Modeling cost-effectiveness of HIV prevention programs

Abstract

There have been important advances in the literature regarding modeling the cost-effectiveness of HIV prevention programs. Frequently, costs and/or effectiveness are modeled with a combination of parameters since they cannot be measured directly. In most HIV prevention interventions, it is not possible to directly measure effectiveness due to changes in HIV incidence. Instead, effectiveness in preventing HIV is usually assessed using mathematical models that translate measurements of behavior change into estimates of the number of HIV infections averted. This measure of HIV prevention effectiveness is dependent upon the risk levels in the communities (e.g., prevalence of HIV, sexual or drug-injection behaviors), and on the behavioral effectiveness of the intervention in changing participants’ risk behaviors. The number of HIV cases averted can be calculated using probabilistic models of HIV transmission, which compare the sexual risk behaviors of individuals before and after the intervention. Moreover, cost data are not typically collected alongside HIV prevention programs. An overview of published research on modeling cost-effectiveness of HIV prevention programs is provided, a list of the important outcomes is presented and methodological issues associated with modeling such programs are discussed in this review.

• Bernoulli model
• cost-effectiveness
• HIV prevention
• Markov model
• model uncertainty
• parameter uncertainty
• probabilistic model
• sexual transmission
• simulation model
• vertical transmission