Possible predictors of outcome for conditionally discharged patients--a preliminary study.

Conditional discharge for restricted hospital order patients is by and large a successful process. The present study aimed to identify variables among a cohort of conditionally discharged patients in the West Midlands that would predict whether an individual was more likely to be readmitted to hospital, involved in a serious incident, to be recalled to hospital or given an absolute discharge. A retrospective case note analysis was undertaken. Logistic regression analysis was used to identify variables that could predict outcome. Patients were six times more likely to be readmitted to hospital if they misused drugs and nine times more likely if they self-harmed. They were also six and four times more likely to be involved in a serious incident if they misused alcohol and drugs respectively. Patients were five times more likely to be recalled to hospital if they did not have close social support and were four and a half times more likely to get an absolute discharge if they lived in supported accommodation. Conditional discharge is an effective model of community care for restricted hospital order patients. Addressing the problems caused by drug and alcohol misuse, self-harming behaviour, recognising the importance of a close social support network and developing more appropriate housing for these individuals will help make the process more successful.