1
Introduction

Promoting human health and safety by reducing exposures to risks and harms through regulatory interventions is a critical responsibility of government. Such efforts encompass a wide array of activities in many different contexts, such as improving air and water quality; safeguarding the food supply; reducing the risk of injury on the job, in transportation, and from consumer products; and minimizing exposures to toxic chemicals. Estimating the magnitude of the expected health and longevity benefits helps policy makers decide whether particular interventions merit the expected costs associated with achieving these benefits and informs their choices among alternative strategies. The results of such cost-effectiveness analyses (CEAs) can be one important contribution to the regulatory decision-making process.

This report of the Committee to Evaluate Measures of Health Benefits for Environmental, Health, and Safety Regulation provides recommendations and guidance on the valuation of health and safety improvements in CEA. More specifically, it considers how health-adjusted life-year (HALY) measures, which combine morbidity and mortality effects in a single index value, should be used to assess the benefits of regulatory interventions.

This introductory chapter first presents the Committee’s charge in the context of recent policy guidance. It then describes the role of economic analysis in the regulatory development process. Next, it introduces CEA in more detail, noting how it differs from benefit–cost analysis (BCA). The chapter concludes with a brief overview of the remainder of the report.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis 1 Introduction Promoting human health and safety by reducing exposures to risks and harms through regulatory interventions is a critical responsibility of government. Such efforts encompass a wide array of activities in many different contexts, such as improving air and water quality; safeguarding the food supply; reducing the risk of injury on the job, in transportation, and from consumer products; and minimizing exposures to toxic chemicals. Estimating the magnitude of the expected health and longevity benefits helps policy makers decide whether particular interventions merit the expected costs associated with achieving these benefits and informs their choices among alternative strategies. The results of such cost-effectiveness analyses (CEAs) can be one important contribution to the regulatory decision-making process. This report of the Committee to Evaluate Measures of Health Benefits for Environmental, Health, and Safety Regulation provides recommendations and guidance on the valuation of health and safety improvements in CEA. More specifically, it considers how health-adjusted life-year (HALY) measures, which combine morbidity and mortality effects in a single index value, should be used to assess the benefits of regulatory interventions. This introductory chapter first presents the Committee’s charge in the context of recent policy guidance. It then describes the role of economic analysis in the regulatory development process. Next, it introduces CEA in more detail, noting how it differs from benefit–cost analysis (BCA). The chapter concludes with a brief overview of the remainder of the report.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis THE CHARGE TO THE COMMITTEE This report responds to a change in the requirements for economic analysis of major regulations developed by federal agencies. Until recently, agencies were required only to conduct BCAs, assessing both net impacts on the national level and their distribution across different subgroups of concern. In 2003 new government-wide requirements for CEA of major health and safety rules were added, and those are now being implemented by the agencies. The charge of this Committee is to provide advice related to these new requirements. Background Federal agencies are directed by executive order (and sometimes by statute) to estimate and report the expected benefits of proposed major regulations in conjunction with their estimated costs.1 For economically significant health and safety rules, these analyses must generally include both a BCA and a CEA. BCA involves estimating the monetary value of benefits (e.g., improvements in human health or in the natural environment), then subtracting regulatory costs (e.g., related to industry compliance) to determine the net benefits of regulatory options. Ideally, the value of benefits is determined from estimates of willingness to pay (WTP) for the risk reductions or other impacts. In contrast, CEA involves dividing regulatory costs by a nonmonetary benefit measure (e.g., number of lives saved, cases of illness avoided, years of life gained, or tons of pollution reduced) to determine the unit costs of achieving the benefits. When regulations reduce the risks of different kinds of illness or injury, or of both fatal and nonfatal effects, analysts may use measures that reflect impacts on health-related quality of life (HRQL) and longevity as the indicator of effectiveness. These measures are generally referred to as HALYs; the quality-adjusted life year (QALY) is the most common HALY metric. HALY measures convert different types of health effects to a composite unit so that regulatory costs can be compared to a single measure of health-related effectiveness. Until recently, guidelines for regulatory analysis focused on the conduct of BCA. In 2003, however, the U.S. Office of Management and Budget (OMB) within the Executive Office of the President (EOP) issued Circular A-4, Regulatory Analysis, which instructed executive branch agencies to 1   Not all of the activities of regulatory agencies result in regulations subject to requirements for economic analysis. For example, the Food and Drug Administration’s review and approval of new drugs does not involve issuance of regulations and is not subject to requirements for BCA or CEA.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis begin conducting CEAs for economically significant health and safety regulations whenever feasible. The guidance also continues to require preparation of a BCA and assessment of the distribution of the impacts. The guidelines explicitly recognize that BCA and CEA provide different perspectives on the economic effects of regulatory interventions, in combination offering complementary and useful information for decision making. An additional rationale for requiring CEA offered by OMB is that it facilitates cross-program comparisons (see Box 1-1). BOX 1-1 OMB’s Rationale for Requiring CEA as Part of Regulatory Analysis In remarks to a 2003 conference on Valuing Health Outcomes at Resources for the Future, John Graham, Administrator of the Office of Management and Budget’s (OMB’s) Office of Information and Regulatory Affairs, gave the following rationale for requiring the conduct of regulatory cost-effectiveness analyses (CEAs) in addition to benefit–cost analyses (BCAs) (Graham, 2003b): “… CEA is useful because it provides information about which regulatory alternatives will produce the most health gains per unit of resource investment. It is a ‘bang for the buck’ exercise, where the payoff is measured in health units rather than dollars. My experience as both a professor and government administrator is that some people who are skeptical of traditional benefit–cost analysis gain insight from the cost-effectiveness perspective. I think it is instructive that the peer-reviewed medical and public health literature is far more dominated by CEA than BCA…. Since the CEA only provides relative comparisons, we need BCA to determine whether the benefits of any particular alternative justify the costs.” “… In order to promote more consistency, OMB will be sponsoring interagency discussions about the most promising and practical effectiveness measures. We will also request that agencies supply OMB their original data on mortality and morbidity. OMB will then be in a position to compare rulemakings across agencies using similar methods and assumptions. The Administration is moving with determination toward more performance-based budgeting, and a greater focus on cost-effectiveness and net benefits should be helpful in budgeting.” “In BCA, the monetary valuation of lifesaving is as important as it is controversial…. While these issues are crucial in BCA, they may be considered less important for CEA. In the health field, CEA is often defended partly on a social-contract basis rather than on a pure free-market basis. Here is a version of the social-contract argument. In what the late John Rawls called the ‘original position,’ where citizens are blinded by a ‘veil of ignorance’ to their own age, health status and wealth, they might rationally prefer a social contract that would maximize the number of healthy life years saved through public policy, given the resources available. Of course, I am not aware of any interest groups in this town who are prepared to wear this veil of ignorance when they visit Congress or OMB, but that is ‘just’ a practical problem!”

