Table 6. AUROC for tested variables as predictors for death in our group.
Some parameters were not tested because the difference between groups was not significant. We considered a substantial AUROC of at least 0.600. The ROC curve shows the trade-off between sensitivity and specificity of a given model. AUROC is the measure of the classifier's ability to distinguish positive and negative classes. An AUROC of 0.5 indicates random results, while a value of 1 means that the model is perfect. We considered a substantial AUROC of at least 0.600.
Systolic BP: systolic blood pressure; EGFR: estimated glomerular filtration rate; INR: international normalised ratio; CRP: C-reactive protein; CI: confidence interval
| Variable | AUROC | Standard Error | Significance | 95% CI |
| Age | 0.799 | 0.031 | 0.000 | 0.737-0.860 |
| Systolic BP | 0.399 | - | - | - |
| Leucocyte count | 0.606 | 0.041 | 0.011 | 0.526-0.687 |
| Haemoglobin | 0.321 | - | - | - |
| Creatinine | 0.616 | 0.041 | 0.005 | 0.536-0.697 |
| EGFR | 0.343 | - | - | - |
| Albumin | 0.231 | - | - | - |
| Potassium | 0.355 | - | - | - |
| INR | 0.760 | 0.042 | 0.000 | 0.677-0.842 |
| CRP | 0.663 | 0.045 | 0.014 | 0.577-0.749 |
| Fibrinogen | 0.616 | 0.044 | 0.012 | 0.529-0.729 |