represents the sum total of individual risks within
the exposed population. Two important population risk descriptors should be estimated and
presented (EPA, 1992):
« the probabilistic number of health effect cases estimated in the population of
interest over a specified time period: and
5-3
-------
• the percentage of the population, or the number of persons, above a specified level
of risk or range of health benchmark levels.
Sensitive or Susceptible Subpopulations. Risk descriptors to describe sensitive
subpopulations are a subset of population risk. Unlike population risk, sensitive subpopulations
consist of a specific set of individuals who are particularly susceptible to adverse health effects
because of physiological (e.g., age, gender, pre-existing conditions), socioeconomic (e.g,
nutrition), other demographic variables, or significantly greater levels of exposure. Recent
guidance states: "Highly exposed (susceptible) subgroups can be identified, and where possible,
characterized and the magnitude of risk quantified. This descriptor is useful when there is (or
is expected to be) significantly different exposures or doses (or the sensitivity or susceptibility)
from that of the larger population. (In cases of susceptibility, it may be necessary to use a
different dose-response relationship)." (EPA, 1992)
Central Tendency Estimates of Risk. The central tendency risk descriptors are intended
to give a characterization of risk for the typical situation in which an individual is likely to be
exposed. Recent guidance states: "The risk descriptor addressing central tendency may be either
the arithmetic mean risk (average estimate) or the median risk (median estimate), either of which
should be clearly labeled. Where both the arithmetic mean and the median are available but they
differ substantially, it is helpful to present both.... Because of the skewness of typical exposure
profiles,... the median estimate is usually a valuable descriptor for this type of distribution, since
half the population will be above and half below this value."(EPA, 1992)
High-end Estimates of Risk. The "high end" risk descriptor is intended to estimate the
risk that is expected to occur in a small but definable segment of the population. The intent is
to "convey an estimate of risk in the upper range of the distribution, but to avoid estimates which
are beyond the true distribution. Conceptually, high end risk means risk above about the 90%
percentile of the population distribution, but not higher than the individual in the population wno
has the highest risk." (EPA, 1992)
5.2.3 Professional Judgment
The call for more extensive characterization of risk has obvious limits. For example, the
risk characterization includes only the most significant data and uncertainties from the assessment
(those that define and explain the main risk conclusions) so that decision-makers and the public
are not overwhelmed by valid but secondary information.
5-4
-------
The degree to which confidence and uncertainty are addressed depends largely on the
scope of the assessment and available resources. When special circumstances (e.g., lack of data,
extremely complex situations, resource limitations, statutory deadlines) preclude a full assessment,
such circumstances should be explained.
5.3 QUANTIFICATION OF CANCER RISKS
5.3.1 Overview
This section presents methods of developing and presenting numerical estimates of cancer
risks. The section on technical background and issues (5.3.2) describes the commonly used
measures for characterizing individual risks and aggregate (population) risks. A discussion of
the need for consistency when combining dose-response and exposure information is included.
•V-
Finally, scientific issues and methods for estimating cancer risks from multiple chemicals and
multiple exposure pathways are discussed. To aid in selecting cancer risk characterization
methods, Section 5.3.3 discusses the objectives, time, resources, and expertise required for each
method.
5.3.2 Technical Background and Methods
5.3.2.1 Cancer Unit Risk Estimates. Risks are generally expressed as either
individual risks or aggregate population risk. The distribution of individual risks within a given
population can also be presented. For air toxic emissions, individual or aggregate cancer risks
can be calculated by multiplying the exposure estimate by the unit risk estimate ^URE). The
URE usually represents an upper bound of the increased nsk of contracting cancer for an
individual exposed continuously for a lifetime (70 years) to a specific concentration ('e.g.. 1 ppm
or 1 ug/m3) of a pollutant in the air as was described in Section 3.3.4. The URE is based on the
assumption of iow-dose linearity. If a nonlinear low-dose-response extrapolation model were
used, the unit risk would differ at different dose levels, and the dose-response assessment output
could be expressed as a dose corresponding to a given levei of risk, analogous to the Risk
Specific Dose, rather than as a single URE.
The relationship of cancer UREs and weight-of-evidence classification because it also
influences the risk characterization and risk communication approach. Weight-of-Evidence
attempts to answer the question how likely an agent is to be a human carcinogen under some
circumstance of exposure while the URE attempts to estimate the possible impact upon an
5-5
-------
exposed population assuming the agent is a hazard to humans. In those cases where the weight-
of-evidence is below Group A, "known" human carcinogen, (i.e. Group B1/B2/C), the derviation
of URE's is predicated on a what if assumption, that is what might the impact be if the agent
really is a human carcinogen. The mere ability to do the calculation has not modified the
certainty about the weight-of-evidence.
The what if nature of the assumption suggests that UREs of agents that have weights-of-
evidence spanning a range of "A-B-C" need to be identified so that a decisionmaker appreciates
this range of qualitative uncertainty.
From a dose-response data perspective, the notion that human data is always more
accurate than animal data is not always clearly true. On the one hand, using human data negates
the need to use assumptions to estimate how animal does and responses extrapolate to humans.
>-
On the other hand, the accuracy of the human exposure can be highly uncertain. While in the
case of animal dose response data the ability to measure dose is very good, the issue of
extrapolation to humans remains a major uncertainty.
Each risk assessment has its own distinctive weight-of-evidence and URE relationship.
At a minimum its very important to carry weight-of-evidence designations along with the UREs
in some manner.
5.3.2.2 Expression of Cancer Risks. Maximum individual lifetime risk (MIR) is
commonly used to express individual risks. MIR is defined as the probability of contracting
cancer following exposure to a pollutant at the maximum modeled long-term ambient
concentration assuming a 70-year (lifetime) duration of exposure. Estimates of MIR are usually
expressed as a probability represented in scientific notation as a negative exponent of 10. A risk
of contracting cancer of i chance in 10,000 is written as IxlO"4. Table 5-1 provides a simplified
example that can be used to demonstrate the concepts of calculating MIR (and aggregate risk).
In this example the URE for the chemical of concern is 4x10° per ug/m3 ambient concentration
and the maximum modeled ambient concentration is 2 ug/mj. so the MIR is 8x10° or 8 in
100,000 [2 ug/m3 x 4xlO-5/(ug/m3)] (EPA/NATICH, 1987).
As stated in Section 3.3.4, uncertainties in the unit risk estimate include uncertainties m
the data from animal or human studies, in study selection, and in low-dose extrapolation models,
as well as uncertainties involved in the exposure assessment (including uncertainties in emission
estimates, dispersion modeling, and population characterization). All these uncertainties affect
5-6
-------
the estimate of MIR. Therefore, even a quantitative risk value such as an MIR is not a
measurement. It is an estimate based on many previous estimates. The risk characterization
should always include discussions of weight of evidence, assumptions, and all the uncertainties,
along with presentation of the MIR as an estimate.
Recent EPA Guidelines recommend that estimates of individual risk be presented for the
central tendency (mean or median) and high end portions of the risk distribution (e.g., 90%, 95%,
97.5%, 99%).
In addition to expressions of individual risk, a distribution of individual risks should be
presented as part of risk characterization. Risk distribution is an estimate of the number of
people exposed to various levels of risk. Table 5-1 provides an example of a risk distribution.
To produce a risk distribution, the exposure assessment must provide estimates of the population
exposed to various modeled concentration levels. If only a screening exposure assessment with
a single point estimate of maximum exposure is available, a risk distribution cannot be calculated.
Aggregate risk, also referred to as population risk, is usually an upper bound estimate that applies
to the entire population within the given area of analysis. The aggregate population risk is often
expressed as annual cancer incidence, which is the average number of excess cancer cases
expected annually in the exposed population. The simplified example in Table 5-2 shows the
calculation of population risk for a population exposed to three different modeled ambient
concentration levels. Each modeled ambient concentration level is multiplied by the number of
people exposed to that level and by the URE. This provides an estimate of risk for each group
after a 70-year exposure (assumed human lifespan). The risks are summed to provide the total
estimated excess cancer cases in the exposed population. This 70-year risk estimate can be
divided by 70 to calculate annual incidence in units of cancer cases per year (EPA/NATICH,
1987).
In reality, the population around an emission source would be exposed to a greater range of
concentration levels than shown in the example, and the aggregate risk would be calculated by
computer.
Risk distributions can be used to show the cancer incidence (cases per year) associated
with each risk level (see Tables 5-1 and 5-2). This type of presentation provides such
information as whether the aggregate risk is due mainly to a large population exposed to a low-
risk level or a small population exposed to a higher risk level. As with individual risk measures.
5-'
-------
there is significant uncertainty in aggregate population risk estimates due to uncertainties in the
URE and in the exposure assessment.
The preceding examples dealt mainly with inhalation risks from single pollutants for
which a URE was available. Individual and aggregate risks may be calculated for pathways other
than inhalation of a chemical in ambient air. When other pathways are considered, dose-response
assessment outputs are not in the form of an inhalation URE and exposure model outputs are not
in the form of long-term average ambient concentration in ppm or ug/m3. However, the general
risk characterization methodologies and ways of expressing risks as either individual or aggregate
risks are similar to those presented for the inhalation pathway. The following sections discuss
more complex risk characterization scenarios.
5.3.2.2 Consistency in Risk Calculation. Certain key assumptions are common to the exposure
assessment outputs and dose-response outputs. Before calculating individual or aggregate risk,
it is necessary to check the consistency and validity of key assumptions such as:
• the averaging period for exposure,
• the exposure route,
• absorption adjustments, and
• spatial consistency.
Averaging Period. In the case of cancer risks, the dose-response output (e.g., the URE)
is typically expressed as risk resulting from a lifetime exposure to a given dose or concentration.
Unless there is evidence to the contrary in a particular case, the cumulative exposure or dose
received over a lifetime, expressed as average daily exposure prorated over a 70-year lifetime,
is recommended as the appropriate measure of exposure to a carcinogen. (The assumption is
made that a high-dose over a short period of time is equivalent to a corresponding iow-dose
spread over a lifetime). (EPA, 1986)
Exposure Route. Exposure route consistency is also important. For example, if the
exposure assessment predicts an inhalation exposure pathway, then a URE for inhalation
(expressed as risk per ug/mj or ppm concentration in air) should be used and matched with the
average ambient concentration. In cases where the exposure assessment provides a concentration
in drinking water, a water unit risk ^expressed as risk per ,ug/L concentration in water; is
commonly used. IRIS contains both air and water UREs. The air URE assumes that since most
of a person's lifetime is spent as an adult from a physiologic perspective, an average adult person
5-8
-------
TABLE 5-1. EXAMPLE RISK DISTRIBUTION
Individual Risk Level
>1 x 10"3
1 x10'4to 1 x10'3
1 x10"5to 1 x10"4
1 x 10"6to 1 x 10"5
<1 x 10'6
Population
Exposed at Each
Level
0
10
1,000
80,000
500,000
Cumulative
Population At Risk
0
10
1,000
81 ,000
580,000
Incidence for Each
Risk Group
(Cases/year)
0
0.005
0.05
0.40
0.25
TOTAL CASES 0.71
5-9
-------
co
cc
or
LLJ
o
1
3
o
s
QL
<
04
in
LU
in
.r
CO
CM
,_
£
en
"en x"^~
(£ <" X
c. "o co
-2 en CU Cl)
o
5
O
"5
1 £
§ i
o C}
UU 0
cc
CO
o
0
1
in
o
^
X
CD
CO
I
C\J
T—
^
^-^;
in
ci
X
^.
ej
o"
m
b
•*••
X
T
1
§
O
o"
o
**"
in
b
^~
X
CNJ
CD
H
in
o
in
CM
£
73
en
CD
1^
CTJ
CA)
CC
-------
weighs 70 kg and inhales 20 m3 of air per day. The water unit risk assumes that the average
person weighs 70 kg and drinks 2L of water a day. Slope factors may be used instead of media
concentrations to express the dose-response relationship. These are expressed as risk per dose -
- mg of chemical per kg of body weight per day (risk per mg/kg-day). If slope factors are used,
the exposure or dose would be calculated in units of mg/kg-day. If dose-response information
is not available for an exposure pathway of concern (e.g., there is an oral slope factor but no air
inhalation URE), case-by-case evaluation and expert judgment can be used to determine whether
route-to-route extrapolation is legitimate. (See Section 3 for a discussion of route-to-route
extrapolation). When dose-response information is inadequate, a qualitative rather than
quantitative risk characterization may be preferred.
Absorption Adjustments. Slope factors are most commonly expressed as intake doses but
may be expressed as absorbed doses if there is good information on the relationship between
inhalation or ingestion and absorption. The exposure and dose-response estimate must both be
expressed as administered intakes, or both must be expressed as absorbed doses. Usually, both
are expressed as intakes. Dermal exposures are an exception since there are no slope factors for
dermal exposures in IRIS. In some cases, exposures via dermal routes can be calculated and
expressed as adsorbed doses, and compared with an oral toxicity value that has been adjusted so
it too is expressed as an absorbed dose (EPA/OSW, 1989). This would be inappropriate in the
case of carcinogens that cause skin cancer through direct action at the point of application.
Spatial Consistency. As described in Section 4.3, ambient rate-and transport models often
estimate concentration using a grid. Because the grid of points where concentration is estimated
often does not match the locations provided by the population data source, interpolation of
concentration or population is necessary to predict exposure and risk levels. If exposure
concentrations for different pathways are calculated at different locations, the situation becomes
even more compiicated. The location of concentration estimates for different media and
population estimates should be taken into account when estimating total risks from multiple
exposure routes.
5.3.2.3 Multiple Chemical Mixtures. The major issues related to risks from
chemical mixtures are discussed in detail in Section 6.2 but the methods are summarized briefly
below. The following equation estimates the incremental individual cancer risk for simultaneous
exposures to several carcinogens:
5-11
-------
Risky = Z Riskj (5-1)
where:
TO
T = the total cancer risk (expressed as a probability of contracting cancer over
a lifetime)
Rj = the risk estimate for the i substance.
In other words, the cancer risks predicted for individual chemicals can be added to estimate total
risk. This equation is an approximation of a more precise equation for combining risks. The
precise equation is consistent with the assumption of dose additivity and accounts for the joint
probabilities of the same individual developing cancer as a result of exposure to two or more
carcinogens (see EPA, 1986, 51). The differences between equation 5-1 and the precise equation
are negligible for total individual cancer risks less than 1 x 10-1 (or 0.1). (EPA/OSW, 1989)
The EPA guidelines for chemical mixtures (EPA, 1986 and EPA 1988) provide further detail on
mathematical models for multiple chemical risk estimation.
Two other approaches have recently been developed for chemical mixture risk assessment:
the comparative potency approach and the toxic equivalency approach. These approaches are
described in Section 3. If dose-response information is developed using one of these approaches,
then exposure assessment output should be in a format consistent with their use.
5.3.2.4 Sensitive Subpopulations. Certain groups within a population are more
sensitive to carcinogenic exposure than other groups". A quantitative characterization of
subpopulation risk requires both aose-response and exposure information for the subpopulation.
The hazard identification and dose-response assessment steps of a risk analysis may identify
sensitive subpopulations and develop dose-response information for them. Subpopulations can
be defined using age, race, sex, and other factors. If enough information is available, a
quantitative risk estimate for a subpopulation can be developed. If not, then any qualitative
information about subpopulations gathered during hazard identification should be summarized as
part of the risk characterization. In the absence of contrary data, it is assumed that susceptibility
to cancer is uniform throughout the population.
5.3.2.5 Multipathway Risks In some exposure scenarios, the same
individual or population may be simultaneously exposed to risks from multiple pathways. For
example, emissions from a stationary source may be directly inhaled, but they may also be
5-12
-------
deposited on crops or in water bodies and lead to public exposures through consumption of
contaminated food. Various pathways are described in Section 4. Total cancer risks can be
estimated by adding the cancer risks calculated for each pathway if the risks apply to the same
individuals or populations and are for the same time period (usually a lifetime average). This
addition is a simplification of a more complex equation (see Section 5.3.2.3) but is appropriate
for total individual cancer risks less than 1 x 10'1 (or 0.1) (EPA/OSW, 1989).
In estimating multipathway exposures, care should be taken to develop plausible exposure
scenarios. The location of the modeled maximum exposure for two different pathways may not
be the same. When conducting a detailed risk assessment it would be appropriate to examine
whether the same individual or subpopulation within a study area is likely to face the maximum
modeled exposure from more than one pathway. Exposure pathways may be added if they have
the potential to expose the same individuals, but not if different individuals or subpopulations
would be affected. A multipathway risk characterization should clearly document the rationale,
assumptions, and uncertainties involved in developing multipathway exposure and risk estimates.
5.3.2.6 Summary of Methods for Carcinogens
Basic Methods. Table 5-3 summarizes the various cancer risk characterization methods.
The first method of quantitative risk characterization is estimation of the maximum individual
lifetime risk (MIR). The MIR is typically calculated as the maximum modeled annual average
ambient concentration multiplied by the URE. The URE is used for single pollutant inhalation
exposure scenarios, but other measures of individual risk are also possible. The MIR usually
assumes exposure to the maximum modeled concentration, regardless of how likeiy it is that
someone would be continuously exposed at the predicted location of maximum annual average
concentration; but an alternative is to determine the maximum annual average concentration to
which an actual individual is likely to be exposed using site-specific population information.
Recent guidelines also emphasize the need to state the risk for the central tendency (i.e., mean
or median) as well as the high end of the risk distribution (e.g., 90 or 95 percent).
The second method of risk characterization is estimation of aggregate population risk.
This is typically expressed as the annual average excess cancer incidence for the exposed
population within the study area. A risk distribution can also be used to provide a more detailed
characterization of risk.
5-13
-------
o
o
I
oc
s
cc
o
cc
cr
UJ
o
1
LL.
O
>-
cc
in
LU
—I
CQ
Other Characteristics of Method |
•3
O-
o
3
-5
03
O
a.
c:
o
"co
3
o
CO
O
,
« i 1 i t s 1 1
"c5cgQ-icq'5gw»J
C ^ V *v* {n c^.
liif ill?
03
-§ .i5
-^
01 .§
_a3_ -o
o 'Z
m ^
~ 2
X g ^n
13^
o o 5
x" o 5
m o co
To — *
^ 03 x
^ O. v-
^> 0 J
T3 Ul '(5
.£ x 3
If
":o ""•
if
"3 "55
75 ~ ^
"8 s
.c -—
'as "5
^> —
«c
o £
^H c
'S x
0 "!
co 2
o .^
w- ^ ^, n*"~
to c r— c" Q}
? -8 I S S
'53
03
O3
If
11
Q. CO
C 13
'~ §
E ^o.
co -«
r- to
•1 §
-52 CD
8 "2
-^ "O —
111
"rt ^- -Jg
0) O -^
03 »_
-------
to
ID
_a
CQ
Other Characteristics of Method |
"3
Q.
5
1
-s
ca
o
co>-/gCL
Si o f S "8 S -i el S-
5 -9
"c ?
10 s ZLs
a: 2 cc .2
mm
*— C ,
• In developing exposure scenario,
consider whether same
individual(s) are likely to be
exposed to risk from each pathway
• Degree of conservatism depends
on exposure assumptions. These
should be described as part of
characterization.
~s
en
c
5™
CD
-^ ™ •£.
-I 1
CO O £
-o In 2.
'> •= »
~E -2 "™
•= co •=
-= en o-
3 g o
O O5 Q- xv
*~ CT> Q_
• CO CO
-1
« s
-== E 2
y 3 __
^3 ^ T3
CD C CU
— 5 Q.
CO = •«
^. __ .2 -2 >,
c S 13 js 2
« 1 8 i. 1
^ O3 — J O CO
• O. • CL O.
g
S
o
CD CO
5 >*
« OJ
0 -5
cu ^% !^ ^- ~*
^r . 03
"7^ (O "~ '^
^ ^ i -i e
o ~ := -a .2
O3 03 C T3 t^
as o _ « o
o §-— o «
•i= 9- ill o c
lll^§
1 1 S o
_3 Co ^ — •
,— . -^^ — J —
V3 J> , ^
co ja
CO
OJ
a CO
^ *~ ^.
S E o
•^ ^ ^
O c/) ^
• • CO
CA
^.
to
CC ;
j>^
1
to •
Q.
O
i n£
c^ |§
* 0 •
-id
LO
tr
T3
C
O3 — -
- OJ
^ -a
(D ••=-
«^ '-^
- «= S
ra o 2
ill
-^ o. —
^ 0 .«
-£ Q. -a
•
_
o •^-
'"& S3 _
"E 9- 8
cu JS -p
-o o 5
S g-5
C to ^r
o — o
"« <3 fe
o -^ 2
O CO O
— ™ ca
S CD
•o 3
_C Q
TJ E
CD O
cn 4i
•
cy^
2
UJ
0
1
5-15
-------
For risk assessments that deal with multiple chemical mixtures, the most common risk
characterization approach is to calculate the risks for each chemical compound and then add them
together. This approach is most valid when dose-response information on the mixture as a whole
is not available, when exposures to each chemical are low (within the linear range of the dose-
response curve), and when there are no known interactions among the compounds. If information
is available on the mixture of interest or a similar mixture, or if studies indicate synergistic or
antagonistic affects among compounds within the mixture, then this information should be
qualitatively or quantitatively considered in the risk characterization.
Risk assessments often deal with multipathway risks by calculating risks for each pathway
and then adding the risks together. A key consideration is the likelihood that the same
individuals will be exposed to risks from more than one pathway. There are also uncertainties
associated with development of a plausible exposure scenario.
All risk characterizations should include discussion of key assumptions and uncertainties
in the exposure assessment and dose-response assessment or URE. Because these affect the
estimates of individual risk, aggregate risk, and risk distribution, they should always be presented
as part of the risk characterization.
Computer Models. Some of the computer models discussed in Section 4 also include
cancer risk characterization elements. For example, the HEM-II output includes MIR, annual
incidence, and risk distribution. The dispersion modeling, population characterization, and
exposure assessment portions of the HEM-II and the model inputs are described in Section 4.
The risk characterization part of the HEM-II calculates MIR for each chemical and each
emission source by multiplying the maximum modeled ambient concentration by the URE'. The
HEM-II model can also calculate risks to individuals resulting from inhalation exposure to the
same chemical from more than one emission source (such as two plants located near each other).
The HEM-II also calculates aggregate risk (annual incidence) and risk distributions. As
discussed in Section 4, the population for each census data block group/enumeration district
(BGED) is matched to the ambient concentration predicted at the population centroid of the
BGED, and this exposure (population x. concentration) is multiplied by the URE to estimate risk.
These risk values are summed to give aggregate risks for the population within the study area.
An example of risk report output from HEM-II is shown in Table 5-4. A HEM-II Users Guide
is being developed to provide further information (EPA/OAQPS, 1990). Using HEM-II, risk
5-16
-------
CO
LU
O
o
o
cc
<
o
cc
o
LL
C/3
13
C73
LU
CC
LU
X
LL
O
>
CC
03
TT
in
LU
5
in
b
/laximum
jal Risk >
2
^
_c
E ir
3 _
E ™
X "O
^ ^
_c
0)
co
Z
CO
3
"o
a.
^
E
^
3
3
"o
Q.
»
-^
o
03
LL
O
Z
01
b
X
CM
*-
O
"c
3
"o
Cu
T-
<
O
CO
0)
IT)
b
X
r-
cvi
O
3
jO
3
"o
Q.
•r—
<
,_
•^«
CO
>
b
X
CO
•^:
Q
"c
_CO
"o
Q.
OJ
CD
in
OJ
CO
b
X
o
CO
O
3
_CO
"6
Q.
