FIG. 3.
(A) Two-sample MR in CKDGen of hypothyroidism on eGFRcrea. (B) Two-sample MR in CKDGen of hypothyroidism on CKD. The causal MR estimates of the various hypothyroidism exposures are multiplied with 0.693 (i.e., loge2) to reflect a doubling (twofold increase) in the odds of the binary exposure. IVW-FE, inverse-variance weighted fixed effects; IVW-RE, inverse-variance weighted random effects; SIMEX, simulation extrapolation algorithm for the MR-Egger model when the “No Measurement Error” assumption is violated, that is, I2GX below 0.9.