Table 3.
Association of heavy metals and eosinophil count using multivariable linear regression analysis
| Heavy metals | Multivariable (Eosinophil count) | |
|---|---|---|
| Unstandardized coefficient β (95% CI) | p | |
| Blood | ||
| Pb (log per 1 mg/L) | 0.732 (0.594, 0.869) | < 0.001 |
| Urine | ||
| Ni (log per 1 μg/L) | -0.006 (-0.068, 0.056) | 0.840 |
| Cr (log per 1 μg/L) | 0.005 (-0.215, 0.226) | 0.962 |
| Mn (log per 1 μg/L) | 0.013 (-0.056, 0.083) | 0.706 |
| As (log per 1 μg/L) | 0.133 (0.025, 0.240) | 0.015 |
| Cu (log per 1 μg/dL) | 0.181 (0.031, 0.330) | 0.018 |
| Cd (log per 1 μg/L) | 0.139 (0.052, 0.226) | 0.002 |
Values expressed as unstandardized coefficient β and 95% confidence interval (CI). Abbreviations are the same as in Table 1.
Covariates in the multivariable model included age, sex, diabetes, hypertension, body mass index, systolic and diastolic blood pressures, fasting glucose, log triglyceride, total cholesterol, HDL-cholesterol, LDL-cholesterol, hemoglobin, platelet, eGFR and uric acid.