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis These remarks suggest that OMB is interested in the use of CEA for two purposes: to inform the decisions regarding the selection of a regulatory approach to a particular problem and to aid in cross-program comparisons of performance. Such comparisons, sometimes referred to as “league tables” or “scorecards,” might, for example, involve comparing the cost per life saved across programs that address air pollution, foodborne disease, and automobile accidents. In contrast, the selection of regulatory approaches involves within-program comparisons, such as different options for reducing the emission of a particular air pollutant. To support these applications of CEA, OMB asked the Institute of Medicine (IOM) to evaluate various cost-effectiveness methodologies and assess their theoretical soundness, feasibility, and ethical underpinnings and implications, as well as to provide recommendations and guidance to federal agencies regarding the use of these measures in regulatory analysis. This report aims to help agencies understand the strengths and weaknesses of various health measures used in CEA, and demonstrates their application. It also provides advice on the appropriate consideration of cost-effectiveness within the larger decision-making context. The report builds on the work of the U.S. Panel on Cost-Effectiveness in Health and Medicine (PCEHM), which published its recommendations in 1996 (Gold et al., 1996b), and on the work of an earlier IOM Committee on Summary Measures of Population Health (IOM, 1998). It also benefits from the work of the National Research Council Committee on Estimating the Health-Risk-Reduction Benefits of Proposed Air Pollution Regulations (NRC, 2002). The Task OMB’s Office of Information and Regulatory Affairs and a consortium of federal agency sponsors charged the IOM Committee to identify existing and proposed measures of benefits for use in regulatory CEA, to describe and evaluate these measures, to recommend a subset of measures to be considered for use by federal regulatory agencies, and to identify areas of research needed to enhance the development and consistent use of such benefit measures by the agencies. Specifically, the Committee was asked to: Describe current practices among federal agencies for evaluating the costs and outcomes of regulatory measures. Review measures of health benefits currently used in CEA, including specific instruments and assessment techniques applied to value health within these composite measures. Develop criteria for choosing among measures potentially useful in the evaluation of health outcomes of regulatory actions.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis Assess the various measures of health benefits in terms of their data requirements and feasibility of application, theoretical validity, appropriateness for special populations (children, ethnic minorities), and ethical implications. Recommend measures appropriate for use by federal agencies in evaluating the health outcomes of regulatory actions in light of the Committee’s criteria. Construct CEA case studies, using information from published BCAs, to illustrate the impact of alternative measures of health benefits within CEA and to compare the CEAs and BCAs. Discuss criteria for identifying regulations for which CEA of regulatory impacts would be informative. Recommend research that would improve the measurement of health benefits in regulatory analyses. The sponsors’ charge specifically directed that aspects of CEA other than the measurement and valuation of health benefits—discounting, for example—be considered only as they relate to the question of valuing health benefits.2 The charge also specified that the Committee should refrain from examining methods for assigning a monetary value to HALY measures. The Committee does, however, comment more generally on the practice of using monetized HALY measures in BCA. To respond to the charge, in March 2004 the IOM constituted a 16-member committee of experts in health services research, environmental health, economics, psychology and survey research, statistics and epidemiology, decision analysis, ethics, law and regulatory policy, and disability policy. This report represents the Committee’s findings, conclusions, and consensus recommendations. THE ROLE OF ECONOMIC ANALYSIS IN REGULATORY DEVELOPMENT The process for developing major regulations is complex and often spans several years. It involves numerous individuals ranging from technical experts within the agencies to political appointees in the executive branch, and may include members of Congress and the judiciary, as well as a large number of interest groups and individual citizens. The types of 2   Throughout this report, we use the terms “value” and “valuing” in their broader, everyday senses, that is, to represent the worth, importance, or usefulness of something (such as a particular state of health). Wedo not use these terms in their narrower, economic meaning of monetary equivalent.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis information considered are diverse, including factors such as technical feasibility and equity considerations as well as economic effects. Below, we briefly summarize the regulatory development process and the role of economic analyses (BCA and CEA) in related decision making as context for the remainder of this report. Chapter 2 builds on this discussion and provides more information on the types of regulations affected, the requirements for economic analysis, and current agency practices. Regulatory Development Process The starting point for the federal regulatory process is usually a statute passed by Congress that requires the development of regulations. These statutes identify, in general terms, the issues and goals that Congress expects the agencies to address and the goals of related regulations. Although the statute may also legislate some specific aspects of the regulations (as discussed in more detail in Chapter 2), executive branch agencies have primary responsibility for determining the detailed requirements. Although the statute may include deadlines for regulatory action, or deadlines may be established by the courts if the action has been subject to litigation, the executive branch usually has some discretion in setting regulatory priorities. Thus the schedule for regulatory activity is determined by a combination of legal requirements, Presidential priorities, and Administration and individual agency preferences. This schedule is reflected in the regulatory agendas developed by each agency, which are published twice a year in the Federal Register and list all recently completed and planned regulatory actions. Once an agency decides to move ahead with a regulation, it often convenes an internal agency workgroup to study the problem and identify regulatory options (see, e.g., EPA, 2003); generally a particular office has lead responsibility. The workgroup is likely to include personnel with expertise in related technical, scientific, economic, legal, health, and other issues. This workgroup or responsible office commissions related studies; the economic analysis is often one of several research efforts undertaken to support major rules. For example, separate studies may address the technical feasibility or effectiveness of different regulatory requirements, as well as the health risks associated with the hazard. Agency staff also engage in both informal and formal public outreach activities; public involvement may, in some cases, include public hearings or negotiations among directly affected groups and other interested parties. Based on the results of these efforts, agency staff prepare suggested regulatory language and supporting analyses, which are reviewed internally and then by OMB if the regulation is significant. Other agencies are involved in this review if the regulation has implications for their activities.