A mortality study of lung cancer among swiss professional drivers: accounting for the smoking related fraction by a multivariate approach.

The mortality due to lung cancer among 'chauffeurs', who have a presumably long-term exposure to diesel exhaust fumes, was analysed. As controls, men in industrial occupations of similar socio-economic status were selected. Cases were drawn from the Swiss mortality register for the years 1979-1983. Person-years were obtained using data from the 1980 census records. These two data files were combined by occupation, age class and socio-economic status. Age adjusted incidence rates were calculated applying Poisson regression. To control for tobacco related lung cancer mortality an indirect adjustment was undertaken. Using information about the smoking habits of the people in the occupations under study, smoking-attributable lung cancer mortality was accounted for by incorporating Axelson's technique into multivariate regression modeling. The mortality ratio for lung cancer for chauffeurs with respect to the controls was 2.27, which is significantly in excess of 1:95% CI (1.99, 2.58). Other tobacco related diagnoses such as bladder cancer, esophageal cancer and ischemic heart diseases showed excess risks as well. After accounting for smoking, a slight but significant increase in lung cancer mortality remained among chauffeurs (mortality ratio 1.48, 95% CI: 1.30, 1.68). In summary, the present results do support the hypothesis that diesel exhaust is a significant cause of lung cancer.