Guidance Manual for Health Risk Assessment of
Chemically Contaminated Seafood
PB90-197880
Tetra Tech, inc., Bellevue, WA
Prepared for:
Environmental Protection Agency, Seattle, WA
Jun 86
i
Ui!>:
-------
30217-101
REPORT DOCUMENTATION
PAGE
I. REPORT NO.
EPA 910/9-33-182
I 3. Recipient*» Accesilon No.
90 1) q iyg
4. Tttto aad Subtitle
Guidance Manual for Health Risk Assessment of CP.emically
Contaminated Seafood
| 5. Report Data
June 1986
7. Authors)
8. Performing Organization Rapt No
9L Performing Organization Name and Add ran
Tetra Tech, Inc.
11820 Northup Way, Suite 100
Bellevue, Washington 98005
10. Proiect/Task/Work Unit No.
11. Contract(C) or Crant(G) No.
(C)
(G)
ti. Sponsoring Organization Nam* and Addrau
U.S. Environmental Protection Agency
Region 10, Office of Puget Sound
1200 Sixth Ave.
Seattle, Washington 98101
13. Typo of Report & Parted Covered
14.
15. Supplementary Note*
16. Abstract (Limit 200 words)
This report was written to assist in the evaluation and
interpretation of the human health risks associated with chemical
contaminate levels in seafood. High concentrations of toxic
chemicals have been found in sediments and marine organisms in
parts of Puget Sound. Since heavy consumption of contaminated
seafood may pose a substantial human health risk, it's important
that assessments of the risk associated with seafood consumption
be conducted in a consistent, acceptable manner. This report
provides an overview of risk assessment, and describes hazard
identification, dose-response assessment, exposure assessment
and risk characterization. Guidance is provided on presentation
and interpretation of results.
17. Document Analysis a Dasc-lptors
Idmtlflera/Open-Ended Tarmr
COSATI Field/Group
1*. Availability Stat.mant
19. Security Class (This Report)
20. Security Clasi (This Page)
21. No. of Pages
22. Prlci
(SMANSI-Z39.18)
See Instruction! on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-35)
Department of Commerce
-------
P890-197880
EPA 910/9-8S-182 »»««
Puget Sound Estuary Program
GUIDANCE MANUAL FOR
HEALTH RISK ASSESSMENT
OF CHEMICALLY
CONTAMINATED SEAFOOD
FINAL REPORT
PREPARED BY:
TETRA TECH, INC.
PREPARED FOR:
US. ENVIRONMENTAL PROTECTION AGENCY
REGION 10 — OFFICE OF PUGET SOUND
JUNE 1986
REPRODUCED BY
U S DEPARTMENT OF COMMERCE
NATIONAL TECHNICAL INFORMATION SERVICE
SPRINGFIELD. VA 22161
-------
ACKNOWLEDGMENTS
This document was prepared by Tetra Tech, Inc., under the direction
of Dr. Robert A. Pastorok, for the U.S. Environmental Protection Agency
(EPA) in partial ft-1 fi 1 linent of Contract No. 68-03-1977. Ms. Sally Hanft
of U.S. EPA was the Project Officer and Dr. Thomas C. Ginn of Tetra Tech
was the Program Manager. Portions of this work were initiated under U.S. EPA
301(h) post-decision technical support Contract No. 68-01-6938, Ms. Allison
Duryee, Project Officer.
The primary author of this report was Dr. Robert A. Pastorok. Dr. Leslie
Uilliams and Mr. Jonathan Shields of Tetra Tech, Inc., provided technic.il
assistance. The following individuals provided valuable comments on the
draft report from which this report was developed: Dr. John Armstrong
of U.S. EPA, Mr. Pieter Booth of Tetra Tech, Dr. Alan Ehrl ich of U.S. EPA,
Dr. Henry Lee of U.S. EPA, Mr. Jerry Leitch of U.S. Food and Drug Admini-
stration, Dr. Gerald Pollock of California Department of Health Services,
Ms. Patricia Storm of U.S. EPA, and Mr. David Tetta of U.S. EPA. Ms. Marcy
Brooks-McAuliffe assisted in technical editing and report production.
This report benefited from discussions at a workshop on "Approaches
to Ecological and Human Health Risk Analysis for Disposal of Contaminated
Sediments and Human Consumption of Contaminated Seafood" held December
16-17, 1985, in Seattle, Washington. The workshop was jointly sponsored
by U.S. EPA Region 10 and the Seattle District of the U.S. Army Corps of
Engineers. Primary participants at the workshop are listed below:
Participants
Dr. Jack Gentile
Dr. Thomas Di1 Ion
Dr. Peter Melln.yer
Dr. Elaine Faustman-Watts
Dr. Curtis Brown
Dr. David Eaton
Dr. Eugene Stakhiv
Dr. Richard Peddicord
Dr. Alan M. Ehrlich
Or. Gerald Pollock
Dr. David Rosenblatt
Dr. Lawrence Barnthouse
Dr. Richard Branchflower
Mr. Stephen Norsted
Dr. Michael Watson
Ms. Diane Martin
Ms. Jane Lee
Dr. Thomas Ginn
Dr. Michael Dourson
Dr. John Armstrong
Mr. Keith Phillips
Affillation
U.S. Environmental Protection Agency
Waterways Experiment Station, Corps
of Engineers
Battelle Northwest Laboratory
University of Washington
Bureau of Reclamation
University of Washington
U.S. Army Corps of Engineers
Waterways Experiment Station, Corps
of Engineers
U.S. Environmental Protection Agency
California Department of Health Services
U.S. Army Medical Bioengineering R&D
Laboratory
Oak Ridge National Laboratory
Tacoma General Hospital
Washington Department of Social and
Health Services
U.S. Environmental Protection Agency
Envirosphere Company
Seattle-King County Health Department
Tetra Tech, Inc.
U.S. Environmental Protection Agency
U.S. Environmental Protection Agency
U.S. Army Corps of Engineers
11
-------
CONTENTS
Page
ACKNOWLEDGMENTS ii
LIST OF FIGURES vi
LIST OF TABLES vii
EXECUTIVE SUMMARY 1
INTRODUCTION 1
THE PISK ASSESSMENT PROCESS 2
Hazard Identification 3
Dose-Response Assessment 3
Exposure Assessment 3
Risk Characterization 5
PRESENTATION AND INTERPRETATION OF RESULTS 6
INTRODUCTION 9
OBJECTIVES 9
ORGANIZATION 10
OVERVIEW OF RISK ASSESSMENT 11
MAJOR STEPS IN RISK ASSESSMENT 12
NEED FOR RISK ASSESSMENT APPROACH 12
USES OF RISK ASSESSMENT AND MANAGEMENT 13
HAZARD IDENTIFICATION 16
CONTAMINANTS OF CONCERN 16
TOXICITY PROFILES 21
SOURCES OF INFORMATION 24
DOSE-RESPONSE ASSESSMENT 26
EXPOSURE AND DOSE 26
GENERAL DOSE-RESPONSE RELATIONSHIPS 26
HI
-------
CARCINOGENIC POTENCY FACTORS 29
REFERENCE DOSES 33
SOURCES OF INFORMATION 33
Carcinogenic Potency Factors 35
Reference-Risk Doses 35
EXPOSURE ASSESSMENT 36
TISSUE CONCENTRATIONS OF CONTAMINANTS 36
Selection of Analytical Detection Limits 37
Selection of Target Species 37
Statistical Treatment of Data 40
EXPOSED POPULATION ANALYSIS 41
Comprehensive Catch/Consumption Analysis 42
Assumed Seafood Consumption Rate 45
INTEGRATED EXPOSURE ANALYSIS 47
RISK CHARACTERIZATION 49
CARCINOGENIC RISK 49
NONCARCIMOGEN 1C EFFECTS 51
CHEMICAL MIXTURES 52
PRESENTATION AND INTERPRETATION OF RESULTS 53
PRESENTATION FORMAT 53
Summary Tables 53
Summary Graphics 55
RISK COMPARISONS 56
SUMMARY OF ASSUMPTIONS 62
UNCERTAINTY ANALYSIS 64
Sources of Uncertainty 65
Approaches to Uncertainty Analysis 67
SUPPLEMENTARY INFORMATION 68
REFERENCES 70
tv
-------
APPENDICES
A: Sources of information for toxicity profiles A-l
B: Example database summary for Reference doses [RfDs}
derived by U.S. EPA 8-1
C: Regulatory limits on chemical contaminants in seafoods C-l
-------
FIGURES
Page
Hypothetical example of dose-response curves for a
carcinogen and a noncarcinogen 28
Concaptual structure of quantitative health risk assess-
ment model 50
Example graphic format for display of quantitative risk
assessment results for hypothetical study area and reference
area 57
Plausible-upper-limit estimate of lifetime cancer risk
associated with mean contaminant concentrations in seafood
species A vs. rate of seafood consumption 58
Plausible-upper-limit estimate of lifetime cancer risk vs.
concentration of a chemical contaminant in seafood (ppm wet
wt) at selected ingestion rates 59
-------
TABLES
Number
1 Organic priority pollutants with established toxi-
cological indices ranked in order of their octanol-
water partition coefficients (KQw) 17
2 Inorganic priority pollutants with established toxi-
cological indices 19
3 Toxicity profile for mercury and PCBs 22
4 Carcinogenic priority pollutants ranked by potency factors 31
5 Reference Oose (RfD) volues for priority pollutants 34
6 Example tabular format for display of quantitative risk
assessment for consumption of seafood 54
7 Example of cancer risks from common carcinogens 61
8 Summary of assumptions and numerical estimates used in
risk assessment approach 63
-------
EXECUTIVE SUMMARY
INTRODUCTION
This guidance manual was prepared under the Puget Sound Estuary Program
in response to the potential problem of toxic chemical accumulation in
marine organisms. The objectives of this guidance manual are to:
• Describe the steps of a procedure for assessing potential
human health risks associated with consumption of contam-
inated seafood
a Provide guidance on presenting and interpreting risk assessment
results for public understanding
o Summarize assumptions and uncertainties of the recommended
procedure for risk assessment
e Summarize standard model coefficients and criteria used
in risk assessment, and information sources for updating
these values.
The risk analysis process consists of two distinct phases: risk assess-
ment and risk management. Risk assessment entails estimating the scientific
probability of incurring an adverse health effect from exposure to a toxic
agent. Risk management entails interpreting risk assessment results to
formulate public policy. Socioeconomic, technical, and political factors
are considered in risk management.
Risk assessment uses predictive models for two principal reasons:
-------
o Direct measurements of human health risks associated with
seafood contamination are rarely available due tc the difficulty
and high cost of conducting epidemiological studies
o Regulatory agencies use predictive models to develop criteria
such as U.S. FDA action levels (i.e., maximum allowable
contaminant'concentrations in food) to prevent health problems.
Risk assessment techniques are inexact and yield uncertain results. However,
they are virtually the only predictive tools available with which to formulate
public policy regarding toxic contamination in seafood.
In the Puget Sound Estuary Program, risk assessment will be used to:
o Identify problem areas, problem chemicals, and problem species
(and possibly weight classes/length within species)
o Develop guidelines for contaminant concentrations in seafood
or for consumption limits
o Provide public information or issue public health advisories.
The results of risk assessment are useful in toxic chemical management
by regulatory agencies. Risk assessment of seafocd contamination is also
important because it will provide the public with information on which
to base individual decisions regarding where to harvest seafood, what species
to harvest, and how much to consume.
THE RISK ASSESSMElfT PROCESS
Tne risk assessment process has four major components: hazard identifi-
cation, dose-response assessment, exposure assessment, and risk character-
ization.
-------
Hazard Identi Meat ion
Hazard identification involves defining toxicolugical hazards posed
by individual chemical contaminants in seafood samples. Factors that determine
whether a cor.tarninant should be evaluated include persistence, bioaccumulatior
potential, presence on the list of Puget Sound contaminants of concern
(Konasewich et al. 1982), and status as a U.S. EPA priority pollutant.
P'jget Sound contaminants of concern for which toxicological potency factors
hav? been defined are shown in Tables 1 and 2 in the text. Toxicity profiles
are constructed for the selected chemicals based on physical, chemical,
metabolic, and pharmacokinetic properties, and toxicologic effects (Table
3 in the text).
Dose-Response Assessment
Dose-response data are used to determine the toxicological potency
of a substance (Figure 1 in the text). A measure of toxijcological potency
is derived from the dose-response relationship for the most sensitive species
tested (usually a laboratory strain of rats or mice). Results of laooratory
experiments are then extrapolated to humans.
The toxicological index used for carcinogens is the Carcinogenic Potency
Factor. For carcinogens, there is an implied finite risk of cancer even
at low doses (Figure 1 in the text). For noncarcinogens, there is usually
a threshold dose below which adverse biological effects are not observed
in animal bioassays (Figure 1 in the text). The toxicological index used
for noncarcinogens is the Reference Dose (RfO), wfnch is the highest average
daily exposure over a lifetime that would not be expected to produce adverse
effects.
Exposure Assessment
Exposure assessment is the process of characterizing the populations
exposed to the chemicals of concern, the environmental transport and fate
pathways, and the magnitude and duration of the exposure dose (U.S. EPA
1984b). A risk assessment of contaminated seafood involves:
-------
9 Evaluation of tissue concentrations of contaminants in
organisms
• Characterization of the exposed population, including catch-
consumption characteristics and consumption rate.
An integrated exposure analysis is performed to estimate the exposure dose
(chemical intake by human) for each seafood species.
Tissue Concentration Analysis--
In determining tissue concentrations of contaminants in marine biota,
it is important to select study components carefully, e.g.: 1) the detection
limits of laboratory analytical procedures and equipment, 2) target species,
and 3) statistical methods for data analysis. In general, statistical
summaries of tissue concentration data should include at a minimum the
arithmetic mean concentration and a measure of variance.
Exposed Population Analysis--
The analysis of exposed populations includes four steps: 1) identifying
the potentially exposed population by fishery harvest area, 2) describing
the demographic and seafood harvesting activities of the population, 3) charac-
terizing catch patterns and consumption patterns, and 4) estimating average
consumption rates. Consumption rates can be calculated for each seafood
species and each human subpopulation if extensive catch/consumption data
are available. Where these data are lacking, standard values for consumption
are assumed.
It is appropriate to use standard consumption rates when site-specific
data are not available, differences among areas (or times) are expected
to be small, or a thorough catch/consumption analysis is unnecessary to
meet the study objectives. Three standard values of seafood consumption
are recommended in this manual:
-------
o 6.5 g/day to represent a low estimate of average seafood
consumption for the entire U.S. population (U.S. EPA 1980b)
o 20 g/day to represent a high estimate of the same average
rate (U.S. Department of Agriculture 1984)
o 165 g/day to represent average seafood consumption for the
small portion (0.1 percent) or the U.S. population consuming
the most seafooo (Finch 1973).
The consumption rate values correspond to 16, 48, and 400 meals/yr, assuming
one mpdl equals 150 g (0.33 Ib).
Integrated Exposure Analysis—
In the integrated exposure analysis, estimated contaminant concentrations
and seafood consumption rates are combined to estimate exposure dose by
seafood species. The exposure dose is expressed in terms of mg'kg body
weight" day" averaged over a 70-yr lifetime.
Risk Characterization
In the risk characterization stage, the probability and extent of
adverse effects associated with consumption of contaminated seafood (Figure 2
in the text) are estimated from results of the exposure and dose-response
assessments. Carcinogens and noncarcinogens are treated separately.
Numerical estimates of carcinogenic risk can be presented as a unit
risk score (i.e., risk per unit dose) or as a maximum allowable dose or
concentration. Risk estimates can also be presented on an individual or
population basis. The general model for estimating a plausible-upper limit
to excess lifetime risk of cancer for an individual is:
R.!
-------
where:
"ijkm = Plausible-upper-limit risk of cancer associated with chemical m
in species i for subpopulation j in area k (dimensionless)
* 111
Qlm = Carcinogenic potency factor for chemical m [(mg'kg 1-day i) x]
estimated as the upper 95 percent confidence bound on the
slope of a linear dose-response curve
Eijkm a Exposure dose of chemical m from species i for subpopulation j
in area k (mg'kg"1-day"1).
An index of noncarcinogenic risk may be approximated as the ratio
of the estimated exposure dose to the Reference Dose. This index is compared
to a value of 1.0 to evaluate noncarcinogenic hazard.
Because data on chemical interactions are limited, estimated risks
for individual chemicals are usually summed to obtain an approximate estimate
of total risk for a chemical mixture.
PRESENTATION AND INTERPRETATION OF RESULTS
The results of risk assessment may be presented in both tabular and
graphic format. All risk estimates should be interpreted as plausible-
upper-1 unit values for the stated assumptions and exposure conditions.
Because risk estimates for a given area and seafood species vary with con-
sumption rate and because consumption rates vary greatly among individuals,
plots of plausible-upper-limit risks vs. consumption rate are recommended
as the primary means of presenting results (Figures 3 and 4 in the text).
Such graphic representation can be a valuable aid for the lay public and
for risk management because it can incorporate additional information such
as comparative risks.
