Table 2.
Discrimination and calibration performance statistics for risk score models in estimating the 10-year risk of CVD death
| FPR | SCORE-Low | SCORE-High | ||||
|---|---|---|---|---|---|---|
| Quintiles | Actual | Predicted | Actual | Predicted | Actual | Predicted |
| 1 | 1* | 0.1 | 0 | 0.0 | 0 | 0.0 |
| 2 | 0 | 0.3 | 0 | 0.1 | 0 | 0.2 |
| 3 | 2 | 0.8 | 2 | 0.6 | 2 | 1.0 |
| 4 | 4 | 3.6 | 5 | 3.5 | 5 | 5.6 |
| 5 | 21 | 29.7 | 21 | 28.3 | 21 | 42.4 |
| Total | 27 | 34.5 | 28 | 32.5 | 28 | 49.1 |
| Hosmer-Lemeshow X2 | 4.74 | 6.09 | 12.06 | |||
| Hosmer-Lemeshow p-value | 0.1918 | 0.1074 | 0.0072 | |||
| Calibrated model? | Yes | Yes | No | |||
| ROC AUC (95% CI) | 0.858 (0.786 – 0.929) | 0.877 (0.827 – 0.927) | 0.877 (0.827 – 0.927) | |||
*An outlier was excluded from quintile 1 (1 death) in the calculation of the Hosmer-Lemeshow test, owing to its undue influence on the test. Including this outlier would result in the Hosmer-Lemeshow X2 = 14.35, p-value = 0.0025, and the risk score model being assessed as not calibrated. This outlier did not represent a woman with elevated risk factors associated with increased CVD risk, on further examination.
Abbreviations: FPR, Framingham 10-year predicted risk for CVD death; SCORE-Low, SCORE-Low 10-year predicted risk for CVD death; SCORE-High, SCORE-High 10-year predicted risk for CVD death; ROC, receiver operating characteristic curve; AUC, area under the curve; CI, confidence interval.