Performance of the risk model at different cut-off values in the validation population (n=1 826 293)
| Cut-off value | Patients at risk of lung cancer based on risk model | Patients not requiring a chest x-ray based on risk model | No. of true positives | No. of true negatives | Sensitivity (%)* | Specificity (%)† |
|---|---|---|---|---|---|---|
| −3 | 737390 | 1088903 | 1624 | 1088799 | 93.98 | 59.67 |
| −2.5 | 541074 | 1285219 | 1526 | 1285017 | 88.31 | 70.43 |
| −2 | 388040 | 1438253 | 1375 | 1437900 | 79.57 | 78.81 |
| −1.5 | 255788 | 1570505 | 1182 | 1569959 | 68.40 | 86.05 |
| −1.25 | 192433 | 1633860 | 1063 | 1633195 | 61.52 | 89.51 |
| −1 | 144523 | 1681770 | 917 | 1680959 | 53.07 | 92.13 |
| −0.5 | 72883 | 1752292 | 610 | 1752292 | 35.30 | 96.04 |
| 0 | 30994 | 1795299 | 367 | 1793938 | 21.24 | 98.32 |
| 0.5 | 11860 | 1814433 | 174 | 1812879 | 10.07 | 99.36 |
*Sensitivity=true positives / (true positives+false negatives).
†Specificity=true negatives / (true negatives+false positives).