欢迎访问《中国临床药理学与治疗学》杂志官方网站,今天是

中国临床药理学与治疗学 ›› 2002, Vol. 7 ›› Issue (3): 257-262.

• 药理试验设计与统计 • 上一篇    下一篇

临床试验中评价处理效应的协变量调整问题

刘玉秀, 姚晨1, 陈峰2, 陈启光3, 苏炳华4   

  1. 南京军区南京总医院信息科, 南京 210002;
    1解放军总医院医学统计教研室, 北京 100842;
    2南通医学院医学统计教研室, 南通 226001;
    3东南大学流行病学与生物统计系, 南京 210009;
    4上海第二医科大学生物统计教研室, 上海 200025
  • 收稿日期:2002-03-31 修回日期:2002-04-18 发布日期:2020-12-01
  • 通讯作者: 刘玉秀, 男, 副主任医师, 医学统计硕士, 主要从事临床试验设计与统计分析研究及信息管理工作。Tel:025-4826808-58346 Email:liu-yuxiu@163.com

Statistical adjustment of treatment effect for covariates in clinical trials

LIU Yu-Xiu, YAO Chen1, CHEN Feng2, CHEN Qi-Guang3, SU Bing-Hua4   

  1. General Hospital of Nanjing Command PLA, Nanjing 210002;
    1Department of Medical Statistics, General Hospital of PLA, Benjing 100853;
    2Department of Medical Statistics, Nantong Medical College, Nantong 226001;
    3Department of Epidemiology and Biostatistics, Southeast University, Nanjing 210009;
    4Department of Biostatistics, Shanghai Second U-niversity of Medical Sciences, Shanghai 200025
  • Received:2002-03-31 Revised:2002-04-18 Published:2020-12-01

摘要: 临床试验设计和分析中常常包括影响疾病结果的非处理因素, 这些因素被统称为协变量。对处理效应的评价, 若不合理考虑协变量的影响, 常常会导致结论的偏倚乃至错误。因此, 在临床试验中要审慎地对待这一问题, 在评价处理效应时应该考虑用协变量对其进行适当的调整, 以达到处理效应的正确评价。试验设计阶段的协变量调整称为“事先调整”, 常用的方法有分层法、动态分配最小化法等。试验分析阶段的协变量调整称为“事后调整”, 常用的方法有分层检验法、回归模型法等。通过协变量调整, 一方面阐明病人特征在处理组间的不均衡并没有影响到处理效应的正确评价, 提高临床试验结果的可信度;另外, 协变量调整方法的合理运用还可提高统计效能。本文对处理效应评价协变量调整涉及的概念、方法、步骤等进行介绍和探讨, 尤其强调了应用中的一些统计学事项。

关键词: 临床试验, 处理效应, 协变量, 分层, 最小化, 分层检验, 回归模型分析, ICH

Abstract: Somenon-treatment variables that affect the outcomeof a disease are often called covariates. These covariables should be consideredin the design andanalysis of clinical trials to obtain unbias conclusion.To ensure that any observed treatment effect is not influenced by an imbalances in baseline characteristics, both pread-justment and postadjustment are providedin the design stage andanalysis stage of the trials respectively.They can improve the credibility of the trial results andincrease the statistical efficiency.Based on a few papers published about adjustment for covariates and somedocumentations of the International Conference on Harmonization (ICH), wereview the concepts, methods and procedures for ad-justment of treatment effects for the influence of covari-ates.The statistical issues on the application of adjust-ment are especially discussedin great depth.

Key words: clinical trial, treatment effect, covari-ate, stratification, minimization, stratified tests, regres-sion model, ICH

中图分类号: