Gender differences in structured risk assessment: comparing the accuracy of five instruments

Area of Science:

• Forensic Psychology
• Criminology
• Risk Assessment

Background:

• Effective clinical risk management relies on structured assessment tools.
• Identifying the most accurate risk assessment instruments for male and female offenders is crucial but remains uncertain.

Purpose of the Study:

• To compare the predictive accuracy of various risk assessment instruments for violence and acquisitive offending in male and female prisoners.
• To determine which instruments offer the highest predictive validity for different types of offending and genders.

Main Methods:

• Prospective assessment of prisoners using the Psychopathy Checklist-Revised (PCL-R), Historical, Clinical, Risk Management-20 (HCR-20), Risk Matrix 2000-Violence (RM2000[V]), Violence Risk Appraisal Guide (VRAG), Offenders Group Reconviction Scale (OGRS), and total previous convictions.
• Comparison of predischarge measures with subsequent offending data.
• Multivariate regression analysis to rank instrument accuracy.

Main Results:

• Most instruments showed moderate predictive accuracy for offending.
• The OGRS was the most accurate predictor of violence among men.
• The PCL-R and HCR-20 H subscale showed the highest accuracy for predicting violence among women.
• The OGRS and total previous acquisitive convictions were most accurate for predicting acquisitive offending in both men and women.
• Actuarial instruments performed comparably to structured assessments and were superior for predicting male violence.

Conclusions:

• The choice of risk assessment tool should consider the type of offending and the offender's gender.
• Simpler, training-free actuarial measures can be as effective as more complex instruments.
• The OGRS and conviction history are valuable for predicting acquisitive offenses across genders.