Filling Gaps in the PrEP Cascade: User-centered Counseling About HIV Pre-exposure Prophylaxis and Linkage to Care for Gay, Bisexual and Other Men Who Have Sex with Men

Our research questions are: (1) Is a training program for PrEP acceptable for healthcare providers, measured by perceived advantage of applying the recommendations learned in the program, compared to current practice? (2) Is a training program for PrEP feasible for healthcare providers, measured by intention to continue applying the recommendations after the study conclusion? And (3) What is the anticipated effectiveness of a training program at increasing the proportion of PrEP-eligible GBM and TGNC who start PrEP within one month of a counseling session with a healthcare provider, compared to passive referral to existing resources?

In this open-label cluster randomized pilot trial, the investigators will randomize consenting providers to undergo an online self-paced training on brief user-centered PrEP counseling vs only receiving a list of available resources. The goal is to enroll a range of different healthcare provider types working in both urban and suburban settings, and in different work settings. Recruitment of user-participants will be facilitated by participating providers, but done by a research coordinator.

Randomization and inclusion of provider-participants is expected to start in October 2024 and will continue through February 2025. Data collection (questionnaires and interviews) will continue beyond that period for provider-participants, as well as for user-participants whose enrolment occurs during the specified recruitment dates.

The qualitative component will explore implementation considerations from the perspective of provider-participants. The investigators will do focus groups (of approximately one hour, variable size) and offer the possibility of doing individuals interviews (of approximately 30-45 minutes) if they cannot attend the focus groups.

In the analysis stage, counts and proportions will be used to describe categorical and dichotomous variables. Mean and standard deviations or median and interquartile range will be used to describe continuous normally and non-normally distributed variables respectively. For the anticipated effectiveness outcome, the intra-cluster correlation coefficient (ICC) among patient-participants will be calculated. For that purpose, our analysis will be a generalized linear mixed model (GLMM) with logit link to estimate the effect of using the training program on the probability of user-participants starting (or re-starting) PrEP, while accounting for clustering of user-participants within the respective providers.