T-
CQ
in
OJ
o
10
b
X
CO
Q
3
_CO
"o
Q.
OJ
o
CO
^
03
Z
LU
CD
O
o
cc
o
z
o
z
cc
o
LL
03
03
LU
CC
LU
LL
O
cc
03
c
o
'2
Is
§2
0 CO
13
E
'x
CO
O
m~**
c E
3 "l!
"5
0.
OJ
O)
CO
OJ
CO n3
< o =L
E o
1°
X
CO
1 i
— CO
o Z
a.
d Q
CO _Q
3 s
o .3
a. •£•
o|
03 Q
tQ
to ~
LL
O
q
T~
b
a
2
-^
-^—
o
a.
CO
<
o
0
q
1~
co
b
3
3
"o
CL
co
<
O
q
•^
lj^
P^
b
O
3
—
^
-^—
0
a.
CO
CO
OJ
CO
0)
in
r-
cvi
"c
2
^
-^r
a.
03
CQ
tn
OJ
CO
0
o
c\i
00
c\i
X
c
2
3
"5
a.
r^
O
oo
-------
/ reports can be produced for a single source and a summary report can be produced for all the
sources considered in a national study.
If HEM-II is used for cancer risk characterization, the user should be familiar with the
assumptions inherent in the URE and exposure assessment procedures. As always, the risk
characterization should present a discussion of these assumptions and uncertainties along with
the numerical risk estimates.
The HEM-II model does not calculate total risks from multiple chemicals or risks from
indirect (non-inhalation) exposure pathways.
The GEMS model can also generate a risk report. The "lifetime risk/cases of cancer"
option under Exposure and Risk Estimation will generate risk distribution tables of the population
exposed and the excess lifetime cases of cancer associated with each risk level. (Risk levels are
given as orders of magnitude.) The lifetime risk is calculated from the average modeled ambient
concentration in each "sector segment" (defined by a wind direction and distance from the source)
using- the following equation:
Lifetime Risk (i) = conc(i) x q*/BW x IR + 0.001/mg/ug
where:
conc(i) = average concentration (jag/m3) from the ith sector segment
q* = potency slope factor (risk per mg/kg/day)
BW = body weight of aduit in kg, default 70
IR = daily inhalation rate (nrVday), default 20.
The default values for BW and IR are those typically used in developing an air URE from a
slope factor (see Section 3). The GEMS approach is similar to multiplying a concentration times
a URE. However, it produces an average concentration for the sector segment rather than a
maximum. The lifetime cancer cases in the popuiation(i) exposed to each risk level(i) is
computed in GEMS by multiplying the nsk(i) times the population(i) number of people in the
ith sector segment (General Sciences Corp., 1988).
Some other computer exposure models produce an exposure or dosage, but do not include
a calculation of risk within the model. The exposure output of these models (in units of ug/nr,
5-18
-------
ppm, or mg/kg/day) could be multiplied by an appropriate URE or slope factor to calculate
individual or aggregate risk.
5.3.3 Selection of Methods
Table 5-5 presents information to aid in selection of quantitative risk characterization
methods. The selected risk assessment method will depend on the amount of information
available, the level of detail needed, and a number of practical considerations including the time,
resources, and expertise available for the project. If there is insufficient information on the health
effects or emissions of a chemical, a quantitative risk characterization may not be possible. Even
if quantitative risk estimation is used, it may vary from a quick, rough screening study to a
detailed site-specific analysis. Section 7 presents some case studies that illustrate various levels
of analysis.
Many screening studies concentrate on MIR or a similar estimate of individual risk. The
results can be used to decide whether more detailed analyses would be useful. It is relatively
simple to calculate individual risk because only a URE and a "maximum" modeled concentration
are needed; a detailed exposure pattern is not necessary. However, more refined estimates of
individual risk are possible. For example, site-specific emission measurements and site-specific
population data could be used to develop more refined estimates of individual exposure and risk.
Methods for estimating aggregate risk and risk distribution are more time and resource-
intensive than those for estimating MIR. A computer model is usually needed because exposure
levels will vary greatly with direction and distance from a source, exposing the study population
to a wide range of risk levels. The ievei of expertise, time, and resources for an aggregate risk
estimate varies depending on the model used and the simplifying assumptions made.
An estimate of multiple chemical and/or multiple pathway risks, requires more resources
and expertise than a single chemical, single pathway risk assessment. The exposure and dose-
response assessment steps will take longer since information for multiple chemicals and pathways
must be generated. Furthermore, the risk characterization step will require careful consideration
of the scientific issues described in Section 5.3.2 to ensure that an appropriate methodology is
used. If the assumption of additivity is used, the calculations will be fairly simple. If the
available information indicates chemicals or pathways do not have additive effects, quantification
of risks will be more complex and may require expert judgement.
5-19
-------
CO
0
o
M
oc
Ul
o
:*;
co
cc
cc
LU
o
o
cc
2
QC
UJ
DC
O
o
LU
LU
CO
LL_
O
>-
co
LU
CQ
CD
.g-g
CD ™
••= i~
CO f«£
cc
Resources Needed
2
-a
2?
cr
CD
CC
CD
CD
a.
.=
>j ca
S •= !
^ a.
_O 0
_^* Q_
3 <
CO .22
~0 0
"ai £
CD ^>
-a
0
o
~=
2
3
5
•May use HEM-II computer model.
•Hand-calculation also possible if
ambient concentration estimate
and URE are available.
"o
CD
-§>
.32
J*£
« c
— CO
03 CO
rj co
-a co
1 2
jlj
S
CD
*— '
0
g1
CD
CD
O
CO
^-~
_i£
CO
CC
5
•3
>
JZ
|
.1
x
15
S
E
Z3
T3
Q3
0
>
°
-53
1-1
o To
— o
° 1
ca <2
Q. JC
—
CD
cn
-a
CD
C
03 "j;
r" CD
'S |
03 CD
3 «
-o co
•> ro
-5 -*z
1 =
"O
u
2
CD
^-^
^
g>
'c
»
T3 CT3
il
— a.
to -^
CD C
s c
4; ^
a>
^5 "^
O -£-
E
,3
CD
2
CD
2?
O
n» 03
5J O
O3 Q) Q
II 1 -
— 03 — ^
ca "?=
-U r- §
i 'I §
>- CO
CD O3 CD
1 is
i i-i
"3
03
tu
•b
o
OJ
"ra
S
2
CJ
C
•D
"C!
O
is
^j
s
-—^
— ^
if
en
en
o
O)
E
_3
CD
S
^CD
•Computer model to generate
exposure information (See Chapte
•Computer model may also calcula
risk (e.g., HEM-II).
"o
CD
CD
1 1
CD CO -B
o £ §
^ 3 co
>- — CO
CD CO CD
•3 rj CO
1 il
-o
£ |"i
S | E _
"ca CD Jo "^
I | If
"S !
"s
"53
-2 i
CD
5
5
o
2 !
i
CO '•
,^>
LU
C3
o
1
LU
X
5-20
-------
In estimating the time and resource demands of a risk characterization, time should be
allotted to clearly document the methods, assumptions, and uncertainties involved and to include
qualitative as well as quantitative information. This documentation is an essential part of a valid
cancer risk characterization.
5.4 QUANTIFICATION OF NONCANCER RISKS
5.4.1 Overview
This section presents information on methods of developing and presenting estimates of
noncancer risks. Section 5.4.2 describes commonly used measures for characterizing individual
and population risks. Methods of estimating noncancer risks for exposure to multiple compounds
and via multiple pathways are also discussed. To aid in selecting noncancer risk characterization
X-
methods, Section 5.4.3 discusses the study objectives, time, resources, and expertise required for
each method.
5.4.2 Technical Background and Issues
5.4.2.1 Comparison of Exposures to Reference Concentrations or Doses. The
concepts of individual and aggregate (population) risks introduced in the cancer risk
characterization section are also applicable to noncancer risks. As in the calculation of MIR or
MEI, a study can characterize the risks to the individual(s) predicted to receive the maximum
modeled exposure including both those around the central tendency and those at the high end of
the risk distribution. A distribution of exposures and risks for the study population can aiso be
presented.
Unlike cancer risk characterization, noncancer risks typically are not expressed as a
probability of an individual suffering an adverse effect. Instead, the potential for noncancer
effects is evaluated by comparing an exposure level over a specified period of time (e.g., lifetime)
with a reference dose, representing the highest "safe" dose, derived for a similar exposure period.
A ratio of exposure to oral reference dose or inhalation reference concentration (E/RfD or E/RfO
is commonly calculated in Superfund and other EPA risk assessments (EPA/OSW, 1989;
EPA/ECAO. 1989). The maximum modeled exposure is typically used in this comparison.
especially in a screening study although recent guidance suggests a central tendency (e.g., mean
or median) and the high end should also be used. However, a range of exposure levels predicted
for the study population can aiso be used to determine how close various subgroups are to the
5-21
-------
RfC, or to see how many people are exposed to concentrations above the RfC. The derivation
of inhalation reference concentrations (RfC) and oral reference doses (RfD) are described in
Chapter 3. The RfC and RfD are useful reference points for gauging the potential effects of
other doses. Doses below the RfC or RfD are not likely to be associated with adverse health
effects. As the amount and frequency of exposures exceeding the RfC or RfD increase, the
likelihood of adverse effects increases. However, it is not categorically true that exposures below
the RfC are of no concern or that doses above are likely to cause health effects (EPA/ECAO,
1990). Furthermore, the E/RfC ratio should not be interpreted as a probability. Cross-substance
comparisons of the E/RfD ratio may not be valid, and the level of concern does not increase
linearly as exposures approach or exceed the RfC. This is because RfCs have varying levels of
accuracy and precision and are based on different types of health effects. Also, the slope of the
>>
dose-response curve above the RfC can vary widely depending on the substance (EPA/OSW,
1989).
As in cancer risk characterization, temporal and spatial consistency are desirable when
comparing or combining outputs of the exposure and dose-response assessments. The inhalation
RfCs are all derived for chronic exposures via inhalation of compounds in the ambient air. There
are also oral RfDs for chronic exposures via ingestion pathways. The RfC or RfD should match
the exposure pathway. IRIS contains both inhalation RfCs and oral RfDs. Recently, some
subchronic oral RfDs have been developed for use in risk assessments for Superfund sites where
exposures may last for shorter periods of between 2 weeks and 7 years (EPA/OSW, A989).
However, the Agency has not determined the appropriateness of these subchronic RfDs to other
types of risk assessments.
Although a compound may produce several types of noncancer effects, the RfCs and RfDs
are generally based on the critical, or most sensitive, effect. To address developmental effects
in particular, EPA and some States have begun deriving developmental RfCs (RfCDT) to evaluate
the potential effects on a developing organism following a single exposure event (EPA/OSW,
1989; CDHS, 1989). The derivation and use of developmental RfCs are described ;n the
"Proposed Amendment to the Guidelines for the Health Assessment of Suspect Development
Toxicants" (EPA, 1989, 54 FR 9386).
In any risk characterization where an RfC or RfD is used, information about its derivation,
assumptions, and uncertainties should be presented. For example, the critical effect associated
5-22
-------
with the RfC, the type of epidemiologic or toxicologic studies considered, the uncertainty and
modifying factors used in deriving the RfC, and the uncertainties and degree of confidence in the
RfC should be discussed.
Margin of Exposure. The EPA has also used a similar measure of noncancer risks, the
margin of exposure (MOE). The MOE is the magnitude by which the No-Observed-Adverse-
Effect-Level (NOAEL) of the critical effect exceeds the estimated exposure, where both are
expressed in the same units (EPA/ECAO, 1990). The general equation is:
MOE = NOAEL (given as the experimental dose)/E (given as the human dose).
If the ratio is greater than the product of the uncertainty factors and modifying factors that would
be used in deriving an RfC, then there is relatively little likelihood of human health effects.
•\-
Again, a discussion of the studies, critical effect, methods and assumptions used to derive the
NOAEL, and the associated uncertainties should be included in a risk characterization.
5.4.2.2 Hazard Index for Chemical Mixtures. While some potential
environmental hazards may involve significant exposure to only a single compound, exposure to
a mixture of compounds that may produce similar or dissimilar cancer and/or noncancer health
effects is more common. The EPA has developed guidelines for chemical mixture risk
assessment, which are discussed in more detail in Section 6.2. In a few cases, toxicity studies
may be available for a chemical mixture of concern or for a sufficiently similar mixture. In such
cases, risk characterization can be conducted on the total mixture using the same procedures used
for a single compound.
However, noncancer health effects data are usually available only for individual
compounds within a mixture. In such cases, a hazard index (HI) approach is typically used
(EPA, 1986; EPA/OSW, 1989). This approach is based on the assumption of dose additivity (as
opposed to synergism or antagonism). When RfCs are used as the measure of noncancer effects,
the HI is calculated as:
HI =
E, + E2 + .... + E,
RfC, RfC, RfC,
Where:
E, = exposure level to the ith toxicant, and
RfC, = reference concentration for the i!h toxicant.
5-23
-------
As the HI approaches unity, concern for the potential hazard of the mixture increases
(EPA, 1986). If the HI exceeds unity the mixture has exceeded the equivalent of the RfC for the
mixture. The HI approach assumes that simultaneous exposures to several compounds (even at
subthreshold levels) could, in combination, result in an adverse health effect. Even if no single
compound exceeds its RfC, the HI for the overall mixture may exceed 1. However, the HI
should not be interpreted as a probability of risk, nor as strict delineation of "safe" and "unsafe"
levels (EPA, 1986; EPA/OSW, 1989). Rather the HI is a rough measure of likely toxicity and
needs to be interpreted carefully.
For ambient exposures, the inhalation RfCs represent chronic exposures, and chronic
exposure levels would be used in calculating the HI. In Superfund risk assessments, where both
chronic and subchronic exposures may be encountered and subchronic as well as chronic oral
RfDs may be used, separate His are calculated for chronic and subchronic effects (EPA/OSW,
1989).
The hazard index approach has some limitations. The assumption of dose additivity is
most properly applied to compounds that induce the same effect by similar modes of action
(EPA, 1986). Consequently, application of the HI equation to compounds that may produce
different effects, or that act by different mechanisms, could overestimate the potential for effects.
However, this approach may be appropriate for a screening-level study (EPA/OSW, 1989).
From a scientific standpoint, it is more appropriate to calculate a separate hazard index
for each noncancer endpoint of concern (EPA, 1986), when mechanisms of action are known to
be the same. Segregation of His by effect (and/or by mechanism of action) is complex and
should be performed by a toxicologist. If not done properly, risks could be underestimated, fa
the risk characterization using His, there should be explicit discussion on the uncertainties of the
assumption of additivity, knowledge of interactions, differences in confidence for individual
acceptable level, and mixture composition.
5.4.2.3 Multiple Pathway Hazard Index. His can be calculated for multiple
exposure pathways. For example, a chronic hazard index for ambient exposures (using inhalation
RfCs) could be calculated; a chronic hazard index for a drinking water or ingestion exposure
route could be calculated (using chronic oral RfDs); and the two His could then be added.
However, the assumptions of this approach should be examined carefully.
5-24
-------
It is important to develop a reasonable exposure scenario. If the location of the maximum
modeled exposure for two different pathways is not the same, it would be unreasonable to
combine them. For a detailed study, it would be appropriate to consider whether the same
individual or subpopulation within a study area is likely to face the maximum modeled exposure
from more than one pathway.
A second consideration is whether the noncancer endpoints and mechanisms of action are
similar or dissimilar for the compounds and exposure routes associated with the various
pathways. If not, the hazard index approach may not be appropriate, since it is based on the
assumption of dose additivity.
5.4.2.4 Dose-Response Modeling. The dose-response modeling approach typically
used in cancer risk assessment has recently been applied to noncancer risk assessment to predict
the probability of an adverse noncancer health effect at various dose levels (EPA, 1991).
Noncancer dose-response modeling is still generally theoretical and is not typically used to
support any regulations for air toxics. This approach is feasible when sufficient quantitative
epidemiologic or toxicologic study data are available and has been conducted for criteria air
pollutants (e.g., lead, ozone, sulfur dioxide) and used to support decisions on national ambient
air quality standards. Furthermore, the scientific issues associated with the threshold concept,
animal to human extrapolation, and any low-dose extrapolation outside of the range of the
experimental data must be carefully considered to ensure that the selected dose-response model
is appropriate. In risk characterization, the output of the exposure assessment (the estimated
human exposure or dose ievel(s)) can be combined with dose-response modeling to generate a
quantitative health risk estimate. Individual risk, aggregate population risk, or risk distributions
can be estimated, as described in Section 5.3.2.
Where dose-response modeling is used, the risk characterization should include a
discussion of trie observed dose-response behavior of the critical effect(s), data such as the shapes
and slopes of the dose-response curves for the various toxic endpoints over the range of
experimental data, pharmacokinetic information, and a discussion of how this information was
used to determine the appropriate dose-response model. The uncertainties inherent in dose-
response modeling (including interspecies extrapolation, and low-dose extrapolation) should also
be discussed along with methods and uncertainties in the exposure assessment.
5-25
-------
5.4.2.5 Decision-Analysis Approach. When a decision-analysis approach is taken
in the dose-response and/or exposure assessments, this also has implications for risk
characterization. In this approach, major uncertainties are explicitly characterized and
probabilities are used to convey the degree of uncertainty. These probabilities are estimated
either by Bayesian statistical techniques when there is sufficient dose-response data to
characterize the relationship or by obtaining probabilistic judgements from selected exposure
and/or health experts using a format interview process, referred to as probability encoding
(Whitfield, et al., 1993), (see Section 3 and EPA/ECAO, 1991). Using this approach, the risk
characterization not only presents the best estimates of individual and/or population risk, it also
explicitly conveys the degree of uncertainty. For example, the 50 percent credible interval range
or the 90 percent credible interval range can be presented. The 90 percent credible interval range
would be the numerical range (above and below the best" estimate) of individual or population
risk within which the experts are 90 percent certain the "true" risk lies.
As with the other approaches, the risk characterization should include discussions of data,
methodology, assumptions, and uncertainties in the dose-response and exposure assessments. In
particular, information on the selection of experts and the process of eliciting their judgements
on the magnitude of uncertainties and converting these to probabilities should be presented
whenever a decision-analysis approach is used.
5.4.2.6 Summary of Methods. Table 5-6 summarizes the various methods of
noncancer risk characterization and Table 5-7 presents information to aid in the selection of
noncancer risk characterization methods. The selected method will depend on the amount of
information available, the level of detail needed, and a number of practical considerations
including the time, resources, and expertise available for the project. Therefore, even when a
quantitative risk characterization is performed, the level of detail and complexity can vary greatly.
Insufficient information on the noncancer health effects or emissions of a compound may
preclude a quantitative risk characterization. One method is the comparison of exposure to either
the inhalation RfC or oral RfD, depending on the exposure pathway. The output is only a
comparison of the estimated exposure level with a benchmark level unlikely to cause an adverse
health effect; it is not a probability of a health response. The cautions on interpretation of the
exposure to RfC comparison contained in Section 5.4.2.1 should be heeded. This approach is
5-26
-------
CO
Q
LU
CO
uS
THER CHARACTERISTICS OF METHOD
O
t—
Q.
O
Q.
—
o
5~
CC
a.
OL
cr
o
<
3
o
i
o
EPA-approved RfC's and RfD's available
in IRIS.
RfC's and RfD's are based on health data
with uncertainty factors and modifying
factors.
If E/RIC <1, the exposure level is unlikely
to cause adverse effects.
E/RfC ratio is not a probability of a
noncancer effect or a "bright line" of
acceptablity
.
o
To
Q ~a
£ -g
LU §
5-5
.2 -S3
2 c5
o "OT
2 ™
u3 p
0
E "f
(/> ™ 3 6
1 l~l
2* ,| j5 ^ -=*
CC g -o £ = .2
O _Q 03 Q :S "5
CC ~o «S cu ^ CL
* •
31
~>^
ca ^*
•s M
a. ro
§_.
^ 1
^o o
~S 2^
•= ~
O Q
cc o:
, UJ U]
CO
UJ
Q
Q
2
CC
Q
CO
Q
0
LU
^
§ 0 CO
la y -2
^ fT* --"^
-C -— *• !^
2 5 7T
2: 03
3 -2i ca —
0 2 o «
Provides index of how human exposure
compares to the NOAEL from
experimental studies.
Uncertainties in species differences, length
and quality of studies, etc. should be
considered in interpreting MOE.
If NOAEL is not available from EPA, user
could review health studies and derive
NOAEL. Would need to describe data
procedures and assumptions.
• * •
.Q
2""*
S
LU
O
^^
^r"
03
—
tn ^
S 0
CO CD
cr =
E |
J= J=
~^ T3
O50
• *
UJ
UJ
••X
O
ijj"
0
^.
2
o
CL
X
LU
"o
£
CO
2
Could be used where no EPA-approved
RfC is available.
Value derived would not be an approved
RfC unless it underwent EPA peer review
Procedures, assumptions, and
uncertainties in derivation should be
presented.
Ratio is not a probability of noncancer
effect.
• • • •
u
2
.2
•-
~§v.
1 0
•o *—
5> ^
It
CD
^
O
2o |-
— ? c
•I 1. i
0 E jl
58 .0
CJ 03 o
f'~ ~ J
5 "
r- C ~°
E « E j> -5-
QJ O 03 (jj ^Xi,
*j "o y ~o QJ
°- — o .g ?,
LU ui 0 2 -2
* •
"H
'o
a.
o
03
o5
*a5
T3
.>
al
a5
(7)
ill
a
i ™?
o5 (^ o
3 f o
0 •§ QC
03 O- O
^5 CD "*~
?n CJ tn
0 "=
-------
I
CD
in
UJ
3
THER CHARACTERISTICS OF METHOD |
o
i—
Q-
•5
O
1
^_
o
rr
Q.
a.
oc
o
5
— J
o
o
en
UJ
O
2
CC
0
C/3
Q
§
h—
UJ
S
If Hl<1, there is less likelihood on
noncancer effect
The HI is not a probability of a noncancer
effect and not a "bright line' of
acceptability
Assumes dose additivity, which is most
appropriate for compounds with similar
effects and mechanisms of action
Effecl-spedfic Hi's should be calculated by
a toxicologist or may understate risk
.
CD
1
ZS
.0
X
"*2
•o
1
X
3;
1 1
Q §" -5
?8 I
O -C X "O
o
i
QJ
•35 >p
C >
8|
x S.
•
2. ~~ ™ -o
f 1 1 f
2 « §•"
S. = s "S
§ §. 1 °
" §• f= g-
•^ C CO CO CO
O -55 CD cj O o ^
uj cu cu ro •— Q. 3
O S ^ " O T3 £
CC -3- 2 -2 iZ .£ 2 .2 -J co 0.
•
o
o" ^
LU O -^3*
*" a. -—
•*• 2. S-
O ^ c
_ ,._ CD
=t[ QJ w_
UJ 0) 8
illl
£ u c
~--_ ^ o
UJ 'J C
11 ca "o
x 1-3
CO
a.
J
O CO
"~ fl)
X -§
^, ^5
x 2
•o -o
^ i
i — ,— i
3 £
T3 O
X 0
• « •
C/I
ca
^
a.
"H
CU
cu
1 53
0 -^
**"* iU
= s»
-*- o
Ji
^ cO
O S
X
^^
ca
a.
^s
2
Output is a probability of (an) effect(s)
Requires sufficient epidemiologic or
toxicologic data
Scientific issues (such as spedes
differences, thresholds, pharmacokinetics,
mecharrisms of action) must be addressed
in selecting dose response model and
described as part of risk characterization
. .
