Table 3.
Mendelian randomization estimates of the causal effect of epilepsy on anthropometric traits using pan-UKB data.
| Method | Number of SNPs | MR analysis | Heterogeneity test | P of MR-Egger intercept | ||||
|---|---|---|---|---|---|---|---|---|
| β | SE | P | Cochran’s Q | I2 | P | |||
| Epilepsy (N = 64,629, Ncase = 27,559, Ncontrol = 42,436) on height (N = 4,19,596) | ||||||||
| IVW | 34 | 0.029 | 0.018 | .109 | 69.1 | 53.7 | .000 | .761 |
| MR-Egger | 34 | 0.048 | 0.067 | .475 | 69.3 | 52.4 | .000 | – |
| Weighted Median | 34 | 0.020 | 0.020 | .319 | – | – | – | – |
| Epilepsy (N = 64,629, Ncase = 27,559, Ncontrol = 42,436) on weight (N = 4,19,316) | ||||||||
| IVW | 35 | 0.004 | 0.014 | .779 | 34.9 | 5.5 | .377 | – |
| MR-Egger | 35 | 0.017 | 0.047 | .725 | 35.0 | 2.9 | .420 | .777 |
| Weighted Median | 35 | -0.002 | 0.020 | .904 | – | – | – | – |
| Epilepsy (N = 64,629, Ncase = 27,559, Ncontrol = 42,436) on BMI (N = 4,19,163) | ||||||||
| IVW | 37 | 0.013 | 0.018 | 0.472 | 55.2 | 36.6 | .016 | – |
| MR-Egger | 37 | 0.032 | 0.062 | 0.616 | 55.4 | 35.0 | .020 | .759 |
| Weighted Median | 37 | 0.010 | 0.022 | 0.646 | – | – | – | – |
| GGE (N = 49,388, Ncase = 6952, Ncontrol = 42,436) on height (N = 4,19,596) | ||||||||
| IVW | 13 | 0.002 | 0.014 | .902 | 21.6 | 49.1 | .028 | – |
| MR-Egger | 13 | −0.211 | 0.122 | .112 | 27.6 | 56.6 | .006 | .107 |
| Weighted Median | 13 | 0.026 | 0.016 | .089 | – | – | – | – |
| GGE (N = 49,388, Ncase = 6952, Ncontrol = 42,436) on weight (N = 4,19,316) | ||||||||
| IVW | 16 | 0.023 | 0.017 | .172 | 43.2 | 67.6 | .000 | – |
| MR-Egger | 16 | 0.013 | 0.081 | .877 | 43.3 | 65.4 | .000 | .903 |
| Weighted Median | 16 | 0.026 | 0.017 | .122 | – | – | – | – |
| GGE (N = 49,388, Ncase = 6952, Ncontrol = 42,436) on BMI (N = 4,19,163) | ||||||||
| IVW | 15 | 0.042 | 0.017 | .015 | 33.1 | 60.8 | .002 | – |
| MR-Egger | 15 | 0.072 | 0.086 | .418 | 33.5 | 58.2 | .002 | .731 |
| Weighted Median | 15 | 0.078 | 0.018 | .000 | – | – | – | – |
| FE (N = 57,375, Ncase = 14,939, Ncontrol = 42,436) on height (N = 4,19,596) | ||||||||
| IVW | 22 | 0.022 | 0.020 | .270 | 47.6 | 58.0 | .000 | – |
| MR-Egger | 22 | 0.068 | 0.077 | .390 | 48.5 | 56.7 | .001 | .544 |
| Weighted Median | 22 | 0.017 | 0.021 | .413 | – | – | – | – |
| FE (N = 57,375, Ncase = 14,939, Ncontrol = 42,436) on weight (N = 4,19,316) | ||||||||
| IVW | 25 | 0.040 | 0.015 | .008 | 27.7 | 17.1 | .226 | – |
| MR-Egger | 25 | 0.061 | 0.060 | .323 | 27.9 | 14.0 | .264 | .724 |
| Weighted Median | 25 | 0.030 | 0.020 | 0140 | – | – | – | – |
| FE (N = 57,375, Ncase = 14,939, Ncontrol = 42,436) on BMI (N = 4,19,163) | ||||||||
| IVW | 27 | 0.034 | 0.019 | .082 | 43.5 | 42.5 | .012 | – |
| MR-Egger | 27 | 0.067 | 0.079 | .405 | 43.8 | 40.7 | .016 | .667 |
| Weighted Median | 27 | 0.027 | 0.023 | .232 | – | – | – | – |
BMI = body mass index, FE = focal epilepsy, GGE = generalized genetic epilepsy, IVW = inverse-variance weighted, MR = Mendelian randomization, SE = standard error, SNP = single nucleotide polymorphism.