They compared the predictive accuracy of two risk-assessment methods that are easy to use in routine clinical settings—brief risk-assessment tools and patients' self-perceptions of risk—in "high-risk" inpatients who had a mental illness other than schizophrenia and had a comorbid substance use disorder. During interviews at a large psychiatric hospital with the 86 patients, the researchers used the Clinically Feasible Iterative Classification Tree and the Modified Screening Tool. They also had the patients indicate their own perceptions of their risk for violence. Two months after hospital discharge, the participants completed a follow-up interview in which they were asked about violence since discharge. The patients' self-perceptions turned out to be more accurate predictors than did the two assessment tools.
The researchers noted that they do not know if these findings will generalize from a research to a clinical context and that using self-perceptions should be one tool in violence risk assessment and "not a replacement" for other assessment methods.
Read more about the study in the new issue of Psychiatric Services here. Also see Psychiatric News here.
(image: Minerva studio/Shutterstock.com)