PB89-222640
Health Risk Assessment of Chemical Mixtures
(U.S.) Environmental Protection Agency, Cincinnati,
OH
1989
L
J
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
NT1S
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
ii
-------
TECHNICAL REPORT DATA
locate read Instructions on the reverie before completing/
1. REPORT NO.
T.PA/600/D-89/028
3. RECIPIENT'S ACCESSION NO
4. TITLE ANDSUBTITLE
Health Risk Assessment of Chemical Mixtures
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
PB89-2226ttO
|7. AUTHOR(S)
J.F. Stara, «.'. Patterson, K. blackburn, R.C. Hertzberg
and C.T. UeRosa
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NCV
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Criteria and Assessment Office
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati. OH 45268
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/22
1t. SUPPLEMENTARY NOTES
In: Trace Substances in Environmental llealth-XIX. Proc. Univ. of Missouri's
19th Ann. Conf., U.O. Hemphill, Ed., June 3-6, 1985
16. ABSTRACT
The implementation of Superfund requires a methodology for estimating health risk from
multi-chemical contamination at ambient levels. Most often, the chemical composition
of these mixtures is poorly characterized, exposure data are uncertain and toxicologic
data on trie known components of the mixture are limited. However, a potential human
nealth nazaru may exist ana the U.S. uPA, state and local governments need to be able
to assess the total hazard in order to make decisions on appropriate action. This
paper describes a procedure for assessing the n'sks from chemical mixtures that
includes options when different kinds of data are available. Good-quality information
on the mixture of concern or a similar mixture should always be used. A less
desirable, Lu>£"stiil useful approach, is to utilize data on components and their
interactions. The quality of exposure and toxicity data must be determined and the
uncertainties involved in each risk assessment must be thoroughly discussed. Water
contamination is briefly discussed since it is of vital concern as the primary
exposure medium for chemical mixtures. The methodology for estimating the human
health risk froiii single chemicals, both carcinogens and systemic toxicants, is
reviewed as it fonns the basis for the assessment of mixtures. . •
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
|18. DISTRIBUTION STATEMENT
Public
b. IDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (This Report 1
Unclassified
20. SECURITY CLASS (This page/
Unclassified
c. COSATI Field/Group
21. NO. OF PAGES
//
22. PRICE
M3
PA Fo»w 2220-1 (R»r. 4-77) PREVIOUS COITION is OBSOLETE
-------
EPA/600/D-89/028
Health Risk Assessment of Chemical Mixtures
J.F. Stara, J. Patterson. K. Blackburn, R.C. Hertzberg and C.T. OeRosa
U. S. Environmental Protection Agency
Environmental Criteria and Assessment Office
Cincinnati, Ohio
ABSTRACT
The Implementation of Superfund requires a methodology for
estimating health risk from mult 1-chemical contamination at
ambient levels. Most often, the chemical composition of these
mixtures Is poorly characterized, exposure data are uncertain
and toxlcologlc data on the known components of the mixture are
limited. However, a potential human health hazard may exist
and the U.S. EPA. state and local governments need to be able
to assess the total hazard In order to make decisions on appro-
priate action. This paper describes a procedure for assessing
the risks from chemical mixtures that Includes options when
different kinds of data are available. Good-quality Information
nation on the mixture of concern or a similar mixture should
always be used. A less desirable, but still useful approach.
Is to utilize data on components and their Interatlons. The
quality of exposure and toxlc'ty data must be determined and
the uncertainties Involved In each risk assessment must be
thoroughly discussed. Water contamination Is briefly discussed
since It 1s of vital concern as the primary exposure medium for
chemical mixtures. The methodology for estimating the human
health risk from single chemicals, both carcinogens and systemic
toxicants. Is reviewed as It forms the basis for the assessment
of mixtures.
