Correcting Misinterpretations of the E-Value

Editorials15 January 2019Correcting Misinterpretations of the E-ValueTyler J. VanderWeele, PhD, Maya B. Mathur, PhD, and Peng Ding, PhDTyler J. VanderWeele, PhDHarvard T.H. Chan School of Public Health, Boston, Massachusetts (T.J.V., M.B.M.), Maya B. Mathur, PhDHarvard T.H. Chan School of Public Health, Boston, Massachusetts (T.J.V., M.B.M.), and Peng Ding, PhDUniversity of California, Berkeley, Berkeley, California (P.D.)Author, Article, and Disclosure Informationhttps://doi.org/10.7326/M18-3112 SectionsAboutFull TextPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinkedInRedditEmail We thank Ioannidis and colleagues (1) for raising several important issues concerning the potential misuse of the E-value (2). These issues are not new, and we have addressed them previously and provided guidance as to how to avoid misuse (2–4). Let us begin with where we think Ioannidis and colleagues are correct (1).First, biases other than confounding, such as measurement error, selection bias, and publication bias, can also undermine estimates of causal effects. We agree, we have noted this previously (2–4), and we have now developed similar techniques to address other biases (5, 6).Second, the E-value is monotonically ...References1. Ioannidis JPA, Tan YJ, Blum MR. Limitations and misinterpretations of E-values for sensitivity analyses of observational studies. Ann Intern Med. 2019;170:108-11. doi:10.7326/M18-2159 LinkGoogle Scholar2. VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med. 2017;167:268-74. [PMID: 28693043]. doi:10.7326/M16-2607 LinkGoogle Scholar3. Ding P, VanderWeele TJ. Sensitivity analysis without assumptions. Epidemiology. 2016;27:368-77. [PMID: 26841057] doi:10.1097/EDE.0000000000000457 CrossrefMedlineGoogle Scholar4. VanderWeele TJ, Ding P, Mathur M. Technical considerations in the use of the E-value. J Causal Inference. 2019;7. Preprint available at: https://biostats.bepress.com/harvardbiostat/paper215. Google Scholar5. Smith LH, VanderWeele TJ. Bounding bias due to selection. Epidemiology. 2019;30. Preprint available at: https://arxiv.org/abs/1810.13402. Google Scholar6. VanderWeele TJ, Mathur MB, Chen Y. Outcome-wide longitudinal analyses for causal inference: a new template for empirical studies. Stat Sci. 2019;34. https://arxiv.org/abs/1810.10164. Google Scholar7. Mathur MB, Ding P, Riddell CA, VanderWeele TJ. Web site and R package for computing E-values. Epidemiology. 2018;29:e45-7. [PMID: 29912013] doi:10.1097/EDE.0000000000000864 CrossrefMedlineGoogle Scholar8. Benjamin DJ, Berger JO, Johannesson M, Nosek BA, Wagenmakers EJ, Berk R, et al. Redefine statistical significance. Nat Hum Behav. 2018;2:6-10. CrossrefMedlineGoogle Scholar Author, Article, and Disclosure InformationAffiliations: Harvard T.H. Chan School of Public Health, Boston, Massachusetts (T.J.V., M.B.M.)University of California, Berkeley, Berkeley, California (P.D.)Disclosures: Authors have disclosed no conflicts of interest. Forms can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M18-3112.Corresponding Author: Tyler J. VanderWeele, PhD, Department of Epidemiology, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston MA 02115; e-mail, [email protected]harvard.edu.Current Author Addresses: Dr. VanderWeele: Department of Epidemiology, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston MA 02115.Dr. Mathur: Harvard University, 655 Huntington Avenue, Building 2, Boston MA 02115.Dr. Ding: University of California, Berkeley, 425 Evans Hall, Berkeley, CA 94720.This article was published at Annals.org on 1 January 2019. PreviousarticleNextarticle Advertisement FiguresReferencesRelatedDetailsSee AlsoLimitations and Misinterpretations of E-Values for Sensitivity Analyses of Observational Studies John P.A. Ioannidis , Yuan Jin Tan , and Manuel R. Blum Metrics Cited byAre Greenland, Ioannidis and Poole opposed to the Cornfield conditions? A defence of the E-valueE-values for effect heterogeneity and approximations for causal interactionHow to report E-values for meta-analyses: Recommended improvements and additions to the new GRADE approachAssociation of Alcohol Use Diagnostic Codes in Pregnancy and Offspring Conotruncal and Endocardial Cushion Heart DefectsAn approach to quantifying the potential importance of residual confounding in systematic reviews of observational studies: A GRADE concept paperEffects of hearing impairment and hearing aid use on the incidence of cognitive impairment among community-dwelling older adults: evidence from the Taiwan Longitudinal Study on Aging (TLSA)The Importance of Making Assumptions in Bias AnalysisBias Analysis Gone BadPharmacoepidemiology for nephrologists (part 2): potential biases and how to overcome themUncontrolled ConfoundersOn the use of the E-value for sensitivity analysis in epidemiologic studiesLong-term exposure to NO2 and O3 and all-cause and respiratory mortality: A systematic review and meta-analysisControversy and debate on credibility ceilings. Paper 2: Using credibility ceilings to explore skepticism about observational evidenceE Values and Incidence Density SamplingSensitivity analysis for publication bias in meta‐analysesImpact of neuraxial anaesthesia on short-term outcomes after elective open abdominal aortic aneurysm repair. Comment on Br J Anaesth 2020; 124: 544–52Commentary: Quantifying the unknown unknownsCommentary: Developing best-practice guidelines for the reporting of E-valuesCommentary: An argument against E-values for assessing the plausibility that an association could be explained away by residual confoundingCommentary: Cynical epidemiologyCommentary: Continuing the E-value's post-publication peer reviewRejoinder: The Future of Outcome-Wide StudiesAttrition from specialised rehabilitation associated with an elevated mortality risk: results from a vital status tracing study in Swiss spinal cord injured patientsDeath and long-term disability after gun injury: a cohort analysisEvaluation of Bedside Delivery of Medications Before Discharge: Effect on 30-Day ReadmissionBeta-blocker therapy and risk of vascular dementia: A population-based prospective studyComparison of bias adjustment methods in meta-analysis suggests that quality effects modeling may have less limitations than other approachesThe Role of Genetic Variant rs13266634 in SLC30A8/ZnT8 in Postoperative Hyperglycemia After Major Abdominal SurgeryTWO AUTHORS REPLYHospital Readmission Rates in Medicare Advantage and Traditional Medicare A Retrospective Population-Based AnalysisOrestis A. Panagiotou, MD, PhD, Amit Kumar, PT, MPH, PhD, Roee Gutman, PhD, Laura M. Keohane, PhD, Maricruz Rivera-Hernandez, PhD, Momotazur Rahman, PhD, Pedro L. Gozalo, PhD, Vincent Mor, PhD, and Amal N. Trivedi, MD, MPH 15 January 2019Volume 170, Issue 2Page: 131-132KeywordsConflicts of interestDisclosureEpidemiologyPrevention, policy, and public healthRisk ratio ePublished: 1 January 2019 Issue Published: 15 January 2019 Copyright & PermissionsCopyright © 2019 by American College of Physicians. All Rights Reserved.PDF downloadLoading ...