Abstract
INTRODUCTION: Oral pre-exposure prophylaxis (PrEP) has been scaled up; however, data from real-world settings are limited. We studied oral PrEP preference, uptake, adherence and continuation among adolescent girls and young women (AGYW) vulnerable to HIV in sub-Saharan Africa. METHODS: We conducted a prospective cohort study among 14- to 24-year-old AGYW without HIV who were followed for 12 months in Kampala, Uganda. Within at least 14 days of enrolment, they received two education sessions, including demonstrations on five biomedical interventions that are; available (oral PrEP), will be available soon (long-acting injectable PrEP and anti-retroviral vaginal ring) and in development (PrEP implant and HIV vaccine). Information included mode and frequency of delivery, potential side effects and method availability. Volunteers ranked interventions, 1 = most preferred to 5 = least preferred. Oral PrEP was "preferred" if ranked among the top two choices. All were offered oral PrEP, and determinants of uptake assessed using Poisson regression with robust error variance. Adherence was assessed using plasma tenofovir levels and self-reports. RESULTS: Between January and October 2019, 532 volunteers were screened; 285 enrolled of whom 265 received two education sessions. Mean age was 20 years (SD±2.2), 92.8% reported paid sex, 20.4% reported ≥10 sexual partners in the past 3 months, 38.5% used hormonal contraceptives, 26.9% had chlamydia, gonorrhoea and/or active syphilis. Of 265 volunteers, 47.6% preferred oral PrEP. Willingness to take PrEP was 90.2%; however, uptake was 30.6% (n = 81). Following enrolment, 51.9% started PrEP on day 14 (same day PrEP offered), 20.9% within 30 days and 27.2% after 30 days. PrEP uptake was associated with more sexual partners in the past 3 months: 2-9 partners (aRR = 2.36, 95% CI: 1.20-4.63) and ≥10 partners (aRR 4.70, 95% CI 2.41-9.17); oral PrEP preference (aRR 1.53, 95% CI 1.08-2.19) and being separated (aRR 1.55, 95% CI 1.04-2.33). Of 100 samples from 49 volunteers during follow up, 19 had quantifiable tenofovir levels (>10 μg/L) of which only three were protective (>40 μg/L). CONCLUSIONS: Half of AGYW preferred oral PrEP, uptake and adherence were low, uptake was associated with sexual behavioural risk and oral PrEP preference. Development of alternative biomedical products should be expedited to meet end-user preferences and, community delivery promoted during restricted movement.