Effect of a Peer Support Intervention for Oral Pre‐Exposure Prophylaxis Uptake and Adherence Among Adolescent Girls and Young Women at Risk of HIV Acquisition in Kampala, Uganda: A Randomized Controlled Trial
Yunia Mayanja, Andrew Abaasa, Kenneth Roger Katumba, Matthew Spinelli, Bernard Kikaire, Monica Gandhi, Kyriaki Kosidou, Anna Mia Ekstrom, Eugene RuzagiraABSTRACT
Introduction
Daily oral pre‐exposure prophylaxis (PrEP) use remains low among adolescent girls and young women (AGYW). We evaluated whether a peer support intervention improved daily oral PrEP uptake and adherence among AGYW at high risk of HIV acquisition in Kampala, Uganda.
Methods
We conducted an open‐label randomized controlled trial (January 2023−June 2024) among 14‐ to 24‐year‐olds engaged in sex work. Eligible participants were randomized 1:1 to receive either health worker counselling only (control) or supplemented with peer support (intervention). Peer leaders delivered co‐created peer education sessions starting 1 month after enrolment. Participants were followed for 12 months. Co‐primary outcomes were PrEP uptake (yes/no) and adherence, evaluated using two separate measures; optimal if: ≥85% pill‐counts (primary measure); positive point‐of‐care urine tenofovir test. Both arms received adherence counselling based on urine results. Secondary outcomes: PrEP persistence (time to discontinuation) and cost‐effectiveness (cost per disability adjusted life year/DALY averted). Analysis used logistic regression (uptake); generalized estimating equations (adherence), Cox regression (persistence) and incremental cost‐effectiveness ratio (ICER) below US$537 per DALY averted was considered cost‐effective.
Results
We enrolled 394 participants (197 per arm) with a mean age of 19.4 (SD ±2.3) years. The intervention did not impact PrEP uptake (OR 1.15; 95% CI 0.77−1.71) but improved adherence, pill count (OR 2.20; 95% CI 1.23−3.95) and urine assay (OR 2.18; 95% CI 1.22−3.91). Optimal adherence was low but significantly higher in the intervention by both measures: pill count (intervention = 54.8%, control = 39.6%, p = 0.008), urine assay (intervention = 54.8%, control = 40.8%, p = 0.001). Median duration before PrEP discontinuation was 38 days (IQR 0–319 days); intervention (56 days), control (29 days). The intervention had no effect on persistence (HR 0.86; 95% CI 0.64−1.14). It was cost‐effective with the urine tenofovir assay (ICER = $459) but not with pill counts (ICER = $581).
Conclusions
This peer‐led intervention improved PrEP adherence among AGYW in Kampala and is recommended in high‐incidence settings. Urine tenofovir testing can be used to assess adherence if affordable. Enhanced peer support strategies are needed until long‐acting PrEP becomes accessible and acceptable. Future research should optimize timing, intensity and delivery models of peer support interventions and integrate them with long‐acting PrEP to improve sustainability.