A Proposal for a National Risk
Assessment Clearinghouse

Paulette L. Stenzel*

I. INTRODUCTION

Over the past two decades, the focus of the environmental pro-
tection movement has shifted from efforts to clean up visible air
and water pollution to efforts to deal with the even greater
hazards' posed by toxic chemicals.2 Although toxic chemicals are
often invisible to the human eye, they pervade our daily lives,
presenting hazards in the workplace, in our homes, and out-of-
doors. Our current efforts to clean up both visible and invisible
pollution in the environment, protect workers and the general
population, and prevent the creation of future hazards are re-
ferred to as "risk management."3 Risk management decisions
are based on "risk assessments"4 prepared by toxicologists. The
methods used by toxicologists in the risk assessment process,
however, are riddled with uncertainties and assumptions, and the
data produced is often inconclusive as to the toxicity of the chem-
ical being assessed. Therefore, risk management decisionmakers
must supplement the information provided by the science of toxi-
cology with careful choices of public policy.

* Associate Professor, Business, Law and Public Policy, Michigan State University. The
author wishes to acknowledge the assistance of Michigan State University's Center for En-
vironmental Toxicology, which provided partial funding for this research.

1. It is important to distinguish between the concepts of "risk" and "hazard." A "haz-
ard" is a threat to us as humans and to what we value. "Risk," which is often expressed in
quantitative terms, is a conditional probability of suffering harm. See Hohenemser, Kates
& Slovic, The Nature of Technological Hazard, 220 SCIENCE 371 (1983).

2. In 1972 public opinion polls on environmental issues showed that most people were
worried about water and air pollution - especially smog. Now, even though those
problems remain largely unsolved, polls show that as a public concern air and water pollu-
tion runs behind a new environmental threat - toxic chemicals. Commoner, A Reporter at
Large: The Environment, THE NEW YORKER, June 15, 1987, at 46, 52.

3. Risk management is the process through which steps are taken to "reduce or elimi-
nate the risk that has been revealed by the risk assessment." M. KAMRIN, ToXICOLoGY-A
PRIMER ON TOXICOLOGY PRINCIPLES AND APPLICATIONS 137 (1988).

4. A risk assessment, usually expressed in quantitative terms, is an attempt to determine
"the potential toxic effects due to chemical exposure in a particular situation." Id.

549



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

Because risk management is a complicated process, Congress
has delegated authority to make risk management decisions to
powerful administrative agencies including the Environmental
Protection Agency (EPA), 5 the Occupational Safety and Health
Administration (OSHA), 6 the Food and Drug Administration
(FDA) 7, and the Nuclear Regulatory Commission (NRC)8 . But, as
citizens in a democracy, the public retains the right to be in-
formed about the nature of specific risks, and demands a right to
participate in risk management decisions. Former EPA adminis-
trator William Ruckelshaus observed, "We have decided, in an
unprecedented way, that the decisionmaking responsibility in-
volving risk issues must be shared with the American people, and
we are very unlikely to back away from that decision."9

Yet, most laypeople do not understand general concepts of
risk, and they have not been educated with respect to the risk
management process. Further, they are uninformed or poorly in-
formed about specific risks which they confront personally or
which are the subject of agencies' risk management decisions.

The consequences of poor risk communication are costly to our
society. Many individuals become unduly alarmed with respect to
minimal risks while ignoring risks of much greater magnitude.
The public distrusts administrative agencies and contests or, at a
minimum, does hot readily accept their risk management deci-
sions. Further, agencies do not receive the informed public input
they need.

In order to earn the public's trust and gain its input, adminis-
trative agencies must be compelled to share risk information in a
meaningful, consistent manner. In The Need for a National Risk
Assessment Communication Policy, I proposed a National Risk Assess-
ment Communication Policy Act (RACPA) designed to promote
that goal in two phases.'0 First, the Act would set up an Advisory
Board and guidelines to be used by the Board. The Board would
establish a format to be used by agencies in describing specific
risks to the public. Second, the Act would include an educational

5. National Environmental Policy Act of 1969, 42 U.S.C. § 4321 (1982).
6. Occupational Safety and Health Act of 1970, 29 U.S.C. §§ 651-678 (1982).
7. Federal Food and Drugs Act, 21 U.S.C. §§ 1-15 (1906), amended by Federal Food,

Drug and Cosmetic Act, 21 U.S.C. §§ 301-392 (1938).
8. Energy Reorganization Act of 1974, 42 U.S.C. §§ 5801-5879, 5891 (1982).
9. Ruckelshaus, Communicating about Risk, in RISK COMMUNICATION 5 (1987).
10. Stenzel, The Need for a National Risk Assessment Communication Policy, II HARv. ENvL.

L. REV. 381 (1987).

550



National Risk Assessment Clearinghouse

component through which Congress would require administra-
tive agencies to actively disseminate risk assessments to the public
through the news media, pamphlets, workshops, and similar edu-
cational programs.

This article expands on the second phase of the proposed Act.
In addition to, and especially in the absence of affirmative risk
education programs, the public needs and is entitled to open ac-
cess to the risk assessments used by administrative agencies in
their risk management decisions. This article first reviews the
public's general need to understand risk and examines its specific
need for access to risk assessments. Secondly, it demonstrates
that risk assessments are not readily available even to those who
actively seek them. The process of obtaining a risk assessment is
time-consuming and cumbersome. Further, risk assessments are
conveyed in a form which is difficult for laypeople to understand.
A centralized source, through which risk assessments would be
made readily accessible to the public, is needed.

To this end, I propose a National Risk Assessment Clearing-
house. This article describes the operation of the proposed
Clearinghouse and cites precedent for such a program. What the
Clearinghouse can accomplish is described; potential criticisms
are evaluated; and limitations of a Clearinghouse are discussed.
Overall, it is illustrated that establishment of a Risk Assessment
Clearinghouse is one of a series of steps we must take as we strive
for open, two-way risk communication between administrative
agencies and the public.

II. BACKGROUND-THE PUBLIC's GENERAL NEED TO

UNDERSTAND RISK AND SPECIFIC NEED FOR ACCESS TO

RiSK ASSESSMENTS

The public's need to understand the nature of risk and the risk
assessment and management processes is crucial. Risk-related
decisions are of great consequence: they can determine the state
of our health and even life and death. Risk-related decisions per-
vade our daily lives as we choose food and water, select an occu-
pation, accept or reject employment, choose a place in which to
live, and purchase consumer products. To varying degrees, de-
pending on their willingness and ability to do research before un-
dertaking such activities, people make many risk-related decisions
on an individual basis. For that reason alone, individuals need to

1989] 551



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

be well-informed about the nature of risk and the significance (in-
cluding limitations) of the data presented in a risk assessment.

We rely on administrative agencies to make many risk manage-
ment decisions for us, however, because the information needed
to make informed decisions is complicated and lengthy. Thus, to
some extent, we rely on the Food and Drug Administration to
keep "unsafe" foods off the grocery shelves, and we rely, again to
some extent, on the Occupational Safety and Health Administra-
tion to monitor our work environment.

One of the reasons frequently cited for creating agencies and
for delegating decisions to them is that an agency has the re-
sources which enable it to focus on a specific problem in detail
and provide expert advice and decisions." This ideal might be
achieved in the case of risk assessment if scientists within agencies
could provide us with "facts" about toxic effects of chemicals.
However, as discussed infra,12 the "data" produced by toxicolo-
gists is riddled with uncertainties and, moreover, such data does
not exist for many chemicals. Therefore the risk management
decisions of regulatory agencies rely substantially on public policy
to supplement the incomplete scientific "data."

The policy choices, made by agencies as they supplement the
gaps in information resulting from uncertainties, are made both
consciously and unconsciously. Such policy choices are involved,
for example, when our country encourages one industry over an-
other concern by granting tax incentives or through its regulatory
structure. In recent years the United States government has
given greater incentives to the development of nuclear power
plants than to the development and implementation of solar
power. Currently, our government gives greater encouragement
and more assistance to trash-to-electricity incinerators than to re-
cycling in an attempt to deal with the disposal of toxic and non-
toxic garbage. Similarly, value choices are involved as agencies,
courts, and other governmental decisionmakers distribute bur-
dens and benefits among various populations and among present
and future generations, and when these decisionmakers place
"dollar" values on health and life.' 3 For example, a decision to

11. K. DAVIS, DISCRETIONARY JUSTICE 46 (1969).

12. See infra text accompanying notes 30-43 (discussing the limitations of data produced
through a quantitative risk assessment).

13. See Bazelon, Science and Uncertainty: A jurist's View, 5 HARv. ENvrt. L. REV. 209, 211
(1981) (discusses conscious and unconscious policy choices made in environmental law-

552



National Risk Assessment Clearinghouse

allow higher levels of asbestos in the workplace atmosphere may
save two million dollars for an employer and save a hundred jobs.
The "cost" of these "benefits" may be ten additional employee
deaths from lung disease over the next ten years. Thus, the em-
ployers (including their managers and shareholders) benefit from
the hi'gher levels of asbestos. However, the benefit to others is
questionable in view of the risks of illness and death imposed on
employees and the hardships imposed on their families. Such
cost/benefit analyses are usually the basis for our agencies' risk
management decisions.'4 Yet, the use of such cost/benefit analy-
ses has been strongly criticized by various researchers because
this method of analysis provides an illusion of certainty and fails
to acknowledge the value of choices that were made as the
cost/benefit result was compiled.'5

Because of the prevalence of value-laden decisions in the area
of risk management, we should therefore limit our reliance on
agencies and their experts in that area. In the alternative, it is
crucial that individual citizens participate in those agencies' deci-
sions. When value choices are involved, the opinion of a layper-
son is as valuable as that of any expert or scientist. With respect
to the policy decisions in risk management, any layperson can
form an opinion about those policy decisions based on values,
morals, ethics, or other internal beliefs. No "expert" has the abil-
ity to tell another person whether it is better to allow higher levels
of exposure and save a hundred jobs at the expense of ten lives,
or to save those lives at the expense of lost jobs and lost corpo-
rate earnings. Any layperson may contribute a valid and valuable
perspective to the decisionmaking process.

Further, our democratic political system is based on the prem-
ise that citizen participation in such policy-based decisions is es-

suits and emphasizes that those choices affect all members of society, not just the parties to
the lawsuit).

14. See generally K. SHRADER-FRECHETTE, SCIENCE POLICY, ETHICS, AND ECONOMIC METH-

ODOLOGY (1984) (provides in-depth discussion of risk/cost-benefit analysis, and illustrates

that this analysis pervades our risk management decisions, and criticizes its use as a deci-

sionmaking methodology in risk management).

15. In spite of the prominent role of cost-benefit analysis in regulatory decision-making
today, placing a dollar value on the loss of human life or health is both morally objectiona-
ble and impossible. See Kennedy, Cost-Benefit Analysis of Entitlement Problems: A Critique, 33
STAN. L. REV. 387 (1981). See also Shaw & Wolfe, A Legal and Ethical Critique of Using Cost-
Benefit Analysis in Public Law, 19 Hous. L. REV. 899 (1982) (criticizes the use of cost-benefit
analysis as a decisionmaking methodology.)