INTRODUCTION
Public awareness of environmental Issues In the U.S. Increased dra-
matically In the 1970s, and was accompanied by continuously expanding
scientific Investigations of pollution by various toxic chemicals. Anew
facet was added In the 1980s with the passage of the Comprehensive Environ-
mental Response, Conservation and Liability Act (CERCIA), the so-called
"Superfund" legislation. While previous statutes (e.g. Clean A1r Act, 1963;
Clean Water Act, 1977) were largely focused on hazards due to single chemi-
cals, the Superfund Act required the hazard evaluation of waste sites, that
Is, the assessment of health hazard from human exposures to mixtures of
chemicals.
Since 1979, the staff of the Environmental Criteria and Assessment
Office In Cincinnati of the U.S. EPA has been developing health risk
assessment guidelines and methodologies to be used In deriving "acceptable
dally Intakes" or 'risk-specific Intakes" for environmental pollutants.
Specific guidelines for health risk assessment were developed for use In
preparing the 65 Anblent Water Quality Criteria Documents, mandated by
by the Clean Water Act of 1977 (14,17).
These efforts were directed primarily towards predicting exposure
levels for Individual chemicals which would be below the theoretical
population threshold for adverse effects In the case of systemic toxicants
effects, or would estimate an Intake for carcinogens corresponding to an
upper-bound lifetime risk. Since these methodologies were originally de-
veloped to address the Issue of water criteria for various toxic species.
exposure across the entire Hfespan was a basic assumption In these esti
mates. In addition, there was no necessity for defining risk associate
with supra-threshold exposures to non-genotoxlc chemicals since the emph
sis was the definition of acceptable :xposure levels. One of the strH'
differences between Superfund and the other laws enforced by EPA Is that
under Superfund, there Is a need to estimate the hazard associated with
existing contamination levels. Thus, we had to shift from a "protective
to a "predictive" approach In the development of guidelines.
Since Superfund deals with site-specific Issues, a new spectrum of '
assessment questions have arisen. In addition to estimates of risk foil
Ing lifetime exposures, we now must consider the consequences of partial
lifetime exposures. In order to evaluate various types of remedial act*
the risk associated with exposures of varied duration at varied levels ^
be evaluated. In addition, the estimate of risk associated with exposur
to multiple chemicals at potentially supra-threshold levels has emphaslz
the need to consider Interactions between constituents. Obviously, the
goal of limiting exposure to virtually safe levels still remains. Howev
definition of existing hazard plays a vital role In determining pr1or1tt-
and In planning remedial actions. In dealing with site-specific sltuatt
many options related to exposure control exist, 1n contrast to criteria
levels where unrestricted lifetime exposure, by necessity, formed a basl'
premise.
A primary concern In the evaluation of these sites Is the potential •
migration of site constituents Into ground and surface waters. Once thr
has occurred, options for controlling exposure are severely compromised •
to current or potential use of these contaminated waters as drinking wat>
sources. It 1s then necessary to characterize Incremental exposure as a
function of site-specific factors; a difficult and highly uncertain ende?
or.
In this paper the major emphasis 1s placed on a review of the approai
dealing with health risk assessment of chemical mixtures. The approach <
health risk assessment of single chemicals Is briefly discussed because 1
forms the basis of the evaluation of mixtures. The problem of water
contamination Is briefly discussed since It 1s of vital concern as the
primary exposure medium for chemical mixtures.
WATER POLLUTION
The burning Cuyahoga River In 1969 Illustrated a serious nationwide
water pollution problem. Water basins In the U.S. have been shown to be
polluted by Increasingly complex municipal and Industrial discharges (2) •
In addition to the contamination of surface waters In this manner, ground
water pollution has recently become a general public concern. About half
of the U.S. population uses ground water as Its source of drinking water.
Ground water from Individually-owned wells represents a major drinking
water resource In many rural areas. Regionally, the middle West and West
use more ground water relative to surface water than other regions (2).
Some of these states with high ground water usage (e.g.. Nebraska,
Kansas, Oklahoma) are also characterized by agricultural activity. 1ncre«
1ng the possibility of contaminating ground water sources with pesticides
and fertilizers. A more recent and growing public concern Is the potent 1
for ground water to be contaminated from hazardous waste disposal sites.
