This study seeks to characterize the effects of undertreatment of pain in HIV positive substance abusers in terms of aberrant drug-taking behaviors and examine the correlates of this behavior. This will include contrasting the frequencies of these behaviors with those of cancer pain patients who have a lower baseline rate of substance abuse and who are either adequately or inadequately managed for their pain. Previous research has described the problem of undertreatment of pain in AIDS patients. Among the risk factors for undertreatment is a history of intraveneuous drug use as an HIV-transmission factor. Findings indicate that substance abusers with pain were being differentially undermedicated despite the fact that there are no differences in the characteristics of their pain reports and problems. The undermedication of active substance abusers is often driven by fears of contributing to addiction and diversion and fears of """"""""re-addiction"""""""" of those in recovery. However, an equally important contribution to addiction and aberrant drug-related behaviors may stem from undertreatment of pain; this is an understudied problem, as clinical experience suggests that untreated pain can lead to illicit drug use, diversion of prescribed medications and therefore a significant contribution to addiction. Research has found that the use of illicit drugs for symptom management was not uncommon in women with AIDS, many of whom had a history of substance abuse. Cancer pain is also undertreated in the United States. The lower rate of pre-existing substance use in this population and the greater likelihood that such patients receive adequate pain medication renders cancer patients an excellent comparison group for exploring the range and nature of aberrant behavior set in motion by untreated pain. The drug taking behavior of cancer pain patients may also change due to untreated pain, though it is likely these behaviors are less markedly aberrant. In this study, the investigators will work closely with experts in the field of substance abuse and cancer and AIDS pain management to devise a structured interview in phase one and then use this interview to assess the problem. After obtaining this information the consulting group will help to develop a coding system so that results of the interview can be correlated with measures of pain, other physical symptom distress, psychological distress and social desirability and addiction status. The information obtained from this investigation will not only describe the frequency and extent of this problem and possibly be influential in policy making and improvement of pain management for patients with cancer and AIDS, but will also be used to help in our efforts to develop instrumentation for the study of aberrant drug-taking in pain patients.