Misconceptions on the use of MR-Egger regression and the evaluation of the InSIDE assumption
Author:
Open Access
- 12 September 2017
- journal article
- research article
- Published by Oxford University Press (OUP) in International Journal of Epidemiology
- Vol. 46 (6) , 2097-2099
Abstract
In their letter to this journal, Slob et al.1 attempt to derive the bias of the MR-Egger regression2 estimate for a Mendelian randomization (MR) analysis. They show that its bias can be larger than that of the inverse variance weighted (IVW) estimate when the instrument strength independent of direct effect (InSIDE) assumption is violated, and suggest a method for assessing the magnitude of InSIDE violation in any given data set. Slob et al. conclude by cautioning against placing undue reliance on the MR-Egger estimate in practice.Keywords
This publication has 11 references indexed in Scilit:
- A note on the use of Egger regression in Mendelian randomization studiesInternational Journal of Epidemiology, 2017
- Difficulties in Testing the Instrument Strength Independent of Direct Effect Assumption in Mendelian RandomizationJAMA Cardiology, 2017
- Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumptionInternational Journal of Epidemiology, 2017
- A framework for the investigation of pleiotropy in two‐sample summary data Mendelian randomizationStatistics in Medicine, 2017
- Why internal weights should be avoided (not only) in MR-Egger regressionInternational Journal of Epidemiology, 2016
- Response to Hartwig and DaviesInternational Journal of Epidemiology, 2016
- Assessing the suitability of summary data for two-sample Mendelian randomization analyses using MR-Egger regression: the role of the I2 statisticInternational Journal of Epidemiology, 2016
- Association of Lipid Fractions With Risks for Coronary Artery Disease and DiabetesJAMA Cardiology, 2016
- Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median EstimatorGenetic Epidemiology, 2016
- Multivariable Mendelian Randomization: The Use of Pleiotropic Genetic Variants to Estimate Causal EffectsAmerican Journal of Epidemiology, 2015