目的: 通过对实施DRG付费后四川省攀枝花市某三甲医院医保患者住院费用的主要影响因素进行分析,为医院合理控制住院费用增长、降低亏损风险,以及医保部门制定DRG付费标准提供决策依据。方法: 研究对象为该院2021年纳入DRG付费的全部医保住院患者,收集其病案首页信息,采用BP神经网络建模,并与多元线性回归结果对比,分析医保患者住院费用的影响因素。结果: BP神经网络模型和多元线性回归模型均能较好地拟合数据。模型结果显示,住院天数、转科、输血和手术是影响DRG付费下医保患者住院费用的主要因素。结论: BP神经网络模型应用于住院费用影响因素的研究取得较好效果,具有很好的现实意义。针对住院费用的主要影响因素,医院应采取措施合理控费,降低DRG付费下可能出现的亏损风险,医保部门应适当考虑将影响因素纳入DRG进行细分组,使DRG付费更加科学合理。

Objective: The paper analyzes the main influencing factors of inpatient costs of medical insurance patients in a tertiary hospital in Panzhihua City of Sichuan Province after the implementation of DRG payment to provide decision-making basis for hospitals to reasonably control the growth of inpatient costs and reduce the risk of possible loss and medical insurance departments to develop DRG payment standards. Methods: The objects of study were all the medical insurance inpatients who were included in DRG payment in the hospital in 2021, and the home page information of their medical records was collected, modeled by BP neural network, and compared with the results of multiple linear regression to analyze the influencing factors of medical insurance patients' hospitalization costs. Results: Both the BP neural network model and the multiple linear regression model fit the data well. The results of both models showed that hospitalization days, transfer, blood transfusion, and surgery were the main influencing factors on the hospitalization cost of patients under DRG payment. Conclusions: The application of BP neural network model in the study of factors affecting hospitalization expenses works well and has great practical significance. For the main influencing factors of hospitalization cost, hospitals should take measures to reasonably control the cost and reduce the risk of possible loss under DRG payment, meanwhile, the medical insurance departments should consider incorporating the influencing factors into DRG for the subdivision group to make the payment more scientific and reasonable.