THE COST-EFFECTIVENSS OF ALTERNATIVE HIV INTERVENTION PORTFOLIOS IN SOUTH AFRICA

Purpose: With 5.7 million people living with HIV in South Africa and 1,000 AIDS-related deaths occurring daily, reducing new infections and increasing access to highly active antiretroviral therapy (HAART) is a national health priority.  We aimed to evaluate the cost-effectiveness of programs to increase HIV screening, HAART, male circumcision, or a combination of the three interventions.

Method: We implemented a dynamic HIV transmission model, calibrated to epidemiologic data for a generalized heterosexual HIV epidemic in South Africa, to evaluate alternative combinations of three HIV interventions: HIV screening and counseling, treatment with HAART, and male circumcision.  Using available data, we included each intervention’s effect on transmission probabilities and sexual behavior estimates.  For each portfolio, we estimated HIV prevalence, new infections, discounted quality-adjusted life years (QALYs) and costs (2007 USD), and incremental cost-effectiveness ratios, assuming a 20-year time horizon.  Uncertainty on behavioral responses to each intervention was examined with one-way sensitivity analysis.

Result: With no increase in prevention or treatment levels beyond current levels, more than 10 million new infections occur over 20 years.  A program that optimistically expands HAART coverage from a current level of 28% to 75% prevents 250,000 infections at a cost of $433 per QALY gained.  Aggressively expanding HIV screening prevents 800,000-1.4 million new infections, at approximately $500 per QALY gained.  Circumcision programs reaching 10-20% of uncircumcised men each year could prevent 1.5-2.2 million infections, substantially reduce HIV prevalence among men and women (due to reduced secondary transmission), and cost less than $60 per QALY gained.  Even with a 30% increase in circumcision-related behavioral disinhibition, this program prevents a modest number of infections and remains cost-effective. Combination strategies prevent the greatest number of infections, but result in diminishing returns as additional programs are included.

Conclusion: In South Africa, a comprehensive portfolio of expanded HIV screening, HAART, and male circumcision is the best strategy for mitigating the virus’ spread and improving health outcomes.  Increased male circumcision is more cost-effective than exclusively expanding HIV screening and treatment, although the benefits may be attenuated if men significantly increase risky behavior.  We find that the costs and benefits of implementing multiple HIV interventions simultaneously are non-additive, and a dynamic epidemic model is necessary to accurately estimate the effect of compound interventions on epidemic and economic outcomes.