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.
1 Introduction The Clean Air Act requires the U.S. Environmental Protection Agency (EPA) to issue and review periodically the National Ambient Air Quality Stan- dards (NAAQS) for each of six criteria pollutants. One of those pollutants is ozone in the lower atmosphere.1 Although ozone was originally viewed as an urban pollutant (associated largely with such locations as Los Angeles), it is now recognized that ozone formation and transport over much larger areas result in increased exposures to humans nationwide (e.g., NRC 1991). International transport of ozone exacerbates the problem (Jacob et al. 1999). Efforts to mitigate ambient ozone involve controlling precursor emissions (nitrogen oxides [NOx] and volatile organic compounds [VOCs]) from a wide range of stationary sources (such as factories, electricity-generating facilities, and gasoline stations) and mobile sources. Although EPA is not allowed to con- sider costs and benefits when setting NAAQS, economic analyses are carried out for the proposed and final NAAQS to estimate costs and benefits expected to result from the standards. And federal agencies deciding on (generally national) actions expected to cost more than $100 million per year are required to carry out a cost-benefit analysis of alternative regulatory strategies and to assess the distribution of their impacts in different segments of society. In past assessments of the benefits of regulations to mitigate ozone, EPA has addressed the relationship between ozone exposure and respiratory disorders. EPAâs reviews since the Regulatory Impact Analysis for the 1997 NAAQS (EPA 1997a) have included the results of epidemiologic studies that link ambi- ent ozone concentrations with premature deaths in sensitivity analyses, but EPA has not included these mortality results in their primary estimates of the benefits 1 In this report ozone is used to refer to the broad array of photochemical oxidants pre- sent in ambient air. 17
18 Ambient Ozone and Mortality: Estimating Risk-Reduction Benefits of reductions in ambient ozone concentrations because of uncertainty about the appropriate interpretation of these results. Studies published since 1990 have yielded evidence in support of a rela- tionship between short-term exposure and premature death. They include city- specific time-series studies, many of which have been included in recent meta- analyses (Bell et al. 2004, 2005; Ito et al. 2005; Levy et al. 2005). However, interpretation of the evidence is complicated by ozoneâs occurrence in mixtures of other pollutants whose concentrations fluctuate in a similar manner and can be influenced by short-term meteorologic changes. It is important to understand whether it is ozone or a copollutant that is causing health effects, because steps to control the wrong agent will be both expensive and ineffective. Interpretation of the health studies results is also complicated by uncertainties that result from relying on ozone measurements from outdoor monitoring sites to estimate expo- sures of people who spend most of their time indoors. If changes in risks of premature mortality are attributed to changes in am- bient ozone concentrations in regulatory benefits assessments, EPA also needs an estimate of the monetary value of such changes in mortality risk. In previous sensitivity analyses of mortality risk reduction benefits from ozone reductions, EPA used the same value of statistical life (VSL) as that used for analysis of other mortality risks, such as those associated with particulate matter with a di- ameter less than or equal to 2.5 microns (PM2.5). Available VSL estimates are drawn largely from studies of working-age adults with average remaining life expectancies. There are many questions about the applicability of these esti- mates to mortality risks associated with ozone, which may fall disproportion- ately to an older population with shorter remaining life expectancy and more frail health status. CHARGE TO THE COMMITTEE In light of the recent evidence on ozone mortality risk and questions about its implications for benefit analysis, EPA asked the National Research Council for scientific advice on how the ozone-mortality research findings could be used in the context of health-benefit analyses associated with regulatory assessments. In response, the National Research Council established the Committee on Mortality Risk Reduction Benefits from Decreasing Tropospheric Ozone Exposure (see Appendix A). The Statement of Task to the committee (see Box 1-1) includes evaluation of the scientific and technical bases of approaches used by EPA for estimating reductions in mortality risk and associated benefits of health-based ozone stan- dards over time. The committee was to assess methods for estimating reductions in premature death due to diminished short-term exposure to ozone, increases in life expectancy, economic valuation of the increased life expectancy, and asso- ciated uncertainties and their general implications for decision-making. The
