TABLE 3.
Statistical power in Mendelian randomization analyses of body shape and the risk of Alzheimer’s disease.
| Each AT | R2 | Statistical power | Minimal/Maximal detectable OR |
| BMI | |||
| Positive/negative correlated with AD risk | 2.14% | 81% | 1.19/0.83 |
| Previous association (Razay et al., 2006) | 2.14% | >99% | 1.80∗ |
| WHR | |||
| Positive/negative correlated with AD risk | 0.77% | 80% | 1.33/0.73 |
| Previous association (Razay et al., 2006) | 0.77% | >99% | 2.00∗ |
| WHRadjBMI | |||
| Positive/negative correlated with AD risk | 1.13% | 82% | 1.27/0.77 |
| WC | |||
| Positive/negative correlated with AD risk | 1.33% | 80% | 1.24/0.79 |
Based on our sample size (case/control: 17,008/37,154, total: 54,162), minimal/maximal detectable odds ratio (OR) was calculated at a significance level of 0.05. R2 refers to the proportion of variance explained by IVs in each anthropometric trait. The statistical power of previous association in each AT means the power to detect previous association in this MR analysis. AD, Alzheimer’s disease; AT, anthropometric traits; BMI, body mass index; WC, waist circumference; WHR, waist-to-hip ratio; WHRadjBMI, waist-to-hip ratio adjusted for body mass index. ∗Association provided by a previous observational study.