Estimated health risks for the study area should be interpreted by
comparison with:
0 Health risks for consumption of similar seafood species
from a reference area
-------
• Health risks for consumption of alternative foods or other
activities
• Acceptable risk levels defined by agency policy.
The results should include a summary of assumptions and an uncertainty
analysis. The sunmary should note general assumptions inherent in risk
assessments (e.g., extrapolation of effects from laboratory animals to
humans), specific assumptions adopted for the risk analysis in question,
and estimates of model coefficients (Table 8 in the text).
Uncertainties in the risk assessment approach presented in this manual
arise from the following factors:
• Estimating carcinogenic potency factors or RfDs
• Estimating seafood consumption rates
• Estimating the efficiency of assimilation (or absorption)
of contaminants by the human gastrointestinal system
• Variation of exposure factors among individuals
• Model uncertainty.
Uncertainty ranges (e.g., 95 percent confidence intervals) around estimates
of mean risk may typically span 3-4 orders of magnitude. The approach
taken by U.S. EPA (1980b, 1985a) and followed herein is to estimate a plausible-
upper limit to risk for specified exposure conditions. This is accomplished
by using the upper bound of the 95 percent confidence limit on the estimate
of the carcinogenic potency factor. In this way, it is unlikely that the
risk associated with the stated exposure dose will be underestimated sub-
stantially. Moreover, the plausible-upper-limit estimate serves as a consistent
basis for relative risk comparisons. Uncertainty in estimates of exposure
-------
dose may be addressed by one of several methods discussed in the text.
The approach selected will depend on the available data and the study ob-
jectives.
-------
INTRODUCTION
Under the Puget Sound Estuary Program, the U.S. Environmental Protection
Agency (EPA), Region 10, has identified accumulation of toxic chemicals
in marine organisms as a potential problem. High concentrations of toxic
chemicals have been found in sediments and in some marine organisms from
urban bays, such as Commencement Bay and Elliott Bay, relative to those
from remote locations of Puget Sound (Mai ins et al. 1980, 1982; Tetra Tech
1985, 1986c). Heavy consumption of contaminated seafood may pose a substantial
human health risk. This concern has prompted recent studies of catch and
consumption patterns for recreational fisheries in urban bays (e.g., Landolt
et al. 1985; McCallum 1985) and associated health risks (Versar 1985; Eagle
Harbor Ad Hoc Committee 1985).
One goal of the Puget Sound Estuary Program is to protect the health
of local seafood consumers by providing information on relative health
risks associated with various edible marine species, geographic locations,
and seafood consumption rates. Diverse models have been used in the past
to estimate human health risks from exposure to toxic substances in food
(e.g., U.S. Office of Technology Assessment 1979; Food Safety Council 1980,
1982; Connor 1984). A standardized procedure is needed for risk assessment
of chemically contaminated seafood.
OBJECTIVES
The purpose of this report is to provide guidance for risk assessment
of contaminated seafood based on U.S. EPA approaches (e.g., U.S. EPA 1980b;
1984a,b,c; 1985a,c). The objectives of this guidance manual are to:
o Describe the steps of a risk assessment procedure for contam-
inated seafood
-------
o Provide guidance on presentation and interpretation of risk
assessment results
o Summarize assumptions and uncertainties of the recommended
procedure for risk assessment
o Summarize standard model coefficients (e.g., carcinogenic
potency factors) and criteria [e.g., U.S. Food and Drug
Administration (FDA) action levels)] used in risk assessment,
and information sources for updating these values.
ORGANIZATION
An overview of risk assessment is provided in the next section. The
overview includes a discussion of the distinction between risk assessment
and risk management and a review of their possible uses under the Puget
Sound Estuary Program. Each major step of the risk assessment process is
described in subsequent sections. Guidance is provided on general mathematical
models to be used. Sources of information on toxic chemicals and model
variables are noted. Finally, suggestions for presenting and interpreting
risk assessment results are provided. Uncertainties and assumptions of
the assessment approach described in this manual are summarized.
10
-------
OVERVIEW OF RISK ASSESSMENT
Risk assessment is a scientific procedure to.determine the probability
of adverse health effects frcm a specific exposure to a tdxic agent. Risk
assessment differs from risk management, although both are components of
regulatory decision-making (National Research Council 1983). Risk assessment
provides the scientific basis for public policy and action. In risk management,
risks are interpreted in light of legislative, socioeconomic, technical,
and political factors, and appropriate controls are determined.
Direct measurement of human health risks is possible in certain limited
circumstances. Such circumstances generally involve a single high exposure
or repeated moderate exposures to a specific chemical, and a clear cause-effect
relationship. For example, direct measurement of cancer risks might be
possible in a population of workers exposed to an industrial chemical spill.
In contrast, it is virtually impossible to directly measure the health
risks of eating seafood harvested from Puget Sound during recreational
activities. Models that predict health risks are therefore needed. Risk
assessment procedures discussed in this manual focus on predicting health
risks from long-term exposure to relatively low levels of contamination.
The following sections provide an overview of the steps in risk assessment,
the need for risk assessment, and the potential uses of risk assessment
in the Puget Sound Estuary Program. The general format for risk assessment
and all definitions of terms used in this report are consistent with t^ose
provided by National Research Council (1983) and U.S. EPA (1984a,b,c; 1985c).
Background information on food safety evaluation by federal and state agencies
is provided by the U.S. Office of Technology Assessment (1979) and Food
Safety Council (1980, 1982).
11
-------
MAJOR STEPS IN RISK ASSESSMENT
A complete risk assessment Includes the following steps:
• Hazard identification: Qualitative evaluation of the potential
for a substance to cause .adverse health effects (e.g., birth
defects, cancer) in animals or in humans
• Dose-response assessment: Quantitative estimation of the
relationship between the dose of a substance and the probability
of an adverse health effect
• Exposure assessment: Characterization of the populations
exposed to the toxic chemicals of concern; the environmental
transport and fate pathways; and the magnitude, frequency,
and duration of exposure
f Risk characterization: Estimation of risk for the health
effect of concern, based on information from the dose-response
and exposure assessments.
Because uncertainties are pervasive in risk assessment, uncertainty analysis
is a key element of each stage of the assessment process. Assumptions
and uncertainties are summarized in the risk characterization step.
NEED FOR RISK ASSESSMENT APPROACH
Scientific knowledge of the effects of toxic chemicals on humans is
still rudimentary. Much of our present information is extrapolated from
results of laboratory tests on animals (e.g., rats, mice). Toxicologists
are thus faced with many uncertainties when evaluating the potential human
health risks associated with intake of toxic chemicals. Regulatory decisions
must be made despite these uncertainties. Many assumptions and subjective
judgments may enter into an evaluation of human health risk. The risk
assessment approach provides a framework for consistent, systematic estimation
of health risks, with clear statements of assumptions and uncertainties.
12
-------
As noted by Kneip (1983) and Peddicord (1984), many Investigators
have evaluated bioaccumulation data relative to human health concerns simply
by comparing tissue concentrations of selected chemicals to action levels
or tolerances established by U.S. FDA (1982, 1984). This approach is severely
limited for the following reasons:
• U.S. FDA limits are available for only a few chemicals (mercury
and approximately 13 organic compounds).
« U.S. FDA has not established regulatory limits for some
of the most potent suspected human carcinogens (e.g., 2,3,7,8-
tetrachlorodibenzodioxin) or for some of the cannon contaminants
in Puget Sound (e.g., PAH, As).
t Action levels and tolerances were intended to be used only
for regulation of interstate commerce of food products.
• When setting regulatory limits, U.S. FDA considers economic
impacts of food regulation as well as potential human health
risk (U.S. FDA 1984). When using U.S. FDA limits to interpret
bioaccumulation data, investigators implicitly adopt economic
policies of U.S. FDA. Thus, risk management issues are
not clearly separated from risk assessments.
Use of regulatory limits on toxic chemicals in food products established
by other countries (Nauen 1983) would suffer from many of the limitations
listed above for U.S. FDA values. Moreover, a concise review of the basis
for each of these limits is not available.
USES OF RISK ASSESSMENT AND MANAGEMENT
Uses of risk assessment and risk management in the Puget Sound Estuary
ogram may include the following:
13
-------
• Toxic chemical problem identification and ranking
• Environmental criteria or guidelines development
• Public information and advisories.
The first two uses fall within the general category of regulatory decision-
making. In this context, one goal of the U.S. EPA is to define, identify,
and set priorities for reducing unacceptable risks. Risk assessment and
management provide a framework for balanced analysis of environmental problems
and consistent policies for reducing health risks.
In the Puget Sound Estuary Program, risk assessment can be used to
identify and rank environmental problems in several ways. First, locations
can be ranked according to the risks associated with consuming seafood
from them. Such assessments have already been conducted for the Commencement
Bay waterways (e.g., Nicola et al. 1983; Tetra Tech 1985d; Versar 1985).
Extension of the analysis to multiple bays and the main basins of Puget
Sound would provide a broad geographic overview of the condition of recrea-
tionally harvested seafoods. Second, priority chemicals can be identified
according to associated health risks. Finally, different seafood species
and weight classes within species can be ranked according to relative risks.
Risk assessment is an important analytical method for developing environ-
mental criteria and guidelines. For example, water quality criteria derived
by U.S. EPA (1980b) are based in part on human health risk assessment.
U.S. FDA considers potential human health risks as well as economic factors
in developing tolerance levels for chemical contaminants in fishery products
(U.S. FDA 1984). Guidelines on maximum advisable contaminant concentrations
in recreational ly harvested species were established for use in the Puget
Sound Estuary Program by using risk assessment models (Tetra Tech 1986c).
The results of risk assessments conducted under the Puget Sound Estuary
Program will be used to inform the public about the relative health risks
of various seafood species and geographic locations. Providing the recreational
public with such information allows for individual choice in determining
14
-------
harvest area, target species, consumption rates, and other factors based
on relative risk. Furthermore, risk management by federal, state, or local
agencies may include:
o Reducing exposure potential by implementing pollution controls
a Prohibiting seafood harvests by geographic area or by species
o Issuing public advisories or controls to limit:
Geographic area of harvesting
Harvest season
Harvest methods
Species harvested
Catch number
Size range harvested
Consumption rate.
The risk management option selected should depend on the specific problem
and the estimated level of risk (Pollock, G., 13 June 1986, personal com-
munication) .
15
-------
HAZARD IDENTIFICATION
Tha first step in the risk assessment process is to define toxicological
hazards posed by the individual chemical contaminants in the seafood samples.
These hazards are defined by constructing a toxicity profile for each contam-
inant of concern. To
-------
TABLE 1. ORGANIC PRIORITY POLLUTANTS WITH ESTABLISHE&
TOXICOLOGICAL INDICES RANKED IN ORDER OF THEIR
OCTANOL-WATER PARTITION COEFFICIENTS (K )
Priority
Pollutant No.
Ill
73
107
92
110
129
94
91
106
112
93
53
90
9
100
39
68
64
98
109
52
66
108
12
103
102
7
105
21
97
96
95
26
27
25
113
38
62
31
28
89
37
85
60
6
Substance
PCB-1260
benzo(a)pyrene
PCB-1254
4, 4 '-DOT
PCB-1248
TCOD (dloxin)
4, 4' -ODD
chlordane
PCB-1242
PCB-1016
4, 4 ''DOE
hexachlorocycl open tad iene
dieldnn
hexachlorobenzene
heptachlor
fluoranthene
di-n-butyl phthalate
pentachlorophenol
endrin
PCB-1232
hexachl orobutad iene
bis(2-ethylhexyl)phthalate
PCB-1221
hexachl oroethane
beta-HCH
alpha-HCH
chlorobenzene
gamma-HCH
2,4,6-trichlorophenol
endosulfan sulfate
beta-endosulfan
alpha-endosulfan
1 , 3-d ichl orcbenzene
1,4-dichlorobenzene
1,2-dir.hlorobenzene
toxaphene
ethylbenzene
N-nitrosodiphenylamine
2,4-dicnlorophenol
3,3'-dichlorobenzidine
aldrin
1,2-diphenylhydrazine
tetrachloroethene
4,6-dinitro-o-cresol
tetrachloromethane
Puget Sound
Contaminant
of Concern3
yes
no
yes
yes
yes
no
yes
no
yes
yes
yes
no
no
yes
no
no
yes
no
no
yes
yes
yes
yes
no
no
no
yes
PO
no
no
no
no
yes
yes
yes
no
no
no
no
no
no
no
no
no
no
Tox icological — «
Index"
CPF
CPF
CPF
CPF
CPF
CPF
CPF
CPF
CPF
CPP
CPF
RfD
CPF
CPF
CPF
RfD
RfO
RfD
RfD
CPF
CPF
RfO
CPF
CPF
CPF
CPF
RfO
CPF
CPF
RfD
RfD
RfD
RfD
RfO
RfD
CPF
RfD
CPF
RfD
CPF
CPF
CPF
CPF
RfD
CPF
logK
6.91
6.50
6.48
6.19
6.11
6.10
6.02
6.00
6.00
5.88
5.69
5.51
5.48
5.47
5.45
5.33
5.15
5.00
4.56
4.48
4.28
4.20
4.00
3.93
3.35
3.85
3.79
3.72
3.69
3.60
3.60
3.60
3.48
3.38
3.38
3.30
3.15
3.13
3.08
3.02
3.01
2.94
2.88
2.85
2.64
17
-------
TABLE 1. (Continued)
Priority
Pollutant No.
42
11
87
15
86
14
4
35
33
23
56
5
54
71
59
29
65
10
70
44
3
18
46
2
45
88
61
Substance
bis(2-chloroisopropyl) ether
1,1,1-trichloroethane
trichloroethene
1,1,2,2-tetrachloroethane
toluene
1,1,2-trichloroe thane
benzene
2,4-dinitrotoluene
1,3-dichloropropene
chloroform
nitrobenzene
benzidine
isophorone
dimethyl phthalate
2,4-dini trophenol
1,1-dichloroethene
phenol
1,2-dichloroethane
diethyl phthalate
dichloromethane
acrylonitrile
bis(2-chl oroethy 1 ) ether
bromomethane
acrolein
chloromethane
vinyl chloride
N-nitrosodimethylamine
Puget Sound
Contaminant
of Concern3
no
no
yes
no
no
no
no
no
no
no
no
no
no
yes
no
yes
no
no
yes
no
no
no
no
no
no
no
no
lexicological
Indexb
RfD
RfD
CPF
CPF
RfD
CPF
CPF
CPF
RfD
CPF
RfD
CPF
RfD
RfD
RfD
CPF
RfD
CPF
RfD
CPF/RfD
CPF
CPF
RfD
RfD
RfD
CPF
CPF
log K
3 ow
2.58
2.47
2.42
2.39
2.21
2.18
2.11
2.00
1.98
1.90
1.83
1.81
1.67
l'.=>3
1.48
1.46
1.45
1.40
1.30
1.20
1.12
1.00
0.90
0.90
0.60
-0.58
a As determined hy Konasewich et al. (1982).
b Carcinogenic potency factors (CPF) and Reference Doses (RfD) published by the U.S. EPA i!98Cc
1985a; 1986). See Tables 4 and 5 below for values.
18
-------
TABLE 2. INORGANIC PRIORITY POLLUTANTS WITH
ESTABLISHED TOXICOLOGICAL INDICES
Priority
Pollutant No
115
118
119
119
119
123
124
127
114
117
121
124
125
126
Puget Sound
Contaminant
•Substance of Concern9
arsenic
cadmium
chromium VI
chromium VI
chromium III
mercury
nickel
thai 1 mm
antimony
beryllium
cyan ide
nickel (subsulfide, refinery dust)
selenium
silver
yes
yes
no
no
no
yes
no
no
no
no
no
no
yes
yes
Toxicological
Index6
CPF
CPFd
CPFd
RfO
RfO
RfD
RfO
RfO
RfD
CPFd
RfD_
CPFd
RfO
RfO
Log BCFC
2.544
2.513
2.190
2.190
2.104
2.000
1.699
1.176
NO
NO
NO
NO
NO
NO
a A* determined by Konasewich et al. (1982).
b Carcinogenic potency factors (CPF) and Reference Doses (RfD) published by the U.S. EPA (198C:
1985a; 1986,. See Tables 4 and 5 below for values.
c BCF = Bioconcentration Factor (U.S. F.PA 1980b; Tetra Tech 1985a).
NO = No data.
These metals are not considered carcinogenic by the dietary route of exposure.
19
-------
The initial list of contaminants considered in this evaluation included
all U.S. EPA priority pollutants and additional chemicals of concern in
Puget Sound identified by Konasewich et al. (1982). Each contaminant on
this initial list was evaluated in terms of bioaccumulation potential and
availability of lexicological indices (Reference Oose or carcinogenic potency
factor).