Preventing and cleaning up contaminated ground water presents many dif-
ficulties not encountered with surface water pollution. Once an aquifer
contaminated. Its restoration as a usable drinking water supply Is extre^
difficult and/or expensive. With an Increasing rel 1 ancejn ground water.
-------
It Is important that to be able to Identify and characterize the health
risk from contaminants In sources of drinking water (16).
SINGLE CHEMICALS: RISK SPECIFIC INTAKE LEVELS AND ACCEPTABLE DAILY INTAKES
Health risk assessment deals with estimates of exposure to environ-
mental pollutants and associated health hazards. Such an assessment In-
cludes the basic toxicologic concept of dose-response relationships; for
systemic toxicants we compare actual exposure to levels which do not pre-
sent a human health hazard. For carcinogens, only the Incremental risks
associated with a pollutant level 1n a specific environmental medium are
considered. In agreement with the National Academy of Sciences (7), EPA
assumes that carclnogenesls 1s a non-threshold phenomenon, whereas other
toxic effects have thresholds. I.e., doses below which no adverse effects
will occur. As a result, the first step In single chemical risk assessment
Involves a determination of the potential carclnogenlclty of the chemical.
Depending upon this determination, the risk assessment proceeds utilizing
one of two parallel methodologies which have been designed to address
non-threshold or threshold effects.
Carcinogens (Northreshold Effects) (15)
After a compound has been deten»
MA
le
Le
animal potency (mg/kg/day)'1
assumed human weight, kg
animal weight, kg
length of exposure
length of experiment or observation period
llfespan of the animal .
The cube root of the ratio of body weights Is used to adjust for
species differences on the assumption that metabolic rate 1s proportions
to body surface area, which is proportional to the 2/3 power of body weU
The factor 1 e/Le adjusts the actual dose to a dally dose averaged over U
length of the experiment. The third factor, (Le/L)3, Is used to estlmat*
risk from lifetime exposure when the animal experiment Is only partial
lifetime. This adjustment Is necessary to allow for positive responses
that would have occurred had sufficient time been allowed for the tumors
develop (17).
After the human potency has been calculated, the Intake rate (I, 1n
mg/day) associated with a specific lifetime risk (e.g., 10*5 or 1 In 100,
Is determined:
1
70 (10-5)
potency
(2)
This risk-specific Intake rate Is easily converted Into a media cone
tration by dividing by the appropriate Intake rates for the exposure medi
For example, assuming an Intake of 2 1 water/day, the risk specific water
concentration (C, In gm/1 ) Is:
I
7
(3)
The prediction of cancer risk at a given exposure level uses the saw
basic approach outlined above, and Involves similar assumptions. When hu'
data are adequate, the observed human potency Is used directly to predict
the upper bound of risk. When animal data must be used, and particularly
when higher exposure levels are Involved, the potency alone 1s not suffi-
cient, and the complete model should be used. The risk (r) at exposure d
using the multistage model is (3):
Uexp(-q0-q1d-q2d2-...) (4)
where the qt's are parameters In the model to be estimated by curve-flttl'
procedures. The Incremental risk (or "excess risk") 1s then:
R. r(d)-r(0)
l-r(O)
(5)
An estimated upper confidence limit on the excess risk R Is used as
the lifetime risk projection at exposure level d, suitably modified is
above for species differences and for duration If the animal study was
for only partial lifetime.
Systemic Toxicants (Threshold Effects) (1«)
Five types of response levels are considered for deriving criteria
based on noncarclnogenic responses:
NOEL - No-Observed-Effect Level
NOAEL - No-Observed-Adverse-Effect Level
LOEL - Lowest-Observed-Effect Level
LOAEL - Lowest-Observed-Adverse-Effect Level
FEL - Frank-Effect Level
-------
Adverse effects are define^ as any effects that result in functional
Impairment and/or pathological lesions that may affect the performance of
the whole organism or that reduce an organism's ability to respond to an
additional challenge. Frank effects are defined as overt or gross adverse
(e.g., severe convulsions, lethality).
