Combines Wald ratios (or Ratio estimates) together in an inverse variance weighted meta-analysis, adjusting for heterogeneity.
All genetic IVs can be invalid due to pleiotropy as long as the pleiotropy is balanced (it has a zero mean and satisfies the Instrument Strength Independent of Direct Effect – InSIDE - assumption). This means that the association of genetic instrument with exposure of interest should not be associated with the path between instrument and outcome that does not go via the exposure of interest. Asymptotically equivalent to TSLS with uncorrelated variants.
- Bowden J, Del Greco M F, Minelli C, Davey Smith G, Sheehan N, Thompson J. A framework for the investigation of pleiotropy in two-sample summary data Mendelian randomization. Statistics in medicine 2017;36:1783-1802.