Brachial-Ankle Pulse Wave Velocity Predicts New-Onset Hypertension and the Modifying Effect of Blood Pressure in a Chinese Community-Based Population

Table 3

Multivariate regression for the effect of baPWV on SBP measured at revisit in participants without antihypertensive treatment in the follow-up survey.

Variables	N	Crude model		Adjusted model 1		Adjusted model 2
		β (95% CI)	value	β (95% CI)	value	β (95% CI)	value
baPWV continuous, per 1 m/s increase
	1771	1.97 (1.77, 2.17)	<0.001	0.94 (0.72, 1.17)	<0.001	0.98 (0.75, 1.20)	<0.001
baPWV categories
Q1 (<1311)	443	1 (reference)		1 (reference)		1 (reference)
Q2 (≥1311 − <1451)	438	4.75 (3.19, 6.31)	<0.001	1.64 (0.24, 3.04)	0.022	1.48 (0.08, 2.88)	0.039
Q3 (≥1451 − <1622)	447	8.78 (7.22, 10.33)	<0.001	3.06 (1.59, 4.53)	<0.001	2.95 (1.48, 4.42)	<0.001
Q4 (≥1622)	443	13.97 (12.41, 15.53)	<0.001	5.75 (4.1, 7.39)	<0.001	5.92 (4.26, 7.58)	<0.001
for trend		<0.001		<0.001		<0.001

baPWV, brachial-ankle pulse wave velocity; SBP, systolic blood pressure; OR, odds ratio; CI, confidence interval. Model 1, adjusted for age, sex, and baseline SBP; Model 2, adjusted for variables in model 1 and body mass index, current smoking, current drinking, baseline estimated glomerular filtration rate, diabetes mellitus, dyslipidemia, history of cardiovascular disease, hypoglycemic agents, and lipid-lowering agents.