Young men who have sex with men (YMSM) have unique needs for targeted and tailored HIV prevention messages. HIV infection rates among MSM remain high, particularly among men under the age of 30 and for Black YMSM, whose HIV infection rates are over four times those of White YMSM. The use of drugs and alcohol has been linked to increased sexual risk for HIV among MSM yet no intervention addressing substance use and sexual risk within this population has been rigorously evaluated. Philadelphia Health Management Corporation (PHMC) proposes to develop, implement, and evaluate a community-level intervention for Black and White substance-using MSM between the ages of 15 and 29. The intervention, implemented with the collaboration of a community service organization, the Mazzoni Center, will be designed to change both social norms and risk behavior using persuasive media communication and interpersonal networking as primary intervention strategies. The proposed project aims to: 1) examine patterns and contexts of episodic substance use among YMSM; 2)compare patterns of substance use for Black and White YMSM, and differences by age, sexual identity and socio-economic status; 3) clarify the association between patterns and contexts of substance use and sexual risk behaviors; 4) evaluate the efficacy of a community-level intervention for substance-using YMSM to decrease sexual risk for HIV, through strengthening social norms, increasing perceived risk, and enhancing self-efficacy and positive attitudes to HIV prevention; 5) measure the differential impact of intervention messages on sexual risk behavior for subgroups of YMSM; 6) determine if a reduction in substance use leads to a decrease in sexual risk behaviors. During the formative research phase, qualitative data will be obtained through focus groups and in-depth interviews to refine the conceptual model and the media materials. The intervention evaluation will collect quantitative data from YMSM in both Philadelphia PA and the comparison community of Baltimore MD, using geocoding and census tract data to control for demographic and socio-economic differences. A Solomon four-group design will be used to collect data from panel and cross-sectional samples in intervention and comparison communities. Four annual waves of interviews (baseline, 12, 24, and 36 months) will enable us to examine exposure to the intervention, changes in norms and behavior, and long-term intervention effects. A total of 650 men will be interviewed in Philadelphia and 400 in Baltimore. The data will be analyzed using multivariate techniques, including generalized linear model (GLM) repeated measures analysis and cumulative logit models. The study results will provide needed information on substance use and sexual risk of Black and White YMSM and have direct implications for implementing effective interventions to this high-risk population.