The InSIDE assumption is a further assumption for MR-Egger and related MR methods (additional to core and additional MR assumptions).
The assumption states that the association between the genetic instrumental variable (IV) and exposure is not correlated with the (pleiotropic) path from the genetic IV to the outcome that is independent of the exposure of interest. If unbalanced pleiotropy is present and the InSIDE assumption is violated, 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 (unmeasured) confounder(s) of the exposure-outcome association, as the genetic IV-exposure and genetic IV-outcome associations would be, in part, via those confounders. The latter (whereby the genetic variants influence the outcome via confounders of the exposure-outcome association) is sometimes referred to as correlated pleiotropy. If the genetic variants influence the outcome in a way that is independent of the exposure, this is sometimes referred to as uncorrelated pleiotropy.
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; 48: 728-742.
Other terms in 'Sources of bias and limitations in MR':
- Assortative mating
- Canalization
- Collider
- Collider bias
- Conditional F-statistic for multiple exposures
- Confounding
- Exclusion restriction assumption
- F-statistic
- Harmonization (in two-sample MR)
- Homogeneity Assumption
- Horizontal Pleiotropy
- Independence assumption
- Intergenerational (or dynastic) effects
- Monotonicity assumption
- MR for testing critical or sensitive periods
- MR for testing developmental origins
- No effect modification assumption
- NO Measurement Error (NOME) assumption
- Non-linear MR
- Non-overlapping samples (in two-sample MR)
- Overfitting
- Pleiotropy
- Population stratification
- R-squared
- Regression dilution bias (attenuation by errors)
- Relevance assumption
- Reverse causality
- Same underlying population (in two-sample MR)
- Statistical power and efficiency
- Vertical pleiotropy
- Weak instrument bias
- Winner's curse