Violence risk assessment for young adults receiving treatment for early psychosis.

Abstract Aim Although the absolute risk of violence is small for individuals with mental illnesses, a specific subgroup of individuals who appear to be at increased risk for violence includes young people experiencing emerging or early psychosis. Prior research has identified risk factors for violence in this population, though no prior studies using a formal risk assessment tool have been identified. This study used the Historical Clinical Risk Management-20, version 3 (HCR-20) to identify risk of future violence among a sample of young adults with early psychosis and relevant predictors of risk unique to this population. Methods The HCR-20 was administered to a sample of young adults with early psychosis (N = 53) enrolled at one OnTrackNY site, part of a statewide program providing early intervention services to young adults presenting with a first episode of non-affective psychosis. A Confirmatory Factor Analysis (CFA) was conducted to explore the relative importance of the HCR-20 items for this population. Results The average age of participants was 21.9 years (SD 3.6 years) and most were male (69.8%, n = 37). Most patients were assessed to be at low risk for future violence based on the Case Prioritization summary risk rating (67.9%, n = 36). The CFA identified 4 items that were not of relative predictive value in identifying the risk of violence in this sample: history of substance use (item H5), history of major mental disorder (item H6), living situation (item R2), and personal support (item R3). Conclusion This study presents a formal approach to assessing violence risk in a population at elevated risk of violence, demonstrates the feasibility of using a standardized risk assessment tool in early intervention services, and identifies factors of particular importance associated with predicting violence in this population. Future research should implement violence risk assessment with a structured tool such as the HCR-20 and assess its accuracy in predicting future violent behavior in this setting.