Meta-analytic evidence of HIV behavioral prevention interventions for drug users

Background: The first two decades of the HIV epidemic produced scores of primary research studies of risk reduction interventions for drug users. Meta-analysis allows for assessing and summarizing effects of interventions. Several meta-analyses (MAs) of such interventions for drug users have been published allowing for assessing the meta-analytic evidence.

Methods: We reviewed all 6 MAs of behavioral HIV interventions for drug users published during 2000-2007. We describe the purpose of these MAs and included primary studies (number and publication period) and assess the meta-analytic results.

Results: Each MA comprised 18 to 47 primary studies published during 1970-2004. The 6 MAs covered different primary studies because they covered different research questions or publication periods. The MAs assessed the effect of risk reduction interventions for in- (3 MAs) or out-of (3 MAs) substance-abuse-treatment drug users, using needle exchange programs (1 MA), and enrolling in contingency management of rewards for stopping/avoiding drug use (2 MAs). Summary effects indicated a 20%-30% change in odds of drug and sex risk, measured as increase in protective behaviors or decrease in risky behaviors. More efficacious interventions included skills-training for avoiding risk, and providing interventions to in-treatment drug users later in treatment.

Conclusion: The 6 MAs covered various research questions, interventions, and recruitment and delivery settings, and provided a wide perspective and moderate favorable effect on risk reduction. Meta-analytic evidence highlights the need for developing stronger interventions, linking risk reduction to reduction in HIV transmission, assessing cost-effectiveness of interventions, and implementing evidence-based interventions.