Matthew F. Shaw

Shaw builds upon a study tracking 73 Not Guilty by Reason of Insanity (NGRI) acquittees over an eighteen year period. He integrates retrospective chart reviews with arrest and hospitalization data and assesses strengths and community variables along with individual factors. Stepwise logistic regression models form the basis for predictions. The acquittees resembled prior samples in terms of pre-trial characteristics and demonstrated clinical gains during hospitalization. Nonetheless, one third of the sample was re-arrested and fifteen percent were re-hospitalized after release. Clinical, community, and system factors predicted recidivism. The models developed were correct in roughly 85% of the cases. The study concludes with policy implications and directions for future research.