Table 4: Longitudinal multivariate analysis of body lead burden and other predictors of progressive change in the estimated glomerular filtration rate (D-GFR), using generalized estimating equations, during the 24-month longitudinal study period ().
Estimate (interactive effect)*
Age (each increment of 1 yr)
Gender (female versus male)
Smoking (no versus yes)
Body-mass index (each increment of 1 kg/m2)
History of cardiovascular diseases (no versus yes)
MAP (mmHg) (each increment of 1 mmHg)
Cholesterol (mg/dL) (each increment of 1 mg/dL)
Triglycerides (mg/dL) (each increment of 1 mg/dL)
HbA1c (%) (each increment of 1%)
Serum creatinine (mg/dL) (each increment of 1 mg/dL)
Body lead burden (μg) (each increment of 1 g)
Daily protein intake (g/kg) (each increment of 1 g/kg)
Daily protein excretion (g) (each increment of 1 g)
The interactive effect of variables was calculated by a generalized estimating equation. Negative values for the interactive effect indicate a decline in the glomerular filtration rate, and positive values indicate an increase. Cardiovascular diseases included ischemic heart disease, congestive heart failure, stroke, and diabetic foot. MAP: mean arterial pressure.