The core objectives of this study were to document the process by which a community-based organization replicated and adapted an experimentally developed intervention to its own use and to explore the effectiveness of that HIV prevention program for male prostitutes and other patrons in New York City "hustler" bars. The intervention model employed was based on previous research with gay men (Kelly, St. Lawrence, Diaz, et al., 1991; Kelly, St. Lawrence, Stevenson, et al., 1992) and inspired by diffusion of innovation theory (Rogers, 1995). The effects of the current intervention were assessed on a sample of 1,741 male prostitutes and bar patrons. Analyses indicated significant reductions in paid, unprotected sexual intercourse and oral sex following the intervention. Analyses further indicated that the data were partially consistent with the program's model, which specified that norms were the putative mediator of behavior change in the intervention. Also, the intervention's effects varied by bar and by participants' race/ethnicity. Data support the utility of the intervention model for an urban sample of men at high risk for HIV infection. The importance of exploring the mechanisms that underlie the intervention is discussed.