Table 2.
Performance of the models for predicting hospital mortality; N = 36,632 patients
| Models | AUC (CI 95%)* | Brier score (CI 95%) | Ĉ-statistic (CI)* | Ĉ-statistic p-value |
|---|---|---|---|---|
| APACHE IV – model | 0.851 (0.851–0.851) | 0.019 (0.019–0.019) | 27.0 (24.1–36.4) | < 0.0001 |
| APACHE II – model | 0.830 (0.830–0.830) | 0.020 (0.19–0.20) | 16.3 (12.6–24.8) | 0.0308 |
| SOFA - model | 0.809 (0.808–0.810) | 0.020 (0.019–0.20) | 43.7 (31.5–61.1) | < 0.0001 |
| SAPS-II - model | 0.850 (0.850–0.850) | 0.019 (0.019–0.019) | 19.4 (11.0–33.5) | 0.009 |
| MPM24-II - model | 0.801 (0.801–0.801) | 0.020 (0.20–0.020) | 30.3 (28.7–37.6) | < 0.0001 |
*AUC: area under the receiver operating characteristic curve; Ĉ-statistic: Hosmer and Lemeshow goodness-of-fit Ĉ-statistic; CI: 95% confidence interval