Abstract
Background. Trauma and Injury Severity Score (TRISS) methodology has become a standard tool for evaluating the performance of trauma centres and identifying cases for critical review. Recent work has identified several limitations and questioned the validity of the methodology in certain types of trauma. Methods. The usefulness and limitations of the TRISS methodology were evaluated in an urban trauma centre. Trauma registry data of 5445 patients with major trauma were analysed with respect to 30 demographic, prehospital, injury severity and hospitalization attributes. The performance of TRISS was measured primarily by the percentage of misclassifications, including false positives and false negatives, comparing the survival status predicted by TRISS with the true status. Sensitivity, specificity, and positive and negative predictive values were also measured for subgroups defined by the 30 attributes. Logistic regression analysis was used to identify significant independent factors related to the performance of TRISS. Results. The overall misclassification rate was 4.3 per cent. However, in many subgroups of patients with severe trauma the misclassification rate was very high: 34 per cent in patients older than 54 years with Injury Severity Score (ISS) greater than 20; 29 per cent in those with fall injuries and ISS above 20; 29 per cent in patients with injuries involving four or more body areas and ISS greater than 20; 28.6 per cent in patients with injuries needing admission to the intensive care unit (ICU) and ISS greater than 20; 26.4 per cent in patients in severe distress before reaching hospital with ISS greater than 20; and 26.1 per cent in patients whose ISS score was above 20 and who had complications in hospital. Conclusion. The TRISS methodology has major limitations in many subgroups of patients, especially in severe trauma. In its present form TRISS has no useful role in major urban trauma centres. Its use should be seriously, reconsidered, if not abandoned.