中国药物警戒 ›› 2010, Vol. 7 ›› Issue (3): 142-144.

• 药品不良反应信号检测 • 上一篇    下一篇

药品不良反应信号检测中混杂因素的控制方法

钱维, 叶小飞, 王超, 贺佳*   

  1. 第二军医大学卫勤系卫生统计学教研室,上海 200433
  • 收稿日期:2016-02-19 修回日期:2016-02-19 出版日期:2010-03-08 发布日期:2016-02-19
  • 通讯作者: 贺佳,女,博士生导师,药物流行病学及新药评价的统计分析。E-mail:hejia@smmu.edu.cn
  • 作者简介:钱维,女,在读硕士研究生,数据挖掘在药物警戒中的应用。
  • 基金资助:
    国家自然科学基金资助项目(No.30872186),上海市优秀学科带头人计划(A类)(No.09XD1405500)

Methods of Controlling Confounding Factors in Adverse Drug Reaction Signal Detection

QIAN Wei, YIE Xiao-fei, WANG Chao, HE Jia*   

  1. Department of Health Statistics, Faculty of Health Service, Second Military Medical University, Shanghai 200433, China
  • Received:2016-02-19 Revised:2016-02-19 Online:2010-03-08 Published:2016-02-19

摘要: 上市后药品不良反应监测无法实现用药人群的随机化,因此无法控制药品不良反应信号检测中数据的混杂偏倚。通过介绍三种运用较为广泛的混杂因素控制方法,简单分析了各个方法的优缺点和在药品不良反应数据中的应用情况,以期为上市后药品不良反应的准确监测提供借鉴。

关键词: 信号检测, 混杂, 分层, Logistic回归, 倾向性评分法

Abstract: It is impossible to randomize the allocation of people using the drug under the post-marketing condition, so hard to control the confounding bias in signal detection of adverse drug reaction(ADR) databases. This article briefly introduces three widely-used Methods of controlling confounding bias. It give a brief description of the advantages and limitations of each method, and their applications in ADR signal detection in order to provide some effective information for the post-marketing surveillance.

Key words: signal detection, confounding, stratification, logistic regression, propensity score

中图分类号: