Table 3.
The predictive performance of prediction model, SOFA and SAPSII
| Models | Sets | Accuracy (95% CI) | AUC (95% CI) | Sensitivity (95% CI) | Specificity (95% CI) | PPV (95% CI) | NPV (95% CI) |
|---|---|---|---|---|---|---|---|
| Established model | Testing set | 0.705 (0.692–0.718) | 0.812 (0.798–0.826) | 0.798 (0.773–0.823) | 0.682 (0.668–0.697) | 0.385 (0.364–0.407) | 0.931 (0.922–0.940) |
| Training set | 0.732 (0.724–0.740)*** | 0.811 (0.802–0.820) | 0.762 (0.745–0.779)* | 0.724 (0.714–0.733)*** | 0.422 (0.407–0.437)** | 0.920 (0.914–0.926) * | |
| SOFA | Testing set | 0.723 (0.711–0.736) | 0.539 (0.518–0.559)*** | 0.232 (0.206–0.258)*** | 0.846 (0.835–0.857)*** | 0.273 (0.243–0.304)*** | 0.815 (0.803–0.827)*** |
| SAPSII | Testing set | 0.609 (0.595–0.623)*** | 0.609 (0.589–0.629)*** | 0.555 (0.524–0.586)*** | 0.623 (0.607–0.638)*** | 0.269 (0.249–0.288)*** | 0.848 (0.835–0.861)*** |
AUC The area under of curve; CI Confidence interval; SOFA Sequential organ failure assessment; SAPS II Simplified acute physiology score II; PPV Positive predictive value; NPV Negative predictive value
Taking established model-testing set as reference, the predictive performance of established model training set, SOFA-testing set and SAPSII-testing set was compared;
*represents P < 0.05; **represents P < 0.01; ***represents P < 0.001