Our proposed study is in response to Program Announcement #95-057 HIV Risk Behavior, Determinants and Consequences from NIDA Division Of Epidemiology and Prevention Research. Trauma histories among chemically dependent populations are just beginning to be systematically studied in an attempt to understand the interrelationships among the effects of early trauma, drug use, and HIV infection. The prevalence of sexual abuse within drug using populations has already been well documented. There is evidence which suggests a link between a sexual abuse history and participation in subsequent HIV risk behaviors. However, it remains unclear as to why some drug users with sexual abuse histories go on to engage in HIV risk behaviors and others do not. In order to understand a relationship between sexual abuse and HIV risk behaviors, we will explore hypothesized psychosocial pathway variables, which may be activated as a result of sexual abuse. We hypothesize that the severity of sexual abuse will be predictive of participation in HIV risk behaviors through a number of possible psychosocial pathway variables. By examining the severity of sexual abuse, as well as the psychosocial processes that may contribute to risk taking, it may be possible to discriminate groups with varying levels of risk based on indices of severity of sexual abuse. This, in turn, allows us to examine plausible causal pathways to HIV risk behaviors for the development of HIV prevention strategies. The currently proposed research will use two existing cohorts Non-injecting Heroin Users (NIHUs) and younger/ newer injectors (CIDUS II). By combining current and past trauma and psychosocial functioning data from the proposed cross sectional research with longitudinally collected behavioral data from an existing cohort study, the investigators will be able to efficiently and economically (1) examine the relationship between sexual abuse severity and HIV risk; through the use of a nested case control design; (2) empirically test the role in the development of HIV risk; behaviors of three hypothesized psychosocial pathway variables (i.e., trauma symptoms, depressive symptomatology, and dissociative behaviors) which may result in cases of severe sexual abuse; and (3) determine the stability of the relationship between the severity of sexual abuse and HIV risk behaviors over time. The cohort subjects who will participate in our proposed research are at very high risk for HIV and other viral infections if they become IDUs or if they have sex with IDUs, or if they practice risky injection. Thus, if sexual abuse is an important risk factor among women and/or men affecting these behaviors, then understanding the processes that may be activated by certain characteristics of sexual abuse may be critical to reducing HIV risk; behaviors and thereby minimizing the spread of HIV infection.