5531989)



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

sential. Policy'6 choices involve the selection of goals and
guiding principles from among various available alternatives.
Thus, the exercise of discretion is integral to the policymaking
process. In the United States, we adhere to the premise that a
primary task of our government and its institutions is to provide a
forum for the voicing of divergent viewpoints. Through our
political process, competing viewpoints can be aired before the
exercise of discretion takes place.'7 The more important the pol-
icy being chosen, the more crucial it is that there be an open,
vigorous airing and discussion of the perspectives and desires of
individual citizens. Therefore, it is particularly offensive for pol-
icy decisions directly involving human health and life to be made
without full discussion and. consideration of citizens'
viewpoints. 18

When agencies do not encourage citizen participation and do
not engage in two-way communication with the public with re-
spect to policy decisions, citizens disrespect or contest the agen-
cies' actions. The EPA's handling of the 1984 controversy over
the pesticide EDB (ethylene dibromide) presents an example of
the public's antagonism toward an agency resulting from the
agency's poor risk communication. When the EPA banned most
agricultural uses of EDB, the EPA was operating "in an area of
enormous scientific uncertainty."'9 That uncertainty, however,
translated into a public perception of evasiveness. Further, the
EPA failed to inform the public of the options available in terms
of siting, technology, or other means of waste "disposal" and
failed to discuss publicly who would bear the costs of various op-
tions.20 Risk perception expert Baruch Fischhoff says that the
EPA's failure to discuss options and the consequences of those

16. Policy can be defined as "the general principles by which a government is guided in
its management of public affairs, or the legislature in its measures. . . . This term, as
applied to a law ordinance, or rule of law, denotes its general purpose or tendency consid-
ered as directed to the welfare or prosperity of the state or community." BLACK'S LAW
DICTIONARY 1041 (5th ed. 1979).

17. Felix Frankfurter said, "[Tihe essential task of law - and its greatest triumph - is
to devise peaceful accommodations, expressive of the dominant ideals of western demo-
cratic civilization, for the clash of interests and feelings in a dynamic society." Frankfurter,
Foreward, 47 YALE L. J. 515 (1938).

18. See generally Stenzel, Toxic Substance Regulation: A Compelling Situation for Revival of the
Delegation Doctrine, 24 AM. Bus. L.J. 1, 15-17 (1986) (argues that citizen involvement is
crucial in making toxic substance regulatory policy).

19. Press, Science and Risk Communication, in RISK COMMUNICATION 14 (1987).

20. Fischhoff, Managing Risk Perceptions, 2 IssUES Sci. & TECH. 83, 95 (1985).

554



National Risk Assessment Clearinghouse

options left "a confused impression of the issues and the agency's
concern for the public's welfare." 21 A panel of experts reviewing
the controversy agreed that the EDB episode was not a public
health crisis, but, rather, "a crisis in communication."22

The consequences of poor risk communication can be seen in
an ongoing controversy in Michigan. In April 1986, a public out-
cry arose in response to the city of Detroit's plans to build the
world's largest trash-to-electricity incinerator. Initially, public
awareness and protest were provoked when the Michigan Depart-
ment of Natural Resources (DNR) released a study which it pre-
pared in an effort to persuade the City of Detroit to change its
original plans and employ new technology for air filtering.23

Among other concerns, citizens were outraged because they
learned of the plant only upon the release of that report even
though the plant had been planned for over ten years. The EPA
became involved and threatened to file a lawsuit to stop construc-
tion of the plant on the grounds that the incinerator would not
meet federal Clean Air Act requirements.24 The EPA dropped its
plans for litigation when it realized that it had given its tacit ap-
proval to the project through an EPA audit conducted in 1985,25
but a lawsuit was filed by several environmental groups and a
community group. They sought an injunction to stop construc-
tion of the incinerator or at least force the City to employ the new
air filtering technology available. Although the lawsuit was dis-
missed in federal court, an appeal is pending and active citizen
opposition to the project continues.26

Access to risk assessments and better risk communication in
general will not eliminate public opposition to agency decisions
relating to the siting of a trash incinerator or a waste dump, or
decisions about the types and quantities of pesticides that will be
allowed in our foods. If the public is well informed about risk in

21. Id.
22. Johnson, EDB (Ethylene Dibromide), in RISK COMMUNICATION 83-85 (1987).
23. The new technology for air filtering was not available in 1984 when the DNR origi-

nally granted a permit for the incinerator. Detroit Free Press, Apr. 4, 1986, at 1, col. 4.
24. On May 12, 1986, the EPA's regional administrator announced that the incinerator

would not meet Federal Clean Air Act requirements and that the EPA would go to court to
force the City of Detroit to reduce pollution levels if it did not do so voluntarily. Detroit
Free Press, May 13, 1986, at 1, col. 5.

25. Detroit Free Press, Sept. 23, 1986, at 1. col. 5.
26. See also Stenzel, supra note 10, at 382-85 (discusses examples of agencies' failure to

satisfy the public's desire for information about risks and its desire for influence over their
management).

19891 555



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

general, however, it will be capable of evaluating new risk infor-
mation more rationally. Moreover, if the public can readily ob-
tain the risk assessments available to administrative agencies, it
can more quickly "get down to the business" of evaluating and
dealing with the risks involved instead of struggling to obtain that
information from an agency which appears to be evasive or even
antagonistic.27

Therefore, in keeping with our democratic tradition, it is crucial
that administrative agencies communicate with the public about
the nature of risk and specific risk decisions. Meaningful citizen
input in the regulatory process can only be achieved if the public
understands the risk assessment and risk management processes.
Understanding of these processes requires further comprehen-
sion of the nature and limitations of the data used. In addition, in
order to participate in specific risk management decisions, the
public must be provided with "enabling information:" 2 8 the risk
assessments which the administrative agency is using in the deci-
sionmaking process.29

A. Risk Communication is a Complicated Process

Due to the nature of the risk assessment process, it is particu-
larly difficult to educate and inform the public about risk. The
complex subject matter cannot be adequately conveyed to the
public in a haphazard or unplanned fashion due to a number of
complicating factors including: (a) uncertainties inherent in the
risk assessment process; (b) the manner in which people evaluate
risk information; and (c) popular distrust of scientists, the risk
assessors.

27. Id. at 410-12 (discusses benefits of open communication about risk in managing spe-
cific hazards).

28. The work of sociologist James S. Coleman is relevant to this discussion. He asserts
that until recently our society virtually ignored the information rights of individuals. Re-
cently, those rights have been addressed in an ad hoc, inadequate manner through such
legislation as the Federal Freedom of Information Act (5 U.S.C. § 552 (1982)). Professor
Coleman calls for the development of a political theory of information rights to guide us as
we attempt to provide citizens with the information they need to participate in the political
process in a meaningful way. See generally J. S. COLEMAN, THE ASYMMETRIc SOCIETY
(1982); J. S. COLEMAN, POWER AND THE STRUCTRE OF SocIETY (1974).

29. "[B]efore access can become meaningful, there must be sufficient information in the
hands of the public for citizens to be effective participants. The process of supplying this
information is at the heart of risk communication." Ruckelshaus, Communicating About Risk,
-in RISK COMMUNICATION 6 (1987).

556



National Risk Assessment Clearinghouse

Quantitative risk assessments (which are compiled as the result
of a toxicity assessment30 and an exposure assessment3 ) are the
basis for risk management decisions.32 Yet, uncertainties pervade
the process of compiling quantitative risk assessments. In order
to proceed in spite of those uncertainties, scientists make numer-
ous assumptions as they assess the toxicity of a chemical and as
they assess the exposure of a human population.

There are four main types of tests used to assess toxicity, three
of which are based on animal bioassays. A single exposure test is
used to determine the LDso (lethal dose for 50 percent of the ani-
mals studied).33 In the "subacute" test, the effects of daily expo-
sure of a group of animals to a chemical are studied.34 To test for
chronic (long-term) toxicity, the most commonly used test is the
carcinogenesis bioassay in which high doses are used to produce
cahcer in groups of about fifty animals.35 The high-dose animal
study data is used to estimate a low-dose value for humans. The
data produced using each of these three tests is based on a series
of assumptions. For example, society-is concerned about a risk to
one human in 100,000 or a million, yet finances limit tests to
thousands or hundreds of animals. Therefore, toxicologists use
data from small numbers of animals to estimate effects on large
numbers of people. The species used may be more or less sensi-

30. Toxicity assessment is the "evaluation of the toxicity of a chemical based on all
available human and animal data." M. KAmRIN, supra note 3, at 138.

31. Exposure assessment is the attempt to determine the risks of a chemical to a particu-
lar human population. Id. at 59.

32. It should be noted that the EPA views the risk assessment as consisting of four com-
ponents:

(1) Hazard identification: in which hazards and potential toxic effects of a contaminant
are identified.

(2) Dose-response assessment: determines the dose or amount of a contaminant
which may produce an adverse effect in humans.

(3) Human exposure assessment: assesses the potential for humans to come in contact
with the contaminant.

(4) Risk characterization: combines the above three components to estimate the po-
tential health impacts from the contaminant.
Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. 33,994-99 (1986). The EPA's
"hazard identification" and "dose-response assessment" correspond to the process de-
scribed in this article as "toxicity assessment." It is worthwhile to note that the EPA's "risk
characterization" provides niore than a quantitative risk assessment by presenting some of
the uncertainties associated with the particular risk assessment.

33. M. KAMRIN, supra note 3, at 45-48; M.A. OTrOBONI. THE DOSE MAKES THE POISON: A
PLAIN-LANGUAGE GUIDE TO TOXICOLOGY 61-62 (1984).

34. M. KAMRIN, supra note 3, at 48-50; M.A. OrrOBONI, supra note 33, at 33, 78.
35. M. KAMRIN, supra note 3, at 51-53; M.A. OrOBONI, supra note 33, at 77-82.

1989] 557



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

tive than humans to the substance being tested. Various dose-
response models are used to extrapolate from high-dose animal
study data to an estimated low-dose value for humans.36 But
these extrapolations produce predictions, not "facts."

The fourth main tool for "testing" toxicity is epidemiology: the
retrospective or prospective study of human populations to estab-
lish correlations between a particular toxic substance and a spe-
cific effect such as cancer.37 Retrospective studies, which depend
on recollections of people being studied or on medical or occupa-
tional records, are problematic.38 Recollections may be selective
or otherwise unreliable. Prospective studies can eliminate
problems of recall, but they involve other problems.39 Records
may lack necessary data. People taking part in a study may be-
have differently once they become subjects and, for example, be-
come more health or exercise conscious than the norm. Because
ours is a mobile society, study subjects may also be difficult to
track over long periods of time. An additional result of the long-
term nature of these studies is that it takes years before data and
conclusions, if any, will be available. Scientists admit,
"[E]pidemiology is at best a crude science"40 and that, overall,
"[T]oxicity testing is done in the face of basic biological and
chemical ignorance."4 '

The exposure assessment, the second stage in compiling a risk
assessment, is similarly riddled with uncertainties and assump-
tions.4 2 Present concentrations of a chemical in the air or water
can be measured, but, generally, data needed to determine past
concentrations is not available. The scientist extrapolates back-
ward in time making the assumption that conditions in the past
were identical to present conditions. Further, for present or past
exposure, assumptions about length of exposure (hours per day
and years) are made for a hypothetical average person. Thus,
under its Carcinogen Risk Assessment Guidelines, the EPA as-
sumes that an individual has been exposed to the chemical being

36. M. KAMRIN, supra note 3, at 51-53.
37. Id. at 53;J. URQUHART & K. HEILMANN, RISK WATCH: THE ODDS OF LIFE 31 (1984).

38. M. KAMRIN, supra note 3, at 54. M.A. OrroBONI, supra note 33, at 81-82.
39. M. KAMRZN, supra note 3. at 55: M.A. OrroBoNI, supra note 33, at 81-82.
40. Leape, Quantitative Risk Assessment in Regulation of Environmental Cardnogens, 4 HARV.

ENvrT. L. REV. 92 (1980).
41. M. KAMRIN, supra note 3, at 57.
42. See generally M. KAMRIN, supra note 3. at 59-61 (discusses uncertainties inherent in

the exposure assessment).

558



National Risk Assessment Clearinghouse

assessed for twenty-four hours daily for seventy years.43 (It
should be noted that the EPA is considering the adoption of the
new assumptions that a person is exposed two to sixteen hours
per day for ten to thirty-five years.)4 4 Whichever set of assump-
tions is used, however, the assumptions seldom if ever, will accu-
rately summarize the exposure of an actual exposed individual.
Yet, there is no way the agency can do anything but generalize,
because individuals' exposures do vary widely. Therefore, it is
imperative that the agency convey the assumptions it has used as
it communicates a risk assessment. Then, an individual will be
able to compare his or her situation to the assumptions and begin
to reframe the risk assessment into some sort of individualized
perspective.