3
Id c
-i „ 5 -2
« SO ^ OJ
* s ^^.
"in -a cu o
•= co 3= a.
gj — CD g
•— s
S s1 y g
<= J3 0 -S
3 2 §=5
O -S £ .E
•
1 - -5-2
"5 O (D (13 "^
° -^ S. 5 ^
111 11
<2 -§ co £ at
S" o> * "55 f; -o
2 -£ V 5 co !2
« S ¥ » 1 "i
S =5 .2 S ^ ~
Q .= "a -a 2 ,2
* <•
o t
-^ "o ^
"5 "3 "^
§- f-s.
f-20-2
O 3 CO "Jo
"S O 3 ^§
-a x"c» E
.> * *- c
1 § '1 --S
t5 S s. 1 -i
<>> c co •• _^ ^Q
S !s S 5 <"
•S co w ^ 2
O S:^ 2 O "5
. . .
o
i i
Is!
* £- «
a> r: -c
(n o a?
§*- Q, W
C X «D
8-i •'€
to § o '5
§ $ 1 1
O ca CL S
* " i
1
cn 53 -rr
Q. "aj
1 , Iff
•5 ~ a. c cu ^2
2=2^^3
o -g « .2 2 E
s 1 a s s i-
^= tl X =** ° T3 -.
^3 Q. OJ CO CD .3, ^
-^ 03 "o .- -2 .12 3
i. i S 2 ^ -^ «"
liIHl!
O CO OJ CO CO O_ "O
. 1
3 i
a
^
o
Ss
ea
c:
g
o
O
m
Q.
_OJ
to
o
O
JO
o
<
CO
5-28
-------
Q
O
M
CC
LU
co
cc
CC
LU
O
O
O
z
CC
o
cr
LU
H;
CC
O
Q
i—
O
UJ
CO
>-
cr
<
LU
CQ
<
LU
p2
LU fC
^— ^^
-£
Q oZ
P <
CO W
VJJ —
U- Q
O O
_j ^Z
UJ 1—
S 2
Q
g
}_
LU
S
i
O3
«
1
Q3
=3
0) <"
CO Q.
CO x
X3. OJ
•° 1 —
CC en ^3
o .2 9-
§ £ 03
Q O3 03
< 2 -2.
* *
03
_03
O
5 T^
• Individual with a 1
of risk assessmel
1
5
13
-•-'
O)
c:
'c
03
O3
O
CO
.
o
CJ
cr S
2 E
O3 _9;
5 °
§.-§
X .£
O3 CO
03 CO
^ O
O =i
O CC
|
03
_
O
~*
to IRIS to other sources of
information (see Chapter 3)
to generate exposure estimate
CO — 1 T3
to LLI 0
03 <; jz
0 O 3
*
^1
LU
13 O
e-f
3 .S
.£ S3
o "o .
•- €
o uj"
"D
O3
-2
~§
o
-*-
O3
C
'c
Q3
O
CO
TJ
Q
•*•"
u?
o
2,
2
55
8.8
LU 'E
^_ CD
O -C
<~~
o>-2
5 c
^^ '"
«£? V}
s
c
o
-a
1
u
c
o
CO
03
1
"ca
O3
JC
o
11
< .E
*
• Toxicologist
T3
Oj
3
O3
Q
Q
._. ^^
, cr
(U W3 O
= - o
— 'o
03 Q- 0
5 03 "~
Tf, o .«
"O T3
J3J 03 <=
§ |-s
5 •= o S
• Individual with a 1
of risk assessmei
• Toxicologist if cal
effect-specific ha,
TJ
3
O3
O
O3
"c
O3
03
^j
CO
_o
CJ
CO
2
O-
x
03
"3
~
-a 33
CO 3
CO —
x 2
±
o
E
1
2
« 1-
||
S> °-
ssf
o i E
S »- °
OJ **~
co E .E
cr R "c/5
— " as
o o 03
CO T3 3
(O O CO
03 JT o
lilt
• Toxicologist
-a
-03
c5
"a3
"^
,2
"«
•a
o
2
X
03
-a
CO
^j
i
_£Z
'to
g.
^
2
.c
•^
to epidemiologic or toxicologic
sponse model (computer)
or model to estimate exposure
to co S -o
(O 03 ' O
•• § 1 § i
< to Q 2
• Toxicologist
•o
.2>
'5
^3
t~—
. .
C
03
CO
03
CO
C/5
**-
1
UJ
OJ
OC
^
Q
"§,
^p
sponse model
or model to estimate exposure
) elicit expert characterization of
nties and present them as
ities
*r o "^1_ t ~S
(11 r— ^* ^ CO
% & -a g ^
o ^ = g o
Q 2 co =i o.
"O
03 Si £
CO 3 ra
oo c £•
Q. Q. — 0
co x -a <"
c 03 03 03 -S
ill It
§_ -S ^- — s_ "g
15 ,2 to co c o ^"
53 8 o3 a3 S :« 1
JT 'i^ CO C/3 Jrr O ^T
O3 55 (O CO O3 O3 iri
co S co ca Q. -o S
-a
O3
.2
'S
a :
o
CO
O
a.
Q.
o
"ca
c
0
10
"o
O3
Q ;
M
c
I
o
5-29
-------
relatively quick and straight-forward and does not require a toxicologist. The RfCs and RfDs,
along with information on their uncertainty and level of confidence, can be found in IRIS.
For exposure to mixtures of noncarcinogenic compounds, a hazard index approach is most
commonly used. This involves dividing the exposure concentration by the acceptable level of
exposure (e.g., RfC/RfD). To characterize the noncancer risk for a mixture, the hazard index can
be computed from the ratio of exposure to RfC/RfD for each compound. Because this approach
is based on the assumption of dose additivity, it is most appropriate for compounds with the same
effect or endpoint and similar mechanisms of action.
A screening approach might calculate one hazard index, not differentiated by the type of
noncancer effect caused; however, this may under- of overestimate risk. Any available evidence
as to whether effects of two or more compounds are additive, synergistic, or antagonistic, should
X-
be considered. Instead of calculating a hazard index for a mixture, effect-specific hazard indices
can be derived, but expert judgment of a toxicologist is required to avoid underestimating the
risks. Mixture risk assessment is more time consuming than single-compound risk assessment
because information must be gathered on a number of compounds.
Two recently developed approaches to noncancer risk assessment, dose-response modeling
and the decision analysis approach can also be used. The advantage of these approaches is that,
where they can be used appropriately, the risk characterization can provide a probability of
noncancer health effects occurring in the study population. However, these approaches usually
require detailed experimental data, expert judgement, an understanding of the exposure/RfD ratio
or hazard index approach and of mechanisms of action and pharmacokinetics, as well as the use
of computer modeling.
If no RfC or RfD is available in IRIS, there are several options for characterizing
noncancer risk. One is to perform a qualitative rather than quantitative risk characterization.
This involves presenting the evidence gathered during hazard identification on the potential
noncancer effects of the compound without attempting to determine the likelihood that the human
exposures modeied in the study will cause an adverse health effect. Another option ;s to
determine a NOAEL from the available epidemioiogic or toxicologic data and calculate a margin
of exposure (MOE). In this case, the selection of the critical study, critical effect, and NOAEL
would need to be justified. A third option is to use the available data and follow the procedures
for developing a new RfC or RfD (see Section 3 and EPA/ECAO, 1989). The risk
5-30
-------
characterization would ultimately compare exposure to this user-derived value. This would not
be an officially-approved RfC or RfD unless it went through the EPA peer review process. A
toxicologist would need to be consulted during the dose-response and risk characterization steps.
When exposure is compared to a user-derived NOAEL or user-derived reference dose, the risk
characterization step is somewhat more complex than when an EPA-approved RfC or RfD is
used, because a more detailed discussion of the derivation of the reference dose may need to be
presented.
5.5 CHARACTERIZATION OF UNCERTAINTY
Recent EPA guidance calls for a full characterization of risk, not just the single point
estimate which has become synonymous with risk characterization. Critical to full
characterization of risk is a frank and open discussion of the uncertainty in the overall assessment
and in each of its components. Numerical estimates should always be accompanied by
descriptive information carefully selected to ensure an objective and balanced characterization
of risk (EPA, 1992).
Uncertainty can be introduced into a health risk assessment .at every step in the process.
It occurs because risk assessment is a complex process, requiring the integration of:
• the fate and transport of pollutants in a variable environment by processes that are
often poorly understood or too complex to quantify accurately;
• the potential for adverse health effects in humans as extrapolated from animal
bioassays; and
• the probability of adverse effects in a human population that is highly variable
genetically, in age, in activity level, and in life style.
Even using the most accurate data with the most sophisticated models, uncertainty is inherent in
the process.
Finkel (1990) classified all uncertainty into four types which are summarized in Table 5-8.
The first two, parameter uncertainty and model uncertainty, are generally recognized by risk
assessors as major sources of uncertainty.
Parameter uncertainty occurs when variables cannot be measured precisely either because
of equipment limitations or because the quantity being measured varies spatially or temporally.
Random, or sample errors, are a common source of parameter uncertainty that is especially
5-31
-------
Ul
2
en
CO
LLJ
CO
CO
CO
cc
z
>
cc
LLI
O
CO
111
o
en
CD
<
Comments/Examples
C
r:
03
O
5
"o
fl3
3
o
CO
.0
'o
03
Q.
CO
03
Q.
h-
"m
0)
03
CD
|
03
• include limitations of equipment, methodology, and human
• some processes impossible to measure exactly
£
0
03
urement
C/)
O3
O
2
^^
£Z
'ra
•c
CD
Q
^
03
03
E
2
03
Q.
sampling errors
• can be minimized increasing sample size
<73
^_
o
03
P
O
-o
c
eg
CC
• nonrandom errors
• result of inherent flaw in data gathering processes
• minimize by exernal peer review
o
03
O
'5
03
~y3
•>,
co
CO
c
e.g., use of animal bioassays to determine effect on huma
.2
"^t
CO
03
"to
O)
O
^
~^
co
2j*
c
03
o
—
"3
-a
o
2
co
03
N
• may be a result of model simplification or failure to recogn
important variable
CO
0)
.a
to
to
•a
03
T3
73
X
LU
• for example, failure to recognize importance of episodic
meteorological events
CO
g
.±±
o
u
"5
o
c
<
CO
• for example, choice of dose-response model for carcinoge
E
,0
03
0
"o
03
^
0
o
c
,
:^
CO
CO
o
O3
cn
o>
JiT
C
LL
T3
03
B.
to
5-32
-------
critical for small sample sizes. More difficult to recognize are nonrandom or systematic errors
that result from bias in sampling, experimental design, or choice of assumptions. Since the risk
assessor is unlikely to detect his or her own bias, it is wise to use external peer reviewers on any
risk assessment protocol.
Model uncertainty is associated with a variety of models used in all phases of a risk
assessment. These include the animal models used as surrogates for testing human
carcinogenicity as well as the computer models used to predict the fate and transport of chemicals
in the environment. The use of rodents as surrogates for humans introduces uncertainty into the
risk factor since different species do not respond to toxins in exactly the same way. Computer
models are simplifications of reality and some variables are excluded. The risk assessor needs
to consider the importance of excluded variables on a case-by-case basis, because a given
variable may be important in some instances and not in others. A similar problem can occur
when a model that is applicable under average conditions is used for a case where conditions are
abnormal. Large bodies of water, for example, can cause meteorological conditions that are not
adequately modeled by air dispersion models such as ISC. Finally, choosing the correct model
form is often difficult because conflicting theories seem to explain a phenomenon equally well.
Cothern (1988) gives a good example of this in contrasting dose-response models; four equally
reasonable models give risk estimates that vary by four orders of magnitude.
The third type, decision-rule uncertainty, is probably of more concern to risk managers.
This type of uncertainty arises out of the need to balance different social concerns when
determining an acceptabie level of risk, for example. The risk assessor needs to understand the
rationale for setting certain acceptable levels because the rational can affect the choice of model,
data, or assumptions. Finkel (1990) provides a complete discussion of decision-rule uncertainty.
Variability, the fourth type of uncertainty, is often used interchangeably with the term
"uncertainty," but this is not strictly correct. The variability of a characteristic may be known
with absolute certainty. For example, the age distribution of a population may be known and
represented by the mean age and its standard deviation. The fact that ages do vary introduces
uncertainty into characterizing risk for that population. On the other hand, if the age distribution
may not be known, then the variability associated with the population's age is in itself an
uncertainty.
5-33
-------
The degree to which all types of uncertainty need to be quantified and the amount of
uncertainty that is acceptable varies. For a screening level analysis, a high degree of uncertainty
is often acceptable, provided that conservative assumptions are used to bias potential error toward
protecting human health. Similarly, a region-wide or nationwide study will be more uncertain
than a site-specific one. In general, the more detailed or accurate the risk characterization, the
more carefully uncertainty needs to be considered.
Figure 5-1 depicts the factors contributing to uncertainty and the orders of magnitude of
uncertainty associated with each factor. The bar by each uncertainty factor shows the likely
direction of uncertainty. For example under Unit Risk Estimate, if synergisms are present, the
maximum individual risk could be an order of magnitude higher than the EPA estimate.
However, if antagonistic interactions occur, the maximum individual risk could be an order of
magnitude lower than the EPA estimate.
Often, the sources of uncertainty in a risk assessment can be determined, but they cannot
be quantified. This can occur when a factor is known or expected to be variable, but no data are
available (e.g., the amount of time people at a specific site spend out of doors). In this case,
sometimes default data are available that can be useful for estimating a possible range of values.
Uncertainty often arises out of a complete lack of data. A process may be so poorly understood
that the uncertainty cannot be quantified with any confidence. When uncertainty can only be
presented qualitatively, the possible direction and orders of magnitude of the potential error
should be considered.
Knowledge of experimental or measurement errors can also be used to introduce a aegree
of quantitative information into a qualitative presentation of uncertainty. For example, standard
laboratory procedures or field sampling methods may have a known error level that can be used
to quantify uncertainty. Cothern (1988) describes how the well-known experimental and
laboratory procedures used in animal bioassays can be used in a practical way to estimate order-
of-magmtude uncertainties for toxicity studies.
In many cases, the uncertainty associated with particular parameter values or for the
estimated risks can be expressed quantitatively. Finkel (1990) identified a six-step process to
producing a quantitative uncertainty estimate. Initially, the measure of risk should be defined
(e.g., deaths, life-years lost, maximum individual risk, population above an "unacceptable" level
of risk). More than one measure of risk may result from a particular risk assessment: however.
5-34
-------
Probable Impact of Major Assumptions on
EPA's Risk Assessment—MIR-Model plant
-2
Orders of Magnitude
EPA
— 1 Estimate +1
L
Unit Risk
Estimate
HHAG/CAG
API/CMA
GARB
Sensi Sub-Pop
Synergism
Antagonism
Other D/R Models
Emissions &
Pource
arameters
Fugitive
Stock
Eff Release Ht
Pfont Life
Disp Model
Complex Terrain
Urban Release
Dev wl bldg wake
Bevno bldg woke
Exposure
Assessment
HEM Meteorology
Plant Property
Exp at Pop Cant
Indoor « Outdoor
Migration
Human Activity
Urban/Rural Met
Lot/Long Urban
Lot/Long Rural
Area Emis—Point
Figure 5-1. Example of a Semi-Quantitative Presentation of Uncertainty
5-35
-------
the uncertainty should be quantified for each individually. Second, one or more "risk equations"
should be specified which present the mathematical relationships that express the risk measure
in terms of its components. This step is used to identify the important parameters in the risk
estimation process. Third, for each parameter or equation component, an uncertainty distribution
is generated. These uncertainty distributions may be generated by the use of analogy, statistical
inference techniques, or elicitation of expert opinion, or some combination of these. These will
be described in more detail below. Fourth, the individual distributions should be combined into
a composite uncertainty distribution. To carry this out, Monte Carlo simulation is frequently
used; this method is discussed in greater detail below. Fifth, the uncertainty distributions should
be "recalibrated". Inferential analysis could be used to "tighten" or "broaden" particular
distributions to possibly account for dependencies among the variables and/or to truncate the
distributions to exclude extreme values. Finally, the output should be summarized in a manner
which is clear and highlights the important risk management implications. Specific aspects
should be addressed including: the implication of supplanting a point estimate produced without
considering uncertainty, the balance of the costs of under- or over-estimating risks, unresolved
scientific controversies, and implications for research.
When a detailed quantitative treatment of uncertainty is required, statistical methods are
employed. A statistician should be consulted, as many of the methods used require expert
knowledge. Two approaches to a statistical treatment of uncertainty with regards to parameter
values are described here and can be applied to any particular step in the risk assessment process.
The first is simply to express ail variables for which uncertainty is a major concern using an
appropriate statistic (see Table 5-9). For example, if a value used is from a sample (e.g., hourly
emissions from a stack), both the mean and standard deviation should be presented. If the
sample size is very small, it may be appropriate to give the range of sample values and use a
midpoint in the model; or, both the smallest and largest measured value could be used to get two
estimates that bound the expected true value. The appropriate statistic co use depends on the
amount of data available and the degree of detail required. The propagation of uncertainties can
be done using analytical or numerical methods. A common analytical method is first-order
analysis; it is appropriate when only a few parameters are of interest, their distributions are
known, and all relationships are linear.
5-36
-------
TABLE 5-9. SOME STATISTICS USEFUL FOR QUANTIFYING UNCERTAINTY'
Statistic
Sample
Variance
Sample
Standard
Deviation
Coefficient of
Variation
Probability
Density
Function (pdf)
Cumulative
Distribution
Function (cdf)
Expression
8«.rJ&ta£
^ (n-1)
s= yirf
N^(n-1)
rv - s ' 100
y
-(x-^1
1 2o2
fM et
rw • . e
v/2^2
F(x) = P(X < x)
Definition/Derivation
• mean square error
• a measure of the spread of values of a
sample
• square root of the variance
• also a measure of spread but has the
same units as the mean
• data are often summarized ± one standard
deviation
• the standard deviation expressed as a
percentage of the mean
• useful for comparing samples or
populations with different means,
particularly if the means are different
orders of magnitude
• specifies likelihood of occurrence of a
value within the distribution defined by the
specified mean (u.) and variance (cf)
• the "bell curve" for normal data
• specifies the likelihood of a result greater
than any value in the range of a
distribution j
'Formulae shown are for normally distributed data.
5-37
-------
A second approach is to use the probability distributions of major variables to propagate
parameter value uncertainties through the risk analysis. A probability distribution of expected
values is developed for each parameter value. These probability distributions are typically
expressed as either probability density functions (PDF) or as cumulative density functions (CDF).
The PDF presents the relative probability for discrete parameter values while the CDF presents
the cumulative probability that a value is less than or equal to a specific value. These probability
distributions may be developed using statistical methods or through the use of judgmental
probability methods.
The judgmental probability approach requires that experts use subjective judgments to
develop quantitative estimates (encoded probability) of uncertainty for a particular parameter
value. This approach may be applied to parameter values for either the exposure assessment or
dose-response assessment of risk. This method was described in some detail in Section 3.4.3.4.
As described in that section, a main distinguishing characteristic of the judgmental probability
approach is the emphasis on explicitly characterizing and representing uncertainty using
probability as the language to convey the degree of uncertainty. A variety of quantitative
schemes are possible. Experts may be asked to assign order-of-magnitude error bounds about
a data point or model estimate; or, they may be asked to describe the probability distribution of
a variable. Experts are selected to represent the range of credible scientific opinion and,
therefore, to implicitly represent the major uncertainties. As a result, experts may diverge,
sometimes widely, m their estimates. Disagreement among experts needs to be recognized by
the risk assessor in any discussion of uncertainty.
The propagation of uncertainties is accomplished by developing a composite uncertainty
distribution by combining the individual distributions. Numerical methods are often employed
for this phase with Monte Carlo simulations gaining wide acceptance for this purpose. In Monte
Cario simulations, a computer program is used to repeatedly solve the model equations to
calculate a distribution of exposure (or risk) values. Each time the equations are calculated,
values are randomly sampled from the specified distributions. The end result is a distribution
of exposure for risk). These can again be expressed as PDFs, or more appropriately as CDFs.
The distribution allows the risk assessor to choose the value corresponding to the appropriate
percentile in the overall distribution. For example, an exposure level or risk level can be selected
5-38
-------
which corresponds to the 95th percentile of the overall risk distribution rather than relying on a
point estimate of risk based on the 95th percentile values for each parameter.
Numerical methods require the use of a computer, and until recently, were generally too
slow and too difficult to program to be widely used. However, the use of Latin hypercube
sampling and Monte Carlo simulation has increased the speed of these models considerably.
Also, several computer packages are available that minimize (or eliminate) the programming
required (Salmento et al., 1989; Salhotra et al., 1988).
A complete discussion of these statistical techniques is beyond the scope of this document.
For more general information and references to more detailed discussions, refer to Section 8.4
of the Risk Assessment Guidance for Superfund (U.S. EPA/OSW, 1989). Various uncertainty
modeling techniques are compared in Iman and Helton (1988). Monte Carlo simulations have
been used in risk assessments related to exposures of cartion monoxide and ozone (Johnson et
al., 1992a and 1992b), and used to identify greatest sources of model uncertainty in
photochemical models (Derwent and Hov, 1988) and in food chain analysis (McKone and Ryan,
1989). The encoded probability approach has been used to estimate the risks and uncertainties
from exposure to ozone (Whitfield et al., 1993) and lead (Whitfield and Wallsten, 1989).
5-39
-------
5.6 REFERENCES
U. S. Environmental Protection Agency, Environmental Criteria and Assessment Office.
November 1988. Technical Support Document on Risk Assessment for Chemical Mixtures.
Final Draft, ECAO-CIN-572.
U. S. Environmental Protection Agency, Environmental Criteria and Assessment Office. April
1989. Interim Methods for Development of Inhalation Reference Doses. EPA 600 8/88-066F.
Cincinnati, OH.
U. S. Environmental Protection Agency, National Air Toxics Information Clearing House. June
1987. Qualitative and Quantitative Carcinogenic Risk Assessment. Research Triangle Park, NC.
U. S. Environmental Protection Agency, Office of Air Quality Planning and Standards. March
1990. HEM-II Users Guide. Draft. Research Triangle Park, NC.
U. S. Environmental Protection Agency, Office of Solid W^ste and Emergency Response. July
1989. Risk Assessment Guidance for Superfund, Human Health Evaluation Manual Part A.
Interim Final. Chapters 7 and 8. Washington, DC. EPA-540/1-89-002.
U. S. Environmental Protection Agency. Office of Health and Environmental Assessment,
Cincinnati, Ohio. Integrated Risk Information System.
U. S. Environmental Protection Agency, Environmental Criteria and Assessment Office. February
1991. General Quantitative Risk Assessment Guidelines for Noncancer Health Effects. Second
External Review Draft, ECAO-CIN-538. Cincinnati, Ohio.
U. S. Environmental Protection Agency. Guidelines for Developmental Toxicity Risk
Assessment. 1991, 56FR 63798 - 63826.
U. S. Environmental Protection Agency. Office of the Administrator, Guidance on Risk
Characterization for Risk Managers and Risk Assessors, 1992, memorandum from F. Henry
Habicht, II, Deputy Administrator.
Whitfield, R.G., H.M. Richmond, S.R. Hayes, A.S. Rosenbaum, T.S. Wallsten, R.L. Winkler.
M.L.G. Absil, and P. Narducci. (1993) Health Risk Assessment of Ozone, in Tropospheric
Ozone: Human Health ana Agricultural Impacts. David J. McKee (ed.) Lewis Publishers, Boca
Raton, Florida.
