验证Caprini和Padua风险评估模型联合使用在内科住院患者中筛选静脉血栓栓塞症（VTE）的有效性。

采用单中心回顾性研究，分析从2017年8~12月青岛大学附属医院内科住院患者资料，以Caprini和Padua风险评估模型对所有患者进行回顾性血栓风险评分，并通过Logistic二分类回归分析得到Caprini和Padua联合使用后的预测概率模型，分别绘制3种模型的受试者工作特征曲线（ROC），采用Delong法比较3种模型的曲线下面积，以Youden指数最大的分界点作为最佳诊断分界点。

Caprini和Padua联合使用后的预测概率模型ROC曲线下面积高于Caprini模型和Padua模型（0.927±0.009 vs 0.739±0.020，0.927±0.009 vs 0.817±0.017），差异具有统计学意义（Z=10.659，Z=8.952，P均<0.001）。并且预测概率模型灵敏度及特异度均大于Caprini或Padua（83.7% vs 79.1% vs 72.1%；87.9% vs 66.7% vs 82.7%）。

Caprini和Padua风险评估模型联合使用可提高筛选内科住院患者VTE的准确性。

To verify the efficacy of Caprini and Padua risk assessment models in screening of venous thromboembolism in patients in the department of internal medicine.

A single center retrospective study was performed in the inpatients at the Affiliated Hospital of Qingdao University from August 2017 to December. The Padua and Caprini risk scores were record retrospectively for each patient. Meanwhile, logistic bivariate regression analysis was performed to assess the predictive ability of combined Caprini and Padua risk assessment models for diagnosis of VTE. The operating characteristic curves (ROC) were plotted to calculate the area under the curve and pairwise comparisons were then performed. The cutoff point was chosen based on the Youden index.

The area under the ROC curve of the combination model was significantly higher than that of Caprini or Padua model alone (0.927±0.009 vs 0.739±0.020, Z=10.659, P<0.001; 0.927±0.009 vs 0.817±0.017, Z=8.952, P<0.001).The sensitivity and specificity of the combination model were also higher than those of Caprini or Padua model alone (83.7% vs 79.1% vs 72.1%; 87.9% vs 66.7% vs 82.7%).

The combination of Caprini and Padua risk assessment models can improve the accuracy of screening for VTE in patients in the department of internal medicine.