Table 6. Comparison of batches with best predictors of specific type of infection alone.
| Adjusted OR* | CI (95%) | p-value | AUC | Wald-p** | |
| Batches 1: WBC+CRP+Copeptin | |||||
| Any infection | |||||
| WBC | 3.70 | 2.26–6.08 | <0.001 | 0.86 | <0.001 |
| CRP | 1.66 | 1.24–2.21 | <0.001 | ||
| Copeptin | 1.53 | 1.07–2.18 | 0.019 | ||
| Pneumonia | |||||
| WBC | 4.12 | 1.63–10.39 | 0.003 | 0.92 | <0.001 |
| CRP | 1.92 | 1.30–2.84 | 0.001 | ||
| Copeptin | 2.06 | 1.19–3.57 | 0.010 | ||
| Urinary Tract infection | |||||
| WBC | 3.11 | 1.55–6.24 | 0.001 | 0.85 | 0.058 |
| CRP | 1.62 | 1.01–2.61 | 0.047 | ||
| Copeptin | 1.26 | 0.73–2.19 | 0.411 | ||
| Other Infections | |||||
| WBC | 6.84 | 3.00–15.60 | <0.001 | 0.90 | 0.43 |
| CRP | 1.29 | 0.87–1.93 | 0.208 | ||
| Copeptin | 1.18 | 0.69–2.03 | 0.550 | ||
| Batch 2: WBC+CRP+PCT | |||||
| Any infection | |||||
| WBC | 3.67 | 2.42–5.58 | <0.001 | 0.84 | <0.001 |
| CRP | 1.56 | 1.16–2.11 | 0.003 | ||
| PCT | 1.25 | 0.99–1.57 | 0.064 | ||
| Pneumonia | |||||
| WBC | 4.25 | 2.27–7.97 | <0.001 | 0.90 | <0.001 |
| CRP | 1.87 | 1.39–2.52 | <0.001 | ||
| PCT | 1.36 | 1.00–1.85 | 0.052 | ||
| Urinary Tract Infection | |||||
| WBC | 2.89 | 1.69–4.95 | <0.001 | 0.82 | 0.014 |
| CRP | 1.59 | 0.90–2.81 | 0.114 | ||
| PCT | 1.19 | 0.82–1.70 | 0.359 | ||
| Other Infections | |||||
| WBC | 5.74 | 3.33–9.87 | <0.001 | 0.89 | 0.25 |
| CRP | 1.53 | 0.92–2.55 | 0.103 | ||
| PCT | 0.85 | 0.59–1.24 | 0.407 | ||
WBC: white blood cells; CRP: C-reactive protein; PCT: procalcitonin. AUC: Area under the curve to predict infection using the combined model of all predictors.g.
*
adjusted for all predictors in the respective model.
**
Wald-p: refers to the comparison of the combined model with the model of the strongest predictor, alone which always was WBC.