- Email: [email protected]
Strengthening Cost-Effectiveness Analysis for Public Health Policy
Strengthening Cost-Effectiveness Analysis for Public Health Policy Louise B. Russell, PhD,1 Anushua Sinha, MD, MPH2 Although the U.S. spends more on medical care than any country in the world, Americans live shorter lives than the citizens of other high-income countries. Many important opportunities to improve this record lie outside the health sector and involve improving the conditions in which Americans live and work: safe design and maintenance of roads, bridges, train tracks, and airports; control of environmental pollutants; occupational safety; healthy buildings; a safe and healthy food supply; safe manufacture of consumer products; a healthy social environment; and others. Faced with the overwhelming array of possibilities, U.S. decision makers need help identifying those that can contribute the most to health. Cost-effectiveness analysis is designed to serve that purpose, but has mainly been used to assess interventions within the health sector. This paper briefly reviews the objective of cost-effectiveness analysis and its methodologic evolution and discusses the issues that arise when it is used to evaluate interventions that fall outside the health sector under three headings: structuring the analysis, quantifying/measuring benefits and costs, and valuing benefits and costs. (Am J Prev Med 2016;50(5S1):S6–S12) & 2016 American Journal of Preventive Medicine. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction
A
lthough the U.S. spends more on medical care than any country in the world, Americans live shorter lives than the citizens of other highincome countries. U.S. life expectancy has been falling in the international rankings for decades: As medical spending rose from 8.9% of national income in 1980 to 17.4% in 2013, life expectancy fell from about average for high-income countries to nearly the lowest.1,2 What can be done to reverse this situation? Public health has come to realize, or, perhaps more accurately, has returned to the realization,3 that health is not solely or even primarily a matter of medical care, but of the total environment in which human beings live and work. Many opportunities to improve U.S. life expectancy lie outside the health sector: safe design and maintenance of roads, bridges, train tracks, and airports; control of environmental pollutants; occupational safety; healthy buildings; a safe and healthy food supply; safe manufacture of consumer products; a healthy social environment; and others. IOM From the 1Institute for Health, Health Care Policy, and Aging Research, and Department of Economics, Rutgers University, New Brunswick, New Jersey; and 2Department of Health Systems and Policy, Rutgers School of Public Health, Rutgers University, New Brunswick, New Jersey Address correspondence to: Louise B. Russell, PhD, Institute for Health, Rutgers University, 112 Paterson Street, New Brunswick NJ 08091. E-mail: [email protected]. This article is part of the supplement issue titled The Use of Economics in Informing U.S. Public Health Policy. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2015.11.007
defines the scope of environmental health as covering “the natural and built environments at local, national, and global levels” and has issued reports addressing health topics as varied as the food supply chain, occupational health, and the relation between oceans and human health.4 Good health is not just a matter of opportunities for improving health, but also of making good choices among those opportunities. Faced with so many possibilities, Americans and the decision makers who represent them need help identifying those that contribute the most to health. Cost-effectiveness analysis (CEA), which has primarily been applied to interventions in the health sector, has proven a useful guide. The goals of public health and CEA are one and the same: to maximize the health of a population. CEA’s methodology is designed to identify investments that contribute the most to better health,5,6 and is useful for investments from prevention to treatment.7–9 Perhaps not coincidentally, CEA plays a larger part in decision making about health in many other high-income countries, all of which have longer life expectancies, than it does in the U.S.6 How can CEA be made more useful to U.S. decision makers as they consider interventions outside the health sector? Why does it need to be? This paper explores how the design of CEA might be modified to address such interventions. It begins by providing some brief background on CEA: its goal and methodologic evolution over the last several decades and the development of the concept of the reference case. It then examines the issues that arise in using CEA to evaluate potential health investments that fall largely or entirely outside the health
S6 Am J Prev Med 2016;50(5S1):S6–S12 & 2016 American Journal of Preventive Medicine Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
sector. The paper concludes with some principles to help the field move forward so that CEA can more usefully contribute to good decisions about public health investments.
Goal and Evolution of Cost-Effectiveness Analysis Both public health and CEA aim to improve the health of populations. CEA evaluates the costs and health effects of interventions and summarizes the results in costeffectiveness ratios, for example, the cost per year of life gained. A general measure of health outcome, such as life-years or quality-adjusted life-years (QALYs), makes it possible to compare very different health interventions. The attention to costs has led some to see CEA as a method of cost control, but two things required for cost control are missing: CEA does not set budgets or spending limits; and it does not tell decision makers what they can or cannot choose, but only suggests priorities. Budgets or spending limits control spending. It is easier to see that health is the goal if cost-effectiveness ratios are used to calculate the number of QALYs that could be purchased for a given outlay, say $1 million. Smokingcessation programs, for example, at $6,200 per QALY gained (2013 dollars), would yield 161 healthy years for $1 million.6,10 At $36,000 per QALY, mammograms every 2 years to screen for breast cancer in women aged 50 79 years would yield 28 healthy years.6,11 Both are productive investments in health—they yield better health. If the budget will cover both, both should be offered. CEA only suggests that, if the budget will not cover both, and if the goal is the most health for the population, it is worthwhile to consider providing smoking cessation rather than breast cancer screening. CEA suggests priorities for situations in which not everything can be purchased. In the course of its development over the last 40 years, CEA has built on advances in many other fields. New developments in simulation modeling, one of CEA’s basic tools, and standards for simulation models, including uncertainty analysis, have been important to improving the quality of analyses.12,13 The methodologies of systematic review and meta-analysis have contributed to better estimates of the parameters needed for CEA models. Preferred Reporting Items for Systematic Reviews and Meta-Analyses14 and the Cochrane Community Reviews15 have been particularly influential, as have reports like IOM’s Finding What Works in Health Care: Standards for Systematic Reviews.16 One of the most important advances in CEA has been the organizing concept of the reference case. As the example of smoking cessation and mammograms illustrates, the point of CEA is to compare alternatives and choose those that May 2016
S7
contribute most to health. Analyses will necessarily differ because of the nature of the disease, the outcomes associated with it, the nature and costs of possible interventions, and the ways in which the features of the disease and the interventions vary from one jurisdiction, or one group of patients, to another. But they should not differ because of methodologic choices that are not necessary to reflect those real and important differences. A consistent and comprehensive set of standards for CEAs helps decision makers reach the right decisions. The reference case concept has been developed to meet the needs of decision makers around the world. A partial list of groups that have established reference case standards for the conduct of CEAs includes: the U.S. Panel on Cost-Effectiveness in Health and Medicine,5 the United Kingdom’s National Institute for Health and Clinical Excellence,17 the Canadian Agency for Drugs and Technologies in Health,18 the Australian Pharmaceutical Benefits Advisory Committee,19 WHO,20 and the Bill and Melinda Gates Foundation.21 No two reference cases are the same, but they share common elements: 1. Reference cases typically focus on the costs of the formal health sector—sometimes the entire formal health sector, more often the costs of a particular payer or payers. Though some recommend including costs and benefits outside the health sector, they provide little guidance on how to do so. 2. Reference cases typically recommend the use of a summary measure of health that allows the health consequences of different diseases and interventions to be compared. High-income countries use one of several systems for calculating QALYs gained from an intervention.5 Low- and middle-income countries generally use disability-adjusted life-years averted.22 3. Because modeling is an essential tool in CEAs, reference cases usually discuss standards for modeling, such as the range of interventions to be modeled, the time horizon for the analysis, and methods for exploring uncertainty in the data used to inform model parameters. 4. The groups that have developed reference cases agree that CEA is one element in decision making. Decisions may involve other factors—ethical issues or feasibility and logistic factors—that lead to choices that do not follow the strict ordering of options presented in a CEA.