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis BOX 1-2 Key Requirements of the Administrative Procedure Act Under the Administrative Procedure Act, the key process requirements that must be met are to provide notice to the public and to solicit public input. With exceptions only for emergency situations, all regulations must first be issued as a Notice of Proposed Rulemaking and published in the Federal Register. As part of the process, agencies establish a time period (often 90 days) during which interested citizens may submit comments. When the final regulation is published in the Federal Register, the agency’s response to the comments is included. The final regulation is then incorporated into the Code of Federal Regulations. The proposed regulatory language is then published in the Federal Register, along with a preamble that discusses the rationale for the overall regulation and its specific provisions and that summarizes the economic analysis. Related reports, including the complete economic analysis, are placed in a docket (and generally on the agency’s website), where they can be reviewed by the general public. This process is governed by statutory guidelines as well as by agency administrative directives. Under the Administrative Procedure Act (APA, 5 U.S.C. Art. 500 et seq.), agencies are required to provide supporting information and request comment from the public before finalizing the regulation.3 These requirements are rooted in notions of due process and fairness and, if not strictly followed, the courts will rule that the regulation has not been appropriately issued. The APA requirements are summarized in Box 1-2. Once the agency receives comments, the workgroup may reconvene to review the comments, determine whether additional analysis or public outreach is needed, and develop the final regulation and supporting analyses. The final regulation is then published in the Federal Register after review within the issuing agency as well as by OMB and other agencies. The economic analysis is generally updated to reflect any new information received as well as any changes in the regulatory requirements, and placed in the docket along with other supporting documentation. This documentation includes the comments received by the agency and the agency’s detailed responses. The Federal Register preamble to the regulation generally summarizes the comments received and the agency’s responses, provides information supporting and describing the final regulatory provisions, and sum- 3   See Jacobson and Hoffman (2003) for more detailed information on these requirements.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis marizes the economic analysis. The final rule may later be overturned by congressional action, the results of litigation, or new regulatory efforts. Requirements for Regulatory Analysis and Decision Making Within the legal framework and administrative process discussed above, the results of economic analyses (including both BCA and CEA) are one of many types of information considered by policy makers. Executive Order 12866, “Regulatory Planning and Review,” establishes the Administration’s basic approach to regulatory development and decision making. This Executive Order was issued by President Clinton in 1993 and reaffirmed by the current Bush Administration. Box 1-3 provides a brief review of its evolution. BOX 1-3 History of Administrative Guidance on Regulatory Analysis According to the Office of Management and Budget, or OMB (1997), requirements for regulatory analysis and centralized review can be traced back to the early 1970s. Initially, assessment efforts were directed narrowly at particular types of impacts. The Nixon Administration concentrated on reducing the burden of environmental regulations on business. Under President Ford, the focus shifted to inflation; ultimately, the members of Ford’s Council on Wage and Price Stability concluded that a regulation would not be truly inflationary unless its social costs exceeded its benefits. As a result, benefit–cost analysis was required for major regulations; this requirement has persisted in modified form through today. President Carter built on the Ford legacy, issuing Executive Order 12044, Improving Government Regulations (1978). This Executive Order established general principles for regulatory development and required analysis of regulations with major economic impacts. President Carter also initiated a process for centralized review of significant regulations. President Reagan then made regulatory relief a cornerstone of his economic program and developed a more centralized system involving OMB review of agencies’ major regulatory actions (1981). He issued Executive Order 12291, Federal Regulation, which required agencies to prepare benefit–cost analyses for major rules and instructed them to issue only those regulations that maximize net benefits. The first Bush Administration continued to adhere to these requirements. In 1993, President Clinton issued Executive Order 12866, Regulatory Planning and Review. This Executive Order continued the Reagan requirements for centralized OMB review with some modifications, for example, to provide more public information on the review process. This Executive Order retained the requirements for benefit–cost analysis as well as the general principle that regulations should be issued only if their benefits justify the costs. The current Bush Administration has affirmed its commitment to Executive Order 12866, amending it only to change the roles and responsibilities of some of the officials involved in regulatory review.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis Executive Order 12866 sets out basic principles for decision making and establishes a regulatory analysis and review process for ensuring that these principles are met. Some of its key provisions include: Ensuring that regulations are promulgated only when necessary. Establishing a regulatory development process that includes interagency coordination, centralized review, and public involvement. Requiring consideration of a wide range of impacts, including both those that can be expressed in quantitative terms and those that cannot. The general regulatory philosophy supported by the Executive Order is expressed in its first section as follows: Federal agencies should promulgate only such regulations as are required by law, are necessary to interpret the law, or are made necessary by compelling public need, such as material failures of private markets to protect or improve the health and safety of the public, the environment, or the well-being of the American people (EOP, 1993, p. 1). Subsequent sections of the Order implement this philosophy through a variety of requirements for analysis and review, which are further reinforced in OMB’s 2003 Circular A-4, Regulatory Analysis. (This Circular is included as Appendix C.) The Circular expands the discussion of the types of market failures that may lead to the need for federal regulation, including externalities (e.g., actions that impose uncompensated benefits or costs on others, such as industrial pollution), market power related to imperfect competition or natural monopolies, and inadequate or asymmetric information. Both the Executive Order and the Circular direct that agencies should assess nonregulatory approaches (e.g., economic incentives or information dissemination) as well as a variety of types of regulatory options when considering how to address these problems. They also emphasize the need to consider the impact of regulations on different levels of government. Much of Executive Order 12866 is focused on establishing a process for centralized review of regulations to ensure they meet the goals of the Order and of the Administration. This process includes interagency review and review by state, local, and tribal governments, as well as by OMB staff. In addition, the Executive Order requires agencies to involve the public in regulatory development, for example, by disseminating information on regulatory plans, providing opportunities for public comment, considering the use of consensual approaches such as negotiated rulemakings, and publishing information on the rationale for the regulation as well as the supporting analyses. Executive Order 12866 also establishes the basic requirements for economic analysis of major regulations. These requirements concentrate on the

