paperKB
coga / coga-kb
Help
Sign in

Chunk #31 — EXAMPLE: CAUSAL EFFECT OF HDL‐C ON CHD RISK — Varying the orientation of the genetic variants

Source
Extending the MR-Egger method for multivariable Mendelian randomization to correct for both measured and unmeasured pleiotropy.
Embedded
yes

Text

All of the causal odds ratios for HDL‐C from the multivariable MR‐Egger models indicated that HDL‐C is not causally associated with CHD risk. Significant adverse effects of LDL‐C on CHD risk were reported from the multivariable IVW (1.45, 95% CI: 1.34‐1.58) and multivariable MR‐Egger (1.52, 95% CI: 1.37‐1.69) methods. Orientating the variants with respect to the risk increasing alleles for HDL‐C and triglycerides had little impact on the causal estimates for LDL‐C from multivariable MR‐Egger. The multivariable IVW method suggested a significant adverse effect of triglycerides on CHD risk with a causal odds ratio of 1.19 (95% CI: 1.07, 1.33), this estimate was attenuated to the null in the multivariable MR‐Egger method (1.09, 95% CI: 0.96, 1.23). The causal odds ratios for triglycerides remained significant, however, when the variants were orientated with respect to HDL‐C and LDL‐C in the multivariable MR‐Egger models.