Abstract: OBJECTIVE To establish the mathematical prediction model for postoperative surgical site infection（SSI） in the elderly fracture patients by multivariate logistic regression analysis and analyze its efficiency in prediction of infection. METHODS A total of 1 034 elderly fracture patients who were treated in Affiliated Hospital of Jiangnan University from Jan 2015 to Dec 2019 were enrolled in the study and divided into the infection group and the control group according to the status of postoperative infection. Multivariate logistic regression analysis was performed for risk factors for the postoperative SSI, the prediction model was established, and the efficiency of the prediction model was evaluated by means of receiver operating characteristic（ROC） curve. RESULTS Of the 1 034 patients, 60 had postoperative SSI, with the incidence rate 5.80%. Multivariate logistic regression analysis showed that the no less than 65 years of age, operation duration no less than 3 hours, type II/III surgical incision, time interval between injury and surgery no less than 8 hours, drainage tube indwelling time no less than 5 days and length of hospital stay no less than 7 days were the independent risk factors for the postoperative SSI（P＜0.05）. The prediction model for the postoperative SSI in the elderly fracture patients was Logit（P）=2.438+0.641×（age）+0.581×（operation duration）+0.504×（type II/III surgical incision）+0.482×（time interval between injury and surgery）+0.653×（drainage tube indwelling time）+0.543×（length of hospital stay）,Hosmer-lemeshow goodness of fit test showed that there was no significant difference between the prediction probability and the actual incidence rate. Verification test indicated that the area under ROC curve was 0.840,95%CI was 0.760～0.919, totally 100 cases were randomly selected for the verification of the Logistic regression prediction model, and the total accuracy rate was 76.00%. CONCLUSION The no less than 65 years of age, operation duration no less than 3 hours, type II/III surgical incision, time interval between injury and surgery, drainage tube indwelling time no less than 5 days and length of hospital stay no less than 7 days are the risk factors for the postoperative SSI in the elderly fracture patients. The efficiency of the prediction model established based on Logistic regression analysis is high in prediction of the postoperative SSI.