A case control study of multiple myeloma at four nuclear facilities

PURPOSE: Reported elevations of multiple myeloma among nuclear workers expo sed to external penetrating ionizing radiation, based on small numbers of c ases, prompted this multi-facility study of workers at US Department of Ene rgy facilities. METHODS: Ninety-eight multiple myeloma deaths and 391 age-matched controls were selected from the combined roster of 115,143 workers hired before 1979 at Hanford, Los Alamos National Laboratory, Oak Ridge National Laboratory, and the Savannah River site. These workers were followed for vital status through 1990 (1986 for Hanford). Demographic, work history, and occupationa l exposure data were derived from personnel, occupational medicine, industr ial hygiene, and health physics records. Exposure-disease associations were evaluated using conditional logistic regression. RESULTS: Cases were disproportionately African American, male, and hired pr ior to 1948. Lifetime cumulative whole body ionizing radiation dose was not associated with multiple myeloma, however, there was a significant effect of age at exposure, with positive associations between multiple myeloma and doses received at older ages. Dose response associations increased in magn itude with exposure age (from 40 to 50) and lag assumption (from 5 to 15 ye ars), while a likelihood ratio goodness of fit test reached the highest val ue for cumulative doses received at ages above 45 with a 5-year lag (X-2 = 5.43,1 df; relative risk = 6.9% per 10 mSv). Dose response associations per sisted with adjustment fur potential confounders. CONCLUSIONS: Multiple myeloma was associated with low level whole body pene trating ionizing radiation doses at older ages. The exposure age effect is at odds with interpretations of A-bomb survivor studies but in agreement wi th several studies of cancer among nuclear workers. Ann Epidemiol 2000;10:1 44-153. (C) 2000 Elsevier Science Inc. All rights reserved.