Representatives Klein, Zimmer and 14 co-sponsors introduced H.R. 4306 in April 1994. It would have established a Risk Assessment Program within EPA to develop a process for independent peer review of risk assessments, developing and issuing risk assessment guidelines designed to incorporate state-of-the-art science, and providing the opportunity for public input. It would have required the White House Office of Science and Technology Policy (OSTP) to establish a Comparative Risk Assessment pilot program to test risk-ranking methodology and to develop interagency coordinating mechanisms for communicating state-of-the-art practices to states. OST P was to evaluate risk assessment research and training needs and to prioritize research according to need and the largest uncertainties in risk assessment.

The original bill had the support of environmentalists, industry and the Administration.15 However, Rep. Walker added amendments that would have required the judicial review of contested risk assessments and that "objective" and "unbiased" science were to be used by EPA in its risk assessments. Walker's stated objective was to indicate that the guidelines were to be considered "binding rules" and not general statements of policy; risk assessments, therefore, were to be "conducted, applied and practiced throughout the agency in accordance with the guidelines." 16 Courts could then look at contested risk assessments for departure from the guidelines.

Brown claimed the amendments destroyed the coalition supporting the bill and altered its intent,17 and EPA Administrator Browner withdrew support in July 1994.18

Introduced by Rep. Moorehead in February 1994, the RCA contained principles for risk assessment and characterization. It sought to ensure that risk assessments are as scientifically objective and inclusive of all relevant data as possible. It would have required EPA's risk assessments to distinguish scientific findings from other considerations; to consider and discuss all available data; to describe the selection of any significant assumption, inference or model, to explain the basis for the choice and to identify any policy or value judgments.

EPA also would have had to characterize populations or resources at risk and provide the best estimate and the range of uncertainty; provide meaningful comparisons to other risks; set out a process for reviewing assessments in view of new information; and ensure public participation in formulating risk assessment and characterization guidelines.

Senator Moynihan introduced this bill to require EPA to set risk-based priorities. It would have established two committees within the EPA: one to rank risks to human health and the environment; another to determine the benefits of eliminating such risks. Its stated objective for forming these committees was to make science, not politics, the basis for selecting health and environmental priorities. The bill would have required the EPA to create risk assessment guidelines that would set minimum standards for different risk assessment approaches.

Senator Lott would have required EPA to develop uniform general procedures governing risk assessments and incorporate relevant guidelines. His bill would have required assessments to be "consistent," of "high technical quality," "scientifically sound" and "unbiased." Moreover, EPA must "disclose all significant uncertainties regarding facts, scientific knowledge, the validity of analytical techniques, or numerical risk estimates" in terms "readily understandable" to the public. In a cover letter, the Senator urged his colleagues to support it because "(r)isk assessments must be accountable to the public and Congress; not just subject to the discretion of invisible bureaucrats."

To S. 171, Senator Johnston attached an amendment requiring each final rule to contain what became known as "Johnston-type" analysis. The amendments would have required regulations to contain an estimate of human or ecological health addressed by the rule and the rule's anticipated impact on the target risk; a comparative analysis of the risk addressed by the rule compared to other risks to which the public is exposed; and an estimate of the costs associated with implementation of and compliance with the rule.

Industry lauded Johnston's amendment,19 since it required EPA to estimate risk and conduct comparative risk and cost-benefit analyses for every regulation. EPA also would have to certify that a proposed regulation would significantly advance public protection. The original bill failed to leave committee, partly because of the risk language.

Amendments to SDWA reauthorization carried Johnston-type analyses but were only to apply to major regulations (exceeding $100M), and would have required only that benefits justify, rather than exceed, cost (as required in the original). This bill came close to passing.

A bipartisan group in committee drafted an alternative to the reauthorization of the CWA, but their version contained risk assessment provisions. For all CWA regulations, EPA would have been required to perform risk assessments. A significant departure from the original Johnston amendment was a requirement that contested risk assessments undergo judicial review. The amendment kept this bill in committee.

Representative Zimmer amended the EPA cabinet bill to create an Office of Environmental Risk (OER). It was to provide an organizational structure within EPA where the director would develop and implement a strategy for risk reduction. The amendment also created an Advisory Committee on Relative Risks to advise the director. Like the Moynihan bill (S. 110), EPA would set its priorities based on relative risk ranking.

The Environmental Technologies Act directs the OSTP to set federal environmental research and development priorities by using scientifically-objective information, data, and assessments of risk. The Walker amendment established guidelines on the necessary elements for a sound risk assessment and the methods for communicating those assessments. It added fifteen criteria for conducting risk assessments and language on risk comparisons and substitution risk, judicial review, and public notice and comment. The House approved the amendment, assuring that other environmental legislation, particularly the CWA reauthorization and Superfund reform, would not be reauthorized without comparable risk assessment provisions.20

Provisions in this bill required that, for any hazard at a Superfund site, the EPA would include central estimates of risk, as well as upper and lower bound estimates.

