Utility of urine drug screening: a clinical audit

Abstract

OBJECTIVE: To assess requesting patterns and clinical utility of urine drug screens (UDS). METHODS: All UDS requests received between March 2002 and October 2002 were identified. Two trained assessors reviewed case histories and categorized the utility of the UDS and gave a utility score using a Visual Analogue Scale. RESULTS: Of 171 UDS requested, 67.3% were men. Age range was 15-92 (mean 36) years. In total, 70.8% were received from accident and emergency centre (AEC). Clinical utility was identified in only 85 of 171 UDS. Sixty-one per cent of requests from AEC were defined as having no clinical utility. Twenty-four per cent of requests originating from other specialities were identified as having no clinical utility. Overall, a test was more likely to be of utility if it was requested from non-AEC than from AEC (Pearson's chi(2) two-sided t-test, P < 0.0001). Indications for UDS were toxic ingestion (33.3%), psychosis (15.2%), impaired consciousness (13.5%), behavioural disturbance (4.1%), confusion with HIV infection (3.5%), alleged adulteration (3.5%), seizure (9.4%), cardiac arrest (1.8%) and miscellaneous (15.8%). CONCLUSIONS: Our data show excessive requesting for UDS in routine overdose, contrary to international guidelines. Requests for UDS originating from AEC are more likely to have no utility than those from other specialities. Availability of point of care UDS might increase use of UDS, but cost-effectiveness and utility must be assessed.