Multiple logistic regression analysis was used to identify risk
factors for urinary cadmium excretion Z1 μg/g creatinine and to
determine associations between urinary cadmium and prevalence
of renal impairments, after adjusting for other variables. The regression
analysis was separately performed for the prevalence of
each renal effect marker above the normal/average level and that
above the clinically relevant level. Adjusted odds ratios and their
95% confidence intervals were calculated to indicate the strength
of association. Multiple linear regression analysis was used to
determine associations between urinary cadmium and renal biomarkers
and blood pressure, after adjusting for other variables.
We also used similar bivariate and multivariate analyses to
determine associations between blood cadmium, an indicator of
recent cadmium exposure, and renal dysfunctions and blood
pressure.