Introduction 19 BOX 1-1 Statement of Task An NRC committee will evaluate the scientific and technical bases of approaches used by EPA for estimating ozone mortality risk reduction and associated benefits of health-based standards over time. It will assess methods for estimating reductions in premature mortality due to diminished short-term exposure to ozone, increases in life expectancy, economic valuation of the increased life expectancy, and associated uncertainties and their general implications for decision making. In addition, the commit- tee will recommend approaches for characterizing and communicating each of those aspects in regulatory health benefit analyses. Specifically, the committeeâs evaluation will include consideration of the following aspects: The committee will take into account the relevant reports of past NRC and IOM committees, as well as risk assessment for particulate mat- ter (PM) carried out by the EPA. (1) Relative contributions of recent studies (especially those pub- lished since 2003) for characterizing the size of the ozone-mortality effect in the context of benefits analysis. (2) Potential implications of methods used in the recent studies on reported benefits estimates (e.g., selection of data considered in the stud- ies, selection of mortality effect estimates from within the considered data, control for effect moderators [such as temporal trends], treatment of poten- tial publication bias, or other factors). Include consideration of the likely direction and magnitude of any influences that may result from choice of methods. (3) Available data and methods to account for the relative roles of potential confounders, such as copollutants, and the influence they have on estimates of ozone mortality effects. (4) The most appropriate exposure metrics for use in developing a concentration-response function designed to quantify the number of pre- mature mortalities associated with specific regulatory options. How does the choice of exposure metrics affect benefits estimation? Include consid- eration of dose estimation, particularly for sensitive groups. (5) Adequacy of a basis for estimating the likely impact on life expec- tancy from reductions in short-term daily exposures to ozone. If there is an adequate basis, provide specific recommendations, if possible, on ap- proaches for how EPA should quantitatively express the magnitude and associated uncertainties of this impact. (6) Strengths and weaknesses of the data, methods, and assump- tions employed by EPA for sensitivity analyses in the context of an ozone- mortality effect relationship. (7) Quantitative approaches for estimating the degree of uncertainty in premature mortality estimates associated with reductions in ambient (Continued)
20 Ambient Ozone and Mortality: Estimating Risk-Reduction Benefits BOX 1-1 Continued ozone concentrations. Aspects to be considered will include both parame- ter and model uncertainty in the concentration-response function, differ- ences within and between the epidemiologic and toxicologic data bases, reliance on surrogates of personal exposure, identification of sensitive populations, and other factors. Identify methods of presenting the quantita- tive uncertainty estimates. (8) Scientific approaches for assigning economic values to reduc- tions in mortality risk when the reduction in risk results in changes in life- expectancy of varying lengths. If there is an adequate basis for quantifying those changes, consider: (a) How the current understanding of premature mortality associated with short-term ozone exposure informs the most appropriate scien- tific approach to economic valuation of benefits. Consider methods to account for baseline health status, baseline and the magnitude of the change in life-expectancy, differences across various sectors of the population, uncertainty, and other relevant factors. (b) Applicability of the economics literature on mortality risk valuation and the associated methodologies developed for economic valuation of premature mortality reductions in the context of the health effects of ozone. (c) Issues that are specific to ozone and those which may be more broadly applicable to benefits assessments for other risk reduction strategies. (9) Major gaps in knowledge about ozone-mortality benefits analysis and the most promising research strategies to close those gaps. Identify any additional data, analyses, or research needed to separate the relative contributions of ozone and other gaseous or particulate components of the air pollution mix to the total short term premature mortality effect docu- mented in the literature. In assessing the methods for estimating ozone mortality risk, the committee will not develop its own estimate of such risk. In assessing the methods for economic valuation of reductions in mortality risk, the commit- tee will not itself judge the appropriateness of values assigned to a life saved or life years added. Sponsor: U.S. Environmental Protection Agency. committee was asked in the Statement of Task to consider the relative contribu- tions of recent studies (especially those published since 2003) to characterization of the size of the ozone-mortality effect in the context of benefit analysis. The