Cothern, C. R., W. A. Coniglio, and W. L. Marcus. 1984. Uncertainty in Population Risk
Estimates for Environmental Contaminants, pp. 265-286 in Covello, V. T., L. B. Lave, A.
Moghissi, and V. R. R. Uppuluri (eds). Uncertainty in Risk Assessment, Risk Management, and
Decision Making. Plenum Press, New York. 535 pages.
Derwent, R. and O. Hov. 1988. Application of Sensitivity and Uncertainty Analysis Techniques
to a Photochemical Ozone Model. J. Geophys. Res., 93: 5185-5199.
5-40
-------
Finkel, A.M. 1990. Confronting Uncertainty in Risk Management. A Guide for Decision
Makers. Center for Risk Management, Resources for the Future, Washington, D. C.
General Sciences Corporation. December 1988. PCGEMS Users Guide Release 1.0. Draft.
Prepared for U. S. Environmental Protection Agency, Office of Pesticides and Toxic Substances,
Washington, D.C. Contract No. 68-02-4281.
Iman, R. L. and J. C. Helton. 1988. An Investigation of Uncertainty and Sensitivity Analysis
Techniques for Computer Models. Risk Analysis 8: 71-90.
Johnson, T., J. Capel, R. Paul, and L. Wijnberg, 1992a. Estimation of Carbon Monoxide
Exposures and Associated Carboxyhemoglobin Levels in Denver Residnets Using A Probabilistic
Version of NEM, Prepared by International Technology Air Qulaity Services for US
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC, Contract No. 68-DO-0062.
Johnson, T., J. Capel, E. Olaguer, and L. Wijnberg, 1992b. Estimation of Ozone Exposures
Experienced by Residents of the ROMNET Domain Using A Probabilistic Version of NEM,
Prepared by International Technology Air Qulaity Services for US Environmental Protection
Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC, Contract No.
68-DO-0062.
McKone, T. E. and P. B. Ryan. 1989. Human Exposures to Chemicals through Food Chains:
An Uncertainty Analysis. Environ. Sci. Technol., 23: 1154-1163.
Morgan, M. G., M. Herion, S. C. Morris, and D. A. L. Amaral. 1985. Uncertainty in Risk
Assessment. Environ. Sci. Technol., 19: 662-667.
Salhotra. A. M., R.Schanz, and P. Mineart. 1988. A Monte Carlo Simulation Shell for
Uncertainty Analysis. Prepared for U. S. EPA, Environmental Research Laboratory, Athens, GA,
under contract no. 68-03-6304.
Salmento, J. S., E. S. Rubin, and A. M. Finkei. 1989. A Review of @RISK. Risk Analysis.
9: 244-257.
U, S. Environmental Protection Agency. i986a. Guidelines for Carcinogen Risk Assessment.
51 FR 33998. September 24, 1986.
U. S. Environmental Protection Agency. I986b. Guidelines for the Health Risk Assessment of
Chemical Mixtures. 51 FR 34014 - 34025. September 24, 1986.
U. S. Environmental Protection Agency, Environmental Criteria and Assessment Office.
November 1988. Technical Support Document on Risk Assessment for Chemical Mixtures.
Final Draft, ECAO-CIN-572.
5-41
-------
U. S. Environmental Protection Agency, Environmental Criteria and Assessment Office. April
1990. Interim Methods for Development of Inhalation Reference Doses. EPA 600 8/90-066A.
Cincinnati, OH., August 1990
U. S. Environmental Protection Agency, National Air Toxics Information Clearing House. June
1987. Qualitative and Quantitative Carcinogenic Risk Assessment. Research Triangle Park, NC.
U. S. Environmental Protection Agency, Office of Air Quality Planning and Standards. March
1990. HEM-II Users Guide. Draft. Research Triangle Park, NC.
U. S. Environmental Protection Agency, Office of Solid Waste and Emergency Response. July
1989. Risk Assessment Guidance for Superfund, Human Health Evaluation Manual Part A.
Interim Final. Chapters 7 and 8. Washington, DC. EPA-540/1-89-002.
U. S. Environmental Protection Agency. Office of Health and Environmental Assessment,
Cincinnati, Ohio. Integrated Risk Information System.
•\»
U. S. Environmental Protection Agency, Environmental Criteria and Assessment Office. February
1991. General Quantitative Risk Assessment Guidelines for Noncancer Health Effects. Second
External Review Draft, ECAO-CIN-538. Cincinnati, Ohio.
U. S. Environmental Protection Agency. Guidelines for Developmental Toxicity Risk
Assessment, 1991, 56FR 63798 - 63826.
U. S. Environmental Protection Agency. Office of the Administrator, Guidance on Risk
Characterization for Risk Managers and Risk Assessors, 1992, memorandum from F. Henry
Habicht, II, Deputy dministrator.
Whitfieid. R.G. and T.S. Wallsten, 1989. A Risk Assessment for Selected Lead-Induced Health
Effects: An Example of a General Methodology, Risk Analysis. 9(2): 197-208
Whitfieid, R.G., H.M. Richmond, S.R. Hayes, A.S. Rosenbaum, T.S. Wailsten, R.L. Winkler,
M.L.G. Absil. and P. Narducci, 1993. Health Risk Assessment of Ozone, in Troposphenc Ozone:
Human Health and Agricultural Impacts, D.J. McKee (ed.), Lewis Publishers, Boca Raton, FL.
5-42
-------
6.0 EMERGING ISSUES
6.1 ECOLOGICAL RISK ASSESSMENT
The EPA's mandate requires that it protect human health and the environment.
Although a broad interpretation of that mandate would require protection of nonhuman
populations and ecosystems, legal requirements of legislation and funding realities have
resulted in the emphasis being placed on human health. Recently, several factors have led to
a renewed interest on the part of EPA in ecological risk assessment. In the past, it was
assumed that the environmental levels that were protective of human health would be
sufficient to protect the environment. This has not proved to be true. EPA's Science
Advisory Board (SAB) recognized the importance of ecological systems in their report
Reducing Risk: Setting Priorities and Strategies for Environmental Protection. The SAB
•v-
stated that there are strong linkages between human health and the health of wetlands, forests,
oceans, and estuaries. Most human activities that pose significant ecological risks ~ for
example, the effects of agricultural activities on wetlands pose direct or indirect human health
risks as well. Healthy ecosystems are a prerequisite to health humans and prosperous
economies. Ecosystems, and the plant and animal populations that compose them, are
important as food resources, as bioindicators of environmental degradation, and as an
influence on climate.
Guidelines on ecological risk assessment were not included in the original set of EPA
risk guidelines pubiisned in 1986. Subsequently, the EPA has developed guidelines m other
risk assessment areas (e.g.. developmental toxicity). To date, no EPA guidelines have been
developed for ecological risk assessment. However, individual EPA programs have generated
program-specific guidance for ecological effects and the EPA is in the process of developing
Agency-wide guidelines as well. iMany programs at EPA have always had an ecological risk
assessment component to some degree; (Norten et ai. 1988; Bascietto et ai., 1990). For
exampie. EPA programs in pesticides and water have traditionally been concerned with
impacts on aquatic nd terrestrial organisms. Much of their work has been used, in part, to
develop much of the methodology described below. Examples of program-specific guidance
related to ecological risk assessment include: Interim Report on Data and Methods for
Assessment of 2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin Risks to Aquatic Life and Associated
-------
Wildlife (EPA, 1993a) Proposed Water Quality Guidance for the Great Lake System.(EPA,
1993b) Wildlife Criteria Portions of the Proposed Water Quality Guidance for the Great
Lakes System. (EPA, 1993c) Derivation of Proposed Human Health and Wildlife
Bioaccumulation Factors for the Great Lakes Initiative (EPA, 1993d), Great Lakes Water
Quality Initiative Criteria Documents for the Protection of Wildlife (Proposed) -- DDT,
Mercury; 2,3,7,8-TCDD, PCBs (EPA, 1993e), Bioaccumulation of Selected Pollutants in
Fish (EPA, 1990), Technical Support Document for the Determination of the Need to
Regulate Pulp and Paper Mill Sludge Landfills and Surface Impoundments, (EPA, 1991)
Preliminary work on Agency guidelines for ecological effects began in 1988. As part
of this work, EPA studied existing assessments and identified issues to help develop a basis
for articulating guiding principles for the assessment of ecological risks (EPA, 1991). The
V-
EPA's Science Advisory Board urged the EPA to expand the consideration of ecological risk
issues to include both chemical and nonchemical stressors under programs and laws
administered by the EPA (EPA, 1990). As a result, a new program was initiated to develop
guidelines for ecological risk assessment. The EPA Risk Assessment Forum (RAF),
responsible for EPA's Agency-wide risk assessment guidance initiated three ecological risk
guidance projects: (1) compilation of case studies to illustrate "state-of-the-practice" in
ecological assessments, (2) preparation of long-term plan for developing specific ecological
risk assessment guidelines, and (3) development of a framework to describe the basic
principles for ecological risk assessment and provide a flexible structure conducting and
evaluating ecological risk assessments. Several documents have been published as a result of
these efforts (EPA, 1992 a,b,c) although the case studies have yet to be published.
Section 6.1.1 will introduce the framework for ecological risk assessment. The
methodologies developed by these programs are, in part, the basis for the methodology
discussion in Section 6.1.2. Major issues facing the future development of ecological risk
assessment are also discussed in Section 6.1.3. The field of ecological risk assessment is
rapidly evolving. Therefore, the discussion here will be limited to the activities and products
of the RAF to provide a general overview. Additional references will be mentioned which
can be used to obtain additional details.
6-2
-------
6.1.1 FRAMEWORK FOR ECOLOGICAL RISK ASSESSMENT (EPA, 1992a)
As defined by the framework (EPA, 1992a) ecological risk assessment is a process
that evaluates the likelihood that adverse ecological effects may occur or are occurring as a
result of exposure to one or more stressors. A risk does not exist unless (1) the stressor has
the inherent ability to cause one or more adverse effects and (2) the stressor impacts an
ecological component1 (i.e., organisms, populations, communities, or ecosystems) long
enough and at sufficient intensity to elicit an adverse effect. Ecological risk assessment may
evaluate one or many stressors and/or ecological components.
Ecological risk may be expressed in a variety of ways. While some ecological risk
assessments may provide true probabilistic estimates of both adverse effects and exposure
elements, others may be deterministic or even qualitative in nature. In these cases, the
likelihood of adverse effects is expressed through a semiqiiantitative or qualitative comparison
of effects and exposure.
The distinctive nature of the ecological risk framework results from three differences
in emphasis relative to previous risk assessment approaches. First, ecological risk assessment
may consider effects beyond those on individuals of a single species and may examine
population, community, or ecosystem impacts. Second, there is no single set of assessment
endpoints (i.e., environmental elements to be protected) that can be generally applied in all
ecological risk assessments. Instead the assessment endpoints are selected from a large
number of possibilities based on both scientific and policy considerations. Finally, a
comprehensive approach may go beyond the traditional emphasis on chemical effects to
consider the effects of nonchemical stressors. The ecological risk assessment framework as
proposed by the RAF is shown in Figure 6-1. The framework consists of three major
elements, problem formulation, analysis and risk characterization.
6.1.1.1 Problem Formulation. The first phase is problem formulation, a preliminary
characienzation of exposure and effects. In this phase, a wide range of data and policy issues
must be considered including the examination of scientific data and data needs, policy and
lRecent trends have indicated support and consensus on the use of the term
ecological receptor to replace the term ecological component as proposed by the
RAF.
6-3
-------
Discussion
Between the
Risk Assessor
and
Risk Manager
(Planning)
Ecological Risk Assessment
PROBLEfv
1 FORMULATION
A
N
A Characterization Characterization
L of o'
Y Exposure Ecological
' Effects
1
S,,._
^
RISK CHA
\7 \7
RACTERIZATION
4
IP
o
&
|
55°
=v
on: Verification and Monitoring
4
Discussion Between the
Risk Assessor and Risk Manager
(results)
4
ir
Risk Management ^
Figure 6-1. FRAMEWORK FOR ECOLOGICAL RISK ASSESSMENT
Source: EPA. 1992
6-4
-------
regulatory issues, and site-specific factors to define the feasibility, scope, and objectives of
the ecological risk assessment. The level of detail and the information needed to complete
the assessment are also determined. Ecological risk assessments tend to be of higher
complexity than other traditional risk assessment where data needs are clearly defined. The
high degree of complexity and variability in ecological risk assessments are due to site-
specific differences in environmental conditions, species, nutrient cycling, and community
structure. As a result, ecological component response may vary dramatically among sites.
This systematic planning phase is proposed because ecological risk assessments often address
the risks of stressors to many species as well as risks to communities and ecosystems. In
addition, there may be many ways a stressor can elicit adverse effects (e.g., direct effects on
mortality and growth, and indirect effects such as decreased food supply). Therefore, this
>-
planning phase is critical in defining the approach and to consider all potential issues, inputs,
and effects.
The major elements of the problem formulation phase are characterizing the stressors,
defining the ecosystem potentially at risk, defining the range of ecological effects and
selection of the endpoints to be analyzed. Stressors should be characterized in terms of type
(chemical or nonchemical), intensity (concentration or magnitude), duration, frequency (single
event, episodic, or continuous), timing (occurrence relative to biological cycles), and scale
(spatial heterogeneity and extent). In this context, problem formulation is used in place of the
analogous hazard assessment in human health risk assessments to define the range of potential
ecological effects wnich may occur as a result of the stressor. A wide range of endpoints
may be considered, including the death of a population of one species or a disruption of
nutrient cycling in an ecosystem. The appropriate end points will vary depending on the site
and the chemicals of interest. A useful end point, however, should have the following
characteristics:
i i) have ecological relevance.
(2) be of importance to society, and
(3) have demonstrated susceptibility to the stressor
Endpoints can be divided into two basis categories, measurement endpoints and
assessment endpoints. Ecological effects are difficult to measure directly and would require
6-5
-------
extensive field studies. As a result, measurement endpoints are selected because they are
easily defined and quantified. Ideally these measurement endpoints would be related to some
environmental impact, which is represented by an assessment endpoint. The relationship
between measurement endpoints and an assessment endpoint (ecological impact) should be
evaluated during the problem formulation phase and the most efficient and ecologically
plausible endpoints selected. The selection of endpoints would likely vary from site to site
and depend on both the type of stressor under consideration and the environmental of
concern. Some types of endpoints and examples are given in Table 6.1 for three levels of
biological organization the population, community, and ecosystem levels. The lower levels--
cells or individual organisms—are generally not relevant to an ecological risk assessment,
though few individuals of an endangered species may present legitimate concern.
6.1.1.2 Analysis. The second phase of the framework is termed analysis and consists
of two activities, characterization of exposure and characterization of ecological effects. The
purpose of exposure characterization is to predict or measure the spatial and temporal
distribution of a stressor and its co-occurrence with the ecological components of concern.
The purpose of ecological effect characterization is to identify or quantify the adverse effects
elicited by a stressor and, to the extent possible, evaluate cause-and-effect relationships.
Exposure characterization also has two major components, stressor characterization and
ecosystem characterization. Stressor characterization involves determining the stressor's
distribution or pattern of change. Many techniques can be applied to stressor characterization
including fate and transport modeling similar to that used in human risk assessments.
Ecosystem characterization defines the spatial and temporal distributions of the ecological
component, and the ecosystem attributes that influence the distribution and nature of the
stressor. Characteristics of the ecosystem can greatly modify the ultimate nature and
distribution of a particular stressor through biotransformation by microbial communities or
through other environmental fate processes, such as photolysis, hydrolysis, and sorption.
Characteristics of ecological components may influence their exposure to and response due to
a particular stressor. These characteristics should be defined and may include habitat needs.
food preferences, reproductive cycles, and seasonal activities.
6-6
-------
TABLE 6-1. SOME POTENTIAL ENDPOINTS OF AN ECOLOGICAL RISK
ASSESSMENT
Level of
Organization
Population
Community
Ecosystem
Characteristic
Density/Natality, Mortality
B amboos/Productivity
Genetic Composition/Evaluation
Spatial Pattern
Diversity
Physical Structure
Trophic Structure/Food Webs
Bamboos/Energy Flow
Nutrient Pools/Biogeochemical
Cycling
Potential Adverse Effect
Decrease
Increase (if undesirable species)
Decline in yields of a crop
Selection of resistant species (e.g.,
insects resistant to pesticides)
Change in range or dispersion pattern
Loss of species
Decrease in the number of animals
Decrease or change in complexity
resulting in loss of habitat
Loss of top predators
Decrease in vegetation
Decrease in bamboos per unit area
Disruption of nutrient cycles
6-7
-------
The characterization of ecological effects describes the relationship between the
stressor and the assessment (and/or measurement endpoints) identified during the problem
formulation phase. The characterization begins with the evaluation of effects data that are
relevant to the stressor, ranging from mortality and reproductive impairment in individuals
and populations to disruptions in community and ecosystem function such as primary
productivity. During ecological response analysis, the relationship between the stressor and
endpoints of concern is quantified and the ecological effects elicited are evaluated. In
addition, extrapolations from measurement endpoints to assessment endpoints are conducted
during this phase. The product is a stressor-response profile that quantifies and summarizes
the relationship between (1) the magnitude, frequency, or duration of the stressor in an
observational or experimental setting and (2) the magnitude of response. The stressor-
response profile is then used as input to risk characterization.
In the past, most efforts of evaluating environmental toxicity have focused on
determining the toxicity of an agent for particular species. This usually relied on laboratory
(acute) toxicity studies which yield little information on how that toxicity would be expressed
in natural ecosystems (e.g., community structure, species population, diversity). Furthermore,
these efforts usually focused on direct exposure to a particular contaminated media and did
not address inter-media transfer and more importantly bioaccumulation and food chain
pathways. Therefore, there is a paucity of existing data by which to evaluate ecological
effects of particular stressor. Ideally, fieid testing or chronic testing provides more useful
information.
6.1.1.3 Risk Characterization. As with human health risk assessment, risk
characterization is the final phase of ecological risk assessment. During this phase, the
likelihood of adverse effects occurring as a result of exposure to a stressor are evaluated.
Risk characterization contains two major steps: risk estimation and risk description.
The stressor-response profile and the exposure profile from the analysis phase serve as
input to risk estimation. Three general approaches are suggested to integrate the stressor-
response and exposure profiles: d) comparing the single effect and exposure values ('also
known as the quotient approach discussed below); (2) comparing distributions of effects and
exposure; and (3) conducting simulation modeling. The final choice as to which approach.
6-8
-------
will be selected depends on the original purpose of the assessment as well as time and data
constraints. The uncertainties identified during all phases of the risk assessment should be
examined for the four major areas of uncertainty: conceptual model formulation, information
and data, natural variability, and error. To conform to recent EPA guidance on risk
characterization (discussed in Section 5), uncertainty analyses should be explicit in all
ecological risk assessments.
Risk description has two primary elements: ecological risk summary and ecological
significance. The ecological risk summary can be divided into three components: summary of
risk estimation and uncertainty, weight of evidence, and identification of additional analyses.
First, the results of the risk estimation are summarized in a quantitative statement and the
uncertainties associated with problem formulation, analysis, and risk characterization are
discussed quantitatively and/or qualitatively. Secondly, the confidence in the risk estimates is
expressed through a weight-of-evidence discussion. The weight-of-evidence discussion should
consider the sufficiency and quality of data, corroborative information, and evidence of
causality, and the need for additional analysis.
The interpretation of ecological significance evaluates risk estimates in the context of
the types and extent of anticipated effects. Ecological significance may be defined in terms
of the nature and magnitude of effects, the spatial and temporal patterns of effects, and the
potential for recovery once a stressor is removed. These three aspects of ecological
significance are inextricably linked. A single stressor may cause several effects, or a
particular effects may be affected be several stressors. Therefore it is important ro define the
nature of the assessment or measurement endpoints which are of concern. The magnitude of
effects may be influenced both by ecological context (e.g., reduction in reproductive rate in a
population that reproduces rapidly or slowly) or by spatial or temporal patterns. For example,
a spotted owl requires old growth forests for habitat, and loss of that habitat would have
much greater magnitude for the spotted own than for other species which may have the ability
to adapt to other habitats. Spatial patterns may also be influenced by environmental fate and
transport, and bioaccumulation. Temporal patterns may be influence magnitude of effects due
to the persistence of the stressor as well as how often the stressor occurs, and how it relates
to critical life stages of organisms. The recovery potential for a particular ecological effect
6-9
-------
may be dependent on the nature, duration and, extent of the stressor and may require
additional analyses which should be explicitly discussed in the final risk assessment. As
indicated above, there may be a higher degree of complexity than traditionally found in
human health risk assessment and may require more qualifications in the presentation of risk.
6.1.2 DEFINING ECOLOGICAL RISK ASSESSMENT METHODS
The definition of ecological risk assessment used in this discussion is the qualitative
and/or quantitative appraisal of the potential effects of a pollutant on ecological receptors.
This definition does not include damage assessment which generally applies when the
pollutant has been released in the environment. While some the references used in this
discussion include damage assessment as a part of ecological risk assessment (e.g.,
USEPA/OERR 1989) there are different in that ecologicarrisk assessment is predictive while
damage assessments are after-the-fact assessments of environmental damage. The reader who
refers to those references for more information needs to bear this distinction in mind.
Fate and transport of chemicals can be determined using the same models and methods
used for estimating human exposure in the food chain. The main differences are in the end
points chosen and the methods used to determine risk. The methods used to assess effects are
somewhat dependent on the end points chosen. Unlike human health risk assessment, cancer
is not generally an effect of concern. Acute effects on nonhuman populations include the
death of all or part of a population, increases in undesirable species (such as blooms of blue-
green algae), or reductions in yield of agricultural crops or umber. Two types of assessment
methods exist. Qualitative methods are useful in situations where data and/or resources are
limited. Qualitative methods may be used for a screening level analysis, however, they are
rarely sufficient to support regulatory development.
Most quantitative methods can be classified as either quotient (or ratio) methods or
continuous (or exposure-response methods) (Morten et al. 1988). Barnthouse and Suter (1986)
also include ecosystem uncertainty analysis as a quantitative method. Although
characterization of uncertainty should be a part of any risk assessment, in some cases (e.g.,
where ecosystem effects are the endpomts and a suitable model is available) the uncertainty
analysis itself may be an appropriate assessment method.
6-10
-------
6.1.2.1 Qualitative Methods. A qualitative or semi-quantitative approach is
often used at the screening level of an assessment. At the least, some inventory of the site is
made (or existing studies on the site are used) and important species or site characteristics are
identified. The presence of rare or endangered species or commercially important species
should be noted. Any relevant studies that might help estimate the potential for adverse
effects should be consulted. If sufficient data are available, a hazard assessment should be
conducted. Fate and transport models could then be used to estimate the potential for
exposure. Risk characterization may consist of classifying each chemical into categories
based in a subjective appraisal of the likelihood of an adverse effect occurring. An example
of a categorization scheme is (1) no effect on any species, (2) some effect on sensitive
species, (3) effects on most species, or (4) effects on all species. Categorization schemes may
also be developed for ecosystems depending on the sensitivity of those ecosystems to
stressors. Several qualitative approaches are described in more detail in Norten et al., 1988,
and USEPA/OERR, 1989.
Semiquantitative methods have been adapted to ecological risk assessment from other
disciplines (Barnthouse and Suter 1986; Suter et al., 1987). One of these, fault tree analysis,
is used in engineering safety assessments to determine events and system states that have the
potential to cause the system to fail. If ecological communities are treated as systems, it is
possible to identify potential ecological failures, such as population extinctions, or failure of a
crop to set fruit. However, it is usually impossible even to assign probabilities to the
likelihood of an event occurring.