Issues Raised by Interventions Outside the Health Sector Reference cases, which currently focus on the health sector, need to be revised to provide guidance when CEA is used to evaluate broader public health interventions.
S8
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
The natural scope for CEAs of public health interventions is a comprehensive perspective that attempts to measure all benefits and costs, no matter whether they occur inside or outside the health sector. Outside the health sector, this kind of comprehensive analysis has mainly been conducted using cost benefit analysis (CBA). CBA differs from CEA only in that it applies a monetary valuation to all outcomes, not just resources used; any CEA can be turned into a CBA by multiplying the health outcomes by a monetary valuation. There is a large literature on the methods used to derive such monetary values. The final summary of a CBA is net monetary benefit: the monetary value of all benefits minus the monetary value of all resources used.23,24 Within the field of CEA, however, with its tradition of focusing on health and the health sector, there is no general consensus on what this comprehensive perspective is or how to apply it. The U.S. Panel on CostEffectiveness in Health and Medicine, for example, recommended that benefits and costs outside the health sector be included in CEA, but gave little guidance how to do that. IOM, the Second Panel on Cost-Effectiveness Analysis in Health and Medicine, and the National Institute for Health and Clinical Excellence, among others, have debated more specific guidance for a comprehensive perspective.25,26 A comprehensive perspective raises issues at every step of an analysis: structuring the analysis, measuring costs and benefits in natural units, and valuing costs and benefits. Keeping the steps conceptually separate is important to developing good guidance. Combining steps can lead to errors, particularly at the later steps, where measuring and valuing can sometimes be confused with each other.
Structuring the Analysis Structuring an analysis encompasses tasks familiar to any CEA analyst, among them specifying the target population, the alternatives to be evaluated, the costs and benefits to measure, and the time horizon over which to evaluate them.5 Identifying costs and benefits outside the health sector, unfamiliar territory for CEA analysts, can be difficult. At the identification step, analysts can use the strategy that has served them well for health interventions: review the literature and consult a range of subject experts. A 2012 IOM report, An Integrated Framework for Assessing the Value of Community-Based Prevention,25 addresses the issue of identifying all important consequences. It states that: “[t]he valuation of communitybased prevention interventions should be done with a comprehensive perspective; that is, the measurement of
benefits, harms, and resources should include impacts on all members of the community as well as on stakeholders who may be outside the community.” The first recommendation identifies three broad and mutually exclusive categories of benefits and costs: “The committee recommends that those seeking to assign value to communitybased prevention interventions take a comprehensive view that includes the benefits and harms in the three major domains of health, community well-being, and community process as well as the resource use associated with such interventions.” The IOM report explains: Community-based prevention can create value not only through improvements in the health of individuals but also by increasing the investment that individuals are willing and able to make in themselves, in their family and neighbors, and in their environment. Furthermore, community-based prevention involves decisions among groups of people about how to live in society, how the physical environment is built, what food is served in schools, and so on. Thus, the process by which interventions are decided upon and undertaken needs to be treated as a valued outcome. If a community decides to tell people what they can or cannot do, or what they should or should not do, the decisions need to have the legitimacy—the added value—that comes from an open and inclusive group decision-making process. No accepted measures exist for the second two classes of benefit. The committee recommended that the Centers for Disease Control and Prevention (CDC) develop some. A simple example can illustrate the three domains identified by the IOM report.25 Suppose a community invests in a park with trails and night-time lighting to encourage people to walk and engage in other physical activity for their health. The list of possible benefits certainly includes health, which the IOM report recommended be measured in QALYs, but the park can also bring aesthetic benefits (people can enjoy the park’s beauty whether they walk or not) and safety benefits (if the lighting reduces crime in the area). If decisions about the park are made with community involvement, to ensure that the community contributes to the plans, the benefits can include a greater sense of cohesion and pride on the part of community members. Cost benefit analysts routinely face the same issue. As noted, the perspective usually taken in a CBA is comprehensive—the societal perspective—and encompasses all significant benefits or costs that occur in the jurisdiction of the decision maker. For example, in its 2011 evaluation of the Clean Air Act Amendments of 1990, which estimated costs and benefits over the www.ajpmonline.org
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
1992 2020 period, the Environmental Protection Agency valued reductions in mortality and illness; improvements in visibility at recreational sites and in residential areas; benefits to commercial timber, agricultural crops, and recreational fishing; and reduced materials damage.27
Measuring Benefits and Costs One of CEA’s major contributions to decision making is that it not only lists the benefits and costs of an intervention, but also measures their magnitude. Quantifying them gives a clearer idea of their size and importance. The measurement process uses “natural units,” such as familiar health metrics—cases of disease, deaths prevented, and life-years and QALYs gained by an intervention. Thus, the next step is to select metrics that allow each cost or benefit to be measured. Interventions that take place largely or entirely outside the health sector can involve metrics new to CEA analysts as they seek to quantify the intervention’s effects on the aesthetic environment, crime and safety, and others. At this stage, costs (resources used by the intervention) are also best measured in physical units if data are available. As noted, existing measures for many benefits and costs, such as levels of smog and air pollution, rates of property and violent crime, test scores and years of schooling, housing quality, and traffic safety, can be identified by experts in each subject area. Analysts also need to measure all the resources used by an intervention. Perhaps even more than interventions that take place in the health sector, those that take place largely or entirely outside it can involve substantial amounts of donated resources. The temptation is to consider these resources as costing nothing—as free— because no check changes hands for them. Donated time, goods, and services are not free unless they would have gone completely unused had this project not come along. The Community Preventive Services Task Force makes this point, correctly noting that all resources, whether purchased or donated, should be included in costs.28 Consider the park example again. People may donate trees and plants for the landscaping. They may volunteer their time for planning, carrying out the plans, and maintaining the park once it has been created. Some businesses encourage employees to spend paid work time on community activities. Donated time and goods and services are as much resource costs of the project as are purchased labor and materials. Valuing and Summarizing Once benefits and costs are measured in natural units, how should they be valued? Why should they be valued? May 2016
S9
