Socio-behavioral correlates of pre-exposure prophylaxis use and correct adherence in men who have sex with men in West Africa.

BACKGROUND: Multiple barriers compromise pre-exposure prophylaxis (PrEP) engagement (i.e., use and adherence) in men who have sex with men (MSM). In low/middle-income countries, little is known about PrEP engagement in this population. In West Africa, the CohMSM-PrEP study was one of the rare interventions providing PrEP to MSM. We estimated PrEP use and correct adherence rates in CohMSM-PrEP, together with associated factors over time.  METHODS: CohMSM-PrEP recruited MSM in four community-based clinics in Mali, Côte d'Ivoire, Burkina Faso, and Togo. Quarterly follow-up included collecting socio-behavioral data, and providing a comprehensive HIV prevention package, PrEP (daily or event-driven), and peer educator (PE)-led counselling. Using repeated measures, multivariate generalized estimating equations models were used to identify factors associated with self-reported i) PrEP use and ii) correct PrEP adherence during participants' most recent anal intercourse (defined as four pills/week for daily users and 2 + 1 + 1 for event-driven users). RESULTS: Five hundred twenty participants were included with a median follow-up time of 12 months (IQR 6-21). Of the 2839 intercourses declared over the follow-up period, PrEP use was self-reported for 1996 (70%), and correct PrEP adherence for 1461 (73%) of the latter. PrEP use was higher in participants who also attended participating clinics outside of scheduled visits (adjusted odds ratio (aOR) [95% Confidence Interval, CI], p-value; 1.32[1.01-1.71], 0.040), and in those who practiced condomless anal sex (1.86[1.54-2.24], < 0.001). Correct adherence was higher in those who often contacted PE outside of scheduled visits (2.16[1.01-4.64], 0.047) and in participants who adopted receptive/versatile sexual positions with stable partners (1.36[1.03-1.81], 0.030). Instead, after an interaction effect between financial situation and regimen was tested, it was lower in event-driven users with a difficult/very difficult financial situation (comfortable/just making ends meet & daily, 4.19[2.56-6.86], < 0.001; difficult/very difficult & daily, 6.47[4.05-10.30], < 0.001; comfortable/just making ends meet & event-driven, 1.63[1.22-2.17], 0.001), and in participants who felt alone (0.76[0.58-0.99], 0.042). CONCLUSIONS: Community-based clinic attendance and PE contact outside of scheduled visits were both associated with higher PrEP engagement, but some socially and economically marginalized participants struggled with adherence. As scale-up continues in West Africa, we recommend implementing community-based interventions and providing extra support for vulnerable users to ensure adequate PrEP engagement.