Assessment of radiological and non-radiological exposures for epidemiological follow-up of the French uranium enrichment workers

Introduction
In France the assessment of regulations respect is based on external dosimetry and uranium bioassays of uranium exposed workers. In epidemiology these data are useful for studying radiotoxic effects but need to be completed with chemical-physical characteristics of uranium when its chemical toxicity is considered. Since regulations on non-radiological exposure in nuclear industry are recent; no individual exposure data exist. Our objective was to assess the amount of radiological and non-radiological exposure in the uranium enrichment plant in France by elaborating a plant specific job exposure matrix (JEM) for epidemiological purpose.

Methods
A group of 29 experts realised an inventory of all uranium compounds and other chemical and physical pollutants known as carcinogens, mutagens or toxics, used in the plant. Jobs were sorted in homogeneous exposure groups. For each job, active and retired workers were asked for assessing the level of exposure for all categories of pollutants. Final levels of exposures and internal consistency of the JEM was validated by the experts and compared to available measurement data (uranium atmospheric concentration (Bq/m3) and noise (dB)) based on Spearman’s correlation ρ test.

Results
In total, 227 workers participated in the exposure level assessment of 26 categories of pollutants in 102 occupational groups through 1978 to 2008. Only 7% of exposure levels were rectified by experts for internal consistency reasons. Noise, heat, trichloroethylene and soluble uranium compounds were the most prevalent exposures at the plant although their levels tended to decrease across time. JEM exposure levels were fairly well correlated with uranium atmospheric measurements especially for soluble natural uranium compounds (rho=0.46) and noise (rho=0.43).

Discussion
In the context of missing information on individual radiological exposure and absence of non-radiological exposure data, our JEM presents a valid alternative for exposure assessment in epidemiological follow-up of the French uranium enrichment workers.