Quantitative risk assessments can be conducted only for the chemicals
with toxicological indices (Tables 1 and 2). The organic chemicals are
listed in Table 1 in descending order of bioaccumulation potential, according
to octanol-water partition coefficient (Tetra Tech 1985a). Metals are
listed in Table 2 in descending order of bioaccumulation potential, according
to bioconcentration factor (see Tetra Tech 1985a). Toxicological indices
are not available for any of the nonpriority pollutants identified by Konasewich
et al. (1932) (i.e., polychlorinated dibenzofurans, chlorinated butadienes,
methylated naphthalenes, methyl benzo(a)anthracene, benzo(ghi)fluoranthene,
benzo(i)fluoranthene, and methyl fluoranthene).
Further screening of the list of high priority chemicals is possible
based on preliminary risk analysis. For example, some of the chemicals
listed in Tables 1 and 2 have relatively low toxicity. Only extremely
high concentr?fions (e.g., >100 ppm) in seafood would cause concern, assuming
a very high seafood consumption rate of 150 g/day (0.33 Ib/day) for 70
yr (for discussion of consumption rates, see section on "Exposure Assessment,
Exposed Population Analysis"). Chemicals falling into this category include:
• Toluene
o 1,1,1-trichloroethane
• Chloromethane
o Nitrobenzene
o Bis(Z-ethylhexyl) phthalate
20
-------
t Di-n-butyl phthalate
• Chromium III
• Dimethyl phthalate
• Diethyl phthalate.
Lower consumption rates would further restrict the list of chemicals of
concern. However, further screening of contaminants of concern should
be done on a case-by-case basis during preparation of actual risk assessments.
Also note that carcinogenicity of some phthalates on the list above is
presently being evaluated.
TOXICITY PROFILES
Toxicity profiles are constructed for the selected chemicals of concern
by summarizing the following information:
• Physical-chemical properties (e.g., vapor pressure, octanol-water
partition coefficients)
• Metabolic and pharmacokinetic properties (e.g., metabolic
degradation products, depuration kinetics)
• lexicological effects for specific uptake routes (e.g.,
target organs, cytotoxicity, carcinogenicity, mutagemcity).
Toxicity profiles are available for approximately 195 chemicals from U.S. EPA
(Office of Waste Programs Enforcement and Office of Environmental Criteria
and Assessment; see Appendix A, Table A-l).
The key elements of a hazard identification should be summarized in
a concise tabular format. The examples shown in Table 3 illustrate the
21
-------
TABLE 3. TOXICITY PROFILE FOR MERCURY AND PCBSa
Property
Mercuryb
PCBs
CAS Number
Physical-Chemical
Molecular Weight
Vapor Pressure (mm Hg)
Solubility (mg/L)
Loo K
Log Bioconcentration Factord
Carcinogenic Status
Acute Toxicity
Human (mg/kg body wt)
Mammal (mg/kg bodj wt)
Aquatic (mg/L)
Chronic Toxicological Effects
Humans
Mammals
Aquatic Organisms
7439-97-6
200.6-318.7
0.012-0.028
0.056-400,000
N/Ae
2.0-4.6
None arcinogen
299
1.0-40.9
0.015-32.0
1336-36-3
154.2-498.7
2.8 x 10-9 - 7.6 x 10-5
<0.001-5.9
4.0-6.9
1.9-5.2
Probable human careinogenf
-- Sufficient animal evidence
— Inadequate human evidence
1,010-16,000
0.001-61.0
Motor and sensory impairment Skin lesions, liver dysfunctions
leading to paralysis, loss and sensory nsuropaeny.
of vision and hearing, and
death. Kidney dysfunction.
Reproductive impairment and
teratogenic effects.
Hepatotoxicity, fetotoxic ity , s<
lesions, and hepatocel lular car
Developmental and structural Reproductive and developmental
anomalies, suppression of impairment.
growth and reproduction,
impairment of behavior.
« This is an example toxicity profile and is not intended to be comprehensive.
b Mercury may occur in its elemental form, as inorganic salts, or as organic complexes. Conse-
quently, the chemical and toxicological properties vary tremendously depending on the degree
of complexation or metal speciation.
-------
TABLE 3. (Continued)
c Physical-chemical properties and toxic ity vary according to the degree of chlorine substitution
the number of adjacent unsubstituted carbons and steric configuration.
d Tetra Tech (1985a).
e N/A - not applicable.
f U.S. EPA (1980b, 1985a); IARC 1978.
9 For mercury (II) choride via oral route of exposure (Tatken and Lewis 1983).
23
-------
kinds of information used to evaluate toxicological hazards. A suggested
format for display of key information is also illustrated. Neither toxicity
profile is intended to be complete.
Information in a toxicity profile is used to establish the weight
of evidence for how likely a chemical is to cause a given health effect.
U.S. EFA is developing a weight-of-evidence classification scheme for carcino-
gens (see U.S. EPA 1984a). The U.S. EPA classification scheme will be
an adaptation of che approach used by the International Agency for Research
on Cancer and will include the following categories:
o Group A - Carcinogenic to Humans
o Group B - Probably Carcinogenic to Humans
o Group C - Possibly Carcinogenic to Humans
o Group D - Not Classifiable as to Human Carcinogenicity
o Group E - No Evidence of Carcinogenicity for Humans.
Criteria for each category are given by U.S. EPA (1984a). At present,
a general evaluation of evidence for carcinogenicity is available for each
chemical assigned a carcinogenic potency factor by U.S. EPA (1985a) (see
below, "Dose Response Assessment," "Carcinogenic Potency factors").
SOURCES OF INFORMATION
The primary sources of toxicity profiles are the U.S. EPA Office of
Waste Programs Enforcement and Office of Health ?nd Environmental Assessment
(e.g., Appendix A, Table A-l). Additional sources are shown in Appendix A,
Table A-2.
Supplementary information for toxicity profiles may be obtained from
bibliographic or chemical/toxicological databases. DIALOG, a comprehensive
bibliographic database system (Dialog Information Services, Inc., 3460
24
-------
Hillview Avenue, Palo Alto, CA 94304), offers access to databases such
as Pollution Abstracts, National Technical Information Service, and ENVIROLINE.
Chemical and toxicological Information can be obtained from the databases
listed in Appendix A, Table A-3.
Supplementary information for toxicity profiles may also be obtained
from Important references such as Lyrnan et al. (1982) and Callahan-et al.
(1979). Other key sources tnat are periodically updated are the Registry
of Toxic Effects of Chemical Substances (e.g., Tat ken and Lewis 1983) and
the Annual Report on Carcinogens (e.g.. National Toxicology Program 1982).
25
-------
DOSE-RESPONSE ASSESSMENT
After the potential hazard associated with each contaminant of concern
is characterized, the relationship between dose of a substance and its
biological effect is determined. Dose-response data are used to determine
the toxicological potency of a substance, a quantitative measure of its
potential to cause a specified biological effect. The concepts of exposure,
dose, dose-response relationship, and index of toxicological potency a're
discussed in the following sections.
EXPOSURE AND DOSE
The concepts of exposure and dose, as defined below, are central to
risk assessment:
• Exposure: Contact by an organism with a chemical or physical
agent
• Dose: The amount of chemical uptake by an organism over
a specified time as a consequence of exposure.
The "ingested dose," or amount of chemical ingested, is distinct from the
"absorbed dose," or amount of chemical actually assimilated by absorption
across the lining of thp gastrointestinal system. Exposure level or exposure
concentration is used to denote the concentration (mg/kg wet weight) of
contaminant in seafood. As shown later, the absorbed dose is estimated
from seafood consumption rate, the exposure concentration, and an absorption
coefficient (see "Exposure Assessment").
GENERAL DOSE-RESPONSE RELATIONSHIPS
The form of the dose-response relationship for carcinogens is fundamentally
different from that for noncarcinogens. Examples of general dose-response
26
-------
relationships are shown in Figure 1. The lack of a demonstrated threshold
in dose-response relationships for carcinogens (U.S. EPA 1980b, 1984a;
U.S. Office of Science and Technology Policy 1985) implies a finite risk
of cancer even at very low doses of the carcinogen. Therefore, the dose-
response relationship is used to predict an upper-limit estimate of the
probability (risk) that a given exposure level will result in cancer.
For noncarcinogenic effects, there is usually a threshold dose below which
no adverse biological effects are observed in the animal bioassay. This
threshold dose is termed the "No-Observed-Adverse-Effect-Level" (NOAEL),
as shown in Figure 1.
A measure of toxicol ogical potency is derived from the dose-response
relationship for the chemical of interest using a data set for the most
sensitive species. Data are evaluated by U.S. EPA to ensure high quality
(e.g., U.S. EPA 1980b; 1985a). Toxicological potency indices for two broad
categories of toxicants are defined as follows:
t Carcinogens are each characterized by a Carcinogenic Potency
Factor, a measure of the cancer-causing potential of a substance
estimated by the upper 95 percent confidence limit of the
slope of a straight line calculated by the linearized multistage
model or another appropriate model
• Noncarclnogens are each characterized by a Reference Dose
(RfD) value, the highest average daily exposure over a lifetime
that would not be expected to produce adverse effects.
RfDs were previously known by U.S. EPA (1980b) and others as Acceptable
Daily Intakes.
Carcinogenic potency factors, RfDs, and methods for deriving them
are presented in the following sections. U.S. EPA Region 10 will rely
on carcinogenic potency factors and RfD values derived by U.S. EPA program
offices concerned specifically with human health risk assessment. At present,
values for these toxicol ogical indices are being standardized for agency-
27
-------
crt
1C
o
u.
O
U
UJ
o
Ul
DC
LOW-DOSE
REGION OF
CONCERN
DOSE OF CARCINOGEN
OBSERVED DATA POINTS
• CHEMICAL A
A CHEMICAL B
• CHEMICAL C
MODELS
———— Low dose
extrapolation
—— Models lit within
observed data range
o
i
Ul
ZX
iu o
3 H
o
Ul
Rfd
,»... up.
NOAEL
DOSE OF NONCARCINOGEN
Frequency •
RID.
Proportion of
animals tested
Reference Risk
Dose
UF - Uncertainty Factor
NOAEL - No Observed
Adverse Effects
Level
DOM • Ingested Dose
Figure 1. Hypothetical example of dose-response curves for a
carcinogen and a noncarcinogpn.
28
-------
wide use. The brief overview of derivation methods below is presented
for background information only.
CARCINOGENIC POTENCY FACTORS
The Carcinogen Assessment Group of U.S. EPA currently uses the linearized
multistage model (U.S. EPA 1980b, 1984a, 1985a) to derive carcinogenic
potency factors. The multistage model assumes that carcinogenesis results
from a series of interactions between the carcinogenic chemical and ONA,
with the age-specific rate of interactions linearly related to dose. For
example, a chemical may induce a mutation in the DNA of a cell to initiate
carcinogenesis. However, a series of further interactions between DNA
and the same chemical (or another one) may be necessary to promote carcino-
genesis and induce a tumor. The multistage model is one of several biologically
realistic models. It is the model most frequently used when there is no
convincing oiological evidence to support application of an alternative
model. Other models include the logit, probit, single-hit, and Weibull
models (Food Safety Council 1980, 1982; Hogan and Hoel 1982; Cothern et
al. 1986). Pt high doses (corresponding to lifetime risks greater than
about 10~2), all currently used models yield similar risk estimates. Below
risks on the order of 10"2, the models diverge increasing1" as dose declines.
In general, the linearized multistage model predicts risks similar to the
single-hit model. For many data sets, both of these models yield higher
estimates of low-dose risk than do other models (U.S. EPA 1980b, 1984a;
Hogan and Hoel 1982; U.S. Office of Science and Technology Policy 1985).
The mathematical form of the multistage model for a specified carcinogen
is:
R(d) = 1 - exp [-(qjd + q2d2 + ••• + qkd><)] (1)
where:
R(d) = Excess lifetime risk of cancer (over background at dose d)
(dimensionless)
qi values = Coefficients [kg-day-mg~l (i.e., the inverse of dose units)]
29
-------
d = Dose (mg.kg'1-da., ^)
k = Number of stages in carcinogenesis.
In general, a linearized form of the multistage model is appropriate for
risks less than approximately 10~2 (i.e., one excess tumor per 100 exnosed
individuals). The linearized multistage model is:
R(d) = qi.d (2)
with terms defined as above.
To derive a carcinogenic potencv factor, either the original multistage
model or its linearized form is fit to dose-response data (e.g., Figure 1).
The upper 95 percent confidence limit of the first coefficient (qi*) ™
Equation 1 is then used as a plausible-upper-limit estimate of carcinogenic
potency (i.e., the carcinogenic potency factor) (Table 4). The use of
these values in estimating a plausible upper-limit to cancer risk is discussed
below (see "Risk Characterization").
If a potency factor is derived from nonhuman data, as is usually the
case, it must be extrapolated to humans. Before being applied to humans,
carcinogenic potency factors derived from animal data are corrected using
surface area differences between bioassay animals and humans (U.S. EPA
1980b; 1984a). The rationale for using surface area extrapolations is
detailed in Mantel and Schneider-man (1975).
The main source of dose-response data for carcinogens is lifetime
cancer bioassays performed on rats or mice. Because most of these experiments
are designed for cost-effective assessment of tumor incidence, doses in
bioassays may be orders of magnitude above those encountered in the human
environment. High doses are used in laboratory bioassays for several reasons:
1) to reduce the time required to produce a response and thus overcome
problems related to cancer latency periods (e.g., rat lifetime is about
2 yr, human lifetime is assumed to be about 70 yr), 2) to obtain sufficient
statistical power to detect tumors, and 3) to decrease thu absolute number
of ammalc ed and thereby reduce costs. Doses in animal bioassays
30
-------
TABLE 4. CARCINOGENIC PRIORITY POLLUTANTS
RANKED BY POTENCY FACTORS
PP# Pollutant
129 TCOO (dioxin)
5 benzidine
119 chromium VIC
90 dieldrin
61 N-nitrosodimethylamine
115 arsenic
73 benzo(a)pvrene
89 aldrin
102 alpha-HCH
118 cadmium0
106 PC8-1242
107 PCB-l2l54
108 PCB-1221
109 PC8-1232
110 PCB-1248
111 PCB-1260
112 PCB-1016
100 heptachlor
117 beryllium0
103 beta-HCH
28 3,3'-dichlorobenzidine
9 hexachlorobenzene
91 chlordane
105 gamma-HCH
29 1,1-dichloroechene
18 bis(2-chloroet'iyl)ether
113 toxaohene
124 nickel (subsulfide,
refinery dust)c
37 1,2-diphenylhydrazine
92 4,4'-ODT
93 4, 4 '-ODE
94 4.4'-ODD
3^ 2,4-diriitrotoluene
3 acrylomtrile
15 1,1,2,2-tetrachloroethane
6 tetrachloromethane
10 1,2-dichloroethane
52 hexachlorobutadiene
23 chloroform
14 1,1,2-trichloroethane
85 tetrachloroethene
4 benzene
21 2,4,6-trichlorophenol
88 vinyl chloride
12 hexachloroethane
CAS Number
1746-01-6
92-87-5 .
7440-47 -3a
60-57-1
62-75-9 .
7440-38-2d
50-32-8
309-00-2
319-84-6 .
7440-43-9d
53469-21-9
11097-69-1
11104-28-2
11141-16-5
12672-29-6
11096-82-5
12674-11-2
76-44-8 .
7440-41 -7d
319-85-7
91-94-1
118-74-1
57-74-9
58-89-9
75-35-4
111-44-4
8001-35-2
*j
7440-02-0a
122-66-7
50-29-3
72-55-9
72-54-8
121-14-2
107-13-1
79-34-5
56-23-5
107-06-2
87-68-3
67-66-3
79-00-5
127-18-4
71-43-2
88-06-2
75-01-4
67-72-1
Level of Evidence0
Potency3 Humans Animals
156000.00000
234.00000 (W)
41.00000 (W)
30.40000
25.90000 (B)
15.00000 (H)
11.50000
11.40000
11.12000
6.10000 (W)
4.34000
4.34000
4.34000
4. 34000
4.34000
4.34000
4.34000
3.37000
2.60000
1.84000
1.69000
1.67000
1.61000
1.33000
1.16000 (I)
1.14000
1.13000
1.05000 (W)
0.77000
0.34000
0.34000
0.34000
0.31000
0.24000 (W)
0.20000
0.13000
0.09100
0.07750
0.07000
0.05730
0.05100
0.02900 (W)
0.01990
0.01750 (I)
0.01420
I
S
s
I
I
s
I
I
I
L
I
I
I
I
I
I
I
I
L
I
I
I
I
I
I
I
I
S
I
I
t
i
I
L
I
I
I
I
I
I
I
S
I
S
I
s
s
s
s
s
I
s
L
S
s
s
5
S
S
S
S
s
s
s
L
S
s
L
L
I.
S
S
S
s
s
5
S
i
s
L
5
S
L
S
L
L
S
S
S
L
31
-------
TABLE 4. (Continued)
PP# Pollutant
87 trichloroethene
62 N-nitrosodiphenylamine
44 dichlcromethane
CAS Number
79-01-6
36-30-6
75-09-02
Potency3
0.01100
0.00492
0.00063 (I)
Level of
Human s
I
I
I
Evidence"
Animal s
L/S
S
L
a From U.S. Environmental Protection Agency (198Sa), Table 9-66. All slopes calculate:
upper 95 percent confidence limit of slope (qj.*) based on animal oral data and multistage T.oc
except:
(B) = slope calculated from 1-Hit model
(U) = slope calculated from occupational exposure
(H) = slope calculated from human drinking water exposure
(I) = slope calculated from animal inhalation studies.
b S = Sufficient evidence; L = Limited evidence; I = Inadequate evidence.
c Chromium (VI), cadmium, beryllium, and nickel are not considered to be carcinogenic via ci=:i
exposure.
