Abstract
Men who have sex with men (MSM) with history of childhood sexual abuse (CSA) are at high risk for HIV acquisition. One reason is posttraumatic responses compromise ability to accurately appraise risk for danger/vulnerability. Health behavior change models and related interventions assume risk perception can be changed in an enduring manner. Given paucity of studies examining how risk perception changes or sustains over time post-intervention, this underlying assumption is not confirmed. Among this particularly high-risk group who struggle with perceiving risk due to trauma-related cognitions, it may be accuracy of risk perception is fluid. The study primarily aimed to examine accuracy of HIV risk perception over time post-HIV prevention behavioral intervention. Leveraging data from a larger RCT, N = 190 MSM in Boston, MA and Miami, FL USA completed a psychosocial baseline assessment, an intervention aimed to increase awareness of personal HIV risk level, then four follow-up assessments three months apart for a year. Linear mixed effect models were used to examine the degree to which accuracy of HIV risk perception (vs. traditional construct of risk perception with no information about accuracy) predicts sex risk behavior over time delineated by between-person (trait level) and within-person (state level) effects. Majority (92%) of participants fluctuated in HIV risk accuracy over time post-intervention. Within-person risk accuracy (one's accuracy at any given timepoint) predicted sex risk behavior (condomless sex not protected by adherent PrEP) over time, but not between-person (one's average of accuracy). Findings have implications for intervention and counseling related to specific HIV prevention strategies.