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

-------