The Rakai Project Research Group (RPRG) proposes to collaborate with the NIMH/ACTs/DCC by implementing a randomized, community-based trial of the Opinion Leaders (OL) intervention in rural Rakai District, Uganda. Rakai provides a unique setting for the trial: we have conducted multiple behavioral, epidemiological, and intervention studies in the district since 1989, resulting in detailed demographic, behavioral and STD/HIV data needed or trial design. We have extensive experience in combining behavioral data and biological specimen collection in the community, and have achieved high enrolment compliance and extended follow up greater than 75 percent. Over 80 percent of the African population is rural, and faces substantial risk of HIV. In Rakai, approximately 13 percent of all adults greater than or equal to age 13 years are HIV+. We propose to target and enrol all consenting persons aged 12-49 resident in 30 communities and to randomly assign communities to an intervention and control arm. Prior to implementing the intervention, we will conduct ethnographic and social networks research in the intervention arm to identify opinion leaders and develop intervention messages, including promotion of consistent condom use and reductions in numbers of sexual partners. Given that sexual behavior change and condom use in Rakai are constrained by lack of communication about sexual matters between partners, marriage between HIV discordant individuals, fertility desires, transactional factors governing sexual relations between young girls and older men, alcohol use and coercive sex, we propose to address such factors in the OL intervention messages, as guided by the ethnographic research. The OL intervention will be evaluated via a baseline and two follow up surveys at 12 and 24 months, conducted in the home in all study communities. We will measure intermediate behavioral variables (condom use, reported numbers of partners, other behaviors) and STD/HIV endpoints, including trichomonas prevalence and incidence, and HIV and HSV-2 prevalence and incidence. We will also compare the results of ethnographic research to network surveys as a means of identifying OLs, monitor OL turnover, and assess whether the intensity of message provision remains stable over time. Such data will assist in future programmatic replications of the OL intervention. The research design, target population and messages can be modified (ex: to include bars in Rakai trading centers) if desirable for the design of the overall OL collaborative study or for metanalysis of data from multiple sites.
