Validation cohort: multivariable, logistical regression results using in-hospital mortality as the dependent variable
| B | Significance | OR (95% CI) | |
| Chest radiograph consolidation | 0.358 | 0.089 | 1.43 (0.95–2.16) |
| Glasgow Coma Scale ≤14 | 0.658 | 0.002 | 1.93 (1.26–2.95) |
| Atrial fibrillation | 0.842 | <0.0001 | 2.32 (1.45–3.71) |
| pH <7.25 | 0.961 | <0.0001 | 2.61 (1.69–4.04) |
| Time to acidaemia >12 h | 1.289 | <0.0001 | 3.63 (2.33–5.64) |
| eMRCD 5a | 1.425 | <0.0001 | 4.16 (2.46–7.02) |
| eMRCD 5b | 1.960 | <0.0001 | 7.10 (4.05–12.46) |
Intercept −2.832, R2 0.285, Hosmer and Lemeshow 0.130. Leverage values, Cook's distances and Studentised residuals all within acceptable limits. eMRCD: extended Medical Research Council dyspnoea scale.