Assessing dynamic violence risk: Common language risk levels and recidivism rates for the Violence Risk Scale

Area of Science:

• Forensic Psychology
• Criminology
• Psychometrics

Background:

• The Violence Risk Scale (VRS) is a crucial tool for assessing violence risk and guiding treatment in forensic and correctional settings.
• Existing VRS risk categories and recidivism estimates require updating to reflect current data and improve clinical utility.
• Accurate risk assessment is vital for effective violence prevention strategies and offender management.

Purpose of the Study:

• To develop new risk categories and recidivism estimates for the Violence Risk Scale (VRS).
• To identify specific rates of violent recidivism associated with VRS scores.
• To generate updated, evidence-based VRS risk categories with external validation for improved accuracy.

Main Methods:

• Utilized a multisite North American sample (N=1,338) of adult offenders with complete VRS scores and police-recorded outcomes.
• Employed logistic regression with VRS pretreatment and change scores (N=472) to model 2-, 3-, and 5-year violent and general recidivism estimates.
• Applied Council of State Governments' guidelines to create five risk levels using percentiles, risk ratios, and absolute recidivism estimates.

Main Results:

• Developed a VRS recidivism rates calculator based on logistic regression algorithms.
• Generated five distinct VRS risk levels with validated external construct validity.
• Established clear associations between VRS scores and rates of violent and general recidivism over 2-, 3-, and 5-year periods.

Conclusions:

• The updated VRS risk categories and recidivism estimates provide a more accurate and evidence-based tool for violence risk assessment.
• These enhancements can improve risk communication, treatment planning, and violence prevention efforts for offenders.
• The study underscores the importance of regularly updating risk assessment tools with contemporary data to ensure their effectiveness.