. Author manuscript; available in PMC: 2011 Nov 16.

Published in final edited form as: J Am Coll Cardiol. 2010 Nov 16;56(21):1720–1727. doi: 10.1016/j.jacc.2010.04.064

Table 3.

Regression Outcomes Predicting changes in CTX to 10nM response to exercise; BDI and NE as a potential moderators

Model		Coefficients

	R² delta	t	p	β
Risk factors	.024
Age		.487	.628	.001
Gender		.181	.857	.011
BMI		1.361	.178	.005
6-mw		.302	.764	.0001
HF status		−3.014	.004	−.985
Predictors	.146
NE		3.125	.003	.171
BDI		3.119	.003	.091
Interactions	.153
HF × NE		2.205	.031	.118
HF × BDI		3.446	.001	.105

The final regression model and coefficients are shown. The regression models were as follows: Cardiovascular risk factors (Step 1: age, gender, BMI, 6 min walk, HF status), Predictors of CTX-I (Step 2: resting NE and BDI), and Interaction Variables (Step 3: HF status × NE and HF status × BDI). BMI = body mass index, HF = heart failure, NE = norepinephrine, BDI = Beck Depression Inventory.