Introduction 21 recent studies include Bell et al. (2004); Bell et al. (2005); Ito et al. (2005); and Levy et al. (2005). The committee was also to consider the potential implica- tions of the methods (such as criteria for selection of data) used in the recent studies. The committee was asked specifically not to develop its own estimate of ozone mortality risk or to judge the appropriateness of values assigned to a life saved or life years added by decreases in ozone exposure. Because the commit- tee was asked to focus on human mortality risk reduction and associated benefits, it considered changes in human morbidity as an end point only to the extent that they illuminate issues related to mortality. The committee was not asked to as- sess changes in other effects of ozone (referred to as secondary effects), such as effects on agricultural crops and forests that would result from changes in ozone concentrations. In carrying out its charge, the committee considered presentations made during its public sessions and relevant scientific and technical documents pre- pared by EPA and other organizations. The committee sought to benefit from and build on the work of other advisory groups. For example, the National Re- search Council report titled Estimating the Public Health Benefits of Proposed Air Pollution Regulations evaluated methods used by EPA to estimate health benefits, primarily the analysis of mortality associated with exposure to airborne particulate matter (NRC 2002). The Institute of Medicine report Valuing Health for Regulatory Cost-effectiveness Analysis (IOM 2006) addressed analytic and policy issues, including the use of cost-effectiveness analysis to estimate health- related effects of regulatory actions. EPAâs Science Advisory Board (SAB) has completed several reviews of ozone mortality in EPAâs benefit-analysis frame- work. An SAB developed an advisory report on mortality risk valuation and how the valuation may be affected by differences in life expectancy (EPA SAB 2007). The journal Review of Environmental Economics and Policy published an issue containing a series of articles on mortality risk and valuation (Review of Environmental Economics and Policy 2007, Vol. 1). The report of a confer- ence on Critical Considerations in Evaluating Scientific Evidence of Health Ef- fects of Ambient Ozone summarized scientific issues the conference participants viewed as central for decisions to be made in setting the ozone NAAQS (Brauer et al. 2007). The committee also considered a EPA staff paper that summarized staff recommendations on the ozone NAAQS to the EPA administrator (EPA 2007a) and EPAâs regulatory impact analysis to accompany the agencyâs recent proposal for revising the NAAQS for ozone (EPA 2007b). Figure 1-1 presents a sequence of analytic methods used by EPA to assess mortality risk reduction and associated health benefits expected to result from setting and implementing ozone NAAQS (see Chapter 2). Estimation of the po- tential health effects of actions taken to improve air quality, and the economic value of those effects, requires estimation of the likely changes in emissions, ambient ozone concentrations, population exposures, and health risk and then the economic valuation of the change in risk. Figure 1-1 shows which compo- nents of the committeeâs Statement of Task correspond to specific steps in the
22 Ambient Ozone and Mortality: Estimating Risk-Reduction Benefits sequence of analytic methods. It also indicates which chapters of this report cover specific topics. Chapter 6 integrates the committeeâs overall conclusions and recommendations concerning ozone exposure, mortality risk, and benefit as- sessment. FIGURE 1-1 Methods used by EPA to assess the effects of NAAQS and control strate- gies developed to implement the standards. The left portion of the figure indicates the chapters of this report that deal with each method. Numbers in brackets correspond to relevant portions of the committeeâs statement of task, shown in Box 1-1. NOTE: The committee was not asked to and did not consider the cost aspects of benefit-cost analysis and cost-effectiveness analysis. Abbreviations: â = change; NAAQS = National Ambient Air Quality Standard.