OBJECTIVES:

Pre-exposure prophylaxis (PrEP) is effective in preventing HIV transmission and acquisition. However, poor adherence may limit its efficacy. Thus, we examined the association between adherence trajectories of PrEP and the risk of HIV acquisition among PrEP users.

METHODS:

We conducted a retrospective cohort study, using data from the 2011-2020 MarketScan database of commercially-insured individuals in the U.S. aged 12-64 prescribed ≥ 30 days of PrEP use (i.e., tenofovir-disoproxil-fumarate with emtricitabine or tenofovir-alafenamide with emtricitabine). To ensure PrEP use, we excluded persons with any diagnosis codes or prescriptions for HIV or hepatitis B virus (HBV) during the year before the first PrEP prescription date and prescriptions (<30 days) for post-exposure prophylaxis use. We used group-based trajectory modeling (GBTM) to identify distinct adherence patterns of PrEP using 15-day proportion of days covered over 1 year, starting from day 31 after PrEP initiation. We examined the association between the adherence trajectories and the risk of HIV acquisition during the GBTM measurement period, using inverse probability treatment weighted (IPTW) Cox proportional hazard models.

RESULTS:

The study cohort consisted of 20,757 new PrEP users (96% males; 52% aged 18-34 years; 38% from the South). The final GBTM models identified four distinct PrEP adherence trajectories including nonadherent (23% of the cohort), rapidly-declining (12%), gradually-declining (20%), and consistently-adherent (44%) groups, with HIV incidence rates of 73, 38, 59, and 39 per 10,000 person-years, respectively. In IPTW Cox regression models, compared with consistently adherent group, nonadherent group had an increased HIV risk (HR 2.17 [1.06-4.44]). There were no statistically-significant differences in HIV risk between the rapidly-declining (HR 0.92 [0.29-2.93]) or gradually-declining (HR 1.72 [0.79-3.75]) adherence groups and the consistently adherent group.

CONCLUSIONS:

We found an increased risk of HIV among those who were nonadherent with PrEP, suggesting the need for interventions targeting barriers to adherence and PrEP access continuity.