These concepts are Illustrated In Figure 1. They represent landmarks
that help to define the threshold region In specific experiments. Thus, If
an experiment yle'.ds a NOEL, a NOAEL, a LOAEL, and a clearly defined FEL In
relatively closely spaced doses, the threshold region has been relatively
well-defined. Such data are very useful In deriving Acceptable Dally
Intakes (AOIs). On the other hand, a clearly defined TEL Is of little use
in establishing criteria when it stands alone because such a level gives no
Indication of how far removed It Is from the threshold region. Similarly,
a free-standing NOEL has little utility because there Is no Indication of
Its proximity to the threshold region.
'•o _
s
*
A (LIGHT •OOV WEIGHT
DfCRtASE
• LIVEMMECMOSIS
C MORTALITY
? 10
OOtt (AMm«AflY UNITS)
FIGURE 1. RESPONSE LEVELS CONSIDERED IN DEFINING THRESHOLD
REGIONS IN TOXICITY EXPERIMENTS. DOSES ASSOCIATED WITH THESE
LEVELS ARE AS FOLLOWS: 3 - NOEL, NOAEL; 4 - LOEL. NOAEL; 5 -
NOAEL (HIGHEST); 7 - LOAEL; 10 - FEL; 20 - FEL. (14)
Based on the above dose-response classification system, the following
guidelines for deriving criteria from toxicity data can be used.
0 A free-standing FEL is unsuitable for the derivation of criteria.
0 A free-standing NOEL is unsuitable for derivation of criteria. If
multiple NOELs are available without additional data on LOEL's.
NOAEL's, or LOAEL's, the highest NOEL should be used to derive -.
criterion.
0 A NOAEL. LOEL. or LOAEL can be suitable for criteria derivation. A
well-defined NOAEL from a chronic (at least 90-day) study can be
used directly, dividing by the appropriate uncertainty factor, For
a LOEL, a Judgment must be made as to whether It actually corresponr
to a NOAEL or a LOAEL. In the case of a LOAEL, an additional
uncertainty factor is applied; the magnitude of the additional
uncertainty factor Is judgmental and should lie 1n the range of 1 tc
10. Caution must be exercised not to substitute Frank-Effect Levels
for Lowest-Observed-Adverse-Effect Levels.
0 If—for reasonable closely spaced doses—only a NOEL and a LOAEL of
equal quality are available,the appropriate uncertainty factor 1s
applied to the NOEL.
In using this approach, the selection and Justification of uncertainty
factors are critical . The National Academy of Science (7) has provided
guidelines for using uncertainty factors. "Safety factor" or "uncertainty
factor" Is defined as a number that reflects the degree or amount of
uncertainty that must be considered when data from animal experiments are
extrapolated to humans. When the quality and quantity of experimental data
are satisfactory, a low uncertainty factor Is used; when data are Judged to
be Inadequate or equivocal, a larger uncertainty factor Is used. In those
cases where the data do not completely fulfill the conditions for one cate-
gory—or appear to be intermediate between two categories—an intermediate
uncertainty factor Is used. Such intermediate uncertainty factors may be
developed based on a logarithmic scale. These Issues were reviewed by
Dourson and Stara In 1983 (5). In order to determine the acceptable
exposure level In water, the highest NOEL or NOAEL, or the lowest AEL
(depending on the data available) is divided by one or more uncertainty
factors (17). The ADI 1s then substituted for ! In equation 3 above.
APPROACH TO TOXICITY DATA EVALUATION
For the purposes of projecting hazard associated with a defined exposure
to a single toxicant, all toxicity data should be used along with the ADI.
This dictates that the data need to be rated for severity of effect and that
the exposure level and duration be converted to equivalent human values.
This procedure allows all available toxicity data on a chemical to be used
In predicting the results of a given exposure to humans. One method for
incorporating all the data in a predictive assessment is demonstrated by
Figure 2, a graph of study dose and duration versus effect severity.
-------
ioo.ono
KXOOO
woo
-------
TABU I. RISK ASSESSMENT APPROACH FOR CHEMICAL MIXTURES*(19)
1. Assess the quality of the data on Interactions, health effects and
exposure.
a. If adequate, proceed to Step 2
b. If inadequate, proceed to Step 14.