Overall, administrative agencies and their scientists cannot
eliminate the uncertainties inherent in the risk assessment pro-
cess. Agencies and their scientists must identify and openly ac-
knowledge the existence of the uncertainties. At the same time,
citizens must learn to recognize and deal with the uncertainties in
order to participate in the risk management process in a meaning-
ful way.

Yet, many non-scientists do not understand the uncertainties
and are unaware of the assumptions used in the risk assessment
process. Frequently, their understanding of the character and
magnitude of a specific risk is uninformed. When people do not
have statistical evidence with which to evaluate risks, they rely on
general inferential rules known as "heuristics" to help them re-
duce difficult mental tasks to more manageable ones.45 The ap-
plication of one such rule results in a tendency to simplify
complex decisions.46 This is illustrated when people demand that
risk managers state whether a food, chemical, or technology is
safe or unsafe. They fail to realize that safety is a continuous vari-
able.4 7 With reference to research on the health effects of pollu-
tion, Senator Edmund Muskie said he was looking for a "one-
armed" scientist who does not say, "On the one hand, the evi-

43. Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. 33,992 (1986).
44. Remarks made at a symposium on February 12, 1988 by Peter Preuss, Director of

EPA's Office of Technology Transfer and Regulatory Support, as reported in 1988 Chem.
Reg. Rep. 1806.

45. Slovic, Fischhoff & Lichtenstein, Perceived Risk: Psychological Factors and Social Implica-
tions, A376 PROCEEDINGS ROYAL SOCIETY LONDON 17, 18 (1981).

46. Fischhoff, Managing Risk Perceptions, 2 ISSUEs ScI. & TEcH. 83, 86 (1985).
47. Id. at 86.

1989] 559



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

dence is so, but on the other hand. .. ."48 This desire for simple
answers is reflected in the inclusion of the Delaney Clause in the
Federal Food, Drug and Cosmetic Act,49 which prohibits the use
of any food additive which in any amount has been found to in-
duce cancer in animals or humans.50 This provision is an attempt
to force regulators to categorize a food as being "safe" or
"unsafe."5 '

Another example is California's controversial "Proposition 65"
which states, "No person in the course of doing business shall
knowingly and intentionally expose any individual to a chemical
known to the state to cause cancer or reproductive toxicity with-
out first giving clear and reasonable warning to such individ-
ual. . .. "52 The consumer warning requirements will apply to any
chemical on a list put out (and periodically updated) by Califor-
nia's governor.53 The warnings for consumer products will read,
as appropriate, "WARNING: This product contains a chemical
known to the State of California to cause cancer," or "WARNING:
This product contains a chemical known to the State of California
to cause birth defects or other reproductive harm."54 Similar
wordings will be required for alcoholic beverages, fresh fruits,
vegetables or nuts.55 Critics of Proposition 65 argue that it fails
to distinguish between major causes of human cancer as com-
pared to minimal risks. In view of the fact that our conclusions

48. As quoted in Allman, Staying Alive in the 20th Century, 85 SCIENCE 30, 33 (1985).
49. 21 U.S.C. H§ 301-392 (1982).
50. 21 U.S.C. § 348(c)(3)(A) (1982).
51. It should be noted, however, that the Food and Drug Administration currently ap-

plies a "de minimis" (negligible risk) standard as it interprets the Delaney Clause. Under
the "de minimis" standard the FDA permits the use of "a carcinogenic food additive when
those uses are shown to present a potential carcinogenic risk that is so trivial, based on
conservative statistical analyses, as to be the functional equivalent of no risk at all." 50
Fed. Reg. 51,551 (1985) (to be codified at 21 C.F.R. Pt. 700) (explains "de minimis" stan-
dard and applies it to use of methylene chloride in decaffeinated coffee, concluding that
action with respect to that use is not necessary).

52. Safe Drinking Water and Toxic Enforcement Act of 1986, CAL. HEAL-u & SAFETY
CODE § 25,249.6 (West 1986).

53. Id. § 25,249.8.
54. CAL. ADMIN. CODE tit. xxii, § 12,601 (b)(4). For consumer products other than alco-

hol, for which there is a separate list of options, the regulations allow businesses to choose
among three types of warnings. The three options for consumer products are:

(1) Labels affixed to the product or its packaging;
(2) identification at the retail outlet through signs, shelf labels, or other means; or
(3) a system of advertising, signs and toll-free information services, or "any other sys-

tem, that provides clear and reasonable warnings." Id. § 12,601(b)(1).
55. Id § 12,601(b)(4).

560



National Risk Assessment Clearinghouse

about causes of human cancer are frequently based on data ob-
tained from animal testing, one critic of Proposition 65 empha-
sizes, "No human diet can be entirely free of mutagens or agents
that can be carcinogenic in rodent systems."56 By asking that a
food or other product be labeled as cancer producing or not, or
as causing birth defects or not, people are trying to compel regu-
lators (or product manufacturers) to respond to complex ques-
tions with simple "yes or no" answers. However, such simple
"solutions" may only complicate matters by raising unfounded
health scares, obscuring priorities, and causing limited resources
to be spent in unwise ways, leaving other more serious health
risks unaddressed.5 7

People use other general inferential rules which further distort
or inhibit their understanding of risks. The "availability heuris-
tic" is a rule stating that people remember what they see or expe-
rience or what is familiar to them. Thus, they overestimate causes
of death which are dramatic and sensational (such as fires, toxic
spills, and homicides) and they underestimate the frequency of
less spectacular events (stroke, diabetes, and smallpox
vaccinations.)5 8

56. Biochemist Bruce Ames commenting on Proposition 65, as quoted in Wall St. J.,
June 15, 1987, at 26, col. 1. Dr. Ames' opposition to Proposition 65 has been outspoken
and well publicized. This reference to Dr. Ames does not indicate my concurrence with his
position with respect to Proposition 65. However, further discussion of Proposition 65
and of Dr. Ames' views is beyond the scope of this article.

57. Proposition 65 may cause harm as well as good. Philip Abelson, Deputy Editor of
Science magazine says, "Labeling a large number of items as carcinogens because they con-
tain parts per billion of something of doubtful carcinogenicity will not enable the public to
act more judiciously in safeguarding health." He cites Milton Russell, a former assistant
administrator for Policy Planning and Evaluation at the EPA who said in a different but
similar context:

Real people are suffering and dying because they don't know when to worry, and
when to calm down. They don't know when to demand action to reduce risk and
when to relax, because health risks are trivial or simply not there. I see a nation on
worry overload . . . . Anxiety and stress are public health hazards in themselves.
When the worry is focused on phantom or insignificant risks it diverts personal atten-
tion from risks that can be reduced.

Abelson, Caifornia's Proposition 65, 237 SCIENCE 1553 (1987).
Scientists are concerned because the public devotes more attention to some risks than is

appropriate. For example, "[M]any scientists feel that the recent and continuing preoccu-
pation with risk management of dioxins represents an exaggerated response to a risk
which is of lesser importance than others which have received less attention." M. KAMRIN,
supra note 3, at 67.

58. Press, Science and Risk Communication, in RISK COMMUNICATION (1987); Slovic,

Fischhoff & Lichtenstein, supra note 45, at 18.

561



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

In addition, once people (both laypeople and scientists in-
cluded) make up their minds, it is difficult to change them. They
are unaware of the uncertainties and tenuous assumptions on
which their own assessments of risk are based,59 and they fail to
look for and examine contradictory information. One risk per-
ception expert commenting on this queried, "[H]ow many envi-
ronmentalists read Forbes and how many industrialists read the
Sierra Club's Bulletin to learn about risks (as opposed to reading
those publications to anticipate the tactics of the opposing
side)?"so

Laypeople need more education about risks in general and in-
formation about specific risks sd that they will not have to rely on
heuristics to evaluate them. Regulators and their scientists must
communicate more openly about risks with the public in order to
learn what the public has to say about risks.61

Even conscious efforts by scientists and regulators to communi-
cate risk information to the public face an additional barrier
which can be characterized as "popular" distrust of scientists.
People doubt the conclusions reached by scientists. They de-
mand certainty where there is none62 and generally expect more
of scientists than they are able to provide.63 This tendency can be
alleviated, however, by educating the public about the risk assess-
ment process and by communicating openly about risks.64

59. Slovic, Fischhoff & Lichtenstein, supra note 45, at 20-2 1.
60. Fischhoff, supra note 46, at 87.
61. Risk communication must be a two-way process. The regulatory community and

scientists have much to learn from the public about factors the public uses in defining risk
and in evaluating specific risks. "Experts" tend to define risk and rank a risk according to
the number of fatalities it has caused in the past. Laypeople, on the other hand, consider a
variety of other factors as they evaluate the severity of a risk. It is crucial that "experts" be
informed about criteria which are important to laypeople as they evaluate risks. However,
in-depth discussion of educating and informing experts about laypeople's perspectives is
beyond the scope of this article. See generally Stenzel, supra note 10, (discusses need for
two-way risk communication and proposes mechanisms to promote it).

62. See supra text accompanying notes 46-57.
63. M.A. OrroBONI, supra note 33, at 178.
64. Irrational distrust of science transcends the more particularized problems created

by inadequate communication about risks and risk assessments. A better general science
education in the public schools would help to alleviate irrational distrust of science and
would help reduce the public's unrealistic expectations of scientists. This applies with
respect to the public's expectations for risk assessments as well as their expectations of
other "data" obtained through scientific study.

562



National Risk Assessment Clearinghouse

B. Current Avenues for Risk Communication in General Are
Inadequate.

Problems of risk communication have been actively discussed
among toxicologists, medical doctors, sociologists, and psycholo-
gists for the past decade. But, within our society in general, and
within regulatory agencies, risk communication has only recently
been recognized as an important subject.6 5 Thus, we are only in
the very early stages of devising effective means of risk
communication.

The public is not being formally educated about the nature of
the risk assessment process. Toxicology, which provides the data
used in preparing a risk assessment, is a relatively new science
which has developed since the 1960s.6 6 Most adults have not
studied toxicology as a part of their general public school educa-
tion. Further, even today there is no comprehensive program
designed to include such materials in the public schools' curricu-
lum. Therefore, education of laypeople about concepts integral
to the risk management process is left to more informal mecha-
nisms, if any.

The public relies on the news media for environmental and risk
information. "[Tihe media often play the role of transmitter and
translator of information between government agencies and citi-
zens . . . ."67 For example, in a recent study, New York residents
were questioned about their sources of environmental informa-
tion. That survey found that the three most extensively used
sources were, first, television; second, radio; and third,
newspapers.68

The news media, however, are inadequate sources of risk infor-
mation. Experts from various fields criticize the news media for
presenting risk information in a manner which frightens laype-

65. Davies, Introduction, in RISK COMMUNICATION 1 (1987).
66. The first national recognition of toxicology as a distinct science came in 1961 with

the formation of the Society of Toxicology. In academic institutions, departments having
toxicology as part of their title were created starting in the mid-1970s and 1980s. As a
result, there are few scientists specifically trained in toxicology. M. KAMRIN, supra note 3,
at 2.

67. Covello, von Winterfeldt & Slovic, Risk Communication: An Assessment of the
Literature on Communicating Information about Health, Safety and Environmental Risks,
Draft Preliminary Report to the Environmental Protection Agency Uan. I1, 1986) (avail-
able from the National Science Foundation, Washington, D.C.).

68. P. Yarbrough & F. Yarbrough, Pesticides and Related Environmental Issues: A
Study of the Opinions and Behaviors of New York Adults 46 (1985) (available from the
Department of Communication Arts, Cornell University).

