Cancer mortality in a Chinese population surrounding a multi-metal sulphide mine in Guangdong province: an ecologic study

Table 5 Estimation of the age-adjusted mortality risk for the selected specific cause associated with blood heavy metals levels for both sexes using simple, multiple linear and ridge regression in the nine villages evaluated in this study

^a using simple linear regression.
^b Estimate of the regression coefficient of log-transformed for blood cadmium, lead and zinc levels by using simple linear regression.
^c P < 0.05.
^d Estimate of the regression coefficient of the relationships between log-transformed for blood cadmium and lead levels and log-transformed for age-adjusted mortality rates from all-cancer and stomach cancer using multiple linear regression and estimate of the regression coefficient of the relationships log-transformed for blood cadmium, lead and zinc levels and the original age-adjusted mortality rates from all-cancer and stomach cancer using ridge regression.
^e A tolerance <0.10 indicates that collinearity must be considered.
^f Variance inflation factor; If >10, collinearity must be considered.
^j Ridge regression coefficients; a Ridge k of 0.85 and RSQ of 0.60815 were caiculated using ridge regression.
^h Ridge regression coefficients; a Ridge k of 0.40 and RSQ of 0.66049 were caiculated using ridge regression.

ISSN: 1471-2458