We evaluated the Revised Trauma Score (RTS) for the rapid identification of severely injured patients on their arrival at the accident and emergency department. A total of 1407 consecutively injured patients admitted to the Accident and Emergency Department of the Royal Victoria Hospital, Belfast, had their RTSs calculated on arrival. A trauma team, consisting of experienced senior doctors, was summoned for all patients with an abnormal RTS of 11 or less. Each patient also had their Injury Severity Score (ISS) calculated later, and this was compared with their RTS on admission. It was found that 53 patients had an ISS of 16 or above, indicating severe injury. The RTS identified 42 of these. Of the remaining 11, five had incurred penetrating trauma and three spinal cord injuries. Clinical examination revealed actual or potentially serious injuries in all eight patients. However, serious injuries were not recognized initially in three patients due to physiological compensation and/or a short time lapse between injury and arrival at hospital. On admission 49 patients had an abnormal RTS, but their final ISS was less than 16. Forty had injuries warranting urgent resuscitation or a period of observation. Of the remaining nine patients, five had fractures with an elevated respiratory rate, presumably due to pain, and the other four had a depressed level of consciousness which could be accounted for by alcohol intoxication alone. Our results reinforce some well-known points. First, alcohol alone should not be assumed to be the only cause for a depressed level of consciousness in patients with head injuries. Second, systolic hypotension following trauma is an important sign of serious injury. We recommend the use of the RTS as an aid to junior doctors in the recognition of seriously injured patients in the accident and emergency department. Furthermore, the score should be recalculated at frequent intervals while the patient remains in the department.