Epidemiological Studies of Low-Dose Radiation. A. Ekbom, Department of Cancer Epidemiology, Uppsala University, S-751 85 Uppsala, Sweden and Department of Epidemiology, Harvard School of Public Health, Boston, MA, USA
Background
The thyroid gland is believed to be especially susceptible to radiation carcinogenesis. Diagnostic radiology is the largest man-made source of exposure to ionizing radiation for the general population. It is therefore not surprising that the impact of diagnostic radiology on the risk of thyroid cancer repeatedly has been in the focus in different studies on the etiology of this cancer form.
Most studies have used a case-control approach and utilized interviews in order to reconstruct histories of exposure to diagnostic radiology. This approach has a potential to introduce a substantial underascertainment of exposure and also the possibility of differential recall of X-rays by cases and controls. An alternative approach is to ascertain exposure through medical records. Although this is a cumbersome process, and probably also will lead to some underascertainment of exposure, differential misclassification will be a non-issue.
During the 1990s two Swedish papers have been published analyzing exposure to diagnostic X-rays and thyroid cancer. The first one from Linköping consisting of 104 cases and 387 controls used interviews in order to assess the history of X-ray exposure, The second from Uppsala consisting of 484 cases and 484 controls assessed the exposure from the X-ray archives all over Sweden, The differences in the results and conclusions, serve as a good example of the pitfalls when relying on interviews.
Results
In the study from Linköping, the authors found a positive trend by dose for risk of thyroid cancer and a point estimate of 2.8 in the group exposed to more than 0.8 mGy compared to those never exposed. Exposure to diagnostic radioexami nations of the arms and lower legs where associated with the highest risk.
It the study from Uppsala, no association could be found by dose or by any single X-ray examination after taking into account previous diseases of the thyroid gland. Moreover, the lifetime dose was higher in the study from Uppsala both for cases and controls compared to that from Linköping, indicating that underascertainment occurred to a higher degree when utilizing interviews compared to retrieval through medical charts.
Conclusion
Retrospective interviews in order to assess exposure to low-dose radiation such as diagnostic radiology have a potential to be highly unreliable and can result in false positive associations.
Go to:
Table of Contents of the 1997
SRA-Europe Conference
Program of the 1997 SRA-Europe Conference
RiskWorld Welcome Page
Tec-Com, Inc.