End-user input on early stage design of new HIV prevention approaches is critical to yielding products that overcome adherence challenges and ultimately achieve high uptake and use. In iPrevent, a novel, multi-phased study designed to engage youth to inform the product pipeline, we examined preferences among key attributes of long-acting PrEP. We conducted a discrete choice experiment (DCE) with 608 sexually active, PrEP-naïve male and female youth aged 18-24 residing in two townships in Cape Town, South Africa, recruited through population-based sampling of residential plots. With an experimental design, participants chose between two hypothetical products in 9 choice questions, each composed of 5 attributes. Attributes included: form (injection, implant); dosing frequency (2, 6 or 12 months); where to obtain (clinic, pharmacy, community distribution, mobile clinic); pain involved with injection or insertion; and delivery location (arm, buttock, thigh). We used a random-parameters logit model to estimate preference weights and relative importance scores (based on coefficient value ranges) for each attribute. Median age was 21 (IQR: 19-22) and, by design, 66% were female. Nearly all were heterosexual (99%), had tested for HIV (95%) and were HIV negative (91% of testers). Duration between doses was the most important attribute (relative importance [RI]=2.0, 95% CI: 1.7, 2.3), with strong preference for less frequent dosing. The second most important attribute was product form (RI=0.8, 95% CI: 0.6, 0.9): injection was favored over implant. Participants also preferred delivery of a dose to the arm and disliked the thigh, and preferred mild over moderate pain. They disliked pharmacy distribution and exhibited no preference among other distribution locations. Youth indicated strong preferences for longer duration products and injections. Nonetheless, each attribute influenced preferences, offering nuanced insight into trade-offs that inform long-acting PrEP development.
