Predicting and understanding risk of re-offending: the Prisoner Cohort Study

Abstract

Risk assessment and risk management are key components of the Government’s
\nproposals to detain and treat individuals with Dangerous and Severe Personality
\nDisorder (DSPD). Accuracy in risk assessment plays a major role in identifi cation of
\nthe small group of individuals thought to pose a very high risk of harm to society and
\nin monitoring their level of risk during and after treatment (Douglas et al., 2005). The
\nPrisoner Cohort Study was a research project originally commissioned by the Home
\nOffi ce as part of the DSPD programme to evaluate the predictive accuracy of a range
\nof currently available risk assessment instruments for future violent and sexual reoffending.
\nThe main aims of the study were to test the accuracy in a UK population
\nof the risk assessment devices and instruments currently being piloted for use in the
\nDSPD centres in predicting serious re-offending, and to identify the best instruments in
\nterms of their accuracy in prediction. Most risk assessment instruments included in this
\nstudy were previously validated on US/Canadian male prisoners or forensic patients
\nwithout further differentiation. The study also examined the prevalence of offenders
\npotentially classifi able as having DSPD on the basis of the currently available risk
\ninstruments and personality assessments, and their dangerousness in terms of reoffending
\nafter release into the community. This report focuses on male offenders and
\nviolent re-offending. Data collection for sexual re-offending and for female offenders
\nwas ongoing at the time of the preparation of this report. Findings are therefore
\npresented for men serving determinate sentences for violent or sexual index offences
\ninterviewed in the fi rst phase of the study (N=1396).