Sexual mixing and the risk environment of sexually active transgender women: data from a respondent-driven sampling study of HIV risk among transwomen in San Francisco, 2010

Background: Research on the sexual networks of transwomen is central to explaining higher HIV risk for this population. This study examined HIV risk behaviors and sexual mixing patterns of transwomen by demographic and HIV-related risk behaviors. Methods: Data were obtained from a 2010 study of HIV risk for transwomen in San Francisco. Assortativity by race, partner type, HIV serostatus, and IDU across sexual networks was calculated using Newman's assortativity coefficient (NC). Multivariable generalized estimating equations (GEE) logistic regression models were used to evaluate associations between unprotected anal intercourse with race and HIV serostatus, partner-IDU status and relationship type discordance while adjusting for the HIV status of transwomen. Results: There were 235 sexually active transwomen in this study, of whom 104 (44.3%) were HIV-positive and 73 (31.1%) had a history of injection drug use. Within the 575 partnerships, African American/black and Latina transwomen were the most racially assortative (NC 0.40, 95% CI 0.34-0.45, and NC 0.43, 95% CI 0.38-0.49, respectively). In partnerships where the partner's HIV status was known (n = 309, 53.7%), most transwomen were in sexual partnerships with people of their same known serostatus (71.8%, n = 222). In multivariable analyses, unprotected anal intercourse was significantly associated with primary partners, having a sexual partner who was an injection drug user, and sexual partner seroconcordance. Conclusions: Public health efforts to reduce transwomen's HIV risk would likely benefit from prioritizing prevention efforts to risk reduction within IDU-discordant and primary partnerships, determining risks attributable to sexual network characteristics, and actively addressing injection drug use among transwomen.