Abstract: Objective To quantitatively evaluate the treatment quality of trauma integration treatment system in the patients with severe trauma. Methods Records of patients with severe trauma hospitalized in our department from January 2010 to December 2012 were extracted from trauma database and analyzed, including gender, age, basic situation after admission [including systolic pressure, diastolic pressure, axillary temperature, heart rate, Injury Severity Score (ISS), Acute Physiology and Chronic Health Evaluation (APACHE) Ⅱ score, Glasgow Coma Score, and urine volume on the first day], the first time determination values of physiological and biochemical indexes after admission (including pH value, base excess, PaCO₂, PaO₂, standard bicarbonate ion, leucocyte count, neutrophile granulocyte, hemoglobin, platelet count, albumin, urea nitrogen, lactic acid, blood glucose, and blood sodium), surgical situation, length of ICU stay, occurrence of major complications [including infection, acute respiratory distress syndrome (ARDS), multiple organ dysfunction syndrome (MODS)/multiple organ failure (MOF)], and death. Single factor analysis was used to screen death-associated exposure factors, then the exposure factors were brought into multivariate Logistic regression to establish adjustment mortality models to calculate observation/expectation (O/E) ratio of adjustment mortality of patients in these three years, and Poisson distribution was used to calculate the 95% confidence interval (CI) of O/E ratio. Data were processed with Student t test, Wilcox test, chi-square test and or Fisher's exact test. Results A total of 536 patients with severe trauma were enrolled in these three years, with 438 male (81.72%) and 98 female (18.28%). There were no statistically significant differences in gender, age, and basic situation of patients after admission among these three years (χ²=0.16, with t values from 0.05 to 104.50, W values from 0.008 to 104.500, P values above 0.05). There were no statistically significant differences in the first time determination values of physiological and biochemical indexes after admission including pH value, base excess, PaCO₂, PaO₂, hemoglobin, platelet count, and blood sodium of patients among these three years (with t values from 0.80 to 29.10, W values respectively 0.110 and 5.450, P values above 0.05), while there were statistically significant differences in standard bicarbonate ion, leucocyte count, neutrophile granulocyte, albumin, urea nitrogen, lactic acid, and blood glucose of patients among these three years (with t values from 1 542.00 to 500 000.00, W values from 637.000 to 500 000.000, P<0.05 or P<0.01). There were no statistically significant differences in surgical situation, length of ICU stay, and occurrence of major complications including infection, ARDS, and MODS/MOF in patients among these three years (with χ² values from 0.48 to 2.43, W =2.100, P values above 0.05). The mortality of patients in 2010, 2011, and 2012 were 11.9% (19/159), 11.2% (21/187), and 7.4% (14/190), respectively, showing a trend of decline, but there was no statistically significant difference (χ²=2.43, P>0.05). Death-associated exposure factors were age, ISS, APACHE Ⅱ score, urea volume on the first day, platelet count, albumin, and blood sodium. The O/E ratio of adjustment mortality (95%CI) in 2010, 2011, and 2012 were 0.727 (0.460-1.180), 0.718 (0.460-1.230), and 0.460 (0.270-0.840), respectively, showing a trend of decline each year. Conclusions The trauma integration treatment system can improve the treatment quality of patients with severe trauma.