Examining the utility of strengths and protective factors in violence risk assessment measures in a tertiary civil psychiatric population

Research on risk assessments utilizing protective factors with civil psychiatric populations remains limited. Additionally, there has been some debate regarding the generalizability of risk assessment measures to female populations given that many of these measures were developed with male populations. Yet, no known studies exist that have made direct comparisons between male and female civil psychiatric patients on protective factors and violence prediction. The Structured Assessment of Protective Factors for Violence Risk (SAPROF; de Vogel et al., 2008) is a structured professional judgment risk assessment measure intended to be used in conjunction with a measure of risk factors, (i.e., HCR-20, Webster et al., 1997). The Short-Term Assessment of Risk and Treatability (START; Webster et al., 2004) is a multidisciplinary tool that is intended for use in inpatient and community settings and for which clinicians code the measure by considering a client’s strengths and vulnerabilities. Until now there has been no empirical comparison of the START and SAPROF. To address these gaps in the current body of knowledge on the utility of protective factors in violence risk assessment, we conducted a prospective study, utilizing interview and file review to investigate strength based risk assessments in a male and female civil psychiatric population. Participants included 102 civil psychiatric patients residing at a large tertiary psychiatric hospital who were being transferred to community-based tertiary inpatient settings. Baseline file reviews and interview-based assessments were conducted prior to patient transfer to community based treatment facilities. Outcome data (i.e., verbal, sexual, and physical aggression) was collected every 6 months over a 12-month period. We found evidence that supports the use of strength based risk assessments with civil psychiatric populations. On the whole, the psychometric properties for all the measures included were good. Protective factors demonstrated incremental validity over the risk factors alone, as did summary risk judgments over actuarial assessments. We found a number of potentially interesting gender differences in the predictive validity of all the risk measures included. Generally, these measures performed better with the males when predicting physical and verbal aggression, and better with the females when predicting sexual aggression.