Men's Behavior Predicts Women's Risks for HIV/AIDS: Multilevel Analysis of Alcohol-Serving Venues in South Africa.
South Africa has among the highest rates of HIV infection in the world, with women disproportionately affected. Alcohol-serving venues, where alcohol use and sexual risk often intersect, play an important role in HIV risk. Previous studies indicate alcohol use and gender inequity as drivers of this epidemic, yet these factors have largely been examined using person-level predictors. We sought to advance upon this literature by examining venue-level predictors, namely men's gender attitudes, alcohol, and sex behavior, to predict women's risks for HIV. We recruited a cohort of 554 women from 12 alcohol venues (6 primarily Black African, and 6 primarily Coloured [i.e., mixed race] venues) in Cape Town, who were followed for 1 year across four time points. In each of these venues, men's (N = 2216) attitudes, alcohol use, and sexual behaviors were also assessed. Men's attitudes and behaviors at the venue level were modeled using multilevel modeling to predict women's unprotected sex over time. We stratified analyses by venue race. As predicted, venue-level characteristics were significantly associated with women's unprotected sex. Stratified results varied between Black and Coloured venues. Among Black venues where men reported drinking alcohol more frequently, and among Coloured venues where men reported meeting sex partners more frequently, women reported more unprotected sex. This study adds to the growing literature on venues, context, and HIV risk. The results demonstrate that men's behavior at alcohol drinking venues relate to women's risks for HIV. This novel finding suggests a need for social-structural interventions that target both men and women to reduce women's risks.
Pitpitan, EV; Kalichman, SC; Eaton, LA; Sikkema, KJ; Watt, MH; Skinner, D; Pieterse, D
Volume / Issue
Start / End Page
Electronic International Standard Serial Number (EISSN)
International Standard Serial Number (ISSN)
Digital Object Identifier (DOI)