Table 3. Sex-specific AUC by different anthropometric indices for the prediction of hypertension.

		Men (n 1441)			Women (n 1812)
Model	Independent variable	AUC1	95% CI	P value2	AUC	95% CI	P value
Crude model3	BMI	0.65	0.61, 0.68	-	0.63	0.59, 0.67	-
	WC	0.63	0.59, 0.66	0.184	0.62	0.58, 0.66	0.662
	WHR	0.57	0.54, 0.61	<0.001	0.60	0.57, 0.64	0.349
	WSR	0.63	0.60, 0.66	0.285	0.63	0.59, 0.67	0.938
	BMI + WC	0.65	0.61, 0.68	0.669	0.64	0.60, 0.67	0.874
	BMI + WHR	0.65	0.61, 0.68	0.879	0.64	0.61, 0.68	0.637
	BMI + WSR	0.65	0.61, 0.68	0.490	0.64	0.60, 0.68	0.723
Age-adjusted4	BMI	0.68	0.65, 0.71	-	0.72	0.69, 0.76	-
	WC	0.67	0.64, 0.70	0.250	0.71	0.67, 0.74	0.528
	WHR	0.63	0.59, 0.66	<0.001	0.70	0.66, 0.74	0.357
	WSR	0.66	0.63, 0.70	0.097	0.71	0.67, 0.74	0.506
	BMI + WC	0.68	0.65, 0.71	0.413	0.72	0.69, 0.76	0.997
	BMI + WHR	0.68	0.65, 0.71	0.573	0.72	0.69, 0.76	0.996
	BMI + WSR	0.68	0.65, 0.71	0.284	0.72	0.70, 0.75	0.987

AUC, area under the receiver operating characteristic curve; BMI, body mass index; WC, waist circumference; WHR, waist:hip ratio; WSR, waist:stature ratio.

¹ AUC values were estimated by logistic regression models; range from 0.5 (no prediction) to 1.0 (perfect prediction).

² Models compared with a model with BMI.

³ Crude models: include independent variables in the list.

⁴ Age-adjusted models: include independent variables and age.