Table 5. Results of multiple linear regression analysis showing the relative contributions and extents of the relation of CETP-TaqIB genotype, selected lifestyle factors, and relative parameters to serum HDL-C levels.

Dependent Variable	Independent Variable	Men			Women
		β	PR2	P	β	PR2	P
HDL-C, mg/dl	Age (years)	-0.1 (0.1)	0.5	0.14	0.0 (0.1)	0.0	0.75
	Log BMI	-20.5 (6.0)	3.0	<0.01	-25.0 (6.3)	4.1	<0.01
	CETP-TaqIB Genotype
	B1B1	reference			reference
	B1B2	4.1 (1.6)	1.7	<0.01	2.3 (1.7)	0.6	0.15
	B2B2	6.9 (2.3)	2.2	<0.01	4.3 (2.3)	1.0	0.06
	Smoking
	Nonsmokers	reference			reference
	Smokers	-4.1 (1.5)	2.0	<0.01	2.8 (3.8)	0.1	0.45
	Alcohol Drinking
	Nondrinkers	reference			reference
	Drinkers	9.6 (1.9)	6.9	<0.01	4.4 (1.6)	2.0	<0.01
	Tertiles of AII (METs)
	Tertile 1	reference			reference
	Tertile 2	1.6 (1.8)	0.2	0.39	1.6 (1.8)	0.2	0.37
	Tertile 3	-0.1 (1.7)	0.0	0.93	1.8 (1.9)	0.2	0.36
	Log TG	-11.3 (1.5)	14.0	<0.01	-12.0 (1.7)	13.4	<0.01

β: regression coefficient (SE); PR²: partial R² expressed as percentages; TG: fasting triglyceride; AII: active intensity index (METs). CETP-TaqIB genotype and tertiles of AII are used as dummy variables to evaluate the effects of B1B2, B2B2, tertile 2, and tertile 3 to HDL-C, respectively.

Status of smoking or alcohol drinking is coded as 0, 1.