This study examined demographic, clinical, and criminal histories as they related to whether 244 youth (ages 12-18) referred for mental health services at a juvenile corrections facility were repeat or first-time offenders. The population of 244 patients included 144 first-time offenders and 100 repeat offenders. Results of the study found that the referred youth in the sample differed significantly from youth in the overall population at the juvenile hall in almost all of the categories measured. The sample included fewer first-time offenders and Hispanics, it included more females, and included more youth with a history of serious violent offenses. Findings also indicated that the arrest patterns for youth referred for mental health services were more likely to be affected by the juvenile justice history of the youth than by clinical factors, including drug use and mental health service use. The number of years of delinquency appeared to be the strongest predictor of repeat offending within the group of referred youth. Lastly, the study found violent offenders were half as likely to be repeat offenders as non-violent offenders. (Contains 13 references.) (CR)