Estimation of the past and future burden of mortality from mesothelioma in France.

OBJECTIVES Firstly to evaluate future mortality from mesothelioma in France with an age-period-cohort approach and evaluate different hypotheses on risk of mesothelioma for the most recent birth cohort. Secondly to compare the results with a British and an American study. Thirdly to study if any trends were detectable on data for women which would be consistent with the consequences of increasing environmental exposure to asbestos. METHODS Estimates of mortality from mesothelioma among men and women in France from 1950 to 1995 were based on the analysis of the pleural cancer mortality data coded 163 in the ninth revision of the international classification of diseases (ICD-9). Correction factors were used to derive the mortality from mesothelioma from these data, based on two regional registries. The analysis of the past mortality data has been performed by an age-cohort model (with a maximum likelihood technique). Predictions of deaths from mesothelioma over the next 50 years were based on four different assumptions on the risk of death from mesothelioma in future birth cohorts. RESULTS The predicted lifetime probability of dying from mesothelioma increases until the last birth cohort 1964-8 among men whereas it decreases strongly from the 1954-8 birth cohort among women. The projected numbers of deaths from mesothelioma in France until 2020 are similar, whichever hypothesis is considered: around 20,000 deaths from mesothelioma might occur among men and 2900 among women from 1996 to 2020. CONCLUSIONS French data show an increasing lifetime probability of death from mesothelioma in the more recent male cohorts. Although the mortality burden can be predicted until 2020, and is intermediate between the United Kingdom and United States estimates, there is still high uncertainty on the figures after 2020. No increase is found in women, and this does not support the hypothesis that current environmental exposure to asbestos could be associated with a detectable risk of death. Specific surveillance should be set up to monitor future trends or their absence.