Table 2.
MR estimates from each method of assessing the causal effect of dried fruit intake on the risk of osteoarthritis.
| MR method | nSNP | Beta | SE | OR | 95% CI | Association P value | Cochran Q statistic | I 2 | Heterogeneity P value |
|---|---|---|---|---|---|---|---|---|---|
| MR Egger | 41 | −0.020 | 0.043 | 0.980 | 0.900–1.066 | .640 | 57.81 | 0.325 | .027 |
| Weighted median | 41 | −0.026 | 0.012 | 0.974 | 0.951–0.997 | .025 | – | – | –− |
| Inverse variance weighted | 41 | −0.020 | 0.009 | 0.981 | 0.963–0.999 | .039 | 57.81 | 0.309 | .034 |
I2 = (Q − df)/Q.
Beta = beta coefficient, nSNP = number of single nucleotide polymorphism, SE = standard error.