Table 3.
Multivariate linear regression models for risk factors of total urine mercury and blood methylmercury exposure, Madre de Dios, Peru, 2010
| Variable | Exp (β coefficient) | p value | Adjusted model R 2 |
|---|---|---|---|
| Total urine mercury exposure | 0.08 | ||
| Drinking water source | |||
| Public system | 1.00 | – | |
| Nonpublic systemb | 1.51 | 0.091 | |
| Heat gold–mercury amalgams | |||
| No | 1.00 | – | |
| Yes | 1.85 | 0.030 | |
| Blood methylmercury exposure | 0.20 | ||
| Highest education attained | |||
| None/some elementary | 1.00 | – | |
| Secondary | 1.31 | 0.168 | |
| University | 2.32 | 0.006 | |
| Drinking water source | |||
| Public system | 1.00 | – | |
| Nonpublic systemb | 1.51 | 0.035 | |
| Mix mercury with gold | |||
| No | 1.00 | – | |
| Yes | 0.59 | 0.011 | |
| Fish consumption | |||
| No | 1.00 | – | |
| Yes | 1.61 | 0.012 |
aOther drinking water sources constituted well (n = 30), spring (n = 10), river (n = 7), and water truck (n = 1)
bNonpublic drinking water systems constituted well (n = 30), spring (n = 10), river (n = 7), and water truck (n = 1)