Table 4a. Prognostic Utility of 6MWD vs. CPX Indices in Predicting All-Cause Hospitalization/Mortality.

Model	Parameter	Chi Squire statistic	P value	Hazard Ratio* ( 95% confidence interval)	C-Index (95% confidence interval)	IDI****
Unadjusted Univariate predictors	6MWD*** (Z<1)	99	<.0001	0.75 (0.70,0.79)	0.58 (0.57, 0.60)
	Peak VO2	158	<.0001	0.69 (0.65,0.73)	0.61 (0.59, 0.62)
	VE/VCO2 Slope	85	<.0001	1.27 (1.21, 1.33)	0.56 (0.55, 0.58)
Adjusted**	6MWD*** (Z<1)	48	<.0001	0.78 (0.73, 0.84)	0.62 (0.60, 0.64)	0.019
	Peak VO2	80	<.0001	0.72 (0.67, 0.77)	0.63 (0.61, 0.65)	0.043
	VE/VCO2 Slope	19	<.0001	1.15 (1.08, 1.22)	0.61 (0.59, 0.62)	0.009

6MWD-6 minute walk distance; VO₂-oxygen consumption; VE/VCO₂-minute ventilation-carbon dioxide production; IDI-Integrated Discrimination Improvement

^*Hazard Ratio based on Z score

^**--All-Cause Hospitalization/Mortality Model adjusted for Gender, Region (US vs. Non-US), Mitral Regurgitation, ECG Ventricular Conduction Abnormality, Blood Urea Nitrogen (BUN), Left Ventricular Ejection Fraction (LVEF), Carvedilol Equivalent Dose, and Kansas City Cardiomyopathy Questionnaire Symptom Stability Score --All-Cause Mortality Model adjusted for Gender, BMI, Loop Diuretic Dose, Angina Class, ECG Ventricular Conduction Abnormality, LVEF, and Creatinine

^***--6MWD (normalized) is truncated at 1 standard deviation in the model of Hospitalization/Mortality because of its lack of relationship with this endpoint beyond that point. Truncation in this case implies that the Hazard Ratio for values of 6MWD>1 is set to 1.

--Other truncated covariates are carvedilol equivalent dose-truncated above 50 mg/day; BMI-body mass index-truncated above 25 kg/m²; Cr-truncated above 2.3 mg/dl.

^****IDI Model includes N=2013 patients with non-missing values for 6MW, Peak VO₂, and VE/VCO₂