People with severe mental illness (SMI) have been found to be at extremely elevated risk for human immunodeficiency virus (HIV) -- 4 percent to 23 percent seroprevalence -- in the U.S. To address the special needs of this population, several HIV prevention interventions have been tested and found to be efficacious for the SMI. Despite a well-documented AIDS crisis in Brazil and the Brazilian government?s commitment to HIV prevention for the whole society, Brazilian SMI are not regularly offered HIV prevention interventions. Two small studies in Brazilian psychiatric settings have documented sexual risk-taking among SMI patients. Tailoring efficacious prevention interventions to the SMI of Brazil is likely to facilitate integration into the mental health care system of HIV prevention programming. The goal of this study is to adapt, refine, and pilot a provider-delivered HIV prevention intervention for men and women with SMI in Brazil in order to: (1) impart skills to mental health care provider about (MHCPs) in Brazil to deliver an HIV prevention intervention for their SMI clients; and (2) help these clients reduce their HIV risk behaviors. This study will be undertaken collaboratively by a US-Brazil team of experienced investigators. The study will be administered by ABIA, a community-based organization dedicated to ethnographically-based HIV prevention in Brazil, and carried out at the Instituto de Psiquiatria da Universidade Federal do Rio de Janeiro (UFRJ/IPUB), an academic psychiatric institute with direct-care responsibilities for a large number of psychiatric outpatients. We propose to conduct formative studies with Brazilian MHCPs and psychiatric patients at the UFRJ/IPUB, to consist of ethnographic observations/assessment and focus groups with key informants (MHCPs and male and female psychiatric outpatients), which will be used to tailor extant efficacious interventions to the needs of the SMI population in Brazil. The refined intervention will be pilot-tested with men and women with SMI using process measures and psychiatric assessments following every intervention session, as well as risk behavior assessment at baseline and after 3 months. An attention control intervention will be similarly adapted, implemented with a group of outpatients, and assessed using the same measures. Results will be disseminated in 2-day conference in Brazil to MHCPs, researchers, and mental health policy makers (i.e., local opinion leaders). This project will lay the groundwork for designing a clinical trial of the adapted intervention in Brazil.