Or is a detailed table listing benefits and costs in natural units enough? Valuation involves weighting benefits and resources after they have been measured in natural units so that they can be combined and summarized. To make summarization possible, it attaches a second number to each that reflects people’s preferences: A higher number indicates a higher value. Using a consistent principle for valuation, and a common unit of measure for the values, as CBA does when it uses the principle of willingness to pay (WTP) to translate benefits and costs into dollars, allows benefits and costs to be combined in an overall summary. For a given decision, the detailed table of costs and benefits in natural units can and should be presented to decision makers, but they may also find summary measures helpful for comparing choices. In CEA, there are two summary measures, one for health and one for resources used. In CBA, a single summary measure, net benefit, is possible because both benefits and resources are valued in monetary terms. Valuing benefits and costs in order to summarize them can be the most difficult part of an analysis. It can be easy to value some benefits twice, such as the portion of health used for paid work, which then biases the summary in favor of interventions that produce particular kinds of health or health for particular kinds of people. The tendency to count donations as costing nothing biases the summary in another way, toward interventions that involve donated time, goods, and services. Finding or developing appropriate measures of value, measures that reflect how the community values benefits and costs, can also be difficult. In thinking about how to value benefits and costs, CEA can learn from the experience of CBA. For example, CBA used to value health as the additional earnings healthier people could make.29 That tradition is still widely influential outside of economics, but economists began to realize in the 1960s and 1970s that labor market earnings were the wrong way to value health because that is not how people themselves value it. For example, earnings assign no value to other activities made possible by better health— taking care of children or the elderly, maintaining the house, socializing, sleeping—although people clearly do value these activities. By the 1980s, economists agreed that the basic economic principle of WTP was the right way to approach valuing life and health and the concept “that social decisions should, so far as possible, reflect the interests, preferences and attitudes to risk of those who are likely to be affected by the decisions”—had been “extensively developed in the literature.”30 The Environmental Protection Agency was already applying the principle in the mid-1980s.31 In its CBA standards, the U.S. Office of Management and Budget states: “In
S10
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
monetizing health benefits, a WTP measure is the conceptually appropriate measure as compared to other alternatives (e.g., cost of illness or lifetime earnings), in part because it attempts to capture pain and suffering and other quality-of-life effects.”32 How can WTP be used to value life and health? The answer begins with the recognition that people act on their values all the time, revealing them through the choices they make. Some of those choices occur in markets, where people reveal how much they value a good or service through the amount they are willing to spend for it. Life and health are not sold in markets, but economists realized that goods and services sold in markets often reflect people’s valuation of non-market goods. Purchases of safety equipment to prevent accidents, for example, reveal a WTP for better health and longer life. Similarly, higher pay is required for dangerous jobs because workers value life and health.33 Economists have used this “risk premium,” the extra money per hour of work, to calculate the “value of a statistical life” (VSL; the value of a statistical life-year, VSLY, can be derived from VSL32) and have found that it implies a high valuation on life and health. Based on wage-risk studies, the Environmental Protection Agency currently uses a VSL of $7.4 million (2006 dollars) to value reductions in mortality due to less pollution. The WTP principle explains an apparent paradox first noted when labor market earnings were still used to value health: People were often willing to contribute large sums to rescue others in distress, sums much larger than the earnings used to value life saving in CBA. The explanation proposed at the time was that people valued “identified” lives—real people in real crises—more highly than they valued “statistical” lives, the future and unidentified lives that would be saved by, say, introducing safety gear on the job or repairing bridges. Once, however, economists began to estimate what people were willing to pay for life saving, VSL turned out to be much higher than lost earnings and in line with the amounts they contribute to save identified lives.34 Another principle borrowed from CBA is that the resources used by an intervention should be valued at their opportunity cost, which reflects the value of the goods and services that would be produced if the resources were put to their next-best use. In effect, valuation of resources is also based on the principle of WTP. In perfectly functioning markets—many buyers and sellers, each fully informed about the product, where benefits accrue only to buyers, costs only to sellers— prices would reflect this opportunity cost. Even imperfect market prices are often good proxies for opportunity costs. When prices are not available, as for resources that are not sold in markets, market prices for other goods
can, again, often offer evidence about the value people place on those resources. This explains economists’ practice of valuing time spent in activities other than paid employment at the wage for paid work: By choosing unpaid activities, people indicate that they value time spent on those activities at least as highly as the wage they could earn with that time.35 A distinction is sometimes drawn between the valuation of things sold in markets and those not sold in markets,28 but the economic principle does not make such a distinction. Both kinds of resources are valued at their opportunity costs and have thus been translated into comparable terms using a consistent principle.
Conclusions Where does this point us? Just as CEA has adopted helpful CBA practices in the past, it can benefit from considering the way CBA handles the valuation of benefits and costs that fall outside the health sector as it begins to develop standards for this kind of application. There is no ready-made solution available and, in any event, the solution must be developed out of a consensus among analysts and decision makers. The following principles, however, may help lead to a good solution: 1. In democratic societies, outcomes should be valued as the affected people value them. CEA, which derives QALY weights from surveys, and CBA, which derives monetary values from market choices (and, when those are lacking, surveys), agree on this principle. 2. As it has in the past, the experience of CBA can help CEA analysts develop ways to value more outcomes in monetary terms and do so in a consistent way. But, it is important that CEA analysts inform themselves of current thinking in CBA and consider its implications before developing general recommendations. 3. CBA has the strength that everything can be monetized, so that the sum of costs can be deducted from the sum of benefits to show whether an intervention is beneficial when the potentially wide array of benefits and costs involved in non health sector interventions is considered. CEA has the strength that it focuses directly on health, the outcome of primary interest, using a measure of health. For now, it seems best to follow the advice of the U.S. Office of Management and Budget and do both, in order to give decision makers a full pictures of the consequences of an intervention.32,36 4. In either form of analysis, benefits and costs should always be presented in a disaggregated www.ajpmonline.org
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
way, so that decision makers can clearly see the components. 5. CEA summarizes results with cost-effectiveness ratios, CBA with net monetary benefit. Both can calculate those summaries in a variety of ways when there are many benefits and costs involved. For example, CEA can calculate the health sector cost per QALY, the total cost per QALY, and the net cost (costs less monetized non-health benefits) per QALY. CBA can calculate net benefits for health effects and health sector costs or for wider sets of benefits and costs; for example, in the case of the park, including only aesthetic benefits in one summary and both crime reduction and aesthetic benefits in another. CEA analysts and decision makers need to explore the range of possible summary measures before arriving at a consensus, if one is possible, about those that are conceptually consistent and best serve the needs of public health decision makers.
Publication of this article was supported by the Centers for Disease Control and Prevention. We thank the four reviewers and guest editors of the supplement for helpful comments on an earlier version of the paper. Preparation of this paper was not supported by any grant or contract. Both authors contributed to the conceptualization of the paper, development of ideas, and critical review and final approval of the draft; LBR took primary responsibility for drafting the paper. The authors have no financial conflicts of interest in connection with this paper and received no external funding to write it. Dr. Russell is a member of the leadership group of the Second Panel on Cost-Effectiveness in Health and Medicine.
