Table 3.

Logistic regression analyses of the influence of BMI on hyperlipidemia prevalence.

Model	BMI	B c	S.E. d	Wald	p	OR (95% CI)
Univariate	<18.5	−0.792	0.078	104.196	<0.001	0.453 (0.389–0.527)
	18.5~					1.000
	24.0~	1.008	0.031	1025.644	<0.001	2.741 (2.577–2.916)
	30.0~	1.233	0.063	379.395	<0.001	3.430 (3.030–3.883)
					p value for trend p < 0.001
Model I a	<18.5	−0.695	0.081	73.860	<0.001	0.499 (0.426–0.585)
	18.5~					1.000
	24.0~	0.951	0.032	865.523	<0.001	2.587 (2.428–2.756)
	30.0~	1.285	0.065	393.025	<0.001	3.614 (3.183–4.104)
					p value for trend p < 0.001
Model II b	<18.5	−0.316	0.086	13.337	<0.001	0.729 (0.616–0.864)
	18.5~					1.000
	24.0~	0.501	0.042	142.217	<0.001	1.651 (1.520–1.793)
	30.0~	0.539	0.083	42.235	<0.001	1.714 (1.457–2.017)
					p value for trend p < 0.001

^a Adjusted age and gender; ^b Adjusted age, gender, region, district, ethnicity, education, marital status, main occupation, monthly family income per capita, smoking, drinking, exercise, central obesity, waist, hip; ^c B represents the logistics regression coefficient; ^d S.E. represents the standard error.