The primary goal of the proposed study is to assess the impact of housing type, drug use, social isolation, PTSD, dissociation, depression, manic episodes and HIV serostatus on the risk behavior and victimization of poor and marginally housed women. In addition to the 105 HIV-infected women recruited during the first study phase, we will recruit an additional 45 HIV-positive women and 150 HIV-negative women in the same manner. We will do this in order to identify whether victimization and risk behavior patterns differ between women who are at risk to become HIV-infected and those who are already infected. We will no longer collect data on homeless and marginally housed men. Guided by Chu's model of revictimization, we will test whether predictors of victimization differ between women who do and do not report a history of childhood sexual abuse; we will also test whether the impact PTSD and dissociation on victimization are modified by depression, manic episodes, or housing status. Guided by the Rhodes approach to HIV risk as a function of structural barriers, we will conduct qualitative interviews regarding SRO hotels as potential high-risk environments, additional structural barriers regarding the context in which risk occurs, and intentional social isolation. This will be one of the first longitudinal studies of victimization and risk behavior to focus on indigent women and structural barriers. Understanding changes over time will facilitate the development of structural interventions and sustainable trauma service provision that will assist service providers in finding or creating safer housing for their clients, helping clients understand how their current situation still leaves them vulnerable, and offering alternate options