References 1. NHE Summary, including share of GDP, CY 1960-2013, Table 1. http:// www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trendsand-Reports/NationalHealthExpendData/nationalHealthAccountsHis torical.html. Accessed July 9, 2015. 2. Woolf SH, Aron L. Panel on Understanding Cross-National Health Differences Among High-Income Countries; Committee on Population; Division of Behavioral and Social Sciences and Education; National Research Council; Board on Population Health and Public Health Practice; Institute of Medicine. U.S. Health in International Perspective: Shorter Lives, Poorer Health. Washington, DC: The National Academies Press; 2013. 3. Teutsch S, Fielding JE. Rediscovering the core of public health. Front Public Health Serv Syst Res. 2013;2(3):1–7, Article 1.
May 2016
S11
4. IOM, National Academy of Sciences, Engineering, Medicine. Topic: environmental health. http://iom.nationalacademies.org/Global/Topics/ Environmental-Health.aspx. Accessed July 15, 2015. 5. Gold MR, Siegel JE, Russell LB, Weinstein MC, ed. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996. 6. Russell LB. The science of making better decisions about health: costeffectiveness and cost-benefit analysis. In: Kaplan RM, David D, Spittel M, eds. Emerging Behavioral and Social Science Perspectives on Population Health. Washington, DC: U.S. Government Printing Office; 2015. 7. Russell LB. Is Prevention Better than Cure? Washington, DC: The Brookings Institution; 1986. 8. Cohen JT, Neumann PJ, Weinstein MC. Does preventive care save money? Health economics and the presidential candidates. N Engl J Med. 2008;358(7):661–663. http://dx.doi.org/10.1056/NEJMp0708558. 9. Russell LB. Preventing chronic disease: an important investment, but don’t count on cost savings. Health Aff (Millwood). 2009;28(1):42–45. http://dx.doi.org/10.1377/hlthaff.28.1.42. 10. Cromwell J, Bartosch WJ, Fiore MC, et al. Cost-effectiveness of the clinical practice recommendations in the AHCPR guideline for smoking cessation. JAMA. 1997;278(21):1759–1766. http://dx.doi. org/10.1001/jama.1997.03550210057039. 11. Mandelblatt J, Saha S, Teutsch S, et al. The cost-effectiveness of screening mammography beyond age 65 years: a systematic review for the U.S. Preventive Services Task Force. Ann Intern Med. 2003;139(10):835–842. http://dx.doi.org/10.7326/0003-4819-139-10-200311180-00011. 12. National Cancer Institute. Cancer Intervention and Surveillance Modeling Network: Modeling to guide public health research and priorities. http://cisnet.cancer.gov/about/. Accessed March 10, 2015. 13. Caro JJ, Briggs AH, Siebert U, Kuntz KM. Modeling Good Research Practices—Overview: A Report of the ISPOR-SMDM Modeling Good Research Practices Task Force 1. Med Decis Making. 2012;32(5): 667–677. http://dx.doi.org/10.1177/0272989X12454577. 14. PRISMA (Transparent Reporting of Systematic Reviews and MetaAnalyses). www.prisma-statement.org/. Accessed July 7, 2015. 15. Cochrane Community Reviews. www.cochrane.org/cochrane-reviews. Accessed July 7, 2015. 16. IOM. Finding What Works in Health Care: Standards for Systematic Reviews. www.iom.edu/Reports/2011/Finding-What-Works-in-HealthCare-Standards-for-systematic-Reviews.aspx. Published 2011. Accessed July 7, 2015. 17. National Institute for Health and Care Excellence (NICE). Guide to the methods of technology appraisal. Published 2013. http://publications. nice.org.uk/pmg9. April 4, 2013. 18. Canadian Agency for Drugs and Technologies in Health (CADTH). Guidelines for the economic evaluation of health technologies: Canada. 3rd ed. Ottawa: Canadian Agency for Drugs and Technologies in Health; 2006. 19. Australian Government, Department of Health and Ageing, Pharmaceutical Benefits Advisory Commission (PBAC). Guidelines for Preparing Submissions to the Pharmaceutical Benefits Advisory Committee (Version 4.3). http://www.pbac.pbs.gov.au/content/infor mation/archived-versions/pbac-guidelines-v4.3-2008.pdf. Published December 2008. Accessed November 19, 2015. 20. WHO. WHO guide to cost-effectiveness analysis. Published 2003. www.who.int/choice/publications/p_2003_generalised_cea.pdf. 21. Bill and Melinda Gates Foundation. Methods for Economic Evaluation Project (MEEP) Final Report. A partnership between Bill and Melinda Gates Foundation, NICE International, the Health Intervention and Technology Assessment Program (Thailand), and the University of York, Centre for Health Economics. https://www.nice.org.uk/Media/ Default/About/what-we-do/NICE-International/projects/MEEP-report. pdf. January 2014. Accessed November 19, 2015. 22. Salomon JA, Vos T, Hogan DR, et al. Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):2129– 2143. http://dx.doi.org/10.1016/S0140-6736(12)61680-8.
S12
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
23. Boardman AE, Greenberg DH, Vining AR, Weimer DL. Cost-Benefit Analysis: Concept and Practice. New York NY: Prentice Hall, 2011. 24. Freeman III. The Measurement of Environmental and Resource Values: Theory and Methods. Second Edition, Washington, DC: Resources for the Future; 2003. 25. IOM. An Integrated Framework for Assessing the Value of CommunityBased Prevention. Washington, DC: The National Academies Press; 2012. 26. Social Value Judgements: Principles for the Development of NICE Guidance. Second Edition. National Institute for Health and Clinical Excellence website. http://www.nice.org.uk/proxy/?sourceUrl=http%3a% 2f%2fwww.nice.org.uk%2fmedia%2fC18%2f30%2fSVJ2PUBLICAT ION2008.pdf. Published 2008. Accessed November 19, 2015. 27. U.S. Environmental Protection Agency, Office of Air and Radiation. The Benefits and Costs of the Clean Air Act from 1990 to 2020: Summary Report. Washington, DC: U.S. Environmental Protection Agency, March 2011. 28. Stone GA, Hutchinson AB, Corso PS, et al. Chapter 11: understanding and using the economic evidence. In: The Guide to Community Preventive Services: What Works to Promote Health? www.thecommu nityguide.org/library/economics.pdf. Accessed June 8, 2015. 29. Jone-Lee MW. The Value of Life: An Economic Analysis. Chicago, IL: The University of Chicago Press; 1976.
