A Severity-Matching Strategy Illustrated Using the Alternative Model for Personality Disorders in Violence Risk Assessment on College Campuses

Use of the alternative model for personality disorders (AMPD) in risk assessment entails measuring personality functioning (PF) impairment and personality traits. However, instruments designed to measure the maladaptive end of the personality trait continuum might not be ideal when evaluating violence risk in populations showing relatively low levels of personality pathology. Rather, more accurate assessment may be achieved by using instruments that best match the individual's expected severity. This severity-matching strategy might prove especially beneficial for violence risk assessment in populations exhibiting low levels of personality pathology. One such population is college students, and literature suggests a growing need for evaluating violence risk on college campuses. The present study takes an initial step in investigating the severity-matching strategy, specifically in the context of using the AMPD when evaluating violence risk among college students. Normal-range personality traits were found to explain unique variance in college students' history and frequency of physical violence after accounting for PF impairment and pathological personality traits. However, the same was not found for pathological personality traits when the procedure was reversed. In light of these initial findings, we contend that the severity-matching strategy is an approach worthy of further research and that it is logical and appears tentatively valid to assess level of PF impairment and normal-range personality traits, à la the severity-matching strategy, as part of violence risk assessments in college settings where personality pathology is less common and less severe than commonly seen in forensic or clinical populations. Nevertheless, additional research is necessary. Public Significance Statement Preliminary examination of the severity-matching strategy, which has the potential to increase the precision and efficiency of violence risk assessment, produced encouraging results. Results also suggest that measures of normal-range personality traits may be more appropriate than pathological trait measures when evaluating violence risk among college students given this population's low levels of personality pathology.