[Objective] To analyze the risk factors of perioperative lower limb DVT in elderly hip fractures, establish a prediction model, and guide clinical prediction of the risk of DVT. [Methods] A retrospective review of the clinical data of 384 elderly patients with hip fractures admitted to our hospital from August 2015 to September 2021. Divided into DVT group and non-DVT group, single factor and multivariate analysis were used to obtain the risk factors of lower limb DVT. Establish a predictive model and evaluate the effectiveness of the model. [Results] Among the 384 patients, the perioperative rate was 67 (17.4% 67/384), 33 cases (8.6%) before operation, 34 cases (8.9%) after operation, and 10 cases remained unchanged before operation to after operation. (14.9%), 37 cases (55.2%) were aggravated, and 20 cases (29.9%) were relieved. Compared with the non-DVT group, compared with the non-DVT group, the patients in the DVT group had a longer waiting time before surgery, a higher proportion of blood transfusion and combined fractures, a longer operation time, more intraoperative blood loss, and a longer APTT. The differences were statistically significant. Academic significance (P<0.05). Multivariate binary logistic regression analysis showed: long preoperative waiting time (OR=1.110, CI=1.050～1.173), perioperative blood transfusion (OR=0.390, CI=0.221～0.688), combined with ≥3 medical diseases (OR =0.532, CI=0.300～0.944) is an independent risk factor for DVT of lower limbs during perioperative period. Based on the logistic regression results, the accuracy of the lower limb DVT prediction equation was 82.3%, the critical value was 0.319, the sensitivity was 71.6%, and the specificity was 60.3%; the area under the ROC curve was 0.705. [Conclusion] Long waiting time before operation, perioperative blood transfusion, combined with ≥3 kinds of medical diseases are independent risk factors for perioperative lower limb DVT. The DVT prediction model is helpful to clinically judge the risk of thrombosis and prevent it in advance.