30. Jones-Lee MW, Hammerton M, Philips PR. The value of safety: results of a national sample survey. Econ J. 1985;95(377):49–72. http://dx.doi. org/10.2307/2233468. 31. Nichols AL. Lead in gasoline. In: Morgenstern RD, Economic Analyses at EPA: Assessing Regulatory Impact. Washington, DC: Resources for the Future, 1997:49–86. 32. U.S. Office of Management and Budget, Circular A-4, Subject: Regulatory Analysis, September 17, 2003. www.whitehouse.gov/omb/ circulars_a004_a-4/. Accessed July 14, 2015. 33. Robinson LA. How US government agencies value mortality risk reductions. Rev Environ Econ Policy. 2007;1(2):283–299. http://dx.doi. org/10.1093/reep/rem018. 34. Russell LB. Do we really value identified lives more highly than statistical lives? Med Decis Making. 2014;34(5):556–559. http://dx.doi. org/10.1177/0272989X13512183. 35. Russell LB. Completing costs: patients’ time. Med Care. 2009; 47(7 Suppl 1):S89–S93. http://dx.doi.org/10.1097/MLR.0b013e31819bc077. 36. Miller W, Robinson LA, Lawrence RS. Committee to Evaluate Measures of Health Benefits for Environmental, Health, and Safety Regulation, eds. Valuing Health for Regulatory CostEffectiveness Analysis. Washington, DC: The National Academies Press; 2006.
www.ajpmonline.org
Introduction
A
lthough the U.S. spends more on medical care than any country in the world, Americans live shorter lives than the citizens of other highincome countries. U.S. life expectancy has been falling in the international rankings for decades: As medical spending rose from 8.9% of national income in 1980 to 17.4% in 2013, life expectancy fell from about average for high-income countries to nearly the lowest.1,2 What can be done to reverse this situation? Public health has come to realize, or, perhaps more accurately, has returned to the realization,3 that health is not solely or even primarily a matter of medical care, but of the total environment in which human beings live and work. Many opportunities to improve U.S. life expectancy lie outside the health sector: safe design and maintenance of roads, bridges, train tracks, and airports; control of environmental pollutants; occupational safety; healthy buildings; a safe and healthy food supply; safe manufacture of consumer products; a healthy social environment; and others. IOM From the 1Institute for Health, Health Care Policy, and Aging Research, and Department of Economics, Rutgers University, New Brunswick, New Jersey; and 2Department of Health Systems and Policy, Rutgers School of Public Health, Rutgers University, New Brunswick, New Jersey Address correspondence to: Louise B. Russell, PhD, Institute for Health, Rutgers University, 112 Paterson Street, New Brunswick NJ 08091. E-mail: [email protected]. This article is part of the supplement issue titled The Use of Economics in Informing U.S. Public Health Policy. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2015.11.007
defines the scope of environmental health as covering “the natural and built environments at local, national, and global levels” and has issued reports addressing health topics as varied as the food supply chain, occupational health, and the relation between oceans and human health.4 Good health is not just a matter of opportunities for improving health, but also of making good choices among those opportunities. Faced with so many possibilities, Americans and the decision makers who represent them need help identifying those that contribute the most to health. Cost-effectiveness analysis (CEA), which has primarily been applied to interventions in the health sector, has proven a useful guide. The goals of public health and CEA are one and the same: to maximize the health of a population. CEA’s methodology is designed to identify investments that contribute the most to better health,5,6 and is useful for investments from prevention to treatment.7–9 Perhaps not coincidentally, CEA plays a larger part in decision making about health in many other high-income countries, all of which have longer life expectancies, than it does in the U.S.6 How can CEA be made more useful to U.S. decision makers as they consider interventions outside the health sector? Why does it need to be? This paper explores how the design of CEA might be modified to address such interventions. It begins by providing some brief background on CEA: its goal and methodologic evolution over the last several decades and the development of the concept of the reference case. It then examines the issues that arise in using CEA to evaluate potential health investments that fall largely or entirely outside the health
S6 Am J Prev Med 2016;50(5S1):S6–S12 & 2016 American Journal of Preventive Medicine Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
sector. The paper concludes with some principles to help the field move forward so that CEA can more usefully contribute to good decisions about public health investments.
Goal and Evolution of Cost-Effectiveness Analysis Both public health and CEA aim to improve the health of populations. CEA evaluates the costs and health effects of interventions and summarizes the results in costeffectiveness ratios, for example, the cost per year of life gained. A general measure of health outcome, such as life-years or quality-adjusted life-years (QALYs), makes it possible to compare very different health interventions. The attention to costs has led some to see CEA as a method of cost control, but two things required for cost control are missing: CEA does not set budgets or spending limits; and it does not tell decision makers what they can or cannot choose, but only suggests priorities. Budgets or spending limits control spending. It is easier to see that health is the goal if cost-effectiveness ratios are used to calculate the number of QALYs that could be purchased for a given outlay, say $1 million. Smokingcessation programs, for example, at $6,200 per QALY gained (2013 dollars), would yield 161 healthy years for $1 million.6,10 At $36,000 per QALY, mammograms every 2 years to screen for breast cancer in women aged 50 79 years would yield 28 healthy years.6,11 Both are productive investments in health—they yield better health. If the budget will cover both, both should be offered. CEA only suggests that, if the budget will not cover both, and if the goal is the most health for the population, it is worthwhile to consider providing smoking cessation rather than breast cancer screening. CEA suggests priorities for situations in which not everything can be purchased. In the course of its development over the last 40 years, CEA has built on advances in many other fields. New developments in simulation modeling, one of CEA’s basic tools, and standards for simulation models, including uncertainty analysis, have been important to improving the quality of analyses.12,13 The methodologies of systematic review and meta-analysis have contributed to better estimates of the parameters needed for CEA models. Preferred Reporting Items for Systematic Reviews and Meta-Analyses14 and the Cochrane Community Reviews15 have been particularly influential, as have reports like IOM’s Finding What Works in Health Care: Standards for Systematic Reviews.16 One of the most important advances in CEA has been the organizing concept of the reference case. As the example of smoking cessation and mammograms illustrates, the point of CEA is to compare alternatives and choose those that May 2016
S7
contribute most to health. Analyses will necessarily differ because of the nature of the disease, the outcomes associated with it, the nature and costs of possible interventions, and the ways in which the features of the disease and the interventions vary from one jurisdiction, or one group of patients, to another. But they should not differ because of methodologic choices that are not necessary to reflect those real and important differences. A consistent and comprehensive set of standards for CEAs helps decision makers reach the right decisions. The reference case concept has been developed to meet the needs of decision makers around the world. A partial list of groups that have established reference case standards for the conduct of CEAs includes: the U.S. Panel on Cost-Effectiveness in Health and Medicine,5 the United Kingdom’s National Institute for Health and Clinical Excellence,17 the Canadian Agency for Drugs and Technologies in Health,18 the Australian Pharmaceutical Benefits Advisory Committee,19 WHO,20 and the Bill and Melinda Gates Foundation.21 No two reference cases are the same, but they share common elements: 1. Reference cases typically focus on the costs of the formal health sector—sometimes the entire formal health sector, more often the costs of a particular payer or payers. Though some recommend including costs and benefits outside the health sector, they provide little guidance on how to do so. 2. Reference cases typically recommend the use of a summary measure of health that allows the health consequences of different diseases and interventions to be compared. High-income countries use one of several systems for calculating QALYs gained from an intervention.5 Low- and middle-income countries generally use disability-adjusted life-years averted.22 3. Because modeling is an essential tool in CEAs, reference cases usually discuss standards for modeling, such as the range of interventions to be modeled, the time horizon for the analysis, and methods for exploring uncertainty in the data used to inform model parameters. 4. The groups that have developed reference cases agree that CEA is one element in decision making. Decisions may involve other factors—ethical issues or feasibility and logistic factors—that lead to choices that do not follow the strict ordering of options presented in a CEA.