*
Even when qualitative methods are used, the assessment needs to be performed by
someone who is trained in a relevant discipline. In fact, the absence of good quantitative data
makes the use of an expert even more important when important decisions depend on the
outcome of professional judgment.
6.1.1.2 Quantitative Methods.
Quotient Methods. Although these are considered quantitative methods, they are still
generally used as screening level methods. A benchmark concentration is used to compare
against expected environmental concentrations to assess the possibility of an effect. These
benchmark concentrations are indicative of some toxic endpoint or stressor. The LD50 (the
6-11
-------
dose lethal to 50% of the test population), the LC50 (the concentration lethal to 50% of the
test population), or the chronic no-effect level for a sensitive indicator species are typical
ecotoxicological hazard data. If the ratio of the estimated environmental concentrations
(EEC) to one of these hazard data exceeds certain fixed criteria, a potential adverse effect is
inferred (Bascietto et al. 1990).
Quotient methods are useful for setting standards or priorities, or for identifying the
potential adverse effects for further study. However, quotient methods give only a "risk" or
"no risk" result; they do not give the degree of risk for quotients greater than one. They also
can not be used to estimate the magnitude of effects associated with pollutant concentrations.
A good discussion of the application of the quotient method is found in Barnthouse and Suter
(1986).
Exposure-Response Methods. When an estimate of the magnitude of the risk is
needed, exposure-response or continuous methods are used. These methods use a continuous
curve relating an effect (e.g., reductions in crop yield) to an estimated exposure concentration.
These curves are analogous to dose-response curves used in toxicological studies in human
health risk assessments.
Continuous methods provide much more information to the policymaker or risk
manager. This allows the decision-maker to weigh the costs and benefits associated with
different levels of pollution. The limitation of this method is that it requires more data than
other methods, and that data are often unavailable for many species and ecosystems. In some
cases, concentration-response functions for different life stages of both tested and untested
species can be extrapolated from laboratory tests by analysis of extrapolation uncertainty
(Barnthouse and Suter, 1986).
Ecosystem Uncertainty Analysis. Characterization of uncertainty is especially
important in ecological risk assessment because of the compiexity and variability of the
system being analyzed. The number of unknowns in an ecological risk assessment is
generally much greater than in a human risk assessment. The methods of characterizing
uncertainty used in human risk assessments apply to ecological ones as well.
If reliable ecological models ('See Suter, 1993V exist for the situation of interest, those
models can be used to generate probabilistic estimates of risk. The effects of pollutants or
other stressors on individual organisms (as determined by laboratory or field experiments) can
be extrapolated to populations, trophic levels, or ecosystems by use of ecosystem simulation
6-12
-------
models. Monte Carlo (or other numerical methods) are used to propagate the uncertainties in
the baseline data as well as any uncertainties associated with the model to give probabilistic
estimates of risk for the desired endpoints.
One advantage of this approach is that it allows for assessment of risk on endpoints
for which data may be difficult or costly to obtain. These models allow for the extension of
analysis to the ecosystem population or community levels. Models may also be useful for
assessing the interaction of natural processes such as species competition or predation with
the contaminant effects. The results are expressed as risk with attendant uncertainty, which is
more realistic than single number risk estimates. However, the results are only as good as the
model, and the confidence that can be placed in the model is very important to consider if
this method is chosen. The application of ecosystem uncertainty analysis is described in
detail in Barnthouse and Suter (1986).
6.1.3 MAJOR ISSUES
As scientists and policymakers attempt to define ecological risk assessment and
develop acceptable methods, many issues have been raised (Norten et al. 1988). For the most
part, these issues relate to the lack of information on natural populations and ecosystems
suitable for making important risk management or regulatory decisions.
Most ecological risk assessors choose end points related to population dynamics for
only one of a few of the species that are potentially affected. This is due, in part, to the
significance attached to commercially-important species. However, the choice is frequently
determined by the paucity of data available on community- or ecosystem-level effects.
Scientists recognize that ecosystem response may ultimately be the most important er'fect to
assess, as ecosystem integrity affects ultimately both human and ecological receptors. The
complexity of ecosystems, the difficulty of defining and delimiting an ecosystem, and the
long delays associated with ecosystem responses (decades or centuries) ail contribute to the
difficulty m assessing ecological risks.
Another issue is the lack of sufficient exposure-response data for most species. This
necessitates the use of quotient methods in quantitative risk assessments which provide very
limited information on which to base important regulatory or risk management decisions.
Also, the information needed to treat uncertainty quantitatively is often lacking.
Since most of the information on lethal effects comes from laboratory studies, very
little is known about the interaction of physical and biological processes in nature with
6-13
-------
pollutants. Bioaccumulation plays a critical role in determining the extent of stressor impacts
and introducing potential impacted species not previously considered in laboratory
evaluations. A certain amount of error is associated with extrapolation from laboratory to real
ecosystems and that error is potentially very large. The use of mesocosms (artificial
assemblages of organisms in chambers used to study ecosystem processes such as nutrient
cycling) may help bridge this gap.
A critical issue in ecological risk assessment is what are relevant endpoints of concern
and what level of risks are significant. In human health risk assessment, endpoints (e.g.,
cancer) and level or measure of risk (e.g., 10"6) are usually clearly defined. However, given
our inexperience in ecological risk assessment, the endpoints (e.g., death v. reduced fertility)
and measure of risks (e.g., number of deaths, single deaths v. 150 deaths in a population) are
not clearly defined.
6-14
-------
6.2 CHEMICAL MIXTURES
While some potential environmental hazards involve significant exposure to only a
single compound, most instances of environmental contamination involve concurrent or
sequential exposures to a mixture of compounds that may induce similar or dissimilar effects
over exposure periods ranging from short-term to lifetime (EPA, 1986). The EPA has long
recognized the importance of chemical mixtures in environmental risk assessment. When
EPA developed their first set of risk assessment guidelines in 1986, chemical mixtures was
one of five topics for which guidelines were developed (EPA, 1986).
Mixtures have been defined in the EPA Guidelines for the Health Risk Assessment of
Chemical Mixtures as "any combination of two or more chemical substances regardless of
source or spatial or temporal proximity." In some cases, mixtures may be highly complex
consisting of scores of compounds that are generated simultaneously from a single source or
process (e.g., coke oven emissions and diesel exhaust). In other cases, mixtures may consist
of related compounds produced by commercial products (e.g., PCBs, gasoline, and pesticide
formulations). Other mixtures may consist of unrelated chemicals (chemically or
commercially) which are placed in proximity to one another and eventually may mix and
released into the environment (e.g., hazard waste disposal sites).
The quality and quantity of pertinent information available for risk assessment varies
considerably for different mixtures. Some mixtures are well characterized or have been well
studied, and have well-defined levels of exposure and toxicoiogical properties well defined.
However, in most cases there is limited information on the mixture especially if the
composition of the mixture is in question or toxicologic data on its constituents are iirmted.
As a result, how risk assessments are conducted can be anticipated to vary widely from
mixture to mixture with a high degree of case-by-case decision-making. The EPA guidelines
present a scheme for overall method selection and analysis depending on the nature and
quality of the data. In addition, the EPA has developed the Technical Support Document on
Risk Assessment of Chemical Mixtures (EPA, 1988) which contains a thorough review of
available information on the toxicity of chemical mixtures and a discussion of research needs.
The major points of these documents are summarized below.
6.2.1 Overall Approach
No single approach can be recommended for risk assessments of multiple chemical
exposures. However, general guidelines can be recommended depending on type of mixture,
6-15
-------
known toxic effects of mixture components, availability of toxicity data on the mixture or
similar mixtures, interactions among components, and exposure data. Given the variety of
data available for mixtures, emphasis must be placed on flexibility, judgement, and a clear
articulation of the assumptions and limitations in any risk assessment (EPA, 1986). As a
result, the EPA has proposed an approach to method selection, summarized in Figure 6-2,
A general hierarchy of analysis has been suggested when conducting risk assessments
on chemical mixtures. When possible, the risk assessment should be based on the mixture of
concern. Some highly complex mixtures that are generated in large quantities (e.g., coke
oven emissions, diesel exhaust) have been well studied using toxicity tests conducted directly
on the mixture. If no data are available for a particular mixture, the risk assessment should
focus on data available from similar mixtures. Mixtures can be considered similar if they
have the same components but in slightly different ratios, or if they have several common
components but lack one or more additional components, or have one or more additional
components. Whether mixtures are "sufficiently similar" to justify using the data must be
decided on a case-by-case basis, considering not only the uncertainty of using a data from a
dissimilar mixture but also the uncertainties using other approaches such as component
additivity. Consideration should be given to any information on the components that differ or
are contained in markedly different proportions. Particular emphasis should be placed on any
toxicologic or pharmacokinetic data on the components of the mixture that would be useful in
assessing the significance of any chemical differences between the mixtures. If no data are
available for the mixture of concern or for any similar mixtures, then the risk assessment
should focus on evaluating the mixture components. Consideration should be given to
potential interactions between components (see Section 6.2.3). If no data are available for
any of the above approaches, then no quantitative risk assessment can be conducted.
An alphanumeric classification scheme has also been developed for ranking the quality
of data used in risk assessment and the overall quality of the risk assessment. This scheme is
outlined in Table 6-2.
6.2.2 Existing Methods
Typically, data are not available on an identical or reasonably similar mixture.
Therefore, most of the commonly used methods focus on analysis of individual constituents
and assume dose additivity. Dose additivity is based on the assumption that the components
in the mixture have the same mode of action and elicit the same effects. In actuality,
6-16
-------
1. Assess quality of data using Table 2.
I. Interactions
((.Health Effects
III. Exposure
Adequate
2. Data on mixture of concern?
3. Risk assessment using data
on mixture of concern.
Inadequate
4. Data on similar mixture?
5.Mixtures sufficiently similar?
6. Risk assessment using data
on similar mixtures
12. Compare risk assessment from
steps 3.6. 10.11 as appropriate
Identify preferred assessment
14 Qualitatively assess hazard
No quantitative risk assessment.
7 Data cm mixture components
8. Indices of acceptability and
risk based on component data.
9 Sufficient information to
quantify interactions?
1! Y
10 Risk assessment with interactions
quantified where appropriate
Use addrtivity for all components.
Optional
11. Risk assessment
for all components
using additivity
13. Develop integrated summary including
discussion of uncertainties.
Figure 6-2. FLOW CHART FOR RISK ASSESMENT OF CHEMICAL MIXTURES
6-17
-------
Table 6-2. CLASSIFICATION SCHEME FOR THE QUALITY OF THE RISK ASSESSMENT OF THE MIXTURE
Classification Description
Information on Interactions
I Assessment is based on data on the mixture of concern.
II Assessment is based on data on a sufficiently similar mixture.
Ill Quantitative interactions of components are well characterized.
IV The assumption of additivity is justified based on the nature of the health effects and on the number of
component compounds.
V An assumption of additivity cannot be justified and no quantitative risk assessment can be conducted.
Health Effects Information
A Full health effects data are available and relatively minor extrapolation is required.
B Full health effects data are available but extensive extrapolation is required for route or duration of
exposure for species differences. These extrapolations are supported by pharmacokinetic considerations,
empirical observations, or other relevant information.
C Full health effects data are available but extensive extrapolation is required for route or duration of
exposure for species differences. These extrapolations are not directly supported by information available.
D Certain important health effects data are lacking and extensive extrapolations are required for route or
duration of exposure of for species differences.
E A lack of health effects information on the mixture and its components in the mixture precludes a
quantitative risk assessment.
Exposure Information
1 Monitoring information either alone or in combination with modeling information is sufficient to accurately
characterize human exposure to the mixture or its components.
2 Moaenng information is sufficient to reasonably characterize human exposure to the mixture or its
components.
3 Exposure estimates for some components are lacking, uncertain, or variable. Information on health
effects or environmental chemistry suggest that this limitation is not likely to substantially affect the risk
assessment.
4 Not all components in the mixture have been identified or levels of exposure are highly uncertain or
variable. Information on health effects or environmental chemistry is not sufficient to assess the effect of
this limitation on the risk assessment.
5 The available exposure information is insufficient for conducting a risk assessment.
6-18
-------
compounds have the ability to interact or have different receptors or modes of action.
Interactions will be discussed in greater detail in Section 6.2.3. While dose additivity may
not be the most biologically plausible approach, several studies have demonstrated that dose-
additive models often predict reasonably well the toxicities of mixtures composed of a
substantial variety of both similar and dissimilar compounds. Basic methods to assess
mixtures have been developed for both noncarcinogens (systemic toxicants) and carcinogens.
Most methods used in the Agency focus on the dose additivity of individual constituents. The
Office of Emergency Response and Remediation (OERR), responsible for Superfund cleanups,
has modified this approach to include only a subset of the chemicals in the mixture to act as
indicator compounds. For each mixture the analysis if noncarcinogen and carcinogens is
carried out independently.
6.2.2.1 Noncarcinogens. The approach to assess noncarcinogenic mixtures is based
•s*
on the current methodology used by the Agency for single compounds, i.e., the derivation of
an exposure level that is not anticipated to cause significant adverse effects. The level may
be expressed in a variety of ways depending on the route of exposure, media of concern, and
legislative mandate. As described in Section 3.0, the noncancer risk standard for the
inhalation route is the Reference Concentration (RfC). As described in Section 5.0, risks for
noncarcinogens are evaluated by dividing the exposure level by the RfC, with values greater
than one indicating an exceedance of the Hazard Index (HI). The HI has been developed to
accomplish the same function for mixtures that the RfC does for single compounds. The HI
of a mixture, based on the assumption of dose addition, is defined as:
HI = E,/RfC, + Ej/RfC, + ... + E/RfC,
Where:
E, = exposure level of the ith toxicant, and
RfC. = Reference concentration for the ith toxicant
The HI provides a rough measure of likely toxicity, not a direct measure of incidence. -As
with single chemicals, an HI below 1 is not anticipated to result in adverse health effects. As
the HI approaches 1 concern for the potential hazard of the mixture increases. If the HI
exceeds I, the concern is the same as if an individual chemical exposure exceeded the RfC by
the same level. His exceeding 1 do not necessarily imply that adverse health effects will
occur, only that the potential exists and the likelihood of an effect increases as the HI
increases. The HI is not a direct estimate of risk as it does not define a dose-response
6-19
-------
relationship, or yield risk estimates as probability, nor is there, a strict delineation of "safe"
and "unsafe" levels.
Dose addition, that mixture components induce the same effect by similar modes of
action. Ideally, a separate HI should be generated for each endpoint to support this
assumption. In cases where data clearly indicate that dissimilar modes of action and effects
exist, then the use of the HI should be questioned. In any event, if a hazard index is
generated, the quality of the experimental evidence supporting the assumption of dose
addition should be explicitly discussed as part of the risk estimate.
6.2.2.2 Carcinogens. For carcinogens, the increase in risk can be assumed to be
additive whenever linearity of the individual dose-response curves has been assumed. This is
usually restricted to low doses such as those used with the linearized multistage model. The
risk from simultaneous exposures to several carcinogens in such a mixture can be estimated
>••
from the following equation:
P = Idft
where:
P = excess cancer risk
' dj = exposure level of the i* component
B; = carcinogenic potency of the im component
This equation assumes an independence of action and is assumed to be equivalent to
assumption of dose addition. The equation is an approximation of a more precise equation
for combining risks described in detail in 51 FR 34014. The precise equation is consistent
with the assumption of dose additivity and accounts for the joint probabilities of the same
individual developing cancer as a result of exposure to two or more carcinogens (See EPA.
1986, 51 FR 34014). The difference between equation 5-1 and the precise equation are
negligible for total individual cancer risks less than 1 x 10"1 (or 0.1). (EPA/OSW, 1989) The
•EPA Guidelines for Chemical Mixtures (EPA, 1986, 51 FR 34014) and EPA/ECAO (1988)
provide further detail on mathematical models for multiple chemical risk estimation.
The risk summation technique assumes exposures are in the low-dose range where
responses are linear. At higher risk levels, nonlineanty may need to be considered. The
approach also assumes independence of action by the compounds involved (i.e., that there are
no synergistic or antagonistic chemical interactions and that all chemicals produce the same
effect, i.e., cancer through independent mechanisms of action). If these assumptions are
6-20
-------
incorrect, over- or under-estimation of the actual multiple-substance risk could result
(EPA/OSW, 1989).
There are several other limitations to this approach that must be acknowledged. First,
because each slope factor or URE is typically an upper 95th percentile estimate of potency,
and because upper 95th percentiles of probability distributions are not strictly additive, the
total cancer risk estimate might become artificially more conservative as risks from a number
of different carcinogens are summed. If one or two carcinogens drive the risk, however, this
problem is not of concern. Second, it often will be the case that substances with different
weights of evidence for human carcinogenicity are included. The cancer risk equation for
multiple substances sums all carcinogens equally, giving as much weight to class B or C as to
class A carcinogens. In addition, slope factors or URE's derived from animal data will be
given the same weight as slope factors derived from human data (EPA/OSW, 1989).
6.2.3 Interactions
Those methods described above assume dose additivity, that the toxic action of the
mixture as a whole is equal to the sum of the toxicity of its individual components.
However, the possibility exists that interactions may occur. Compounds may interact and
result in synergistic or antagonistic interactions. A synergistic interaction is a pharmacologic
or toxicologic interaction in which the compounds reinforce or magnify the toxic effect. The
resulting toxicity is much greater than would be predicted by the sum of the individual
toxicities. In an antagonistic interactions are those where a compound may interfere with or
cancel the toxic activity of another compound and the resulting toxicity is much less than
would be predicted by the sum of the individual component toxicities. Data should be
reviewed to determine if there are known interactions between any of the compounds in the
mixture.
Most of the data available on toxicant interactions are derived from acute toxicity
studies using experimental animals exposed to mixtures of two compounds, often in only a
single combination. The use of information from two-component mixtures to assess
interactions in mixtures containing more than two components is difficult from a mechanistic
perspective. Studies of mixtures with more than two chemicals are rare and difficult to
interpret. The EPA's Environmental Criteria and Assessment Office in Cincinnati has
developed a computerized database and data retrieval system called MIXTOX, which contains
summaries of available literature on toxicologic interactions between environmental chemicals
6-21
-------
(EPA, 1992). The database reflects the literature published prior to 1988. Once a
determination of potential interactions has been made, the relevance of the data to chronic
exposures should be assessed, especially with regard to the use of acute data and the presence
of other compounds in the mixture.
6.2.4 Other Mathematical Models
The common methods described above use simple additivity to assess the risks from
chemical mixtures. More detailed models can be used to describe or quantify toxicity of
chemical mixtures. Two basic types of models are available, the dose addition model and the
response addition model. The dose addition model assumes that toxicants in a mixture
behave as if they were dilutions of one another, that the true slopes of the dose-response
curves for individual compounds are identical, and that the response elicited by the mixture
can be predicted by summing the individual doses after adjusting for differences in potency.
The adjustment for differences in potency is defined as the ratio of equitoxic doses. Probit
transformation, introduced in Section 3, typically makes this ratio constant at all doses when
parallel straight lines are obtained. This model is based on the assumption that each
compounds behaves similarly in terms of mechanism of action and target site.
The other model, response addition, assumes that the two toxicants act on different
receptor systems and that the correlation of individual tolerances may range from completely
negative to completely positive. Response addition assumes that the response of a given
concentration of a mixture of toxicants is completely determined by the response to the
components and the pairwise correlation coefficient.
Detailed descriptions and mathematical forms can be found in the Guidelines for
Chemical Mixtures, 51 FR 34021-34022. Each of the above models assume no interactions.
However, in cases where quantitative data are available for interactions, any of the
mathematical models described above can be modified to reflect the interaction, either
synergisms or anatagomsms. Descriptions of mathematical modifications can also be found in
the Guidelines.
6.2.5 Uncertainties
As with ail risk assessments, the uncertainties should be clearly discussed and the
overall quality of the risk assessment should be classified. As described above in Table 6-1,
an overall classification scheme has been developed to classify the quality of a risk
assessment on chemical mixtures. There are several categories for sources of uncertainty,
6-22
-------
including health effects, exposure, and the composition of the.mixture. In many cases, data
will not be available for mixtures of concern or similar mixtures and the assessment must rely
on data for individual compounds. An assumption of dose additivity is typically used unless
there are data on interactions. Interactions among chemicals are likely though information on
the type and magnitude are typically not available. This has been described above in some
detail and it should be explicitly discussed. Other sources of uncertainty include exposure
and composition uncertainties. General uncertainties related to exposure have been described
in Section 4. However, for chemical mixtures these uncertainties may be increased as the
number of compounds of concern increases. If the chemicals in a mixture are not from a
single source or have different chemical and physical properties, then environmental fate,
transport, and environmental concentrations may vary with time. Therefore, care must be
taken in estimating exposure levels for each component and any differences should be
incorporated into the dose-response assessment as well. Furthermore, in many cases the
identity of all constituents in a mixture may not be known or may change with time due to
changes in the source. As a result, unidentified individual components or those without
quantified exposure levels may represent significant risk. When possible, efforts should be
made to identify all components, and conduct a hazard assessment on each of these
compounds. However, when data are limited, major constituents can be used in the risk
assessment, as long as the uncertainties of unidentified components should be explicitly
discussed.
6-23
-------
6.4.1 References For Section 6.1. Ecological Risk Assessment
Barnthouse, L. W. and G. W. Suter, II (eds.). 1986. User's Manual for Ecological Risk
Assessment. Oak Ridge National Laboratory, Oak Ridge, TN. ORNL-6251.
Bascietto, J., D. Hinckley, J. Plafkin, and M. Slimak. 1990. Ecotoxicity and ecological risk
assessment. Environ. Sci. Technol. 24:10-15.
Norten, S., M. McVey, J. Colt, J. Durda, and R. Hegner. 1988. Review of Ecological Risk
Assessment Methods. Prepared for Office of Policy Planning and Evaluation, USEPA,
Washington, D.C. EPA/230-10-88-041.
Suter, G.W., II,L.W. Barnthouse, S.M. Bartell, T. Mill, D. Mackay and S. Paterson.
Ecological Risk Assessment. Lewis Publishers, Chelsea, MI
Suter, G. W., II, L. W. Barnthouse, and R. V. O'Neill. 1987. Treatment of Risk in
Environmental Impact Assessment. Environ. Management, 11:295-303.
US Environmental Protection Agency, 1989. Risk Assessment Guidance for Superfund,
Volume II. Environmental Evaluation Manual. Office of Emergency and Remedial
Response, Washington, D.C. EPA/540/1-89/001.
US Environmental Protection Agency, 1990. Reducing Risks: Setting Priorities and Strategies
for Environmental Protection., Science Advisory Board, Washignton, DC.
U.S. EPA, 1990: Bioaccumulation of Selected Pollutants in Fish. A National Study, Vol. II.
Office of Water Regulations and Standards (WH-552), Washington, DC. EPA 506/6-90/00 Ib.
US Environmental Protection Agency, 1991, Summary Report on Issues in Ecological Risk
Assessment, Risk Assessment Forum, Washington, DC., February 1991, EPA/625/3-91/018.
U.S. Environmental Protection Agency, 1991: Technical Support Document for the
Determination of the Need to Regulate Pulp and Paper Mill Sludge Landfills and Surface
Impoundments.
US Environmental Protection Agency, 1992a, Framework for Ecological Risk Assessment,
Risk Assessment Forum, Washington, DC., February 1992, EPA/630/R-92/001.
US Environmental Protection Agency, 1992b, Report on the Ecological Risk Assessment
Guidelines Strategic Planning Workshop, Risk Assessment Forum, Washington, DC., February
1992, EPA/630/R-92/002.