Representative Waxman introduced this bill calling for EPA's assessments of radon dangers to include estimates of population risk and the reasonable range of uncertainty about such estimates. It passed the House, but its companion (S. 657) stalled in the Senate.

Representative Waxman and Senator Kennedy introduced into their respective chambers the Pesticide Food Safety Act. Each established a single "bright line" standard of one cancer in a million for pesticide residues in raw and processed foods, if the pesticide is found to cause cancer in animals or humans.

Representatives Lehman, Bliley and Rowland introduced competing legislation. In contrast, the Food Quality Protection Act would have established a single negligible risk standard for pesticide residues in both raw and processed foods. In setting a standard, EPA would have had to consider the validity, completeness, and reliability of available studies; the nature of toxic effects; the dietary exposure; and the effects on major subgroups of consumers. Dietary risk levels must consider the percentage of food actually treated, and actual residue levels. Obviously, this would have required EPA to consider more information to offset health considerations before setting a standard.

Either bill would have overturned the Delaney Clause that governs pesticide tolerances: It bars the EPA from granting tolerances for pesticide residues found to induce cancer in animals that concentrates in processed foods.21 This prohibition made the Clause an important symbol of environmental protection and food safety to environmental and other public interest groups but also a symbol of excessive zeal to agricultural and chemical industries.22 Either bright line or negligible risk standards would allow EPA to consider other information, even if an animal assay of a pesticide demonstrated carcinogenicity.

ERDDA required that EPA identify personnel needs for research, its Science Advisory Board (SAB) review EPA research activities and that EPA submit Congressional reports to identify, prioritize and describe major research areas that would reduce risk assessment uncertainties. The bill also established a separately-identified risk assessment research program and sought to improve methods for assessing economic impacts.

Both Senate and House reports contained language that sought to improve the quality of supported research. The conferees agreed on the importance of"credible" science and the need for EPA to improve peer-review. They required EPA to work with the National Science Foundation (NSF), the National Research Council, and the Carnegie Commission in the merit review process. The Senate report required that half of EPA's extramural research be subject to the NSF peer-review process. It also required EPA to report to Congress on research to be conducted, addressing not only its nature but also how it would affect EPA's priorities. The Act was signed in September 1994.

Several analysts have assessed the major issues legislators were trying to address. For the House Committee on Science, Space, and Technology, Schierow provided a side-by-side comparison of provisions related to environmental risk analysis.25 Agency and Congressional sources also analyzed risk bills.26 From these sources and the legislation, I discerned five major themes that kept reappearing, including good science, cost-benefit analysis, transparency and accountability, risk comparisons, and administrative prominence.

These themes seem to be based on several common perceptions relating to whether environmental laws are primarily protect human and environmental heath or as create unwarranted economic burdens. Most recent proposals seem to be driven by the second view.

Bills focused on several aspects of risk assessment to promote good science. Common proposals included mandates for peer-reviewed risk assessment guidelines that EPA would periodically update. The RAIA required regular development and publication of EPA risk assessment guidelines; similarly, the USDA Reorganization Act requires the USDA to develop a strategy for consistent risk/benefit analysis using state of the art scientific methods. Superfund reform bills required a national risk protocol with guidelines for assessments related to Superfund sites.

The RAIA required EPA to provide independent peer-review of its risk assessment guidelines and mechanisms for the Director of the Risk Assessment Program, established in the bill, to consult with assessment experts. Similarly, the amendment to the EPA elevation bill required it to publish and use peer-reviewed guidelines. The RAIA also required EPA to review all of their risk characterizations annually to determine compliance with guidelines and report to Congress. Also, several bills required EPA to develop a process for reviewing prior assessments.

Repeatedly, language required EPA to provide Congress with reports on research and training needs. The RAIA required the director of a Risk Assessment Program it created to evaluate and report them. Similarly, ERDDA required EPA to identify research personnel needs, and its SAB to review research activities. Both the RAIA and ERDDA required EPA to submit Congressional reports that identify, prioritize, and describe major research areas to reduce risk assessment uncertainties. The ETA required specific procedures to develop research priorities for environmental technologies.

To achieve state-of-the-art science, several provisions facilitated peer review and communication with the scientific community. This is probably best demonstrated in the Appropriations conference reports.27 Conferees required EPA to subject extramural research to a peer-review system guided by the NSF and other science organizations.

Requiring such an analysis in any environmental bill engendered particularly sharp debate. Rep. Waxman claimed that the original Johnston amendment would contradict several environmental statutes, such as the SDWA, that explicitly forbid cost-benefit analysis.28 This proved especially troublesome since his subcommittee had jurisdiction over many major environmental statutes.

