Evaluation And Analysis Of The Prognosis Of Traffic Injury Patients With Two Comprehensive Scoring Methods

Objective:(1).Comparison of reliability and judgment ability between TRISS and ASCOT methods in predicting the outcome of traffic injury treatment;(2).Combined with two scoring characteristics,the combined score was used to predict the outcome of traffic injury treatment and to evaluate its predictive value.Methods:This study collected 930 cases of traffic injury patients who were treated from June2015 to July 2017 in Shanxi Dayi Hospital Affliated to Shanxi Medical University.The physiological and anatomical data were imported into the trauma database system V3.0for prediction and evaluation.The rule of evaluation is survival probability Ps ≥ 50%to predict survival and Ps < 50% to predict death.Clinical observation outcome is the gold standard.The combined scoring methods was evaluated according to the evaluation rules by taking the average value of Ps(TRISS)and Ps(ASCOT)for the same patient.The prediction results of TRISS and ASCOT scores were obtained by statistical analysis of the score results.ROC curves were drawn to evaluate the predictive effectiveness of TRISS,ASCOT and combined scoring methods.Results:(1)Among 930 traffic injury patients,there were significant differences between the survival group and the death group(P < 0.05).The TRISS score survival error rate was 0.5%,and death error rate was 26.5%.The survival error rate of ASCOT score was2.7%,and that of death error rate was 5.9%.(2)The Kappa test was used to analyze the consistency of TRISS,ASCOT and thecombined score for survival prediction.The Kappa values were 0.807,0.810,0.946 respectively.The area under the ROC curve of the TRISS,ASCOT and the combined scoring method were 0.865,0.957,0.982 respectively.Conclusion:The TRISS,ASCOT and the combined scoring method increased the consistency of survival prediction in turn.The prediction accuracy of ASCOT method is higher than that of TRISS method.In this study,the overall predictive value of the combined score is better than that of single score,and the prediction value of the combined score still needs to be verified by multi-center and large sample.