1989] 563



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

ople and overwhelms them with exaggerated predictions of risk.
For example, toxicologist M. Alice Ottoboni cites headlines in-
cluding, "Pesticide Poisonings, Harvest of Illness," "Now, Cancer
in the Crib," and "Cancer in a Spray Can."69 Doctors John Ur-
quhart and Klaus Heilmann assert that the media's "faddish" way
of reporting risk-related stories makes it difficult for the public to
rationally evaluate risks.70 Studies confirm that the news media
devote a disproportionate amount of coverage to catastrophic
causes of death.7' The research of risk perception experts con-
firms that biased news coverage leads to distorted perceptions of
risks in our society.72

Yet, the news media should not receive all blame for the phe-
nomena labelled "news media toxicology." 73 When there is a
toxic leak from a factory or a community learns that its water sup-
ply is contaminated with chemicals, significant press coverage is
an appropriate response to the public's desire for information.
Further, it is unrealistic to expect the news media to be the pri-
mary source of risk education and risk information in our society.

Many responsible reporters welcome the opportunity to obtain
clear, consistently-presented risk information from administrative
agencies.74 It is their job to talk to experts and "get the story."
But, as one researcher observes, "the difficult part of getting most
environmental stories is that no one has them to give."7 5

There are some small scale programs designed to answer the
public's questions and facilitate access to risk information. They
include, for example, the Center for Environmental Toxicology at
Michigan State University and the New England Community En-

69. M.A. OTrOBONI, supra note 33, at 10.
70. J. URQUHART & K. HEILMANN, supra note 37, at 98-100.
71. W. LOWRANCE, OF ACCEPTABLE RISK 118 (1976); Combs & Slovic, iVewspaper Coverage

of Causes of Death, 56 JOURNALISM Q. 837 (1979).
72. See generally Combs & Slovic, supra note 71.
73. Stenzel, Book Review, 13 ECOLOGY L. Q. 361, 364 (1986) (reviewing J. URQUHART &

K. HEILMANN, RISK WATCH: THE ODDS OF LIFE (1984)).
74. When faced.with a decision as to what arsenic air pollution standard should be im-

posed on a Tacoma, Washington air smelter, the EPA made a special effort to involve the
public. "The media were really very cooperative, not on the editorial page but in the news
stories, in trying to present information as accurately as possible." Ruckelshaus, supra note
9, at 7.

75. Fischhoff, Protocols for Environmental Reporting: What to Ask the Experts, [Winter 1985]
JOURNALIST 11.

564



National Risk Assessment Clearinghouse

vironmental Education Project.76 However, such programs are
not widely available within individual states or across the United
States.

Therefore, a substantial portion of the task of risk education
and information dissemination must be assumed by the adminis-
trative agencies charged with managing risks. Administrative
agencies should convey risk information directly to the general
public and to intermediaries such as the news media, which, in
turn, will convey it to the public. Yet, our administrative agen-
cies' approaches to risk communication are uncoordinated and in-
consistent from one situation to another. In spite of the fact that
risk communication may be the most crucial element in risk man-
agement today, our agencies have no overall policy or program
for educating the public about risks. Therefore, in The Need for a
National Risk Assessment Communication Policy,77 I proposed legisla-
tion through which Congress would require agencies to actively
disseminate risk information to the public. A primary objective of
the Risk Communication Policy would be to provide citizens with
background understanding of the nature of risk and an informa-
tion base which would assist them in understanding and evaluat-
ing specific risks.

In addition to the general understanding of the nature of risk,
the public needs access to specific risk assessments. When a citi-
zen learns of a toxic spill or of a chemical in his or her drinking
water, that person wants to know, "What is the chemical?" Then
he or she asks, "Is it harmful?" "What can it cause?" "To
whom?" "In what amounts?" "After how much exposure?" This
is the kind of information which administrative agencies seek to
provide in a risk assessment. This information is necessary so
that individuals may translate information about "macro" risks
(generalized societal risks) into "micro" risk information (risks to
an individual human being).78 Individuals want and deserve to

76. For information about these programs, contact the Center for Environmental Toxi-
cology, Holden Hall, Michigan State University, E. Lansing, Michigan, and the New Eng-
land Community Environmental Education Project, West Hartford, Connecticut.

77. Supra note 10.
78. For example, one commentator states, "It is macro language that estimates the

global likelihood that a given chemical will lead to cancer. The pronouncements are not in
micro language that would tell the housewife whether she should buy a cake powder
tinged with traces of EDB." Press, Science and Risk Communication, in RISK COMMUNICATION
I1, 16 (1987).

1989) 565



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

know, to the extent feasible, "What does this risk assessment
mean to me and to my family members?"

C. The Public's Need for Access to Risk Assessments

Laypeople need access to risk assessments. Ellen Silbergeld, a
Senior Scientist with the Environment Defense Fund, emphasizes
that in order to wield power in the area of environmental politics,
citizens need opportunities for choice and equal access to re-
sources. She explains that such resources include the technical
information provided in risk assessments.79

The need for access to risk assessments is clear in the "crisis"
situation in which the public learns of a toxic spill on a highway,
suspects that its community water supply is contaminated, or
wishes to respond to proposals for new pollution-emitting facili-
ties in their communities. For example, in the early 1980s in
Woburn, Massachusetts, citizens became aware of an abnormally
high incidence of childhood leukemia in certain neighborhoods in
their community.80 Upon investigation, the families of victims
learned that their water supplies were contaminated with chemi-
cals including trichloroethylene (TCE) and tetrachloroethylene
(perchlorethylene) (PCE). One of their next steps, of course, was
to seek risk assessments for those chemicals."' Similarly, citizens
involved in the Detroit, Michigan incinerator controversy82 need
access to specific risk assessments for the dioxins, furans, heavy
metals and other pollutants found in the ash residue produced by
such incinerators.83

Beyond discussion provoked by a crisis, the public is showing
increasing awareness of the existence of risks in daily life. One
cannot pick up a newspaper without reading about some risk-re-
lated situation, whether it is asbestos in workplace buildings, air-
borne toxins in an urban area, or pesticides in foods. Awareness
leads to increased demands for information.

79. Silbergeld, Panel on Responjibilities of Risk Communicators, in RISK COMMUNICATION 27,
34 (1987).

80. Wald,Jury in Cancer Death Suit Says Factory Polluted Eells, N.Y. Times,July 29, 1986, at
A8, col. 1; TIME, Apr. 7, 1986, at 27.

81. Id.
82. See supra text accompanying notes 23-26.
83. See generally Detroit Trash Incinerator: We Say No! (undated 16 page collection of

essays and articles opposing the incinerator, distributed April, 1988 by The Evergreen
Alliance, c/o P.O. Box 02455, Detroit, Michigan, 48202.)

566



National Risk Assessment Clearinghouse

Such awareness has also led to the enactment of new legislation
designed to require that the existence of risks be revealed to the
public. Implementation of such legislation, in turn, increases the
public's demands for access to assessments of those risks.

One example of such legislation is OSHA's 1983 "Hazard
Communication Standard" which requires chemical manufactur-
ers, importers, and distributors to provide to employers evalua-
tions of all hazardous or toxic materials they sell or distribute.4

This information is conveyed in a Material Safety Data Sheet
(MSDS) for each chemical involved. (Recent amendments to the
Hazard Communication Standard make the requirements applica-
ble to non-manufacturing employers as well as those in manufac-
turing.85) Employers must, in turn, make the MSDS available to
their employees.86 The MSDS must identify the hazardous chem-
ical, describe its physical hazards (such as potential for fire and
reactivity), list health hazards of the chemical, and identify OSHA
exposure limits. These data sheets, however, are necessarily
brief. Further, they depend on information supplied by the
chemical manufacturer, which has a strong incentive to present
information which deemphasizes the risks involved.

The Superfund Amendments and Reauthorization Act of 1986
(SARA)8 7 includes new "Emergency Planning and Community
Right-to-Know" provisions which extend right-to-know concepts
to communities in which businesses are located.8 The provi-
sions, which were adopted in response to the December, 1984,
toxic leaks which caused mass death and injury in Bhopal, India,89

require businesses to reveal the presence and identities of hazard-
ous chemicals to the communities in which the businesses' facili-
ties are located. Among its other provisions, SARA requires that
the community be provided with access to Material Safety Data

84. In addition, over forty cities and over twenty states have passed workers' right-to-
know laws. Wall St. J., Dec. 14, 1984, at 22, col. 5.

85. OSHA Hazard Communication Standard, 52 Fed. Reg. 31,852; 29 C.F.R.
§ 1910.1200 (1987).

86. OSHA Hazard Communication Standard, 48 Fed. Reg. 53,280 (1983); 29 C.F.R.
§ 1910.1200 (1987).

87. Comprehensive Environmental Response, Compensation and Liability Act of 1980,
42 U.S.C. §§ 9601-9675 (1982 & Supp IV 1986); as amended by Superfund Amendments
and Reauthorization Act of 1986, Pub. L. No. 99-499, 100 Stat. 1613 (1986).

88. Emergency Planning and Community Right-to-Know Act of 1986, 42 U.S.C.
§ 11001-11050 (Supp. IV 1986).

89. ENVIRONMENTAL LAw REPORTER, SUPERFUND DESKBOOK 13 (1986).

5671989]



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

Sheets which are identical to those required by OSHA under its
Hazard Communication Standard.co

Thus, under SARA citizens will be informed of the presence
and identity of large quantities of chemicals in their communities.
They will also receive a brief account, prepared by the manufacturer of
the chemical, of the hazards each chemical may present. Neither
OSHA's Hazard Communication Standard nor SARA requires
businesses to perform toxicological studies. SARA does not re-
quire businesses to assess the risks to which they expose the peo-
ple living near their facilities. But SARA and OSHA's Hazard
Communication Standard do give workers, communities, and
health professionals access to information on which they may
base further studies to determine exposure and compile risk as-
sessments. One commentator notes, "This is clearly an intended
and welcome consequence in a world where chemical manufactur-
ers have not always been honest about the true dangers created
by their products."9 '

When communities are made aware of the presence and iden-
tity of a chemical such as trichloroethylene, its members will want
more detail about its potential effects. They will ask, "What can
it cause?" "What amounts cause such effects?" "What are the
chances that this will happen?" "To whom?" "Over how much
time?" So, community and worker right-to-know laws are only a
first step in informing the public about chemical risks in their
neighborhoods and workplaces. The public needs more detail
than is provided in the MSDS. In addition, citizens need access to
risk assessments prepared by someone other than the chemical's
manufacturer.

California's Proposition 65 and February 1988 regulations im-
plementing it will raise similar demands for access to risk assess-
ments. Proposition 65 will require manufacturers and retailers in
California to warn consumers about any product containing a
chemical known by the state to cause cancer or reproductive tox-
icity.9 2 That warning, however, will simply be in the form of a
label stating that the product contains a chemical known by the
state of California to cause "cancer," or "birth defects or other

90. Emergency Planning and Community Right-To-Know Act of 1986, 42 U.S.C.
§ 11021 (Supp. IV 1986).

91. Tobey, Right-to-Know and the Chicken Little Syndrome, 7 MICH. ENvrT.. L.J. 6, 7 (1987).
92. Safe Drinking Water and Toxic Enforcement Act of 1986, CAL. HEALTH & SAFETY

§ 25,249.5 (West 1986).

568



National Risk Assessment Clearinghouse

reproductive harm." As under the Delaney Clause, the conclu-
sion that a chemical causes toxic effects will often be based on
data obtained from animal tests. Also like the Delaney Clause,
the law obscures the fact that risks should be viewed on a contin-
uum: most questions of product safety are not "black and white"
questions of a product being "safe" or "unsafe."9 3 Proposition
65 warnings will alert consumers to the existence of risks of can-
cer or reproductive harm, but the warnings will not provide any
information about the magnitude of those risks. Thus, Proposi-
tion 65 warnings will give the public even less information than is
provided to workers and communities under federal "right-to-
know" laws. Because twenty or more states may be considering
similar laws,94 the situation in California should not be viewed as
unique. If similar legislation is enacted elsewhere, the public's
needs and demands for access to risk assessments will continue to
increase.

