Utilization and Decision-Making Accuracy on Antithrombotic Prophylaxis by Caprini and Padua Risk-Assessment Models for Predicting Venous Thromboembolism in Hospitalized Patients

Background: Risk-Assessment Models may not accurately predict Venous Thromboembolism (VTE). This study prospectively assessed the use and decision-making accuracy of the Caprini and Padua risk-assessment models for antithrombotic prophylaxis in predicting VTE. Methods: Prospective study was conducted on 1075 Baghdad Teaching Hospital patients from December 1, 2022, until January 1, 2024. All patients and the subgroup examined for RAM discrimination using baseline medical and demographic data. Assessing Caprini and Padua scores' sensitivity and specificity. We analyze each score using the ROC curve and AUC. Results: The study observed 37 VTE cases. Individuals with VTE events had a significantly higher mean age (±SD) of 60.62 (15.8), p = 0.001. VTE occurs in 54% of acutely infected and/or rheumatologic patients P 0.002. Survival time and event-free period mean was 83.099 days (95% CI: 81.877-84.322) for patients without thromboprophylaxis and 74.085 days for those utilizing it. The Caprini RAM predicted VTE with 0.652 AUC, lower than Padua's 0.724. The Caprini RAM had 54.1% sensitivity and 61.7% specificity, whereas the Padua prediction score had 97.3% and 36.0%. Conclusion: The Caprini scores and Padua Prediction Score may improve hospital patients' thromboembolic risk classification compared to current practice.