Purpose: The aim of this study was to evaluate the quality of the trauma care system of our hospital, in which emergency physicians care for major trauma patients in the emergency intensive care unit (ICU) in consultation with intervention radiologists and surgeons. Methods: This was a retrospective observational study conducted in an emergency ICU of a tertiary referral hospital. We enrolled consecutive patients who had been admitted to our emergency ICU with major trauma from March 2007 to September 2010. We collected data with respect to demographic findings, mechanisms of injury, the trauma and injury severity score (TRISS), emergency surgery, angiographic intervention, and 6-month mortality. Then, we compared the observed and predicted survivals of the patients. The Hosmer-Lemeshow test and calibration plots by using 10 groups, one for each decile, of predicted mortality were used to evaluate the fitness of TRISS. P-values of greater than 0.05 represent a fair calibration. Results: Among 116 patients, 12 (10.34%) were dead within 6 months after admission to the ICU, and 29 (25.00%) and 38 (32.80%) patients received emergency surgery and angiographic intervention, respectively. The mean injury severity score and revised trauma score were $36.97{pm}17.73$ and $7.84{pm}6.75$, respectively. The observed survival and the predicted survival of the TRISS were 89.66% (95% confidence interval [CI]: 84.03~95.28%) and 69.85% (95% CI: 63.80~75.91%), respectively. The calibration plots showed that the observed survival of our patients was consistently higher than the predicted survival of the TRISS ($p$ <0.001). Conclusion: The observed survival for the trauma care system of our hospital, in which emergency physicians care for major trauma patients in the emergency ICU in consultation with intervention radiologists and surgeons, was higher than the predicted survival of the TRISS.