论文部分内容阅读
目的分析不同病例组合下非重症社区获得性肺炎患者的住院费用。为DRGs的实行、合理控制医疗费用提供理论依据。方法采用R语言中的glm()函数进行Logistic回归,筛选出影响非重症社区获得性肺炎患者住院费用的所有因素,再使用R语言rpart包中的rpart()和rpart.lpot()函数建立决策树模型的和绘制决策树图形。结果经logistic回归分析,影响非重症社区获得性肺炎患者住院费用的主要影响因素为年龄、住院天数、入院病情、有无手术。通过决策树分析,纳入住院天数、年龄和入院病情三个变量,最终确定7种病例组合。结论决策树和logistic回归结合能够评价不同病例组合下非重症社区获得性肺炎患者的住院费用。
Objective To analyze the hospitalization costs of patients with non-critical community-acquired pneumonia under different combinations of cases. Provide a theoretical basis for the implementation of DRGs and reasonable control of medical expenses. Methods Logistic regression was performed using the glm () function in R language to screen out all the factors influencing hospitalization costs in patients with nonspecific community-acquired pneumonia and to use the rpart () and rpart.lpot () functions in the R language rpart package to make decisions Tree model and draw decision tree graphics. Results By logistic regression analysis, the main influencing factors of hospitalization expenses of patients with non-critical community-acquired pneumonia were age, days of hospitalization, hospitalization, and surgery. Through the decision tree analysis, including the number of hospital days, age and admission of three variables, and ultimately determine the combination of seven cases. Conclusion The combination of decision tree and logistic regression can evaluate the hospitalization costs of patients with nonsmall community-acquired pneumonia under different combinations of cases.