Purpose: Thirty-five candidate HIV vaccines are undergoing evaluation in clinical trials and more are in development. How best to use a vaccine, should one become available, is not known. We aimed to evaluate the cost-effectiveness of implementing mass HIV vaccination programs targeted to the general population, high-risk groups, or both.

Methods: We developed a deterministic compartmental model of the HIV epidemic to estimate the effects and cost-effectiveness of providing preventive HIV vaccines to different risk groups. We estimated all health costs and benefits, in terms of quality-adjusted life years (QALYs). We allowed for variations in vaccine cost, duration, efficacy (i.e., reducing HIV acquisition in uninfected individuals), and potential therapeutic effects (i.e., reducing infectivity in infected individuals). We evaluated different vaccination strategies that target high-risk groups (injection drug users, men who have sex with men, and commercial sex workers) and the general population. We applied our model using U.S. data; however, the modeling framework can be extended to other geographic regions with potentially different risk groups and epidemic stage. Our base case assumed a preventive vaccine with 75% efficacy, lifetime duration, and price of $1000.

Results: All vaccination strategies targeted to high-risk groups were cost-saving; they reduced total costs while increasing total QALYs. Vaccinating 50% of uninfected high-risk individuals prevented 2.2 million infections over 20 years, saved $50 billion (discounted) and added 2.6 million QALYs. The vaccine price must exceed $6,500 in order for this strategy to no longer be cost-saving. Strategies that vaccinate 10-90% of the uninfected general population prevented 300,000-550,000 infections and cost $270,000-310,000 per QALY gained. For a strategy that vaccinates 50% of the general population, the vaccine price must decrease to $235 for the cost-effectiveness to reach $50,000 per QALY gained. Vaccinating 50% of both high-risk individuals and the general population prevented 2.4 million infections and cost $53,000 per QALY gained, or $15,000 per QALY gained if the vaccine price was halved to $500.

Conclusions: Targeted HIV vaccination of high-risk groups in the U.S. prevents more HIV infections and is significantly more cost-effective than exclusively vaccinating the general population. Strategies that vaccinate all uninfected individuals regardless of risk group prevent the greatest number of infections but are cost-effective only for low vaccine prices.