CAS numbers for these substances vary depending on whether they occur in cheir 9l
form, as inorganic- salts, or as organic complexes.
32
-------
for oral uptake of contaminants are usually the administered (ingested)
dose, not the absorbed dose (i.e., uptake across the lining of the gastro-
intestinal system).
REFERENCE DOSES
Current methods for predicting human health effects from exposure
to nonearcinogenic chemicals rely on the concept of a Reference Dose (RfD)
(Vettorazi 1976, 1980; U.S. EPA 1980b). The RfD is the highest average
daily dose that is considered safe or acceptable over a lifetime of exposure.
The RfD is derived from an observed threshold dose (e.g., No-Observed-Adverse-
Effect-Level) in a chronic animal bioassay by applying an uncertainty factor,
which usually ranges from 1 to 1,000 (Dourson and Stara 1983). Derivation
of an example RfD from dose-response data is illustrated in Figure 1.
The uncertainty factor accounts for differences in threshold doses among
species, among intraspecies groups differing in sensitivity, and among
toxitity experiments of different duration. Dourson and Stara (1983) discuss
the methods for deriving RfDs and the criteria for selecting uncertainty
factors.
P.fD values are provided in Table b. Note that these values are largely
current published values. An effort is underway at U.S. EPA to standardize
RfDs throughout the agency. Some of these current values may be revised
and published as part of new rule-making (e.g., U.S. EPA 1985b).
SOURCES OF INFORMATION
Current values for carcinogenic potency factors are given in Table 4.
Current values for RfDs are given in Table 5. Before using these values,
investigators should verify that they are still the most current values.
Verification sources are discussed in the following sections. Addresses
for information sources are given ir, Appendix A, Table A-4.
33
-------
TABLE 5. REFERENCE DOSE (RfD) VALUES FOR PRIORITY POLLUTANTS
PP# Pollutant
126 silver
123 mercury
60 4,6-dimtro-o-cresol
127 thallium
42 bis(2-chloroisopropyl)ether
98 endrin
59 2,4-dinitrophenol
33 1 ,3-dichloropropene
119 chromium VI
95 alpha-endosul fan
96 beta-endosulfan
97 endosulfan sulfate
114 antimony
39 fluoranthene
53 hexachlorocycl open tad iene
125 selenium
25 1,2-dichlorobenzene
26 1,3-dichlorobenzene
27 1,4-dichlorobenzene
7 chlorobenzene
2 acrolein
45 bromomethane
124 nickel
38 ethylbenzene
64 pentachlorophenol
31 2,4-dichlorophenol
65 phenol
121 cyanide
54 isophorone
44 dichloromethane
86 toluene
11 1 ,1,1-tr ictiloroethane
45 chlorometnane
56 nitrobenzene
66 bis(2-ethylhexyl)phthalate
68 di-n-butyl phthalate
119 chromium III
71 dimethyl phthalate
70 diethyl phthalate
CAS #
7440-22-4*
7439 -97 -6a
534-52-1
563-68-8a
39638-32-9
72-20-8
51-28-5
10061-02-6
7440-47 -3a
115-29-7
115-29-7
1031-07-8
7440-36-0*
206-44-0
77-47-4
7782-49-2
95-50-1
541-73-1
106-46-7
108-90-7
107-82-8
74-83-9
7440-02-0*
100-41-4
87-86-5
120-83-2
108-95-2
57-12-5a
78-59-1
75-09-02
108-88-3
71-55-6
74-87-3
98-95-3
117-81-7
87-74-2
7440-47-3*
131-11-3
84-66-2
RfD
nig/ day
0.016
0.1
0.027
0.04
0.070
0.070
0.14
0.175
0.175
0.28
0.28
0.28
0.29
0.4
0.418
0.7
0.94
0.94
0.94
1.008
1.100
1.5
1.5
7
2
7.0
7
2
10.5
4
20
37.5
38
0.03
42
88
125
700
875
RfO
mg/kg/day
0.0002
0.002
0.0004
0.0004
0.001
0.001
0.002
0.002
0.002
O.C04
0.004
0.004
0.004
0.006
0.006
0.01
0.01
0.01
0.01
0.01
0.02
0.02
0.02
0.1
0.03
0.1
0.1
0.02
0.150
0.06
0.3
0.5
0.5
0.0005
0.6
I
2
10
10
Criteria
Page
C-125
*
C-93
*
C-61
B-12
C-92
C-27
C-34
C-87
C-87
C-87
C-70
C-47
C-63
C-66
C-64
C-61
C-64
C-20
C-53
*
*
C-32
*
*
C-20
*
*
C-77
*
C-57
C-57
C-57
C-57
* CAS numbers for these substances vary depending on their specific form (e.g., inorganic 5.= Its
or organic complexes.
Asterisk indicates that values art .erified RfDs from U.S. EPA (1986).
Reference: U.S. EPA (1980b). Priority pollutant numbers are shown ir firs: column of tio'f.
For each RfD, page citation for corresponding Acceptable Daily Intake value from a Water Quali;..
Criteria document is shown in last column. Blanks in page citation column indict'? £*>»; --'~
values are errata to water quality criteria (U.S. EPA, 8 August 1984, personal communicaf-on;
34
-------
Carcinogenic Potency Factors
Most of the carcinogenic potency factors calculated by the U.S. EPA
Office of Health and Environment are published in each health assessment
document produced by the office (e.g., U.S. EPA 1985a). The U.S. EPA Carcinonen
Assessment Group determines these carcinogenic potency values _and updates
them periodically. Information on U.S. EPA careinogenic potency values
is presently being compiled into a database. Therefore, the sources just
mentioned must be contacted to obtain new information as it is released.
Reference Doses
A review of RfOs is in progress by a U.S. EPA work group (U.S. EPA
1986). RfOs and supporting data will be entered into a computerized database
accessible to U.S. EPA regional offices through an electronic mail system.
An example of an RfD data sheet for the pesticide aldrin is shown in Appendix
B. The data sheet provides information on endpoint (biological effect),
experimental data sets, doses, uncertainty factors, additional modifying
factors, confidence in the RfD, reference documentation, and dates of agency
RfD reviews.
35
-------
EXPOSURE ASSESSMENT
Exposure assessment is the process pf-eh"aracter\z.ing-the populations
exposed to the chemicals of concern, the environmental transport and fate
pathways, and the magnitude and duration of the exposure dose (U.S. EPA
1984b). For risk assessment of contaminated seafood, the following factors
should be considered:
o Concentrations of contaminants in marine biota
o Food chain transfer of contaminants from marine species
to humans
o Characteristics of exposed human populations
o Numerical model coefficients (e.g., seafood consumption
rate, contaminant absorption efficiency).
An exposure assessment is performed in three stages. First, average
concentrations of contaminants in seafood are estimated, usually based
on chemical analyses of tissue samples. Second, the exposed population
is characterized (including seafood consumption rate). Third, informa-
tion on contaminant concentrations and the exposed population is combined
in an integrated analysis to construct an exposure profile.
TISSUE CONCENTRATIONS OF CONTAMINANTS
For each contaminant of concern, data on concentrations in edible
tissues of marine biota are obtained. Selection of analytical detection
limits, selection of target species, and statistical treatment of tissue
concentration data are discussed in the following sections.
36
-------
Selection of Analytical Detection Limits
Guidance on method detection limits for analytical protocols may be
developed using risk assessment models explained later (see "Risk Character-
ization"). The analytical chemistry methods should be sufficient to detect
a chemical concentration associated with a certain minimum risk level (e.q.,
10"5 or 10"6 individual lifetime-risk; see below "Presentation and Interpre-
tation of Results, Risk Comparisons"). Other factors may dictate choice
of a lower detection limit. For example, routine analytical methods may
attain much lower limits than required by the specified risk level. Also,
lower detection limits may be desired if an objective of the study is to
develop baseline bioaccumulation data as well as health risk data. In
some cases (e.g., 2,3,7,8-tetrachlorodibenzodioxin, benzidine, dieldrin,
N-nitrosodimethylamine) the minimum detection limit that can be achieved
with current technologies corresponds to a plausible-upper-1imit risk that
is substantially above risk levels of concern (e.g., 10"5 to 10"6).
Selection of Target Species
Ideally, the set of species selected for contaminant analysis would
include all harvested species. Because available data and funds for collecting
new data are often limited, selected marine species may be used for human
health risk assessment. The particular marine species selected for a risk
assessment will depend on the study objectives. An example of approaches
and guidance on selection of target species is given below.
Dominant Harvested Species--
If adequate data are available for characterizing fisheries catches
and consumption from field surveys (e.g., Landolt et al. 1985; McCallum
1985), then selected species could include the dominant members of the
catch on a wet-weight basis. For Puget Sound embayments studied by Landolt
et al. (1985), dominant components of the recreational harvest in rank
order fron largest to smallest catch are listed below. Numbers in parentheses
are the percent weight distribution for the total catch (4.013.6 kg).
37
-------
• Unidentified species (26)
0 Market squid, Loligo opalescens (12.5)
0 Chinook salmon, Oncorhynchus tshawytscha (12.4)
• Coho salmon, Oncorhynchus kisutch (8.3)
• Unidentified salmon, Oncorhynchus spp. (7.0)
• Pacific hake. Merluccius-jroductus (6.5J
t Pacific cod, Gad us macrocephaVus ("6.5)
• Pile perch, Rhacochilus vacca (4.5)
t Walleye pollock, Theragra chalcogramma (3.1)
• Striped perch, Embiotoca lateralis (2.5)
• Sablefish, Anoplopoma fimbria (2.4)
• Unidentified flatfish, Pleuronectidae, Bothidae (1.2)
• Unidentified rockfish, Sebastes spp. (1.0)
• Pacific tomcod, Microgadus proximus (1.0)
Other species groups each contributed less than 1 percent of the weight
of the catch (Landolt et al. 1985). In general, the species listed above
were also found to be romponents of the Puget Sound recreational catch
surveyed by McCallum (1985). Other species known to be important recreational
resources, at least in selected areas of Puget Sound, are bivalve molluscs,
Cancer crabs, and various algal species.
The advantages of choosing the dominant harvested species for risk
assessment are that:
• The risk estimates will be based on realistic conditions
in terms of relative weight of species in the diet, presuming
that catch data reflect consumption patterns
• Adequate numbers of organisms for chemical analyses will
be easy to obtain.
The disadvantages of this approach are that:
• Minor components of the diet by weight that are highly
contaminated may be overlooked
38
-------
o Dominant species often vary spatially, making it difficult
to compare risk estimates for different sites-
o Extensive- species-specific data on catch, consumption, and
contamination patterns are needed (costly to obtain if not
already available)
o A major component of the catch may be unidentifiable because
the catch is sometimes cleaned before being surveyed.
Indicator Species--
Use of selected indicator species is an alternative to the use of
dominant harvested species. Indicator species are chosen to represent
the average (or maximum) contamination levels in the harvest. Use of indicator
species is appropriate for investigations with multiple objectives (e.g..,
assessment of bioaccumulation in marine species and human health risks
for specific areas within a bay). If small-scale discrimination of spatial
patterns of contamination is a concern, target species should include non-
migratory biota or species that show minimal movement within the nearshore
area (e.g., bivalve molluscs and English sole). Phillips (1980) and Tetra
Tech (1985b) provide criteria for selecting target species for bioaccumulation
surveys.
In many cases, the selected target species may be associated with
soft-sediment substrates. Contact with sediments by such species may lead
to body burdens of contaminants that are high relative to those in pelagic
organisms of similar lipid content and size. However, the relative contam-
ination on a wet-weight basis is difficult to predict without extensive
data. As shown by Tetra Tech (1986c), English sole may be used as an indicator
of the order-of-magnitude contaminant levels that would be expected in
edible tissues of pelagic fish species. However, relative contamination
among species may vary among bays. For example, in Commencement Bay, the
average PCS level in muscle of English sole was about twice that in the
recreationally harvested pelagic species (based on data from Gahler et
39
-------
al. 1982). In Elliott Bay, the average PCB concentration in English sole
was about 0.4 times that in the harvested pelagic species (based on data
from Landolt et al. 1985). More data are needed to evaluate relative contam-
ination of potential indicator species in Puget Sound.
The use of a few selected species for risk assessment is appropriate
for initial screening of geographic areas before more detailed risk assessments
are conducted. If no potential health problems are identified in an initial
analysis, then further data collection may not be warranted (except for
long-term monitoring purposes). If, on the other hand, selected species
reveal substantial health risks, then further field surveys may be needed
to perform a detailed risk assessment based on consumption patterns and
contaminant concentrations for a wider variety of harvested species.
Further guidance on sampling strategies is beyond the scope of this
report. Phillips (1980) and Tetra Tech (1985b,c; 1986b] should be consulted
for detailed information on choice of species, analytical detection limits,
and other sampling considerations (e.g., samples of individual organisms
vs. composite samples).
Statistical Treatment of Data
Statistical analyses of data will depend on specific study objectives.
For each species, statistical summaries of tissue concentration data should
include sample size, estimates of arithmetic mean concentration, range,
and a measure of variance (standard error or 95 percent confidence limits).
For small sample sizes, a display of the mean and the distribution of individual
observations is sufficient. Geometric mean concentrations are appropriate
measures of central tendency when only estimates of tissue burden of contam-
inants or exposure dose are desired. However, arithmetic means are needed
to compare exposure estimates with RfDs and to calculate health risk.
Mean tissue concentrations and variances may be calculated for mixed-species
diets if data are available on the proportion of each species in the diet.
Data on concentrations of contaminants of concern in seafood tissue
samples will often contain observations below detection limits. Means
-------
and variances for tissue concentrations should be calculated twice: once
using detection limits for undetected observations and once using 0 for
undetected observations. According to the U.S. EPA Exposure Assessment
Group, calculations of plausible-upper-limit risk estimates based on detection
limits should generally be avoided. However, risk estimates based on detection
limits may occasionally be useful to indicate that particular chemicals,
species, or geographic locations are not problems, even assuming undetected
contaminants are present at concentrations just below their respective
detection limits.
EXPOSED POPULATION ANALYSIS
The second stage of the exposure assessment, analysis of exposed popula-
tions, includes the following steps:
• Identify potentially exposed human populations and map locations
of fisheries harvest areas
0 Characterize potentially exposed populations
Subpopulations by age, sex, and ethnic composition
Population abundance by subpopulation
• Analyze population activities
Harvest trip frequency
Seasonal and diel patterns of harvest trips
Time per harvest trip
General activity (e.g., clamming, crabbing, fishing)
e Analyze catch/consumption patterns by total exposed population
and subpopulation
Proportion of successful trips
Catch by numbers and weight according to species
Time since last meal of locally harvested seafood
41
-------
Number of consumers sharing catch
Parts of organisms eaten
Method of seafood preparation (e.g., raw, broiled,
baked)
o Estimate arithmetic average seafood consumption rate by
species and by total catch for the total exposed population
and for subpopulations. For seasonal fisheries, consumption
rates may be estimated on an annual and a seasonal basis.
Only selected steps may be performed in a given exposure assessment,
depending on d-ita availability, study objectives, and funding limitations.
Two approaches are outlined below. In the first approach, a comprehensive
analysis is performed based on extensive catch/consumption data for the
exposed population. In the second approach, no such data are available
and estimates of seafood consumption rates are based on standard values
for the U.S. population or other assumed values.
Comprehensive Catch/Consumption Analysis
Appropriate field survey forms, data analyses, and format for presentation
of results for a comprehensive catch/consumption analysis are provided
by Landolt et al. (1985) and McCallum (1985). Details of methods will
not be presented here, except to emphasize some important considerations
for calculating seafood consumption rates. Harvest weights should generally
be determined directly rather than from length measurements. However,
for shellfish and crabs, it may be necessary to establish tissue weights
from weight-length regression analysis. Supplementary information on seafood
consumption uata analysis can be found in SRI (1980). Lindsay (1986) reviewed
altar-native survey methods, including use of food diaries and dietary recall.
The average seafood consumption rate is the key exposure variable
for use in subsequent steps of risk assessment. Consumption rates should
be expressed in terms of g/day and meals/yr [assuming an average seafood
meal equals about 150 g (0.33 lb)]. Average consumption rate for each
harvest species is calculated according to the following steps:
42
-------
• For each successful angler trip, calculate the weight of
harves'. by species based on number and total weight harvested
per household
t Calculate mean harvest weight consumed per person per time
by
Dividing the total harvest weight for each species
by the number of consumers in household and by the
days elapsed since last meal from the same area
Multiplying the value obtained in the preceding amputation
cy a factor to account for the proportion of cleaned
weight to total weight [factor equals 0.49 for squid
and crabs, 0.3 for fish, and 1.0 for shucked clams
(Landolt et al. 1985)]
o Calculate mean consumption rate per person by geographic
harvest area, by subpopulation, and by total exposed population.
