Mortality data from 9609 workers at two asbestos mining areas in Quebec were analyzed to assess the effects of the intensity and timing of exposure on lung cancer risk. Summary exposure measures based on differing assumption were computed for lung cancer cases and matched controls and were fitted to the data using conditional logistic regression. A non-linear relationship between intensity and risk fit both mining areas, but risk was greater at one area than the other. At the mine with lower risk, exposure occurring more than 30 years prior to death had little effect, while at the other mine risk did not vary with time since exposure and men starting employment before 1924 were at elevated risk. The results point to differences in dust composition at the two areas and illustrate the difficulties in estimating risk.
From the Department of Medical Biostatistics, University of Vermont College of Medicine, Burlington, Vt.
Address correspondence to: Pamela M. Vacek, PhD, Department of Medical Biostatistics, University of Vermont College of Medicine, Hills Building, Burlington, VT 05405.