BACKGROUND: More mentally disordered offenders (MDOs) are referred to secure psychiatric care settings than are accepted for admission. Psychiatrists working in different care settings may disagree on the appropriate level of security for MDOs, resulting in treatment delay. A pre-admission structured assessment of security needs for MDOs may facilitate agreement and access to care. AIMS: To assess the predictive validity and reliability of a structured assessment of security need (OPRISK) in a prospective cohort of referrals to high security hospital. METHOD: Operationalized criteria describing risk factors related to security need were used to develop OPRISK, an 18 item checklist. The predictive validity of OPRISK was assessed prospectively on the outcome of 140 referrals to Broadmoor high security hospital. RESULTS: Receiver operating characteristic curves of the predictive validity of OPRISK yielded an area under the curve of 0.765 (p < 0.001, 95% CI: 0.686-0.844). Internal consistency (>0.75) and inter-rater reliability (0.925) were high. CONCLUSION: OPRISK makes the evidence for an MDO's security needs explicit, aiding communication across service settings and improving access to care. Copyright 2008 John Wiley & Sons, Ltd.
BACKGROUND: More mentally disordered offenders (MDOs) are referred to secure psychiatric care settings than are accepted for admission. Psychiatrists working in different care settings may disagree on the appropriate level of security for MDOs, resulting in treatment delay. A pre-admission structured assessment of security needs for MDOs may facilitate agreement and access to care. AIMS: To assess the predictive validity and reliability of a structured assessment of security need (OPRISK) in a prospective cohort of referrals to high security hospital. METHOD: Operationalized criteria describing risk factors related to security need were used to develop OPRISK, an 18 item checklist. The predictive validity of OPRISK was assessed prospectively on the outcome of 140 referrals to Broadmoor high security hospital. RESULTS: Receiver operating characteristic curves of the predictive validity of OPRISK yielded an area under the curve of 0.765 (p < 0.001, 95% CI: 0.686-0.844). Internal consistency (>0.75) and inter-rater reliability (0.925) were high. CONCLUSION: OPRISK makes the evidence for an MDO's security needs explicit, aiding communication across service settings and improving access to care. Copyright 2008 John Wiley & Sons, Ltd.