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis comparison of benefits and costs, although the Order also notes that regulations should be designed to meet their goals in a cost-effective manner. More specifically, in its discussion of general principles, it notes: In deciding whether and how to regulate, agencies should assess all costs and benefits of available regulatory alternatives, including the alternative of not regulating. Costs and benefits shall be understood to include both quantifiable measures (to the fullest extent that these can be usefully estimated) and qualitative measures of costs and benefits that are difficult to quantify, but nevertheless essential to consider. Further, in choosing among alternative regulatory approaches, agencies should select those approaches that maximize net benefits (including potential economic, environmental, public health and safety, and other advantages; distributive impacts; and equity), unless a statute requires another regulatory approach (EOP, 1993, p. 1). For economically significant regulations (defined in Box 1-4), the Executive Order establishes a number of specific analytic requirements. Consistent with its overall philosophy, the Order defines both costs and benefits broadly to include both economic impacts and other social welfare concerns such as fairness or equity, and indicates that both quantified and nonquantifiable impacts must be considered. Examples of benefits include “the promotion of the efficient functioning of the economy and private markets, the enhancement of health and safety, the protection of the natural environment, and the elimination or reduction of discrimination or bias”; examples of costs include the direct cost both to the government in administering the regulation and to businesses and others in complying with the regulation, and any adverse effects on the efficient functioning of the economy, private markets (including productivity, employment, and competitiveness), health, safety, and the natural environment” (EOP, 1993, p. 7). BOX 1-4 Definition of “Economically Significant” Regulations According to Executive Order 12866, regulatory actions are identified as economically significant if the resulting rule is likely to: Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities (EOP, 1993, p. 4).

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis This focus on both quantitative and qualitative benefits and on the distribution of impacts is reinforced by two additional executive orders that require agencies to address risks that disproportionately affect children and to address adverse human health or environmental effects on minority and low-income populations. The provisions of these executive orders are summarized in Box 1-5. The analytic requirements established by these Presidential executive orders and defined more extensively in OMB Circular A-4 are discussed in more detail in Chapter 2 and in a background paper commissioned by the Committee (Robinson, 2004). Although the requirements apply specifically to significant regulations subject to OMB review, such analyses are often carried out for rulemakings that are not economically significant, as well as BOX 1-5 Executive Orders Executive Order 13045: Protection of Children from Environmental Health Risks and Safety Risks In April 1997, President Clinton issued an Executive Order directing all federal agencies to (1) identify and assess health and safety risks that may disproportionately affect children, and (2) ensure that agency activities address such risks (EOP, 1997, p. 1). To meet these goals, the Order establishes a task force to address these issues, requires agencies to coordinate related research, and creates a forum to measure progress. In addition, the Order directs agencies, when proposing and promulgating regulations concerning these types of risks, to submit to the Office of Management and Budget (OMB) an evaluation of the regulation’s effects on children and an explanation of why the regulation is preferable to other feasible alternatives considered. Executive Order 12898: Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations This Executive Order was issued in 1994 by President Clinton, and establishes achieving environmental justice as part of the mission of executive branch agencies. Each agency is required to identify and address “disproportionately high and adverse human health or environmental effects of its programs, policies, and activities on minority populations and low-income populations” (EOP, 1994, p. 1). Specifically, the Order establishes an interagency working group to provide coordination and guidance, requires each agency to develop an environmental justice strategy, and establishes research priorities. Agency activities must include provisions for improving related research and data collection efforts, for ensuring greater public participation, and for identifying differential patterns of natural resource consumption among minority and low-income populations.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis based on market data. However, the benefits of regulatory requirements, such as improvements in human health, generally reflect outcomes that are not normally bought or sold. Thus analysts must rely on other valuation approaches. As noted earlier, the starting point for estimating monetary values is typically the concept of individual WTP: the maximum amount of money an individual would voluntarily exchange to obtain the improvement (e.g., reduced health risks or better health status), given his or her budget constraints.7 For outcomes that are not directly traded in markets, researchers may estimate monetary values by asking individuals what they would be willing to pay for the improvement. These types of approaches, which include contingent valuation surveys, conjoint analyses, and similar research strategies, are generally referred to as “stated preference” methods. Alternatively, researchers may estimate WTP based on market behavior for related goods; these approaches are generally referred to as “revealed preference” methods. Wage-risk studies, which estimate the change in wages required for riskier jobs (using statistical methods to separate out other factors that affect wage levels), are one example of a revealed preference approach.8 These methods focus on individual WTP and may not fully capture the value an individual places on risk reductions or other benefits that accrue to fellow members of society. Although WTP studies could, in theory, be designed to include altruistic considerations, the extent to which this is accomplished in practice is debatable and may raise double-counting issues when WTP is summed across individuals (Jones-Lee, 1992; Viscusi et al., 1988). In addition, in some cases it may be appropriate to add components of a cost-of-illness (COI) measure to a WTP estimate to provide a more complete accounting of social welfare impacts, particularly if the WTP measure excludes costs borne by others (e.g., employers and insurers).9 7   The application of this concept to the valuation of health impacts was introduced by Schelling (1968) and Mishan (1971). 