Other provisions mandated EPA to characterize and quantify regulated risks, including estimates of existing risks and their "significance." By comparing those to estimated risks under proposed exposure levels, the agency could calculate benefits in terms of "lives saved" or "health effects averted." The RRA creates a Committee within EPA to determine such benefits for each rule. Some bills also required agencies to characterize and quantify costs, e.g., the USDA and SDWA reorganization.

A clear objective of some legislation, e.g., the EPA elevation and SDWA reauthorization bills, was that EPA identify reasonable opportunities to achieve significant risk reduction, i.e., to reduce the greatest risks and avoid regulating "minimal" ones.

There also appeared a willingness to promote research into developing new methods for estimating costs and benefits. ERDDA contained language to establish a separately identified risk assessment research program and to improve methods to assess economic impacts. SDWA reauthorization required EPA to establish methods for determining costs and benefits of environmental regulations.

Most risk provisions required EPA to develop and publish risk assessment guidelines, designed to make risk assessments "transparent" to the public. The RAIA required EPA to disclose the degree of conservatism, as determined by the default assumptions used in assessment. Further, the agency had to explain its choice of assumptions, inferences and models and provide a representative list of alternatives. Similarly, the Superfund reform bill calls for EPA to develop a national risk protocol that makes the models used and the selection criteria for the models transparent. The language indicated that its authors wanted these guidelines to be "binding."30

The RAIA required EPA to release a public notice of its intent to publish and to report to Congress on the status of risk assessment guidelines. SDWA required EPA to report to Congress on risk management decisions that present an inherent and unavoidable choice between competing risks.

Certain provisions also would mandate that contested risk assessments undergo judicial review. By requiring compliance with guidelines, those provisions were intended to increase agency accountability.31 In contrast, EPA Administrator Browner claimed that this provision would make guidelines and risk assessments "unreasonably vulnerable to judicial challenges."32

For several years, analysts have considered setting EPA's regulatory priorities based on relative risks. Relative to other risks that most individuals confront, some analysts believe that many regulated environmental risks are small.33 They have elaborate mechanisms for comparing fatalities from one activity or substance to those for another. EPA's SAB released two reports that examined risk-based decision making within the agency.34 More recently, EPA commissioned a conference on the topic and published its proceedings.35 Congress also looked into EPA priority-setting. The RRA established two committees within EPA to rank risks to huma n health and environment. The Zimmer amendment to the EPA elevation bill created a committee on relative risk to advise EPA on its strategic direction. The SDWA reauthorization required EPA to identify and rank sources of pollution with respect to the relative degree of risk of adverse effects on human health, the environment, and public welfare.

Others argue that risk comparisons furnish a poor foundation for risk communication, decision making or priority setting--claiming that quantifiable consequences alone are only one of the legitimate lenses through which to view threats, safety, and the environment. They also argue that those involved in risk ranking must not only consider risk magnitude but also other dimensions, e.g., fear of risk.36

Nevertheless, relative risk appears in the Walker and Johnston amendments, and the Risk Communication Act, all mandating comparisons of EPA-regulated risks with those familiar to ordinary citizens and ostensibly putting them into a meaningful perspective.

In furtherance of these objectives, the EPA appropriations bill for example, requires EPA to report to Congress on research it will conduct on comparative risk assessment.

Not only were these offices to conduct risk assessments, but also to make expertise accessible to administrators. The language of the RAIA required the program director to advise the EPA administrator; and the Zimmer amendment required the director of the OER to advise EPA offices. Similarly, the director of ORA&C-BA will advise USDA offices and agencies about risk-related issues.

Some provisions even elevated risk assessment to the Executive Office of the President (EOP). An RAIA provision promoted interagency coordination by requiring OSTP to survey risk assessment practices and advise the President and Congress. The survey would examine and continuously monitor methods and uses of risk assessment within agencies.

"Risk assessment" and "comparative risk" became political buzzwords. Congressman Brown complained that most legislators and staff did not fully understand the science or the implications of pending legislation.37 Certainly, some risk language was highly prescriptive and very technical. This and the level of Congressional interest provides a window of opportunity for professional societies and trade organizations that might see value in mobilizing their members to help educate the current Congress.

By controlling committee agendas, chairs will likely facilitate debate on risk analysis. A sign of their desire appears in a seldom discussed codicil of the Republican "Contract with America," used during the midterm election and the transition to the 104th Congress. It describes bills that would, if that party prevailed, be introduced within the first 100 days of the session. Under the "Job Creation and Wage Enhancement Act," risk assessment and cost-benefit analysis are part of a bill to "create jobs and raise worker wages." Given this time pressure, one should expect Congress to introduce some of the proposals discussed here, either unchanged or as parts of new legislation. Also, the attention devoted to such issues in the last Congress ensures they will be central in future environmental debates.

Top of page

Index to Science Court papers

Risk Articles Index

Franklin Pierce Law Center Home Page