Table 1.
Sensitivity analyses of serum iron biomarkers in 2014 datasets with epigenetic aging accelerations
| Exposure | Outcome | Weighted median | MR-egger regression | Heterogeneitya | MR-PRESSO outlier detectb | Pleiotropyc | |||
|---|---|---|---|---|---|---|---|---|---|
| Beta (95% CI) | P value | Beta (95% CI) | P value | Beta (95% CI) | P Value | ||||
| Ferritin | Grim | 0.76 (0.23, 1.29) | 4.82E− 03 | 0.76 (− 0.33, 1.85) | 3.04E−01 | I2 = 39.6%; Cochrane Q = 5; P = 0.174 | No significant outliers | Intercept = -0.031; P = 0.603 | |
| Ferritin | Hannum | 0.82 (0.33, 1.31) | 1.04E−03 | 0.83 (− 0.25, 1.91) | 3.74E−01 | I2 = 0%; Cochrane Q = 2; P = 0.380 | No significant outliers | Intercept = -0.018; P = 0.812 | |
| Ferritin | IE | 0.75 (0.27, 1.24) | 2.29E−03 | 0.98 (0.14, 1.82) | 1.51E−01 | I2 = 1.6%; Cochrane Q = 3; P = 0.384 | No significant outliers | Intercept = -0.028; P = 0.554 | |
| Ferritin | Pheno | 0.80 (0.18, 1.41) | 1.18E−02 | 0.77 (− 0.43, 1.97) | 3.34E−01 | I2 = 6.7%; Cochrane Q = 3; P = 0.360 | No significant outliers | Intercept = 0.014; P = 0.820 | |
| Iron | Grim | 0.30 (0.06, 0.55) | 1.33E−02 | 0.49 (0.09, 0.90) | 9.77E−02 | I2 = 14.5%; Cochrane Q = 5; P = 0.322 | No significant outliers | Intercept = -0.045; P = 0.309 | |
| Iron | Hannum | 0.31 (0.08, 0.53) | 8.00E−03 | 0.54 (0.06, 1.01) | 1.15E−01 | I2 = 32.1%; Cochrane Q = 6; P = 0.208 | No significant outliers | Intercept = -0.043; P = 0.396 | |
| Iron | IE | 0.27 (0.02, 0.51) | 3.16E−02 | 0.65 (0.25, 1.06) | 5.09E−02 | I2 = 0%; Cochrane Q = 3; P = 0.480 | No significant outliers | Intercept = -0.064; P = 0.180 | |
| Iron | Pheno | 0.55 (0.24, 0.85) | 4.81E−04 | 0.60 (− 0.40, 1.6) | 3.24E−01 | I2 = 66.5%; Cochrane Q = 12; P = 0.018 | No significant outliers | Intercept = -0.027; P = 0.787 | |
| Transferrin saturation | Grim | 0.27 (0.10, 0.44) | 1.62E−03 | 0.28 (0.03, 0.52) | 7.80E−02 | I2 = 0%; Cochrane Q = 3; P = 0.856 | No significant outliers | Intercept = -0.011; P = 0.678 | |
| Transferrin saturation | Hannum | 0.28 (0.12, 0.44) | 7.46E−04 | 0.32 (0.01, 0.63) | 9.98E−02 | I2 = 30.3%; Cochrane Q = 9; P = 0.196 | No significant outliers | Intercept = -0.013; P = 0.716 | |
| Transferrin saturation | IE | 0.28 (0.11, 0.44) | 9.04E−04 | 0.41 (0.16, 0.66) | 2.36E−02 | I2 = 0%; Cochrane Q = 2; P = 0.870 | No significant outliers | Intercept = -0.035; P = 0.238 | |
| Transferrin saturation | Pheno | 0.36 (0.14, 0.58) | 1.49E−03 | 0.33 (− 0.19, 0.85) | 2.67E−01 | I2 = 56.8%; Cochrane Q = 14; P = 0.031 | No significant outliers | Intercept = 0.010; P = 0.869 | |
| Transferrin | Grim | − 0.05 (− 0.21, 0.12) | 5.68E−01 | − 0.18 (− 0.35, 0.01) | 8.77E−02 | I2 = 4%; Cochrane Q = 9; P = 0.403 | No significant outliers | Intercept = 0.009; P = 0.633 | |
| Transferrin | Hannum | − 0.12 (− 0.29, 0.05) | 1.79E−01 | − 0.15 (− 0.34, 0.05) | 1.82E−01 | I2 = 29.5%; Cochrane Q = 14; P = 0.164 | No significant outliers | Intercept = -0.012; P = 0.530 | |
| Transferrin | IE | − 0.03 (− 0.19, 0.13) | 6.71E−01 | − 0.17 (− 0.34, 0.02) | 9.29E−02 | I2 = 0.7%; Cochrane Q = 9; P = 0.432 | No significant outliers | Intercept = 0.011; P = 0.547 | |
| Transferrin | Pheno | − 0.17 (− 0.36, 0.03) | 9.24E−02 | − 0.19 (− 0.44, 0.07) | 1.83E−01 | I2 = 27.4%; Cochrane Q = 14; P = 0.184 | No significant outliers | Intercept = -0.005; P = 0.830 | |
aHeterogeneity in the random effect IVW methods was reported
bMR-PRESSO (NbDistribution = 10,000, P < 0.05)
cMR-Egger was used to detect Pleiotropy. There is no pleiotropy was observed among all analyses (P > 0.05)
CI, confidence interval, MR-PRESSO, Mendelian Randomization Pleiotropy RESidual Sum and Outlier, IE Intrinsic epigenetic