Decision tree analysis of hospitalization expenses for ischemic heart disease patients with multiple chronic diseases
Mao Ying1，Sun Lili1，Liu Fang1，Yang Shumei1，Guo Lingli1，Wang Lei2
1 Department of Medical Record Statistics, The Fifth Hospital of Xinjiang Medical University, Urumqi 830011, China;
2 Department of educational administration, Xinjiang Medical University, Urumqi 830011, China
Abstract：Objective To analyze the influencing factors of hospitalization expenses for ischemic heart disease patients with multiple chronic diseases in Xinjiang, and to predict the hospitalization expenses in patients with ischemic heart disease and the key differences.Methods The E-CHAID method was used to analyze the influencing factor data of hospitalization expenses for patients with multiple chronic ischemic heart diseases in Xinjiang, and the data were performed optimal segmentation to establish a case combination model. The structure and scale of the decision tree were controlled by pruning to predict the influencing factors, optimal combination path, and hospitalization expenses of ischemic heart patients with multiple chronic diseases.Results The 5 factors were treatment method, actual length of hospital stay, number of complications, the type of diseases, and chronic kidney disease or not. Through the decision tree model, 14 optimal combination paths are formed. Conclusion The decision tree model can be used to analyze the key factors affecting hospitalization expenses, but also help the health management departments to find the optimal grouping path, so as to provide ideas and assistance for optimizing clinical pathways and reforming medical insurance payments in the future.
［7］IZ-N G M, PARDINI C A. Association between medicare’s mandatory hospital value-based purchasing program and cost inefficiency［J］. Appl Health Econ Health Policy, 2018, 16(1):79-90. DOI: 10.1007/s40258-017-0357-3.