US Environmental Protection Agency, 1992c, Peer Review Workshop Report on a Framework
for Ecological Risk Assessment, Risk Assessment Forum, Washington, DC., February 1992,
EPA/625/3-91/022.
6-24
-------
U.S. Department of Commerce, NTIS, 1993: Derivation of Proposed Human Health and
Wildlife Bioaccumulation Factors for the Great Lakes Initiative. PB93-154672.
U.S. EPA, 1993: Interim Report on Data and Methods for Assessment of 2, 3, 7, 8-
Tetrachlorodibenzo-p-dioxin Risks to Aquatic Life and Associated Wildlife. Office of
Research and Development Washington, DC. EPA/600/R-93/055.
U.S.EPA, 1993: Great Lakes Water Quality Initiative Criteria Documents for the Protection
of Wildlife (Proposed) - DDT, Mercury; 2,3,7,8-TCDD, PCBs. Office of Water (WH-586),
Office of Science and Technology, Washington, DC. EPA-822-R-93-007.
U.S.EPA, 1993: Proposed Water Quality Guidance for the Great Lake System. Federal
Register, Vol 58, No. 72. RIN 2040-AC08 [FRL 4205-5] Proposed Rules.
U.S.EPA, 1993: Wildlife Criteria Portions of the Proposed Water Quality Guidance for the
Great Lakes System. Office of Water (WH-586), Office of Science and Technology,
Washington, DC. EPA--822-R-93-006.
6.4.2 References for Section 6.2, Chemical Mixtures
U.S. Environmental Protection Agency, 1986. Guidelines for the Health Risk Assessment of
Chemical Mixtures, Office of Research and Development, 51 FR 34014-34025.
U. S. Environmental Protection Agency, 1992. MDCTOX Version 1.5, An Information
System on Toxicologic Interactions for the MS-DOS Personal Computer, Environmental
Criteria and Assessment Office. Cincinnati, Ohio.
U. S. Environmental Protection Agency, Environmental Criteria and Assessment Office.
November 1988. Technical Support Document on Risk Assessment for Chemical Mixtures.
Final Draft, ECAO-CIN-572.
6-25
-------
-------
7.0 USES OF RISK ASSESSMENT METHODOLOGIES
7.1 INTRODUCTION
The objective of this chapter is to illustrate the selection and use of the risk
assessment methodologies described in this report. Performing a risk assessment may be
fairly simple or it may be extremely complicated. Risk assessments differ greatly from one
another in scope, depth, approach, and level of uncertainty. The approaches chosen will
depend on the goals of the risk assessment, the level of uncertainty that is acceptable, and the
time, resources, and expertise available. As described in Chapters 2 through 5, there is a
wide range of methods for performing each step of a risk assessment. It would be impossible
to describe exactly when and how one might use each of the methods presented. Therefore,
this chapter illustrates the selection and use of various methods by presenting three example
case studies.
The case studies are hypothetical, but were designed to reflect realistic situations in
which States may use risk assessments to reach decisions on control of toxic air emissions.
The example case studies include a range of methodologies from simple screening approaches
to very detailed approaches. Case I describes a screening study to create a relative ranking of
industries in order to prioritize further risk assessment and regulatory activities. The ranking
is based on a rough estimate of the maximum cancer risks posed by the higher emitting plants
within each industry group. Case II presents a more refined estimate of risk which uses more
site-specific information and estimates aggregate population risk as well as maximum
individual risk. Noncancer risks are also addressed. Case III illustrates a very detailed site-
specific risk assessment considering multiple pollutants and indirect as well as inhalation
exposure pathways. This type of assessment might be used in a permitting decision. The text
of each case study discusses why particular methods were chosen and how the goals of the
risk assessment and the level of uncertainty deemed acceptable influenced the choice of
methods.
Each case study is organized into several sections. The first section describes the
objectives and scope of the study. The sections that follow describe each of the four steps of
risk assessment — hazard identification, dose-response assessment, exposure assessment, and
7-1
-------
risk characterization. Within each of these sections there is a discussion of the methods used
and why they were selected, the type of output obtained, and the level of expertise and
resources needed. Finally, there is a short section identifying a few of the possible extensions
or alternative methods that could have been used in the case study. It should be stressed that
there is no single correct way to conduct a risk assessment, and these case studies are not
intended to prescribe the specific approach that should be used in a given situation. Rather
they are to illustrate how methods could be selected and used.
7-2
-------
7.2 CASE I
7.2.1 Study Objectives
The purpose of this example case study was to rank the relative cancer risks posed by
various industrial source categories within a particular state. The results of the ranking were
intended to provide a conservative basis for prioritizing the industrial source categories for
further, detailed risk assessment and potential state regulation. Thus, the level-of-effort and
resource expenditure for this study were desired to be kept at a minimum. The result of the
study was a table showing the source categories ranked relative to one another by scores of
estimated maximum cancer risk. To additionally aid decision-makers in setting priorities, the
number of facilities in and total emissions for each source category were also presented.
Since the objective of the study was to estimate the source category rank order,
emphasis was placed on using a consistent, conservative methodology for assessing the
relative risks, without regard to the actual values predicted. The actual values generated
incorporate a high degree of uncertainty, due to the screening nature of the study. This level
of uncertainty was judged to be acceptable, since the study would only be used to prioritize
source categories. More detailed risk assessments with lower uncertainties would be
conducted for high priority sources before developing any potential regulations.
7.2.2 Scope
A simple screening technique for estimating the relative risks associated with the
various source categories would achieve the study objectives without requiring large amounts
of time, resources and expertise. Information on source category emissions and emission
release characteristics and on pollutant toxicity was obtained from resources that are readily
available. Given that the ranking was only intended to provide a basis for establishing
priorities, a high level of accuracy was not needed. For the purpose of this case example, it
was decided that the dispersion technique and emissions estimates should be conservative in
order to avoid missing source categories that could impact public health. Screening
dispersion modeling was conducted to differentiate the impacts of point and fugitive (area)
sources.
The source categories involved in the study were identified as those of primary
Standard Industrial Classification (SIC) codes 20 through 39. Facilities with these SIC codes
7-3
-------
are those for which toxic chemical release inventory data are reported under Section 313 of
the Emergency Planning and Community Right-to-Know Act. Only facilities located within
the state boundary were included in the assessment. Emissions estimates for both point and
fugitive sources were used, based on the 1989 reporting year only.
The exposure media was assumed to be ambient air. Indirect exposure pathways were
not considered. Chemicals identified as carcinogens by EPA were included in the study.
7.2.3 Selection and Use of Risk Assessment Methods
7.2.3.1 Hazard Identification .
Methods. This study considered only those chemicals for which EPA had already
performed a hazard identification. Only those chemicals classified as human carcinogens
(Class A) or probable human carcinogens (Classes B1 and B2) were included. Carcinogenic
classification was obtained from the EPA Integrated Risk information System (IRIS). The
IRIS data base can be searched to isolate those chemicals for which some carcinogenicity
information is available. Since the objective of the study was to rank source categories by
maximum cancer risk, noncancer health effects were not considered.
It was decided not to search other literature sources or perform hazard identifications
for chemicals not yet reviewed by EPA or included in IRIS. Further information gathering
and assessment would have required more level of resources and scientific expertise than
were available for the study. Also, the State considered it reasonable to concentrate first on
regulation of those chemicals already identified as carcinogens.
Output. The output of the hazard identification step was a list of chemicals along with
their weight-of-evidence classification (i.e.. Class A, Bl or B2, as defined in Chapter 2 of this
report).
Expertise and Resources. Since a State computer account with the National Library of
Medicine already existed, the IRIS data base was readily accessed via personal computer
communications software. (The information contained in the data base is also available on
hard copy through the National Technical Information Service.) This step and the dose-
response step that follows were performed by a health scientist in approximately one day.
7-4
-------
7.2.3.2 Dose-Response Assessment.
Methods. To determine maximum cancer risk, dose-response information for the
carcinogenic chemicals was required. The IRIS data base was accessed to obtain the
inhalation unit risk estimates for each chemical. The unit risk estimate (URE) represents the
increased cancer risk from a lifetime (70 year) exposure to an ambient air concentration of 1
ug/m3. The IRIS data base contains updated UREs and hazard identification information and
is part of the National Library of Medicine's Toxicology Data Network (TOXNET).
Output. The output of the dose-response step was a listing of carcinogenic chemicals
and their associated UREs and weight-of-evidence classifications. A sample of IRIS output
pertaining to the carcinogenicity of a given pollutant is provided in Appendix A.
Expertise and Resources. As mentioned in the hazard identification step, the IRIS data
base was readily accessed via personal computer communications software, since a State
computer account with the National Library of Medicine already existed. (Information
contained in the data base is also available on hard copy through the National Technical
Information Service.) This step and the hazard identification step were performed by a health
scientist in approximately one day.
7.2.3.3 Exposure Assessment.
Emission Characterization. Point and area (fugitive) source emission rates were
obtained through the Toxic Chemical Release Inventory System (TRIS), which is also
available through the National Library of Medicine's Toxicology Data Network (TOXNET).
Data contained in the inventory are submitted to EPA by industrial facilities on a yearly basis,
in compliance with Section 313 of the Emergency Planning and Community Right-to-Know
Act (Title III of the Superfund Amendments and Reauthorization Act of 1986), Public Law
99-499. Non-point (area) and point (stack) air emissions are required to be reported under the
rule.
Point source emissions in TRIS are reported as the total of all releases to the air that
are released through stacks, vents, ducts, pipes, or any other confined air stream. Air
emissions not so released are considered fugitive emissions, and include: fugitive equipment
leaks, evaporative losses from surface impoundments and spills, releases from building
ventilation systems. Each release estimate is accompanied by a letter indicator of the
7-5
-------
principal method used to determine the amount of release reported. The four codes used
represent the following methods: monitoring, mass balance calculations, emission factors and
engineering calculations. Only a single code is entered per emissions estimate. Therefore,
the code identifies the method applying to the largest portion of the total estimated release
quantity. The emission estimation techniques (and associated uncertainties), while varying
from plant-to-plant, were assumed to be consistent between standard industrial source
categories as a whole.
The Toxic Chemical Release Inventory was searched to isolate the facilities located
within the State of interest. Reports submitted for calendar years 1987, 1988 and 1989 may
exhibit reporting of emissions in ranges, for releases to an environmental medium (e.g., air)
that are less than 1,000 pounds for the year. Three ranges are specified: 0, 1-499 and 500-
999 pounds/year. Emissions for those facilities reporting a range in emissions and not a
specific emissions estimate for the year were assumed to be equal to the midpoint of the
range indicated.
Facility emissions were entered into a computerized spreadsheet and sorted by
emission rate within each SIC. The facility with the highest point emissions and with the
highest area emissions of each chemical were identified within each SIC code. Since the
ranking was to be based on maximum risk by SIC, assessments of every facility within an
SIC was not performed. Given the methodology used in the exposure modeling (see the
following steps), facilities emitting smaller amounts would be predicted to produce lower
exposures than those emitting larger amounts. Therefore, the higher emitting sources were
selected for dispersion modeling. It should be noted, however, that emissions alone are not
an accurate measure of risk. Ground level concentrations may be affected by local
meteorology and dispersion characteristics. Exposure may be affected by the proximity of
populations to the facility fenceline or maximum ground level concentration. It may be
possible that the highest emitting facility may not result in the maximum risk. Therefore.
several of the top emitters should be assessed and then compared.
In order to model the facilities, as described in the fate and transport step,
specification of the following point and area emission release parameters was required.
7-6
-------
Emission Release Parameters
Point Sources Area Sources
Stack height Release height
Stack inner diameter Length of side of square area
Stack gas exit velocity
Stack gas exit temperature
Identifying a single representative set of emission release parameters for each SIC
code is a difficult undertaking, and unquestionably introduces more uncertainty in the study
results. The emission release parameters indicated above differentiate and dictate how the
pollutant initially disperses in the atmosphere, and greatly influence the prediction of ambient
concentration to which people are assumed to be exposed. Determining a representative set
of release parameters is complicated by the fact that a wide range of source characteristics
exists within a given SIC code. For example, the majority of facilities within SIC code "A"
could be typified by short stacks and iow-to-moderate gas exit velocities. The average value
of stack height and gas exit velocity calculated over the population of facilities within that
SIC code would reflect these characteristics. While facilities with these stack parameters
could be the most numerous, it is possible that they may account for only a small percentage
of the total emissions within the SIC. If the stack heights of facilities with high emissions are
much taller than the average value selected, the impacts of these facilities may be greatly
overestimated, and consequently bias the source category ranking. Essentially, if dispersion
characteristics are to be introduced into the assessment, representative emission release
parameter data of good quality are required in order to produce meaningful results.
For the purpose of this case study, it was assumed that source parameters by SIC code
were obtained from the State's computerized emission inventory system. Since it was not
7-7
-------
intended that the unique characteristics of each facility be modeled in this case study, a single
set of worst case (i.e., producing the worst atmospheric dispersion) release parameters per SIC
code was determined.
Another product of the emissions characterization step was the total number of
facilities for the State, and the total emissions by chemical for each SIC. While this
information was not used in the modeling, it was presented in the final report.
Fate and Transport Analysis. Pollutant transport was estimated with the EPA
SCREEN model. As its name implies, the SCREEN model is a screening technique that is
used for estimating the air quality impact of stationary sources conservatively. Selection of
the SCREEN model was therefore consistent with the objectives of the study; furthermore it
is a model that is readily available and easy to use.
For receptors in simple terrain (terrain below stack top), SCREEN predicts ambient air
concentrations using standard bi-variate Gaussian dispersion model assumptions. No chemical
fate mechanisms are employed. Facility emissions were converted to units of grams/second
for input to the model.
The highest emitting point and area sources associated with each SIC were modeled in
SCREEN. For the purpose of this study, the following set of options were selected for
modeling each SIC facility.
SCREEN Model Options
Simple Terrain
Full Meteorology
Rural Mode
No Building Downwash
Automated Receptor Array: 200 m to 2000 m
In actuality, the terrain and land use classifications and occurrence of building
downwash will vary from facility to facility. Specification of these variables on a facility-by-
facility basis was beyond the scope of this study; hence they were kept constant. Predicted
7-8
-------
concentrations therefore reflect the magnitude of the emission rate and the emission release
parameters.
Population Characterization. No population characterization was required for this
study, as the source category ranking was based on maximum risk to the individual.
Exposure Calculation. For the modeled facilities within each SIC, the highest
predicted concentration was identified. An individual was simply assumed to be exposed to
this maximum ambient air concentration.
Output. The final products of the exposure assessment step were the rough estimates
of maximum carcinogenic pollutant concentration for the facilities with the highest point and
area emissions within each SIC. In the risk characterization step, these values were used to
estimate maximum risks to individuals for each SIC.
Other outputs included information on the number" of facilities and the estimated
quantity of emissions for each facility, sorted by SIC. The total emissions for each SIC were
also calculated.
Expertise and Resources. Accessing the TRIS data base required a personal computer
with communications software and a computer account with the National Library of Medicine.
The selection and determination of facilities with the highest emissions within each SIC was
accomplished by a permit engineer/data base specialist within a couple of days.
Determination of the representative emission release parameters for each SIC was performed
by a permit engineer/data base specialist and a scientist familiar with air quality modeling and
was performed within approximately two weeks. The SCREEN modeling and summarization
of SCREEN results was performed within a week by a scientist familiar with air quality
modeling. Modeling was performed on a personal computer.
7.2.3.4 Risk Characterization.
Methods. The maximum predicted concentrations from SCREEN were multiplied by
the respective pollutant UREs to obtain indicators of maximum cancer risk. Although UREs
are associated with long-term (annual) average concentrations, they were multiplied by the
hourly average concentrations from SCREEN for the purpose of this case study. Since the
objective of the study was to provide a relative ranking of the source categories, the actual
7-9
-------
value of the risk estimate was not important. Uniformly adjusting the hourly average
concentrations to annual averages would not change the relative ranking.
To produce a maximum cancer risk score for each SIC, the maximum point and area
source concentrations were summed and multiplied by the appropriate UREs. This provides
an indication of maximum risk, but due to the many simplifying assumptions and high degree
of uncertainty in the exposure assessment, this value should not be used as an estimate of
actual risk, but only be used for the purpose of ranking. This value was then scaled by the
total of all SIC maximum cancer risk estimates, as follows.
For each carcinogen:
Indicator of
1.1 (Max Pt Src Cone) + (Max Area Src Conc)l x URE = Maximum Risk
1 (ug/m3) (ug/m3) J (ug/m3)-1 for SIC A
Indicator of 1 Sum of Maximum Risk
2. I Maximum Risk J -^ Indicators for all SICs = Cancer Risk Score
for SIC A
Example calculations for arsenic and benzene are shown in Tables 7-1 and 7-2 at the back of
this case study. Note that generic SIC codes were used in the tables, which are for
illustrative purposes only.
This scheme produced a source category ranking for each pollutant, as illustrated in
Table 7-3. To obtain a cumulative ranking, the maximum cancer risk indicators for all
carcinogens were summed for each SIC and then scaled by the sum of maximum cancer risk
estimates over ail carcinogens and SIC codes, as follows.
Sum of Maximum Risk Sum of Maximum Risk Cumulative
Indicators for all I ^ { Indicators for all 1 = Cancer Risk
carcinogens for J I carcinogens and I Score
SIC A ail SICs
A sample format of the cumulative ranking table is given in Table 7-4.
7-10
-------
Output. A report of the results of this case study would include a summary table
indicating the top ranked SIC for each individual pollutant and for all pollutants collectively.
The entire focus would be on ranking and hence on the cancer risk scores produced, rather
than on the estimates themselves. The accompanying discussion would clearly state the scope
and limitations of the study and would indicate how the results might be used to define a
strategy for further, more refined risk assessment. It should be emphasized, for example, that
the ranking was based only on pollutants classified as either human carcinogens or probable
human carcinogens. A particular SIC may rank low with regard to its emissions of
carcinogenic pollutants, but may rank high with respect to its potential emission of highly
toxic (noncarcinogenic) pollutants.
Aside from the ranking based on maximum cancer risk, the study report should also
indicate, on an SIC basis, the total number of facilities and total emissions of each chemical.
The UREs of each chemical should be listed to indicate their relative toxicities. A State
could decide to regulate sources with high total emissions (statewide) or to regulate all
sources of a chemical with UREs of a certain magnitude, even if those sources did not rank at
the top based on the modeled maximum individual risk.
The major sources of uncertainty in the case study should be reported. In this
example, a high degree of uncertainty was considered acceptable, given the study objectives.
Generally, any assumption made in the methodology will introduce a measure of uncertainty.
Uncertainty in the chemical carcinogemcity is indicated by the weight-of-evidence
classification as is given in the hazard identification step. Uncertainties associated with the
UREs are discussed in Chapters 2 and 5 and include, for example, extrapolation of toxicity
data from animais to humans and from high experimental doses to low doses encountered in
the ambient air. In this case, uncertainty exists in the accuracy of the emissions estimates
from TRIS and in the assumption of emissions from facilities reporting their emissions in
ranges.
Uncertainty is introduced in the modeling by the selection of model options and by
selecting only the facility, from each SIC, with the highest point and area source emissions
for input. The degree to which facility emissions vary within an SIC should be identified
with simple statistics. A high degree of uncertainty is associated with the screening model
7-11
-------
results, given the selection and use of a single set of point and area source emission release
parameters for all SICs in the study. The available source parameter data for each SIC
should be reviewed to assess the variability within and between each SIC.
This report would be given to decision-makers within the State agency for their use in
prioritizing sources for further risk assessment and regulation.
Expertise and Resources. Risk calculations (exposure x URE) were done by a scientist
on a personal computer with spreadsheet software. Chemical specific and summary tables
were also generated by computer spreadsheet. Report preparation, including drafting
discussions of the methodologies used, results and uncertainties, was performed by the various
persons responsible for the study. The report was subsequently reviewed and edited by
supervisory personnel. The total time for performing the risk characterization, reporting and
review was approximately two months.
7.2.4 Other Considerations
Several possible variations to the methods described in this case study could be
implemented. These would generally require additional resources and level-of-effort. Some
of the potential options are given below.
• Instead of conducting screening dispersion modeling, emissions could be multiplied
directly by the respective UREs to produce an alternate indicator ranking based
solely on the magnitude of emissions and the chemical carcinogenicity. This
procedure would save resources, if performed alone, and would provide an
alternative rank order if done in combination with our approach.
• Emissions couid be estimated based on production data and established emission
factors for the SIC as opposed to data from TRIS.
• A more refined method for assigning emission release parameters couid be adopted
by incorporating facility specific information.
• This ranking scheme does not consider whether sources are located in high-
population areas. Population distribution information could be gathered to
supplement the ranking.
• The ranking could be based on a greater number of modeled facilities within each
SIC.
7-12
-------
I
X
CO
en
uu
c/3
cc
O
r~-
LU
1
DC
E »— fc—
| g 8
c§ <5 -*
s if
_ = 75 I"
0 P =1 OC
75 -| s i:
.0 (§ .> cu
— 75 ~" o
E c
g -2 — .
.£ 75 <=P* CD_
JS 7z -~, +
!5 CU 2> ^ ^
&—
03 C
en 2
2 75
> 7=
< i
•c c
o 0 S"
X 03
||
S||
'o
Q.
03
>. °C .
^C O to
*= f— , — .
-— * ^ ~f^ <^
1 I 3="
X o
E §
§ 0
S CD
x o
TJ 3
O
CO
CO
CM
O
O
co
g
3
C\J
o
o
IO
<
CM
O
in
CO
o
o
o
C\4
CD
CO
o
§
s
o
o
CO
00
o
CM
CM
O
CO
g
CO
CD
O
o
in
a
S
75
CO
75
-C
o
Q.
_>,
QJ
5
a.
CD
ra
CO
cu
75
>
"33
T3
o
'IJ
LU
QC
0
co
03
.£
(=
O
In
o
75
c
Q
o
O
ca
E
in
cu
._
^i—
'E
CA
c
~
S
c
o
o
£
^
E
"x
ca
S
75
II
^
>—
^
o
T3
O
75
-g
^
—
E
£
'§
E
o
73
o
-5
_c
en
CO
CM
JD
-a
cu
•a
^
C/3
S
ca
o
15
T3
^
C
' =
5
E
X
£
"o
^
O
ca
o
•o
.S
11
8
en
c/3
•~
0
ca
o
|
1
x
ca
^
"o
a.
1
^
,0
en
o
u
I
^
1
i
X
ca
E
=
CO
"o
JH
J2
7-13
-------
x
CO
CO
II
co
LLJ
co
DC
LLJ
LLJ
LLJ
CQ
O
a_
CM
r>»
LLJ
<
1
53
fill
— o
C 'f ^ , - — -
'x .fc "g • — •
3> co §>
is = — <
.So ~"
£0
Si
CO -J2
a> i=
< 0)
f|is
m CO
« o>
o> "=
en 2
2 2
> c
,^ O ^ |
"§ o "c .
X ^ "1^
11
x — -
CO C
?S O
Q.
O
CO
CO
CM
O
O
O
O
g
CM
§
CM
0
in
o
<
*f
0
o
§
in
CM
o
g
0
o
0
C\J
m
•^
fl^
S
o
g
CO
0
o
o
03
0
CM
CM
CO
O
tfi
O
C3
O
O
CO
g
CO
o
o
Q
r—
S
0
1
CO
.a
"3
o.
>,
O)
5
a.
