The distinctive contribution of this study lies in its modeling of drug and crime involvement variables as predictors of three leading health problems: chronic disease and illness, HIV/AIDS and sexually transmitted diseases (STD's), and mental illness. Data for this study were obtained from a mid-1990's research project on the barriers to drug treatment for cocaine-dependent women in the Miami metropolitan area. Although the overall goal of the study was to obtain information on obstacles to receiving drug treatment, data were also collected on numerous physical and mental health problems that ultimately warranted special attention. The sample consisted of 851 women between the ages of 18 and 34, who were interviewed in Miami between 1994 and 1996. All study respondents were involved in crime and were cocaine dependent. Approximately half were recruited on the streets of Miami by experienced field workers. The other 427 respondents were recruited at residential drug treatment programs or in outpatient programs associated with the Miami Drug Court. Data were collected by using a face-to-face interview format. The bivariate models indicated that the lengths of criminal and drug careers had a small but statistically significant effect on the odds of all three health outcomes. In the multivariate models, introduction of control variables rendered the effects of criminal involvement and drug use insignificant for the most part. Respondents with health insurance and those who had in the past been to a hospital for a drug overdose were substantially and significantly more likely to report a mental illness, HIV/AIDS and STD's, or chronic disease and illness. Since 54.2 percent of the respondents did not have health insurance and had not experienced a drug overdose, the true rates of health problems were probably considerably higher than those reported in this study. These findings point to the need for additional research into the causes of health problems among female drug offenders. 4 tables and 43 references