D. Risk Assessments Are Difficult to Obtain and, Once Obtained,
Difficult to Understand

The next relevant question is, where and how does the public
obtain risk assessments or the information in them? The answer,
in short, is that risk assessments are not readily available.

As was discussed supra,95 the news media are a major source of
environmental information. Yet, media coverage is crisis-ori-
ented. Further, when risk assessments are presented, they are
presented in inconsistent formats. Thus, one news reporter
speaks of "doubling your risk of cancer" while another tells us
that we will see "five additional cases of cancer per million peo-
ple." Yet, the news media is only conveying information it re-
ceives elsewhere.

93. See supra notes 52-57 and accompanying text.

94: "Currently 21 states are believed to have similar laws on their legislative agen-
das. . . ." Best & Nelson, Is Our Food Supply Safe?, 1987 PREPARED FOODs 154. The Sierra
Club claims that Arkansas, Colorado, Louisiana, Missouri, New York, and Wisconsin have

shown an interest in adopting laws similar to Proposition 65. Marshall, California 's Debate
on Carcinogens, 235 SCIENCE 1459 (1987). Carl Pope, an official of the Sierra Club, says,
"[w]e're working on a generic version of Proposition 65 that other states can use, and
we're getting questions from everywhcre," as quoted by Lindsey, Many States Move to Curb

Disposal of Chemicals. N.Y. Times, Dec. 21, 1986, at A26, col. 1.
95. See text accompanying notes 67-68 (discussing media's prominent role in communi-

cating risk information).

1989] 569



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

The general public and news media alike can obtain risk assess-
ments from administrative agencies. Before seeking information,
however, the "researcher" must identify the agency which has
conducted a risk assessment. Various agencies in federal govern-
ment conduct research on and regulate the same toxic substances
independent of each other. For example, OSHA, the EPA, and
the Consumer Product Safety Commission (CPSC)96 have each
studied and taken action to regulate the chemical formaldehyde.97

Each of the fifty states has administrative agencies which may con-
duct their own studies of various chemicals independent of the
federal agencies. President Reagan's "New Federalism" ap-
proach to environmental regulation, under which more and more
responsibility for environmental regulation was transferred to the
individual states, intensified the dispersion of research and
information.

After identifying the source of information, a researcher must
obtain the risk assessment from that source. Occasionally the risk
assessment or a summary is available from an agency's public in-
formation office.98 Portions of some risk assessments have been
published in the Federal Register.9 But a Federal Register
search is tedious even for lawyers; it is too much to expect of the
layperson. When those means fail, a researcher's primary tool for
obtaining a risk assessment from an agency is a request based on
the federal Freedom of Information Act (FOIA) 00 or a state level
equivalent. However, a response to an FOIA request often takes
months. A news reporter wants the story today, not in several
months. Similarly, a citizen concerned about arsenic or trichloro-
ethylene in his or her water wants as much information as is avail-

96. For information on this agency, see Consumer Product Safety Act, 15 U.S.C. § 2053
(1982).

97. See M. KAMRIN, supra note 3, at 87-93 (gives a "case history" summarizing toxicolog-
ical studies conducted on formaldehyde and regulations promulgated by various federal
administrative agencies).

98. For example, the EPA has prepared a "Summary of Risk Assessment and Proposed
Risk Management Actions" (fact sheet dated April 1988, available from U.S. EPA, Office of
Public Affairs, Region 5, 230 S. Dearborn, Chicago, IL 60604). The fact sheet summarizes
the EPA's risk assessments for Midland, Michigan, the site of numerous facilities operated
by chemical manufacturers.

99. For example, see Cosmetics: Proposed Ban on the Use of Methylene Chloride as an
Ingredient of Aerosol Cosmetics, 50 Fed. Reg. 51,551 (1985) (to be codified at 21 C.F.R.
pt. 700) (summarizes toxicity studies including animal bioassays and epidemiological stud-
ies and presents risk assessments for methylene chloride in cosmetics and as a food addi-
tive for decaffeination of coffee).

100. 5 U.S.C. § 552 (1982).

570



National Risk Assessment Clearinghouse

able as soon as possible. A Freedom of Information Act request
is a tool that is too cumbersome and time-consuming to ade-
quately serve the needs of laypeople seeking risk assessments.

Further, once a risk assessment is obtained, it is presented in a
format which is long and difficult for a non-scientist to under-
stand.o'0 Although members of the public should be given access
to the "unabridged" materials, most need assistance in under-
standing those materials.

In summary, the public's needs and demands for risk informa-
tion are not being met. Further, those needs and demands can be
expected to increase significantly in the near future as community
right-to-know legislation and similar programs are implemented.
We need to assist citizens in understanding the nature of risk and
the risk management process through affirmative risk education
and communication programs.0 2 In addition, we must facilitate
access to risk assessments for those who actively seek them. At
the present, obtaining a risk assessment is a time-consuming,
cumbersome task, and the information, once obtained, is difficult
for the layperson to understand. Therefore, in the next section of
this article, I recommend that the federal government establish a
centralized source for risk assessments and provide summary,
plain-language versions of those assessments to the public.

III. ANALYSIS-MAKING RISK ASSESSMENTS MORE READILY
AVAILABLE TO THE PUBLIC.

In this section, I propose a national "Risk Assessment Clearing-
house" to be established by Congress. This section outlines the
format for the Clearinghouse and explores its operative provi-
sions. Also, discussion of specific benefits, responses to potential
criticisms, and acknowledgement of the limitations of a Clearing-
house are presented.

A. A Risk Assessment Clearinghouse

Congress should establish a centralized office charged with col-
lecting risk assessments from administrative agencies. The Risk
Assessment Clearinghouse would be independent of the major

101. For example, the EPA's draft toxicological profile for trichlorethylene is 140 pages
long. One hundred and four of those pages present toxicological data, an exposure assess-
ment, and related data. AGENCY FOR Toxic SUBSTANCES AND DISEASE REGISTRY, U.S. PUB-
uc HEALTH SERVICE, TOXICOLOGICAL PROFILE FOR TRICHLOROETHYLENE (1988).

102. Such programs are discussed in Stenzel, supra note 10.

1989] 571



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

regulatory agencies involved in risk management such as the EPA,
OSHA, and the FDA. This is in order to avoid the impression of
"approving" the risk communication or risk management meth-
ods of one over the other and to avoid rivalry between and among
such agencies.

The Clearinghouse could become a part of the Centers for Dis-
ease Control. The Centers in Atlanta, Georgia are a group of
nine operating components under the U.S. Public Health Ser-
vice,'0 an agency within the U.S. Department of Health and
Human Services.'0 4 Each of the CDC's nine components has a
distinct assignment related to the CDC's overall mission: preven-
tion of unnecessary illness and death, and enhancement of the
health of the American people.0 5 The nine components include,
for example, the National Institute for Occupational Safety and
Health (NIOSH), the Center for Environmental Health, the Epi-
demiology Program, the National Center for Health Statistics and
the Center for Health Promotion and Education.06 The names
of the divisions listed here indicate the substantial involvement of
the Centers in the risk assessment process and in public health
education. NIOSH and The Center for Environmental Health
employ toxicologists who conduct risk assessments for OSHA and
the EPA. NIOSH maintains a clearinghouse for receiving, stor-
ing, and disseminating technical information on occupational
safety and health.'07

One possible obstacle to public acceptance of the CDC as a
source of this information is the fact that CDC personnel conduct
toxicity and exposure assessments, both of which are processes
employing value-laden assumptions. However, the public policy
choices used to supplement risk assessment data in risk manage-
ment decisions will continue to be made in proceedings before
the EPA, OSHA, and similar agencies. To minimize potential
problems of credibility, the Risk Assessment Clearinghouse
should not be a part of one of the CDC's divisions which engages
in risk assessment.

103. See generally CENTERS FOR DISEASE CONTROL, ORGANIZATION, MISSION AND FUNC-
TIONS (1987) (available from Centers for Disease Control, Atlanta, Georgia, 30333).

104. Id. at 1.
105. Id.
106. Id.
107. Id. at 19.

572



National Risk Assessment Clearinghouse

The Centers for Disease Control have earned a high level of
respect in the public's eyes by responding to communities' needs
and requests rather than imposing itself on them.08 The Cen-
ters' Atlanta, Georgia location, being removed from the political
atmosphere of Washington, D.C., also enables them to maintain a
high reputation and credibility.'0 9 Such physical removal and au-
tonomy would also be an asset for the Risk Assessment Clearing-
house, since its effectiveness will be contingent upon its ability to
develop and maintain an impartial, non-political image.

1. Collecting Risk Assessments for the Clearinghouse

Congress should establish conditions under which any federal
agency or any state agency compiling a risk assessment using fed-
eral funding would be required to submit that risk assessment to
the Clearinghouse. Risk assessments prepared by state agencies
without federal funding could be submitted voluntarily by the
state or its agency. I propose that Congress require that a risk
assessment be submitted to the Clearinghouse when two criteria
are met. First, the acting agency conducts a risk assessment or
obtains an assessment from an outside source pursuant to re-
quirements of a state or federal statute or regulation. Second, the
agency relies on that risk assessment either to justify promulga-
tion of a rule or regulation or uses it in an enforcement proceed-
ing (such as denial of a permit or in an enforcement proceeding
against an alleged violator of an environmental law)."l0

The second criterion recognizes that agencies should reject risk
assessments which are inaccurate or unreliable because of defec-
tive testing procedures, inconclusive data, or other similar rea-
sons."' Therefore, to insure a standard of consistency and
reliability with respect to information in the Clearinghouse, risk
assessments which do not meet these two criteria, even if volunta-
rily submitted by a state or federal agency, should not be ac-
cepted. Inclusion of unreliable risk assessments would
complicate an already difficult communication task." l 2

108. Stacey, CDC Centers its Eforts on Controlling Disease, 26 AM. MED. NEWS 3, 47 (1983).
109. Id.
I10. These conditions are identical to those proposed in The Need for a National Risk

Assessment Communication Policy, supra note 10, at 396-97.

111. Id. at 397.
112. This assumes that "an administrative agency is acting in good faith when it chooses

to reject or ignore a given risk assessment. If an agency were to ignore a risk assessment
for invalid reasons, such as political motivation, the consequences of such a breach of

5731989]



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW (Vol. 14:549

2. Information to be conveyed to the public

The Clearinghouse would make two versions of each risk as-
sessment available to the public: a complete, unabridged version,
and a summary, plain-language version. Complete unabridged
copies would be available in response to a citizen's request. Rea-
sonable costs for copying the complete risk assessment should be
charged as is done with respect to information requested under
the Freedom of Information Act." 3

The plain-language abbreviated version of each risk assessment
would be supplied to the public free of charge at government ex-
pense. This version must be compiled and conveyed using a uni-
form format. A consistent format is essential, because without it
the public cannot build an "information base" for dealing with
other risks and for prioritizing among risks when risk manage-
ment choices must be made.

Because risk assessments provide only estimates of risk and
deal with very small risks which are difficult to put into perspec-
tive (for example, one out of hundreds of thousands or millions),
the numbers have little value unless the assessment of one risk
can be compared to that of another. As one researcher says,
"The best way to understand the magnitude of a risk is to com-
pare it with other risks ... ."1 4  Thus, a plain-language version of
a risk assessment should compare the risk being studied to other
similar risks with which laypeople are familiar in order to provide
a basis for evaluation."5  The comparison to other familiar risks
should be added to the plain-language version, even if it does not
appear in the original risk assessment.

Parameters for establishing a uniform format to be used in
communicating risk assessments are outlined in The Need for a Na-
tional Risk Assessment Communication Policy. 116 Those same parame-

public confidence would reach beyond the scope of RACPA [proposed Risk Assessment
Communication Policy Act, a statute designed to promote risk communication]. A remedy
for such a breach belongs in a comprehensive statute covering all uses of risk assess-
ments." Id.

