TABLE 4.
Statistics for model fit for the prediction of CV composite outcomes
| Predictive models | Depended variable: CV composite outcomes | |||||
|---|---|---|---|---|---|---|
| AUC | NRI | IDI | ||||
| M (95%CI) | p value | M (95%CI) | p value | M (95%CI) | p value | |
| Model 1 (Based Model: LVEF<40%) | 0.83 (0.80–0.85) | - | Reference | - | Reference | |
| Model 2 (NT-proBNP>2,250 pg/ml) vs. Model 1 | 0.86 (0.83–0.89) | 0.054 | 0.11 | 0.44 | 0.10 | 0.42 |
| Model 3 (Irisin<6.5 ng/ml) vs. Model 1 | 0.85 (0.79–0.90) | 0.055 | 0.14 | 0.18 | 0.13 | 0.12 |
| Model 4 (NT-proBNP>2,250 pg/ml + Irisin<6.5 ng/ml) vs. Model 1 | 0.92 (0.89–0.95) | 0.044 | 0.25 | 0.04 | 0.16 | 0.04 |
Abbreviations: M, median; CI, confidence interval; AUC, area under curve; ACR, adverse cardiac remodeling; IDI, integrated discrimination indices; NRI, net-reclassification improvement.