摘要
通过对敏感性分析新方法E-value的解读,向读者介绍一种简单实用的观察性研究敏感性分析方法。该方法以相对危险度(RR)作为主要研究指标,构建了RR敏感性分析的统计学模型,从而预测能够将研究结果(RR值)解释掉的,未知混杂因素与暴露因素及结局的最小关联强度。该方法还通过统计学转换,将RR推广到优势比、风险比及结局变化的均值的敏感性分析。该方法为观察性研究的敏感性分析提供了一种简单、可靠的分析方法,建议在以后的观察性研究报告及论文发表中提供相应的敏感性分析结果,但须注意,该敏感性分析并不能替代严谨的科研设计。
Through the interpretation of the new sensitivity analysis method E-value,a simple and practical sensitivity analysis method for observational research is introduced to readers.In this method,Relative Risk(RR)was used as the main research index to construct a statistical model of RR sensitivity analysis,so as to predict the minimum correlation strength of unknown confounders with exposure factors and outcomes that can explain away the research results(RR values).The method also extended RR to sensitivity analysis of Odds ratio,Hazard ratio and mean of outcome changes through statistical transformation.This method provides a simple and reliable method for sensitivity analysis of observational research,and it is suggested to provide corresponding results of sensitivity analysis in future observational research reports and papers,but it should be noted that this sensitivity analysis cannot replace rigorous scientific research design.
作者
姜棋予
孙丽君
毕京峰
JIANG Qiyu;SUN Lijun;BI Jingfeng(Clinical Research Management Center,the Fifth Medical Center,Chinese PLA General Hospital,Beijing 100039,China)
出处
《中国医药导报》
CAS
2020年第11期181-185,192,共6页
China Medical Herald
基金
首都卫生发展科研专项项目(首发2018-2-5035)。