Issues Raised by Interventions Outside the Health Sector Reference cases, which currently focus on the health sector, need to be revised to provide guidance when CEA is used to evaluate broader public health interventions.
S8
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
The natural scope for CEAs of public health interventions is a comprehensive perspective that attempts to measure all benefits and costs, no matter whether they occur inside or outside the health sector. Outside the health sector, this kind of comprehensive analysis has mainly been conducted using cost benefit analysis (CBA). CBA differs from CEA only in that it applies a monetary valuation to all outcomes, not just resources used; any CEA can be turned into a CBA by multiplying the health outcomes by a monetary valuation. There is a large literature on the methods used to derive such monetary values. The final summary of a CBA is net monetary benefit: the monetary value of all benefits minus the monetary value of all resources used.23,24 Within the field of CEA, however, with its tradition of focusing on health and the health sector, there is no general consensus on what this comprehensive perspective is or how to apply it. The U.S. Panel on CostEffectiveness in Health and Medicine, for example, recommended that benefits and costs outside the health sector be included in CEA, but gave little guidance how to do that. IOM, the Second Panel on Cost-Effectiveness Analysis in Health and Medicine, and the National Institute for Health and Clinical Excellence, among others, have debated more specific guidance for a comprehensive perspective.25,26 A comprehensive perspective raises issues at every step of an analysis: structuring the analysis, measuring costs and benefits in natural units, and valuing costs and benefits. Keeping the steps conceptually separate is important to developing good guidance. Combining steps can lead to errors, particularly at the later steps, where measuring and valuing can sometimes be confused with each other.
Structuring the Analysis Structuring an analysis encompasses tasks familiar to any CEA analyst, among them specifying the target population, the alternatives to be evaluated, the costs and benefits to measure, and the time horizon over which to evaluate them.5 Identifying costs and benefits outside the health sector, unfamiliar territory for CEA analysts, can be difficult. At the identification step, analysts can use the strategy that has served them well for health interventions: review the literature and consult a range of subject experts. A 2012 IOM report, An Integrated Framework for Assessing the Value of Community-Based Prevention,25 addresses the issue of identifying all important consequences. It states that: “[t]he valuation of communitybased prevention interventions should be done with a comprehensive perspective; that is, the measurement of
benefits, harms, and resources should include impacts on all members of the community as well as on stakeholders who may be outside the community.” The first recommendation identifies three broad and mutually exclusive categories of benefits and costs: “The committee recommends that those seeking to assign value to communitybased prevention interventions take a comprehensive view that includes the benefits and harms in the three major domains of health, community well-being, and community process as well as the resource use associated with such interventions.” The IOM report explains: Community-based prevention can create value not only through improvements in the health of individuals but also by increasing the investment that individuals are willing and able to make in themselves, in their family and neighbors, and in their environment. Furthermore, community-based prevention involves decisions among groups of people about how to live in society, how the physical environment is built, what food is served in schools, and so on. Thus, the process by which interventions are decided upon and undertaken needs to be treated as a valued outcome. If a community decides to tell people what they can or cannot do, or what they should or should not do, the decisions need to have the legitimacy—the added value—that comes from an open and inclusive group decision-making process. No accepted measures exist for the second two classes of benefit. The committee recommended that the Centers for Disease Control and Prevention (CDC) develop some. A simple example can illustrate the three domains identified by the IOM report.25 Suppose a community invests in a park with trails and night-time lighting to encourage people to walk and engage in other physical activity for their health. The list of possible benefits certainly includes health, which the IOM report recommended be measured in QALYs, but the park can also bring aesthetic benefits (people can enjoy the park’s beauty whether they walk or not) and safety benefits (if the lighting reduces crime in the area). If decisions about the park are made with community involvement, to ensure that the community contributes to the plans, the benefits can include a greater sense of cohesion and pride on the part of community members. Cost benefit analysts routinely face the same issue. As noted, the perspective usually taken in a CBA is comprehensive—the societal perspective—and encompasses all significant benefits or costs that occur in the jurisdiction of the decision maker. For example, in its 2011 evaluation of the Clean Air Act Amendments of 1990, which estimated costs and benefits over the www.ajpmonline.org
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
1992 2020 period, the Environmental Protection Agency valued reductions in mortality and illness; improvements in visibility at recreational sites and in residential areas; benefits to commercial timber, agricultural crops, and recreational fishing; and reduced materials damage.27
Measuring Benefits and Costs One of CEA’s major contributions to decision making is that it not only lists the benefits and costs of an intervention, but also measures their magnitude. Quantifying them gives a clearer idea of their size and importance. The measurement process uses “natural units,” such as familiar health metrics—cases of disease, deaths prevented, and life-years and QALYs gained by an intervention. Thus, the next step is to select metrics that allow each cost or benefit to be measured. Interventions that take place largely or entirely outside the health sector can involve metrics new to CEA analysts as they seek to quantify the intervention’s effects on the aesthetic environment, crime and safety, and others. At this stage, costs (resources used by the intervention) are also best measured in physical units if data are available. As noted, existing measures for many benefits and costs, such as levels of smog and air pollution, rates of property and violent crime, test scores and years of schooling, housing quality, and traffic safety, can be identified by experts in each subject area. Analysts also need to measure all the resources used by an intervention. Perhaps even more than interventions that take place in the health sector, those that take place largely or entirely outside it can involve substantial amounts of donated resources. The temptation is to consider these resources as costing nothing—as free— because no check changes hands for them. Donated time, goods, and services are not free unless they would have gone completely unused had this project not come along. The Community Preventive Services Task Force makes this point, correctly noting that all resources, whether purchased or donated, should be included in costs.28 Consider the park example again. People may donate trees and plants for the landscaping. They may volunteer their time for planning, carrying out the plans, and maintaining the park once it has been created. Some businesses encourage employees to spend paid work time on community activities. Donated time and goods and services are as much resource costs of the project as are purchased labor and materials. Valuing and Summarizing Once benefits and costs are measured in natural units, how should they be valued? Why should they be valued? May 2016
S9
