The juveniles, ages 14-18, were recruited for this study between July 2000 and December 2001 from admissions to five adolescent outpatient substance abuse treatment facilities in the county. Data were obtained from 178 youths (150 males and 28 females). As part of the treatment intake process, participants were administered the Global Appraisal of Individual Needs (GAIN). This is a clinical/research instrument that measures eight domains, including background and treatment arrangements, substance abuse, psychological functioning, illegal activity, and family and peer relationships. Three time periods are assessed for these domains: lifetime, past year, and past 90 days. Supplemental questions focused on the ages at which respondents initiated the use of various substances and committed various offenses. Respondents were asked if they had ever engaged in each of 15 types of illegal activity. Respondents were classified according to the most serious level of offense they had ever committed prior to entering treatment. The dependent variable was criminal severity, representing the most serious level of illegal activity the youth had ever committed prior to entering treatment. Severity levels were major, minor, and minimal. The findings indicated that increased severity of crime was related to being a male, using drugs other than alcohol and marijuana, bullying and being physically cruel to people, higher levels of deviant behavior among peers, school problems, and having sex without some protection. Youths who had committed major offenses tended to have patterns of offending that were more precocious, varied, and frequent than those youths who had committed only minor offenses. The major offenders began their offending career slightly before age 11, over a year before the minor offenders. Unlike the findings of previous studies that involved adolescents drawn from other populations, aggressive behavior and lifetime psychological problems were not significantly associated with crime severity in the current study. These findings suggest that drug treatment programs for adolescents can expect to encounter widespread criminal involvement among participants; however, there will be considerable variation in the severity of the offending. Adolescents who have committed major offenses are likely to require interventions that are qualitatively different from interventions needed by other drug-using youths. 2 tables, 3 notes, and 68 references