Table 3.

Hierarchical regression analyses for associations between alcohol consumption and BNP levels in HF patients and non-HF controls

Variables entered	Standardized β coefficient	t	P	R2 change
BNP baseline levels
Group (HF vs. controls)	.62 (.37)a	4.8(1.8)a	<.001 (.09)a	.38 (.05)a
Drinks per month	.22 (.27)a	1.7(2.0)a	.10 (.06)a	.05 (.09)a
Group by drinks per monthb	.18 (.24)a	1.2(1.8)a	.23 (.08)a	.03 (.05)a
BNP stress change
Group (HF vs. controls)	.32 (.32)a	2.6(1.7)a	.015 (.11)a	.10 (.04)a
Drinks per month	.61 (.56)a	4.9 (4.4)a	<.001 (<.001)a	.37 (.28)a
Group × Drinks per Monthb	.49 (.45)a	4.3 (4.6)a	<.001 (<.001)a	.19 (.18)a

Notes: BNP stress change = difference between post-stress level and baseline. BNP = biomarker B-type natriuretic peptide; HF = heart failure.

^aAfter controlling for age, mean arterial pressure, and meters walked in the 6-minute walk test;

^bregression results for the interaction of Group × Drinks per Month after controlling for group and drinks per month.