Detailed self-report data on drug use, crime, and social functioning were obtained from assessment interviews. More than 90 percent of the addicts had at least one conviction for property or drug offenses. The relationship between crime rates and other variables such as gender, age, age at first conviction, and employment status were examined using Poisson regression techniques. Results showed that addict self-reported convictions were moderately correlated with police-recorded convictions. Men were more likely to begin heroin use at or after their first criminal conviction, while women were more likely to begin offending after they used heroin. The rate at which addicts accrued convictions for property offenses declined as they grew older, and the earlier an addict's first exposure to the criminal justice system the higher the rate at which he or she accrued convictions. Overall, young men became involved in heroin use about the same time they first came to the attention of the criminal justice system. Most had little education and poor social integration, as reflected in unstable employment and living situations. In contrast, the involvement of female addicts in crime appeared to follow established heroin addiction. For some women, prostitution provided a means of maintaining an expensive habit without recourse to other forms of property crime. For other women, heroin addiction led to involvement in acquisitive crime, sometimes after a prolonged period. Findings lend support to the complex relationship between drug use, crime, and social dysfunction. 18 references and 6 tables