Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Appendix A Overview: Risk Assessment Risk assessment can be described as an approach to problem analysis that uses available scientific data (often of varying quality) to character- ize the nature of the adverse effects of a substance or activity on human health. It is designed to produce quantitative estimates of the probability that an individual will suffer disease or death as a result of exposure to a substance, expressed in terms of population or individual mortality or morbidity (Carnegie Commission, 1993). Risk management is the proc- ess of evaluating policy alternatives and selecting the most appropriate action, integrating the results of risk assessment with social, environ- mental, economic, and political concerns. Risk assessments are fre- quently employed by risk managers to help make decisions about regulating substances. Risk assessments are intended to provide risk managers with scientifically credible information that is useful for deci- sionmaking (NRC, 1983). In addition to producing quantitative estimates, the risk assessment and risk management processes are useful for at least two other reasons. First, they serve as a way to collect, organize, and evaluate the data sur- 165
166 RISK AND DECISIONS ABOUT TRU AND HLW rounding the hazardous effects of a toxin or activity. Second, when prac- ticed as analytical-deliberative exercises, they can assist interested par- ties and decision makers in establishing a dialogue about the hazard in question and create a neutral framework for discussion and deliberation. The practice of risk assessment has evolved for decades and can en- compass important non-health outcomes, such as ecological impacts (U.S. EPA, 1998; NRC, 1996), but this discussion focuses mostly on human health risk. In 1983, the practice was codified in a formal way by the National Research Council (NRC) with the publication of Risk As- sessment in the Federal Government: Managing the Process, more commonly known as the Red Book (NRC, 1983). The Red Book divided risk assessment into four steps: (1) hazard identificationâdetermination of whether a compound is causally linked to particular health outcomes; (2) dose-response assessmentâdetermination of the relationship between magnitude of exposure and probability of occurrence of adverse health effects; (3) exposure assessmentâdetermination of the extent of human exposure to the substance; and (4) risk characterizationâa description of the nature and magnitude of human risk, including attendant uncertainty. The Red Book explains that the basic problem in risk assessment is the sparseness and uncertainty of scientific knowledge, a problem that has no readily available solution. In each of these steps, the report notes that there are a number of decision points at which risk can only be inferred from available information. In some circumstances, scientific judgment and/or science policy choices may be applied to select from among pos- sible inferences. Among the recommendations in the Red Book was the suggestion that regulatory agencies maintain and establish a clear conceptual dis- tinction between risk assessment and risk management. Scientific find- ings and policy judgments embodied in risk assessments should be distinguished from political, economic, and social considerations that affect regulatory choices. This conceptual distinction allows for interac- tions and iterations between risk assessment and risk management, but not confusion of the two. The report also states that regulatory agencies should develop a set of inference guidelines to structure the interpretation of scientific and technical information. The goal of these guidelines is to promote clarity, completeness, and consistency. The guidelines should also help maintain the distinction between risk assessment and risk man- agement. The U.S. Environmental Protection Agency (U.S. EPA), the U.S. Nuclear Regulatory Commission (U.S. NRC), and other executive agen- cies and departments have published inference rules covering carcino-
APPENDIX A 167 genicity, mutagenicity, developmental toxicity, exposure, and effects of chemical mixtures (see, e.g., U.S. EPA, 1986, 1992). These guidelines, some of which have been updated and reissued over the years, seek to preserve flexibility while building consistency and clarity in the risk evaluation process. Many aspects of these guidelines have been contro- versial (NRC, 1994a). As risk assessment practice has matured and lessons from the Red Book and inference guidelines have been institutionalized, various criti- cisms have arisen. These concerns include (1) the lack of scientific data quantitatively linking chemical exposures to health risks; (2) the lack of uniformity in the type and manner of reporting research results, making it difficult to compare data from different laboratories and different studies; (3) the uncertainties associated with modeling, which is generally used where direct measurement is not possible (including predictions of how systems will behave in the future); (4) the use of conservative default options (i.e., those more likely to overstate, rather than understate, hu- man risk) and when these options should be abandoned in favor of new information; and (5) the qualification and quantification of variability and uncertainty (NRC, 1994a). To address these and other issues and to fulfill a statutory require- ment of the Clean Air Act Amendments of 1990, the NRC again estab- lished a review committee to study risk assessment protocols, specifically those employed by U.S. EPA (NRC, 1994a). After extensive discussion, the review committee confirmed that the Red Book model was an effective risk assessment paradigm. It also concluded that U.S. EPA had acted