This study examined the prediction of outcome in a sample of defendants hospitalized for treatment of incompetency. Defendants' demographic characteristics and scores on 18 scales of the Computer-Assisted Determination of Competency to Proceed instrument (CADCOMP) were used to predict competency restoration and length of stay (LOS). During the period of study, almost 90 percent of the defendants were restored to competency after a mean stay of over 280 days. Demographic characteristics were unrelated to outcome. Several CADCOMP scales, including two scales measuring psycholegal ability and one measuring psychopathology, were correlated with both outcome criteria. Discriminant analysis using the CADCOMP scales accurately classified 76.7 percent of the defendants into short and long stay groups. Although promising, the findings are nevertheless consistent with prior research in suggesting that examiners should exercise caution in providing feedback to the courts concerning competency restoration and the period of time needed for treatment.