No abstract

Studies of violence prediction have typically involved long-range follow-up of institutionalized persons and have shown poor predictive accuracy. In the present study a sample of adult males at risk for violent behavior admitted as inpatients at a community mental health center was followed for six months after their release. Data collected at admission (which included demographics, family background, criminal justice and mental health systems contacts, past violent behavior, life stress, family, friendship, and work environment measures, psychiatric diagnosis, and alcohol and drug use) were used to predict subsequent violent arrests or readmissions using stepwise discriminant analysis. Results yielded 85% of the total sample correctly classified. Of those predicted to be nonviolent 94% were actually nonviolent and of those predicted to be violent 59% were actually violent. Seventy-six percent of the violent subjects were identified. Implications of the findings for clinical predictions of violence are discussed.

A statistical scale to assess the probability of violence in released mental patients was developed and cross-validated. A calibration sample of 251 adult males admitted as inpatients to a community mental health center and considered potentially violent was obtained. Data collected at admission (including demographics, family background, criminal justice and mental health system contacts, past violent behavior, and current situational measures) were used to predict subsequent violent arrests or readmissions during a 1-year follow-up period with multiple regression. A statistical equation of 5 variables was derived and applied to a cross-validation sample of 265 male patients. The correlation between the score and subsequent violence in this sample was .32. Classification using a a cutoff point obtained from the calibration sample yielded 75.8% correctly classified and a false positive rate of 47.6%.

This study examined predictors of violence in schizophrenic and nonschizophrenic mental health patients. Data were obtained for a sample of 304 adult male patients who were admitted for short‐term emergency treatment and consisted of mental hospital admission records, arrest records, and interview data on demographic, early history, and current situational variables. Subjects were classified as violent or non‐violent during a 1‐year follow‐up period based on arrest and readmission records. The sample was divided into schizophrenic and non‐schizophrenic groups, and a discriminant analysis was conducted for each group using intake data as independent variables to predict violent recidivism. The results indicated some patterns of different predictors for the two groups. Life events, substance abuse, and self‐reported arrests for violent crime were predictive of subsequent violence for non‐schizophrenic patients but not for schizophrenic patients. Selfreported incidents of violence other than arrests predicted subsequent violence for schizophrenic but not for non‐schizophrenic patients. Classification results for both groups were excellent, with 88.1% and 92.9% correct classification and 17.6% and 9.4% false positive rates for schizophrenic and non‐schizophrenic patients, respectively.