Or is a detailed table listing benefits and costs in natural units enough? Valuation involves weighting benefits and resources after they have been measured in natural units so that they can be combined and summarized. To make summarization possible, it attaches a second number to each that reflects people’s preferences: A higher number indicates a higher value. Using a consistent principle for valuation, and a common unit of measure for the values, as CBA does when it uses the principle of willingness to pay (WTP) to translate benefits and costs into dollars, allows benefits and costs to be combined in an overall summary. For a given decision, the detailed table of costs and benefits in natural units can and should be presented to decision makers, but they may also find summary measures helpful for comparing choices. In CEA, there are two summary measures, one for health and one for resources used. In CBA, a single summary measure, net benefit, is possible because both benefits and resources are valued in monetary terms. Valuing benefits and costs in order to summarize them can be the most difficult part of an analysis. It can be easy to value some benefits twice, such as the portion of health used for paid work, which then biases the summary in favor of interventions that produce particular kinds of health or health for particular kinds of people. The tendency to count donations as costing nothing biases the summary in another way, toward interventions that involve donated time, goods, and services. Finding or developing appropriate measures of value, measures that reflect how the community values benefits and costs, can also be difficult. In thinking about how to value benefits and costs, CEA can learn from the experience of CBA. For example, CBA used to value health as the additional earnings healthier people could make.29 That tradition is still widely influential outside of economics, but economists began to realize in the 1960s and 1970s that labor market earnings were the wrong way to value health because that is not how people themselves value it. For example, earnings assign no value to other activities made possible by better health— taking care of children or the elderly, maintaining the house, socializing, sleeping—although people clearly do value these activities. By the 1980s, economists agreed that the basic economic principle of WTP was the right way to approach valuing life and health and the concept “that social decisions should, so far as possible, reflect the interests, preferences and attitudes to risk of those who are likely to be affected by the decisions”—had been “extensively developed in the literature.”30 The Environmental Protection Agency was already applying the principle in the mid-1980s.31 In its CBA standards, the U.S. Office of Management and Budget states: “In
S10
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
monetizing health benefits, a WTP measure is the conceptually appropriate measure as compared to other alternatives (e.g., cost of illness or lifetime earnings), in part because it attempts to capture pain and suffering and other quality-of-life effects.”32 How can WTP be used to value life and health? The answer begins with the recognition that people act on their values all the time, revealing them through the choices they make. Some of those choices occur in markets, where people reveal how much they value a good or service through the amount they are willing to spend for it. Life and health are not sold in markets, but economists realized that goods and services sold in markets often reflect people’s valuation of non-market goods. Purchases of safety equipment to prevent accidents, for example, reveal a WTP for better health and longer life. Similarly, higher pay is required for dangerous jobs because workers value life and health.33 Economists have used this “risk premium,” the extra money per hour of work, to calculate the “value of a statistical life” (VSL; the value of a statistical life-year, VSLY, can be derived from VSL32) and have found that it implies a high valuation on life and health. Based on wage-risk studies, the Environmental Protection Agency currently uses a VSL of $7.4 million (2006 dollars) to value reductions in mortality due to less pollution. The WTP principle explains an apparent paradox first noted when labor market earnings were still used to value health: People were often willing to contribute large sums to rescue others in distress, sums much larger than the earnings used to value life saving in CBA. The explanation proposed at the time was that people valued “identified” lives—real people in real crises—more highly than they valued “statistical” lives, the future and unidentified lives that would be saved by, say, introducing safety gear on the job or repairing bridges. Once, however, economists began to estimate what people were willing to pay for life saving, VSL turned out to be much higher than lost earnings and in line with the amounts they contribute to save identified lives.34 Another principle borrowed from CBA is that the resources used by an intervention should be valued at their opportunity cost, which reflects the value of the goods and services that would be produced if the resources were put to their next-best use. In effect, valuation of resources is also based on the principle of WTP. In perfectly functioning markets—many buyers and sellers, each fully informed about the product, where benefits accrue only to buyers, costs only to sellers— prices would reflect this opportunity cost. Even imperfect market prices are often good proxies for opportunity costs. When prices are not available, as for resources that are not sold in markets, market prices for other goods
can, again, often offer evidence about the value people place on those resources. This explains economists’ practice of valuing time spent in activities other than paid employment at the wage for paid work: By choosing unpaid activities, people indicate that they value time spent on those activities at least as highly as the wage they could earn with that time.35 A distinction is sometimes drawn between the valuation of things sold in markets and those not sold in markets,28 but the economic principle does not make such a distinction. Both kinds of resources are valued at their opportunity costs and have thus been translated into comparable terms using a consistent principle.
Conclusions Where does this point us? Just as CEA has adopted helpful CBA practices in the past, it can benefit from considering the way CBA handles the valuation of benefits and costs that fall outside the health sector as it begins to develop standards for this kind of application. There is no ready-made solution available and, in any event, the solution must be developed out of a consensus among analysts and decision makers. The following principles, however, may help lead to a good solution: 1. In democratic societies, outcomes should be valued as the affected people value them. CEA, which derives QALY weights from surveys, and CBA, which derives monetary values from market choices (and, when those are lacking, surveys), agree on this principle. 2. As it has in the past, the experience of CBA can help CEA analysts develop ways to value more outcomes in monetary terms and do so in a consistent way. But, it is important that CEA analysts inform themselves of current thinking in CBA and consider its implications before developing general recommendations. 3. CBA has the strength that everything can be monetized, so that the sum of costs can be deducted from the sum of benefits to show whether an intervention is beneficial when the potentially wide array of benefits and costs involved in non health sector interventions is considered. CEA has the strength that it focuses directly on health, the outcome of primary interest, using a measure of health. For now, it seems best to follow the advice of the U.S. Office of Management and Budget and do both, in order to give decision makers a full pictures of the consequences of an intervention.32,36 4. In either form of analysis, benefits and costs should always be presented in a disaggregated www.ajpmonline.org
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
way, so that decision makers can clearly see the components. 5. CEA summarizes results with cost-effectiveness ratios, CBA with net monetary benefit. Both can calculate those summaries in a variety of ways when there are many benefits and costs involved. For example, CEA can calculate the health sector cost per QALY, the total cost per QALY, and the net cost (costs less monetized non-health benefits) per QALY. CBA can calculate net benefits for health effects and health sector costs or for wider sets of benefits and costs; for example, in the case of the park, including only aesthetic benefits in one summary and both crime reduction and aesthetic benefits in another. CEA analysts and decision makers need to explore the range of possible summary measures before arriving at a consensus, if one is possible, about those that are conceptually consistent and best serve the needs of public health decision makers.
Publication of this article was supported by the Centers for Disease Control and Prevention. We thank the four reviewers and guest editors of the supplement for helpful comments on an earlier version of the paper. Preparation of this paper was not supported by any grant or contract. Both authors contributed to the conceptualization of the paper, development of ideas, and critical review and final approval of the draft; LBR took primary responsibility for drafting the paper. The authors have no financial conflicts of interest in connection with this paper and received no external funding to write it. Dr. Russell is a member of the leadership group of the Second Panel on Cost-Effectiveness in Health and Medicine.
References 1. NHE Summary, including share of GDP, CY 1960-2013, Table 1. http:// www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trendsand-Reports/NationalHealthExpendData/nationalHealthAccountsHis torical.html. Accessed July 9, 2015. 2. Woolf SH, Aron L. Panel on Understanding Cross-National Health Differences Among High-Income Countries; Committee on Population; Division of Behavioral and Social Sciences and Education; National Research Council; Board on Population Health and Public Health Practice; Institute of Medicine. U.S. Health in International Perspective: Shorter Lives, Poorer Health. Washington, DC: The National Academies Press; 2013. 3. Teutsch S, Fielding JE. Rediscovering the core of public health. Front Public Health Serv Syst Res. 2013;2(3):1–7, Article 1.
