Validation of a Violence‐Prediction Tool

Xiuyu Yao, Zheng Li, David Arthur, Lili Hu, Gen Cheng

The Violence Risk Screening-10 is one of the few instruments available for evaluating violence risk in mental health clients during and after hospitalization. This prospective study examined the applicability of this brief instrument with a sample of 289 clients in the 6 months after discharge from a general psychiatric hospital in Beijing. During the research period, 24 of the 289 clients demonstrated aggression. The receiver-operator characteristic curve yielded an area under the curve of 0.62. At the cut-off point of 4.5, its sensitivity/specificity was 79.2%/33.3%, and the corresponding positive/negative predictive value was 9.9%/94.5%. The predictive accuracy of this instrument was lower compared with the results of the original study, and was also less accurate than when it was administered while the clients were in the hospital. While promising in its utility for use beyond the hospital, it deserves further modification prior to its wide use across culturally-diverse China.