| 宋德胜,李长平,刘媛媛,崔壮,胡良平.生存资料回归模型分析——Cox比例风险假设的图形检验法[J].四川精神卫生杂志,2020,33(2):121-125.Song Desheng,Li Changping,Liu Yuanyuan,Cui Zhuang,Hu Liangping,Analysis of regression model of survival data——the test of the Cox’s proportional hazards assumption[J].SICHUAN MENTAL HEALTH,2020,33(2):121-125 |
| 生存资料回归模型分析——Cox比例风险假设的图形检验法 |
| Analysis of regression model of survival data——the test of the Cox’s proportional hazards assumption |
| 投稿时间：2020-03-12 |
| DOI：10.11886/scjsws20200312003 |
| 中文关键词: 生存分析 比例风险假设 SAS软件 Cox回归模型 |
| 英文关键词:Survival analysis Proportional hazards assumption SAS software Cox regression model |
| 基金项目:国家自然科学基金项目（项目名称：贝叶斯生存分析方法在肝细胞癌肝移植患者预后预测中的应用研究，项目编号：81803333） |
|
| 摘要点击次数: |
| 全文下载次数: |
| 中文摘要: |
| 本文目的是介绍目前使用图形检验比例风险的常用方法。经典的Cox比例风险回归模型要求生存资料满足比例风险假设，而在临床资料中，这个假设往往并不成立。鉴于此，本文首先阐述了比例风险假设的概念；然后介绍了一些检验比例风险假设是否成立的常用图示方法，主要包括Kaplan-Meier生存曲线图、ln[-ln(St)]生存时间关系图、缩放Schoenfeld残差与时间的关系图、SAS软件PHREG过程中ACCESS语句的PH和RESAMPLE选项产生的模拟路径图；最后，基于SAS软件并通过实例演示上述方法的实现。 |
| 英文摘要: |
| The purpose of this study was to introduce the current common methods of examining the assumption of proportional hazards using graphs.The classic Cox proportional hazards model requested the assumption of constant proportional hazards, which was often not true in clinical data. Given that, this article firstly interpreted the concept of the assumption of proportional hazards. Then introduced some common methods to test whether the proportional hazards were constant, which including graphical methods (Kaplan-Meier survival curve, plot about ln[-ln(St)] against survival time, scaled Schoenfeld residual against time diagram, simulation path diagram generated by the PH and RESAMPLE options in the ACCESS statement of the PHREG procedure). Finally, the implementation of the above methods in SAS software was demonstrated through an example. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |