Search Results

When prevention and treatment equally save real lives and avoid real suffering, their results appear to have equivalent value. This creates a strong initial case for a policy of priority equivalence between them. After demonstrating that three arguments often made for treatment’s priority quickly dissolve in the face of this equivalent value, I pursue the intricate distinction between identifiable as compared to statistical lives, concluding that because it does not track well the distinction between prevention and treatment, it does not itself provide an argument for giving priority to treatment. Two arguments are more promising: (1) The very value of the lives and health at stake respectively in cases of prevention and treatment is frequently not equal, but variable in the direction of being lower for prevention. (2) In any context where we are about to abide by a new policy of moral equivalence, it is difficult to be fair to those in the transition generation who are already ill. Neither of these promising arguments, however, is compelling. The first is persuasive only within certain limited contexts and thus generates only a selective priority for treatment. The second loses its moral attraction once we recognize its long-term historical consequences.

Experience with cost-benefit analysis in developed countries has produced a wealth of data on individual preferences. Because of a relative dearth of data collected directly in developing countries, values are often transferred from studies conducted in developed countries. While benefit-transfer analysis raises a number of concerns, practical realities often require its use. Focusing on the use of benefit-transfer to value air pollution-related mortality risks in sub-Saharan Africa, the authors conclude that benefit-transfer permits a wide range of estimates for the value of a statistical life. As a partial remedy, they suggest technical innovations useful for transferring values from high- to low-income countries.

For years, debates about the best way to combat the AIDS pandemic have pitted proponents of scaling up antiretroviral treatment for people suffering from AIDS against those advocating for more cost-effective prevention measures. In an important recent article, Dan Brock and Daniel Wikler argue that there is no sound moral basis to privilege the saving of identified lives through antiretroviral treatment, if preventive methods could save more (statistical) lives. This chapter takes issue with Brock and Wikler’s argument. In so doing, it develops a novel account of how the choice between “treatment” and “prevention” intersects the problem of identified versus statistical lives. The chapter concludes with a postscript on “treatment-as-prevention” (TasP), a new avenue of HIV/AIDS research that stresses the preventive benefits of early antiretroviral treatment. It argues that, despite its medical promise, TasP does not transcend the ethical dichotomy between treatment and prevention explored in this chapter.

This chapter introduces and briefly evaluates several arguments commonly thought to support the moral importance of the distinction between statistical and identified lives in the context of saving lives in health care, many of which frequently appear in the literature. It argues that none of them succeeds in doing that. It concludes that when, but only when, a morally important principle like priority to the worse off maps onto the identified/statistical difference, then the identified/statistical difference is morally important—but that is because of the other principle—in this case, priority to the worse off—not the difference between identified and statistical lives

Environmental law focuses on coming probabilities, spread across human populations, rather than on past events, visited upon individuals, and thus it fixes attention on statistical rather than identified lives. The major US environmental laws, as written, offer exceedingly strong protections for human health. In reality, however, environmental law often does not live up to its protective promise. In defending decisions to allow people to die, the government often deploys arguments that, implicitly or explicitly, take advantage of the generally lower status of the statistical life. The effect is to move us away from solicitude and toward nonchalance in our relationship to the statistical life.

When experts write guidelines for patient care for a health condition, they speak about a certain population of patients from a statistical perspective. When clinicians care for patients who have these conditions, they deal with identified persons. The distinction between statistical and identified lives is thought to make it more acceptable for policymakers to say no to certain treatments for a population of prospective patients than for clinicians to prioritize care for their individual patients. But such a view involves too simplistic an understanding of the statistical/identified distinction. This chapter puts forth the argument that in clinical contexts, it is often not known whether an individual patient would actually benefit from treatment. At the point of treatment patient beneficiaries also remain “statistical.” Clinical treatment cannot therefore simply be associated with identified lives while prevention and policymaking are associated with statistical lives. The confusion in doing so is compounded by speaking of ‘rationing’ instead of ‘prioritizing.’

In Chapter 5, psychologist Paul Slovic and legal scholar David Zionts discuss psychological constraints that impede individuals' ability to respond to the plight of large numbers of people. As a result, individuals price lives differently depending on the numbers involved—specifically, they succumb to a numbing effect whereby they register a greater emotional and financial response to the loss of a few than to the loss of many.

In this chapter, Andreas Mogensen discusses the suggestion that one might be morally obligated to let the child drown in Singer’s infamous “Shallow Pond” case, so that one can save a greater number of lives through donations. Intuitively, there would be something morally horrendous about doing this. Yet a moral requirement to let the child drown seems to be the conclusion of reasoning very similar to that used by Singer and his allies to argue for demanding duties to donate on the basis of cases like “Shallow Pond”; what should we make of this? Mogensen attempts to capture both the intuition that our obligations to donate to effective life-saving organizations are as strong as our obligations to save the child in “Shallow Pond” and the intuition that one should not allow the child to drown even if by doing so one could save a greater number of lives through donations.

While it is plausible that HIV/AIDS slows down GDP growth in the longer run, as a relatively smaller population translates into lower GDP, the empirical evidence on the impact of HIV/AIDS on economic growth is weak. In the empirical literature on economic growth, life expectancy is often used as a proxy for ‘health capital’, and studies calibrating the impact of HIV/AIDS on growth through such a growth regression, combined with an estimate of the link between life expectancy and GDP, find large effects of HIV/AIDS on growth. However, studies estimating the link between HIV/AIDS and growth directly, including estimates of the state of HIV/AIDS in a growth regression, tend to find no effect. Studies evaluating the impact of HIV/AIDS on mortality and on living standards, applying concepts like ‘full income’ or the ‘value of statistical life,’ return an estimated loss of up 70 per cent of GDP (Botswana, 2000).

This chapter estimates the value of a statistical life for children in metropolitan Manila, based on adults’ stated willingness to pay for dengue vaccines. Using a novel approach involving play-like activity, which also served to break the monotony of the question-answer sequence, the study estimates the value of reductions in mortality risk, making the estimates useful for a variety of environment-related project and policy assessments. A two-stage estimation procedure is used, consisting of a household vaccine demand model and a random effects probit model of the vaccine purchase decision for individual members. The estimate for the value of a statistical life is in the range of $700,000 to $800,000 (US dollars).

Decisions on investing in health as well as other policies require deciding how to best allocate available resources—recognizing that using labor, materials, and other resources for one purpose means that they cannot be used for other purposes. The analytic approaches discussed in this volume have in common the overarching goal of providing information on policy impacts, so as to provide an evidence base for decisions. What distinguishes benefit-cost analysis is its emphasis on explicitly accounting for all significant outcomes (both health and non-health) and on valuing them in monetary units to facilitate comparison. Benefit-cost analysis makes the relative values of different outcomes explicit. As conventionally implemented, benefit-cost analysis does not address the distribution of impacts within a population, but it can be supplemented to do so.