2. Health effects Information Is available on the chemical mlxture^of
concern.
a. If yes, proceed to Step 3.
b. If no, proceed to Step 4.
3. Conduct risk assessment on the mixture of concern based on health
effects data on the mixture. Use the same procedures as those for
single compounds. Proceed to Step 7 (Optional) and Step 12.
4. Health effects Information Is available on a mixture that Is s1mt-J-ar
to the mixture of concern.
a. If yes, proceed to Step 5.
b. If no, proceed to Step 7.
5. Assess the similarity of the mixture on which health effects data are
available to the mixture of concern, with emphasis on any differences
In components or proportions of components, as well as the effects that
such differences would have on biological activity.
a. If sufficiently similar, proceed to Step 6.
b. If not sufficiently similar, proceed to Step 7.
6. Conduct risk assessment on the Mixture of concern based on health
effects data on the similar mixture. Use the same procedures as
those for single compounds. Proceed to Step 7 (Optional) and Step 12.
7. Cbuplle health effects and exposure Information on'the components of
the mixture.
8. Derive appropriate Indices of acceptable exposure and/or risk on the
Individual components In the mixture. Proceed to Step 9.
9. Assess data on Interactions of components In the mixtures.
i. if sufficient quantitative data are available on the Interactions
of two or more components In the mixture, proceed to Step 10.
b. If sufficient quantitative data are not available, use whatever
Information 1s available to qualitatively indicate the nature of
potential interactions. Proceed to Step 11.
Table 1 (cont'd)
10. Use an appropriate interaction model to combine risk assessments on
compounds for which data are adequate, and use an additivity assump-
tion for the remaining compounds. Proceed to Step 11 (optional) and
Step 12.
11. Develop a risk assessment based on an additivity approach for all
compounds in the mixture. Proceed to Step 12.
12. Comoare risk assessments conducted in Steps 5, 8 and 9. Identify
and Justify the preferred assessment, and quantify uncertainty, if
possible. Proceed to Step 13.
13. Develop an Integrated summary of the qualitative and quantitative
assessments with special emphasis on uncertainties and assumptions.
Classify the overall quality of the risk assessment. Stop.
i4. No risk assessment can be conducted because of Inadequate data on
interactions, health effects or exposure. Qualitatively assess the
nature of any potential hazard and detail the types of additional
data necessary to support a risk assessment. Stop.
*Note that several decisions used here, especially those concerning ade-
quacy of data and similarity between two mixtures, are not precisely
characterized and will require considerable Judgment.
Margins of Safety (MOS), or acceptable concentrations of a pollutant in
various media. For example, an Index for multlroute exposure based on the
ADI is:
HIC
(6)
where E Is equal to exposure expressed In units of daily Intake. The
subscripts denote exposure route, i.e., 0 • oral, I • inhalation, and D -
dermal . For example, Ej and ADlj may be defined in terms of air concentra-
tion. This approach Is taken where route specific "virtually safe" Intake
rates or practical thresholds are estimated. The contribution of each
route Is added to estimate the total hazard index of a single chemical.
Naturally, this approach is most appropriate when route-specific criteria
1 .e based on systemic effects not related to portal of entry.
In order to compare total chemical exposure to an ADI value, exposure
In units of media concentration must first be converted to an absorbed
eg/day dose. For example:
water exposure level (mg/1) x water consumption (1) x absorption
coefficient • water exposure (mg/day); or
air exposure concentration (mg/m3) x dally ventilation volume (a3)
x absorption coefficient « air exposure (mg/day).
-------
Assuming that the circulating blood level Is responsible for the effects,
then the ADI for any route may be used. The circulating threshold can be
expressed 1n terms of any ADI with Its corresponding absorption coefficient
TABLE II. EXAMPLE OF ASSESSMENT Of MULTIPLE TOXICANT EFFECTS RISK
ASSESSMENT
Circulating Threshold • (ADIoHan.) " (ADI])(a|) « (ADIn.)(a0) (7)
A single chemical Hazard Index for all exposure routes would be as follows:
{E0)(ao)*(E,)(a,)*(ED)(aD)
(8)
where the Ej can represent either oral, Inhalation or dermal.