113. The Freedom of Information Act provides that an agency providing information
under that act may charge reasonable costs for document search and duplication. Free-
dom of Information Act, 5 U.S.C. § 552(a)(4)(A) (1982). The costs for a document search
using the Clearinghouse, however, would be minimal and should be not be charged to the
person requesting the information.

114. Cohen, Putting 'Risk' in Perspective, [Mar..Apr. 1987] FusioN 32, 38.
115. E. CROUCH & R. WILSON, RISK/BENEFIT ANALYSIS 165 (1982).
116. It is important to note that the format is not presented in a complete. usable form

in the Risk Communication article. Moreover, it should not be finalized by any one person or

574



National Risk Assessment Clearinghouse

ters should be applied in the format for a plain-language version
of a risk assessment. The criteria for the uniform format fall into
three categories: (1) quantitative data; (2) comparison to other
risks; and (3) assumptions used.

First, the quantitative description of the risk must convey the
uncertain nature of the numbers presented. Descriptions should
not be based on multiples of risk because such a description does
not provide a frame of reference. Risk can be described in nu-
merical terms, provided that the numbers are given as a range
(for example, "a risk of one to twenty in a million"), not as a
"fixed point" (for example, "two in a million")." 7 Conveying
risks as a range acknowledges the uncertainty inherent in the
numbers and is a step toward dispelling the illusion of scientific
certainty created when risk managers talk about measuring very
small risks such as one in a million.

Second, a risk should only be presented with "comparators,"
that is, in comparison to other risks. But those comparators must
be chosen from among risks with similar characteristics. By doing
so risk communicators will deal with the fact that laypeople and
scientists do not agree on the criteria to be used to define risks
and rank risks according to their "magnitude." While scientists
generally rank risks according to the number of fatalities they
have caused," 18 lay people use a range of criteria."t9 Laypeople

regulatory agency. Congress should establish a non-partisan board charged with setting
guidelines for describing risks. The board would include laypeople, experts in risk per-
ception and risk communication, scientists from outside administrative agencies, and rep-

resentatives of federal agencies. In order to prevent representatives of agencies from
dominating the board's deliberations, they should be limited to only a small percentage of
the board's members. The diverse backgrounds of panel members would promote a con-
sideration of value-related and practical concerns from various viewpoints in addition to
use of the scientific "data" available. It is crucial to avoid exclusive reliance on the risk
assessment numbers, because to do so ignores their uncertain character and the value-

laden choices made in compiling them. Further, without citizen involvement, the format
devised by the board is likely to be viewed skeptically by members of the public. Stenzel,
supra note 10, at 398-400.

117. This concept of presenting the numbers as a range rather than as a fixed point was

not described in The Need for a National Risk Assessment Communication Policy, supra note 10.
118. Slovic, Fischhoff & Lichtenstein, supra note 45, at 23-24.
119. One author lists nine such criteria to illustrate this point. They include: (1) origin

(natural or synthetic), (2) volition, (3) effect manifestation (immediate or delayed), (4)
number of people affected per incident ("ordinary" or catastrophic), (5) controllability, (6)
benefit (clear or unclear), (7) familiarity (or lack thereof), (8) exposure (continuous or
occasional), and (9) necessity (as compared to luxury). M. KAMRIN, supra note 3, at 67. See
also, Hohenemser, Kates & Slovic, The Nature of Technological Hazard, 220 SCIENCE 371, (au-
thors classify 93 technological hazards according to twelve different hazard descriptors).

1989] 575



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

deal with risks in differing ways depending on the characteristics
of the specific risk. Such characteristics include, for example, vo-
lition. Many laypeople knowingly accept risks accompanying vol-
untary activities such as cigarette smoking or drinking alcoholic
beverages while they refuse to accept similar or even lower levels
of risk if the risk is imposed on them involuntarily (for example,
through occupational exposure to chemicals). Similarly, clarity of
benefit, real or believed ability to control a risk, and "ordinary"
nature of a risk may cause laypeople to be willing to accept certain
risks while rejecting others which may be of a comparable "mag-
nitude." For example, at least prior to the 1986 nuclear power
plant accident at Chernobyl in the Soviet Union,' 20 scientists
viewed the use of nuclear power as relatively safe because few
people had died as a result of nuclear power plant accidents.
Nevertheless, laypeople viewed (and still do view) the use of nu-
clear power as being extremely risky because of its catastrophic
potential and threat to future generations.121 Meanwhile, they ac-
cept other risks which cause far greater numbers of fatalities, such
as driving automobiles. Therefore, hazards and the risks that
they pose must be compared to others hazards which share simi-
lar characteristics. For example, years of exposure to a chemical
in a factory must be compared to other chronic hazards;'22 risks
of exposure to a major accident at a nuclear power plant must be
compared to other involuntary, catastrophic risks; and eating
chemical contaminated foods should be compared to other volun-
tarily encountered hazards.123

To establish hazard categories from which comparators can be
selected, it is crucial that the advisory board include people with
varying backgrounds and perspectives. A cross-section of view-
points will be necessary for decisions on two levels. First, criteria
defining hazard categories must be chosen. These would include,

120. NEWSWEEK, May 12, 1986, at 20; TIME, May 12, 1986, at 38; U.S. NEWS & WORLD

REPORT, May 12, 1986, at 18.
121. For discussion of risk perceptions in the aftermath of Chernobyl, see Hohenemser

& Renn, Chermobyl's Other Legacy, 30 ENV'T 4 (1988).
122. With respect to workplace hazards, it is important to consider whether occupa-

tional risks are voluntarily or involuntarily encountered. Comparators should be selected
from the same category of volition. See K.S. SHRADER-FRECHETTE, RISK ANALYSIS AND ScI-
ENTIFIC METHOD: METHODOLOGICAL AND ETHICAL PROBLEMS WITH EVALUATING SOCIETAL

HAZARDS 97-122 (1985) (discusses our society's mistaken assumption that most workers
accept workplace risk "voluntarily").

123. See Stenzel, supra note 10, at 399-400 (discusses hazard categories).

576



National Risk Assessment Clearinghouse

for example, origin (natural or synthetic), volition, controllability,
necessity, and other factors relevant to defining risk.124 Second,
in some cases, the board will have to decide how to classify certain
types of risk according to the criteria chosen for describing risks.
Laypeople, agency personnel, and scientists do not necessarily
agree, for example, on what is or is not voluntary exposure (such
as in the case of an asbestos worker's choice of employment),125

the existence and magnitude of benefits, and other similar value-
laden distinctions.12 6 Thus, a panel that did not include laype-
ople, risk perception experts, and people concerned with the ethi-
cal ramifications of risk management decisions would not be
qualified to establish the hazard categories.

Third, the plain-language version of the risk assessment must
clearly and concisely convey the assumptions used in compiling
the risk assessment. This includes assumptions made in the toxic-
ity assessment and the exposure assessment. For example, when a
carcinogenesis risk assessment compiled by the EPA is conveyed,
the public must be told that its calculations are based on an "up-
per bound" of risk, not on an "average." 27 Similarly, assump-
tions made in assessing exposure must be conveyed. Thus, to
interpret an assessment of risks to those living near a trash-to-
electricity incinerator, the public needs to know what assumptions
were made for the hypothetically exposed individual, such as dis-
tance of his or her residence from the incinerator, number of
hours of daily outdoor (separate from indoor) exposure, number
of years of exposure, and similar factors. Thus, an individual
whose exposure is less than or greater than that used in the risk

124. See supra note 10, (lists criteria used by laypeople to define risk).
125. Our society maintains a "double standard" under which we tolerate greater risks

to workers as compared to risks to which we are willing to expose the general population.

For discussion of that double standard and of our society's mistaken assumption that
workplace risks are "voluntarily" accepted by workers, see SHRADER-FRECHETTE, supra note
122, at 97-122.

126. These concerns of laypeople were expressed repeatedly at a workshop which I con-
ducted on risk communication and risk assessment. The workshop was attended by envi-
ronmentally concerned citizens from a variety of backgrounds. 1988 Backyard ECO
Conference, Lake Michigan, May 21, 1988.

127. "Upper bound" describes the EPA's practice of using responses of the most sensi-
tive species as a basis for carcinogenesis risk assessment rather than, for example, a species
whose reaction is predicted to be most like that of a human being. Guidelines for Carcino-
gen Risk Assessment, 51 Fed. Reg. 33,994-98 (1986).

5771989]



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

assessment can begin to put the information into an individual-
ized perspective.128

3. Requests for Information

The plain-language version and the complete risk assessment
would be available upon written or telephoned request. A written
request could be in any reasonable form, without formal require-
ments, as is permitted with Freedom of Information Act re-
quests.129 In addition, a telephone "hot-line" should be
established to receive requests. Either an "800" number, with no
charge to the caller, or a "900" number, which would reduce
costs to the federal government by imposing a minimal charge to
the caller, could be used.

Responses to written requests would be by mail. Responses to
telephone requests could be immediate, by phone, and by mail.
Operators receiving calls would offer to read information con-
tained in the plain-language risk assessment to callers who want
an immediate response, but a copy of the plain-language assess-
ment would be mailed to confirm and complete that information.
Due to their length, the unabridged risk assessments would be
conveyed only in writing.

B. Precedent

1. Precedent for Plain-Language Document

Providing a plain-language version of the risk assessment is
consistent with a growing trend in our legal system. For example,
lawyers are increasingly conscious of the desirability of the use of
plain-language in legal documents such as wills and contracts.
Use of plain-language in leases has been emphasized and has

128. Ideally, standard sets of assumptions should be used for describing types of risks.
Thus, the public could more easily compare the assessment of one risk to that of another.
For example, a standard set of assumptions should be developed and used consistently for
each of the following categories: "(1) occupational exposure to a chemical through either
inhalation or skin contact; (2) exposure due to living near a site emitting chemical air
pollutants; or (3) exposure through ingestion of chemicals found in a drinking water
source." Stenzel, supra note 10, at 401. However, in the absence of a uniform "Risk Com-
munication Policy," at a minimum, the assumptions used should be conveyed as a part of
each risk assessment.

129. The Freedom of Information Act requires that an agency 'upon any request for
records which (A) reasonably describes such records and (B) is made in accordance with
published rules stating the time, place, fees (if any), and procedures to be followed, shall
make the records promptly available to any person." 5 U.S.C. § 552(a)(3) (1982).

578



National Risk Assessment Clearinghouse

played an important role in landlord tenant relations in recent
years. Under federal law, employers who provide pension plans
(such as defined benefit, defined contribution, or profit-sharing)
to their employees must provide each employee with a "Plan
Summary" which summarizes the employee's rights and obliga-
tions under the plan in plain-language.o30

In environmental law, the EPA has taken steps consistent with
the proposal in this article. Beginning in 1983, the EPA con-
ducted a series of environmental studies in the Midland, Michigan
area for dioxins and other pollutants. In April, 1988, the EPA
released a fact sheet summarizing the results of their risk assess-
ment and proposed risk management actions.'3 ' The EPA's ef-
forts are evidence that the agency recognizes the need to convey
risk assessments to the public in a plain-language form. However,
the proposal in this article goes much further by including all risk
assessments used by federal agencies and many of those used by
state agencies, setting a standardized format, and establishing a
central location through which the assessments could be easily
and quickly obtained.

2. Precedent for Clearinghouse

There is also precedent for use of a clearinghouse. The Birth
Defects Program within the Center for Environmental Health,
one of the nine components of the Centers for Disease Control,
has conducted risk assessments to study the effects of drugs taken
by pregnant and nursing mothers.'32 In various texts used by
health professionals, summaries of effects of individual drugs are
published. One such text, for example, includes a "Fetal Risk
Summary" and a "Breast Feeding Summary," each of which de-
scribes the format (number of participants, exposure, length of
study, etc.) and results of studies which have been conducted on
the chemical.'33 A list citing the studies summarized is also in-

130. Employee Retirement Income Security Act of 1974, 29 U.S.C. §§ 1001-1461
(1982); Reporting and Disclosure Under the Employee Retirement Income Security Act of
1974, 29 C.F.R. §§ 2520.102-1 to 2520.102-5 (1987).

