MR Dictionary

Homogeneity Assumption

This is one of a set of additional (to the core) MR assumptions that are required for estimating a well-defined causal parameter. 

The assumption is that either the 1) the effect of the exposure on the outcome is homogeneous (i.e., is the same for everyone in the population) or 2) the effect of the exposure on the outcome does not depend on the genetic instrumental variable (IV). If this assumption holds, the IV estimate is consistent with the average causal effect (ACE) for the population to which inference is being made. Increasing biological knowledge of genetic variant functionality may provide evidence for some genetic IVs fulfilling this assumption. Weaker assumptions (e.g., the no effect modification and monotonicity assumptions) can alternatively be evoked. However, it is difficult in practice to know how important violation of these additional assumptions are in MR studies. Large genome-wide association study (GWAS) collaborations increasingly combine results from many studies (e.g., those across multiple European samples) and show consistency of association between genetic variants and traits across these studies at genome-wide significance (i.e., those variants used in most MR studies). This therefore suggests that homogeneity in effect estimates may exist for several MR studies in European populations. Non-parametric methods that provide bounds of causal effect estimates requiring only the core MR assumptions to be met may be applicable for some MR studies where violations of any of these additional assumptions is possible.

References

Other terms in 'Sources of bias and limitations in MR':