Table 2 Sensitivity analysis for causal association of HbA1c with lipids (mmol/L) by unweighted GRS, IVW, Weighted median, MR-Egger and MR-PRESSO methods.

From: Mendelian Randomization Analysis Support Causal Associations of HbA1c with Circulating Triglyceride, Total and Low-density Lipoprotein Cholesterol in a Chinese Population

	Methods	Expanded GRS				Conservative GRS
	Methods	β	SE	z- or t-statistic	P value	β	SE	z- or t-statistic	P value
Log-TG, (mmol/L)	Weighted GRS	0.19	0.07	2.62	0.009**	0.22	0.10	2.13	0.03*
	Unweighted GRS	0.22	0.07	2.98	0.003**	0.35	0.11	3.05	0.002**
	IVW	0.21	0.06	3.61	0.0003**	0.25	0.10	2.24	0.03*
	Weighted Median	0.13	0.11	1.18	0.23	0.22	0.14	1.57	0.12
	Egger Estimate	0.20	0.15	1.30	0.22	0.47	0.28	1.68	0.10
	Egger Intercept	0.0005	NA.	NA.	0.92	−0.0005	NA.	NA.	0.87
	MR-PRESSO MR Analysis	0.21	0.06	3.46	0.003**	0.24	0.11	2.18	0.07
	Global Test	NA.	NA.	NA.	0.25	NA.	NA.	NA.	0.043*.
TC, (mmol/L)	Weighted GRS	0.42	0.17	2.46	0.01*	0.60	0.24	2.44	0.01*
	Unweighted GRS	0.51	0.17	3.02	0.003**	0.67	0.25	2.66	0.008**
	IVW	0.48	0.15	3.17	0.002**	0.30	0.12	2.47	0.01*
	Weighted Median	0.38	0.22	1.73	0.08	0.30	0.13	2.40	0.02*
	Egger Estimate	0.31	0.37	0.81	0.42	0.32	0.27	1.19	0.24
	Egger Intercept	0.005	NA.	NA.	0.65	0.002	NA.	NA.	0.26
	MR-PRESSO MR Analysis	0.48	0.15	3.24	0.005**	0.30	0.12	2.53	0.04*
	Global Test	NA.	NA.	NA.	0.20	NA.	NA.	NA.	0.29
LDL-C, (mmol/L)	Weighted GRS	0.33	0.13	2.55	0.01**	0.51	0.19	2.70	0.007**
	Unweighted GRS	0.38	0.13	3.03	0.002**	0.53	0.19	2.74	0.006**
	IVW	0.38	0.10	3.62	0.0003**	0.25	0.10	2.41	0.02*
	Weighted Median	0.28	0.13	2.15	0.03*	0.24	0.10	2.45	0.02*
	Egger Estimate	0.33	0.25	1.32	0.21	0.33	0.28	1.57	0.10
	Egger Intercept	0.002	NA.	NA.	0.84	0.009	NA.	NA.	0.47
	MR-PRESSO MR Analysis	0.38	0.10	3.81	0.002**	0.25	0.10	2.53	0.04*
	Global Test	NA.	NA.	NA.	0.27	NA.	NA.	NA.	0.34
HDL-C, (mmol/L)	Weighted GRS	−0.005	0.04	0.10	0.92	−0.02	0.06	0.28	0.79
	Unweighted GRS	−0.02	0.04	0.57	0.58	−0.06	0.06	0.93	0.36
	IVW	−0.02	0.04	0.52	0.61	−0.009	0.04	0.23	0.82
	Weighted Median	0.05	0.09	0.56	0.58	−0.02	0.03	0.67	0.50
	Egger Estimate	−0.007	0.10	0.07	0.95	0.07	0.06	1.11	0.27
	Egger Intercept	−0.0005	NA.	NA.	0.87	−0.005	NA.	NA.	0.16
	MR-PRESSO MR Analysis	−0.02	0.04	−0.48	0.64	−0.009	0.04	−0.23	0.83
	Global Test	NA.	NA.	NA.	0.48	NA.	NA.	NA.	0.78

IVW, inverse-variance weighted method, Egger, MR-Egger method. MR-PRESSO, Mendelian randomization pleiotropy residual sum and outlier. For MR-PRESSO MR Analysis, outcomes of the outlier-corrected analysis were presented if outliers were detected, otherwise it would be the raw MR analysis. NA., not applicable.
The t statistics were calculated for MR-PRESSO while z statistics for the other analysis. P value were calculated from multivariable-adjusted linear regression. *P < 0.05; **P < 0.01. All data were calculated under the full model with age, sex, BMI, systolic and diastolic blood pressure, smoking and drinking status, physical activity status, and diabetes adjusted.

Back to article page

Table 2 Sensitivity analysis for causal association of HbA1c with lipids (mmol/L) by unweighted GRS, IVW, Weighted median, MR-Egger and MR-PRESSO methods.

Search

Quick links