As an example, to estimate a total hazard Index across one route, In
this case Inhalation, we would use the Inhalation ADls and Inhalation
exposures for each chemical :
El E2 EN
(9)
The following equation can be used to determine the simple sum of the route
Indices, as In Equation 6, or It can be based on the absorbed levels, as In
Equation 8.
HIT - Hi! * HI2 * ... + HIN (10)
In the case of carcinogens, a similar additive approach can be used, except
that Instead of AOIs, doses corresponding to fixed risk levels such as 10'^
are used for each chemical .
The Hazard Index allows a qualitative estimate of whether or not a
hazard may exist. I.e.. a Hazard Index greater than 1 for either a single
chemical across routes or lor a group of chemicals determined to have
potentially additive effects would qualitatively Indicate a potential for
hazard and suggest further Investigation.
In order to quantitatively estimate hazard, one conceptually simple
approach Is response addition In which the correlation of Individual re-
sponses within the population is assumed to be zero. The formula for
predicting the total expected response (Pf) from exposure to two chemicals,
using this assumption, can be expressed as: PT • PI»P? (!-PI). This equa-
tion can be generalized for any number of chemicals, as:
1 - 11 (1-P,)
1-1
(11)
For example, using this equation, the total Incidence based on all adverse
responses (PCj) from each of five chemicals Is given in the last column
of Table II, and the total Incidence for each adverse effect caused by the
combination of chemicals (PE{) Is given in the last row of Table II. In
this theoretical case, each of the five chemicals In the table exhibited
two types of adverse responses (15).
Chemical
Effects of concern
Compound
I
II
III
IV
V
"1
A
2xl--2
5x10-3
2.49x10-2
B C
3x10-3
4x10-2
6xlO-4
3.60x10-3 4xlO"2
D E
8xlO*4
1x10-3
9x10-3
9.79x10° 1x10
F
7x10-3
6x10-3
-3 1.30x10-'
PCi
2.08x10-2
4.00x10-3
4.67x10-?
1.39x10-2
6.60x10-3
? 8.9xlO'?
Source: Adapted from Stara, el al., 1985.
The calculation of P£j 1s a straightforward use of the above equation.
The calculation of PCi, the total Incidence of adverse responses caused by
each chemical . is somewhat different In that the assumption Is that the
separate effects Induced by a given chemical are Independent of one another
For some combinations of effects (e.g.. Increased liver weight, MFO Induc-
tion, proliferation of SER In liver cells) this assumption obviously will
be Invalid. For such cases. It may be more reasonable to assume that the
correlation of tolerances approaches unity (15).
Accepting for the moment that the responses of concern have been select>
so that the assumption of Independence among responses is reasonable, the
total Incidence based on all adverse effects from all chemicals (Pj) can be
calculated from equation 11, substituting PE< for Pj. The use of equation
11 1s best Justified when overall risks are small, and may be better justi-
fied for calculation of cancer risk than of toxic risk. However, where
multiple effects In the same Individual are judged to be significant, their
probability can be easily calculated seperately and expressed as an ad-
ditional factor to be considered (15).
While this approach Is conceptually slnple. It Is predicated on the
ability to estimate response rates for the spectrum of effects associated
with the Individual chemical components. Dependent upon the data base
available for the Individual chemicals and the effects of concern, this may
or may not be possible.
With each risk assessment of a chemical mixture there Must be a thorough
discussion of related uncertainties. They should be clearly discussed and
the overall quality of the risk assessment should be determined based on
an expression of the degree of confidence In the quality of the data on
interaction, health effects and exposure.
DISCUSSION AND CONCLUSION
The various Issues in environmental health dealt with In this paper need
to be tied together In order to be understood by scientists who are not
intimately familiar with risk assessment procedures as they relate to the
Implementation of environmental laws. The short discussion dealing with
potential contamination of drinking water supplies In the U.S. Is Important
because of the present emphasis by the Agency on ground water contamination.
