next up previous
Next: ERR Estimates by Up: Dose-Response Analyses. Previous: Results of the

Dose-Response Using Adjusted Doses.

Table IX shows the effect of using the adjusted doses on the score test values and the dose-response coefficients and LRT statistics (for the low dose data) for each cause of death category that was presented in Table VIII. The results suggest that the effect of missing dose is an overestimation of the strength of the dose-response association, and an upward bias in the dose-response coefficients.