Table 3.
Means of judged and expected PPV and NPV (and 95% CI) in full and restricted data sets, with corresponding P values of paired sample t-tests
| PVs | Full data set | Restricted data set | ||||
| Judged | Expected | P value | Judged | Expected | P value | |
| PPV | ||||||
| All | 90.5 (89.3 to 91.6) | 86.5 (84.7 to 88.3) | 0.000 | 93.7 (92.7 to 94.6) | 93.8 (92.8 to 94.9) | 0.760 |
| Prevalence ≤20% | 87.0 (83.9 to 90.1) | 69.8 (64.6 to 75.0) | 0.000 | 92.1 (89.1 to 95.2) | 85.2 (81.4 to 89.0) | 0.002 |
| NPV | ||||||
| All | 81.1 (79.0 to 83.2) | 89.4 (88.0 to 90.9) | 0.000 | 88.5 (87.0 to 90.1) | 92.1 (90.7 to 93.5) | 0.000 |
| 1-prevalence ≤20% | 71.8 (65.1 to 78.5) | 79.0 (73.8 to 84.1) | 0.054 | 81.5 (75.5 to 87.4) | 82.4 (76.6 to 88.1) | 0.778 |
NPV, negative predicted value; PPV, positive predictive value; PVs, predictive values.