Condom use among young women who sell sex in Zimbabwe: a prevention cascade analysis to identify gaps in HIV prevention programming

INTRODUCTION: Adolescent girls and young women (AGYW), including those who sell sex in sub‐Saharan Africa, are especially vulnerable to HIV. Reaching them with effective prevention is a programmatic priority. The HIV prevention cascade can be used to track intervention coverage, and identify gaps and opportunities for programme strengthening. The aim of this study was to characterise gaps in condom use and identify reasons underlying these gaps among young women who sell sex (YWSS) in Zimbabwe using data from enrolment into an impact evaluation of the DREAMS programme. DREAMS provided a package of biomedical, social and economic interventions to AGYW aged 10 to 24 with the aim of reducing HIV incidence. METHODS: In 2017, we recruited YWSS aged 18 to 24 using respondent‐driven sampling in six sites across Zimbabwe. We measured knowledge about efficacy of, access to, and effective (consistent) use of condoms with the most recent three sexual partners, separately by whether YWSS self‐identified as female sex workers (FSW) or not. Among YWSS without knowledge about efficacy of, not having access to, and not effectively using condoms, we described the potential reasons underlying the gaps in the condom cascade. To identify socio‐demographic characteristics associated with effective condom use, we used logistic regression modelling. All analyses were RDS‐II weighted and restricted to YWSS testing HIV‐negative at enrolment. RESULTS: We enrolled 2431 YWSS. Among 1842 (76%) YWSS testing HIV‐negative, 66% (n = 1221) self‐identified as FSW. 89% of HIV‐negative YWSS demonstrated knowledge about efficacy of condoms, 80% reported access to condoms and 58% reported using condoms consistently with the three most recent sexual partners. Knowledge about efficacy of and effective use of condoms was similar regardless of whether or not YWSS self‐identified as FSW, but YWSS self‐identifying as FSW reported better access to condoms compared to those who did not (87% vs 68%; age‐ and site‐adjusted (adjOR) = 2.69; 95% CI: 2.01 to 3.60; p < 0.001). Women who reported experiencing sexual violence in the past year and common mental disorder in the past week were less likely to use condoms consistently (43% vs. 60%; adjOR = 0.49; 95% CI: 0.35 to 0.68; p < 0.001) and (51% vs. 61%; adjOR = 0.76; 95% CI: 0.60 to 0.97; p = 0.029), respectively. CONCLUSIONS: Despite high knowledge about efficacy of and access to condoms, there remain large gaps in self‐reported consistent condom use among YWSS. Addressing the structural determinants of YWSS’ inconsistent condom use, including violence, could reduce this gap. YWSS who do not self‐identify as FSW have less access to condoms and may require additional programmatic intervention.