Cost-effectiveness analysis is the current dominant methodology for health care cost and outcome evaluation. One metric from a cost-effectiveness analysis is the incremental cost-effectiveness ratio—the ratio of the net change in costs to the net change in effects associated with two different programs or therapies. The denominator represents the gain in health (eg, life years gained, number of additional survivors, cases of disease averted), while the numerator reflects the marginal cost in dollars. As the units are different for the numerator and denominator, the expression will take the form of cost per unit of benefit (eg, dollars per life years gained, dollars per additional survivor, dollars per cases of disease averted). Alternatively, the ratio of cost to outcome can be reported for an individual therapy, rather than in comparison to another therapy (this is known simply as the cost-effectiveness ratio).
After calculating the incremental cost-effectiveness ratio, there remains an entirely separate and subjective decision about whether that therapy or program is deemed cost-effective. That determination is based on a spending threshold—the amount that society is willing to pay overall for a given outcome. For many years, this threshold was held as $50,000, derived from an argument made in the early 1980s-1990s that (1) renal dialysis is cost-effective, (2) renal dialysis costs $50,000 per quality-adjusted life year saved, and (3) therefore, $50,000 is cost-effective. Some challenge this threshold,13,14 but there is general consensus that a level somewhere between $50,000 and $100,000 per year of life gained is acceptable in the United States today. Therefore, a new therapy with an incremental cost-effectiveness ratio of $82,000 per year of life gained would be viewed as cost-effective.
To create these ratios, a typical cost-effectiveness analysis requires collecting a significant amount of information on costs and effects for both standard care and the new intervention, often from varying sources. Assimilating this information may be difficult, requiring a decision analysis model to show key clinical decisions and outcomes. These models are represented by trees, where each branch has a probability of occurrence and a cost. At its simplest, the tree will contain only branches for treatment allocation (eg, inhaled nitric oxide or standard therapy) and outcome (eg, alive or dead). To calibrate the tree, we need to know the probability of living or dying based on each therapy, and the average cost of care for survivors and nonsurvivors in the two treatment arms (Fig. 6-2).
Simple decision tree comparing outcome for neonates with respiratory failure treated with inhaled nitric oxide versus standard care. In order to calibrate the tree, we must estimate (1) the probability for a given patient to live or die, given whether they received the new therapy or not and (2) the average costs associated with each of the four branches.
We could expand this model to include other elements that affect morbidity and cost, such as extracorporeal membrane oxygenation (ECMO) use or sequelae other than death. The new therapy, while expensive alone, may offset its own expense with a reduced need for other supportive care, and may therefore be comparatively more cost-effective than standard therapy. This is unlike the cost-benefit analysis, where downstream effects are not accounted for. As additional elements are incorporated in the decision analysis model, additional branches must be added to the tree. For each branch, we must know a patient's likelihood of entering the arm and the average costs (Fig. 6-3). Indeed, this is how inhaled nitric oxide for neonates with respiratory failure was shown to be a dominant strategy—through substantial reduction in the need for the even more expensive ECMO therapy and reduced incidence of patient-centered outcomes such as chronic lung disease.15
Decision tree comparing outcomes for neonates with respiratory failure treated with inhaled nitric oxide versus standard care that incorporates the potential for transfer from an outside hospital, extracorporeal membrane oxygenation, and outcomes with sequelae. In order to calibrate the tree, we must estimate the probabilities and average costs for nine separate trees. (Reproduced with permission from Angus DC, Clermont G, Watson RS, Linde-Zwirble WT, Clark RH, Roberts MS. Cost-effectiveness of inhaled nitric oxide in the treatment of neonatal respiratory failure in the United States, Pediatrics December 2003;112(6 pt 1):1351-1360.15)
Cost-effectiveness analysis is endorsed by both the United States Public Health Service Panel on Cost-Effectiveness in Health and Medicine (PCEHM) and the ATS as the primary method by which to measure the costs and effects of health care programs and medical therapies.7,8