Table 9.
Two-sample MR analysis of hospital admission rate per year in UK Biobank participants of White British ancestry. Rates and 95% confidence intervals (95% CI) are given. Effect estimates are provided per BMI SD (SDBMI=4.6) and per WHR SD (SDWHR=0.07), for MR-Egger (random effects), IVW (random effects, exact weights), weighted median and weighted mode analyses of BMI (64 SNPs), WHR (34 SNPs) and WHRadjBMI (45 SNPs). SNPs with an LD R2 < 0.001 have been retained.
| MR-Egger |
||||||
|---|---|---|---|---|---|---|
| IVW (random effects, exact weights) | Intercept | Slope | Penalized weighted median | Weighted mode | ||
| BMI (SD) | Ratea (95% CI) | 1.098 | 1.004 | 0.973 | 1.095 | 1.095 |
| (1.009–1.194) | (0.997–1.011) | (0.759–1.247) | (0.930–1.290) | (0.915–1.312) | ||
| WHR (SD) | Ratea (95% CI) | 1.199 | 1.000 | 1.185 | 1.236 | 1.204 |
| (1.054–1.364) | (0.987–1.014) | (0.676–2.079) | (0.985–1.550) | (0.867–1.167) | ||
| WHRadjBMI (SD) | Ratea (95% CI) | 1.151 | 1.008 | 0.870 | 1.124 | 1.076 |
| (1.028–1.287) | (0.995–1.021) | (0.541–1.400) | (0.941–1.343) | (0.819–1.414) | ||
BMI = body mass index, CI = confidence interval, IVW = inverse variance weighted, MR = Mendelian randomization, SD = standard deviation, WHR = waist hip ratio, WHRadjBMI = waist hip ratio adjusted for BMI.
a) Adjusted for sex, age and the first 40 genetic principal components. Estimates (with corresponding 95% CIs) represent the fold increase in yearly hospital admission rate per BMI SD (4.6 kg/m2) and per WHR SD (0.07). Estimates per BMI and WHR unit are given in Table S8.