Prediction of violence on a psychiatric intensive care unit

The aim of the present study was to identify patients who were at particula r risk of violent behaviour, over a period of one month in a psychiatric hi gh dependency unit, and to examine whether particular patients could be ide ntified at admission who would subsequently be violent. Medical records, and other documents for 52 inpatients were included. Demog raphic, historical and clinical data available at admission were noted. All violent incidents were recorded and classified. Seventeen of 52 patients c ommitted physical violence, usually to the staff. Most incidents were carri ed out by a small number of individuals. Identifying the violent, using var iables available at admission, was not possible. The ward admitted a group with a subsequent high rate of violence. Predicti on within this highrisk group was not possible. Measures aimed at preventin g violence on a local secure ward should apply to all patients and should n ot he targeted just at those that would appear to be the most likely to be violent.