The Contribution of Butadiene to Smoking Induced Leukemia Mortality. M. R. Schulz, J. E. Korte, I. Hertz-Picciotto, and L. M. Ball, Department of Epidemiology and Department of Environmental Science & Engineering, School of Public Health, University of North Carolina, Chapel Hill, NC 27599
Critics of quantitative risk assessment suggest that standard extrapolation methods overestimate risks at low level exposures. This investigation examines whether these methods predict implausibly large risks for butadiene exposure. Total smoking-induced leukemia deaths are compared with those predicted from butadiene in cigarette smoke. Cigarette smoke is a substantial source of butadiene exposure in the general population, with smokers inhaling about 69 ug/cigarette. Sidestream exposures to the smoker may add as much as an additional 12 ug/cigarette. The potency of butadiene was estimated by fitting a linear multiplicative model to the dose-response data from an occupational study of butadiene-styrene rubber manufacturing workers (Macaluso et al., 1996). The cigarette exposures and estimated potency were used in three lifetables that calculated cumulative risk of leukemia in nonsmokers, and in smokers with and without the butadiene contribution in smoke. These multiple decrement lifetables adjusted for smoking-related competing causes of death. The results indicated that for men, depending on the number of cigarettes smoked daily, 52-87 smoking induced leukemias occur per 100,000 smokers, and that an estimated 29%-60% of these are due to the butadiene in the cigarette smoke. These risks predicted for butadiene are compatible with the observed leukemia mortality among smokers, considering that benzene is also a component of cigarette smoke. Overall this investigation provides evidence that linear extrapolation from occupational exposures does not clearly overestimate risks from butadiene at environmental levels.
Go to . . .