Table 4. Comparison predictive Values in the development and the validation sample.
| Risk of not returning to work | ||||||||||||||||
| Sensitivity (95% CI) | Specificity (95% CI) | Positive Predictive Value (95% CI) | Negative Predictive Value (95% CI) | |||||||||||||
| Threshold | Development | Validation | Development | Validation | Development | Validation | Development | Validation | ||||||||
| > = 0.1 | 100 | (100 to 100) | 100 | (100 to 100) | 2.6 | (1.7 to 3.4) | 1.71 | (0.82 to 2.6) | 51.3 | (48.7 to 53.8) | 50.4 | (47.0 to 53.8) | 100 | (100 to 100) | 100 | (100 to 100) |
| > = 0.2 | 99.5 | (97.8 to 99.1) | 99.3 | (98.7 to 99.9) | 17.5 | (15.5 to 19.5) | 11.2 | (9.7 to 14.2) | 55.0 | (52.4 to 57.6) | 52.9 | (49.5 to 56.4) | 91.8 | (90.4 to 93.2) | 94.2 | (92.6 to 95.8) |
| > = 0.3 | 93.2 | (91.9 to 94.5) | 94.9 | (93.6 to 96.4) | 33.4 | (31.0 to 35.9) | 26.6 | (23.6 to 29.6) | 58.9 | (56.4 to 61.5) | 56.3 | (52.9 to 59.7) | 82.8 | (80.8 to 84.7) | 83.9 | (81.3 to 86.4) |
| > = 0.4 | 85.0 | (83.1 to 96.9) | 85.1 | (82.7 to 87.5) | 50.9 | (48.3 to 53.2) | 45.1 | (41.7 to 48.5) | 64.0 | (62.5 to 66.4) | 60.7 | (57.4 to 64.1) | 76.8 | (74.6 to 79.0) | 75.2 | (72.3 to 78.2) |
| > = 0.5 | 69.6 | (67.7 to 72.1) | 72.4 | (69.3 to 75.4) | 65.2 | (62.7 to 67.6) | 61.2 | (57.9 to 64.6) | 67.2 | (64.8 to 69.6) | 65.1 | (61.8 to 68.3) | 67.8 | (65.3 to 70.2) | 69.0 | (65.8 to 68.3) |
| > = 0.6 | 51.7 | (49.1 to 54.3) | 49.4 | (46.0 to 52.8) | 78.2 | (76.1 to 80.4) | 78.5 | (75.7 to 81.4) | 70.9 | (68.5 to 73.2) | 69.7 | (66.5 to 72.8) | 61.3 | (58.7 to 63.8) | 60.9 | (57.5 to 64.2) |
| > = 0.7 | 30.4 | (28.0 to 32.8) | 26.4 | (23.4 to 29.4) | 89.5 | (87.9 to 91.1) | 92.4 | (90.6 to 94.3) | 74.7 | (72.5 to 77.0) | 77.7 | (74.9 to 80.6) | 55.7 | (53.1 to 58.2) | 55.7 | (52.3 to 59.1) |
| > = 0.8 | 11.4 | (9.7 to 13.0) | 7.6 | (5.8 to 9.4) | 96.4 | (95.5 to 97.4) | 98.3 | (97.4 to 99.2) | 76.6 | (74.4 to 78.8) | 81.6 | (78.9 to 84.2) | 51.5 | (48.9 to 54.1) | 51.6 | (48.2 to 53.0) |
| > = 0.9 | 1.0 | (0.5 to 1.5) | – | 99.7 | (99.4 to 100) | – | 77.8 | (75.6 to 79.9) | – | 49.6 | (46.9 to 52.1) | – | ||||
Compares diagnostic properties in the development sample with the validation sample. Threshold = Chosen cut-off for the dichotomizing in test negatives (i.e. return to work, below thresholds; non return to work, equal or above threshold).