CO
a>
15
">
"a3
o
Lti
LU
QC
o
co
CD
£
0
"•—
C/l
_o
2
"H
CD
CJ
3
O
.2
2
• S
CO
CD
^£
II
'c
CO
.1
c
.0
"co
J=
C
cu
g
o
o
£
3
£
X
ca
£
"5
it
^
•=
CD
8
03
CJ
"ca
"2
-a
|
c
'x
CO
E
"o
_a
ta
CJ
~o
c:
o
o.
OB
5
2
^5
CJ
1
^.
.<£
E
E
15
"o
la
o
^r
o
o
XI
T3
a>
'S
=0
CO
^_
cu
o
tz
CO
CJ
"S5
73
'>
"•^
g
1
X
CO
g
"3
0
"ca
o
-a
.c
n
CD
3
o
en
^"
CD
O
C
CO
CJ
1
i
X
ca
2
7-14
-------
TABLE 7-3. SAMPLE OUTPUT OF SOURCE CATEGORY RANKING FOR SINGLE POLLUTANT
Pollutant: A
Rank Based on
Maximum Individual
Risk
1
2
3
4
SIC
c
D
A
B
Number of
Facilities in SIC
35
20
12
6
Total Pollutant A
Emissions for SIC
(Ibs/year)
90,000
78,000
52,000
50,000
7-15
-------
TABLE 7-4. SAMPLE OUTPUT OF SOURCE CATEGORY RANKING BASED ON ALL POLLUTANTS
Pollutant: Alt
Rank Based on
Maximum Individual
Risk
1
2
3
4
SIC
c
D
A
B
Number of
Facilities
in SIC
35
20
12
6
Total Emissions of
all
Carcinogenic
Chemicals
210,000
170,000
120,000
100,000
Chemicals
Emitted
A, B, C, D
B, C, D, E
A, B,C
A, C, D
7-L6
-------
7.3 CASE II
7.3.1 Study Objectives
The purpose of this case study is to evaluate the impact of a proposed air toxics
regulation that would require control of carcinogenic pollutants to a risk level of 1 x 10"5
maximum individual cancer risk, and control of noncarcinogenic pollutants to exposures
below their respective reference concentrations for chronic inhalation exposure (RfCs). The
primary objective of the study is to identify which source categories, as defined by Standard
Industrial Classification (SIC) code, would be affected by the proposed rule. Sources affected
are identified as those whose ambient impacts, based on source specific dispersion modeling,
are above the proposed carcinogenic or noncarcinogenic risk levels.
The study will provide an indication of the expected permit work load that may result
from adoption of the rule. It will also provide the necessary basis for evaluating the potential
economic impacts of the proposed regulation and for identifying possible compliance options.
Implementation of these additional tasks is not discussed in this case study.
7.3.2 Scope
The proposed regulation would require all facilities within the state boundary that emit
an toxic air pollutant to obtain a permit. In order for a permit to be issued, a facility owner
or operator would have to demonstrate that the ambient air concentrations predicted in the
area surrounding their facility meet the health-based risk criteria of 1 x 10° maximum
individual cancer risk, for carcinogenic air toxics, and that the predicted annual average
concentrations be less than the inhalation RfCs, for noncarcinogenic air toxics. Those air
toxics covered by the proposed regulation are thus those for which EPA has established unit
risk estimates (UREs) and inhalation RfCs.
The exposure media is assumed to be only ambient air, since the regulation is
designed to control the level of toxic pollutants in air. Since the estimates of exposure and
risk will actually be used for comparison with the health risk criteria, uncertainties associated
with the study should be minimized where possible; however, it is accepted that wide
variability in the accuracy of the estimates may exist, depending on the accuracy of facility
emissions and model input parameters. Any assumptions made in the course of the study
7-17
-------
should be conservative. Although the exposure and risk estimates are based on facility-
specific information, representative of a given source category, they are not intended to be
conclusive of the impacts posed by any particular facility. This analysis would not be the
final determination of whether permits would be issued to specific facilities. If the regulation
were promulgated, an affected source could do a more detailed site-specific risk assessment to
verify whether estimated risks were above or below
1 x 10'5.
In addition to maximum individual risks, decision-makers also requested information
on the distribution of risk and the predicted annual cancer incidence in the population
surrounding facilities that could be affected, if such information could be easily generated.
To obtain this information and to calculate the exposure and risk estimates for comparison
with the health risk criteria, the EPA Human Exposure Model II (HEM-II) was selected.
HEM-II is a combination atmospheric dispersion and population exposure/risk model, and has
been used by EPA in setting natipnal emission standards for hazardous air pollutants
(NESHAPs) under Section 112 of the Clean Air Act. Within HEM-II, concentrations
produced from the dispersion model are matched with the population distributed throughout
the modeled region to produce cumulative or aggregate estimates of exposure and risk.
7.3.3 Selection and Use of Risk Assessment Methods
7.3.3.1 Hazard Identification.
Methods. This study considered only those chemicals for which EPA has already
performed a hazard identification. Botfi carcinogenic and noncarcinogenic pollutants were
•
included. Carcinogenic classification was obtained from the EPA Integrated Risk Information
System (IRIS). IRIS contains updated-dose-response information and is available through the
National Library of Medicine's Toxicology Data Network (TOXNET). Noncarcinogenic
pollutants were limited to those for which a Reference concentration for chronic inhalation
exposure (RfC) existed. Chemicals classified as human carcinogens (Class A) or probable
human carcinogens (Classes Bl and B2) were included. The IRIS data base can be searched
to isolate those chemicals for which carcinogenicity information is available.
It was decided not to search other literature sources or perform hazard identifications
for chemicals not yet reviewed by EPA or included in IRIS. Further information gathering
-------
and assessment would have required more resources and scientific expertise than were
available for the study. Also, the State considered it reasonable to concentrate initially on
regulation of those chemicals for which EPA had already performed a hazard identification.
Output. The output of the hazard identification step was a list of carcinogenic and
noncarcinogenic chemicals, with weight-of-evidence classifications (i.e., Class A, Bl or B2,
as defined in Chapter 2 of this report) indicated for carcinogens.
Expertise and Resources. Since a State computer account with the National Library of
Medicine already existed, the IRIS data base was readily accessed via personal computer
communications software. (The information contained in the data base is also available on
hard copy through the National Technical Information Service, and may be purchased on
floppy diskette.) This step and the dose-response step that follows were performed by a
health scientist in approximately two days.
7.3.3.2 Dose-Response Assessment.
Methods. To determine maximum individual and aggregate cancer risk, inhalation unit
risk estimates (UREs) of the carcinogenic chemicals were used. The URE represents the
increased cancer risk from a lifetime (70 year) exposure to an ambient air concentration of 1
ug/m3.
IRIS was accessed to obtain the unit risk estimates. Inhalation RfCs were also
obtained through IRIS and represent ambient air concentrations below which adverse health
effects are generally not expected to occur.
Output. The output of the dose-response step was a table listing the carcinogenic
chemicals, their associated UREs and weight-of-evidence classifications, and the
noncarcinogenic chemicals with inhalation RfCs. A sample of IRIS output pertaining to the
inhalation RfC summary is provided in Appendix A.
Expertise and Resources. As mentioned in the hazard identification step, the IRIS data
base was readily accessed via personal computer communications software, since a State
computer account with the National Library of Medicine already existed. (Information
contained in the data base is also available on hard copy through the National Technical
Information Service.) This step and the hazard identification step were performed by a health
scientist in approximately two days.
7-19
-------
7.3.3.3 Exposure Assessment.
Emission Characterization. To compile a sample set of sources to be used in the
study, regional offices within the State .were asked to identify three major and three minor
facilities in industrial categories as defined by the primary SIC codes. Major sources were
defined as those emitting more than 100 tons per year of any of the six pollutants for which
National Ambient Air Quality Standards exist (NAAQS). Minor sources were considered to
be those emitting less than 100 tons per year of any of these pollutants. The SIC codes
evaluated in this study were limited, based on a previous determination of the SIC codes for
facilities whose air toxic emissions posed the greatest potential risk to human health within
the State.
The facilities identified by the regional offices were mailed survey forms requesting
information on plant location (latitude and longitude coordinates), plant operating schedule,
process descriptions, point and fugitive emissions, control equipment and stack parameters.
Emission rates were requested for only those pollutants identified in the hazard identification
step. Furthermore, the rates requested were maximum hourly and annual average emission
rates. To conservatively estimate exposure, the hourly average rates were modeled as annual
average emission rates in the fate and transport step.
In order to model the facility emissions, specification of the following point and area
emission release parameters was required.
Point Sources Area Sources
Stack height Release height
Stack inner diameter Length of side of square area
Stack gas exit velocity
Stack gas exit temperature
To assess whether building wake effects might be significant, facilities were requested
to indicate if stack heights were exceeded by any building height within the facility. In
7-20
-------
addition, respondents were requested to indicate whether the predominant land use
surrounding their facility was urban (mostly commercial, industrial and compact residential)
or rural (primarily agricultural and residential) for input to the air dispersion modeling.
Approximately 80 percent of the facilities identified responded to the survey. Those
not responding had either closed or claimed no emissions of the pollutants identified. To
limit the amount of modeling required, the facilities were screened by emission rate.
Screening emission levels had previously been determined by the State. The selection of
these screening levels and the conservative modeling analysis conducted to determine them is
not included in the scope of this study. Those sources exceeding the trace levels of emission
for any pollutant were selected for input to the following steps of the study.
Fate and Transport Analysis. Pollutant transport and dispersion of facility emissions
within each SIC were simulated directly using the EPA Human Exposure Model-II (HEM-II).
HEM-II was selected because it is a model that is consistent with EPA guidelines and because
it has the ability to estimate both maximum and aggregate ambient exposure and risk using
built-in meteorological and population data bases. This simplifies the exposure and risk
calculations for the model user.
HEM-II contains the long-term version of the EPA Industrial Source Complex
(ISCLT) model. ISCLT is a Gaussian plume model that calculates annual average
concentrations resulting from continuously emitting point and area sources, such as those
located within facilities included in this assessment. The ISC regulatory default option, which
assigns several model options in accordance with the EPA "Guideline on Air Quality
Modeling" recommendations for regulatory air quality modeling analyses, was used for
modeling all facilities. The urban or rural land use option, which determines the algorithms
used to predict plume dispersion, was selected based on the information provided in the
facility survey.
The default receptor grid was selected in executing ISCLT within HEM-II. With the
default grid, annual average concentrations are predicted at 22.5 degree intervals on
concentric rings of radial distances, in meters: 200, 500, 1000, 2000, 5000, 10000. 20000.
30000, 40000 and 50000. The grid used in the modeling must include receptors near the
facility to identify the maximum concentration and include enough receptors spatially
7-21
-------
distributed throughout the entire modeled region to adequately determine aggregate population
exposure. The meteorological STAR data required to run ISCLT for the selected facilities is
included in an internal HEM-II data base.
Since the majority of facilities in the inventory reported stacks below building height,
building wake effects were assumed to be a potential problem at every facility. Building
heights used as input to ISCLT were conservatively assumed to equal stack height, subject to
the constraint that all buildings were less than 100 feet tall. Building lengths and widths were
calculated by the formula:
D2
4
<+
where Bw is the building width, BL is the building length, BH is the building height and D is
the effective diameter, set equal to the building height to simulate worst-case building
downwash.
In actuality, the influence of terrain and building downwash will vary from facility to
facility. Terrain effects were not considered in the modeling, given the infeasibility of
collecting receptor terrain data and implementing the complex terrain modeling approach
within the study schedule and budget. Terrain can have a significant impact on the predicted
ambient concentrations, as it can bring receptors to heights closer to plume centerline.
To use HEM-II, access to the EPA National Computer Center VAX computer was
required. Since no previous account had been established, an account for invoicing the
computer charges was opened through the National Technical Information Service (NTIS),
and the Pollutant Assessment Branch of EPA, which manages. HEM-II, was contacted on
proper access procedures (the Pollutant Assessment Branch is located in Research Tnangie
Park, NC). Once the account was authorized, HEM-II was readily accessed via personal
computer communications and VAX terminal emulator software.
Population Characterization. Within HEM-II, people are spatially distributed within
the modeled area based on the latest available U.S. Census Bureau Block Group/Enumeration
District (BGED) data. (Note that a Block Group is essentially a combination of contiguous
7-22
-------
city blocks having an average population of near 1,000.) The population data base contains
the number of people associated with the BGED and the latitude and longitude coordinates of
the BGED for mapping with the modeled concentration data.
Exposure Calculation. The maximum individual exposure was estimated for each
source selected for modeling. The maximum individual exposure is simply the maximum
predicted concentration. The presumption in the calculation is that an individual is located at
the point of maximum predicted concentration. Aggregate exposure was estimated within
HEM-II by matching the BGED population data with the modeled concentrations.
Population exposures for given concentration intervals are calculated as the product of the
BGED population exposed to a concentration within the interval and the concentration to
*
which that BGED population is exposed. The sum of these population exposures produces
the overall aggregate or cumulative population exposure.
Output. Within the exposure assessment step, the facility inventory was compiled and
screened by previously determined, trace emission rates to obtain a final source inventory for
the dispersion modeling. A listing of the facilities selected for modeling and of those that
were screened out was also made, sorted by SIC.
The maximum individual exposure and aggregate population exposures from HEM-II
were recorded for each SIC facility modeled. A sample of HEM-II output is provided in
Appendix B. The maximum concentration (maximum individual exposure) for each chemical
was compared to the inhalation RfC for that chemical to identify any exceedances. Results of
this comparison were recorded in a separate table, a sample of which is provided as Table 7-
~j.
Expertise and Resources. Accessing HEM-II required a personal computer with
communications software and VAX terminal emulator software. An account for invoicing the
computer charges was established with NTIS, and the Pollutant Assessment Branch within the
EPA Office of Air Quality Planning and Standards was contacted regarding user support for
HEM-II.
Final determination of the emissions inventory for modeling was accomplished by a
permit engineer and an air quality modeler, working with the State regional contacts, within a
7-23
-------
few weeks, not counting the time elapsed while the completed survey forms were being
returned. The inventory was compiled on a computerized spreadsheet.
The HEM-II modeling of each facility was performed by an air quality modeler and
was completed in approximately two weeks (the amount of time required to input the source
parameters, execute the model and produce the summary reports will naturally vary with the
number of sources involved in the study). Summarization of the HEM-II exposure results
was performed within a few days by the same air quality modeler. Comparison of maximum
concentrations with RfCs was done with a personal computer spreadsheet program.
7.3.3.4 Risk Characterization.
Methods. HEM-II calculates maximum individual risk (MIR) by multiplying the
»
maximum predicted annual average concentration (in ug/m3) by the pollutant unit risk
estimate (risk of cancer incidence per 1 ug/m3). As described in Chapter 5, maximum
individual risk represents the probability that an individual exposed continuously to the
maximum predicted annual average concentration for a 70 year lifetime will develop the
cancer related to that pollutant within his or her lifetime.
Aggregate population risk is also estimated within HEM-II. This aggregate risk is
often expressed as the annual cancer incidence, which is calculated by multiplying the
cumulative exposure (the cumulative product of the number of people exposed to a predicted
pollutant concentration) by the unit risk factor for the pollutant, and dividing by 70 years
[(cumulative exposure x unit risk)/70]. Annual incidence represents the number of cancer
cases expected per year due to emissions from the facility under study, if the population
within the modeled area are exposed continuously to the pollutant for 70 years and the facility
emissions of that pollutant remain unchanged during that period. Note that the exposure and
risk estimates generated for the facilities in this case study are not considered to be absolute
estimates for the modeled facilities, but are to indicate which source categories may oe
affected by the proposed regulation.
HEM-II also generates risk distributions. The output shows the number of indiviauals
exposed to various lifetime risk leveis. as well as the aggregate cancer cases expected in the
segment of the population exposed at each level. The cancer risk is calculated by multiplying
the pollutant unit risk estimate by the exposure for each BGED. Risks for BGEDs within an
7-24
-------
interval are summed to obtain the population risk for that interval. A sample of HEM-II
output is provided in Appendix B.
Output. Results from the HEM-II output were summarized in tables. A sample
format of the table for carcinogens is shown in Table 7-6 to this case study. The tables
shown in Tables 7-5 and 7-6 serve to consolidate the HEM-II output. A final table isolating
those source categories (SICs) which exceeded either the carcinogenic or noncarcinogenic risk
levels was then developed. A sample format is shown in Table 7-7. For carcinogens, the
predicted MIR is expressed as the ratio of the modeled risk to the standard for carcinogens
(10~5 maximum individual risk):
(modeled MIR/10'5).
Similarly, for noncarcinogens, the ratio of the predicted maximum pollutant
concentration (exposure) to the appropriate RfC is calculated as:
(maximum annual average concentration/pollutant RfC).
These values give a rough indication of the amount by which the standards are exceeded (a
ratio of 2, for example, indicates that the estimated value is double the standard).
The results of this case study estimate very generally whether or not facilities within a
given a source category, as defined by SIC code, will be affected by the proposed reguiation.
Although based on actual facility data, the assessment reflects only a sample of the state' s
facility inventory and cannot be conclusive for the entire state-wide population of permitted
air pollution facilities. As in Case I, a full report of the study, including a description of the
methodology, assumptions, and uncertainties was produced as part of the risk characterization.
In addition, an evaluation of the expected work load that may result from adoption of the rule
was presented, along with recommendations for further analysis of the possible compliance
options and associated economic impacts for the facilities affected.
In this example, an attempt was made to limit the amounr of uncertainty introduced by
using actual data, where possible. A moderate degree of uncertainty was considered
7-25
-------
acceptable, however, given the variability in accuracy of model inputs and study objectives
overall. Any assumption made in the methodology will introduce a degree of uncertainty.
Uncertainty is introduced in the modeling by the selection of model options (including
the omission of such items as terrain), and by selecting only a subset of the facilities from
each SIC for input. The degree to which facility emissions vary within an SIC should be
identified with simple statistics. Uncertainty in the chemical carcinogenicity is indicated by
the weight-of-evidence classification it is given in the hazard identification step.
Uncertainties associated with the UREs are discussed in Chapters 2 and 5 and include, for
example, extrapolation of toxicity data from animals to humans and from high experimental
doses to low doses encountered in the ambient air.
Expertise and Resources. Risk calculations from HEM-II were compiled and
summarized by an air quality scientist. Report preparation", including drafting discussions of
the methodologies used, results and uncertainties, was performed by the various persons
responsible for the study. The report was subsequently reviewed and edited by supervisory
personnel. The total time for evaluating the risk results, reporting the outcome for each SIC,
and circulating the study report for peer review was approximately two months.
7.3.4 Other Considerations
Possible variations of the methods described in this case study could be implemented.
These variations are indicated below.
• The study couid be expanded to include a determination of possible compliance
options for the sources affected. For example, if a control device would decrease
emissions by a certain percent, but would also affect the emissions release
characteristics, such as gas exit temperature, the predicted emission conditions
could be modeled to determine the controlled maximum risk level.
• If facility-specific information were not available, emissions could be based on
established production data and emission factors for the SIC.
• The exposure and risk estimation couid be performed through use of the Graphical
Exposure Modeling System (GEMS) instead of HEM II. GEMS also incorporates
the long-term version of the Industrial Source Complex (ISC) Model.
*
• Annual average ambient concentrations couid be estimated using an alternate EPA-
approved air dispersion model. This model output can be used as input to HEM
7-26
-------
II. Possible model options could include receptor terrain heights and building
cavity impacts.
All the facilities in one or more of the source categories could be modeled instead
of just a sample. This would greatly increase the level of effort, but would
indicate how many and which facilities are likely to be affected by the proposed
regulation.
7-27
-------
CO
z
LU
CD
O
O
cc
<
o
z
o
a:
2
CO
CO
LU
QC
LU
X
LL
O
>-
CC
CO
in
K-
LU
_l
m
co "c 5
283
c
o
O
g
n~*
c E
3 o)
"o
Q.
0)
O)
co
OJ
> c
llf
3 CD O)
< o a.
E o
1°
X
CO
IE
« CT3
O Z
a.
11
.=: c
o .5
Q. z
w§
IQ
CO
u_
o
o
•^
s
d
CD
2
j^
"o
Q.
CO
<
O
o
^
co
d
CD
"c
3
1
-------
CO
LU
o
CC
CO
OJ
CC
CC
<
CO
cci
o-
fc>
A
i»
CC
15
_-a
'•5
'x
CO
.la
CC
"75
«
'>
1
^""
1
'x
CO
03
ca
^
c
•5
"o
0.
TJ
-O
E
— j
•^
c
^o
"o
Q-
1)
T3
O
0
CO
Q
><
^^
o
CO
1
o
z
CD
0
X
CJ
'-
o
^
s
"o
Q-
^_
CO
>-
HI
o
X
r--
cg
O
"c
"o
o.
CO
>-
ir
O
x
CO
•*
Q
^
"3
"o
Q.
j
!
.,_
-
T
O
x
o
co
O
^
•2
"o
Q.
T_
CD
in
OJ
I
o
'z-
<*?
o
x
Tf
CO
Q
"c
•i
"o
Q.
CVJ
0
CO
7-29
-------
TABLE 7-7. SOURCE CATEGORIES AFFECTED BY THE PROPOSED REGULATION
SIC
Code
A
B
C
SIC
Description
X Manufacturing
Y Manufacturing
Z Manufacturing
Facility
ID
11
25
18
Pollutant
Pollutant C
Pollutant C
Pollutant D ,
Pollutant I
Pollutant H
Ratio of Maximum Risk
to Proposed
Carcinogen Standard3
2.7
60
43
NA
NA
Ratio of
Maximum
Exposure to
Proposed
Noncarcmogen
Standard
NA
NA
NA
1.4
1.4
aFor carcinogens, the ratio of the modeled risk TO the standard of (10~5 maximum
individual cancer risk is determined. A value of. 2%, for example, means that the
predicted MIR is twice the standard. For noncarcinogens, the ratio of the maximum
exposure to the pollutant RfC is determined.
7-30
-------
7.4 CASE III
7.4.1 Study Objectives
This case study examined the health risks associated with emissions from a waste-to-
energy facility for the purpose of obtaining an air permit. The results of this comprehensive
health risk assessment provided the State agency with the potential health risks (both
carcinogenic and noncarcinogenic) occurring in the surrounding community due to the
operation of the proposed facility. The level of detail required for this assessment was much
greater than that in the first two case studies. As a result, the level-of-effort and resource
expenditure for this type of analysis was high.
Since the degree of accuracy for this type of risk assessment needed to be as high as
technically possible, site-specific estimates for all parameters were required. These estimates
used should be as close to actual or "real life" as possible.
7.4.2 Scope
The risk assessment covers only one plant, but it is to be very comprehensive in
scope. The hazard potential of all pollutants to be emitted from the proposed facility must be
determined in the hazard identification process. Dose-response information was identified
from available sources for all pollutants that were potentially toxic. For compounds where
EPA-approved dose-response parameters (such as UREs and slope factors for carcinogens and
inhalation RfCs and RfDs for noncarcinogens) were not available, review of information
source and derivation of values were necessary.
A comprehensive health risk assessment required assessment of exposure by all
pathways including ingestion, inhalation, and dermal contact. Multipathway exposures were
determined using various fate and transport techniques including air dispersion and deposition
modeling and food chain modeling. The parameters utilized in these models (e.g., panicle
mass fractions for deposition modeling) should be derived from available data and be as close
to the actual value as possible.
The final step in the comprehensive health risk assessment was the quantification of
the overall risk to the exposed population. Maximum cancer and noncancer risks and hazard
indices were calculated in order to evaluate the overall risk to the surrounding population.