It appears that once an aquifer 1s contaminated by various toxic compounds
it is much more difficult to develop appropriate controls than It Is for
-------
surface water pollution. Of course, a possibility exists that we Just know
much more about controlling contamination of surface water than ground
water, and that It will take a few years before we have accumulated the
knowledge and experience to deal with ground water contamination In a
similar, rather satisfactory, fashion. Up" to now, risk assessments for
water pollutants (and other media) have addressed single chemicals. In
order to understand the approaches being used for chemical mixtures, the
single chemical assessments as they relate to a particular environmental
medium have to be reviewed.
The Office of Drinking Water of the U.S. EPA has established In their
risk assessment documents five levels to control adverse effects In the
population due to drinking water. They are: 24-hour health advisories for
10 kg children ano 70 kg adults; 10-day health advisories (10 kg and 70
kg); and a chronic effect criterion to control life-long exposure to pollu-
tants In adult populations. Currently, all the values are determined for
single chemicals. However, as discussed In the body of the paper, when we
deal with hazardous waste sites, water reuse situations, or Instances of
combined effects of various air pollutants, we must consider that the popu-
lations are exposed to mixtures of chemicals. On that basis, we must
develop testing procedures that are less time-consuming than standard
bloassays, and yet, are satisfactory from a standpoint of data requirements
for risk assessment procedures.
Many of the approaches described In this paper do not Include the full
range of Issues typically addressed In health risk assessments performed by
EPA. Normally such reports represent a detailed consideration of a pollu-
tant's chemical and physical properties and a description of Its behavior
both physiologically and environmentally. Perhaps most Important Is the
practice of presenting risk-specific Intake levels for carcinogens and ac-
ceptable dally Intakes for systemic toxicants In such a way as to fully
describe all assumptions and uncertainties associated with these quantita-
tive estimates. Furthermore, this process always entails scientific peer
and policy review.
It Is Important to distinguish between hazard and risk assessment.
Hazard assessments represent a primarily qualitative characterization of
the spectrum of effects associated with exposure In addition to the para-
meters discussed above. Risk assessments, on the other hand, are usually
based on the hazard assessment (4) as 1t relates to the exposed population,
e.g., comparison of exposure to acceptable dally Intake. Ideally, the
assessment predicts the Incidence of effects In an exposed population with
•11 assumptions and uncertainties clearly articulated.
A primary problem encountered In risk assessment Is defining risk at the
low exposure levels typical of environmental human exposures. Despite
data gaps on chemical toxlclty not only for dose (In this Instance low
dose) but for route, duration and species of concern, there Is often a real
end/or perceived need to characterize risk. This need for characterization
Is not only for specific chemicals, but for different combinations of chemi-
cals, multimedia exposures, and different endpotnts of concern. This
characterization effort is contingent upon biologically-based inference and
extrapolation to develop a working hypothesis formulated to project risk.
(4). This type of assessment effort is not Intended to be In 1 leu of bio-
assay assessments, but rather to be complementary to bioassays"These
factors are Important for risk assessment as opposed to risk management
issues such as regulatory Impact analysis or cost benefit analysis.
This brief overview of current approaches to risk assessment of chemi-
cal mixtures and the related methodologlc developments cannot fully reflect
the extent and the comolexitv of the efforts required for these tasks.
while some of the new developments related both to Individual systemic
toxicants and chemical mixtures, such as an Improved approach to inter-
species dose conversion, have been In progress for several years. Others,
such as methodologies for partial lifetime health risk evaluation and the
determination of sensitive population subgroups, are relatively recent.
The proposed guidelines for health risk assessment of chemical mixtures (50
FR 1170) attempt to address the Issues briefly discussed here, to assure a
consistent approach to assessing mixtures and suggest promising areas of
future research. Much effort is needed both
-------
13. Smyth, H.F.. C.S. Wei'. J.S. West, and C.P. Carpenter. 1970. An ex-
ploration of joint toxic action. II. Equltoxlc versus equlvolume
mixtures. Toxlcol . Appl . pharmacol . 17: 498-503.
