Background
Consistent use of daily oral preexposure prophylaxis (PrEP) during periods of HIV risk is essential for optimal HIV protection. This study estimated and compared persistence of oral PrEP among early and more recent cohorts of PrEP users using a national pharmacy database.
Methods
We analyzed data from the IQVIA Real-World Data—Longitudinal Prescription Database (2017−2023) to identify persons aged ≥16 years newly prescribed oral PrEP for >30 days. Two cohorts were constructed based on PrEP initiation periods: 2017−2018 and 2021−2022, each followed for 12 to 36 months post-initiation. Data from 2020 were excluded due to the COVID-19 pandemic that resulted in decreased PrEP use. Persistence was defined as the duration of continuous PrEP prescription fills until a >30 days gap. Non-persistence was the date when the prescribed supply would be depleted if taken daily. Individuals could restart PrEP after discontinuation and have multiple episodes. We reported median persistence duration by initiation cohort and used Kaplan-Meier survival curves to estimate time to non-persistence. We used Cox proportional hazards models to examine the associations between patient demographic characteristics and non-persistence.
Results
We identified 183,584 new PrEP users in 2017−2018 and 321,224 in 2021−2022. The median duration of initial PrEP use was 4.0 months (interquartile range [IQR] 1.0−12.0) for the 2017−2018 cohort and 3.0 months (IQR 2.0−10.0) for the 2021−2022 cohort. Only 42.4% of the early cohort and 37.9% of the recent cohort persisted for 6 months (Figure). About 35.6% restarted PrEP at least once, with a median second episode duration of 3.0 months across cohorts. Factors associated with non-persistence across cohorts included being female (adjusted Hazard Ratio [aHR] 1.62, 95% CI 1.61−1.64), younger age (16-24 years: aHR 1.65, 95% CI 1.62−1.68) compared to those aged 55-64, Black race (aHR 1.18, 95% CI 1.16−1.20) compared to White, having public health insurance (aHR 1.35, 95% CI 1.33−1.36) or paying cash (aHR 1.63, 95% CI 1.59−1.67) compared to using commercial insurance.
Conclusions
Less than half of new oral PrEP users persisted beyond 6 months, and persistence declined over time. This trend might reflect increased use of on-demand PrEP and transitions to injectable PrEP. Better understanding of on-demand PrEP uptake and adherence is needed to address persistence trends. Future studies can benefit from assessing the impact of switching between PrEP formulations on persistence.
Table or Figure
