Environment Agency put on notice of liability
a letter from the Low Level Radiation Campaign
and what came next

ENVIRONMENT AGENCY
PO Box 446
Bedford
MK42 0ZR

25 August 2000

Dear Sir or Madam,

BNFL//MAGNOX Consultations:

Berkeley Centre, Berkeley Power Station, Bradwell Power Station, Dungeness Power Station, Hinkley Point A Power Station, Oldbury Power Station, Sizewell A Power Station, Trawsfynydd Power Station and Wylfa Power Station.

We are responding to your consultation on BNFL's applications for radioactivity discharge licences in respect of the above sites.

We are hereby giving you formal notice of unequivocal evidence of insecurity in the radiation risk models employed by you and by the National Radiological Protection Board when considering the health effects of exposure to radioactive materials discharged to the environment.

This evidence is the increased incidence of leukaemia in the cohort of infants who were in the womb in 1987 and 1988 and were exposed to the radioactive fallout from the Chernobyl accident. This phenomenon has been reported from five separate countries:- Germany ^[1], Greece ^[2], USA ^[3], Wales ^[4]and Scotland^[4].

The radioisotopes to which the populations of these countries were exposed are the same as those to which the present consultation relates. Exposure doses were well below Natural Background Radiation levels.

Since details of exposure to the exposed cohort of children in Wales and Scotland had been published by the National Radiological Protection Board it was possible in the case of those two countries to establish the ratio between the number of cases predicted by their risk model and the number of cases observed. The ratio is more than 100 to 1 [4] and this represents the scale of the error in the risk model.

An error of this magnitude accounts for the well established childhood leukaemia excess near other sources of the same man-made radioisotopes, namely, Sellafield, Dounreay, La Hague, Aldermaston and Burghfield. Whilst various unconvincing explanations have been advanced for these leukaemia excesses, there can be no other explanation for the infant leukaemias following Chernobyl. As the combined probability that these findings were a chance occurrence was less than 0.0000000001 (or <10-8) we consider it unequivocally established that the present risk model for exposure to fission-product radioactivity is in error by a large value.

In legal cases which established liability for lung cancer consequent upon exposure to asbestos it was argued that the defendants' knowledge of the health consequences of asbestos, and in particular the date on which it could be shown that they had this knowledge, were important determinators of their being found liable. A similar situation existed with the Thalidomide litigation, where the date of knowledge of the effects of the drug was a critical factor in the subsequent jailing of those responsible for continuing to make it available on the market. In both cases individuals involved in the decision making process were held personally responsible.

Similarly you have now been informed of the evidence that the radiation risk models are insecure. This deprives you of any subsequent defence that you were unaware of the likely health consequences of licensing emissions of radioactivity from the nine sites involved in this consultation.

We feel we should warn you that it will not be sufficient merely to rely on the opinion and advice of the National Radiological Protection Board, nor on that of the International Commission on Radiological Protection, with whom NRPB has a significant overlap of key personnel. When dealing with the problem of predicting risks from the type of practice you are now considering these bodies employ deductive methodology based on the assumption that their risk factors are secure, rather than inductive reasoning to examine the manifold observed associations between radioactive contamination and disease. Since the philosophy of science requires inductive rather than deductive reasoning it follows that NRPB and ICRP are not scientists - merely technicians. They are well aware of the shortcomings of their modelling, particularly for internal radiation from incorporated radionuclides. For your benefit we will summarise them :-

The official estimation of radiation hazard at low dose depends on studies of the survivors of the Hiroshima and Nagasaki bombs which ('though NRPB says they are "pivotal") have been widely criticised. Briefly, the group considered to be "exposed" consisted of those who were in the open at the time of the explosion, and so received a single large acute external dose of gamma rays. The control group consisted of people who were elsewhere at the time or were shielded. But both groups lived in the bombed cities, and were therefore exposed to ingesting, inhaling and absorbing fallout. The studies are therefore silent on internal radiation and the very different types of exposure involved - chronic, low dose, low dose rate, internal, alpha and beta emitters.

The model which NRPB and ICRP use is essentially a physical one. As far as the epidemiology is concerned all that has been done is a straight line extrapolation from the exposed group's high dose data points down to the origin. This is supposed to define risks in the low dose region. The extrapolation has been widely criticised: Goodhead calls it "a large region of uncertainty"; Wright at MRC questions the relevance of bomb survivors' data; Sternglass , Busby , Gofman , and Alice Stewart all attack it on various grounds. Early reports of cancer incidence at Hiroshima were seriously out of line with those later used to set risk factors .

Other studies which are supposed to inform on risk are mostly of external x-rays. A couple of small internal studies are included, but as these are of natural isotopes they are of little value in estimating the effects of artificial substances which never existed before the nuclear era and which in biological systems behave like natural substances necessary for building tissue and enzymes.

The scientific literature has several examples of studies which purport to show no correlation between disease and radiation dose, but which do show an effect in populations exposed to radioactive fallout from weapons tests and Chernobyl, and in nuclear workers monitored for internal contamination and in their children. A synopsis of these studies is included in the attached Compendium.

The Compendium also includes reference to a large number of studies which together if not separately provide prima facie evidence of the error in risk factors which we have mentioned above. They are, in other words, the evidence to which inductive reasoning should be applied by any protection agency with scientific integrity.

The Compendium is posted on our website. This notice will be posted there as well by the time you receive this submission.

Yours faithfully

Richard Bramhall

References

1. Michaelis, J, Kaletsch U, Burkart W, Grosche B (1997) ‘Infant leukaemia after the Chernobyl Accident’ Nature 387, 246

2. Petridou E, Tricopoulos D, Dessypris N (1996) ‘Infant leukaemia after in utero exposure to the radiation from Chernobyl’ Nature 382: 352-353

3. Mangano J (1997) ‘Childhood leukaemia in US may have risen due to fallout from Chernobyl’ British Medical Journal 314: 1200

4. Busby C, Scott Cato M (2000) ‘Increases in leukaemia in Infants in Wales and Scotland following Chernobyl’ Energy and Environment 11(2) 127-139