Knowledge of HIV status is a necessary pre-condition for most HIV interventions, including treatment as well as biomedical and behavioral prevention interventions. We used data from a female sex worker (FSW) cohort in three Zambian transit towns to understand the effect that knowledge of HIV status has on FSWs' HIV risk-related sexual behaviors with clients. The cohort was formed from an HIV self-testing trial that followed participants for 4 months. Participants completed three rounds of data collection at baseline, 1 month, and 4 months where they reported their perceived knowledge of HIV status, number of clients on an average working night, and consistent condom use with clients. We measured the effect of knowledge of HIV status on participants' sexual behaviors by using linear regression models with individual fixed effects. The majority of the 965 participants tested for HIV at least once during the observation period (96%) and changed their knowledge of HIV status (79%). Knowledge of HIV status did not affect participants' number of clients, but it did affect their consistency of condom use. Compared with unknown HIV status, knowledge of HIV-negative status significantly increased participants' consistent condom use by 8.1% points [95% confidence interval (CI): 2.7-13.4, p = 0.003] and knowledge of HIV-positive status increased participants' consistent condom use by 6.1% points (95% CI: -0.1 to 12.9, p = 0.08); however, this latter effect was not statistically significant. FSWs in Zambia engaged in safer sex with clients when they learned their HIV status. The expansion of HIV testing programs may serve as a behavioral HIV prevention measure among FSWs.
Keywords: HIV; key and vulnerable populations; sex workers; sexual behaviors; testing; women.