8   Wage-risk studies reflect the market equilibrium that results from workers’ demands for wages and firms’ willingness to supply jobs at these wage rates. See Freeman (2003) for a discussion of the relationship between the resulting estimates and WTP. 9   In some cases, analysts rely on COI estimates as a substitute for WTP, when WTP studies of reasonable quality are not available for the health effect of concern. However, COI estimates are not a preferred measure of value from the perspective of welfare economics. Such estimates usually include medical expenses (e.g., for doctor visits, prescription medicine, hospital stays) and may also include lost work time (e.g., foregone earnings, decreased household production). However, they tend to understate WTP for risk reductions due, at least in part, to the exclusion of the value of lost leisure time and pain and suffering (EPA, 2000b, Appendix B; EPA, 2005b). This use of COI as a substitute for, or supplement to, estimates of WTP differs from its use in CEA, as discussed later in this chapter.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis Calculation of Net Benefits Once the benefits and costs of regulatory options are estimated, the monetary values can be aggregated and compared as part of the decision-making process. Several kinds of criteria can be applied to determine whether a policy is worth pursuing, given the BCA results. Perhaps the simplest formulation is the Pareto Principle, which states that a project is desirable (or economically efficient) if it makes at least one person better off, and makes no person worse off. Although attractive in theory, few policies meet this criterion because regulations impose costs or otherwise adversely affect at least some individuals or organizations. To address this limitation, variations on this standard were developed by Kaldor and Hicks (Hicks, 1939; Kaldor, 1939).10 These variations suggest that a project is desirable if it makes “the winners” (i.e., those who benefit) better off by an amount large enough to compensate “the losers” (i.e., those who are harmed); or, alternatively, that a project should be rejected if the losers could pay the winners to not pursue the policy and not be worse off. These criteria do not demand that actual compensation occur. The Kaldor-Hicks criterion forms the basis of the standard decision framework generally applied in BCA. This framework suggests that policies should not be pursued if costs exceed benefits, and, if more than one policy provides positive net benefits, the preferred choice is the option with the largest net benefits. In practice, regulatory decisions are rarely, if ever, based solely on the results of a BCA. This type of analysis provides a useful framework for organizing and analyzing information, and increases the comparability of costs and benefits. To the extent that the costs and benefits of regulations can be quantified, it promotes the identification of economically efficient interventions. However, as discussed earlier in this chapter, decision makers must also weigh issues related to distributional impacts and equity and the potential effects of nonquantified factors, and they must respond to concerns raised by the public. As discussed in Chapter 2, statutory and other legal requirements also must be met. Cost-Effectiveness Analysis CEA constitutes another major practice for quantitative evaluation of policies for improving human health, safety, and longevity. In CEA, the 10   See Mishan (1988) for a standard treatment of Kaldor-Hicks.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis desirable effects of a policy option are not measured in dollars. Instead, they may be accounted for by single-dimension measures (e.g., cases of disease or injury averted, years of preventable mortality avoided, tons of emissions reduced) or with integrated metrics such as HALYs. The costs of each option are then divided by the effect measure to determine the cost per “unit” of benefit provided. The following section provides general information on CEA and summarizes some of the key differences between CEA and BCA. The subsequent chapters of this report consider the use of CEA and (to a limited extent) BCA in regulatory analysis in more detail. Valuation Approach CEA is based on many of the same principles as BCA, reflecting similar concerns with social welfare and individual choice. Because CEA results in a ratio, consistent definition of what is counted as cost (the numerator) and what is counted as effect (the denominator) is of greater importance than in BCA. To promote comparability across analyses, practitioners have defined a “reference case” that includes recommendations for distinguishing between costs and benefits in the CEA context. Valuation practices for health and safety CEAs have been developed largely in the context of health care policy, not regulation. In 1993, the U.S. Public Health Service appointed a group of 13 experts, the PCEHM, to consider issues related to improving the quality and comparability of CEAs used in health policy and medical decision making. In its 1996 report (Gold et al., 1996b), the PCEHM provides specific recommendations for measuring and distinguishing between costs and benefits. These recommendations, codified as a reference case from the societal perspective, are the current standards for best practices in this field. The starting point for assessing the costs of an intervention is the same in CEA and BCA; both rely on the concept of opportunity costs as the basis for valuation and generally use the same approaches to estimate these costs. Both types of analysis also attempt to address all (nonnegligible) costs that are attributable to the intervention and its current and future consequences. CEA differs only in that analysts must be careful to exclude those costs that are addressed by the effectiveness measure to avoid double counting. For health care programs, PCEHM recommends that, in the reference case, costs include changes in the use of health care resources, treatment-related changes in the use of non-health care resources, changes in the use of informal caregiver time, and changes in the use of patient time due to

