Abstract
HIV pre-exposure prophylaxis (PrEP) is a highly effective biomedical prevention for HIV infections. PrEP persistence is critical to achieving optimal protection against HIV infection. However, little is known about PrEP persistence in the United States. This study utilized the Connecticut All-Payer Claims Database (APCD) to identify PrEP persistence among patients who filled their PrEP prescriptions in the state. The authors identified 1,576 PrEP patients who picked up PrEP prescriptions and extracted medical and pharmacy claims to evaluate a longitudinal cohort during 2012-2018 based on the Connecticut APCD. Patients who did not pick up medication for one consecutive month (ie, 30 days) were defined as discontinuing PrEP. Kaplan-Meier Survival Curve and proportional hazard regression were used to describe PrEP persistence. Of the 1,576 patients who picked up PrEP prescriptions, the median age was 32.0 (interquartile range [IQR]: 22.0-44.0). The majority were male individuals (93%). Of 1,040 patients who discontinued PrEP, 702 (67.5%) restarted PrEP at least once. The median time of PrEP persistence was 3 months (IQR: 1-6 months) for initial PrEP use. The median time on PrEP was also around 3 months in the following episodes of PrEP use. Being female, being on parent's insurance, and having high co-pays were associated with shorter periods of PrEP persistence. PrEP persistence was low among patients who picked up PrEP prescriptions. Although many patients restarted PrEP, persistence remained low during follow-up PrEP use and possibly led to periods of increased HIV risk. Effective interventions are needed to improve PrEP persistence and reduce HIV incidence.