Development and Evaluation of a CDSS-Enabled Early Warning System for Venous Thromboembolism Risk in Advanced Lung Cancer Patients

Abstract

BACKGROUNDː This study establishes the clinical practice library of Venous Thromboembolism (VTE) complications in advanced lung cancer, forms a VTE risk warning system for advanced lung cancer patients based on the Clinical Decision Support System (CDSS) and plan to use it in hospitals in the future, and plays a role in preventing VTE complications in advanced lung cancer patients. METHODSː We summarized the VTE risk factors by searching the literature and constructed a knowledge base of advanced lung cancer complication VTE risk factors based on evidence-based medicine; collected real-world clinical data on lung cancer patients and constructed a real-world best-practice library of advanced lung cancer complication VTE by cleaning, quality control and analyzing clinical data; and constructed an equation for the prediction risk coefficient of advanced lung cancer complication VTE by logistic regression analysis. Finally, the three were combined to construct a CDSS-based VTE risk warning system for advanced lung cancer patients. Statistical analysis was performed using R software, with a test level of α = 0.05, and the difference was considered statistically significant at P < 0.05. RESULTS: A total of 12,222 patients were screened through the hospital's electronic medical record system, and 3,320 patients were included after screening in strict accordance with the exclusion criteria. The patients were divided into 67 cases in the group of lung cancer complicating VTE and 3253 cases in the group of uncomplicated VTE. According to the stepwise regression method and the AIC law, the equation of the predicted risk coefficient of advanced lung cancer complicating VTE was constructed. Combining the three basic factor modules of advanced lung cancer complicating VTE and the CDSS system, a CDSS-based risk warning system for advanced lung cancer complicating VTE was established, and the externally validated results showed that the AUC of 0.756 (95% CI: 0.726–0.785) was better than that of the Caprini risk scoring scale AUC of 0.654 (95% CI: 0.638–0.670). CONCLUSIONSː This study established a CDSS-based early warning system for the risk of concurrent VTE in advanced lung cancer. The results of comparative validation showed that this early warning system can improve the clinical diagnosis and treatment efficiency of advanced lung cancer patients with complications of VTE, and at the same time, it can achieve the purpose of real-time monitoring and timely diagnosis and treatment.