| Close to three million people become infected with HIV annually. The failure to prevent the continued spread of STI and HIV can be attributed, in part, to the failure to incorporate the social context of disease and the community environment into prevention programming.;The Encontros STI/HIV prevention intervention with sex workers in Brazil included both clinical and social strategies designed to address sex workers on the individual, interpersonal, and community level, including modification of the social environment around sex work and HIV. We hypothesized that participation in individual and collective intervention activities would increase condom use and reduce incident STI through receiving and internalizing prevention messages, through improved access to condoms, and by improving negotiation skills. We also hypothesized that behavior change would simultaneously be enabled by changing the environment around sex work and HIV prevention, through enhanced social cohesion, participation in social networks, access to social and material resources, and stigma reduction.;Inverse probability weighting (IPW) was used to evaluate the effects of participation in the program on subsequent behaviors and incident STI. This method mitigates biases due to self-selection, missing data, and confounding in longitudinal, observational studies and permits estimation of causal effects in non-randomized studies.;Participation in the intervention was protective against incident STI and resulted in increased reporting of consistent condom use. Participation was associated with a higher odds of subsequent consistent condom use with new clients (OR; 1.6, 95% CI: 0.9-2.8), regular clients (1.9, 1.1-3.3), and nonpaying partners (1.5, 0.9-1.5). The odds of an incident STI were reduced for participants in the current interval (.50, 0.2-1.0) and in subsequent intervals (.46, 0.2-1.3). While there was little evidence that increases in social cohesion and reported participation in networks directly reduced the incidence of STI or increased condom use, results suggest that social factors acted synergistically with the intervention, such that improved levels of social environmental factors enhanced the impact of the project on some outcomes.;This analysis is one of the first to demonstrate that a multi-level intervention focusing on modifying the social environment can reduce subsequent STIs and change condom use behaviors among sex workers. |
