Predicting violence in female forensic inpatients with substance use disorders – the utility of a gender-responsive assessment

Abstract

Introduction: Given that risk assessment tools are commonly based on male samples, the applicability to justice-involved women remains to be clarified. This study aimed at assessing (1) the predictive validity of the HCR-20 V3, the prevailing, yet primarily male-based violence risk assessment instrument, and (2) the incremental validity of the FAM, a gender-responsive supplement, for both inpatient violence and violent recidivism in justice-involved women. Methods: The sample included 452 female forensic inpatients with substance use disorder discharged from German forensic psychiatric care between 2001 and 2018. Results: ROC analyses revealed good predictive accuracy for the HCR-20 V3 while the FAM failed to provide incremental validity. Further, binary logistic regression determined several predictors of violence including personality disorder, covert/manipulative behavior, suicidal behavior/self-harm, and problematic intimate relationship. Discussion: These findings support the applicability of the HCR-20 V3 in justice-involved women with substance use disorder, while highlighting the clinical relevance of the FAM in supporting a gender-informed risk management.