Persistence on HIV preexposure prophylaxis medication over a 2-year period among a national sample of 7148 PrEP users, United States, 2015 to 2017.

INTRODUCTION: Persistence on preexposure prophylaxis for HIV prevention (PrEP) medication has rarely been reported for periods greater than one year, or in real-world settings. This study used pharmacy fill records for PrEP users from a national chain pharmacy to describe persistence on PrEP medication over a two-year period, and to explore correlates with PrEP medication persistence in a real-world setting.

METHODS: We analysed de-identified pharmacy fill records of 7148 eligible individuals who initiated PrEP in 2015 at a national chain pharmacy. A standard algorithm was employed to identify TDF-FTC use for PrEP indication. We considered three time periods for persistence, defined as maintaining refills in PrEP care: year 1 (zero to twelve months), year 2 (thirteen to twenty-four months) and initiation to year 2 (zero to twenty-four months). Individuals with 16 or more days of TDF-FTC PrEP dispensed in a 1-month period for at least three-quarters of a given time period (e.g. nine of twelve months or eighteen of twenty-four months) were classified as persistent on PrEP medication for the period.

RESULTS: Persistence was 56% in year 1, 63% in year 2 and 41% from initiation to year 2. Individuals aged 18 to 24 had the lowest persistence, with 29% from initiation to year 2. Men had higher persistence than women, with 42% compared to 20% persistent from initiation to year 2. Individuals with commercial insurance and individuals who utilized a community-based specialty pharmacy from the national chain also had higher persistence. Male gender, age >18 to 24 years, average monthly copay of $20 or less, commercial insurance, and utilization of a community-based specialty pharmacy were positively associated in adjusted models with persistence in year 1 and from initiation to year 2; the same correlates, with the exception of utilization of a community-based specialty pharmacy, were associated with higher persistence in year 2.

CONCLUSIONS: We found substantial non-persistence on PrEP medication in both year 1 and year 2. Across the entire 2-year period, only two out of every five users persisted on PrEP. Demographic, financial and pharmacy factors were associated with persistence. Further research is needed to explore how social, structural or individual factors may undermine or enhance persistence on PrEP, and to develop interventions to assist persistence on PrEP.