Cost-Effectiveness of an HIV/AIDS Prevention Intervention for Gay Men

The present study sought to determine the cost per discounted quality-adjusted-life-year (QALY) saved by a small group workshop-format, cognitive-behavioral HIV-prevention intervention for gay men. The methodology employed was a retrospective cost-utility analysis of the behavioral intervention. The ability of the intervention to effect HIV-related behavior change was previously assessed in a randomized controlled trial. In the original trial, clients were recruited from gay bars, health department clinics, and other community settings in metropolitan area of 400,000 residents; the intervention was delivered in a medical school outreach setting. The participants were 104 gay men; 87% of the clients identified their race/ethnicity as White, 13% as ethnic minority. The experimental intervention was comprised of 12 sessions and provided HIV-related risk behavior education, self-management and sexual assertion training, and development of reliable and positive social support networks. The comparison condition was a wait-list control group. The main outcome measure in our retrospective cost-utility analysis was “cost per discounted QALY saved.” Under base case assumptions, the cost of the intervention was $24,000 (rounded to the nearest thousand). The discounted medical costs averted by preventing HIV infection were $42,000. Approximately 5.5 discounted QALYs were saved. Hence the intervention is cost-saving under base case assumptions (i.e., the cost per discounted QALY saved ratio is less than zero). The results are generally robust to changes in cost-utility analysis model parameters and assumptions. Because the intervention is cost-saving under base case assumptions, it compares favorably to other health service interventions in which society currently invests. Behavioral interventions such as the one examined here should receive serious consideration for investment by public health decision makers allocating fiscal resources for health services.