Predicting violent reconvictions using the HCR-20.

Area of Science:

• Forensic Psychiatry
• Clinical Psychology
• Criminology

Background:

• Accurate risk assessment for future violence is crucial for public safety and patient care planning.
• Structured clinical assessments offer a standardized approach to improve risk evaluation accuracy.

Purpose of the Study:

• To evaluate the effectiveness of the Historical, Clinical and Risk Management Scales (HCR-20) risk assessment tool.
• To assess the HCR-20's predictive accuracy in a large cohort of male forensic psychiatric patients discharged from UK medium secure units.

Main Methods:

• A pseudo-prospective study followed 887 male patients for at least 2 years post-discharge.
• HCR-20 assessments were based solely on pre-discharge information.
• Post-discharge offending data (violent and other) were collected from official records.

Main Results:

• The HCR-20 total score demonstrated good predictive capability for both violent and other offenses.
• Historical and risk sub-scales significantly predicted offending, while the clinical sub-scale did not.
• Predictive accuracy was highest for periods under 1 year, with a slight decrease over 5 years.

Conclusions:

• The HCR-20 provides a robust evidence base for predicting violent and non-violent offending in male forensic patients released from medium secure facilities.
• The findings support the HCR-20's utility in risk management for this population.