131. Summary of Risk Assessment and Proposed Risk Management Actions-Midland,
Michigan (fact sheet dated April, 1988, available from U.S. EPA, Office of Public Affairs,
Region 5, 230 S. Dearborn. Chicago, IL. 60604).

132. CENTERS FOR DISEASE CONTROL, ORGANIZATION, MISSION, AND FUNCTIONS (1987)

(available from Centers for Disease Control, Atlanta, Georgia, 30333).
133. G. BRIGGS, DRUGS IN PREGNANCY AND LACTATION (1986). For a similar resource, see

also T. H. SHEPARD, CATALOG OF TETRATOGENIc AGENTS (1986).

5791989]



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

cluded.'3 4 To supplement such publications, personnel at the
CDC respond to telephone calls from health professionals (but
not laypeople) who inquire about recent studies and data which
have not yet been incorporated in the published summaries.'3 5

Thus, the CDC may be a particularly appropriate location for a
Risk Assessment Clearinghouse, since health professionals are al-
ready acquainted with their role as an information source. It is
logical to assume that health professionals (including private phy-
sicians and state and local public health officials) may serve as "in-
termediaries." Such intermediaries may direct members of the
public to the Risk Assessment Clearinghouse, or they may actu-
ally contact the Clearinghouse to obtain information on behalf of
laypeople.

Congress is considering legislation which would establish a
clearinghouse for information about one specific area of risk: in-
door air pollution. In August, 1987, Senator Mitchell introduced
the "Indoor Air Quality Act of 1987."136 Among its other provi-
sions, that bill would establish "a national indoor air quality
clearinghouse to be used to disseminate indoor air quality infor-
mation to other federal agencies, state and local governments,
and private organizations and individuals." The indoor air qual-
ity clearinghouse would be a repository for information collected
from government agencies, private organizations, and individu-
als. The proposed legislation, however, does not provide details
for the design, development and implementation of the clearing-
house.'3 7 If that bill is passed, the format and detail outlined in
this article should be implemented with respect to risk assess-
ments included in the indoor air quality clearinghouse. Further,
if a Risk Assessment Clearinghouse is established after an indoor
air quality clearinghouse has been set up, it would be appropriate
to transfer risk assessments for indoor air pollution to the Risk
Assessment Clearinghouse. Indoor air quality information not
meeting the criteria for inclusion in the Risk Assessment

134. G. BRIGGS, supra note 133.
135. For general information, contact the Centers for Disease Control.
136. S. 1629, 100th Cong., Ist. Sess., 133 CONG. REC. 11,655-61 (1987).
137. The design, development, and implementation of the indoor air quality clearing-

house are assigned to the General Services Administrator who is authorized, in turn, to
provide for them through a contractual agreement with a nonprofit organization. Indoor
Air Quality Act of 1987, S. 1629, 100th Cong., Ist. Sess., 133 CONG. REc. 11,655,
§ 12(a)(3) at 11,660.

580



National Risk Assessment Clearinghouse

Clearinghouse would continue to be provided through separate
channels.

To deal with risk assessments, in 1982 the EPA established a
computer data base to assist its personnel in implementing the
Comprehensive Environmental Response, Compensation, and Li-
ability Act (CERCLA).'"3 (CERCLA is also known as the
"Superfund" legislation.) The data base, called the STARA Tox-
icity Base, includes data from animal toxicity tests for about two
hundred chemicals and epidemiological data for about thirty
chemicals.'39 STARA includes quantitative data as well as de-
scriptive information regarding exposure and types of effects.140

An EPA publication explains why the STARA data base was
developed:

Requests for situation-specific assessments or other technical
assistance occur irregularly and often involve repetitive re-
trieval of toxicity information on a variety of chemicals. The
traditional procedure has been to manually extract and com-
pile the desired data from various "hard copy" sources (re-
search articles, review documents) on a case-by-case basis as
the need arose. This approach was deemed outdated and inap-
propriate on the basis of economy, efficiency and even accu-
racy. The logical solution was to compile this bulk of
information into some form of computer accessible data
base.141

The reasons given by the EPA for its own data base apply with
equal force to illustrate the need for a Risk Assessment Clearing-
house to serve the general public.

C. The Benefits of a Risk Assessment Clearinghouse

Regulatory agencies have a duty to share risk assessments with
the public. To withhold. the information overtly or by maintain-
ing barriers to its acquisition is elitist and undemocratic.14 2 As
Thomas Jefferson said, "[I]f we think [the people are] not enlight-
ened enough to exercise their control with a wholesome discre-

138. 42 U.S.C. §§ 9601-9675 (1982 & Supp. IV 1986).
139. Farren & Hertzberg, THE SARA Toxicrry DATA BASE 1 (1986) (EPA Research

Brief EPA/600/M-86/016, available from the EPA Center for Environmental Research In-
formation, Cincinnati, OH 45268).

140. Id.
141. Id. at 1.
142. See generally Bazelon, supra note 13 (argues that in the interests of human dignity as

well as to preserve democracy, full disclosure about risks, value choices, and risk manage-
ment decisions is essential in our society).

5811989)1



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

tion, the remedy is not to take it from them, but to inform their
discretion." 

4 3

Establishment of a Risk Assessment Clearinghouse is a logical,
necessary "next step" following implementation of federal
worker and community right-to-know legislation and state legisla-
tion such as California's Proposition 65.'4 Citizens and in-
termediaries who serve the public will gain easier, faster access to
existing risk assessments.

Access will be faster and easier for a variety of reasons. Unlike
the present situation, citizens will not need to do research as to
the jurisdiction of various agencies before requesting informa-
tion. Response will be almost immediate, with preliminary infor-
mation given by telephone followed within a few days with more
detailed information by mail. Thus, a citizen will not have to deal
with individual agencies' public information offices, the Federal
Register, or Freedom of Information Act requests.

Faster access to risk assessments will enable citizens to more
quickly identify potential toxic effects of exposure to a chemical
and to begin to evaluate the potential severity of those toxic ef-
fects. Suspicion that an agency is being evasive or general frustra-
tion with lack of response by an agency can be avoided or at least
reduced.

Similarly, keeping in mind that the news media are the public's
primary sources of environmental information, news reporters
will be able to gather greater amounts of information more
quickly. With nearly immediate access to risk assessments, news
reporters can obtain some background before interviewing scien-
tists, regulators, and citizens, and they can include the risk infor-
mation in their stories.14 5

Further, the plain-language version of risk assessments will
provide news reporters with a uniform format for presenting risk

143. Letter from Thomas Jefferson to William Charles Jarvis (Sept. 28, 1820), reprinted
in 7 WRITINGS OF THOMASJEFFERSON 177, 179 (H. Washington ed. 1855), quoted in NRDC

v. NRC, 547 F.2d 633 (D.C. Cir. 1976).
144. See supra text accompanying notes 84-94 (discusses worker and community right-

to-know legislation and California's Proposition 65).
145. There are sources designed to assist reporters in identifying and contacting scien-

tists. For example, the Scientists' Institute for Public Information operates a "Media Re-

source Service." It is a referral service for journalists who are working on stories

involving science and technology. A journalist working on a specific story can call the
service. He or she will be given the names and phone numbers of appropriate experts
along with information about the experts' backgrounds and credentials. McGowan, Panel
on Future Challenges for Risk Communicators, in RISK COMMUNICATION 88 (1987).

582



National Risk Assessment Clearinghouse

assessments in their stories. That will minimize instances in
which reporters appear to be comparing "apples and oranges" by
describing risks using differing measures.

These same arguments for providing greater access to informa-
tion and providing it in a uniform format also apply with respect
to other parties seeking such information. For example, various
environmental groups and concerned citizens seek risk assess-
ments for their own efforts in lobbying agencies and the legisla-
tures. They also act as "watch-dogs" of public health and may, in
turn, convey this information to the news media or through their
own publications.

The Clearinghouse will reveal areas where there is little or no
information available, at least within regulatory agencies. (Admit-
tedly, other risk assessments may exist for many chemicals.) Cur-
rently, a citizen who is unable to locate any information about a
given chemical may fear that he or she has looked in the wrong
place or that an agency is withholding information. Conveying
uncertainty and insufficient data are integral parts of risk
communication.

Similarly, use of the Clearinghouse will enable citizens to begin
to evaluate the strength of evidence available with respect to a
specific chemical. It is likely that the Risk Assessment Clearing-
house will include several risk assessments for a specific chemical.
If the Clearinghouse reports that three agencies have assessed the
same chemical in the past three years, each with similar results,
those results would be more trustworthy than two tests run by
one agency fifteen years ago. Further, if several risk assessments
are available but they reach divergent "conclusions," the uncer-
tain nature of the data on that partikular chemical will be
exposed.

The primary purpose of the Clearinghouse is to assist the gen-
eral public. Federal administrative agencies have the expertise
and personnel available to search for and find the information
they need with respect to risk assessments. However, under the
"New Federalism," the EPA and other regulatory agencies are
turning over many risk management tasks, including responsibil-
ity for conducting risk assessments, to the individual fifty states.
The Clearinghouse might provide preliminary information to as-
sist a state in deciding how to spend its limited resources. If fed-
eral regulatory agencies have extensively assessed the risks of a
chemical and the results of those assessments are consistent with

1989J 583



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

each other, a state may opt to spend its resources to assess the
risks of chemical substances which have not been studied as
extensively.

A similar rationale applies to the utility of information in the
Clearinghouse with respect to business organizations. Large cor-
porations possess the personnel and resources to gain access to
the risk assessments prepared by agencies even if the process is
time-consuming and cumbersome. Further, they regularly par-
ticipate in regulatory rulemaking proceedings in which risk as-
sessments are employed. The Clearinghouse information could
be of assistance to small businesses which are concerned about
complying with worker and community right-to-know laws but do
not have the resources of larger companies. As the small business
responds to inquiries from its workers and the community, it can
use information from the Clearinghouse to supplement the infor-
mation supplied by chemical manufacturers in Material Safety
Data Sheets (MSDS's).

Beyond providing information needed to make specific risk
management decisions, the Risk. Assessment Clearinghouse will
educate the public about risks in general. In turn, citizens will be
able to participate in the risk management process in a more
meaningful way.

By providing ready access to risk assessments and by present-
ing the assessments according to a consistent format, agencies
will improve their public image. A free flow of information is nec-
essary if agencies are to avoid the appearance of being evasive.
Studies by risk perception experts confirm that people must trust
the disseminator of information before they will accept that infor-
mation.' 4 6 As the public begins to trust administrative agencies,
citizens will be more willing to participate in regulatory processes.
An educated public, in turn, can make more substantial and more
meaningful contributions to the formulation of public policy and
to risk management decisions.

As the public feels able to and does participate in risk manage-
ment decisions, the public's antagonism toward administrative
agencies will be lessened. Public respect for and acceptance of
regulatory decisions can be expected to increase.

146. "Successful communication demands credibility." Fischhoff, supra note 20, at 96.
See also, Slovic, Informing and Educating the Public about Risks, 6 RISK ANALYSIS 403, 414
(1986).

584



National Risk Assessment Clearinghouse

D. Anticipated Criticisms of the Risk Assessment Clearinghouse

This proposal for a Risk Assessment Clearinghouse may en-
counter opposition from various groups including agencies,
chemical producers and users, employers, and taxpayers in gen-
eral. Critics may argue that we can't afford it and that it's an un-
needed additional level of bureaucracy. It may also be argued
that information which is inexact and subject to further review
should not be distributed. Those arguments are not well-
founded.

