. 2015 Jun 4;10(6):e0128870. doi: 10.1371/journal.pone.0128870

Table 3. Multivariable linear regression analysis of 10 Years Risk of CHD (Ln transformed) with serum Na⁺ and Glucose as predictor variables (ARIC Study).

			Predictive capability of independent variables						Analysis of variance
Dependent variable:Ln(10 Years Risk of CHD)			Na⁺, mmol/l			Glucose, mmol/l			Analysis of variance
Dependent variable:Ln(10 Years Risk of CHD)		Intercept	Coefficient	t	P	Coefficient	t	P	F	P
All participants (N = 12779)		-1.86	0.017	4.2	<0.001	0.165	34.5	<0.001	598	<0.001
^No BP Medications (N = 10336)*		-2.50	0.020	4.2	<0.001	0.176	29.4	<0.001	432	<0.001
BP Medications (N = 3319)		0.29	0.009	1.6	n.s.	0.082	16.4	<0.001	138	<0.001
^Glucose < 6.5 mmol/l (N = 9360)*		-3.67	0.012	2.4	0.02	0.61	25.0	<0.001	319	<0.001
^*White	Men (N = 3788)	-0.25	0.012	2.9	0.004	0.14	20.1	<0.001	202	<0.001
^*White	Women (N = 4453)	-7.84	0.051	8.6	<0.001	0.23	24.8	<0.001	328	<0.001
^*Black	Men (N = 896)	0.23	0.007	0.9	n.s.	0.09	12.8	<0.001	86	<0.001
^*Black	Women (N = 1199)	-4.14	0.027	2.7	0.006	0.15	17.0	<0.001	145	<0.001

Serum Na⁺ and glucose are significant predictors of the 10 Years Risk of CHD when all study participants are included in analysis and in the following subsets: people who do not take blood pressure medications, people with blood glucose concentration lower than 6.5 mmol/l, white men, white women and black men.

* Participants who took blood pressure medications were excluded from the analysis

Table 3. Multivariable linear regression analysis of 10 Years Risk of CHD (Ln transformed) with serum Na+ and Glucose as predictor variables (ARIC Study).

Table 3. Multivariable linear regression analysis of 10 Years Risk of CHD (Ln transformed) with serum Na⁺ and Glucose as predictor variables (ARIC Study).