Mortality in the cohort of nuclear workers of the French electricity company: 1968-2003 period

Full text: Background: We conducted a mortality study on a cohort of French nuclear workers employed at Electricite de France (EDF). A first mortality analysis relied on the period 1968-1994. Present results consider an updated mortality analysis including 9 additional years of follow-up and rely on the period 1968-2003. Methods: The cohort includes 22,393 workers, 96% of them are males. Employment data were updated using Personnel EDF file. Vital status was ascertained using the French National Registry of individual identification, and further completed using EDF personnel and pension files. Causes of death were obtained through the National file of causes of death. The observed mortality was compared with national rates by indirect standardisation and expressed by Standardised Mortality Ratios (SMR). All causes, all cancers and external causes SMRs were detailed according to sex, age and calendar year. Variations according to date of entry into cohort, duration of employment and socio-economic status (SES) were studied. Comparison with the earlier follow-up is presented. Results: At the present study end point (31/12/2003), 74% of workers are still in active service. Only 0.3% of workers are lost to follow up. The total number of person-years is nearly 450,000. Causes were ascertained for 96 % of deaths. The total number of deaths is 874: 307 are cancer deaths, including 16 leukaemia. SMRs for all causes of death and all cancers show a significant deficit compared to the French national mortality. Significant variations of all causes and all cancers SMRs according to SES were observed. No significant excess compared to French national mortality was observed for any of the 30 different cancer sites studied. Borderline significant excess were observed however for pancreatic, pleural, kidney and brain cancer. SMR for leukaemia is equal to unity. Conclusion: There is a clear evidence for an effect of selection of healthy individuals into the workforce. Although 9 years of follow-up were added, this cohort is constituted by relatively young workers, most of them being still in active service. Regular update of the follow up of this cohort together with inclusion of newly employed workers is planned. This project contributes to a better occupational health surveillance and will be included in large international studies aiming to study low risks in relation to ionising radiation. In this frame, individual radiation exposure data will be updated very soon in order to allow an updated exposure-risk analysis. (author)