论文部分内容阅读
目的:确立科学的分类节点变量,为建立病例分组做准备。方法:在运用正态性检验、核密度分布图以及Box-Cox转换等方法的基础上,用非参数检验Kruskal Wallis方法和多因素回归方法分析选取分组的分类节点变量。结果:分别对住院天数和住院费用这两种统计分类指标的35大类统计,筛选的主要分类节点变量为ICD-10国际疾病编码前一位码、是否手术、疾病危重度、伴随症和转归情况,也有少数大类涉及性别、年龄、入院情况、是否随诊、是否抢救以及是否感染等变量。结论:选取的分类节点变量科学。
Objective: To establish a scientific classification of node variables, to prepare the case group. Methods: Based on the normality test, kernel density distribution and Box-Cox transformation, Kruskal Wallis method and multivariate regression method were used to analyze the classification node variables. Results: The 35 major categories of hospitalization days and hospitalization costs were statistically analyzed. The main classified node variables were ICD-10 International Code of Diseases code, whether surgery, critical illness, complication and transfer There are also a few major categories that deal with variables such as gender, age, admission, follow-up, salvage, and infection. CONCLUSION: The selected classification node variable science.