The predictive validity of the Structured Assessment of Violence Risk in Youth (SAVRY) among institutionalised adolescents

The aim of this study was to examine the short-term predictive validity of the Structured Assessment of Violence Risk in Youth (SAVRY) in a sample of institutionalised adolescents. Subjects were 208 adolescents in general residential adolescent psychiatry, correctional schools, or adolescent forensic units. Demographic features and the information needed to assess violence risk with SAVRY were retrieved from medical files and child welfare records. Information on institutional violence was retrieved from the files for the six months after assessment (or at discharge if this occurred sooner). Validity was measured by comparing risk estimates and violent outcomes. Logistic regression analysis revealed odds ratios of 27.85 for the high risk group and 3.83 for the moderate risk group when compared to the low risk group, when adjusting the analyses for sex, age, diagnosis, service level, and time of follow-up. Based on these findings we can assume that the SAVRY is a useful and valid aid for professionals across several service levels when assessing risk of violent behaviour and intervention needs in youth with severe challenging and aggressive behaviours, even though we need further information on the impact of gender and psychopathology on violence risk in youth samples.