Table 4.

Measures of model fit for prediction of heart failure with heart failure risk variables (HFRV) only or combined with estimated glomerular filtration rate (eGFR), N-terminal pro-brain natriuretic peptide (NT-proBNP) or both among adults with and without Chronic Kidney Disease in the Multi-Ethnic Study of Atherosclerosis (n=5504).

	HFRV	HFRV + eGFR	HFRV + NT-proBNP	HFRV + eGFR + NT-proBNP
Chronic Kidney Disease (eGFR < 60 ml/min/1.73 m2) (n=325)
C-statistic (95% CI)	0.71 (0.64, 0.78)	0.73 (0.64, 0.78)	0.78 (0.71, 0.85)*	0.79 (0.71, 0.86)*
Log likelihood	−204.2	−204.0	−193.4	−192.1
Akaike Information Criterion	422.4	424.0	402.8	402.3
Bayes Information Criterion	448.8	454.2	433.0	436.3
Net Reclassification Index	Referent	6.53%	23.1%	28.8%*
Risk Stratification Capacity	64.3%	63.1%	77.8%	77.2%
No Chronic Kidney Disease (eGFR ≥ 60 ml/min/1.73 m2)(n=5,179)
C-statistic (95% CI)	0.77 (0.74, 0.80)	0.77 (0.74, 0.80)	0.83 (0.80, 0.85)*	0.83 (0.80, 0.85)*
Log likelihood	−1420.4	−1419.6	−1357.6	−1357.5
Akaike Information Criterion	2854.9	2855.2	2731.2	2733.1
Bayes Information Criterion	2900.7	2907.6	2783.6	2792.0
Net Reclassification Index	Referent	−0.46%	10.2%*	15.9%*
Risk Stratification Capacity	96.1%	96.2%	97.7%	96.8%

Abbreviations: eGFR= estimated glomerular filtration rate calculated with serum creatinine and cystatin C. [28] Heart failure risk variables (HFRV) = age, sex, body mass index, smoking, systolic blood pressure, heart rate, and diabetes.

^*P < 0.05 compared with model with heart failure risk variables (HFRV) only