Rule | No of features | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) |
---|---|---|---|---|---|
PPV=positive predictive value; NPV=negative predictive value. | |||||
Derivation set (n=718) | |||||
CURB | ⩾0 | 100.0 | 0.0 | 9.6 | NA |
⩾1 | 95.7 | 33.0 | 13.2 | 98.6 | |
⩾2 | 75.4 | 68.9 | 20.5 | 96.3 | |
⩾3 | 42.0 | 90.3 | 31.5 | 93.6 | |
⩾4 | 1.4 | 99.1 | 14.3 | 90.4 | |
CURB-65 | ⩾0 | 100.0 | 0.0 | 9.6 | NA |
⩾1 | 98.6 | 21.0 | 11.7 | 99.3 | |
⩾2 | 92.8 | 49.2 | 16.2 | 98.5 | |
⩾3 | 68.1 | 74.9 | 22.4 | 95.7 | |
⩾4 | 39.1 | 93.1 | 37.5 | 93.5 | |
⩾5 | 1.4 | 99.1 | 14.3 | 90.4 | |
CRB-65 | ⩾0 | 100.0 | 0.0 | 9.6 | NA |
⩾1 | 97.1 | 25.4 | 12.2 | 98.8 | |
⩾2 | 76.8 | 64.3 | 18.6 | 96.3 | |
⩾3 | 43.5 | 89.8 | 31.3 | 93.7 | |
⩾4 | 2.9 | 98.6 | 18.2 | 90.5 | |
Validation set (n=214) | |||||
CURB | ⩾0 | 100.0 | 0.0 | 9.3 | NA |
⩾1 | 100.0 | 28.4 | 12.6 | 100.0 | |
⩾2 | 75.0 | 70.1 | 20.5 | 96.5 | |
⩾3 | 35.0 | 89.7 | 25.9 | 93.0 | |
⩾4 | 5.0 | 98.5 | 25.0 | 91.0 | |
CURB-65 | ⩾0 | 100.0 | 0.0 | 9.3 | NA |
⩾1 | 100.0 | 18.6 | 11.2 | 100.0 | |
⩾2 | 100.0 | 46.4 | 16.1 | 100.0 | |
⩾3 | 75.0 | 74.7 | 23.4 | 96.7 | |
⩾4 | 30.0 | 91.8 | 27.3 | 92.7 | |
⩾5 | 5.0 | 99.0 | 33.3 | 91.0 | |
CRB-65 | ⩾0 | 100.0 | 0.0 | 9.3 | NA |
⩾1 | 100.0 | 23.2 | 11.8 | 100.0 | |
⩾2 | 80.0 | 61.3 | 17.6 | 96.7 | |
⩾3 | 45.0 | 89.7 | 31.0 | 94.1 | |
⩾4 | 5.0 | 99.0 | 33.3 | 91.0 |