Table 3.

Univariable and multivariable linear regression for the association between pulse pressure and body mass index

Variables	Pulse pressure (mmHg)
	Overalla		Normal blood pressureb		High blood pressurec
	$\mathit{\beta }$ coefficient (95% CI)	p-value	$\mathit{\beta }$ coefficient (95% CI)	p-value	$\mathit{\beta }$ coefficient (95% CI)	p-value
Body mass index, kg/m2
Unadjusted model	0.42 (0.40, 0.45)	< 0.001	0.27 (0.25, 0.30)	< 0.001	0.09 (0.04, 0.14)	< 0.001
Adjusted modelc	0.23 (0.21, 0.25)	< 0.001	0.28 (0.25, 0.30)	< 0.001	0.16 (0.12, 0.21)	< 0.001
Body mass index category, kg/m2
Unadjusted model
18.5–22.9	Ref		Ref		Ref
< 18.5	-1.45 (-2.26, -0.64)	< 0.001	-1.44 (-2.13, -0.75)	< 0.001	0.84 (-1.59, 3.27)	0.497
23.0–24.9	1.52 (1.27, 1.76)	< 0.001	1.02 (0.80, 1.24)	< 0.001	0.02 (-0.60, 0.65)	0.939
25.0.29.9	2.71 (2.49, 2.94)	< 0.001	1.78 (1.58, 1.99)	< 0.001	-0.27 (-0.82, 0.29)	0.345
≥ 30.0	4.88 (4.56, 5.19)	< 0.001	2.81 (2.49, 3.13)	< 0.001	1.00 (0.34, 1.65)	0.003
Adjusted modelc
18.5–22.9	Ref		Ref		Ref
< 18.5	-1.25 (-2.05, -0.46)	0.002	-1.55 (-2.28, -0.82)	< 0.001	1.08 (-1.47, 3.63)	0.407
23.0–24.9	0.83 (0.59, 1.07)	< 0.001	0.89 (0.66, 1.12)	< 0.001	0.41 (-0.23, 1.05)	0.211
25.0.29.9	1.38 (1.15, 1.60)	< 0.001	1.77 (1.55, 1.99)	< 0.001	0.43 (-0.14, 1.00)	0.141
≥ 30.0	2.57 (2.26, 2.88)	< 0.001	2.89 (2.55, 3.23)	< 0.001	1.96 (1.28, 2.65)	< 0.001

High blood pressure (systolic blood pressure $\ge$ 140 mmHg or diastolic blood pressure $\ge$ 90 mmHg) 

^aAdjusting for sex, age, regions, regular exercise, smoking status, alcohol use, high total cholesterol, high fasting plasma glucose, and high blood pressure

^bAdjusting for sex, age, regions, regular exercise, smoking status, alcohol use, high total cholesterol, and high fasting plasma glucose

^cP for interaction (high blood pressure and BMI) < 0.05