Prediction of Sex and the Potential Use of On-Demand PrEP Among Young Men Who Have Sex With Men in the United States

BACKGROUND: On-demand dosing of preexposure prophylaxis (PrEP) requires accurate prediction of sex; however, prediction abilities among young men who have sex with men (YMSM) have not been characterized. SETTING: A nationally recruited prospective cohort of YMSM ages 16-24 years. METHODS: We followed 120 YMSM for 8 weeks using digital daily surveys (DDSs) to measure engagement in and prediction of anal sex over 24 hours, along with condom use and other encounter-level circumstances. Our main outcome, an "unpredicted spontaneous encounter," was defined as an anal sex encounter that occurred without sufficient prior knowledge to (hypothetically) enable protective on-demand PrEP use according to dosing guidelines. We operationalized this outcome as an anal sex encounter for which a participant indicated: (1) on the prior day's DDS that there was a low likelihood of sex occurring in the subsequent 24 hours (unpredicted) and (2) on the current day's DDS that he knew ≤2 hours in advance that the encounter would occur (spontaneous). RESULTS: Approximately one-third of all anal sex encounters during the study period were unpredicted and spontaneous and would not have been protected (hypothetically) by on-demand dosing. More than two-thirds of participants experienced such an encounter and almost three-quarters of all acts were condomless. CONCLUSIONS: On-demand PrEP to prevent HIV acquisition may be challenging for many YMSM. Clinical and public health approaches that account for patients' predictive abilities alongside their dosing preferences may help to optimize selection of and adherence to PrEP dosing strategies.