Table 5.

Model performance statistics in training dataset.

Models	AUC	Sensitivity	Specificity	PPV	NPV	P value
	(95%CI)	(95%CI)	(95%CI)	(95%CI)	(95%CI)
Model	0.722	0.571	0.741	0.317	0.891	/
	0.696-0.748	0.526-0.617	0.722-0.790	0.286-0.349	0.877-0.906
Model+SBP	0.723	0.670	0.643	0.284	0.902	0.605
	0.698-0.749	0.627-0.714	0.622-0.663	0.257-0.311	0.887-0.917
Model+LDL-C	0.726	0.765	0.557	0.267	0.918	0.247
	0.700-0.752	0.726-0.804	0.536-0.578	0.243-0.291	0.903-0.933
Model+HbA1c	0.721	0.563	0.752	0.324	0.891	0.107
	0.695-0.747	0.517-0.608	0.734-0.770	0.291-0.356	0.876-0.905
Model+SBP+HbA1c+LDL-C	0.726	0.776	0.550	0.267	0.921	0.286
	0.700-0.752	0.738-0.814	0.529-0.571	0.243-0.291	0.906-0.936

SBP, systolic blood pressure; LDL-C, low-density lipoprotein cholesterol; HbA1c, glycated hemoglobin; AUC, area under the curve; PPV, positive predictive value; NPV, negative predictive value; CI, Confidence Interval. Symbol "/" means that "Model" is used as the reference object for AUC comparison.