Aggregate cancer risk was also calculated to determine the number of people exposed to
7-31
-------
various risk levels and the total expected cancer incidence in the exposed population. This
information is often useful for risk communication with the public during the permitting
process.
7.4.3 Selection and Use of Risk Assessment Methods
7.4.3.1 Hazard Identification.
Methods. This case study considered all the pollutants to be emitted from the waste-
to-energy facility. The potential of each chemical to cause both carcinogenic and
noncarcinogenic health effects was evaluated. Initially, readily available data sources such as
IRIS, NTP annual reports, IARC monographs, ATSDR toxicologic profiles, and EPA Health
Effects Assessment Use documents were reviewed to determine the hazard potential of the
emitted pollutants. If the data search process revealed that a pollutant exhibits insignificant or
low toxicity (e.g., calcium, sodium) the compounds were eliminated from further evaluation in
the health risk assessment.
For compounds which have not been evaluated by an agency or group, the general
scientific literature was reviewed by a toxicologist to determine carcinogenic and
noncarcinogenic potential. However, instances where a compound had not been evaluated by
some group or agency were rare. Toxicologists also made a determination on the weight-of-
evidence for compounds the general literature identified as potential carcinogens. A scheme
similar to the EPA's weight-of-evidence classification system was used.
Output. The output of the hazard identification step was a series of tables listing the
chemicals emitted, the weight-of-evidence classification for carcinogens, and the
noncarcinogenic health effects endpoints of each chemical. References were also cited on the
tables and kept for use in writing the final report.
Expertise and Resources. Hazard identification was done by a toxicologist and a
scientist. The scientist searched the IRIS data base and some of the other readily available
sources of information. A personal computer was used for searches of the IRIS data bases
and for literature searches.
The toxicologist reviewed the general scientific literature and made the weight-of-
evidence determinations for compounds not previously reviewed by EPA or another Agency.
7-32
-------
This step required a thorough knowledge of toxicology and epidemiology. Another
toxicologist served as a peer reviewer for this step in the risk assessment.
The hazard identification step of Case in involved a lot more time and effort than
were required in Cases I and II because of the large number of sources that were reviewed.
Depending on the number of compounds emitted from the source in question, a thorough
hazard identification could take from one to several weeks.
7.4.3.2 Dose-Response Assessment.
Methods. In order to determine the cancer and noncancer risks associated with the
emissions from the proposed facility, dose-response parameters were required. For
determining carcinogenic risks, inhalation and oral unit risk estimates [(ug/m3)"1] or slope
factors [(mg/kg/day)"'] were necessary. The dose-response parameters required for the
noncarcinogenic risk assessment were RfCs for inhalation'exposures and RfDs for oral
exposures.
The primary source for both carcinogenic and noncarcinogenic dose-response
parameters was the IRIS data base. IRIS contains updated dose-response information and is
available on the National Library of Medicine's Toxicology Data Network through the
TOXNET directory. Once a computer account has been established through the National
Technical Information Service (NTIS), the data base is readily accessible via a personal
computer. The information contained in the data base is also available on diskette through
the NTIS. An example of the output from IRIS is given in Appendix A.
Since a number of compounds have not been reviewed or the dose-response
parameters verified by the EPA CRAVE or RfD work groups, additional data sources may
have to be searched to identify the dose-response values. A common source of dose-response
parameters is the EPA Health Effects Assessment Summary Tables (HEAST) issued quarterly
by the Office of Solid Waste and Emergency Response (OSWER). The HEAST contains
LJREs, slope factors, RfDs. and RfCs that are listed in the following EPA documents: Health
Effects Assessment documents. Health and Environmental Effects profiles. Health and
Environmental Effects Documents. Health Assessment Documents, and Air Quality Criteria
Documents. This quarterly publication also includes the most recent dose-response
parameters verified by either the CRAVE or RfD work groups.
7-33
-------
Once all the relevant EPA documents were reviewed, it was necessary to develop
dose-response parameters from the primary scientific literature for a few chemicals. The
derivation of dose-response parameters required an expert knowledgeable in dose-response
models and interpretation of toxicologic and epidemiologic data. Determination of the study
used to generate an RfD or RfC was of particular importance. Once a study was identified,
the methods used by EPA to develop an RfC or RfD were followed, but the resulting values
were not official RfCs or RfDs because they had not been through the EPA review process.
Development of the RfCs and RfDs required the identification of the No Observed Adverse
Effect Level or the Lowest Observed Adverse Effect Level from the critical study and
application of the uncertainty and modifying factors described in Chapter 3. Particular care
was taken in reviewing the literature for metabolic and pharmacokinetic information to
determine whether studies conducted for non-inhalation pathways could be used to develop
inhalation reference concentrations.
Derivation of UREs and slope factors for carcinogens is more difficult because dose-
response models have to be utilized. For this study it was determined that there was adequate
data in the literature to develop a URE for one compound not previously reviewed by EPA.
Procedures similar to those used in developing EPA's UREs were chosen to provide
consistency. Dose conversions from animals to humans was done using the surface area
approach described in Chapter 2. The linearized multistage model was used to extrapolate
responses for high experimental doses to low environmental doses, after checking to see that
there was no scientific evidence that a different model was more appropriate. See Chapter 2
for further discussion of dose-response modeling. It should be noted that for most
comprehensive risk assessments adequate dose-response parameters are available through the
previously mentioned sources to quantitate human health risks.
Output. The output of the dose-response assessment was a list of inhalation UREs,
slope factors, and drinking water UREs for the carcinogens along with the weight-of-evidence
classification for each compound. For the noncarcinogens, a list of RfCs and RfDs for each
compound. References and descriptions of methods, assumptions, and uncertainties were also
developed for inclusion in the final report.
7-34
-------
Expertise and Resources. To complete the dose-response assessment, a team
comprised of scientists, toxicologists, epidemiologist, and computer modelers were utilized.
A health scientist was used to search the IRIS data base and retrieve the applicable dose-
response parameters. A toxicologist derived the RfCs, RfDs, UREs and slope factors for the
compounds that had not been previously evaluated, with assistance from a computer modeler
in completing the computerized dose-response modeling. The peer review duties were
completed by a toxicologist and an epidemiologist.
The time expenditure for this task was large due to the level of detail and research
required for the dose-response assessment. The IRIS data base research required a day while
the literature review (select critical studies, identify critical endpoints, etc.) took a couple of
weeks to complete. A couple of weeks were also required to assemble the data, make
-v.
judgements on the methodologies and assumptions, and perform the dose-response modeling.
Peer review and revisions to the draft document required approximately one month.
7.4.3.3 Exposure Assessment.
Methods. The exposure assessment for evaluating the waste-to-energy facility
included a multiple pathway analysis with inhalation, ingestion, and dermal absorption
considered. This component of the risk assessment incorporated facility- and area-specific
data to accurately assess exposure in the communities surrounding the facility.
Emission Characterization. Point source emission rates were obtained by direct
measurement of the pollutants in trial burns from the specific waste-to-energy facility. These
trial burns were used specifically to quantitate the organic and inorganic pollutants being
emitted from the stack. EPA test methods were used for measurement of both criteria and
noncriteria pollutants from stacks. These test methods specify both the sampling protocol and
the analytical technique for qualitatively and quantitatively characterizing emissions. The test
burn approach was chosen because it would more accurately predict actual emissions from the
facility than the use of emission factors or data from other facilities.
Since air permits typically require a muitipathway exposure analysis, emissions from
any wastewater effluent must also be determined. Concentrations of pollutants in wastewater
effluent were determined from grab samples during the trial burns. The effluent resulted from
the blow down of the cooling tower.
7-35
-------
Exposure Pathways. Some State agencies require determination of exposure from all
pathways which contribute a significant amount to an individual's overall exposure. Exposure
pathways considered in this case study included inhalation, dermal absorption, soil ingestion,
ingestion of contaminated crops, meat, and dairy products, direct surface water consumption,
and consumption of contaminated fish. Other exposure pathways such as swimming and
showering were excluded since their contribution to overall exposure is generally negligible
compared to the other pathways. Table 7-8 lists the exposure scenarios considered in the
comprehensive health risk assessment.
Fate and Transport Analysis. Air pollutant transport was estimated with the ISCLT air
dispersion model. The ISCLT model was used with representative National Weather Service
meteorological data to estimate long-term (annual average) pollutant concentrations. ISCLT
is a Gaussian plume model that calculates annual concentrations resulting from continuously
emitting point and area sources. For the waste-to-energy facility exposure and risk
assessments, only annual averages were calculated.
Initially, a screening receptor grid was established for use in the ISCLT model to
predict annual concentrations at 22.5 degree intervals on concentric rings of radial distances,
in meters: 200, 500, 1000, 2000, 5000, 10000, 20000, 40000 and 50000. The screening grid
used in the modeling must include receptors near the facility to identify the maximum
concentration and include enough receptors spatially distributed throughout the entire modeled
region to adequately determine aggregate population exposure.
Based on the screening analysis, a refined grid was developed to pinpoint the
maximum ground level concentration for the annual averaging period. For this case study,
the refined grid consisted of a square cartesian grid (1000 meters on a side) with 100
receptors spaced 100 meters apart. Such grids were centered on the maximum predicted
ground level concentrations identified in the screening analysis. Discrete receptors were also
placed at sensitive receptors, such as schools and hospitals, to determine the exposure of these
sensitive populations.
Surface water was another potential source of exposure that had to be addressed.
Pollutants may enter surface waters by two pathways: transport of airborne pollutants to the
water body and effluent discharges to the water body. Airborne pollutants may enter the
7-36
-------
water body through atmospheric transport and deposition directly onto the surface water or
indirectly through soil deposition adjacent to the water body with subsequent transport via
surface runoff.
Since a lake located near the facility was a source of drinking water, estimates of
concentration attributable to the waste-to energy facility were calculated. The rate of
deposition of pollutants on the lake was calculated with the ISCLT dispersion model.
Concentrations of pollutants in the wastewater effluent were determined via grab samples
obtained during the trial burns.
Another route of exposure that was considered was the consumption of local fish, soil,
crops, and livestock which take up pollutants from their environment. For the evaluation of
exposure from fish consumption, chemical-specific bioconcentration factors were required
since fish tend to accumulate pollutants in their tissues. The bioconcentration factors were
used to estimate the equilibrium partitioning of pollutant between the surface water and fish
tissues. The following equation was used to estimate pollutant concentrations in fish:
Cf = Cw x BA
where:
Cf = Contaminant concentration in fish (ug/Kg)
Cw = Contaminant concentration in water (ug/L)
BA = Bioconcentration factor
Pollutant deposition and concentration in the uppermost soil layer were determined to
evaluate exposure from direct soil ingestion and root uptake in crops. Since pollutants may
be !ost from the soil by leaching, chemical and biological degradation, and volatilization, soil
loss constants can be applied. However, insufficient data existed to estimate loss for most
pollutants. For exposure via direct soil consumption and dermal contact, a mixing depth of 1
cm was assumed: for root uptake by crops, it was assumed thai the soil is tilled so that the
mixing depth is 20 cm.
Human exposure to pollutants by consumption of contaminated crops was estimated
for three types of produce grown locally in the surrounding area: root crops, leafy vegetables,
and fruits. Contamination of produce occurs via root uptake and by direct deposition onto
exposed plant parts. For root crops, only root uptake is a factor. To calculate a concentration
7-37
-------
of a pollutant in crops, plant-soil bioconcentration factors were used to measure the plant's
ability to accumulate pollutants within its tissue. Direct deposition was calculated as a
function of interception fraction, weathering, and duration of exposure. Equations for
calculating plant concentrations from both root uptake and direct deposition can be found in
"Methodology for Assessing Health Risks Associated with Indirect Exposure to Combustor
Emissions" (EPA, 1990).
Food chain models have been developed for predicting the concentration of pollutants
in animal tissues by considering the pollutant concentrations in plants and soils, the quantity
of plants and soils that animals consume, and the biotransfer factors for each type of animal
tissue. A biotransfer factor is defined as the ratio of pollutant concentration in animal tissue
to the daily intake of pollutant by the animal. These factors describe the extent to which
contaminants are transferred from the environment to specific animal tissues.
Pollutant levels in livestock were determined based on dietary intake and pollutant
levels in the crops eaten by the livestock. Concentration of pollutants in crops, as described
above, were calculated using pollutant-specific bioconcentration factors for root uptake, and
interception fractions to estimate direct deposition on plant surfaces. Typical dietary
compositions for livestock were obtained from State and county agricultural authorities.
Population Characterization. The locations of nearby communities and residences
were determined by reviewing U.S. Geological Survey topographic maps and local street and
land use planning maps of the area surrounding the facility. Sensitive receptors (e.g., schools
and hospitals) were of most importance since individuals located at these facilities are those
particularly susceptible to the effects of pollutants. After the location of the sensitive
receptors was identified, air modeling was conducted to determine annual pollutant
concentrations at these discrete points. The HEM-II model was used for estimating aggregate
population exposure, [n HEM-II people within the modeled region are distributed based on
the latest available U.S. Census Bureau Block Group/Enumeration District (BGED) data.
Exposure Calculation. Ambient air concentrations of pollutants emitted from the
waste-to-energy facility were estimated using the ISCLT model. Maximum individual
exposure for the annual average was determined for each chemical assuming an inhalation
rate of 22 nrVday. This inhalation rate is based upon 16 hours of light activity and 3 hours of
7-38
-------
rest per day. The dose rates of each chemical were determined by multiplying the inhalation
rate times the modeled concentration. Concentrations of pollutants at sensitive receptor sites
were also estimated with the ISCLT model. Aggregate population exposure was estimated
with the HEM-II model by matching the BGED population data with modeled concentrations.
It should be noted that aggregate exposure was estimated for the inhalation route only, since
HEM-II does not indicate exposure via other pathways.
Estimates of human exposures from drinking water, soil, fish, fruits, vegetables, and
livestock were made using behavioral characteristics from the U.S. EPA "Exposure Factors
Handbook" (1989). For drinking water, it was assumed that an individual consumes 2L/day.
Soil ingestion was assumed to occur at a rate of 0.1 g/day for adults and 0.2 g/day for
children. The fish consumption rate was conservatively estimated at 140g/day. Fruit and
vegetable consumption were assumed to occur at 142 and "201 g/day, respectively. Beef and
dairy product consumption rates were estimated at 100 and 400 g/day, respectively. And,
lastly, dermal exposure estimates were based on an estimated percentage of total body surface
area exposed. It was conservatively assumed that 100% absorption of the pollutants occurred.
All of the consumption rates given above were assumed to apply uniformly across the entire
population.
Output. The annual average exposure levels of the compounds emitted from the
waste-to-energy facility were estimated in the exposure assessment steps as well as the
aggregate population exposure. Through use of the refined air modeling analysis, the points
of maximum exposure were identified. Exposure levels resulting from consumption of
drinking water, soil, fish, fruits, vegetables, and livestock were also determined. Another
output was a list of references used in the exposure assessment and the assumptions and
uncertainties associated with the analysis.
Expertise and Resources. The exposure assessment was performed by an air quality
scientist and an ecoiogist. The air quality scientist estimated the dispersion and deposition of
the pollutants using ISCLT and HEM-II. A personal computer with communications software
was required for the air modeling task. A computer account with EPA was required for
assessing HEM-II. The ecoiogist evaluated the other relevant exposure pathways (drinking
7-39
-------
water, soil, etc.) by obtaining the proper EPA-approved algorithms for estimating exposure.
A personal computer was required for developing spreadsheets to calculate human exposure.
Fate and transport analysis for Case HI involved considerably more time and effort
than for Cases I and II. The multipathway nature of this comprehensive risk assessment
required more details and more actual or "real-life" values. As a result,the exposure
assessment could require anywhere from 1 to 3 months to complete depending upon the
number of compounds and pathways to be considered.
7.4.3.4 Risk Characterization.
Methods. Both cancer and noncancer health risks were estimated for the community
surrounding the waste-to-energy facility. Cancer risk was expressed as a probability and
derived by multiplying the URE or slope factor by the maximum exposure estimate. In order
to determine the overall cancer risk, risk estimates from all pathways and compounds were
added. Table 7-9 presents the results from such an analysis. This approach assumes dose
additivity and that synergistic and antagonistic effects do not occur. The resultant maximum
individual risk reflects the risk associated with inhabiting the point of maximum impact for a
lifetime and consuming all foodstuffs from local sources.
Aggregate population risk was estimated within HEM-II and is expressed as an annual
cancer incidence for the population exposed to the facility emissions. This value is calculated
by multiplying the cumulative exposure (the cumulative product of the number of people
exposed to a predicted pollutant concentration) by the pollutant URE, and dividing by 70
years [(cumulative exposure x unit risk)/70]. Annual incidence represents the number of
cancer cases expected per year due to emissions from the facility under study, if the
population within the modeled area is exposed continuously to the pollutant for 70 years and
the facility emissions of that pollutant remain unchanged during that period.
HEM-II also generates risk distributions. The output shows the number of individuals
exposed to various lifetime risk levels, as well as the aggregate cancer cases expected in ihe
segment of the population exposed at each level. The cancer risk is calculated by multiplying
the pollutant unit risk estimate by the exposure for each BGED. Risks for BGEDs within an
interval are summed to obtain the population risk for that interval.
7-40
-------
In addition to the carcinogenic risks, the noncancer risks associated with facility
emissions were estimated. For assessing noncancer risks, hazard indices were generated by
dividing the exposure estimates by either the RfC or RfD. If the resultant value is less than
unity, it is assumed that the exposure is not likely to result in any adverse cancer effects. As
with the cancer risks, the noncancer risks were summed across all exposure pathways and
chemicals to arrive at an overall noncancer risk estimate. Table 7-10 shows the output that
results from such an analysis.
The results of this case study estimate the potential cancer and noncancer health risks
in the surrounding community due to emissions from the waste-to-energy facility. Cancer
risks were estimated for the point of maximum impact, for the general population as a whole,
and at sensitive receptor sites (schools, hospitals, etc.). Noncancer risk was estimated at the
point of maximum impact only. As in Case I and II, a full report of the study, including a
description of the methodology, assumptions, and uncertainties was produced as part of the
risk characterization.
Uncertainties. A number of uncertainties were associated with each step of this
comprehensive health risk assessment. These uncertainties are introduced for steps in which
assumptions have to be made for particular parameters. In the hazard identification step,
uncertainty exists in the determination of the compound's carcinogenic potential.
Uncertainties associated with the derivation of UREs are discussed in Chapters 2 and 5 and
includes extrapolation of animal data to humans and extrapolation from high to low doses.
Uncertainty is introduced in the modeling by the selection of various model option which
affect the pollutant dispersion and deposition. In the multipathway exposure analysis,
uncertainty arises from the selection of consumption rates via the various pathways. There is
also uncertainty associated with the estimation of uptake by biota in the exposure assessment.
Output. A detailed report presenting the exposure and risk estimates for the waste-to-
energy facility and a summary of the methodology, assumptions, and uncertainties associated
with the health risk assessment.
Expertise and Resources. Cancer risk calculations from HEM-II were compiled and
summarized by an air quality scientist. The multipathway cancer and noncancer risk
calculations were performed by a toxicologist on computerized spreadsheets. The final report
7-41
-------
was reviewed and edited by a senior level toxicologist and air quality scientist. The total
time required for generating the risk estimates and describing the methodology, assumptions,
and uncertainties was two months.
7.4.4 Other Considerations
Alternative methods are available for conducting a comprehensive health risk
assessment than described in the above case study. The following is a list of some of these
alternatives.
• For both qualifying and quantifying emissions from the waste-to-energy facility,
test data from a similar facility could be used if available in place of actual data.
AP-42 emission factors may also be used if available.
• The exposure and risk estimation could be performed through use of the Graphical
Exposure Modeling System (GEMS) instead of HEM-II.
• A multipathway exposure analysis may not be required by all State agencies.
• Dose-response models other than the linearized multistage model, such as the one-
hit and Weiiull models, could be used if the data fit.
• Short-term (24-hr average or less) ambient air concentrations could be modeled to
evaluate short-term health effects for chemicals for which acute or subchronic
dose-response parameters have been derived.
7-42
-------
D
o
o
cc
LJJ
UU
LU
CO
I
Q
LU
CO
O
Q.
O
cc
Q.
CC
O
LL
CO
o
cc
<
LU
O
co
LU
CC
rj
CO
O
CO
rv.
LU
—I
CQ
<
Assumptions
-
05
•ss
o
CC
CD
21
CO
O
Q.
x
LU
"3
73
CD
•^
73
CD
O
CD
^
"c
o
"ca
O
5
"5.
CD
CD
CC
E
3
• Worst-case annual average emission
rates
• Normal meteorologic conditions
• Impacted individual resides at maxim
impact point
• Assumed crops not washed before
consumption
CO
Q.
O
O
"^
Si
co
c
g I
•— c
1 § 1
n — O
C « CO _
O ra CD S
•.= -^ O) 2
•5 £ ••= ^
CO t -= CD
-C CD O ^^
£ Q co S
* * * *
.
o
• Beef/dairy farms assumed to be
located at point of highest concentrat
outside city limits
co
0)
£
73
0)
"55
_c
S. 1
^ Q.
It
V3 *O
^ "O
O) ^
-£ co
•
^-
Q
a.
o
.-a
a.
P
w ! ™
0
Q.
CB
Q.
.i -5
E c
3 CD
£ la _
x £ o
^ < CO
^
CO
ii
3
"5
o
o) 5
1 1
111
?B < CO
^ • •
• Concentration at maximum immpact
point assumed for entire watershed
• All deposited particulate assumed to
reach surface water
CO
73
_0)
CO
c
co
1 I
P °
W 0
§ 1- '
o E
"£3 (/j
0) C
.i O
Q O
•
CD
-X
"CD
£ l_
"3 —
O CO
i 8
1 =
"3 co
5 •
7-43
-------
>•
p
o
CD
OC
LU
LJJ
6
111
LU
1
UL
C/3
^i
CO
cc
cc
LU
O
1
UJ
tr
en
ri
LU
—i
CQ
1
u.
1
en
^
0.
cr
CO
O
X
UJ
„
c.
CO
a_
c/1
|
O_
^
"c5
0
__!
CD
05
(U
CO
O)
CO
"o
X
<-r)
oo
<•>
o
X
O5
0
X
CO
o
X
^f
m
o
X
•^r
CM
O
^D
X
CO
m
O
X
CO
OO
O
X
CO
o
•»~
X
~
o
X
in
"^
0
O
X
CO
CO
0
X
o
I —
0
X
n-
CD
0
X
,_
Q
0
X
CO
evi
o
X
^~
cd
n
O
X
in
CM
O
•r—
X >
CO
at
O
X
»—
CO
CO
o
X
0
X
CO
CO
o
X
co
•—
0
0
X
m
LU
^.
X
CO
o
X
^~
h-
o
X
••—
o
•^~
X
CO
ID
X
0
""
.-,
o
X
o
X
o
X
^
h-« i
o
X
-------
I
LU
g
CO
LU
CO
^
CO
cc
cr
LU
o
a
LU
SS
LU
CC
o
X
<
01
CO
LU
CO
s
o
i
o
o
o
o
o
1
o
o
Si
CO
0
i
o
00
o
o
o
.
o
o
O
CO
8
0
CO
o
s
o
o
o
i
0
55
0
o
o
o
o
o
m
8
o o
1
I
0
CD
8
o
0
s
0
o
§
o
Q
s
Si
o
0
0
§
o
•v-
1
0
o
o
o
8
0
CNJ
o
e\j
cr>
S
o
0
o
CJ
c>
ao
O
CD
O
o
ao
0
0
'
CO
o
0
i
0
at
i
1
1
s
O |
o
C\J
o'
o
I
7-45
------- |