The Risk Assessment Clearinghouse would be a relatively inex-
pensive program to set up and administer. Costs for services of
an advisory board which would establish the format for plain-lan-
guage versions of the risk assessment would be minimal. Even
though the board might be needed to periodically revise guide-
lines based on new knowledge, its members need not become
full-time permanent employees of the federal government. The
Clearinghouse would not require that new risk assessments be
compiled. Clearinghouse personnel would simply gather existing
information from regulatory agencies and put that information
into a computer data base.'4 7 The work of Clearinghouse person-
nel would be assisted by the fact that administrative agencies
would have an affirmative duty to convey certain 148 risk assess-
ments to the Clearinghouse. Beyond the day-to-day administra-
tion, the remaining major task to be accomplished would be
preparation of the plain-language documents drawing from infor-
mation in the risk assessments. Costs of running the Clearing-
house would be minimized through use of computers to store and
retrieve copies and by charging the public reasonable copying
fees, except for the short, plain-language documents.

The costs incurred would be easily justified. The public has a
right to information generated through public tax dollars. The
information should not simply sit on the shelves of our regulatory
agencies. The fact that our country has traditionally spent only a
small percentage of its risk management dollars on public educa-
tion and information programs does not mean that to do so has
been a wise course of action. The National Environmental Protec-

147. It is useful to note that the EPA estimates that the cost of compiling a "toxicity"
table for each of the chemicals in its STARA data base is about $500 to $1,000. Farren &
Hertzberg, supra note 139, at 3.

148. See supra, text accompanying notes 110-112 (describes criteria identifying risk as-
sessments which agencies would be required to convey to the Clearinghouse).

1989]1 585



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

tion Board of Sweden, for example, spends a significant portion
of its budget disseminating information about its work.'49 Risk
communication may be the most important task before us in risk
management today as we attempt to set priorities in dealing with
a multitude of risks with limited financial resources. Therefore, it
is time for the United States also to begin to devote substantial
efforts and financial resources to risk education and information-
sharing.

Further, the costs of the Risk Assessment Clearinghouse may
be recovered elsewhere. Individuals and intermediaries acting on
behalf of the public will spend less effort and fewer resources in
efforts to obtain the information in risk assessments. Further, it is
less likely that they will duplicate the research efforts of other in-
dividuals and groups who seek the same information.

Other objections may come from critics who say that we should
not distribute information which is inexact and will be revised re-
peatedly as a result of future studies. These objections are not
compelling. Regulators and courts reviewing their actions are
making decisions based on risk assessments now, not postponing
decisions for decades waiting for more information. If risk assess-
ments are being used to make decisions affecting citizens' health
and lives, then it is elitist and intolerable to suggest that the infor-
mation is not reliable enough to be conveyed to the public. Lack
of knowledge and scientific uncertainty are important messages to
be conveyed. Further, a computer data bank can easily accommo-
date additional or revised risk assessments.

The messages of uncertainty which will be conveyed through
the Clearinghouse also provide a response to a third potential
criticism: those who are critical of agencies' reliance on risk as-
sessments may argue that the Clearinghouse will serve to legiti-
matize risk assessments and further entrench our reliance on
them.o50 This is not necessarily-true. As discussed supra,1'5 risk
assessments are already thoroughly entrenched in our risk man-
agement process. Through the Clearinghouse, regulatory agen-

149. J. DIMENTO, ENVIRONMENTAL LAW AND AMERICAN BUSINESS: DILEMMAS OF COMPLI-

ANCE 176 (1976).

150. This argument was raised at a workshop which I conducted on risk communication
and risk assessment. 1988 Backyard ECO Conference. Lake Michigan, May 21, 1988.

151. See generally text accompanying notes 1-3 (discussing administrative agencies' exten-
sive reliance on quantitative risk assessment data as they make risk management
decisions).

586



National Risk Assessment Clearinghouse

cies will take a step toward acknowledging openly that they don't
have all the answers the public wants and that the risk assessment
process, although useful, is a limited tool.15 2

As the public becomes more familiar with risk assessments and
the assumptions and value-laden decisions involved, there may be
several consequences. Perhaps public exposure of the nature of
the quantitative risk assessment process will result in more atten-
tion to the procedures used in the process.and, ultimately, risk
assessors will be provided with resources and compelled to re-
verse these procedures. More certainly, the public's awareness of
other important factors which are or should be involved in risk
management decisions will increase.

Value-related choices are only implicit in the risk management
decisionmaking process and are rarely, if ever, expressly dis-
cussed and analyzed. We must examine and consider who re-
ceives the benefits of risk management decisions (corporate
managers? shareholders? workers? customers?) and consider
whether they or other parties should receive those benefits. We
must analyze our assumptions about volition in the workplace.
(Are workers truly choosing to expose themselves to occupational
risk in exchange for higher wages? Or, are the workers in mines,
asbestos factories, and other hazardous sites choosing to remain
employed in these areas out of necessity because they, have no
viable options?)'5 3 Further, regulatory agencies making risk man-
agement decisions generally do not consider available alterna-
tives. For example, the EPA requires a chemical manufacturer to
perform a toxicity assessment to support its application for licens-
ing a new pesticide under the Federal Insecticide, Fungicide, and
Rodenticide Act (FIFRA).' 5 4 The EPA does not, however, re-
quire the manufacturer to consider options such as use of a less
hazardous chemical or use of organic farming methods through

152. Professor Frank P. Grad of the Columbia Law School argues that although we will
not (and should not) abandon the use of risk assessment in public decisionmaking, we
should not let ourselves be overwhelmed by the false impression of certainty conveyed by
numbers. He argues that it is a technique which is subject to challenge and varies in use-
fulness, depending on the context in which it is used. Grad, Risk Assessment and the Tyranny
of Numbers: A Brief Comment, I ENVL. L. & LEGIS. I (1986).

153. See K. SHRADER-FRECHETrE, supra note 122, at 107 (argues that many people en-
gage in a certain work only because they have no other alternatives).

154. 7 U.S.C. §§ 135-136y (1982 & Supp. IV 1986).

1989] 587



COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 14:549

which the use of risk-laden chemicals might be avoided.'5 5 The
EPA considers such alternatives as it determines whether or not
to grant the license allowing the use of pesticides. I am not sug-
gesting that the Risk Assessment Clearinghouse alone will cause
regulatory agencies or our society to reexamine and reformulate
the risk management process. But, increased citizen awareness
and knowledge can be the first steps toward serious efforts to in-
stitute reform.

E. Limitations of this Proposal

It is important to acknowledge the limitations of the Risk As-
sessment Clearinghouse. It would be only one of many steps to-
ward the preliminary goal of better, two-way risk communication
between administrative agencies and the public, and the ultimate
objective of fair and effective risk management.

First, it must be acknowledged that risk assessments have been
conducted on only a few hundred of the approximately 50,000
chemicals in common use in our society.'56 However the public
needs and is entitled to be informed about our lack of informa-
tion concerning the safety of these chemical substances.

It must also be acknowledged that, like the Freedom of Infor-
mation Act, the Risk Assessment Clearinghouse would be passive
in nature.'5 7 It would make information more readily available to
those who seek it, but it would not actively disseminate informa-
tion. Nevertheless, the Clearinghouse would be a substantial im-
provement over the use of the FOIA or less formal requests to
agencies, by providing a centralized source and faster access to
risk assessments, and by making plain-language versions
available.

155. I am not recommending that chemical manufacturers be required to investigate
such alternatives. In most cases there would be no economic incentive, the primary mo-
tivator of most corporations, to do so. Presentation of alternatives would most appropri-

ately come from the regulatory agency or from concerned citizens. However, it is
important to note that systematic, serious consideration of viable alternatives involving
less risk is conspicuously absent from most of our risk management practices.

156. M. KAMNRIN, supra note 3, at 6.

157. Professor James S. Coleman observed, "The Freedom of Information Act in the
U.S. established the right of interested parties to gain access to information to social policy
research, but it does not include procedures to aid that dissemination." J.S. Coleman,
Social Policy Research in a Theory of Society (April 17, 1986) (unpublished manuscript
available from the author at the Department of Sociology, The University of Chicago).

588



National Risk Assessment Clearinghouse

Our society's general need for risk education should be ad-
dressed through a variety of mechanisms in addition to the
Clearinghouse. Additional measures would include affirmative
risk education and outreach by administrative agencies.'58 They
would also include public education in our primary and secondary
schools and universities, and adult education through special pro-
grams such as conferences and seminars.

A further limitation of the Clearinghouse is that it would only
provide one-way communication. Ultimately, risk communication
between regulatory agencies and the public must be two-way.
The public must participate actively in the risk management pro-
cess, and regulatory agencies must listen to citizens and incorpo-
rate their needs and values into the risk management process.
The Risk Assessment Clearinghouse will facilitate public input in
the risk management process by enabling citizens to make more
informed contributions more quickly, but other means are
needed to ensure that agencies listen and respond to citizens'
needs and values. For example, it may be necessary for Congress
to redefine the concept of "risk management" to ensure that
agencies view "risk assessment" as only one limited tool among a
variety of tools to be used by agencies in making risk management
decisions. For example, additional "tools" could include consid-
eration of the values of persons to be affected by the risk manage-
ment decision.'59 Also, Congress might require agencies to
consider practical alternatives to use of the chemical, such as use
of another chemical or no chemicals. Thus, when considering an
application for a license for a new pesticide, the EPA would evalu-
ate alternatives such as other chemicals or use of organic farming
methods and deny the license if it concludes that alternative
means for eliminating the "pests" will create fewer risks to human
health and life.' 60

Laws and regulatory programs alone cannot create a well-in-
formed public, but they can be used to facilitate the process.

158. See Stenzel, supra note 10.

159. See generally M. SAGOFF, THE ECONOMY OF THE EARTH: PHILOSOPHY, LAW AND THE

ENVIRONMENT (1988) (examines and criticizes the economic methodologies, such as cost-
benefit analysis, on which our environmental laws are based, and argues that ethical con-
siderations should be included in such social regulation).

160. Further discussion of means for requiring administrative agencies to consider and
incorporate the public's values into risk management decisions is beyond the scope of this
article. However, I intend to continue such discussion in a future article.

1989] 589



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 14:549

Through the Risk Assessment Clearinghouse the federal govern-
ment can and should facilitate risk education and communication.

IV. CONCLUSION

In the United States, the public relies extensively on adminis-
trative agencies to manage the risks created by the use of chemi-
cals. Toxicologists conduct the risk assessments upon which
agencies base their risk management decisions. Yet, the risk as-
sessment process is riddled with. uncertainties. Further, risk as-
sessments have been conducted on only a small percentage of the
chemicals in common use in our society. But, acting through our
legislatures and courts, citizens in the United States have realized
that we cannot wait for more definite answers before we deal with
the hazards of toxic chemicals.

To deal with gaps in knowledge and uncertainties in the risk
assessment process, risk assessors make value-laden assumptions.
In turn, regulators making risk management decisions rely on
public policy to supplement the "data" produced in the risk as-
sessment process. Yet, whenever public policy is involved in reg-
ulatory decisions, it is crucial that the public participate in those
decisions in a meaningful way. Meaningful participation depends
on a well-informed public.

This article has illustrated that although risk assessments are
the primary tools used by administrative agencies in risk manage-
ment decisions, those assessments are not readily available to the
public. Those who actively seek risk assessments find the search
to be a time-consuming, confusing, and cumbersome process.

Therefore, I have proposed that Congress establish a Risk As-
sessment Clearinghouse. At a minimal cost to government, risk
assessments compiled and used by administrative agencies would
be made available to the public in a summary, plain-language
form and in their entirety. By making risk assessments (or ac-
knowledgment that they do not exist for certain chemicals) read-
ily available to the public, the federal government will provide a
logical and necessary extension to existing right-to-know laws and
to legislation designed to reveal the presence of chemical risks in
consumer products to the public (for example, California's Prop-
osition 65). The Risk Assessment Clearinghouse will facilitate
risk education in general and will enable the public to provide
more informed input to administrative agencies. The Clearing-
house is one of a variety of tools we need to promote effective

590



1989] National Risk Assessment Clearinghouse 591

two-way risk communication between administrative agencies and
the public. Effective means of risk communication, including the
Risk Assessment Clearinghouse, will be indispensable as we strive
toward our overall goal of making well-informed, fair and effec-
tive risk management decisions.