Background Amid Zimbabwe’s decline in overall HIV prevalence, the relative importance of key populations to ongoing transmission might be expected to change. We examine variation in the social organisation of sex work and assess the characteristics associated with HIV infection among sex workers (SW) in rural Zimbabwe.
Methods We collected and analysed cross-sectional data on 545 women who reported sex for payment - including cash-per-act and informal credit-based arrangements - and were recruited using snowball and location-based methods in October-December 2010. We used successive logistic regression models to explore determinants of HIV infection among women reporting venue-based sex work (e.g. bars - includes women recruited onsite) and women reporting non-venue based sex work.
Results HIV prevalence was significantly higher in venue SW compared to non-venue women (55.5% vs. 41.9%, p < 0.01). Numbers of clients reported were low, but more venue SW reported 2 or more clients in the last fortnight (38% vs. 7.6%, p < 0.01); consistent condom-use during commercial acts was similar (60% and 68%, p > 0.1). More venue SW reported ever travelling away from where they live to sell sex (27% vs. 12%, p < 0.01), initiated sex work earlier (age 25 vs. 28 years, p < 0.01), and sold sex for longer (6 vs. 5 years, p < 0.05). Factors associated with HIV infection were type of sex work (aOR: 2.2, 95% CI: 1.3–3.8 vs. non-venue), and SW not requesting condom-use (aOR: 3.0, 95% CI: 1.15–7.85), after controlling for basic demographic differences.
Conclusion Venue SW were more likely to be HIV-positive than non-venue women, although in both groups HIV prevalence was substantial, and consistent condom use low. High risk of HIV among women not requesting condom use highlights the importance to renew prevention efforts in this population. The results emphasise the need to understand SW more broadly.
- Female sex work