INTRODUCTION: Adolescent girls and young women (AGYW) in Kenya have low pre-exposure prophylaxis (PrEP) initiation rates in part because of stigmatizing interactions with health care providers. Our recent randomized clinical trial of a standardized patient actor (SP) training intervention for providers found higher quality PrEP delivery at intervention sites
however, it was unclear whether improved service quality improved PrEP initiation. METHODS: This analysis used routine records from facilities participating in the randomized trial that aimed to improve provider communication and adherence to Kenyan guidelines when offering PrEP to AGYW. We used facility-level PrEP registers from May to December 2019 as the baseline period and December 2020 to June 2021 as the postintervention period. We used linear regression with percentage initiating as the outcome, intervention and baseline initiation levels as covariates, and the number eligible postintervention at each facility as frequency weights. RESULTS: Overall, 1375 AGYW presented to study sites, were eligible for PrEP, and were included in analyses (baseline: n = 706, postintervention: n = 669). Among 669 PrEP-eligible AGYW in the postintervention period (intervention: n = 360, control: n = 309), 591 (88.3%) initiated PrEP (intervention: n = 335, control: n = 256). PrEP initiation was 93.1% at intervention sites (range: 0%-100%) and 82.8% at control sites (range: 0%-100%). Adjusted for baseline initiation rates, initiation was 12.1% higher at intervention sites than at control sites ( P <
0.001, [95% CI: 0.09 to 0.15]). CONCLUSIONS: Our study found significant improvement in PrEP initiation among AGYW who presented to intervention facilities. SP training interventions that improve quality of service delivery for AGYW could lead to higher population-level PrEP coverage.