The model for calculating mean consumption rate (lijk) for species i,
subpopulation j, and area k is therefore:
r i_ y i - L_ V wijki PI
ijk • H... ^ Sjkl " N,,., Z- H.
ijk 7" "jkl
where:
s Mean consumption rate of species i for subpopulation j,
area k, and household 1
Nijk = NumDer of households (successful harvest trips) for spec es i,
subpopulation j, and area k
= Weight of species i harvested by household 1 of subpopulation
j in area k
43
-------
Pi = Proportion of cleaned edible weight of species i to total
harvested weight
Lf
jkl = Number of people in household 1 of subpopulation j in area k
TJ|^ = Time elapsed since last meal by household 1 of subpopulation
j in area k.
When consumption rates (^jkj) are log-normally oistributed, the data should
be log- trans formed before applying Equation 3 to calculate a mean consumption
rate.
Consumption rate data may be summarized further by calculating means
across species, subpopulations, and areas. However, it should be recognized
that means of 1^ -^ across species do not represent actual diet patterns.
To calculate mean consumption rates for mixed-species diets, all I- .,
should be sunrned across species within a household before determining mean
consumption rates across households:
Landolt et al . (1985) summarized the assumptions involved in calculating
mean consumption rates (Iijk1) by household as follows:
• Consumption
Pi values are assumed as noted above
Catch was distributed evenly among consumers in house-
hold
People in household actually ate the entire cleaned
catch
Personal harvest consumption was distributed evenly
over the time interval since the last successful trip
44
-------
e Fishing interval
Fishing frequency (days) is related to seasonal fisheries;
that is, interviewees did not report average time interval
for entire year but only for recent past. Therefore,
calculated consumption rates cannot be directly extrapolated
to a yearly basis. Fishing interval was set to 1 day
if umreported (Landolt et al. 1985).
Despite the limitation noted in'the last item above, calculated consumption
rates can be extrapolated to an annual average rate by multiplying the
^'ikl by a sPecies'5Pec^ic factor equal to the fraction of the year a
fishery is available. Determination of this species-specific factor is
somewhat subjective because of large seasonal fluctuations of the harvest
(e.g., Appendix E of Landolt et al. 1985). These factors should be determined
on a case-specific basis.
Assumed Seafood Consumption Rate
In many cases, comprehensive data on catch and consumption patterns
are not available. For some risk assessment problems (e.g., ranking of
potential problem chemicals in seafood), extensive catch/consumption data
are not needed. Moreover, catch/consumption patterns undoubtedly vary
over time. Extensive long-term monitoring of catch/consumption for all
Puget Sound areas of interest may not be warranted. Despite its obvious
limitations, extrapolating consumption data from one area (or time) to
another may be a suitable approach when:
e Site-specific data are unavailable
• Differences among areas (or times) are expected to be small
• Precise estimation of average seafood consumption is unnecessary
to meet the study objectives.
45
-------
In the past, many risk analysts have simply assumed standard values
for seafood consumption rates based on previous analyses of seafood consumption
patterns by the U.S. population (U.S. EPA 19S04; SRI .198Q). Average values
for "seafood" consumption for the U.S. population range from 6.5 to 20.4 g/day
(Nash 1971; National Marine Fisheries Service 1976, 1984; SRI 1980; U.S.
Department of Agriculture 1984). Most of these estimates include fish
and shellfish (molluscs, crustaceans) in marine and fresh waters, but saltwater
species form the bulk of consumed "seafood." The estimate of 6.5 g/day
was used by U.S. EPA (1980b) to develop water quality criteria based on
human health guidelines. Consumption rates for portions of the U.S. population
(e.g., by region, age, race, and sex) show that average seafood consumption
rates may vary from about 6 to 100 g/day (e.g., Suta 1978; SRI 1980; Puffer
et al . 1982). Finch (1973) determined that approximately 0.1 percent of
the U.S. population consunas 165 g/day of commercial seafood. For recreational
anglers of Puget Sound, the geometric mean consumption rates for individual
seafood species ranged from 11 to 40 g.day"1.person"1 during the respective
season for each species (Landolt et al. 1985). Limitations of seafood
consumption data are discussed by SRI (1980) and Landolt et al. (1985).
Based on existing information, three values of average seafood consumption
rate were chosen to represent a plausible range of standard values to be
assumed when site-specific data are unavailable:
• 6.5 g/day to represent a low estimate of average seafood
consumption for the entire U.S. population (U.S. EPA 1980b)
o 20 g/day to represent a high estimate of the same average
rate (U.S. Department of Agriculture 1984)
• 165 g/day to represent average seafood consumption for the
small portion of the U.S. population consuming the most
seafood (Finch 1973).
46
-------
INTEGRATED EXPOSURE ANALYSIS
In the Integrated exposure analysis, Information on estimated contaminant
concentrations and seafood consumption rate are combined to estimate chemical
intake by seafood species. The general model to calculate chemical intake is:
P _ Cikm ITjk -Xm ,,.
Eijkm ST (5)
where:
Eijkm = ExP°sure d°se of chemical m from species i for subpopulation
j in area k (mg-kg-1-dayl averaged over a 70-yr lifetime)
''ikm = Concentration of chemical m in edible portion of species i
in area k (mg/kg)
I-jjk = Consumption rate of species i by subpopulation j in area k
(kg/day averaged over 70-yr lifetime)
Xm = Relative absorption coefficient, or the ratio of human absorption
efficiency to test-animal absorption efficiency for chemical
m (dimensionless).
W = Average human weight (kg).
Values of subscripted terms above may be estimated means or uncertainty
interval bounds (e.g., 95 percent confidence intervals) depending on the
exposure scenario being modeled (e.g., worst case vs. average case vs. lover-
limit case). Note that E^^ is analogous to the dose "d" in Equations
1 and 2. The term E^^ is introduced here to emphasize that the exposure
dose for humans is a calculated value, whereas the dose "d" in Equations
1 and 2 is usally a known dose administered to bioassay animals.
In most cases, W is assumed to be 70 kg for the "reference man" (U.S. EPA
1980b). Assuming other average values to account for growth from a child's
body weight to adult weight over a lifetime would not change the results
of the risk assessment substantially.
47
-------
Absorption coefficients (Xm) are assumed equal to 1.0 unless data
for absorbed dose in animal bioassays used to determine toxicological indices
(carcinogenic potency or RfD) are available and the human absorption coefficient
differs from that of the animal used In the bioassay. Assuming that Xm
is equal to 1,0 is equivalent to assuming that the human absorption efficiency
is equal to that of the animal used in the tnodssay. In the absence of
'data to the contrary, this is appropriate. Tosticolog.ical indices are determined
from bioassays that usually measure administered (ingested) dose. Therefore,
the estimated chemical intake by humans, E is usually the ingested
dose, not the absorbed dose.
48
-------
RISK CHARACTERIZATION
In the risk characterization stage, results of the exposure and the
dose-response assessments are combined to estimate the probability and
extent of adverse effects associated with consumption of contaminated seafood.
An overview of the risk characterization process Is shown In Figure 2.
In human health risk assessment, carcinogens and noncarcinogens are treated
separately. Indices of risk for these different categories of toxicants
are based on different dose-response models (see above, "Dose-Response
Assessment").
CARCINOGENIC RISK
Numerical estimates of carcinogenic risk can be presented in one or
more of the following ways (U.S. EPA 1984a):
e Unit risk - the risk corresponding to a unit exposure of
mg contaminant per kg body weight per day
• Dose or concentration corresponding to <- oecified level
of risk - for example, a guideline for maximum allowable
contamination of a specified medium may be derived by assuming
a maximum allowable risk
a Individual and population risks - estimates of excess lifetime
cancer risk may be expressed for individuals (as a probability
estimate) or for the exposed population (as an estimate
of the number of cancers produced within a population of
specified size per generation).
Unit risks are useful for ranking chemical contaminants according to their
carcinogenic potencies. As shown later, they are also involved in calculations
of other numerical estimates of risk. Regardless of the option chosen
49
-------
mm
mm*
PHYSICAL -CHEMICAL
CHARACTERIZATION MO
BIOACCUNULATION
POTENTIAL
ENVIROMENTAL PARTf.
TIONING. DEGRADATION.
TRMSPORT MECHANISMS.
AHO POTENTIAL
EIPOSURE MEDIA
METABOLISM MO PHAR-
MCOtlNETIC PROPERTIES
TOIIC EFFECTS III
NIWMS AM LABORATORY
MIHAIS
QUANTITATIVE RELATION 1
SHIPS |
HEIGHT OF EVIDENCE
t g . STRUCTURE-ACTIVITY
RELATIONSHIPS. ««..
eiOCONCCNTUTION FACTORS
f 9 . AIR. WATER. SEDIMENTS.
MO BIOTA
f « . IIPOPNILICITV. BIO-
ACTIVATION. TOIIFICATION/
OETOIIFICATION. TARGET
ORGANS. ClIHINATIOII
• g . ACUTE AND CHRONIC
TOIICITT. CARC1NOCCNIC POTENCY,
EPIDEMIOL06IC EVIDENCE
Ht.g..
CARCII
. . DOSE-RESPONSE RELATION!.
CARCINOGENIC POTENCT
I g . ADEQUACY AND QUAIITT OF
DATA. LIULIHOOO OF SPECIFIC
TOIIC EFFECTS
/ IS SU8STMCE \ kX-\ .f—I
( POTENT IAUT >——W « }—-M STOP
\ HAZARDOUS • / ^\-S ^L-J
/ ME DATA
^SUFFICIENT FOR'
A OUANTITATIVC
V »IS«
\ASSESSMtNT '
SELECT ROUTE
OF EIPOSURE
_ CONCENTRATION OF
" SPECIFIC CONTAMINANT
M
M
ROdTE • ORAL
CONCENTRATIOII IN SEAFOOD
CONTACT UTE
DURATION OF EIPOSURE
t.g.. IN G/DAT OF SEAFOOD
CONSLKO
•.« . TEARS OF EIPOSURE.
FRACTION OF LIFETIME EIPOSED
COEFFICIENT
OF ABSORPTION
• «•• ASSIMILATED
JOnTiCTEa
8001 MICMT OF
EIPOSED INDIVIDUAL
OAILT EIPOSURE »ER >g
BOOT WEIGHT
QUALITATIVE IIS* DETERMINATION
BASED ON TOIICOLOGICAl PROPERTIES
«0 LIMITED E»PQSuRE DATA
QUANTITATIVE
-------
for expressing risk, final numerical estimates should be presented as one
significant digit only (U.S. EPA 1984a).
The general model for estimating a plausible-upper limit to excess
lifetime risk of cancer is:
R - * E t&\
ijkm ~ ^Im " ijkm l '
where:
R... = Plausible-upper-limit risk of cancer associated with chemical m
in species i for subpopulation j in area k (dimensionless)
qjm = Carcinogenic potency factor for chemical m [(mg-kg-1-dayl)" ]
estimated as the upper 95 percent confidence limit of the
slope of a linear dose-response curve
Eijkm = Exposure dose of chemical m from species i for subpopulation
j in area k (mg-kg-l-dayl)
Estimates of qlm are given in Table 4 above. All E.-km are calculated
as discussed above (see "Integrated Exposure Analysis" discussion in "Exposure
Assessment").
NONCARCINOGEN1C EFFECTS
An index of noncarcinogenic risk may be approximated as the ratio
of the estimated chemical intake to the Reference Dose (RfD) as follows:
E,.,
where:
Hiikm = Hazard Index of a health effect from intake of chemical m
associated with species i for ethnic group j in area k (dimension-
less)
RfOm = Reference Dose for chemical m (mg-kg'1-day1)
51
-------
and ^ijkm 1S defined as above. RfDm values are given in Table 5 above.
The hazard index is compared to a value of 1.0 to evaluate the chemical
hazard (Stara et al. 1983; U.S. EPA 1985b). Values of Hijkm above 1.0
indicate that the estimated exposure E^- is potentially of concern.
Above 1.0, increasing values of H^^ indicate increasing hazard.
CHEMICAL MIXTURES
U.S. EPA (1985c) discussed various models for assessment of the upper
limit to risk from chemical mixtures. Because of present data limitations
and the complexity of possible contaminant interactions, it is virtually
impossible at present to predict synergistic or antagonistic effects of
most chemical mixtures. The approach used most frequently for multiple-
chemical assessment is the additive-risk (or response-additive) model.
Thus, total upper-limit risk for a chemical mixture is usually estimated
as the sum of upper-limit risks for carcinogens or of hazard indices for
noncarcinogens. A sum of noncarcinogenic hazard indices should be calculated
only for a group of chemicals acting on the same target organ (Stara et
al . 1983). The numerical estimates obtained using the response-additive
model are useful in terms of relative comparisons (e.g., among fishing
areas or among seafood species). However, risk estimates for chemical
mixtures should be regarded only as rough measures of absolute risk (U.S. EPA
1985c). Because technological limitations preclude analyzing seafood samples
for all potentially toxic chemicals, risk estimates for chemical mixtures
should not be interpreted as estimates of total risk associated with seafood
ingest ion.
52
-------
PRESENTATION AND INTERPRETATION OF RESULTS
Examples of formats for presenting the results of risk assessments
are provided below. These formats are adaptable to any level of summary
analysis (e.g., subpopulation vs. total exposed population, individual
species vs. average across species). Interpretation of the results is
discussed relative to risk comparisons (e.g., fish consumption vs. other
activities such as cigarette smoking, estimated risk vs. "acceptable" risk
defined by policy), assumptions, and uncertainty analysis. The term "acceptable
risk" is used to denote the maximum risk considered tolerable by an individual
or a regulatory agency. Although acceptable risk levels must be defined
on a case-specific basis, past regulatory policies in the U.S. have generally
set allowable levels for environmental risks on the order of 10"5 to 10"
(see below, "Risk Comparisons"). Supplementary information, such as comparisons
of contaminant concentrations with U.S. FDA action levels, is addressed
in the final section below.
PRESENTATION FORMAT
The results of risk assessment may be presented in both tabular and
graphic format. In the supporting text of a risk assessment, all final
estimates of risk should be rounded to one significant digit (or an order
of magnitude if appropriate). The U.S. EPA classification of the qualitative
weight of evidence for carcinogenicity (presently under development) should
be shown in brackets adjacent to final risk estimates for carcinogens (U.S. EPA
1984a). Also, all risk estimates should be interpreted as plausible-upper-
limit values for the stated assumptions and exposure conditions.
Summary Tables
An example format of an integrated exposure analysis for a hypothetical
human population is shown in Table 6. The table format allows storage
of exposure information in a computer spreadsheet. Columns of notes containing
-------
TABLE 6. EXAMPLE L.dULAR FORMAT FOR DISPLAY OF QUANTITATIVE RISK ASSESSMENT FOR CONSUMPTION OF SEAFOOD
Substance
PCBs
PCBs
Hg
HO
p
Concen-
tration
In Hedlun
(ng/kg)0
0.007
0.004
0.010
0.007
0.004
0.010
0.1S7
o.ooa
0.478
0.157
o.ooa
0.478
Contact
Rate
(g/day)«»
6.5
6.5
6.S
20.0
20.0
20.0
6.5
6.5
6.5
20.0
20.0
20.0
Total
Dally
Contact
(og/day)
4.6E-OS
2.6E-OS
6.SE-OS
I.4E-04
8.0E-05
Z.OE-04
l.OE-03
5.2E-05
3.IE-03
3.1E-03
1.6E-04
9.6E-03
ire ueieruu
E xposure
Duration
(years)
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
Absorption •
Coefficient
(0-1. OJ«
.0
.0
.0
.0
.0
.0
.0
.0
.0
1.0
1.0
1.0
Body
Height
(kg)
70
70
70
70
70
70
70
70
70
70
TO
70
'1 I1"
Carcinogens
Exposure Potency
Value Factgr
(og/kg/d) l/(mg/kg/d) Risk
6.5E-07
3.7E-07
9.3E-07
2.0E-06
1.1E-06
2.9E-06
1.5E-05
7.4E-07
4.4E-05
4.5E-05
2.3E-06
1 .4E-04
.34 3E-06
.34 ?E-06
.34 4E-06
.34 9E-06
.34 5E-06
.34 IE-OS
N/A N/A
N/A N/A
H/A N/A
N/A N/A
N/A N/A
N/A N/A
Noncarclnogens
RfD
(ntg/kg/d)
H/A*
N/A
N/A
H/A
H/A
N/A
2.9E-04
2.9E-04
2.9E-04
2.9E-04
2.9E-04
2.9E-04
Hazard Index
N/A
N/A
N/A
N/A
N/A
N/A
SE-02
3E-03
ZE-OI
2E-01
8E-OJ
SE-01
0 Concentration of contaminant In seafood species of concern (mg/kg ° ppn by mass, wet weight).
b Amount of seafood Ingested per day. prior to accounting for absorption efficiency, etc.
e Ratio of g of contaminant absorbed per g of contaminant Ingested.
