Despite the fact that drug-using women have been the group of women most at risk for infection with the human immunodeficiency virus (HIV). Little is known about these women, particularly those who inject drugs and/or are the sexual partners of injection drug user (IDUs). Even less is known about female adolescents and young women in their early 20's who inject drugs and/or are the sexual partners of IDUs. These younger women are likely to be developmentally very different from their older female peers/counterparts. The proposed research is designed to characterize HIV risk and investigate the emotional psychological, familial, and social-environmental contexts in which HIV risk and protective behaviors occur among high risk young women, ages 15 to 23 years. Specific female populations of interest include out-of-treatment IDUs and injecting and noninjecting sexual partners of IDUs. Multimethod research will be conducted in two study phases. During Phase 1, preliminary qualitative and ethnographic research will be conducted to adapt our existing survey instruments and develop new scales which are appropriate for use with the target population; develop methods for conducting the proposed social mapping and social network analysis; develop methods for locating snowball sample referrals in a geographic area as large and diverse as Los Angeles County. During Phase 2, epidemiologic and ethnographic research will be conducted to characterize gender-specific differences with respect to the emotional, psychological, familial, interpersonal, and social-environmental contexts of risk and protective behaviors. A cohort of a minimum of 400 subjects will be recruited into Phase 2 of the study. The research will be conducted with youth ages 15 to 23 years who are either out-of-treatment IDUs, sexual partners of IDUs (who may or may not inject drugs), and their peers. Subjects will be identified and recruited using a combination of sampling techniques including representative and systematic probability sampling and snowball sampling. We will first recruit 200 subjects using representative sampling techniques. Then, a minimum of 200-275 subjects will be recruited using snowball sampling techniques. The proposed research will be conducted with both males and females in an effort to characterize gender-specific risk and protective behaviors and the contexts in which these behaviors occur. The sample will also be ethnically diverse in an effort to investigate culturally influenced contexts for risk and protective behaviors. This information will in turn be disseminated widely to both the research and service communities. Technology transfer products will also be developed and disseminated in an effort to encourage and support on-going research with this population.