观察性研究的敏感性分析：“E-value”解读

Abstract
Figure/Table
References (29)
Related Citation (0)

Download: PDF (796 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract 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.

Key words： Observational study Sensitivity analysis E-value Interpretation

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	JIANG Qiyu SUN Lijun BI Jingfeng▲

Cite this article:

JIANG Qiyu SUN Lijun BI Jingfeng▲. Sensitivity analysis of observational studies: “E-value” interpretation[J]. 中国医药导报, 2020, 17(11): 181-185,192.

URL:

http://www.yiyaodaobao.com.cn/EN/ OR http://www.yiyaodaobao.com.cn/EN/Y2020/V17/I11/181

[1] Bross ID. Spurious effects from an extraneous variable [J]. J Chronic Dis，1966，19（6）：637-647.
[2] Schlesselman JJ. Assessing effects of confounding variables [J]. Am J Epidemiol，1978，108（1）：3-8.
[3] Rosenbaum PR，Rubin DB. Assessing sensitivity to an unobserved binary covariate in an observational study with binary outcome [J]. J R Stat Soc Series B，1983，45：212-218.
[4] Lin DY，Psaty BM，Kronrnal RA. Assessing the sensitivity of regression results to unmeasured confounders in observational studies [J]. Biometrics，1998，54（3）：948-963.
[5] Imbens GW. Sensitivity to exogeneity assumptions in program evaluation [J]. Amer Econ Rev，2003， 93：126-132.
[6] Vanderweele TJ，Arah OA. Bias formulas for sensitivity analysis of unmeasured confounding for general outcomes，treatments，and confounders [J]. Epidemiology，2011，22（1）：42-52.
[7] Bross ID. Pertinency of an extraneous variable [J]. J Chronic Dis，1967，20（7）：487-495.
[8] Lee WC. Bounding the bias of unmeasured factors with confounding and effectmodifying potentials [J]. Stat Med，2011，30（9）：1007-1017.
[9] Robins JM，Rotnitzky A，Scharfstein DO. Sensitivity analysis for selection bias and unmeasured confounding in missing data and causal inference models [M]// Statistical Models in Epidemiology, the Environment, and Clinical Trials. Springer New York，2000：1-94.
[10] McCandless LC，Gustafson P，Levy A. Bayesian sensitivity analysis for unmeasured confounding in observational studies [J]. Stat Med， 2007，26（11）：2331-2347.
[11] Brumback BA，Hernán MA，Haneuse SJ. Sensitivity analyses for unmeasured confounding assuming a marginal structural model for repeated measures [J]. Stat Med，2004，23（5）：749-767.
[12] VanderWeele TJ. Explanation in Causal Inference：Methods for Mediation and Interaction [M]. New York：Oxford University Press，2015，Chapter 3.
[13] VanderWeele TJ. Sensitivity analysis for contagion effects in social networks [J]. Sociol Methods Res，2011，40（2）：240-255.
[14] Ding P，VanderWeele TJ. Sensitivity analysis without assumptions [J]. Epidemiology，2016，27（3）：368-377.
[15] Gilbert PB，Bosch RJ，Hudgens MG. Sensitivity analysis for the assessment of causal vaccine effects on viral load in HIV vaccine trials [J]. Biometrics，2003，59（3）：531-541.
[16] Chiba Y，VanderWeele TJ. A simple method for principal strata effects when the outcome has been truncated due to death [J]. Am J Epidemiol，2011，173（7）：745-751.
[17] Huang TH，Lee WC. Bounding formulas for selection bias [J]. Am J Epidemiol，2015，182（10）：868-872.
[18] Robins J，Orellana L，Rotnitzky A. Estimation and extrapolation of optimal treatment and testing strategies [J]. Stat Med，2008，27（23）：4678-4721.
[19] Hernán MA，Lanoy E，Costagliola D. Comparison of dynamic treatment regimes via inverse probability weighting [J]. Basic Clin Pharmacol Toxicol，2006，98（3）：237-242.
[20] Ding P，VanderWeele TJ. Sharp Sensitivity Bounds for Mediation Under Unmeasured Mediator-Outcome Confounding [J]. Biometrika，2016，103（2）：483-490.
[21] VanderWeele TJ，Ding P. Sensitivity Analysis in Observational Research：Introducing the E-Value [J]. Ann Intern Med，2017，167（4）：268-274.
[22] 于彤彤，宋娇磊，刘双双，等.基于倾向性评分匹配法探讨血清白蛋白对心力衰竭患者院内死亡的影响[J].中华内科杂志，2015，54（11）：959-964.
[23] VanderWeele TJ. On a square-root transformation of the odds ratio for a common outcome [J]. Epidemiology，2017， 28（6）：e58-e60
[24] 许春杰，刘晓宇，尹素凤，等.基于剂量-反应关系的体力活动对血清肌酐水平的影响及其性别差异[J].郑州大学学报：医学版，2018，53（2）：225-229.
[25] Reinisch JM，Sanders SA，Mortensen EL. In utero exposure to phenobarbital and intelligence deficits in adult men [J]. JAMA，1995，274（19）：1518-1525.
[26] Hasselblad V，Hedges LV. Meta-analysis of screening and diagnostic tests [J]. Psychol Bull，1995，117（1）：167-178.
[27] Borenstein M，Hedges LV，Higgins JPT，et al. Introduction to Meta-Analysis [M]. Wiley：2009，Chapter 7.
[28] JPA I，Tan YJ，Blum MR. Limitations and Misinterpretations of E-Values for Sensitivity Analyses of Observational Studies [J]. Ann Intern Med，2019，170（2）：108-111.
[29] von EE，Altman DG，Egger M，et al. The Strengthening the Reporting of Observational Studies in Epidemiology（STROBE）statement：guidelines for reporting observational studies [J]. Lancet，2007，370（9596）：1453-1457.