<> H/A • not applicable.
-------
references to sources of Information can easily be added to the spreadsheet
to further document the exposure analysis.
It should be emphasized that some of these variables are capable of
being measured with great precision (e.g., contaminant concentrations in
fish tissue), whereas others may only be estimated on an order-of-magnitude
basis. The precision and accuracy of the final risk estimates are directly
related to the precision and accuracy of the variables incorporated into
the model equations.
Uncertainty is easily characterized with a spreadsheet format by calcu-
lating exposure estimates for low, mid, and high values of key variables
within their respective plausible ranges. Specification of probability
distributions for key variables is an alternative method of uncertainty
analysis requiring graphical models (see below, "Uncertainty Analysis").
In the example shown in Table 6, the average, minimum, and maximum concentra-
tions of two contaminants [PCBs and mercury (Hg)] are used to estimate
potentia] health risk, thereby accounting for uncertainty in chemical analyses.
Also, risks are estimated for two consumption rate estimates (6.5 g/day
and 20 g/day).
Summary Graphics
Presentation of risk assessment results in graphic form may include:
• Plots of risk vs. consumption rate
o Plots of risk vs. contaminant concentration in seafood
• Summary maps of risk estimates for different locations in
Puget Sound or for different locations within an embayment
• Histograms of risk by species, human subpopulation, or geographic
location.
55
-------
Because estimated risk for a given area and seafood species varies with
consumption rate and because consumption rates vary greatly among Individual
humans, the first approach above Is recommended as the primary means of
presenting risk assessment results. Actual consumption patterns of the
exposed population may or may not be estimated (see above, "Exposure Assess-
ment"). If they are, estimates of average consumption rate can be identified
in a footnote (e.g., Figure 3). Uncertainty can be illustrated by plotting
lines corresponding to the minimum and maximum (or 95 percent confidence
limit) values of contaminant concentrations in seafood, as well as the
mean concentration (e.g., Figure 3). As an interpretive aid, risk assessment
results for a reference area can be presented along with those for' the
study area. Coupled with information on comparative and "acceptable" risk
levels (see below, "Risk Comparisons"), Figure 3 is an appropriate format
for graphic display of results to lay public.
This approach may also be used in risk management. For example, if
a reference (i.e., allowable maximum) risk level, is defined by policy,
then an advisory limit on the consumption rate for each species of seafood
may be determined, as shown by the dotted line in Figure 4. Note that
an individual seafood consumer may eat a mixed-species diet and the additive
risks of the total diet may exceed the reference risk level established
separately for each species. This should be taken into account when estab-
lishing the reference risk level to provide an adequate "margin of safety"
for individuals who eat a mixed diet.
Other approaches noted above can be used to supplement the risk vs.
consumption plots. Summary maps and histograms may be especially useful
for presentation of detailed results of spatial analyses by subpopulation
or by species. Plots of risk vs. contaminant concentration in seafood
aid in rapid interpretation of tissue contamination data for selected seafood
consumption rates (e.g., Figure 5).
RISK COMPARISONS
Interpretation of carcinogenic risk assessment results may be based
on comparison of estimated health risks for the study area with:
56
-------
,0-3 _
cr 10-*-
cc
UJ
, o
10-5-
10-6-
10-7
STUDY AREA
N- 25 BUTTER CLAMS
///
''/'
/ / ' \ REFERENCE AREA
' / N- 25 BUTTER CLAMS
x
/
X
I
10
(25)
I
100 c/day
(250) (meals/yr)
CONSUMPTION RATE
NOTE: 1 g/day . 2.5 meals/yr assuming ISOg (0.33 Ib) par meal.
All cancer risks are plausible upper-hmit-estimates of excess risks due to PCBs
based on linear low-doso extrapolation model and assumptions presented m text.
Solid line represents average tissue concentration of contaminant.
Dashed lines represent uncertainty range (e.g., 95 percent confidence limits) for
averags tissue conc3nlration only, not the total uncertainty range.
Figure 3. Example graphic format for display of quantitative
risk assessment results for hypothetical study area
and reference area.
-------
SEAFOOD SPECIES A AND
CHEMICAL CONTAMINANT i
IS)
"^
ce
ac
"1FERENCE
KISK LEVEL
MAXIMUM
CONSUMPTION
ADVISORY
CONSUMPTION RATE
cigure 4. Plausible-upper-limit estimate of lifetime cancer
risk associated with mean contaminant concentration
in seafood species A versus ••ate of seafood consumption.
58
-------
10'c-
cc
HI
o
R,* = REFERENCE
1 RISK
C* = TISSUE
1 CONTAMINATION
to'
GUIDELINE FOR 6.5 g/day
L_J_LLLLll I I _1 I I I III I I I I ..,,1
0001
001 01 10 10
CHEMICAL CONCENTRATION IN SEAFOOD (ppm)
10
Figure 5. Plauiible-upper-limit estimate of lifetime cancer
cancer risk versus concentration of a chemical con-
taminant in seafood (ppm wet wt.) at selected inges-
tion rates.
59
-------
• Health risks for consumption of seafood from a reference
area
• Health risks for consumption of alternative foods (e.g.,
charcoal-broiled steak, marketplace fish) or other activities
(e.g., cigarette smoking)
• Acceptable risk levels defined by agency policy.
An example of comparison with reference-area risk estimates is shown in
Figure 3 above. Comparative risks for alternative foods or activities
can be summarized in a table (e.g., Table 7) or histogram. These kinds
of comparisons are limited by several factors. First, the risks being
compared may be inherently different (e.g., involuntary vs. voluntary risks,
observed vs. predicted risks). Risks associated with some activities (e.g.,
driving an automobile) may be accounted for completely, whereas chemical
risks are only partially addressed (e.g., focus on single chemicals, ignorance
of synergistic interactions, and inability to measure all carcinogenic
chemicals in foods). Finally, the degree of participation in different
activities and the perception of risk varies among people.
An "acceptable" risk level has not been strictly determined by U.S. EPA.
In general, U.S. EPA decisions have allowed individual lifetime risk estimates
of 10"4 to 10 '8 (Thomas 1984). U.S. EPA (1980b) used lifetime risk levels
of 10"' to 10 as reference values to develop water quality criteria,
but these values were used for reference purposes only and did not represent
an agency determination of acceptable risk. Risks on the order of 7xlO~5
per lifetime (10"6/yr) are commonly accepted by most people, while higher
risks are clearly of concern to environmental regulators (Pochin 1975;
Crouch et al. 1983). In setting standards for benzene exposure, Justice
Stewart of the U.S. Supreme Court argued that lifetime risks of 10"3 were
clearly "unacceptable," whereas those of 10"9 were clearly "acceptable"
(Connor 1983).
60
-------
TABLE 7. EXAMPLES OF CANCER RISKS FROM COMMON CARCINOGENS
Diet soda (saccharin) - 12.5 oz/day
Average saccharin consumption
Peanut butter (aflatoxins) - 4 tbsp/day
Milk (aflatoxins) - 1 pt/day
Miami/New Orleans drinking water - 2 './day
Charcoal broiled steak (PAH) - 0.5 Ib/wk
Average smoker (PAH)b
Person sharing room with smoker
Average
Lifetime
Risk3
7x10-4
ixlO-4
6x10-4
1x10-4
7x10-5
2x10-5
8x10-2
7x10-4
Average
Annual
Risk Uncertainty
1x10-5
2x10-6
8x10-6
2x10-6
1x10-5
3x10-7
Factor
of
10
1.2x10-3 Factor of 3
1x10-5 Factor
of 10
a Average lifetime risks were calculated from average annual risks during preparation
of this report assuming a lifetime of 70 yr.
b Risk estimate based on human data.
Reference: Crouch and Wilson (1984).
61
-------
Risk comparisons should be based on consistent exposure analysis and
risk extrapolation models. Analogous exposure scenarios should be used
for each risk estimate being compared (I.e., either worst case, plausible-
upper limit, average, or lower limit). A single model should be applied
consistently to calculate exposure and risk. A linear extrapolation model,
such as Equations 2 and 6 above, is justified in general if the excess
risk attributed to the contaminant of concern is regarded as a marginal
risk, added to a background of relatively high cancer incidence from all
other causes not being modeled (Crump et al. 1976; Omenn 1985).
SUMMARY OF ASSUMPTIONS
Assumptions underlying the risk assessment model and estimates of
model coefficients should be summarized in a concise format (see Table 8
for summary of some assumptions and numerical estimates used in the approach
presented in this manual). Specific assumptions adopted on a case-by-case
basis should be summarized in a similar fashion.
Additional assumptions involved in applying the risk assessment approach
explained above include the following:
t Adverse effects in experimental animals are indicative of
adverse effects in humans (e.g., lifetime incidence of cancer
in humans is the same as that in animals receiving an equivalent
dose in units of mg per surface area)
• Dose-response models can be extrapolated beyond the range
of experimental observations to yield plausible-upper-bound
estimates of risk at low doses
• A threshold dose does not exist for carcinogenesis
• A threshold dose (e.g., No-Observed-Adverse-Effect-Level)
exists for noncarcinogenic effects
62
-------
TABLE 8. SUMMARY OF ASSUMPTIONS AND NUMERICAL
ESTIMATES USED IN RISK ASSESSMENT APPROACH
Parameter
Assumptions/Estimates
Reference
Exposure Assessment:
Contaminant concentrations
in tissues of indicator
species
Average consumption rate
Gastrointestinal absorption
coefficient
Exposure duration
Human body weight
Risk Characterization:
Carcinogenic risk model
Carcinogenic potency
Acceptable Daily Intake
(ADI) = Reference Risk Dose
(RfD)
No effect of cooking
6.5 g/day
20 g/day
165 g/day
1.0
Assumes efficiency of absorption
of contaminants is same for humans
and bioassay animals.
70 yr
70 kg (= avg. adult male)
Linearized Multistage Model
(linear, no-threshold model).
At risks less than 10-2:
Risk = Exposure x Potency.
Potency factors are based on low-
dose extrapolation from animal
bioassay data.
Upper bound of 95 percent confi-
dence interval on potency is used.
ADIs (or RfOs) for noncarcinogens
are current U.S. EPA values.
Worst case for parent
compounds. Net effect
on risk is uncertain.
Low, moderate, and
high values specified
by regulatory policy
(see text)
U.S. EPA 1980b
U.S. EPA-CAG3
U.S. EPA 1980b
U.S. EPA-CAG
U.S. EPA 1980b
U.S. EPA 1980b, 1985a
U.S. EPA l'J80b; U.S.
EPA Envire"menial
Criteria and Assessment
Office
a U.S. Environmental Protection Agency Carcinogen Assessment Group.
-------
• The most sensitive animal species Is appropriate to represent
the response of humans
• Cumulative incidence of cancer increases in proportion to
the third power of age (this assumption is used to estimate
lifetime incidence when data are available only from less-
than-lifetime experiments)
• Average doses are an appropriate measure of exposure dose,
even if dose rates vary over time
• In the absence of pharmacokinetic data, the effective (or
target organ) dose is assumed to be proportional to the
administered dose
• Risks from multiple exposures in time are additive
• For each chemical, the absorption efficiency of humans is
equal to that of the experimental animal
• If available, human data are preferaole to animal data for
risk estimation
t For chemical mixtures, risks for individual chemicals are
additive. However, the total sum of individual chemical
risks is not necessarily the total risk associated with
seafood ingestion because some important toxic compounds
may not have been identified and quantified.
UNCERTAINTY ANALYSIS
Uncertainty analysis is an integral part of risk assessment. A general
discussion of uncertainties present in the risk assessment approach described
herein is presented in the next section. The U.S. EPA guidelines on exposure
assessment describe general approaches for characterizing uncertainty (U.S. EPA
1984b). Methods for uncertainty analysis are discussed by Cox and Baybutt
64
-------
(1981), Morgan (1984), Whitmore (1985), and Tetra Tech (1986a). A discussion
of procedures is beyond the scope of the present effort. Nevertheless,
general approaches to uncertainty analysis of model coefficients are presented
after the discussion of sources of uncertainty.
Sources of Uncertainty
Uncertainties in the risk assessment approach presented in this manual
arise from the following factors:
1. Uncertainties in estimating carcinogenic potency factors
or RfDs, resulting from
o Uncertainties in extrapolating toxicologic data obtained
from laboratory animals to humans
o Uncertainties in high- to low-dose extrapolation of
bioassay test results, which arise from practical
limitations of laboratory experiments and variations
in extrapolation models
2. Uncertainties in estimates of site-specific consumption
rates and contaminant concentrations
3. Uncertainties in the selection of 6.5 g/day, 20 g/day, and
165 g/day as assumed consumption rates when site-specific
data are not available
4. Uncertainties in the efficiency of assimilation (or absorption)
of contaminants by the human gastrointestinal system (assumed
to be the same as assimilation efficiency of the experimental
animal in the bioassay used to determine a carcinogenic
potency factor or RfD)
65
-------
5. Uncertainties associated with variation of exposure factors
among individuals, such as
• Variation in seafood species composition of the diet
among individuals
• Variation in seafood preparation methods and associated
changes in chemical composition and concentrations
due to cooking.
Variance in estimates of carcinogenic potency or RfDs (#1 above) account
for one major uncertainty component in this study. Chemical potencies
are estimated only on an order-of-magnitude basis, whereas analytical chemistry
of tissues is relatively precise (on the order of+20 percent). The choice
of a low-dose extrapolation model greatly influences estimates of the carcin-
ogenic potency factor and calculated risks. This uncertainty contributed
_9
by the model is substantial when predicting risks below 10 . For example,
the plausible-upper limit to lifetime cancer risk associated with 50 ug/L
tetrachloroethene in drinking water ranges from about 10"6 for the probit
model to 10'2 for the Weibull model (Cothern et al. 1986). Model uncertainty
is important when considering absolute risk estimates (e.g., Cothern et
al. 1986), but less important for relative risk comparisons.
Uncertainty analysis conducted by previous researchers illustrates
the variability of risk estimates and potency factors for a given extrapolation
model. For example, the coefficient of variation for the mean value of
potency within species generally ranged from 2 to 105 percent for each
drinking water contaminant studied by Crouch et al. (1983). This uncertainty
arose mainly from error associated with experimental bioassay data. Among
species, the potency of a given chemical may vary only slightly or up to
approximately 1,000-fold, depending on the chemical in question (Clayson
et al. 1983). Thus, the uncertainty associated with extrapolating potency
factors from laboratory animals to humans may be much greater than the
uncertainty associated with animal bioassay techniques. By comparison,
the range of potencies among carcinogens covers 7-9 orders of magnitude
66
-------
(Clayson et al. 1983; U.S. EPA 1984a, 1985a). Relative risk comparisons
•*ong chemicals can be made more confidently when the range of potency
ractors is broad.
In conclusion, uncertainty ranges (e.g., 95 percent confidence intervals)
around estimates of mean risk may typically span 3-5 orders of magnitude.
The approach taken by U.S. EPA (1980b, 1984a, 1985a) and followed herein
is to estimate a plausible-upper limit to risk. In this way, it is unlikely
that risk will be underestimated substantially. Moreover, the plausible-
upper-limit estimate serves as a consistent basis for relative risk comparisons.
Approaches to Uncertainty Analysis
Approaches to treatment of uncertainty in model coefficients used
in risk analysis include the following (Morgan 1984):
• Perform analysis using single-value-best-estimates for model
coefficients, without uncertainty analysis
• Perform single-value-best-estinate analysis, with sensitivity
calculations and appropriate discussion of uncertainty
• Estimate some measure of uncertainty (e.g., standard deviation)
for each model coefficient and use error propagation methods
to estimate uncertainty of final exposure or risk value
• Characterize subjectively the probability distribution of
each model coefficient and propagate error through stochastic
simulation
• Characterize important model coefficients using a parametric
model and perform risk analysis using various plausible
values of each of the coefficients
67
-------
0 Determine upper and lower bounds on model coefficients to
yield order-of-magnitude estimates and range of possible
answers.
Morgan (1984) refers to the first two approaches as "single-value-best-
estimate analysis," to the second two as "probabilistic analysis," and
to the final two as "parametric/ bound ing analysis." The analytical strategies
listed above are in roughly descending order, based on the amount of uncertainty
in the model coefficients. Single-value-best-estimate analysis is appropriate
when model coefficients are precisely known. Bounding analysis is most
appropriate when model coefficients are not well-known. The techniques
listed above do not address model uncertainty, which must be handled by
exploratory examination of outcomes based on alternative model equations.
The choice of a method for uncertainty analysis will depend on the
amount and quality of exposure data and on the study objectives. In many
cases, data will be sufficient only to use parametric/bounding analysis,
as described above (also, see Morgan 1984). Also, quantitative uncertainty
analysis is applied mainly to exposure variables, such as contaminant concen-
tration in seafood and seafood consumption rate. Following U.S. EPA (1980b,
1984a, 1985a), the upper bound of the 95 percent confidence interval for
the carcinogenic potency factor is always used in risk calculations. Substi-
tution of the mean estimate or the lower bound of the 95 percent confidence
interval for the potency factor in the risk calculations is generally not
done because of the instability of these estimates (U.S. EPA 1980b).