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis treatment.11 PCEHM defines these cost components as follows (Gold et al., 1996b, pp. 179–181): Direct health care costs include those associated with medical services such as the provision of supplies (including pharmaceuticals) and facilities as well as personnel salaries and benefits. Direct non-health care costs include nonmedical resources used to support the intervention; the Panel’s examples include the costs of child care while a parent is undergoing treatment, of dietary changes, and of transportation to and from a medical facility. Informal caregiver time reflects the unpaid time spent by family members or volunteers in providing home care. (Paid time for nursing and other medical care is included in the direct health care component.) Patient time involves the time spent in treatment, but not other changes in time use attributable to the health condition of concern. In particular, lost productivity due to illness is excluded from these costs in the reference case to avoid double counting; ideally the effects of illness on the patient’s usual activities should be part of the effectiveness measure. The first two categories involve goods and services for which payment is generally provided; thus monetary expenditures can be used to estimate related values. The latter two areas involve the use of uncompensated time, which is generally valued at the after-tax wage rate, under the assumption that (at the margin) this rate represents the opportunity cost of not engaging in paid work. The Panel’s recommendations focus on the use of CEA to assess health care interventions; the analysis of costs in a regulatory CEA involves additional considerations. First, the regulation itself imposes costs related to the actions it requires organizations and individuals to undertake. For example, an air pollution regulation may require the installation of emissions control devices to reduce the incidence of respiratory and cardiovascular conditions. Second, because the regulation is preventing health effects from occurring, the health care costs listed above accrue as savings rather than as expenditures. Third, regulations may have both health- and non-health-related benefits (e.g., reduced damage to ecological systems) that are not taken into account in health-related effectiveness measures and hence could 11   The COI measures used in BCA may differ in some respects from these recommendations for CEA, particularly because they often include the effects of illness on lifetime earnings or productivity, rather than focusing only on the impacts directly associated with medical treatment.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis conceivably be included as offsets to costs, as discussed in the Committee’s recommendations. Most health and safety regulations typically lead to more than one type of health benefit, in many cases preventing a range of different types of illnesses or injuries as well as increasing life expectancy. As a result, benefit measurement in regulatory CEAs is likely to rely on integrated HALY measures, such as QALYs, that allow analysts to combine information about various health conditions. These measures provide a common metric for estimating the HRQL impact of different conditions, which includes functioning in domains such as mobility, emotion, social activity, and self-care. When these measures are applied, health states are assigned index values12 that reflect their relative desirability or impact on health-related quality of life. These values are usually placed on a zero-to-one scale, where zero corresponds to death and one corresponds to perfect or optimal health. Nonfatal health impairments (disability and morbidity) are assigned intermediate values, with lower numbers representing more severe impacts. (Most indexes can accommodate negative values, representing impaired health states that have been valued as worse than death.) For public policy decisions, PCEHM recommends that these condition weights be based on the preferences of the general population (“community weights”), rather than those of patients or clinicians, to better reflect societal values (Gold et al., 1996b). These index values are generally based on surveys that are similar in many respects to the stated preference surveys used to estimate WTP. Perhaps the most significant difference is the form of the valuation question. Rather than asking individuals what they would be willing to pay to avert a health risk, HALY surveys ask respondents about the relative desirability of different health states. Chapter 3 discusses the methods for constructing index values and the strengths and weaknesses of alternative approaches. Once index values are estimated, they can be multiplied by the duration of each condition to determine the associated HALYs, multiplied by the number of cases averted for each health state, then added across conditions to create a single effectiveness measure. As discussed in subsequent chapters, these values are generally discounted to reflect the extent to which the impacts are spread over time. The idea of using such summary health measures in CEA was introduced four decades ago (Chiang, 1965; Fanshel and Bush, 1970). In 1977, 12   Throughout the report, we use the term “health state index value” for consistency. These index values are also referred to as “utility weights” or “preference weights” in the research literature.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis Weinstein and Stason offered theoretical grounding and analytical guidelines for using QALYs to evaluate health and medical practices, noting that they combine information about changes in survival and morbidity in a way that reflects individuals’ willingness to trade off between them. First employed to evaluate clinical medical interventions, QALYs compared relatively similar health conditions within an identifiable population, such as alternative surgical techniques to improve blood flow to the heart muscle in patients with coronary artery disease. QALYs have also been applied in assessing interventions that may affect a broader population (including currently healthy individuals), such as disease screening and vaccination programs. In regulatory analysis, CEA is likely to involve the aggregation and synthesis of diverse health conditions, including, for example, acute and chronic effects that result to varying degrees in preventable mortality. In addition, regulatory analysis usually involves predicting outcomes that include small risk reductions spread throughout a large population. As a result, the assessment often focuses on “statistical” cases representing the aggregation of changes in risks across many individuals.13 The most obvious difference between these HALY or QALY measures and the WTP measures used in BCA is that the latter use dollars to represent the value placed on different health outcomes as well as other (nonhealth) impacts, whereas the former focus on index measures of HRQL and longevity. As a result, an individual’s WTP may be constrained by his or her available resources, whereas HALY measures are, by design, independent of individuals’ income or wealth. In practice, however, the income or wealth term is not always statistically significant in WTP studies and HALY measures may be influenced by individuals’ income or wealth.14 These measures also differ in a number of other important respects. One key difference is that WTP indicates the extent to which individuals are willing to make trade-offs between different uses of resources, while HALY measures restrict the trade-offs to alternative health states. As discussed in Chapter 3, some commonly used HALY measures assume that the value placed on a given health state in the abstract does not depend on the duration of that health state; that is, regardless of whether the illness or impairment lasts for one day or several years, the index value is the same. In contrast, WTP studies can be constructed to include duration as one of the attributes addressed in valuation. 13   For example, a regulation that prevents a risk of 1 in 10,000 from affecting a population of 10,000 individuals would prevent one statistical case: 1/10,000*10,000 = 1. 14   More extensive discussion of the differences between HALY measures and WTP is provided in Hammitt (2002) and Krupnick (2004).

