PrEP initiation and discontinuation among transgender women in the United States: a longitudinal, mixed methods cohort study

INTRODUCTION: Transgender women in the United States experience high HIV incidence and suboptimal Pre-exposure prophylaxis (PrEP) engagement. We sought to estimate PrEP initiation and discontinuation rates and characterize PrEP discontinuation experiences among a prospective cohort of transgender women. METHODS: Using a sequential, explanatory, mixed-methods design, 1312 transgender women at risk for HIV acquisition were enrolled from March 2018 to August 2020 and followed through July 2022 (median follow-up 24 months; interquartile range 15-36). Cox regression models assessed predictors of initiation and discontinuation. In-depth interviews were conducted among 18 participants, including life history calendars to explore key events and experiences surrounding discontinuations. Qualitative and quantitative data were integrated to generate typologies of discontinuation, inform meta-inferences and facilitate the interpretation of findings. RESULTS: 21.8% (n = 286) of participants reported taking PrEP at one or more study visits while under observation. We observed 139 PrEP initiations over 2127 person-years (6.5 initiations/100 person-years, 95% CI: 5.5-7.7). Predictors of initiation included identifying as Black and PrEP indication. The rate of initiation among those who were PrEP-indicated was 9.6 initiations/100 person-years (132/1372 person-years; 95% CI: 8.1-11.4). We observed 138 PrEP discontinuations over 368 person-years (37.5 discontinuations/100 person-years, 95% CI: 31.7-44.3). Predictors of discontinuation included high school education or less and initiating PrEP for the first time while under observation. Four discontinuation typologies emerged: (1) seroconversion following discontinuation; (2) ongoing HIV acquisition risk following discontinuation; (3) reassessment of HIV/STI prevention strategy following discontinuation; and (4) dynamic PrEP use coinciding with changes in HIV acquisition risk. CONCLUSIONS: PrEP initiation rates were low and discontinuation rates were high. Complex motivations to stop using PrEP did not consistently correspond with HIV acquisition risk reduction. Evidence-based interventions to increase PrEP persistence among transgender women with ongoing acquisition risk and provide HIV prevention support for those who discontinue PrEP are necessary to reduce HIV incidence in this population.