SUPPLEMENTARY INFORMATION
Additional information to support risk assessment of contaminated
seafood may include:
• Comparisons of tissue concentrations of contaminants with
U.S. FDA action (or tolerance) levels
• Statistical comparisons of mean contaminant concentrations
in seafood among species and among areas of Puget Sound
68
-------
• Statistical comparisons of mean contaminant concentrations
in seafood with those in other foods.
Examples of the first two approaches can be found in Tetra Tech (1985d,
1986c). Examples of the latter approach are provided by Eagle Harbor Ad
Hoc Committee (1985).
U.S. FDA limits on contaminants in seafood products are shown in Appendix
C. Limitations to use of these values for assessing health risk were discussed
earlier (see above, "Overview of Risk Assessment"). For comparison, legal
limits on seafood contaminants established by other countries are also
provided in Appendix C.
69
-------
REFERENCES
Callahan, M.A., M.W. Slimak, N.W. Gable, I.P. May, C.F. Fowler, J.R. Freed,
P. Jennings, R.L. Ourfee, F.C. Whitmore, B. Amestri, W.R. Mabey, B.R. Holt,
and C. Gould. 1979. Water-related environmental fate of 129 priority
pollutants. Volumes I and II. Prepared for U.S. Environmental Protection
Agency. Versar, Inc., Springfield, VA. Available from NTIS. PB80-204373.
Clayson, D.B., D. Krewski, and I.C. Munro. 1983. The power and interpretation
of the carcinogenicity assay. Regul. Toxicol. and Pharmacol. 3:329-348.
Connor, M.S. 1983. Estimating the public health risk of organic carcinogens
in U.S. fish. Paper No. 15. In: International Ocean Disposal Symposia
Series Special Symposium: Ocean Waste Management: Policy and Strategies.
University of Rhode Island, Kingston, RI. 26 pp.
Connor, M.S. 1984. Comparison of the carcinogenic risks from fish vs. ground-
water contamination by organic compounds. Environ. Sci. Technol. 18:628-631.
Cothern, C.R., W.A. Coniglio, and W.L. Marcus. 19&6. Estimating risk
to human health. Environ. Sci. Technol. 20:111-116.
Cox, O.C., and P. Baybutt. 1981. Methods of uncertainty analysis: a
comparative survey. Risk Analysis 1:251-258.
Crouch, E.A.C., and R. Wilson. 1984. Inter-ris'< comparisons, pp. 97-112.
In: Assessment and Management of Chemical Risks. J.V. Rodricks and R.G.
Tardiff (eds). ASC Symposium Ser. 239, American Chemical Society, Washing-
ton, DC.
Crouch, E.A.C., R. Wilson, and L. Zeise. 1983. The risks of drinking
water. Water Resour. Res. 19:1359-1375.
Crump, K.S., D.G. Hoel, C.H. Langley, and R. Peto. 1976. Fundamental
carcinogenic processes and their implications for low dose risk assessment.
Cancer Res. 36:2973-2979.
Dourson, M.L., and J.F. Stara. 1983. Regulatory history and experimental sup-
port of uncertainty (safety) factors. Regul. Toxicol. and Pharmacol. 3:224-238.
Eagle Harbor Ad Hoc Committee. 1985. Report of Eagle Harbor Ad Hoc Ccmnittee
to the Department of Social and Health Services. Washington Department
of Social and Health Services, Olympia, WA. 4 pp.
Finch, R. 1973. Effects of regulatory Guidelines on the intake of mercury
from fish - the MECCA project. Fish. Bull. 71:615-626.
Food Safety Council. 1980. Proposed system for food safety assessment.
Food Safety Council, Washington, DC. 160 pp.
70
-------
Food Safety Council. 1982. A proposed food safety evaluation process.
Food Safety Council, Washington, DC. 142 pp.
Gahler, A.R., R.L. Arp, and J.M. Cummins. 1982. Chemical contaminants
in edible non-salmon id fish and crabs from Commencement Bay, Washington.
Environmental Services Division, U.S. Environmental Protection Agency,
Seattle, WA. 117 pp.
Hogan, M.D., and O.G. Hoel. 1982. Extrapolation to man. pp. 711-731.
In: Principles and Methods of Toxicology. A.W. Hayes (ed). Raven Press,
New York, NY.
International Agency for Research on Cancer. 1978. Working group on the
evaluation of the carcinogenic risk of chemicals to humans. In: International
Agency for Research on Cancer Monographs Vol. 18, Polychlorinated Biphenyls.
Lyon, France.
Kneip, T.J. 1983. Public health risks of toxic substances, pp. 577-610.
In: Ocean Disposal of Municipal Wastewater: Impacts on the Coastal Environ-
ment. Vol. 2. E.P. Myers and E.T. Harding (eds). MITSG 83-33. Massa-
chusetts Institute of Technology, Cambridge, MA.
Konasewich, D.E., P.M. Chapman, and E. Gerencher, G. Vigers, and N. Treloar.
1982. Effects, pathways, processes, and transformation of Puget Sound
contaminants of concern. NOAA Technical' Memorandum OMPA-20. National
Oceanic and Atmospheric Administration, Boulder, CO. 357 pp.
Landolt, M.L., F.R. Hafer, A. Nevissi, G. van Belle, K. Van Ness, and C. Rock-
well. 1985. Potential toxicant exposure among consumers of recreational ly
caught fish from urban embay merits of Puget Sound. NOAA Technical Memorandum
NOS-OMA-23. National Oceanographic and Atmospheric Administration, Rockville,
MD. 104 pp.
Life Systems, Inc. 1985. The endangerment assessment handbook. Draft Report.
Prepared for Planning Research Corporation, Chicago, IL for Office of Waste
Programs Enforcement, U.S. Environmental Protection Agency, Washington, DC.
Lindsay, D.G. 1986. Estimation of the dietary intake of chemicals in
food. Food Additivies and Contaminants 3:71-88.
Lyman, W.J., W.F. Reehl, and D.H. Rosenblatt. 1982. Handbook of chemical
property estimation methods. McGraw-Hill Book Co., New York, NY.
Malins, D.C., B.B. McCain, D.W. Brown, A.K. Sparks, and H.O. Hodgins.
1980. Chemical contaminants and biological abnormalities in central and
southern Puget Sound. NOAA Technical Memorandum OMPA-2. National Oceanic
and Atmospheric Administration, Boulder, CO. 295 pp.
Malins, O.C., B.B. McCain, D.W. Brown, A.K. Sparks, H.O. Hodgins, and S.-L.
Chan. 1982. Chemical contaminants and abnormalities in fish and invertebrates
from Puget Sound. NOAA Technical Memorandum OMPA-19. National Oceanic
and Atmospheric Administration, Boulder, CO. 168 pp.
71
-------
Mantel, N., and M.A. Schneiderman. 1975. Estimating "safe levels": a
hazardous undertaking. Cancer Research 35:1379.
McCallum, M. 1985. Recreational and subsistence catch and consumption
of seafood from three urban industrial bays of Puget Sound: Port Gardner,
Elliott Bay, and Sinclair Inlet. Washington Department of Social and Health
Services, Olympia, UA. 59 pp.
Morgan, M.G. 1984. Uncertainty and quantitative assessment in risk manage-
ment, pp. 113-130. In: Assessment and Management of Chemical Risks.
J.V. Rodricks and R.G. Tardiff (eds). ACS Symposium Ser. 239. American
Chemical Society, Washington, DC.
Nash, D.A. 1971. A survey of fish purchases of socio-economic characteris-
tics. Data Report No. 62. National Marine Fisheries Service, Seattle,
WA.
National Marine Fisheries Service. 1976. Seafood consumption study, 1973-
1974. National Marine Fisheries Service, Washington, DC. p. 146.
National Marine Fisheries Service. 1984. Fisheries of the United States,
1983. Current fishery statistics No. 8320. National Marine Fisheries
Service, Washington, DC. 121 pp.
National Research Council. 1983. Risk assessment in the federal government:
managing the process. The Committee on the Institution of Means for the
Assessment of Crisis to Public Health. Washington, OC.
National Toxicology Program. 1982. Third annual report on carcinogens.
U.S. Department of Health and Human Services, Public Health Service, Washington,
DC. 327 pp. + 5 appendices.
Nauen, C.E. 1983. Compilation of legal limits for hazardous substances
in fish and fishery products. FAO Fisheries Circular No. 764. Food and
Agriculture Organization of the United Nations, Rome, Italy. 102 pp.
Nicola, R.M., R. Branchflower, and D. Pierce. 1983. Assessment of health
risks associated with consumption of bottom fish caught in an industrial
bay. Tacoma-Pierce County Health Department, Tacoma, WA. 18 pp.
Omenn, G. 1985. A framework for risk assessment. In: Risk Assessment
in Occupational and Environmental Health. (Short course text). Northwest
Center for Occupational Health and Safety, University of Washington, Seattle,
WA.
Peddicord, R.K. 1984. What is the meaning of bioaccumulation as a measure
of marine pollution effects? pp. 249-260. In: Concepts in Marine Pollution
Measurements. H.H. White (ed). University of Maryland Sea Grant Program,
College Park, MD.
Phillips, D.J.H. 1930. Quantitative aquatic biological indicators. Applied
Science Publishers, Ltd., London, U.K.
Pochiti, E.E. 1975. Acceptance of risk. Br. Med. Bull. 31:184-190.
72
-------
Pollock, G. 13 June 1986. Personal Communication (Phone to Dr. R.A. Pastorok).
California Department of Health Services, Sacramento, CA.
Puffer, H.W., M.J. Ouda, and S.P. Azen. 1982. Potential health hazards
from consumption of fish caught in polluted coastal waters of Los Angeles
County. N. Am. J. Fish. Manage. 2:74-79.
SRI. 1980. Seafood consumption data analysis. Final Report. Prepared
for Office of Water Regulations and Standards, U.S. Environmental Protection
Agency, Washington, DC. 44 pp.
Stara, J.F., R.C. Hertzberg, R.J.F. Bruins, M.L. Dourson, P.R. Durkin,
L.S. Erdreich, and W.E. Pepelko. 1983. Approaches to risk assessment
of chemical mixtures. Report presented at the Second International Conference
on Safety Evaluation and Regulation, Cambridge, MA. 23 pp.
Suta, B.E. 1978. Human exposures to mirex ard kepone. EPA-600/1-78-045.
U.S. Environmental Protection Agency, Washington, DC.
Tatken, R.L., and R.J. Lewis (eds). 1983. Registry of toxic effects of
chemical substances 1981-1982 edition. 3 volumes. U.S. Department of
Health and Human Services, National Institute for Occupational Safetv and
Health, Cincinnati, OH.
Tetra Tech. 1985a. Bioaccumulation monitoring guidance: 1. estimating
the potential for bioaccumulation of priority pollutants and 301(h) pesticides
discharged into marine and estuarine waters. Final program document prepared
for the Marine Operations Division, Office of Marine and Estuarine Protection,
U.S. Environmental Protection Agency. EPA Contract No. 68-01-6938. Tetra
1-ech, Inc., Bellevue, WA. 61 pp.
Tetra Tech. 1985b. Bioaccumulat ion monitoring guidance: 2. selection
of target species and review of available bioaccumulation data. Final
program document prepared for the Marine Operations Division, Office of
Marine and Estuarine Protection, U.S. Environmental Protection Agency.
EPA Contract No. 68-01-6938. Tetra Tech, Inc., Bellevue, WA. 52 pp. +
5 appendices.
Tetra Tech. 1985c. Bioaccumulation monitoring guidance: 3. Recommended
analytical detection limits. Final program document prepared for the Marine
Operations Division, Office of Marine and Estuarine Protection, U.S. Environ-
mental Protection Agency. EPA Contract No. 68-01-6938. Tetra Tech, Inc.,
Sellevue, WA. 23 pp.
Tetra Tech. I935d. Commencement Bay nearshore/tideflats remedial investi-
gation. Vol. 1. Final Report. EPA-910/9-85-134b. Prepared for the
Washington Department of Ecology and U.S. Environmental Protection Agency.
Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1986a. A framework for comparative risk analysir, of dredged
material disposal options. Draft Report. Prepared for Resource Planning
Associates for U.S. Army Corps of Engineers, Seattle District. Tetra Tech,
Inc., Bellevue, WA. 94 pp. + 5 appendices.
73
-------
Tetra Tech. 1986b< Sioaccumulation monitoring guidance: 5. Strategies
for sample replication and compositing. Final program document prepared
for the Marine Operations Division, Office of Marinp aad Estuarine Protection,
U.S. Environmental Protection Agency. EPA Contract No. 68-01-5938. Tetra
Tech, Inc., Bellevue, WA. 46 pp.
Tetra Tech. 1986c. Elliott Bay toxics action program;. _^i~tial dataL-sunroanes
and problem identification. Fina1 Report. Prepared for'.the"U.S.-Environmental
Protection Agency, Region 10. Tetra Tech, Inc.,"Bellevue, "WA. 181 pp. +
8 appendices and maps.
Thomas, L.M. 1984. U.S. EPA memorandum on determining acceptable risk
levels for carcinogens in setting alternate concentration levels under
RCRA. Published November 23, 1984 by Bureau of National Affairs. Inc.,
Washington, DC.
U.S. Department of Agriculture. 1984. Agricultural statistics. U.S. Depart-
ment of Agriculture, Washington, DC. p. 506.
U.S. Environmental Protection Agency. 1980a. Ambient water quality criteria
for polychlorinated biphenyls. U.S. Environmental Protection Agency, Criteria
and Standards Division, Washington, DC. 200 pp.
U.S. Environmental Protection Agency. 1980b. Water quality criteria documents;
availability. U.S. EPA, Washington, DC. Federal Register, Vol. 45, No. 231,
Part V. pp. 79318-79379.
U.S. Environmental Protection Agency. 8 August 1984. Personal Communication
(letter to Dr. Robert Pastorok). Environmental Criteria and Assessment
Office, U.S. Environmental Protection Agency, Cincinnati, OH.
U.S. Environmental Protection Agency. 1984a. Proposed guidelines for car-
cinogen risk assessment; request for comments. U.S. EPA, Washington, DC.
Federal Register, Vol. 49, No. 227. pp. 46294-46301.
U.S. Environmental Protection Agency. 1984b. Proposed guidelines for
exposure assessment; request for comments. U.S. EPA, Washington, DC.
Federal Register, Vol. 49, No. 227, Part VIII. pp. 46304-46312.
U.S. Environmental Protection Agency. 1984c. Proposed guidelines for
the health assessment of suspect developmental toxicants and request for
comments. U.S. EPA, Washington, DC. Federal Register, Vol. 49, No. 227,
Part X. pp. 46324-46331.
U.S. Environmental Protection Agency. 1985a. Health assessment document
for 1,2-dichloroethane (ethylene dichloride). EPA/600/3-84/006F. Final
Report. Office of Health and Environmental Assessment, U.S. Environmental
Protection Agency, Washington, DC. Table 9-66, pp. 9-253 to 9-256.
U.S. Environmental Protection Agency. 19c!>5b. National primary drinking
water regulations; synthetic organic chemicals, inorganic chemicals and
microorganisms; proposed rule. U.S. Environmental Protection Agency, Washing-
ton, DC. Federal Register, Vol. 50, No. 219, pp. 46936-47022.
74
-------
U.S. Environmental Protection Agency. 1985c. Proposed guidelines for
the nealth risk assessment of chemical mixtures and request for comments;
notice. U.S. EPA, Washington, DC. Federal Registar, Vol. 50, No. 6, Part
III. pp. 1170-1176.
U.S. Environmental Protection Agency. -i»86. Verified Reference Doses
(RfDs) of the U.S. EPA. ^AO-CIN-475. Offtefijof Research ipd Development,
U.S. EPA, Washington, DC.
U.S. Fish and Wildlife Service. 1986. Type B technical information document:
reconmendations on use of habitat evaluation procedures and habitat suitability
index models for CERCLA applications. Draft Repoit. Habitat Evaluation
Procedures Work Group, U.S. Fish and Wildlife Servicp, Fort Collins, CO.
45 pp.
U.S. Food and Drug Administration. 1982. Levels for poisonous or deleterious
substances in human food and animal feed. U.S. FDA, Washington, DC. 13 pp.
U.S. Food and Drug Administration. 1984. Polychlorinated biphenyls (PCBs)
in fish and shellfish; reduction of tolerances; final decision. U.S. FDA,
Rockville, MD. Federal Register, Vol. 49, No. 100. pp. 2l3i4-21520.
U.S. Office of Science and Technology Policy. 1985, Chemical carcinogens;
a review of the science and its associated principles. Federal Register,
Vol. 50. pp. 10372-10442.
U.S. Office of Technology Ass3ssment. 1979. Environmental contaminants
in food. U.S. Office of Technology Assessment, Washington, DC. 229 pp.
Versar, Inc. 1985. Assessment of human health risk from ingesting fish
and crab from Commencement Bay. EPA 910/9-85-l?9. Prepared by Versar,
Inc., Springfield, VA.