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis These and other restrictions mean that, although HALY measures are based on surveys reflecting individual choices, these choices may not fully reflect individual preferences and are not entirely consistent with the tenets of utility theory that underlie welfare economics. Although HALY measures can be constructed in a way that at least partially reflects individual preferences, such measures often do not fully conform to the concept of utility as understood in economic theory.15 Calculation of Cost-Effectiveness In CEA, the result of the analysis of costs and benefits is a ratio: the cost per unit of effect (e.g., the cost per QALY). When multiple strategies achieve the same outcome (e.g., each leads to the same reduction in the number of cases of heart disease per year), the cost-effectiveness ratios can be used to identify the most economically efficient approach, that is, the option that achieves the goal at the lowest cost. However, regulatory options usually differ in the amount of both costs and benefits, and there is no general agreement on a standard for using CEA to select among the options. The one exception is where an option is dominated; that is, at least one other option under consideration is both more effective and less costly. In contrast, the decision criteria for BCA suggest that projects with negative net benefits (costs that exceed benefits) should not be pursued, and that the option that leads to the highest net benefits is the most economically efficient. If the goal of an analysis is to allocate health care or other resources within a fixed budget, CEA can be used to inform the allocation of the budget across health-improving interventions and services. In the context of regulatory programs, there is no “regulatory budget” per se. Instead, decision makers are choosing among different options for pursuing a particular statutory or policy goal. Some have argued for applying a cost limit as a guide when using CEA to allocate resources or to establish a threshold for policy action. For example, a number of values have been advocated as cost-per-QALY dollar thresholds that differentiate a worthwhile expenditure on a new medical technology or therapy from a poor one. Phelps and Mushlin (1991) have argued that, if such a threshold value is applied, CEA performs similarly to BCA; however, the assumptions necessary for complete equivalence are fairly restrictive. A wide range of limiting or threshold investments per 15   CEAs using HALY metrics as the effectiveness measure are sometimes referred to as cost–utility analyses. This report uses the term “CEA” exclusively so as to not rely on assumptions about the consistency of HALY measures with the basic tenets of utility theory.

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis QALY have been proposed or cited as implicitly operative: from $50,000 to $160,000 and much higher per QALY (Hirth et al., 2000). There is no sanctioned, widely accepted threshold, however, at this time (Laufer, 2005; Neumann, 2005). In the regulatory context, the cost-effectiveness ratio is useful in exploring the incremental effects of investing additional resources to address a particular problem or reduce a certain type of risk.16 In addition to providing insights into the selection of regulatory options, these ratios have been used in the types of league tables and scorecards described earlier to identify the relative cost-effectiveness of different regulatory and nonregulatory interventions, particularly in terms of the costs per life or life year saved. These ratios can be reviewed by decision makers to determine whether additional increments of investment are worthwhile, and to identify the types of interventions that appear most cost-effective. Subsequent decisions ultimately require the exercise of judgment, due to the lack of consensus on the “worth” of the additional health benefits achieved, and the need to consider factors not included in the quantitative analysis. CEA is similar to BCA in that it provides a useful framework for collecting, analyzing, organizing, and reporting information on the impacts of regulatory options. It has the advantage of avoiding the need to assign monetary values to health and safety impacts, and thus may be more palatable to those who are uncomfortable with monetization of these types of benefits.17 However, the decision to implement a regulation ultimately involves commitment to a predicted level of expenditure or cost, implicitly assigning a monetary value to the quantified and nonquantified benefits. CEA also faces many of the same challenges as BCA. HALY measures are based on survey techniques that use different types of instruments, but otherwise face some of the difficulties encountered in the use of stated preference studies to estimate WTP, as discussed in Chapters 2 and 3. In addition, as traditionally practiced, both CEA and BCA focus on economic efficiency and confront decision makers with the need for supplemental information on distributional effects and equity issues. As with all assessment methods, CEA and BCA both produce somewhat incomplete and uncertain quantitative estimates because of gaps in the underlying research base. However, well-structured economic analyses of both types can help to identify the sources and extent of uncertainties and indicate the types of research that would be most useful. 16   For more detailed discussions of the calculation and interpretation of incremental cost-effectiveness ratios, see Drummond et al. (1997), Hunink et al. (2001), and Fenwick et al. (2004). 17   For more detailed critiques of BCA, see, for example, Ashford (1980), Lave (1996), and Kopp et al. (1997).

