Table 2.
Comparison of health status classification by each model.
| Predictive model | Health Status: Healthy | ||||
|---|---|---|---|---|---|
| Predictor | AUC | StdError | Lower 95% | Upper 95% | Prob > ChiSq |
| Clinical Risk Model | 0.7350 | 0.0886 | 0.5322 | 0.8712 | |
| Biomarker Model | 0.8272 | 0.0621 | 0.6714 | 0.9181 | |
| Difference | −0.092 | 0.0635 | −0.217 | 0.0323 | 0.1465 |
| Health Status: BE | |||||
| Predictor | AUC | StdError | Lower 95% | Upper 95% | Prob > ChiSq |
| Clinical Risk Model | 0.7435 | 0.0726 | 0.5788 | 0.8595 | |
| Biomarker Model | 0.8405 | 0.0579 | 0.6934 | 0.9247 | |
| Difference | −0.097 | 0.0690 | −0.232 | 0.0382 | 0.1598 |
| Health Status: EAC | |||||
| Predictor | AUC | StdError | Lower 95% | Upper 95% | Prob > ChiSq |
| Clinical Risk Model | 0.8378 | 0.0624 | 0.6774 | 0.9270 | |
| Biomarker Model | 0.9311 | 0.0411 | 0.7936 | 0.9794 | |
| Difference | −0.093 | 0.0432 | −0.178 | −0.009 | 0.0309 * |
AUC, area under the curve; * p < 0.05.