Violence risk assessment: Getting specific about being dynamic.

Abstract

Substantial strides have been made in the field of violence risk assessment. Numerous robust risk factors have been identified and incorporated into structured violence risk assessment instruments. The concepts of violence prevention, management, and treatment have been infused into contemporary thinking on risk assessment. This conceptual development underscores the necessity of identifying, measuring, and monitoring changeable (dynamic) risk factors-the most promising targets for risk reduction efforts. However, empirical investigation of dynamic risk is virtually absent from the literature. In this article, the authors (a) differentiate risk status (interindividual risk level based largely on static risk factors) from risk state (intraindividual risk level determined largely by current status on dynamic risk factors), (b) analyze the relevance of contemporary risk assessment measures for capturing dynamic risk, and (c) distill potentially important dynamic risk factors from the literature in order to facilitate future research. Suggestions for theory development and research design are provided.

The impact of an intensive inpatient violent offender treatment programme on intermediary treatment targets, violence risk and aggressive behaviour in a sample of mentally disordered offenders

“…Although violent offenders with SMI may have additional treatment needs and some unique risk factors (e.g., symptoms of major mental illness) that also impact treatment responsivity, dynamic risk factors are a critical focus of assessment and treatment for violent offenders. Douglas and Skeem (2005) suggest that dynamic risk factors are: (1) antecedent to, and increase the propensity for, violence, (2) changeable, and, (3) predict changes in violent re-offending as a result of treatment. Many widely used structured violence risk assessment instruments comprise well-known dynamic risk factors (e.g., HistoricalClinical-Risk Management-20 version 3, HCR-20 v3 , Douglas, Hart, Webster & Belfrage, 2013;Violence Risk Scale, VRS , Wong & Gordon, 2003) and many of these dynamic risk factors are associated with violent recidivism, thereby satisfying Douglas and Skeem's (2005) first criteria; however, there is (1) little evidence that many of these dynamic risk factors can change in violent offenders (either over time or as a consequence of treatment), and (2) little research that has examined whether changes in dynamic risk factors are associated with reductions in violent re-offending following release from custody (Klepfisz, Daffern & Day, 2015).…”

“…Douglas and Skeem (2005) suggest that dynamic risk factors are: (1) antecedent to, and increase the propensity for, violence, (2) changeable, and, (3) predict changes in violent re-offending as a result of treatment. Many widely used structured violence risk assessment instruments comprise well-known dynamic risk factors (e.g., HistoricalClinical-Risk Management-20 version 3, HCR-20 v3 , Douglas, Hart, Webster & Belfrage, 2013;Violence Risk Scale, VRS , Wong & Gordon, 2003) and many of these dynamic risk factors are associated with violent recidivism, thereby satisfying Douglas and Skeem's (2005) first criteria; however, there is (1) little evidence that many of these dynamic risk factors can change in violent offenders (either over time or as a consequence of treatment), and (2) little research that has examined whether changes in dynamic risk factors are associated with reductions in violent re-offending following release from custody (Klepfisz, Daffern & Day, 2015). There is some support for the proposition that intra-individual change in dynamic risk factors is associated with reduced reoffending in prisoners (Serin, Lloyd, Helmus, Derkzen & Luong, 2013), and there is also evidence that reduction in aggregate dynamic risk factors measured using multi-item structured violence risk assessment instruments is associated with reduced violent recidivism in populations with SMI (De Vries Robbé, de Vogel, Douglas & Nijman, 2015;Lewis, Olver & Wong, 2012).…”

Does reassessment of risk improve predictions? A framework and examination of the SAVRY and YLS/CMI.

“…Many of these tools conceptualize risk as dynamic, and as such, they include factors that can change (Douglas & Skeem, 2005). This is perhaps particularly true of adolescent risk assessment tools (Viljoen, Cruise, Nicholls, Desmarais, & Webster, 2012).…”

Predictive validity of self‐reported self‐control for different forms of recidivism

“…A second limitation concerns the conclusions that may be drawn about our SC-related trait clusters as predictors of recidivism. In order to establish patterns of SC-related trait clusters as a causal dynamic risk factor, future studies would have to demonstrate that these pattern changes spontaneously through intervention and that a change in SC scores predicts changes in the likelihood of reoffending (Douglas and Skeem, 2005). If this were the case, treatment programs tailored to change this pattern of SC-related trait clusters should theoretically be able to influence recidivism rates (Andrews et al,1990).…”