Abstract
Young women who sell sex (YWSS) in Africa are particularly vulnerable to HIV acquisition and may benefit from daily oral pre-exposure prophylaxis (PrEP) but face multiple barriers to uptake. We identified temporal patterns of PrEP continuity/retention and associated characteristics within a cohort of YWSS in Uganda. Cisgender women aged 18-24 years who exchanged sex for money, goods, or services were recruited into a PrEP demonstration study in peri-urban Mukono District. We used group-based trajectory modeling to discern temporal patterns of PrEP continuity over 12 months of follow-up and multinomial logistic regression to identify baseline predictors of PrEP retention trajectories. Among 833 YWSS initiating PrEP, a four-group trajectory model emerged as best-fitting: (1) Rapid PrEP discontinuity (39.7%), characterized by high PrEP program attrition by 3 months; (2) Delayed PrEP discontinuity (24.2%), characterized by PrEP continuation through 6 months; (3) Immediate PrEP discontinuity (19.1%), characterized by high PrEP discontinuation in the month following initiation; and (4) High PrEP continuity (17.0%), characterized by sustained PrEP engagement throughout follow-up. In general, shorter- and longer-term PrEP continuation was associated with older age, higher education, pre-coital alcohol/drug use, and having 10 + paying partners in the past 3 months. Perceived HIV risk and related HIV vulnerabilities helped distinguish observed PrEP retention trajectories, predicting the timing of PrEP discontinuation among YWSS.