Abstract
BACKGROUND: Forensic medium secure services in the UK are a scarce but essential resource providing care for those in the criminal justice system with severe mental disorder. Appropriate allocation of beds to those most in need is essential to ensure efficient use of this resource. To improve decision-making processes in a UK forensic service, an admissions panel utilized the DUNDRUM 1&2 (D1 & D2) triage instruments. METHODS: Demographic, diagnostic and clinical information on a prospective sample of referrals to a UK adult forensic service was gathered (n = 195). D1 and D2 measures were scored by a panel of clinical managers considering referral information and clinician opinion in reaching their ratings; those not admitted were also followed up. RESULTS: Within the sample, D1 ratings were predictive of decisions to admit (AUC = .79) and also differentiated between levels of security (F(4) = 16.54, p < .001). Non-admission was not significantly associated with increased risk of offending at follow-up. Items relating to self-harm and institutional behaviour did not show a predictive relationship with the panel decision to admit. CONCLUSIONS: Use of a structured professional judgement tool showing good predictive validity has improved transparency of decisions and appears to be associated with more efficient use of resources, without increased risk to the public.