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis Summary Prospective analysis of the predicted economic effects of regulatory interventions helps participants in the regulatory development process anticipate and weigh the likely consequences of possible courses of action. Both BCA and CEA provide a structured framework for conducting research on possible impacts and presenting the results. Such research can affect not only the policy decision, but also public perceptions and understanding of both the policy process and the action taken. BCA and CEA both have the potential to improve the accountability of government by presenting information about the costs and benefits of a proposed intervention (and the reasons bearing on the ultimate decision) in comprehensible, transparent, and comprehensive terms. BCA and CEA provide complementary measures of outcomes. BCA focuses on determining the net benefits of an action, that is, which regulatory option provides the greatest increase in welfare, net of any harm or cost imposed. BCAs generally use estimates of WTP to value both health-related impacts and other effects on welfare, such as environmental effects. In contrast, CEA focuses on the cost per unit of benefit, that is, the ratio of costs relative to a unit of improvement. CEAs based on HALY measures account for both the impacts on HRQL and longevity. Both types of analysis usually also report single-dimension measures of impacts, such as cases of illness or injury averted, or years of life extended. As typically implemented, both approaches focus primarily on economic efficiency and must be supplemented by other sources of information. In particular, most policy makers are concerned with the distributional and ethical consequences of their choices. In the case of BCA, the Pareto Principle and the hypothetical compensation criterion take as given the distribution of resources at the regulatory baseline. Some argue that this presumption implicitly endorses that distribution as a just or fair starting point. Although CEA avoids this problem in part, it does not address the distributive implications of a regulatory action. The results of both BCA and CEA can be presented in disaggregate form to indicate the impact on different subgroups of concern. Ultimately, however, decision makers must engage in collective reasoning, consider additional information and the views of the public, and exercise judgment to determine whether the resulting distribution of costs and benefits argues in favor of an option that differs from the approach that provides the largest net benefits or appears most cost-effective across all groups. The incorporation of these sorts of deliberative processes is discussed in Chapter 4. Furthermore, both BCA and CEA inevitably provide somewhat incomplete and imperfect estimates of regulatory impacts. Analysts may find it difficult to quantify all the major impacts of the regulatory options, and the

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis results may be subject to significant uncertainty. Although these difficulties may reflect in part the limited time or resources available for a particular analysis, they often reflect more fundamental problems related to the status of the underlying scientific research base. For example, the relationship TABLE 1-1 Comparison of Key Features of BCA and CEA Approach for Valuing Benefits Willingness to Pay Health-Adjusted Life Year Accounts for health-related impacts Yes Yes Accounts for nonhealth impacts, such as ecological effects Yes Excluded from effect measure, may be included as an offset to costs Accounts for altruistic values or concerns about impacts on others Depends on study design Depends on elicitation method and question May be influenced by income or wealth Yes No Consistent with utility theory Yes Only under restrictive assumptions Summary Measure Net Benefits Cost-Effectiveness Ratio Decision criteria The option where benefits exceed costs by the largest amount is the most economically efficient; options with benefits less than costs should not be selected Compare the incremental cost-effectiveness of different options to determine whether an increase in the effect measure is worth the increase in cost; options that are dominated (i.e., have higher costs and lower benefits than others) should not be selected Incorporates assessment of uncertainty Yes Yes Indicates the equity or fairness of the distribution of impacts No, must be assessed separately Indicates the importance of impacts that cannot be measured in quantifiable terms No, must be assessed separately

OCR for page 15
Valuing Health for Regulatory Cost-Effectiveness Analysis between a specific hazard addressed by a regulation (e.g., air pollution, contaminants in food or water) and health risks may not be well understood by scientists. As a result, both BCA and CEA reports must discuss the uncertainty of the estimates and highlight important impacts that could not be quantified. Table 1-1 summarizes the key similarities and differences between BCA and CEA based on the overview contained in this chapter. Other differences in the implementation of these approaches are discussed in Chapter 2. The table focuses on what is possible under each approach; in reality, deficiencies in the research base or other factors may limit the ability of a particular analysis to include all of the features noted in the table. The Committee encountered some of these challenges in conducting the three case studies, as discussed later in the report. ORGANIZATION OF THE REPORT This chapter introduced the charge of the Committee, provided an overview of the regulatory development process, identified key features of BCA and CEA, and noted their important differences. This last section outlines the organization of the remainder of the report. Chapter 2 discusses current practices for the conduct of regulatory economic analyses and reviews the various approaches used by individual federal agencies. Chapter 3 presents criteria for selecting integrated measures and survey instruments, reviews alternative HALY measures and HRQL survey instruments, and describes different strategies to obtain estimates for regulatory CEA. Chapter 4 reviews the aspects of risk regulation that policy makers need to consider that are not reflected in the cost-effectiveness ratios, including ethical issues related generally to CEA as well as those related to the population and risk characteristics that are not fully captured in the effectiveness measures. Chapter 5 concludes the report by presenting the Committee’s recommendations for valuing health benefits in the economic analysis of regulations, including recommendations for additional research and data collection. The appendixes to this report discuss three case studies of regulatory CEAs that the Committee conducted in collaboration with federal agency staff (Appendix A), include commonly used HRQL survey instruments (Appendix B), and provide the full text of OMB Circular A-4 (Appendix C). A glossary and list of acronyms are also included as Appendix D.