Prevention of sexually transmitted HIV infection: A meta-analytic review of teh behavioral outcome literature.

Social learning theory-based models have recently provided the foundation for a series of twelve controlled human immunodeficiency virus (HIV) risk reduction intervention studies that have examined sexual behavior change. These interventions have been tested with adolescents, gay and bisexual men, inner-city women, college students, and seriously mentally ill adults. We report the first meta-analysis of these intervention studies. We found that, as expected, the mean weighted effect of HIV-risk reduction interventions on behavioral outcomes was positive and strongly significant (d+=0.25). Moreover, the studies' effect sizes were consistently positive, ranging from 0.11 to 0.53, and were largest when the outcomes were measured close in time to the intervention. We discuss other methodological challenges that, if solved, should enhance the success of future HIV-risk reduction interventions.