TRISS methodology is a statistical technique by which the probability of survival in injury can be estimated. It combines an anatomical index of injury severity (the injury severity score, ISS), a physiological index (the revised trauma score, RTS), age and the mechanism of injury. In this study TRISS was used to assess 198 patients with penetrating injury at Baragwanath Hospital. One hundred and sixty-two patients had stab wounds, 26 gunshot wounds, and 10 received other injuries. Fifty patients were seriously injured (ISS > 15). Four patients (2%) died. The respective probabilities of survival for these four patients, as derived by TRISS, were 0%, 0%, 6% and 46%. TRISS demonstrated that the outcome of the patients as a group was comparable to other centres. TRISS is a useful method for monitoring the effectiveness of a trauma unit and objectively identifying cases for peer review.