In 1983, the Washington Division of Juvenile Rehabilitation (DJR) conducted a study of the juvenile offender population, concluding that 67 percent of the population were chemically dependent and another 20 percent were serious substance abusers. As a result, the DJR developed a wide range of services for this client population; each offender was assessed using the Client Substance Index (CSI), and the CSI played a major role in DJR's program development. The CSI was also modified to provide more complete, clinical information to rehabilitation staff. However, several years later, because of concerns that the CSI did not fully explore the complexity of adolescent substance abuse, the DJR offered an offender population of 100 youths to test the PEI. All the subjects had either an extensive criminal history or a serious, violent committing offense. Local substance abuse assessment specialists, using the 22 PEI subscales as guidance, interviewed each client, gathering data on the nature and impact of substance use or abuse in the client's life. Then PEI was administered to each youth. The results indicate that the PEI is a useful tool for substance abuse assessment among an offender population. The PEI scores were significantly correlated to rating of independent clinical specialists and the magnitude of the correlations was greater than alternative measures. The PEI subscales were all validated and the high-risk subscales were strongly correlated with the general PEI chemical dependency score. 4 tables, 1 attachment.