Faire avancer la sûreté nucléaire

La Recherchev2


Lung cancer risk associated with low chronic exposure to radon in the French cohort of uranium miners



Email :

Mot de passe :

National Radiation Environment (NRE - VII), 20-24 May, 2002, Rhodes, Grèce D. Laurier (1), A. Rogel (1), M. Tirmarche (1), B. Quesne (2) (1) Laboratory of Epidemiology, Institut de radioprotection et de sûreté nucléaire. (2) Medical Coordinator of COGEMA


Comprehensive analysis from miners cohorts have demonstrated a clear association between lung cancer risk and radon exposure cumulated over occupational life. However a large proportion of these miners had cumulated high radon exposure during a short period. The French cohort of uranium miners allows the estimation of the risk linked to low radon concentration protracted over long duration. A first analysis has been published in 1993, based on a cohort of 1785 underground miners followed up to 1985. Since then, the cohort has been enlarged and the follow-up has been extended up to 1994. Furthermore, the study will allow considering the modifying effects of confounding factors present in the mining atmosphere. The French cohort of uranium miners is conducted in collaboration with the Occupational Medical Service of COGEMA, the company in charge of the mines in France. It includes 5098 males employed at least one year as a miner since 1947, and followed up to 1994. The total number of person-years is 133 500. For years before 1956, annual individual exposure has been reconstructed retrospectively by a group of experts, both from environmental measurements in the mining atmosphere and from information about the type and place of work. Since, 1956, annual individual exposure to radon, gamma rays and uranium ore dust was recorded systematically. Forced ventilation was introduced in the mines in 1956, which led to a dramatic decrease in the level of exposure to radon (mean annual exposure of 15 to 30 WLM before 1956 to less than 4 WLM after). Causes of death were obtained from the National Mortality Database, which gathers all information from death certificates in France since 1968. Cause-of-death information from the COGEMA Occupational Medicine Department was used as a complementary source. The total number of death is 1162, including 395 cancers and 125 lung cancer deaths. Mean cumulative exposure to radon is 36.5 WLM, protracted over a mean duration of 11.5 years. A significant excess of lung cancer deaths is observed (Standardised Mortality Ratio = 1.5, 95% confidence interval = [1.2 – 1.8]). Regression shows a significant linear increase of lung cancer risk with cumulative exposure to radon (Excess Relative Risk (ERR) = 0.9 per 100 WLM, 95% CI = [0.3 – 1.5]). Period of exposure (before or after 1956) appears as an important modifier of this relationship. After adjustment on this difference, no modifying effect of time since exposure or age at exposure can be shown. No association with radon exposure was observed for other causes of death. Compared to previous results from the literature, the French cohort of uranium miners constitutes a population exposed to low levels of radon during a long duration. Such levels of exposure can be observed in certain households, and the distance is therefore reduced when extrapolating risk model from this cohort to indoor radon exposures. Statistical analyses of this cohort are continuing, especially regarding the sub-cohort of more than 3000 miners who have been employed only after 1955, when mean annual exposure to radon was always below 4 WLM. In this sub-cohort, exposures to gamma rays and to uranium ore dust were also recorded. Nevertheless, the statistical power for analysing the risk associated to such low levels of exposure is reduced. In the near future, power will be increased using two approaches: a joint analysis with data from the Czech cohorts of uranium miners, and an extension of the follow-up of the French cohort up to 1999. This work is included in the framework of a European collaborative research project, aiming to a synthesis of results from epidemiology and from animal data regarding the risk associated to low levels of radon exposure.