Does Stigma Toward Anal Sexuality Impede HIV Prevention Among Men Who Have Sex with Men in the United States? A Structural Equation Modeling Assessment

Abstract Introduction Men who have sex with men (MSM) are suboptimally engaged in efficacious HIV interventions, due in part to stigma. Aim We sought to validate the Anal Health Stigma Model, developed based on theory and prior qualitative data, by testing the magnitude of associations between measures of anal sex stigma and engagement in HIV prevention practices, while adjusting for covariates. Methods We conducted a cross-sectional online survey of 1,263 cisgender MSM living in the United States and analyzed data with structural equation modeling. We tested a direct path from Anal Sex Stigma to Engagement in HIV Prevention alongside 2 indirect paths, 1 through Anal Sex Concerns and another through Comfort Discussing Anal Sexuality with Health Workers. The model adjusted for Social Support, Everyday Discrimination, and Sociodemographics. Main Outcome Measure Engagement in HIV Prevention comprised an ad hoc measure of (i) lifetime exposure to a behavioral intervention, (ii) current adherence to biomedical intervention, and (iii) consistent use of a prevention strategy during recent penile-anal intercourse. Results In the final model, anal sex stigma was associated with less engagement (β = −0.22, P < .001), mediated by participants' comfort talking about anal sex practices with health workers (β = −0.52; β = 0.44; both P < .001), adjusting for covariates (R2 = 67%; χ2/df = 2.98, root mean square error of approximation = 0.040, comparative fit index = 0.99 and Tucker-Lewis index = 0.99). Sex-related concerns partially mediated the association between stigma and comfort (β = 0.55; β = 0.14, both P < .001). Modification indices also supported total effects of social support on increased comfort discussing anal sex (β = 0.35, P < .001) and, to a lesser degree, on decreased sex-related concerns (β = −0.10; P < .001). Clinical Implications Higher stigma toward anal sexuality is associated with less engagement in HIV prevention, largely due to discomfort discussing anal sex practices with health workers. Strength & Limitations Adjustment for mediation in a cross-sectional design cannot establish temporal causality. Self-report is vulnerable to social desirability and recall bias. Online samples may not represent cisgender MSM in general. However, findings place HIV- and health-related behaviors within a social and relational context and may suggest points for intervention in health-care settings. Conclusion Providers' willingness to engage in discussion about anal sexuality, for example, by responding to questions related to sexual well-being, may function as social support and thereby bolster comfort and improve engagement in HIV prevention.