Violence after discharge from forensic units in the safe pilot study: a prospective study with matched pair design.

Objective: This paper reports on a prospective naturalistic study of violent recidivism after discharge from forensic mental health. Main aims were to find predictors of violence and to test the feasibility of a matched pair design for this purpose.

Methods: Patients from the Safe pilot project (n=18) and a group of controls (n=18) were matched on 10 variables, such as diagnosis, seriousness of violence, setting after discharge, and risk management plans.