reasonably in formulating default guidelines. The report suggested that U.S. EPA consider a methodology to assess cumulative riskâmultiple risks from multiple sourcesâand advised the agency to adopt principles for choosing default options and for judging when and how to depart from them. The NRC committee elaborated on two issues that were discussed briefly in the Red Book: variability and uncertainty, recommending further research and better treatment of both issues (NRC, 1994a). In the mid-1990s, two influential publications investigated risk as- sessment and risk management and offered recommendations to improve their application and credibility. In Understanding Risk: Informing Deci- sions in a Democratic Society (NRC, 1996), the NRC focuses on the complex and controversial nature of risk characterization, recasting it as an analytic-deliberative process. The report offers criteria for evaluating the analytic-deliberative process that leads to risk characterization. The criteria, which are set out in Table A.1, can be used to develop realistic
168 RISK AND DECISIONS ABOUT TRU AND HLW goals for the process. The message of the Understanding Risk report is important for the risk analyst, who must be cognizant of the fact that his or her work product will be relied upon to make regulatory decisions. As such, it must be responsive to the problem, reflective of the strengths and weaknesses of the science, and robust enough to withstand critical peer scrutiny. In addition, Understanding Risk emphasizes that to conduct risk assessment and risk management processes well, scientists and technical experts must interact and work with the public meaningfully (NRC, 1996). Responding to public concerns and questions, and changing course if necessary based on stakeholder input, are linchpins of the ana- lytic-deliberative models. The Presidential-Congressional Commission on Risk Assessment and Risk Management (1997a, 1997b) issued a two-volume report con- taining 71 recommendations aimed at improving federal agencies' ap- proaches to environmental and public health threats. These recommendations rest on a framework for risk management that, like the Understanding Risk report, emphasizes the need for a comprehensive, problem-solving, iterative approach to risk management. The commis- sion suggests that federal agencies move from a one-pollutant-at-a-time approach to a multimedia, multisource, multichemical risk assessment methodology that integrates information about many pollutants and di- verse endpoints (PCCRARM, 1997a, 1997b). As the use of risk-based decision making has spread, the practice of risk assessment and risk management has continued to advance in the directions set out in Understanding Risks and the Presidential- Congressional Commission reports. Recent efforts have called for even greater citizen and community interaction and challenged risk analysts to tackle cumulative risks and aggregate exposure. The U.S. EPA has pub- lished a Framework for Cumulative Risk Assessment (2002), which it sees as the first step in a long-term strategy to develop cumulative risk assessment guidelines. The framework defines cumulative risk assess- ment as an analysis, characterization, and possible quantification of the combined risks to human health and the environment from multiple agents or stressors. It recognizes that within communities there are mul- tiple contributors to health and differential susceptibility within popula- tions at risk. The centerpieces of a cumulative risk approach are a focus on a specific, âflesh-and-bloodâ community; involvement of that com- munity from the start in the assessment process; and a full and detailed vulnerability analysis for that community.
APPENDIX A 169 Table A.1. Criteria of Risk Characterization Criterion Measurement procedure Getting the science right Ask risk analytic experts who rep- resent the spectrum of interested parties to judge the technical ade- quacy of the risk-analytic effort Getting the right science Ask representatives of the inter- ested and affected parties how well their concerns were ad- dressed by the scientific work that informed the decision Getting the right participation Ask public officials and represen- tatives of the interested and af- fected parties if there were other parties that should have been in- volved Getting the participation right Ask representatives of the parties whether they were adequately consulted during the process; if there were specific points when they could have contributed but did not have the opportunity Developing accurate, balanced, Ask representatives of the parties and informative synthesis how well they understand the bases for the decision; whether they perceived any bias in the in- formation coming from the re- sponsible organization Source: NRC, 1996. The U.S. EPA framework adopts a three-phase process for risk-based decision making (see Figure A.1). The process begins with a problem analysis, or scoping and planning phase. The framework points out that
170 RISK AND DECISIONS ABOUT TRU AND HLW at this initial stage a risk assessment team should be organized. The team should include technical experts, members of the community or commu- nities, and others. Scientists can play an important role in this phase by collecting, organizing and presenting data for all of the involved parties. However, planning, scoping, and screening also require the input of so- cietal values and stakeholder participation. While scientists can help identify and characterize risks, they are not uniquely qualified to set pri- orities among them. Cumulative Risk Assessment Planning, Scoping, and Problem Formulation Analysis Interpretation and Risk Characterization FIGURE A.1 Three-phase process for risk-based decision-making. SOURCE: U.S. EPA, 2002.