Study Design and Measurement Methods of Sexual Networks in Ghana

Samuel Jenness, University of Washington
Susan L. Cassels, University of Washington

Social networks are central to population-level transmission of HIV, particularly in lower-prevalence regions like West Africa in which there are disparities in infection within the general population. Network-related data, however, are needed to investigate these issues. We recently completed a cross-sectional study of sexually active adults in Agbogbloshie, an urban slum in Accra, Ghana in which we collected egocentric and partial network data. Here we present our unique methodological approach to sampling design, network-based sexual risk measurement, and considerations on selection bias. Statistical estimates from this data will be used to mathematically model the impact of migration and sexual partner concurrency on HIV transmission dynamics. Despite the complexities of working in an urban slum environment, our field work activities were successful in generating a large sample in a scientifically rigorous manner with novel measurements, including an HIV-1/2 diagnostic assay.