Table 2.
Diagnostic accuracy of the score.
| Score cut-off point | % | SN (%) | SP (%) | PPV (%) | NPV (%) |
|---|---|---|---|---|---|
| Training set | |||||
| ≥140 | 72.2 | 99.6 | 36.1 | 32.2 | 99.7 |
| ≥180 | 61.4 | 98.6 | 49.9 | 37.4 | 99.2 |
| ≥220 | 51.3 | 95.4 | 62.0 | 43.3 | 97.8 |
| ≥260 | 39.9 | 90.2 | 75.4 | 52.7 | 96.2 |
| ≥300 | 29.0 | 79.6 | 86.4 | 63.9 | 93.3 |
| ≥340 | 20.0 | 65.3 | 93.7 | 75.9 | 89.9 |
| ≥380 | 12.4 | 44.2 | 97.2 | 82.9 | 85.2 |
| Validation set | |||||
| ≥140 | 71.7 | 97.6 | 36.3 | 32.2 | 98.0 |
| ≥180 | 62.7 | 95.5 | 47.4 | 36.0 | 97.1 |
| ≥220 | 51.7 | 91.0 | 60.5 | 41.6 | 95.6 |
| ≥260 | 42.1 | 83.0 | 70.6 | 46.6 | 93.1 |
| ≥300 | 32.2 | 70.9 | 79.8 | 52.0 | 89.9 |
| ≥340 | 21.2 | 54.7 | 89.2 | 61.0 | 86.4 |
| ≥380 | 13.4 | 36.7 | 93.8 | 64.6 | 82.7 |
%, number of participants with score≥ cut-off point; SN, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value.