Graphic of Conference Logo - The 3rd International Conference - Measuring the Burden of Injury

Keynotes

KEYNOTE 1
Economic Evaluation of Injury Control:
Alternative Perspectives

John D. Graham, Ph.D.

Introduction

Should a proposed injury control intervention, whether preventive, clinical, or rehabilitative, be subjected to an “economic evaluation” to inform the decision about whether the intervention is implemented? If so, what kind of economic evaluation should be performed? There is a range of answers to these questions and thus the purpose of this essay is to describe some of the major alternative perspectives.

An “economic evaluation” means different things to different people. A leading textbook in the field defines it as “a comparative analysis of alternative courses of action in terms of both their costs and consequences” (Drummondet et al., 1997). The most important word in this definition is “comparative” because any particular injury control measure may look good or bad depending upon what it is compared to (e.g., airbags look good compared to no restraint but not quite as good compared to proper use of a lap/shoulder belt). The intervention to be analyzed may be a single technology/program or a combination of technologies/programs but the evaluation cannot proceed without a well-defined “comparator.”

From a technical perspective, there are four types of economic evaluation: cost-minimization analysis (which simply identifies which alternative is least expensive), cost-effectiveness analysis (which identifies the alternatives that maximize the amount of injury reduced subject to a fixed budget constraint), cost-utility analysis (which identifies the alternatives that maximize the amount of “utility” subject to a fixed budget constraint), and cost-benefit analysis (which identifies the alternative that maximizes net benefits, defined as benefits minus costs). Recent progress in measuring the burden of injury has been particularly useful in enhancing the feasibility of cost-utility and cost-benefit studies of injury control interventions.

Perspectives Govern Analytical Choices

An economic evaluation can be undertaken from numerous perspectives. The findings of an evaluation may differ depending upon whose perspective is used in the analysis. For example, a health care provider may be concerned about an intervention’s impact on a patient’s use of health care services but the provider may have no financial interest in a patient’s ability to work and be productive. On the other hand, an employer’s perspective may focus on productivity concerns. A patient’s perspective may ignore both health care costs and foregone earnings (if the patient is insured against these losses, as is common in some European countries) and focus primarily on length and quality of life.

The most popular perspective used in economic evaluation is that of “society” (Gold et al., 1996). With this perspective, the analyst is expected to consider all costs and consequences of an injury control measure, regardless of who in society incurs these consequences. The challenge is to define the “society”, which could be all citizens in a particular town, locality, or state. There may also be occasions where an international perspective is appropriate. It should be noted that a “governmental” perspective is not necessarily the same as a “societal” perspective, particularly in nations that have a significant private sector. Health care costs incurred by the private sector are counted under a societal perspective but may be excluded from a “governmental” perspective. Likewise, time costs to family members that care for patients are counted in a “societal” analysis but may be excluded from a “governmental” perspective. Regardless of how the issue is resolved, the analyst must identify a perspective in order to know which costs and consequences to count.

Efficiency and Fairness

The societal perspective is rooted in a philosophical perspective called utilitarianism, which Jeremy Bentham and John Stuart Mill defined as involving some calculation of “the greatest good for the greatest number.” Cost-benefit analysis (CBA), in particular, arose out of efforts to translate social utilitarianism into a practical tool for public policy makers. Although many people associate “cost-benefit analysis” with the monetary calculation of an American capitalist, the roots of cost-benefit analysis can be traced to the writings of European social scientists interested in defining the public good. These writers include Jules Dupuit (France, 1844), Vilfredo Pareto (Italy, 1906), Nicholas Kaldor (Great Britain, 1939), Sir John Hicks (Great Britain, 1939, 1941), and EJ Mishan (Great Britain, 1971). For appropriate references, see Mishan (1994).

From this body of scholarship, a definition of “economic efficiency” was formulated from the perspective of society. As applied to injury control, the definition is as follows: if those citizens who benefit from an injury control intervention had to bear its entire cost, they would consider it worthwhile. If those who benefit would not be willing to pay for its cost, the intervention is “inefficient.” The beneficiaries are not necessarily required to actually pay for the intervention (if, for instance, public financing is done through some tax mechanism that collects general revenues) but the principle that “beneficiaries would be willing to pay” - sometimes called the principle of consumer sovereignty - underlies the definition of “economic efficiency.”

