TO PREVENT OR TREAT: THE COST-EFFECTIVENESS OF EXPANDED HIV SCREENING OR ANTIRETROVIRAL THERAPY

Purpose:  Determining the optimal allocation of resources among competing HIV interventions can increase population-wide health benefits.  By expanding HIV prevention efforts now, we can also reduce the need for treatment in the future.  We aimed to evaluate the cost-effectiveness of programs offering expanded HIV screening, increased antiretroviral therapy, or a combination of both.

Methods:  We developed a dynamic HIV epidemic model, which captures the effects of HIV transmission, disease progression, and mortality on health outcomes.  We stratified the population based on gender, risk behavior (men who have sex with men, injection drug users, and low-risk individuals), disease state, identification status, and treatment status, if infected.  We included the effects of HIV screening and counseling, and treatment with highly active antiretroviral therapy (HAART) on biological (e.g., infectivity, quality-of-life, mortality) and behavioral (e.g., number of sexual partners) parameters, based on published data.  We applied the model using U.S. data, and we estimated HIV prevalence and new infections over time, and the health benefits and expenditures associated with each HIV intervention strategy.

Results:  An HIV screening program that increased the fraction of unidentified individuals screened each year from 20% to 60% prevented 120,000 new infections over 20 years, and added 2.2 million quality-adjusted life years (QALYs) at a cost of $14,800 per QALY gained relative to the status quo.  A program that instead expanded access to HAART from 50% to 75% prevented 258,000 infections, adding 5.1 million QALYs at a cost of $14,200 per QALY gained relative to the status quo.  A combination strategy of expanded HIV screening and treatment offered the greatest gain in health benefits, at a similar cost-effectiveness ratio.  The cost-effectiveness of each strategy was sensitive to the local HIV prevalence level and the effectiveness of screening and counseling at reducing risky sexual behavior.

Conclusions:  Programs that focus exclusively on expanding HIV screening or treatment result in similar cost-effectiveness ratios.  In order to fully realize the benefits of each intervention, the complementary program must also be present.  Estimating the effectiveness of a counseling program accompanying HIV screening at reducing risky sexual behavior is an important consideration for determining the most efficient intervention portfolio.  Our general framework can be extended to include additional HIV interventions, such as male circumcision and pre-exposure prophylaxis.