Background Women around the world use intra-vaginal practices (IVPs) for varying reasons. IVPs can disrupt the vaginal microbiome and induce inflammation, increasing susceptibility to bacterial vaginosis (BV), a risk for HIV acquisition. Limited research has focused on IVPs and their correlates among adolescent girls and young women (AGYW) in South Africa, a key population for HIV prevention efforts.

Methods We used cross-sectional baseline survey data from 253 HIV-negative or HIV-status unknown AGYW (16–24 years) enrolled in AYAZAZI, a community-based cohort study in Soweto and Durban, South Africa (2014–2016). We measured IVP use in the past 30 days for 11 IVPs (yes vs no) and estimated the prevalence of using ≥= 1 IVP in the past 30 days (yes vs no). Using existing literature, we identified 18 determinants of IVP use. We estimated the total causal effect of each determinant on any IVP use in the past 30 days using separate logistic regression models adjusted for confounding within the counterfactual framework of causal inference.

Results Mean age was 19 years, 67.2% were in school, and 81.8% ever had sex. Overall IVP prevalence was 76.7% (59% vs 81% among those who had never vs ever had sex; p=0.016). The most common IVPs included: washing intravaginally with water only (66%), with soap (41%), and with other products including towels or sponges (41%). AGYW who reported binge drinking (aOR=7.6; 95%CI: 3.2–18.4), a higher monthly income (≥R1601 vs ≤ R400; aOR=7.4; 95%CI: 1.6–33.9), engaged in transactional sex (aOR=4.4; 95%CI: 1.0–19.1), and/or reported any symptoms of genital tract infections (aOR=4.9; 95%CI: 1.0–23.0) had higher adjusted odds of using IVPs.

Conclusion Over three-quarters of South African AGYW in this study reported IVPs with differences by socio-economic, behavioral, and clinical characteristics. Further investigation of IVP motives and implications for BV and HIV acquisition risk among young women are warranted.

Disclosure No significant relationships.