Skip to main content

Table 4 Odds ratios (OR) for the association between per standard deviation decrease in gSOS and increased odds of incident fracture

From: Improved prediction of fracture risk leveraging a genome-wide polygenic risk score

  OR (95% CI)* p value OR (95% CI), adjusted for RFs§ p value, adjusted for RFs
UK Biobank test dataset (N = 80,014)
 Major osteoporotic fracture 1.46 (1.38–1.55) 1.1 × 10−37 1.42 (1.34–1.51) 1.4 × 10−32
 Hip fracture 1.37 (1.19–1.57) 6.3 × 10−6 1.34 (1.16–1.53) 3.7 × 10−5
MrOS US (N = 4663)
 Major osteoporotic fracture 1.38 (1.26–1.51) 2.2 × 10−12 1.31 (1.17–1.45) 1.1 × 10−6
 Hip fracture 1.37 (1.21–1.56) 5.5 × 10−7 1.33 (1.15–1.55) 1.4 × 10−4
MrOS Sweden (N = 1880)
 Major osteoporotic fracture 1.45 (1.28–1.65) 5.3 × 10−9 1.38 (1.17–1.63) 1.3 × 10−4
 Hip fracture 1.31 (1.08–1.58) 5.0 × 10−3 1.35 (1.05–1.72) 1.8 × 10−2
SOF (N = 3615)
 Major osteoporotic fracture 1.35 (1.24–1.47) 1.0 × 10−11 1.34 (1.21–1.48) 7.3 × 10−9
 Hip fracture 1.31 (1.19–1.43) 1.4 × 10−8 1.32 (1.19–1.47) 4.1 × 10−7
CKB (N = 25,034)
 Major osteoporotic fracture 1.26 (1.13–1.41) 5.3 × 10−5 1.22 (1.09–1.36) 6.8 × 10−4
 Hip fracture 1.25 (0.98–1.60) 7.1 × 10−2 1.21 (0.94–1.54) 0.14
  1. *Derived from logistic regression models adjusted for age and sex
  2. §Derived from logistic regression models adjusted for age, sex and other available risk factors (RFs) listed in Table 1