Abstract
According to the U.S Supreme Court decision in Jackson v. Indiana, mental health professionals must determine if a defendant has “substantial” probability of regaining competency through treatment in the “foreseeable” future. Previous research suggests that examiners are relatively poor at predicting which defendants will regain competency given the low base rate of defendants unable to be restored to competency. In addition, few studies have used multivariate statistics to examine the inter-correlations among predictor variables and restorability predictions. The purpose of this study was to utilize multivariate statistics to determine whether a set of predictor variables (demographic, criminal, and psychiatric variables) could accurately predict a defendant's classification as restorable or not restorable. The analyses suggest that none of the three sets of predictor variables—demographic, criminal, and psychiatric—taken alone significantly outperforms the other. However, psychiatric variables performed slightly better than demographic and criminal variables in correctly classifying unrestorable defendants. Ultimately, an equation with two variables—previous criminal history and a current violent charge—was found to be statistically the most optimal combination of predictors accounting for the most variance. International Journal of Forensic Mental Health
Get full access to this article
View all access options for this article.