Abstract

Aims: To assess the cumulative effect of asbestos on lung cancer risk where the exposure is assessed by an expert rating.

Methods: 1678 male cases and controls were enrolled in a population based matched case–control study, focused on occupational risk factors, carried out in West Germany. The exposure to asbestos was computed as lifelong working hours. For a validation subsample of 164 matched pairs from this study the intensity of asbestos exposure was further assessed by a panel of experts in order to obtain an estimate of the cumulative exposure on a time by intensity scale (fibreyears). The information on duration of asbestos exposure in the original study was combined with the fibreyears following the two phase case control study paradigm.

Results: The number of exposed subjects in the validation subsample was 75 cases and 71 controls. The percentage of subjects with a cumulative exposure ≤1, 1 to ≤10, and >10 fibreyears was 16%, 15%, and 15% for the cases and 18%, 16%, and 9% respectively for the controls. The smoking adjusted odds ratios for the fibreyears based on an unconditional logistic regression were 0.81, 1.02, and 1.60 respectively with increasing exposure categories (not significant). The coefficient (beta) for a log transformed trend was 1.156. Applying the two phase paradigm, these odds ratios became 0.86, 1.33, and 1.94; the latter reached significance and the beta coefficient was 1.178.

Conclusions: The two phase paradigm allowed us to obtain a more precise estimate of the effect of asbestos on lung cancer. Results are consistent with a doubling of the lung cancer risk with 25 fibreyears asbestos exposure.