In the most ambitious thinking of early writers, the economic-efficiency test was proposed as a necessary and sufficient condition for public policy intervention. This ambitious position is no longer favored due to growing philosophical and political interests in fairness as well as efficiency. For example, an injury control intervention aimed at a low-income population that is both effective and inexpensive might still be judged “economically inefficient” if the beneficiaries, due to their low ability to pay, do not express sufficient interest in the benefits of the measure. These kinds of fairness concerns have led to the viewpoint that economic efficiency should be a contribution to public choice but efficiency should not be considered a necessary or sufficient condition for public intervention.

One of the most perplexing challenges is to determine how much weight to give to efficiency and fairness in making policy decisions. This is the topic of a well-known book by the late Arthur Okun, “Equality and Efficiency: The Big Tradeoff,” (1975).

Strands of Utilitarianism

There are three major philosophical traditions that have influenced political thinking in modern Western-style democracies: liberalism, utilitarianism, and communitarianism.

“Liberalism” refers to both the libertarians, who seek to protect and expand the private sphere of life by defining those “negative” rights of citizens that government may not violate, and egalitarians, who pursue more fairness and equality in life by defining those “positive” rights of citizens that government is obliged to supply. For strict liberals of either stripe, there is little interest in economic efficiency since the formulations of rights determine what a society should do to advance the welfare of citizens. Strict communitarians, who believe that public policy should evolve out of community-defined values (whether ethically or religiously defined), have also expressed little interest in economic efficiency. Of course, “strict” liberals and communitarians, defined as those who have no interest in other perspectives, are few in both number and influence and thus there is interest throughout the world in practical tools to operationalize utilitarian thinking, at least as a partial contribution to public policy.

There are two competing strands of utilitarianism, subjective and objective, which have influenced the evolution of the field of economic evaluation. Subjective utilitarians believe that the principle of consumer sovereignty should govern value choices. In the case of consumer or patient choice about injury control, the subjectivist believes that the personal perspective of the consumer or patient, assuming they have been fully informed of the issues, should determine what is a “benefit” or “cost” and how much money should be expended to prevent or treat a particular trauma-induced impairment. Hence, strict utilitarians believe that cost-benefit analysis is the appropriate tool for application to injury control interventions. The objectivist utilitarians do not completely trust consumer or patient perceptions in the field of health and thus prefer cost-effectiveness analysis.

Cost-Benefit Analysis

In order to apply CBA to injury control measures, it is necessary to assign monetary values to health outcomes such as nonfatal and fatal injuries that are averted. Substantial progress has been made in this field during the last twenty years. Early studies, pioneered by Selma Muskin (1962), Dorothy Rice (1967), and Burton Weisbrod (1961), applied the “human capital method” to diseases and injury. This method quantifies both the health care costs and foregone earnings attributable to injury, providing an economic basis for quantifying the rate of return from injury prevention measures. The analytical strategy was adapted from the work of economists who were interested in the long-run economic payoff of early investments in the education of children.

Concerns were raised that human-capital studies are theoretically inadequate, since they do not reflect the principle of consumer sovereignty, and do not account for the pain and suffering caused by injuries. Economists, stimulated by the writings of Thomas Schelling, Michael Jones-Lee (1974), and others, argued that injuries should be monetized using willingness-to-pay (WTP) tools. Studies have quantified WTP for health gain based on the revealed preferences of workers and consumers (Viscusi, 1998) or the expressed preferences of survey respondents (Mitchell and Carson, 1989). More recently, Magnus Johannesson and others have suggested that a full accounting of the benefits of injury or disease prevention should include both the personal WTP of people at risk as well as any external costs of injuries (e.g., human capital costs not incurred by the injured) (1996). WTP-based CBA of transport safety policies are now commonplace in both the USA and the UK.

Cost-Effectiveness Analysis

A fundamental limitation of CBA is that it will assign low monetary values to consequences experienced by people who have a low ability to pay (i.e., the poor). This criticism has its roots in egalitarian philosophy and has been quite influential. It has also been argued that “pricing” health through explicit monetary tools is a symbolically and/or politically bad thing to do (Kelman, 1981). This can be considered a communitarian argument in the sense that citizens may feel uncomfortable living in a community where explicit monetary values are assigned to lives and injuries. Despite the analytical progress in monetizing health outcomes, many people are skeptical of or hostile to CBA, including some people who are somewhat sympathetic to utilitarianism.

