The aim of this research was to test the utility of the Addiction Severity Index (ASI) in a rural community substance abuse outpatient treatment center and to predict future alcohol and drug problems based on ASI information. Substance abuse counselors used the ASI to assess the problems of 89 adult clients at intake, and research staff assessed drug use outcomes 3 months later. There were significant improvements between intake and follow-up on the scores on the alcohol, drug, legal, and family sections of the ASI. Scores on the alcohol abuse, drug abuse, and medical sections of the ASI predicted some of the variance (16%) in drug and alcohol use outcomes, but psychiatric, medical, legal, and employment problems were not significant predictors. There was some difficulty in obtaining adequate reliability in counselors' severity ratings for alcohol problems. The relationship of counselors' severity ratings to the more objective composite scores varied substantially by subsection of the ASI. The improvements in alcohol and drug use indices appeared to be more related to subjective appraisals of problems related to use rather than due to changes in drug-using behavior. Implications of these findings and recommendations for further research using the ASI are discussed.