Table 3. Summary Statistics of Models Predicting In-Hospital Mortality of Patients With Acute Myocardial Infarction in Korea, Japan, and Taiwan .
Statistic | Modela | Korea | Japan | Taiwan |
AUC | 1 | 0.703 | 0.657 | 0.674 |
2 | 0.735 | 0.715 | 0.697 | |
3 | 0.732 | 0.713 | 0.698 | |
Difference in AUCs | 2−1 | 0.033 | 0.058 | 0.023 |
3−2 | −0.003 | −0.002 | 0.001 | |
ICC (%) | 2 | 3.54 | 4.99 | 2.22 |
3 | 2.22 | 4.54 | 1.80 | |
Median OR | 2 | 1.39 | 1.49 | 1.30 |
3 | 1.30 | 1.46 | 1.26 | |
PCV (%) | 3−2 | 38.2 | 9.4 | 19.3 |
Abbreviations: AUC, Area under the receiver operating characteristic curve; ICC, Intraclass correlation coefficient; OR, odds ratio; PCV, Proportional change in variance.
aModel 1, logistic regression with patient-level covariates; Model 2, multilevel logistic regression with patient-level covariates and random intercepts for hospitals; Model 3, multilevel logistic regression with patient-level and hospital-level covariates and random intercepts for hospitals.