In June 1990 the Ministry of Health designated eleven hospitals throughout Ontario to be lead hospitals in trauma care. An integral part of a trauma system is the evaluation of care, in particular, outcome of the trauma patients. The TRISS methodology, which offers a standard approach for evaluating outcomes for different populations of trauma patients, was used to determine if there was an improvement in outcomes following the designation of trauma centres of patients involved in motor vehicle crashes (ICD-9-CM E810.0-825.9) with an Injury Severity Score (ISS) > 12 for two twelve month periods, one pre-designation (89/90) and one post-designation (92/93). The Revised Trauma Score (RTS), ISS, age and outcome were calculated or abstracted from the hospital chart of each patient at the trauma centre. The probability of survival of each patient, the z and W statistics of both years were calculated. A measurable improvement was shown in z statistics between the two years from z = -0.40 PRE to z = from TRISS analysis was removed, using a TRISS-like [as per Offner et al, 1992] logistic regression equation which +0.72 POST. When the bias introduced by patients intubated prior to arrival at the trauma centre being excluded allows analysis of intubated patients, the improvement was even greater, with z = + 1.34 PRE and z = +2.97 POST. Only the statistically significant z score of the post-designated year required the W score to be calculated, W = +5.60. This value of W signifies that there were approximately 6 more survivors per 100 patients treated at the trauma centre post-designation than expected from norms. This study demonstrates and quantifies the improvements in patient outcomes associated with trauma centre designation. It also identifies a limitation of TRISS analysis and a method of eliminating the bias in trauma centre populations that are largely referred rather than direct off-the-street.