Modification and validation of the Padua prediction score for assessing the risk of venous thromboembolism in medical inpatients: a case-control study.

AIM: To modify and validate the Padua prediction score in Chinese medical inpatients. DESIGN: A case-control study. METHODS: The clinical data of 663 Chinese internal medicine inpatients provided the basis for optimizing the model. Independent risk factors for Venous Thromboembolism (VTE) were analyzed by multifactorial logistic regression, and a modified risk assessment model was designed. Then 414 patients were selected as the external validation group, and the Receiver Operating Characteristic (ROC) and the McNemar test were used to assess the clinical applicability and discriminative power of the score. RESULTS: Obesity (body mass index ≥ 25 kg/m2), varicose veins, diabetes mellitus, and D-dimer ≥0.5 µg/mL were independent risk factors for the development of VTE in Chinese internal medicine inpatients, with a corresponding risk score of 1. When the modified Padua risk assessment model took the best critical value, the Area Under the Curve (AUC) was 0.80, the accuracy was 0.75, the sensitivity was 0.82, the specificity was 0.69, the positive predictive value was 0.67, and the negative predictive value was 0.83. McNemar test showed that there was a difference in the predictive effect results of the two scores, and, except for specificity, the modified score's predictive effects were better than those before modification. CONCLUSION: The Padua score with the proposed modifications was more effective in predicting the risk of venous thromboembolism in hospitalized Chinese individuals. LIMITATION: This study was based a limited number of cases and only proposed minor adjustments to the established scoring system.