Objective To verify the risk prediction efficacies of the Caprini risk assessment model and the new prediction probability model established based on the Caprini risk assessment model for venous thromboembolism associated with gynecological malignant tumors.Methods A total of 161 patients with gynecological malignancies complicated by VTE in Hunan Cancer Hospital and hospitalized between January 2015 and June 2020 were included in this study (VTE group), and 322 patients with gynecological malignancies in the same period were selected as the control group according to the ratio of 1∶2. The clinical data of patients in the two groups were collected to analyze the risk factors for the occurrence of VTE of patients with gynecological malignancies. A new Caprini thrombosis risk assessment model was constructed, and the receiver operating characteristics (ROC) curve was drawn to compare it with the original Caprini thrombosis risk assessment model.Results The Caprini score of the control group was lower than the VTE group (P<0.001). There were statistical differences in age, cancer stage, hypertension, hyperlipidemia, diabetes, history of blood transfusion, menopausal status, chemotherapy, blood platelet count and D-dimer level between the two groups (P<0.05). Multivariate logistic regression analysis showed that hypertension, hyperlipidemia, diabetes, history of blood transfusion, menopause, chemotherapy, blood platelet ≥300×10⁹ L^-1 and D-dimer ≥0.55 μm·mL^-1 were independent risk factors for VTE in patients with gynecological malignancies (P<0.05). Compared with the Caprini risk assessment model, the new prediction probability model increased the area under the ROC curve by 0.112 (P<0.001).Conclusion The Caprini risk assessment model has a moderate predictive value for the risk of VTE associated with gynecological malignant tumors. The newly constructed Caprini risk assessment model improved the accuracy of VTE risk prediction for patients with gynecological malignancies.