Radiation, Science, and Health

"Low Level
Radiation Health Effects: Compiling the Data"

Revision 2
March 30, 1999

by Radiation, Science, and
Health, Inc.,
Edited by J. Muckerheide

1.2 Human Population Epidemiology

References

"Either a threshold, hormesis, or both are found in most major studies in which low doses of ionizing radiation were used to study radiation induced cancer. Hormesis is the stimulation of physiologic functions by sub-harmful doses of any agent. Hormesis with ionizing radiation has been reviewed (Luckey 1980, 1982, [with 1,269 references]). In the following examples hormesis activates homeostatic mechanisms which provide effective resistance to cancer induction. Minute doses of ionizing radiation give results which are diametrically opposite to those obtained with high doses. This literature review presents data from animal experiments and human subjects to show cancer induction to be generally less than control values when minute doses of ionizing radiation are utilized."

"In a study of 36,000 hyperthyroid patients Saenger et al (1968) found those treated with 131I showed only 60% as much leukemia as those surgically treated. During 20 years patients treated with radioiodine had no excess thyroid tumors, and the total cancers were only one-third as numerous as these found in the surgery cohert (Dobyns et al 1974). Holm (1980) indicated that the use of radioiodine reduced the probability of thyroid cancers in adults. Holm et al (1980) report that radioiodine treatment did not induce any cancers in excess of those expected when compared with a carefully matched set of patients from the same hospitals. The use of 60 mCi 131I resulted in only 1/10th as many cancers as was estimated by the UNSCEAR model.

"Radium dial painters provide the best long term study of low dose ionizing radiation in humans. When the incidence of cancers is plotted against the log dose, the curve shows has a threshold at about 1000 rads. This is equivalent to 2 mCi/kg for 50 years. The same basic result was obtained by Rowland et al (1983) in a study of over 3000 radium dial painters. Evans (1981) comment is pertinent: 'A linear nonthreshold model would have predicted in this group some 15 radiogenic tumors between 0 and 1000 rad. The probability of observing none, if a linear nonthreshold model were correct, is 1 in 5 million. Clearly the conservative linear nonthreshold model used in much 'prudent' radiation protection work is strongly rejected by the actual data in the case of 226Ra and 228Ra in man.'

"In a study of plutonium inhalation by 240 workers, Voeltz et al (1983) noted that about half of the whole body burden of 19 uCi was retained in the lungs for over 30 years. There was only one lung cancer mortality in this cohort. Linear and linear-quadratic models predicted 70-200 lung cancers in this cohort where the average dose rate was 1.01 rad/yr for lung and 1.69 rad/yr for bone surfaces. Five lung cancers were expected by comparison with USA population rates. Total cancer mortality was 8 for the 224 males and zero for the 17 females, compared to 15 expected for the US population averages. These results are compatable with the hormetic model and contradict all of the no threshold models.

"In recent studies of cancer in nuclear workers Gilbert and Marks (1979) found only half as much leukemia in such workers as expected nationally. Boice and Land (1983) found no excess lung cancer when age-adjusted data was evaluated. Webster (1981) found no excess cancer in 25,000 workers at the nuclear powered naval shipyard. When normal procedures were used to study the Hanford workers, Darby and Reissland (1981) found a negative correlation between radiation levels and cancer induction, clear evidence for hormesis. Abbatt et al (1983) found cancer mortality in 18,000 Canadian nuclear power plant workers to be less than that of carefully matched controls in Ontario. Cancer among the nuclear workers was less than that found in thermal energy workers or in other workers within the same company who received the same medical examinations and care.

"Each of the above animal and human studies suggest that minute doses of high LET radiation are hormetic…. Unscientific extrapolation of the data from miners throughout the world suggest that 6.5, 10, or 39 excess lung cancers-would be expected at low doses of radon. Reasonable interpretation of the data … indicate that minute doses would be expected to follow either a threshold or a hormetic model.

"The data reviewed here is overwhelmingly supported by low LET radiation, data (Luckey 1985). The mechanisms are comparable; Tuschl et al (1983) found that radon exposure in humans increased the level of DNA repair enzymes. All this information suggests that the no-threshold models and the ALARA concept are not scientifically acceptable representations. The hormesis model fits most of the data and clearly indicates that cancer incidence and cancer mortality would be decreased by the addition of minute doses of ionizing radiation. Regulation commissions must recognize hormesis or they will continue to promulgate rules which prevent natural homeostatic protective mechanisms at great expense to industry and human health."