Predictors of condom use among repeat HIV-negative testing men who have sex with men

Background: Consistent condom use for anal sex among men who have sex with men (MSM) is critical for meeting CDC's objective of reducing new infections by 50%. This study examined HIV-related risk behaviors and explored predictors of condom use in a sample of males participating in the evaluation of an intervention to reduce HIV infection risk among high-risk, HIV-negative repeat testers in San Diego. Methods: Baseline data for a sub-sample of 85 White (69.4%) and Hispanic males reporting anal sex with male partners in the previous 12 months were analyzed using the Behavioral Ecological Model (BEM; Hovell et al). The BEM hypothesizes that condom use is predicted by multiple behavioral influences in the social and physical environments such as modeling, reinforcement, punishment, and cues to enact behaviors; individuals exposed to higher densities of behavioral influences supportive of condom use would have higher rates of condom use. A sequential multiple regression of four blocks (demographics, sexual practices, condom negotiation, and peer influence) was used to test the significance of 8 independent predictors of condom use. Results: The final regression model accounted for 62.9% of the explained variance in 3-month condom use for MSM anal intercourse, F (8,76) = 18.80, p < .001. The variables that contributed significantly were: the number of times had anal sex (p<.001), condom negotiation (p<.01) and peer influence for condom use (p<.01). Conclusions: Results provide support for the BEM and evidence for extending prevention activities beyond education to establishing skills and engineering peer support for condom use.