Table 2 Performance of the classification trees
| Â | Derivation cohort | Test cohort | Updated model |
|---|---|---|---|
Number of subjects | 220 | 135 | 355 |
Number of true positives | 21 | 16 | 38 |
Number of true negatives | 169 | 75 | 233 |
Number of false positives | 28 | 42 | 81 |
Number of false negatives | 2 | 2 | 3 |
Sensitivity | 91% (70, 98) | 89% (64, 98) | 93% (79, 98) |
Specificity | 86% (80, 90) | 64% (55, 73) | 74% (69, 79) |
Positive predictive value | 43% (29, 58) | 28% (17, 41) | 32% (24, 41) |
Negative predictive value | 99% (95, 100) | 97% (90, 100) | 99% (96, 100) |
+Likelihood ratio | 6.4 (4.5, 9.3) | 2.5 (1.8, 3.3) | 3.6 (2.9, 4.4) |
-Likelihood ratio | 0.1 (0.0, 0.4) | 0.2 (0.0, 0.6) | 0.1 (0.0, 0.3) |
Area under the curve | 0.885 | 0.759 | 0.883 |