TABLE 3.
Association of anthropometric trait with the risk of breast cancer using sex-combined and female-specific instruments.
| Exposure | Sex-combined IVs | Female-specific IVs | u (P) | ||||||
| k0/k1 | PVE (%) | Power (%) | OR (95% CI, P) | k0/k1 | PVE (%) | Power (%) | OR (95% CI, P) | ||
| BMI | 97/92 | 1.51 | 99.9 | 0.845 (0.756∼0.945, 0.003) | 38/36 | 1.16 | 100.0 | 0.763 (0.661∼0.882, 2.52E-04) | 4.795 (1.63E-06) |
| WHR | 39/38 | 0.95 | 2.8 | 1.002 (0.841∼1.194, 0.984) | 34/34 | 1.63 | 6.7 | 0.985 (0.858∼1.131, 0.833) | 0.610 (0.542) |
| WC | 70/66 | 1.42 | 98.0 | 0.866 (0.767∼0.978, 0.020) | 25/25 | 1.04 | 96.1 | 0.858 (0.732∼1.004, 0.057) | 0.299 (0.765) |
| HIP | 89/86 | 2.07 | 6.1 | 0.988 (0.885∼1.102, 0.823) | 41/39 | 1.57 | 93.6 | 0.890 (0.787∼1.006, 0.063) | 9.669 (1.25E-20) |
IV, instrumental variable; BMI, body mass index; WHR, waist-to-hip ratio; WC, waist circumference; HIP, hip circumference; MR, Mendelian randomization; PVE, phenotypic variance explained by IVs that were used in our final MR analysis; power, statistical power; OR, odds ratio; CI, confidence internal.
Note that k0 is the number of candidate instruments, while k1 is the final number of instruments employed in the analysis.