DESCRIPTION (provided by applicant): This application describes a five-year research program that develops and tests a theoretical model of socio-cultural, psychosocial, and contextual influences on unprotected sex in three groups of Latino men who have sex with men (MSM) in the U.S.--Dominicans, Colombians, and Brazilians. These groups represent understudied and growing immigrant populations at high risk for HIV/AIDS. This application involves an innovative approach with a theoretical model that includes variables at two levels. The person level includes between-subjects variables that describe socio-cultural (national origin, acculturation, subscription to beliefs held by many Latinos; contact with gay social groups, and experiences of discrimination) and psychosocial (depression, self efficacy for safe sex, coping, and enjoyment of unprotected sex) characteristics of the person. The situation level includes within subjects variables that describe the circumstances associated with each sexual encounter, such as the relationship with the partner, sero-concordance with the partner, alcohol and drug use, setting, indicated desire for safe sex, and subjective sexual arousal in that encounter. Hypotheses concern not only the contribution of each level to the prediction of unprotected sex, but also moderating effects between levels. By focusing on situational characteristics of sexual encounters, as well as on individual characteristics, this study will provide a greater understanding of factors influencing unprotected sex among Latino MSM, which will aid in efforts to prevent the spread of HIV. There are four phases of this study. The aims of Phase 1 are to elucidate the context of sexual encounters of Dominican, Colombian, and Brazilian MSM in New York City; to identify aspects of the context of social and sexual encounters that lead to various forms of unprotected sex; and, to describe similar and different patterns of beliefs, norms, and behaviors in the three groups. This qualitative phase will involve ethnographic observation, a minimum of 15 key informant interviews, six focus groups, and 30 in-depth interviews. The aims of Phase II are to refine our theoretical model of sexual risk as informed by the qualitative findings from Phase I; and to develop and pilot test a quantitative survey instrument based on the theoretical model. Three focus groups will provide feedback on the survey instrument, which will be revised and then administered via Audio-CASI to a sample of 120 Latino MSM, evenly divided among Dominicans, Colombians, and Brazilians. The aim of Phase III is to further refine the theoretical model of sexual risk and the quantitative survey based on the qualitative and quantitative findings from Phases I and II; to administer the revised survey instrument; to examine similarities and differences in the beliefs, experiences; and behaviors of the three national groups; and to test hypotheses based on the theoretical model of the contextual influences on sexual risk. We will administer the quantitative survey to a sample of 480 Latino MSM, 160 from each of the four groups, using A-CASI methodology. The aim of phase IV is to integrate findings from the multiple sources of the qualitative and quantitative components of the study. We will examine areas of agreement and disagreement across methods. The triangulation of findings will enable us to interpret results and further refine the theoretical model of contextual factors of sexual risk among these three groups of Latino MSM.