May 2016
S11
4. IOM, National Academy of Sciences, Engineering, Medicine. Topic: environmental health. http://iom.nationalacademies.org/Global/Topics/ Environmental-Health.aspx. Accessed July 15, 2015. 5. Gold MR, Siegel JE, Russell LB, Weinstein MC, ed. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996. 6. Russell LB. The science of making better decisions about health: costeffectiveness and cost-benefit analysis. In: Kaplan RM, David D, Spittel M, eds. Emerging Behavioral and Social Science Perspectives on Population Health. Washington, DC: U.S. Government Printing Office; 2015. 7. Russell LB. Is Prevention Better than Cure? Washington, DC: The Brookings Institution; 1986. 8. Cohen JT, Neumann PJ, Weinstein MC. Does preventive care save money? Health economics and the presidential candidates. N Engl J Med. 2008;358(7):661–663. http://dx.doi.org/10.1056/NEJMp0708558. 9. Russell LB. Preventing chronic disease: an important investment, but don’t count on cost savings. Health Aff (Millwood). 2009;28(1):42–45. http://dx.doi.org/10.1377/hlthaff.28.1.42. 10. Cromwell J, Bartosch WJ, Fiore MC, et al. Cost-effectiveness of the clinical practice recommendations in the AHCPR guideline for smoking cessation. JAMA. 1997;278(21):1759–1766. http://dx.doi. org/10.1001/jama.1997.03550210057039. 11. Mandelblatt J, Saha S, Teutsch S, et al. The cost-effectiveness of screening mammography beyond age 65 years: a systematic review for the U.S. Preventive Services Task Force. Ann Intern Med. 2003;139(10):835–842. http://dx.doi.org/10.7326/0003-4819-139-10-200311180-00011. 12. National Cancer Institute. Cancer Intervention and Surveillance Modeling Network: Modeling to guide public health research and priorities. http://cisnet.cancer.gov/about/. Accessed March 10, 2015. 13. Caro JJ, Briggs AH, Siebert U, Kuntz KM. Modeling Good Research Practices—Overview: A Report of the ISPOR-SMDM Modeling Good Research Practices Task Force 1. Med Decis Making. 2012;32(5): 667–677. http://dx.doi.org/10.1177/0272989X12454577. 14. PRISMA (Transparent Reporting of Systematic Reviews and MetaAnalyses). www.prisma-statement.org/. Accessed July 7, 2015. 15. Cochrane Community Reviews. www.cochrane.org/cochrane-reviews. Accessed July 7, 2015. 16. IOM. Finding What Works in Health Care: Standards for Systematic Reviews. www.iom.edu/Reports/2011/Finding-What-Works-in-HealthCare-Standards-for-systematic-Reviews.aspx. Published 2011. Accessed July 7, 2015. 17. National Institute for Health and Care Excellence (NICE). Guide to the methods of technology appraisal. Published 2013. http://publications. nice.org.uk/pmg9. April 4, 2013. 18. Canadian Agency for Drugs and Technologies in Health (CADTH). Guidelines for the economic evaluation of health technologies: Canada. 3rd ed. Ottawa: Canadian Agency for Drugs and Technologies in Health; 2006. 19. Australian Government, Department of Health and Ageing, Pharmaceutical Benefits Advisory Commission (PBAC). Guidelines for Preparing Submissions to the Pharmaceutical Benefits Advisory Committee (Version 4.3). http://www.pbac.pbs.gov.au/content/infor mation/archived-versions/pbac-guidelines-v4.3-2008.pdf. Published December 2008. Accessed November 19, 2015. 20. WHO. WHO guide to cost-effectiveness analysis. Published 2003. www.who.int/choice/publications/p_2003_generalised_cea.pdf. 21. Bill and Melinda Gates Foundation. Methods for Economic Evaluation Project (MEEP) Final Report. A partnership between Bill and Melinda Gates Foundation, NICE International, the Health Intervention and Technology Assessment Program (Thailand), and the University of York, Centre for Health Economics. https://www.nice.org.uk/Media/ Default/About/what-we-do/NICE-International/projects/MEEP-report. pdf. January 2014. Accessed November 19, 2015. 22. Salomon JA, Vos T, Hogan DR, et al. Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):2129– 2143. http://dx.doi.org/10.1016/S0140-6736(12)61680-8.
S12
Russell, Sinha / Am J Prev Med 2016;50(5S1):S6–S12
23. Boardman AE, Greenberg DH, Vining AR, Weimer DL. Cost-Benefit Analysis: Concept and Practice. New York NY: Prentice Hall, 2011. 24. Freeman III. The Measurement of Environmental and Resource Values: Theory and Methods. Second Edition, Washington, DC: Resources for the Future; 2003. 25. IOM. An Integrated Framework for Assessing the Value of CommunityBased Prevention. Washington, DC: The National Academies Press; 2012. 26. Social Value Judgements: Principles for the Development of NICE Guidance. Second Edition. National Institute for Health and Clinical Excellence website. http://www.nice.org.uk/proxy/?sourceUrl=http%3a% 2f%2fwww.nice.org.uk%2fmedia%2fC18%2f30%2fSVJ2PUBLICAT ION2008.pdf. Published 2008. Accessed November 19, 2015. 27. U.S. Environmental Protection Agency, Office of Air and Radiation. The Benefits and Costs of the Clean Air Act from 1990 to 2020: Summary Report. Washington, DC: U.S. Environmental Protection Agency, March 2011. 28. Stone GA, Hutchinson AB, Corso PS, et al. Chapter 11: understanding and using the economic evidence. In: The Guide to Community Preventive Services: What Works to Promote Health? www.thecommu nityguide.org/library/economics.pdf. Accessed June 8, 2015. 29. Jone-Lee MW. The Value of Life: An Economic Analysis. Chicago, IL: The University of Chicago Press; 1976.
30. Jones-Lee MW, Hammerton M, Philips PR. The value of safety: results of a national sample survey. Econ J. 1985;95(377):49–72. http://dx.doi. org/10.2307/2233468. 31. Nichols AL. Lead in gasoline. In: Morgenstern RD, Economic Analyses at EPA: Assessing Regulatory Impact. Washington, DC: Resources for the Future, 1997:49–86. 32. U.S. Office of Management and Budget, Circular A-4, Subject: Regulatory Analysis, September 17, 2003. www.whitehouse.gov/omb/ circulars_a004_a-4/. Accessed July 14, 2015. 33. Robinson LA. How US government agencies value mortality risk reductions. Rev Environ Econ Policy. 2007;1(2):283–299. http://dx.doi. org/10.1093/reep/rem018. 34. Russell LB. Do we really value identified lives more highly than statistical lives? Med Decis Making. 2014;34(5):556–559. http://dx.doi. org/10.1177/0272989X13512183. 35. Russell LB. Completing costs: patients’ time. Med Care. 2009; 47(7 Suppl 1):S89–S93. http://dx.doi.org/10.1097/MLR.0b013e31819bc077. 36. Miller W, Robinson LA, Lawrence RS. Committee to Evaluate Measures of Health Benefits for Environmental, Health, and Safety Regulation, eds. Valuing Health for Regulatory CostEffectiveness Analysis. Washington, DC: The National Academies Press; 2006.
www.ajpmonline.org