Unsupervised Machine Learning to Detect and Characterize Barriers to Pre-exposure Prophylaxis Therapy: Multiplatform Social Media Study.

Both objective and subjective reasons were identified as barriers reported by social media users when initiating, accessing, and adhering to PrEP. Though ample evidence supports PrEP as an effective HIV prevention strategy, user-generated posts nevertheless provide insights into what barriers are preventing people from broader adoption of PrEP, including topics that are specific to 2 different groups of sexual minority groups and racial and ethnic minority populations. Results have the potential to inform future health promotion and regulatory science approaches that can reach these HIV and AIDS communities that may benefit from PrEP.