Bateson-T; Kopylev-L; Sullivan-P; Cooper-G; Vinukoor-L; De Voney-D
Epidemiology 2009 Nov; 20(S6):S78-S79
Background and Objective: In Libby, MT, the mining and milling of vermiculite, which was contaminated with amphibole asbestos, exposed workers and residents to asbestos fibers for decades. Several million homes in the United States and Canada may have vermiculite attic insulation from the Libby, MT mine. USEPA is conducting its cancer risk assessment based on epidemiologic analysis of the occupational cohort assembled by NIOSH. This analysis was based on 880 workers hired in 1960 or later when fiber concentrations were lower and believed to be more consistent with potential environmental exposure. We used extended Cox proportional hazards models to assess the effects of Libby amphibole asbestos on lung cancer mortality and a Poisson model of absolute risk of mesothelioma mortality. Sensitivity analyses used multiple lags for cancer latency and multiple exposure metrics including cumulative exposure, residency-time weighted exposure, and metrics allowing for fiber clearance, translocation or biologic sequestration. We compared results using AIC weights which assign probabilities of each model being the best fit. Results: The best fitting class of models in these analyses for lung cancer mortality used 10-year lags and allowed for simulated clearance over time. These models had 2.6-3.8 times the probability of being the best model compared to cumulative exposure. The best fitting model for mesothelioma mortality had 15-year lag and also allowed for clearance. The two best fitting metrics had 56-167 times the probability of being the best model compared to cumulative exposure. We found that residency-time weighted exposure models had low relative probability for both lung cancer and mesothelioma mortality. Conclusions: The adverse effects of Libby amphibole exposure on cancer mortality in this cohort are clear. Models that mathematically allow for fiber clearance over time provide clearly superior fit to these cancer mortality data.
Mining-industry; Milling-industry; Mineral-processing; Asbestos-dust; Asbestos-fibers; Epidemiology; Statistical-analysis; Lung-cancer; Respiratory-system-disorders; Mathematical-models
Epidemiology. Abstracts of the International Society for Environmental Epidemiology 21st Annual Conference, Dublin, Ireland, August 25-29, 2009