The Instrument Strength Independent of Direct Effect (InSIDE) assumption a further assumption for MR-Egger and related MR methods (additional to core and additional IV assumptions outlined in Chapter 2).
The assumption states that the association between the genetic instrument and exposure is not correlated with path from the genetic instrument to the outcome that is independent of the exposure of interest. If unbalanced pleiotropy is present and the InSIDE assumption is violated, then this is likely to result in a biased MR-Egger effect. Violation of the InSIDE assumption can occur if several genetic variants influence the outcome via the same pleiotropic path or if several variants are related to the same risk factor- outcome (unmeasured) confounder(s), as then the genetic instrument-exposure association is in part via that (those) confounders and it is also related to the outcome via the same confounder(s).
References
- Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. European journal of epidemiology 2017;32:377-389.
- Bowden J, Del Greco M F, Minelli C et al. Improving the accuracy of two-sample summary-data Mendelian randomization: moving beyond the NOME assumption. International Journal of Epidemiology 2018:dyy258-dyy258.
Other terms in 'Sources of bias and limitations in MR':
- Assortative mating
- Canalization
- Collider
- Collider bias
- Confounding
- Dynastic effects
- Exclusion restriction assumption
- Harmonization failure (in two-sample MR)
- Homogeneity Assumption
- Horizontal Pleiotropy
- Independence assumption
- Monotonicity assumption
- MR for testing critical or sensitive periods
- MR for testing developmental origins
- No effect modification assumption (Additional IV assumption)
- Non-linear effects
- Non-overlapping samples (in two-sample MR)
- Overfitting
- Pleiotropy
- Population stratification
- Regression dilution bias (attenuation by errors)
- Relevance assumption
- Reverse causality
- Same underlying population (in two-sample MR)
- Statistical power/efficiency
- Vertical Pleiotropy
- Weak instrument bias
- Winner's curse