Cancer in the offspring of radiation workersBMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7145.1672 (Published 30 May 1998) Cite this as: BMJ 1998;316:1672
Connection between leukaemia and radiation needs to be considered
- Freda Alexander, Reader in cancer epidemiology
- University of Edinburgh, Department of Public Health Sciences, Edinburgh EH8 9AG
- Green Audit, Aberystwyth SY23 1PU
- Leverhulme Trust, University of Birmingham, Department of Public Health and Epidemiology, Birmingham B15 2TT
- Childhood Cancer Research Group, University of Oxford, Oxford OX2 6HJ
- National Radiological Protection Board, Chilton, Didcot OX11 0RQ
- Institute of Occupational Health, University of Birmingham, Birmingham B15 2TT
- Cancer Research Campaign, Cancer Epidemiology Research Group, Department of Public Health, Radcliffe Infirmary, University of Oxford, Oxford OX2 6HE
EDITOR—The results of the study of Draper et al showing an association between worker status of parents and leukaemia in children1 are unlikely to be attributable to chance. Some exposure to a causative agent, either at the workplace of male radiation workers or in the communities to which they belong, must increase the risk of childhood leukaemia.
Draper et al suggest an oncogenic infectious agent that results from high rates of population mixing as the cause of increased leukaemia risk. Their evidence is based on three factors: an association of population mixing with childhood leukaemia; population mixing in the vicinity of certain nuclear sites; and worker migration between sites. These factors do not explain the excess risk of 80%. A review of all the studies published at that time of childhood leukaemia in areas with high population mixing2 gave 50% as best estimate of increased risk, with only one having an estimate as high as 80%.
The effects of population mixing, however, may be diluted in these workers for several reasons. Firstly, only a proportion will have been employed at establishments such as Sellafield. Secondly, the time windows of documented excess mixing near these plants are mostly short relative to the 34 years of this study. Thirdly, radiation worker status of the father before conception of the child is, here, an inaccurate measure of the relevant exposure (which presumably occurs after conception). Nevertheless, given the accumulating evidence that links patterns of exposure to infectious agents with childhood leukaemia, 2 3 an infectious origin for this excess should be considered.
So, I suggest, should a connection with radiation. Maybe an additional factor that applies to communities of radiation workers …