14. Stara, J.F., M. L. Dourson and C.T. OeRosa. 1981. Section 3 - Water
Quality Criteria: Methodology and Applications. In; Conference
Proceedings: Environmental Risk Assessment, "How New Regulations Will
Affect the Utility Industry." EPRI Contract No. WS-80-134.
15. Stara, J.F., R.C. Hertzberg, R.J.F. Bruins, M.L. Dourson. P.R. Ourkln,
L.S. Erdrelch and W.E. Pepelko. 1985. Approaches to Risk Assessment
of Chemical Mixtures. In: F. Homburqer and J.K. Marquis, eds. Chemi-
cal Safety Regulation ana1 Compliance: Proceedings of a course held In
Cambridge, MA. October 24-25, 1983. Basel, Switzerland: S. Karger,
AG.
16. Stara, J.F.. 0. Kello and P. R. Durkln. 1980. Human health hazards
associated with chemical contamination of aquatic environment. Envi-
ronmental Health Perspectives, 34: 145-168.
17. U.S. EPA, 1980. Guidelines and Methodology used In preparation of
health effects assessment chapters of the consent decree water criteria
documents. Federal Register 45: 79347, Nov. 28, 1980.
18. U.S. EPA, 1985. Proposed guidelines for the health risk assessment
of chemical mixtures. Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office. Federal Register 50:
1170. January 9, 1985.
19. U.S. EPA, 1985. Guidelines for the health risk assessment of chemical
mixtures. Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Draft.
20. WHO (World Health Organization). 1981. Health effects of combined ex-
posures In the work environment. WHO Tech. Rept. Series No. 662.
Dose-Response Analysis of Heavy Metal Toxicants in Man:
Direct In vivo Assessment of Body Burden
Kenneth J. Ellis, Ph.D.
Medical Research Center
Brookhaven National Laboratory
Upton, N.y. 11973
ASS TRACT
Differences In uptake, metabolism, and excretion of heavy
metal a makes selection of a suitable biological Media aa a
monitor of body burden very difficult. Exposure assessments
based on body fluid levels can provide, at best, only general
population estimates. The most frequently monitored Media are
blood, urine, nail or hair clippings, sweat, and saliva.
Unfortunately each of these tissues can be Influenced by
recent exposure conditions and are not accurate Indices of the
total dose or body burden. Direct in rtro measurements of
body burden In humans, however, have recently been performed.
This nuclear technique has focused on the measurements of
kidney and liver cadmluai (Cd) by neutron activation analyala
and bone lead (Pb) determination* using x-ray fluorescence.
The dose-response relationship for renal dysfunction based on
the direct In vivo body burden for Cd Is presented. The most
probable Cd ralue for the kidney associated with renal Impair-
ment Is approximately 35 ng. Approximately 10Z of thesubjecta
with 20 ng Cd In the kidney will have moderately elevated
p2-«'croglobulIn, an early Indicator of porentlal renal
functional changes.
INTRODUCTION
There Is llttlt doubt that significant adverse health effccta are
evident for high exposures to heavy metals (16,21,28). Animal studies
have usually been the source of this Information. In many caaes these
studies have been performed In animal species pre-selected for their
enhanced response to a particular adverse effect, independent of the
expoaure agent. The doses used In toxlcologlcal studies may be much
higher than those observed even in the cost significant of industrial
expoaures or accidents. Extrapolation of these experimental findings to
humans, therefore, becomes difficult. Nuch less Is even known about the
effects of low level chronic exposure or "envlrcReental" expoaure to
heavy metals.
Two basic parameters, however, must be obtained in order to
Investigate dose-response or dose-effect retstlonshlps. These are (1) an
adequate quantification of the true dose and (2) Identification of the
biological response. The second requirement has generally received the
most research effort. Measurements of the adverse effects following an
exposure can be performed without knowing the body burden or true dose to
the target organ. Subtle physiological and biochemical changes,
previously undetected, are becoming apparent as more sophisticated
------- |