Cost-Benefit Analysis of Providing Non-Emergency Medical Transportation (2005)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Appendix C: Cost-Effectiveness Analysis and QALYs This study makes heavy use of a number of economic concepts, particularly as they pertain to healthcare. This appendix presents a brief overview of two concepts that may not be well known by all readers of this report. They are cost-effectiveness analysis (CEA) and quality-adjust life years (QALYs). C.1 Economic Evaluations in Healthcare Economic evaluation methods used in healthcare include cost-of-illness studies, cost- minimization analysis, CEA, and cost benefit analysis (Jefferson, et al., 1996). For this project, we believe that the most appropriate method of evaluating the benefits of increased transportation is through CEA. This method employs measures of the effectiveness-per-unit-cost, as opposed to a typical cost-to-cost comparison (Warner and Luce, 1982). Gold writes, “A primary objective of cost-effectiveness analysis is to incorporate a consideration of resource consumption into decisions about healthcare. An explicit examination of resources allows an assessment of costs relative to the health benefits of an intervention” (1996, p.176). Interventions devoted to extending life or increasing its quality may be so successful that net costs (the cost of the intervention minus savings that accrue due to decreased healthcare expenditures) fall – the intervention pays for itself! Most often, however, interventions will have a positive net cost, but the benefit may still be judged worthwhile. This is the proper domain of CEA, as illustrated in Figure C-1. Figure C-1: The Basic Cost-Effectiveness Analysis Matrix Costs Ð Outcomes Ð Costs Ð Outcomes Ï Costs Ï Outcomes Ð Costs Ï Outcomes Ï Black: “Dominance”; Grey: Reject Out of Hand; Textured: CEA Relevant The upper, right quadrant of the figure shows the “dominant” situation, which is characterized by lower net costs after an intervention coupled with better outcomes. The lower left “rejection quadrant” indicates the worst of both worlds: higher costs and worse outcomes. The two textured, diagonal quadrants exemplify the situations relevant to CEA: lower costs/worse outcomes or higher costs/improved outcomes. Because we usually expect better health to warrant an investment, the key question becomes: Is a particular health improvement worth its additional cost? Final Report C-1

C.2 Cost-Effectiveness Analysis Ratios and QALYs The quality-adjusted life year (QALY) concept acts to combine mortality and morbidity into a single measurement for gauging the value of a health-related intervention. Augmenting the conventional CEA framework with the QALY measure elevates the CEA method’s true power: providing a summary measure that specifically integrates quality improvements and facilitates comparisons across therapies – not only can one asthma intervention be compared with another, but each can now be compared with, e.g., a congestive heart failure intervention. The QALY concept depends on a health-related quality of life (HRQL) measurement than can be represented as a continuum bounded by two extremes: 0 representing death and 1 representing perfect, or optimal, health. Thus: 0 < HRQL ≤ 1 The concept begins with a single individual and is then expanded to a population by summing HRQLs across individuals. With estimated HRQLs, the QALY concept can be illustrated as follows. Suppose a randomized control trial of 1,000 heart failure patients is initiated to test an integrated disease management strategy. Assume this intervention leads to an average, per person gain of 0.1 years of life expectancy that is adjusted downward by one-half to reflect its severely impaired HRQL. That is, the representative individual would live 1.2 extra months, but do so at a level that is rated at approximately one-half that of perfect health. The total QALYs obtained from this hypothetical intervention would be: 0.1 * 0.5 * 1,000 = 50 The cost of the intervention results in increased life expectancy or a higher quality of life (or both) for the affected population, as measured by surveys and health-related standards. Comparing the intervention as a discrete alternative to a baseline case yields a cost-effectiveness ratio: All costs are placed in the numerator and the benefit (the sum of QALYs) is shown in the denominator. Continuing from the above illustration, the total cost of the heart failure trial (which would include all incremental costs for treating the experimental group minus any cost savings from reduced healthcare expenditures experienced by this group) might equal $1,000,000. Accordingly, the CEA ratio would be: $1,000,000/50 QALYs = $20,000 / QALY While not based on actual data, this example is germane to this study precisely because an integrated disease management protocol would rely heavily on multiple healthcare encounters for a severely impaired population. This population may also be medically underserved either because of low income or very high utilization, the costs of which are not fully covered by insurance. The potential for missed visits due to transportation is great; the potential for decreases in health status (net health benefits) because of these missed visits likewise is great, especially to the extent that the missed visits are instrumental to the success of the disease-management strategy. An example of a highly cost-effective therapy concerns use of ACE inhibitors by symptomatic patients with heart failure. This treatment has demonstrated a CEA ratio of only $115 per added QALY (Glick et al., 1995). Similar, highly cost- Final Report C-2

effective results have been found for strategies to prevent neural tube defects (Gold et al., 1996). C.3 Relative and Absolute Appraisals in Healthcare Analysis Researchers often anticipate robust results that will show an intervention to be cost- effective, if not actually cost saving – the case of “dominance.” As discussed above, while several examples of highly positive results exist – especially studies that analyze prevention for at-risk populations – most interventions show increases in net cost. Are the benefits worth the cost? While no absolute standard is without controversy, a general convention in the field is that a CEA ratio of less than $50,000 per QALY is regarded as economically attractive. This is based on the cost of providing dialysis to renal failure patients (Evans et al., 2004; Mark and Hlatky, 2002). CEAs that result in a cost per QALY greater than $100,000 are seen as clearly unattractive; the $50,000-to-$100,000 range represents a gray area that requires more subjective judgment by the analyst. Finally, the QALY construct should be seen as producing a relative-absolute measure. The absolute component is a QALY cost estimate for a particular healthcare intervention, and the relative component is the comparison of this estimate to either a different healthcare intervention or to healthcare interventions in general. As above, denoting an intervention as cost-effective can never be assessed in a perfect, non-contentious manner. Both the HRQL assessment and the cost- effectiveness demarcation (e.g., $50,000) will remain controversial. In the health arena, however, this approach is far superior to a strict cost benefit analysis, because the latter forces the researcher to evaluate all the health benefits (as well as the costs) in purely monetary terms. Thus, instead of asserting that different health states can be ranked according to a 0 – 1 scale, and that this scale can be used to adjust life expectancy from a particular intervention, the cost benefit approach requires an estimate of the absolute monetary value of a quality-adjusted life expectancy. Such an estimate is controversial at its core and objectionable to most analysts on ethical and other grounds. Final Report C-3