Background: Because of the need to improve the quality of care of trauma patients in our country, we decided to evaluate the epidemiology and find the most powerful tool for prediction of survival. The Trauma and Injury Severity Score (TRISS) has been known as conventional method for this purpose. We planned to test its ability for prediction of survival of our trauma patients, and also we wanted to compare its ability with the New Injury Severity Score (NISS) in combination with Revised Trauma Score (RTS) and age. We used the most suitable model to evaluate the trauma care in our centers.
Methods: From the Tehran University data registry on trauma patients of three different hospitals during 1 year, we selected trauma patients admitted to hospital for at least 1 day and all those patients who were declared dead at the emergency department. Epidemiologic description of patients has been given and evaluation of TRISS and (NISS + RTS + age) for prediction of survival has been performed. We determined factors affecting mortality and morbidity, evaluated hospitals, and analyzed patients admitted directly and the patients transferred from other hospitals.
Results: A total of 2,662 patients had complete data necessary for the calculation of probability of survival based on the TRISS method. The population at risk for trauma was the young, especially students and industrial workers. The major mechanisms of trauma were road traffic crashes and falls. The time expenditure and means of transportation as well as the time of stay in emergency department all seemed to be far less than optimal. We found that TRISS has higher performance than (NISS + RTS + age).
Conclusion: Based on our descriptive findings, we proposed some suggestions that seem to be necessary for improvement of trauma care in our centers. Among them were improved measures for prehospital service, and emergency department and other health care units of our centers. The findings of this study suggest that conducting trauma surgery training programs and direct transportation to trauma centers can improve the outcome of trauma patients. We conclude that small sample size, mixing penetrating trauma cases with blunt trauma cases, and differences in the mechanism of trauma between study populations may be responsible for the difference between our results and others.
Evaluating processes of care and outcomes of injured patients are important if improvements in the quality of care delivered to injured patients are to be accomplished. Although the characteristics of trauma systems in large North American cities have been evaluated with models, it is less clear which trauma system characteristics are critically important in large cities with unique characteristics different from those in North America. We planned to compare the ability of different methods of scoring of injury severity in combination with other parameters to find an accurate and powerful model for prediction of survival in our patients.
Trauma and Injury Severity Score (TRISS) and New Injury Severity Score (NISS) are tools for making predictions of risk-adjusted hospital mortality, each has limitations and we intended to evaluate the use of these models to determine the relative performance of trauma centers in Tehran for the purpose of identifying what improvements in quality of care could be made. Unfortunately, although ISS and TRISS have been almost universally accepted as the national standard in injury severity grading, there are several limitations that have been recognized. 1–6 TRISS has three major drawbacks: it has been demonstrated to be a less accurate predictor of survival than one would wish, it is designed for the prediction of survival as the only outcome measure in the care of injured patients, and the data are very expensive to collect and use. 7 NISS, which is defined as the simple sum of squares of the three most severe Abbreviated Injury Scale (AIS) injuries regardless of body region, has been shown to outperform ISS. 6
From the Department of Surgery, Section of Vascular Surgery (M.M., M.R.Z.) and Sina Trauma Research Center (H.R.), Sina Hospital, Tehran University of Medical Sciences, Tehran, Iran.
Address for reprints: Majid Moini, MD, Sina Hospital, Hassan-Abad Square, Tehran 11364, Iran; email: firstname.lastname@example.org.
Submitted for publication October 26, 1998.
Accepted for publication December 13, 1999.
This research was supported by Tehran University of Medical Sciences and Health Services.