Abstract
IMPORTANCE: Preexposure prophylaxis (PrEP) is highly effective at preventing HIV-1 when taken as prescribed; however, barriers to uptake persist. OBJECTIVE: To assess clinical practice formulary coverage and PrEP uptake barriers and the association between those factors and cumulative HIV-1 incidence. DESIGN, SETTING, AND PARTICIPANTS: Retrospective observational cohort study from January 2021 to December 2023 using US pharmacy claims data from HIV-1-negative adults who received daily oral brand emtricitabine and tenofovir disoproxil fumarate (F/TDF), generic F/TDF (gF/TDF), or emtricitabine and tenofovir alafenamide fumarate (F/TAF), or intramuscular injection of cabotegravir every 2 months, for PrEP (after a 180-day washout period for previous PrEP use). EXPOSURES: Formulary coverage evaluated by claim rejection reasons, dispensing status (dispensed, rejected, or abandoned), payer types, and copayment costs. MAIN OUTCOMES AND MEASURES: HIV-1 incidence was described based on PrEP dispensing status. Logistic regression was used to estimate odds ratios (ORs) and 95% CIs of dispensing PrEP claims across key demographic and formulary characteristics. RESULTS: Overall, 399 902 individuals with 2 680 545 PrEP claims were analyzed. A total of 348 773 (87.2%) were male; 281 452 (70.4%) were aged 18 to 39 years; 42 549 10.6%] lived in predominantly Hispanic neighborhoods, 4114 [1.0%] lived in predominantly non-Hispanic Asian and Pacific Islander neighborhoods, 20 214 [5.1%] lived in predominantly non-Hispanic Black neighborhoods, and 238 292 [59.6%] lived in predominantly non-Hispanic White neighborhoods). Of 265 833 individuals with 12 or more months of follow-up, 238 271 (89.6%) had 1 or more PrEP claims dispensed (DISP), 17 917 (6.7%) never had PrEP dispensed with 50% or more of claims rejected by payer (ND-R), and 9645 (3.6%) never had PrEP dispensed with more than 50% of claims abandoned (ND-A). Individuals in the ND-R and ND-A groups had higher HIV-1 incidence (662 [3.7%] and 270 [2.8%], respectively) vs the DISP group (4811 [2.0%]). In the DISP group, a 1- to 7-day dispensation delay was associated with higher odds of acquiring HIV-1 (OR, 1.29; 95% CI, 1.13-1.47). Most claims were for gF/TDF (1 400 389 claims [52.2%]); compared with gF/TDF, the likelihood of claims being dispensed was lower for other PrEP regimens (brand F/TDF: OR, 0.36; 95% CI, 0.36-0.37; cabotegravir: OR, 0.52; 95% CI, 0.49-0.55; F/TAF: OR, 0.93; 95% CI, 0.91-0.95). As cost sharing increased, the likelihood of dispensation decreased. The most common reason for claim rejection was that the product was not covered by the formulary (49 690 rejected claims [29.2%]). CONCLUSIONS AND RELEVANCE: In this cohort study, HIV-1 incidence was higher in individuals with abandoned or rejected PrEP claims compared with those with dispensed claims. These findings highlight the importance of targeted strategies to improve PrEP claim dispensation and remove payer barriers to reduce HIV-1 transmission.