The dissatisfaction with CBA, particularly as applied to clinical medicine and public health, has stimulated scientific and practitioner interest in methods of cost-effectiveness analysis (CEA) and cost-utility analysis (CUA). Unlike CBA, which seeks to determine which alternative is most efficient, the objectives of CEA and CUA are more modest; to rank alternatives in terms of their cost-effectiveness ratio. Milton Weinstein and William Stason developed the tools in this field (1977). The ultimate decision about which ratios are acceptable is left for accountable (political) decision-makers to decide either implicitly (by setting a budgetary constraint) or explicitly.

In CEA the task is to compare alternatives according to their cost per physical unit of effectiveness; in CUA the comparison is made in terms of cost per preference-weighted unit of effectiveness. For injury control measures, lives saved or injuries averted are often used as the physical unit of effectiveness. If measures promise reductions in both fatalities and nonfatal injuries, CEA is difficult to implement because CEA requires a uni-dimensional measure of effectiveness.

CUA is an attractive tool because it can be used to combine information on fatalities and injuries into a single unit, most commonly through use of the “quality-adjusted life year” (QALYs) and the “disability-adjusted life year” (DALYs). QALYs, which are often used in developed countries, use citizen-elicited preferences to determine the relative value of nonfatal injuries and fatalities. This approach, which was developed by James Bush, Richard Zeckhauser, and Donald Shepard, assigns each state of nonfatal injury a value on a 0-1 scale, where 1 means perfect health and 0 means death (1976). If a paraplegic values a year of life at 0.7, that means that he/she would be indifferent between 7 years of perfect health and 10 years of life as a paraplegic.

DALYs, which are often used in developing countries and were developed by Christopher Murray, have socially imposed value judgments that reflect an interesting mix of human capital and egalitarian influences (1994). Unlike QALYs, which assume that each healthy year of life has equal value (regardless of age), the most prominent version of DALYs assigns the larger weight to the middle of the life span and less weight to youth and old age. Moreover, DALYs assume the target population will live a life as long as the average Japanese female while QALYs are based on actuarial life expectancies estimated for the specific target population. Despite their differences, QALYs and DALYs have similarities that are at least as important as the noted differences. Some authors, who use CEA to describe any optimization methods based on physical or preference-based measures of effectiveness, do not use the term CUA.

The Promise of Burden-of-Injury Research

For practitioners of economic evaluation, burden-of-injury research is critical to future application of evaluation tools to injury control measures. Ted Miller, for example, has played a critical role in bringing economic tools, thinking, and data to the injury control community (2000). Economic measures of the burden of injury, particularly the measures of health-care cost and productivity loss, play a role in both net-benefit calculations (CBA) and in construction of cost-effectiveness ratios (CEA). The Functional Capacity Index, developed by Ellen MacKenzie and Steve Luchter (1996), has a particularly important role to play in CEA, since it can be employed to express the burden of injuries in terms of foregone QALYs.

If standardized methods are established for conducting CEAs and using burden-of-illness information, it will be feasible to compare the cost-effectiveness of measures of injury control interventions. For example, the cost-effectiveness ratios for driver and passenger airbags have been estimated at $24,000 and $62,000 per QALY saved (1993$), respectively, based on information derived from the FCI. Since QALYs are becoming widely used in other fields of medicine and public health, it will soon be feasible to make systematic comparisons of the cost-effectiveness of measures aimed at both diseases and trauma. The comparisons that have been made to date, published using data from the USA, Japan, and Sweden, are vulnerable to the criticism that they consider only life years as the measure of effectiveness, ignoring burden-of-illness impacts on quality of life. The economic case that these studies make for injury prevention is likely to be even stronger when QALYs instead of life years is used as the measure of effectiveness.

As countries continue to be influenced by practical, market-based, utilitarian thinking, the demand for valid tools of economic evaluation can be expected to grow. The findings from these analytic tools are unlikely to overwhelm political interests in liberalism and communitarianism but they are likely to gradually seep into the decision-making processes employed to formulate injury control policy. Burden-of-injury studies will provide a stronger scientific foundation for the modest yet growing influence of economic evaluation throughout the world.

References