Table 2.
CHARGE-AF | CHA2DS2-VASc | Age alone | |
All (n=111 475; 5264 AF cases) | |||
C-statistic (95% CI) | 0.736 (0.727 to 0.744) | 0.669 (0.661 to 0.677) | 0.716 (0.708 to 0.724) |
Nam-D’Agostino χ2 (p value) | 901.8 (p<0.001) | – | – |
Calibration slope (95% CI) | 0.69 (0.67 to 0.71) | – | – |
Age ≥65 years (n=60 528; 4356 AF cases) | |||
C-statistic (95% CI) | 0.646 (0.637 to 0.655) | 0.581 (0.572 to 0.590) | 0.615 (0.606 to 0.624) |
Nam-D’Agostino χ2 (p value) | 907.2 (p<0.001) | – | – |
Calibration slope (95% CI) | 0.58 (0.54 to 0.61) | – | – |
Age <65 years (n=50 947; 908 AF cases) | |||
C-statistic (95% CI) | 0.706 (0.686 to 0.725) | 0.543 (0.525 to 0.561) | 0.652 (0.633 to 0.671) |
Nam-D’Agostino χ2 (p value) | 38.8 (p<0.001) | – | – |
Calibration slope (95% CI) | 0.93 (0.84 to 1.01) | – | – |
Men (n=52 926; 2692 AF cases) | |||
C-statistic (95% CI) | 0.718 (0.707 to 0.729) | 0.672 (0.661 to 0.683) | 0.702 (0.691 to 0.713) |
Nam-D’Agostino χ2 (p value) | 542.7 (p<0.001) | – | – |
Calibration slope (95% CI) | 0.65 (0.62 to 0.69) | – | – |
Women (n=58 549; 2572 AF cases) | |||
C-statistic (95% CI) | 0.751 (0.740 to 0.763) | 0.706 (0.695 to 0.717) | 0.736 (0.724 to 0.750) |
Nam-D’Agostino χ2 (p value) | 381.6 (p<0.001) | – | – |
Calibration slope (95% CI) | 0.72 (0.69 to 0.76) | – | – |
CHA2DS2-VASc≥2 (n=88 538; 4866 AF cases) | |||
C-statistic (95% CI) | 0.711 (0.702 to 0.719) | 0.636 (0.628 to 0.644) | 0.688 (0.680 to 0.697) |
Nam-D’Agostino χ2 (p value) | 866.0 (p<0.001) | – | – |
Calibration slope (95% CI) | 0.66 (0.63 to 0.68) | – | – |
CHA2DS2-VASc<2 (n=22 937; 398 AF cases) | |||
C-statistic (95% CI) | 0.723 (0.694 to 0.752) | 0.521 (0.501 to 0.541) | 0.680 (0.649 to 0.707) |
Nam-D’Agostino χ2 (p value) | 20.4 (p=0.02) | – | – |
Calibration slope (95% CI) | 0.99 (0.86 to 1.12) | – | – |
CHA2DS2-VASc, congestive heart failure, hypertension, age, diabetes and previous stroke or transient ischaemic attack, vascular disease and female sex category; CHARGE-AF, Cohorts for Heart and Aging Research in Genomic Epidemiology model for atrial fibrillation.