Childhood Leukaemia Incidence and Background Ionising Radiation – a French ecological study

Introduction The leukemogenic effect of ionising radiation has been well established for high-dose exposure. In contrast, the risk associated with natural radiation exposure still needs to be documented. Ecological studies have suggested a positive association between indoor radon concentration and childhood leukaemia, which has not been confirmed by case-control studies. The studies that investigated the relationship between background gamma dose rates and childhood leukaemia have yielded discordant results, sometimes positive, sometimes negative, significant or not significant. Objective The present study aimed at evaluating the ecological association between background ionising radiation (indoor radon, indoor terrestrial gamma radiation, and cosmic rays) and acute leukaemia incidence among children under 15 years of age in France between 1990 and 1998, at the level of “employment areas” (348 geographical units called zones d’emploi, noted ZE), and at the scale of the 95 “départements”. Methods – Cases During that period, 4015 cases were registered by the French National Registry of Childhood Leukaemia and Lymphoma (3270 ALL and 697 AML). Exposure Exposure assessment was based on measurement campaigns, including 13240 measurements of indoor radon concentration, and 8737 measurements of indoor terrestrial gamma dose rates. Cosmic rays were estimated from localisation and altitude in each “commune” of France (about 36 000 “communes” in France), and then, weighted by population counts in order to obtain cosmic rays in each ZE. Indoor radon concentration was measured in 343 ZEs out of 348, and indoor terrestrial gamma radiation in 214 ZEs out of 348. Outdoor environmental gamma measurements were carried out in order to monitor ambient radioactivity due to gamma radiation in each “département”. These measurements were used to complete the measures of the “départements” where no measurement of indoor terrestrial gamma radiation was available. In order to convert indoor radon concentration in Bq/m into annual dose to the bone marrow in mSv per year, we have used a conversion coefficient suggested by Richardson et al. (1991). Then, we have estimated the annual dose to the bone marrow from the three components of background ionising radiation (radon, terrestrial and cosmic gamma rays). Statistical analysis Exposure in each ZE was transformed into a five-category qualitative variable. Each category included approximately a quintile of the expected number of cases. The arithmetic mean of exposure was also considered as a quantitative variable. Analyses were conducted using Poisson regressions, including socio-demographic ecological covariates such as average net income, socio-economic status, level of education and proportion of the inhabitants living in rural areas. Analyses were systematically performed for all cases (0-14 years) and for the complete period (1990-1998), and then, separately, by age