Vettorazzi, G. 1976. Safety factors and their application in the lexicological
evaluation, pp. 207-223. In: The Evaluation of Toxicol og^cal Data for
the Proteccion of Public Health. Pergarcon Press, Oxford, cngland.
Vettorazzi, G. 1980. Handbook of international food regulatory toxicology.
Vol. I: Evaluations, Spectrum Publications, New York, NY. pp. 66-68.
rfhitmore, R.W. 1985. Methodology for characterization of uncertainty
in exposure assessments. Final Report. CHEA-E-160. Office of Health
and Environnental Assessment, Washington, DC. 44 pp. + appendices.
75
-------
APPENDIX A
SOURCES OF INfORMAYION FOR TOXICITY PROFILES
-------
TABLE A-l. TOXICITY PROFILES AVAILABLE FROM U.S. EPA OFFICE
OF WASTE PROGRAMS ENFORCEMENT (OWPE) AND OFFICE OF
EMERGENCY AND REMEDIAL RESPONSE (OERR)
Oat OCU Itealch
CMalul lt£tmlcml Profit* Effect* AIMMMHC
AClUBbttMIM
AcctuphtbjrltM
Ac*tlc arid
AC*t«M
Acroltln
Acrylonlcrll*
Aldrla
Aachrcccnr
Aactaom
Arwnlc
A*b**co*
tmtlum
•••!•••
••UtdtlM
Bmof«1«itlir*c«M
*wiu(*)p7Tiittreb«ni«n«
• K(2-Chlora«tba«T>«tlwa«
Chroalua) (tettl)
analua ((MUMliBC)
Otroeiua (trl*«l«at)
OirrMB*
Ca«l car*
Crtalt
C^t»r
Crcmt
Cf«Bld««
Crawwlc uld
V. -BO
•. -ODD
r! -B«
•. -not
01breoecblar«prop«M
1.2-Olcblerobvnun*
1 ,}-Oietilarob«nMM
1 .*H)lcblarao«iiMiM
l.l-«lcblore*thm«
I.I-Dlcblere€ttun«
1 . l-Olthlaro«hyl«o.«
1 .7-«i*-Dlcblere«th7l«ii«
1 . i-trac-DlehlorocehylcM
2.4-Olehlerepbeael
I.«-01cfalarafiMaar)ru«(lc acid
1 .J-MchloroproBtM
X
X
I
I
I
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
A-l
-------
TABLE A-l. (Continued)
ou
Chmlcal CheBleel
1 .3-Dlchloropropme
1 .3-Dlcbloropropene
Oleofol
Dluldria
Olethyl bensene
Dlethylene elycol
Dlethyl phthelate -
Dlleobutyl ketone
Dluothylnlnoethyl Mtheerylate
DlMtbyl ulllao
DlMthylnlcroMalM
2.4-Oia«thrl p«nt«n«
Z.4-DtB«ih]rlplMBol
D-Dioctyl phthal*t«
1 .4-Oleum
Dlphmyl cthan*
Cndrla
Ethanol
bt«(2-Oilaro«thyl) «th«r
Ether
Ethyl M«cat«
IthjrIbraMn*
Ithylra* glycel
Ethyl iMuiwdlol
bl«-2-CthTlh«xyl phtlulat*
Ethyl to hunt
Fluer«ith«n«
>orBild«hyd«
Clycol (thtrs
Itopiaehlar
M*pt«n«
H*>«eblarob«nMa«
8«uchlarabut*41«a«
H*B«chleroryelah«B«iM
fl«»cblarocyclapontad*.«M
N**«ehlarimthan«
Hnuchloraphm*
H*x«n*
Iron
Irahatyl •Icahol
I«oprepyl b*ni*B«
I»opropyl «th«r
Lead
LlthlOB
HMPMOlUB
H«l(nM««
H*rnry
MthMryllc «e!J
Hathoael
Nithyl chloride
2-Nithyl dodoesiu
IkthylMM eUorld*
Hatbyl atbyl b*ni«tu
Nvthyl othyl kctoiM
IHtocbyl hccoM
Hicbyl loabotyl kotofw
Methyl MthrcryUte
Methyl perethloa
2-Hethyl pentene
3-Methyl oentene
2-«lithyl-l-venteae
2-Rethyl tetredeeane
2-Methyl trldeune
H OEM Hulch
Profile Effect* Aeeeenatnt
Z
I
X
z
X
X
X
X
X
X
X
A-2
-------
TABLE A-l. (Continued)
Ch«»te»l froflU
OEU *t«ltb
Cfftctt M»«noa»nt
HoMchmo Inliw
Njphch*l«n*
Hlck«1
•lirocclluloM
2-MltroptMnol
Ptnt«eh]oroph«nal
P«ii«d«c«i»
PtMiunihrm*
rtMnol
Pll«ayl «th«r
naiptierlc «ctd
Phaiphorai
Picric «cld
rolychlartMC«4 blphvnrU (KB*>
rolyehlorlutcd 41b«nh«n«
, J . 3-Trlchlor«b*tu«M
. 2. 4-TricUorobonMfM
. 3, S-Trichleivbnua*
,3.6-Trlehlorebmialc «eU
.l.l-Trlchl«r
-------
TABLE A-2. U.S. EPA SOURCES OF TOXICITY PROFILES
Document
Criteria Documtnc
Air
Criteria Document
Drinking Watar
Availability
Criteria Document -
Ambient Water Quality
Chemical Hazard
Information Profile
(CHIP)
Chemical Profile
Office of Air Quality
Planning and Standards
(OAQPS)
Office of Drinking Water
(ODW)
Description
Office of Water Regula-
tions and Standards (OWRS)
Office of Toxic Substances
(OTS)
Office of Waste Programs
Enforcement (OWPE)
Summary of the latest scientific knowledge on the
effects of varying quantities of a substance in the
air. Usually prepared for OAQPS bw the Office of
Health and Environmental Assessment (OHEA).
Summary of Important experimental results from the
literature relevant to the chemistry and health
effects of a specific drinking wa£er contaminant.
Serves as a foundation to aupport 'regulatory standards
or guidelines for the acceptable concentration of tht
contaminant in the drinking water.1
Information on the type and extent of Identifiable
toxic effects on health and welfare expected from the
presence of pollutants in any body of water.
Objective of document is to protect most species in a
balanced and healthy aquatic community. To date, 65
have been completed) covering all priority pollutants.
Summary of readily available information concerning
the health and environmental effects and potential
exposure to a chemical.
Brief summary of the chemical/physical properties.
fate and transport, health effects and environmental
toxlcity levels for 202 chemicals identified at
hazardous waste sites. Currently 183 of the planned
Chemical Profiles ere available in draft form.
-------
TABLE A-2. (Continued)
Document
Health Advisory
Availability
Description
ODW
>
Health Asssssment
Documant
Health and Environ*
•ental Effacta
Profile
Health Effecta
Assessments
Office of Health and
Environmental Aaaaaament
(OHEA)
Office of Solid Waste
(OSH)
Office of Emergency and
Renedial Reaponaa (OERR)
Develops toxicological analyses to establish an
acceptable level in drinking water for unregulated
contaainanta for various exposure durations. Used in
transient aituations (spills, accidanta) therefore.
doea not conaider chronic exposure data (e.g.,
carcinogenicity).
Inventorlea the acientific literature and evaluatea
key studies. Discusses dose-response relationships so
that the nature of tha adverae health response is
evaluated in perapective with observed environmental
levels. Uaually prepared by OHEA for another office.
Profiles are "nlni-" criteria documenta prepared
usually aa summaries of existing water quality
criteria documenta. They serve aa a aupport for the
Hating of hazardous wastes In the RCRA program.
Summary of the pertinent health effects information on
58 chemicals found moat often at hazardous waste sites.
Developed by the Environmental Criteria and Assessment
Office (ECAO) for OERR.
Address for all offices listed above:
U.S. Environmental Protection Agency
401 H Street S.W.
Washington, DC 20460
(202) 382-2090
Reference: Life Systems (1985).
-------
TABLE A-3. SELECTED CHEMICAL AND TOXICOLOGICAL DATA BASES
Onin base vnnrlor
Him AIIS (National
I Iliraiy of
Medicine)
CIS
Informal.! on
System)
Ontn base sei.rt.
by vendor
D.ii.i basis contents
Access procedures
loxlIno
ChenIIne
Itrrcs (Registry
or Inxic FfTscts
or Chemical
Substances)
AQUIIU (Aquatic
I n To mo t ion
Petrleval System)
CCSARS (Cliemlciil
Evaluation Snared
and Hotrieviil
System)
1.1 nil I ion references on Contact:
iiiiviroiiiaental and toxlculnglca I
effects of clientcoJs.
An online chemical dictionary of
500.000 records.
flnsic ncntn and chronic toxlclty Tor
more tlinn 57.000 toxic chomica Is.
loxmiiy dnta Tor 2000 chemicals,
cncli cross referenced by CAS number.
lists any studies on hloaccumulatlon.
sublothal Hffects and environmental
rote of the chemical.
Octal lad toxlclty and environmental Contact:
fate Information and evaluation on
150 chemicals of Importance to Great
Lakes.
C1CP (Clinlral Ingredient and product information
toxicology nf foi most conmnre I a 11y available
Conner lea I I'roducts) nonfood items.
^nvlrofate
I SI IOH ( I n format I on
System for
llazaidons Oninnlcs
in Water)
OHM I ADS (Oi I anil
Hazardous MB MM in Is
loclinif-nl
Assistance Unt.i
System)
HfDLARS Management Section
National Library of Medicine
8600 Rockvillc Pike
Bcthnsda. Ml) ?0209
(301) U96-6I93
CIS, Inc.
fein-Harquart Associates
7215 Yoik Road
Baltimore. HO 2121?
(BOO) 24/-8M7
lnrnrin.it inn nn the environmental
lite ol approximately 500 chemicals.
Physical nnil Rlicmiral properties of
I'l.OdO onjanii: comiiniuids and
associated aquatic toxn;lty data.
r i on KM! l>y U.S. f PA super fund.
I nr liiilns i nlo iron I inn on environmental
I'iriM-Ls til II.OOO* lia/.irdous subsinnces
-------
TABLE A-3. (Continued)
On I a base vendor
Da i.a ha so
by vendor
Onm base contents
Access procedures
CAS nnI Inn
((lienien Abstracts)
OOC/III COH
Chemical Abr.c ~c«.s
I'liyslcal and clinnlc.il pni|iortlos
on 6 •!! I Inn client I cat substances.
',? energy-re I.-Hi."!
ond Rnvlronnrninl
data bases ideluding
fnnrgy Data llasc,
Water nosoiirros
Abstracts. Environmental
Hiitagons. anil
tnvlronmcruaI lernlolmiy.
Contact: Chemical Abstracts Office
Custoner Service
P.O. Rox 3012
Columbus. OH 113210
(8001 8U8-6533
Contact: Technical lnfnr»alion Center
U.S. Department of Lnergy
P.O. llox 62
Oak Ridge. IN 37380
(61b)
Reference: U.S. Fish and Wildlife Service (1986).
-------
TABLE A-4. U.S. EPA SOURCES OF CARCINOGENIC POTENCY
FACTORS AND REFERENCE DOSES
Carcinogenic Potency Factors
Carcinogen Assessment Group
RD-689
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency
401 M Street S.W.
Washington, DC 20460
(206) 382-5952
Reference Doses
Regional Risk Assessment Coordinator
U.S. Environmental Protection Agency
Region 10
1200 Sixth Avenue
Seattle, WA 98101
(206) 442-1200
Director
Office of Health and Environmental Assessment
Office of Research and Development
Washington, DC 20460
(202) 382-7317
A-8
-------
APPENDIX B
EXAMPLE DATABASE SUMMARY FOR REFERENCE DOSES (RFDS)
DERIVED BY U.S. EPA
-------
TABLE 8-1. REFERENCE DOSES (RfDs) FQR ORAL EXPOSURE
Chemical: Aldrln CAS *: 309-00-?
CareInogenlcity: CAG Class C; ql* » 11.4/(mg/kgAtey)
Systemic Toxlclty: See below.
Endpolnt
Experimental Doses
UF
MF
RfC (A01)
Fltzhugh et al.
(1964)
Rat chronic Feeding
study
Liver toxIcUy
NOAEL: 0.5 ppm diet 100
(0.025 og/kg/day)
LOAEL: 2 ppn diet
0.0003 mg/kg/day
Oose Conversion: 0.5 rag/kg Food x 0.05 kg food/kg bw
0.02S rag/kg bw/day
Endpolnt and Experimental Doses:
Fltzhugh, O.G., A.A. Nelson, and N.I. QualFe. 1964. Chronic oral toxldty of
aldrln and dleldrln In rats and dogs. Food Cosmet Toxlcol. 2: 551-562.
Groups of 24 rats (12/sex) were Fed aldrln 1n the diet at levels of 0,
0.5, 2, 10, 50, 100 or 150 ppm For 2 years. Liver lesions characteristic of
chlorinated Insecticide poisoning were observed at dose levels oF 2 ppm and
greater. These lesions were characterized by enlarged centrllobular hepatic
cells, with Increased cytoplasmlc oxyphllla. and peripheral migration of baso-
phellc granules. A statistically significant Increase In llver-to-body weight
ratio was observed at all dose levels. Kidney lesions occurred at the hlghett
dose levels. Survival was. markedly decreased at dose levels of 50 ppm and
greater.
Preparation Date: 12/05/85
B-l
-------
TABLE B-l. (Continued)
Uncertainty Factors (UFs):
The composite UP of 100 encompasses the uncertainty of extrapolation from
animals to humans and the uncertainty In the range of human sensitivities.
Kodlfylng Factors (HFs):
None.
Additional Comments:
Additional data are fairly supportive. Effect and no-effect levels are
similar (to those found for rats) for liver effects In dogs after IS months
exposure to aldrln In the diet. Liver effects are observed at slightly higher
doses In several other subchronlc-to-chronlc rat and dog studies. Short-term
exposure to higher doses result In mortality for a number of species.
Confidence In the RfD:
Study: Medium
Data Base: Medium
RfD: Medium
The critical study, designed as a carclnogenesls bloassay. Is strong In
hlstopathnlogleal analysis, but lacks other toxlcologlcal parameters. The
data ba'.e 1s fairly extensive, and generally supportive, but Is not rated
•high" oecause of the lack of NOELs for some studies. Also, no chronic data
exist for the dog. which may be more sensitive than the rat.
Documentation of RfO and Review:
U.S. EPA. 1982. Toxlclty-Based Protective Ambient Hater Levels for Various
Carcinogens. Environmental Criteria and Assessment Office. Cincinnati. OH.
ECAO-CIN-431. Internal review draft.
The RfO has been reviewed Internally by ECAO-Cln.
Agency RfO Review:
First Review:
Second Review:
Verification Date:
72/73/85
U.S. EPA Contact:
Primary: H.L. Dour son
FTS/684-7S44 or 513/569-7544
Secondary: C.T. DeRosa
FTS/6S4-7534 or 513/569-7534
8-2
-------
APPENDIX C
REGULATORY LIMITS ON CHEMICAL CONTAMINANTS IN SEAFOOD
-------
TABLl C-i. COMPILATION OF LEGAL LIMITS FOR HAZARDOUS
METALS IN FISH AND FISHERY PRODUCTS
Metals loam]
Country
Australia*
Brazil
Canada
Chile
Denmark
Ecuador
Finland
France
Germany
Greece
Hong Kong
India
Israel
Italy
Japan
Korea
Netherlands
Hew Zealand
Phil ippines
Poland
Spain
Sweden
Switzerland
Thailand
United Kingdom
United States
u.s.s.a.
Venezuela
Zambia
Range
Minimum
Maximum
As Cd Cr
1.0,1.56 0.2-5.5
3.S
0.12,1.0 0.5
1.0
5.0
0.5
1.4-10 2.0 1.0
1.0
0.05-1.0
1.0 l.G
2.0
4.0
0.1
2.0
1.0
0.1 0.0.1
3.S-S.O
0.1 0 1.0
10 5.S 1.0
Cu
10-70
10
10
10
30
.
10-30
20
20
10
100
10
100
Hg
0.5.1.0
0.5C
0.5
0.5
1.0
l.G
0.5,0.7
1.0
0.7
O.S
O.EC
O.S
0.7C
0.3.0.4C
0.5
l.QC
0.5C
C.5
G.5
l.QC
0.5
0.5
l.OC
0.2-1.0
0.1-0.5
0.2-0.3
0.1
1.0
Pf So Se In
1.5-5.5 1.5 1.0.2.0 40-1.000
0.5
2.0 0.05.0.3 100
5.0
2.0
0.5
6.0 1.0
5.0 SO
2.0
0.5,2.0
2.0 1.0 J.O 43
0.5
1.0-2.0 30-50
1.0-2.0
1.0
1.0
2.0-10 50
2.0
0.5-10 100
0.5 1.0 0.05 30
10 1.5 2.0 1,000
a Limit varies among states.
o Inorganic.
e Total.
References: Nauen (1983); U.S. Food and Drug Administration (1982, 138;).
C-l
------- |