Table 2. Multiple linear regression model of serum creatinine (n = 1074)

Dependent Variable: log serum creatinineaRegression CoefficientStandard ErrorAdjusted R2 = 0.36
Independent VariablesStandardized Regression CoefficientP Value
a Natural logarithm transformations used to satisfy the linearity and homoscedasticity assumptions of linear regression.
Log 24-h urine creatininea0.2540.0190.345<0.001
Log blood urea nitrogena0.3210.0260.307<0.001
Age, yr−0.0050.0006−0.216<0.001
African-American ethnicity0.1600.0210.188<0.001
Male gender0.0570.0180.0790.002
Constant0.7880.119<0.001