|
|
|
POISONING EDEN THE HEALTH EFFECTS OF RADIOACTIVE CONTAMINATION OF THE ENVIRONMENT
Given at the UKAEA/BNES conference: NEDCON'02 Chilton Oxford, April 25-26th
Chris Busby, PhD
Occasional Paper 5/02 Aberystwyth: Green Audit Since a wise man may be wrong, or a hundred men, or several nations and since even human nature, as we think, goes wrong for several centuries on this matter or that, how can we be certain that it occasionally stops going wrong and that in this century it is not mistaken Montaigne (1533-92): The Essays.
1. Introduction This conference is an attempt to discuss how we can make safe what is intrinsically unsafe. How can we do this most effectively? Can we do it at all? If radioactive materials are released to the environment they will end up in the food, the air and water and ultimately living creatures. This includes us and our children. We know this because we have measured weapons fallout isotopes in the bodies of the most isolated tribes on earth, and we have found plutonium from Sellafield in the teeth of children from across the whole of the UK (Priest et al. 1997). Most scientists now take the view that there is no safe dose: this is a consequence of the discrete nature of ionizing track events. Cells are either hit or not hit, even at the smallest dose. And if they are hit, then there is a finite probability of a mutation in the genetic data on the chromosomes, and a finite chance of this eventually leading to serious or fatal illness. The nuclear project was originally countenanced politically at a time when scientists were less knowledgeable (the DNA structure was not even known in 1952). It was developed through a period of Cold War, when atomic bombs were thought to be necessary, and dissenting views about the health effects were not permitted. If anyone doubts this they should be aware that the World Health Organisation signed an agreement in 1958 to leave research on the health consequences of radiation exposure to the International Atomic Energy Agency, whose remit was the advancement of nuclear power. This agreement is still in force (although the European Parliament are trying to have it amended). What has emerged in the last thirty years is a body of evidence that the science which was and is presently used to define the health effects of exposure to internal, novel, man-made radionuclides is in error by a very large amount. This scientific black-box of radiation risk assessment is the model of the International Commission on Radiological Protection (ICRP), a body set up in 1952. The ICRP model (ICRP90, ICRP66) is not valid for internal radiation exposure. The data showing this are now so compelling that the European Parliament passed a resolution in Spring of 2001 asking for a reassessment of the risk model. A similar proposal was passed at the WHO Chernobyl conference in Kiev, in June 2001. On July 31st, the UK government set up a new group, the Committee Examining Radiation Risk from Internal Emitters (CERRIE) in order to examine the evidence and report on the security of the ICRP model. Thus the immediate background to this conference Thus the background to this present conference on nuclear discharges to the environment includes a question mark over the whole basis of calculating the health consequences of such discharges. This paper addresses the evidence that the ICRP model is wrong and that a new model is needed to properly advise on the risks from different types of radiation exposure that follow the various kinds of discharges that are proposed.
2. The ICRP and Radiation Risk Scientific knowledge becomes either refined or altered as a consequence of the ability of its models to explain and predict real world data. There have been many cases of scientific models being discarded and replaced following failures to explain observation. Early theories in chemistry and physics, like the ICRP model, were based on mathematical averaging of large amounts of material: litres of fluid, kilograms of metal, whatever could be conveniently experimented upon. Development of experimental techniques in the 20th century which investigated the microscopic behaviour of the atoms and molecules that made up these kilograms and litres, resulted in falsification of many scientific models. Empirical results were often fatal for theories: this is how science advances. For example, thermodynamics was unable to explain spectroscopic observation of molecular and atomic transitions and had to be discarded for quantum theory. We are at such a crossroads now. The body of knowledge about the biological consequences of low dose radiation from internal emitters is no longer congruent with the simplistic averaging models of the ICRP, which, like their thermodynamic ancestors, use kilograms of matter and litres of fluid. The radiological equivalent of the atoms of Planck, and the biological target for radiation effects, is the living cell. Inadequate recognition of this has resulted in a situation where many people have died and will die, because the ICRP model has been wrongly applied to internal point sources. In the ICRP model, doses are defined as energy per unit mass or DE/DM . The quantity of mass employed is that of an organ or larger. One Gray is the absorption of 1 Joule by 1 kilogram of tissue. Very early on, ICRP had to recognize that this model was inadequate since experiments showed that it was the ionization density that was the important factor in cell killing, and so they added a fudge factor or ‘weighting’ for this to the Gray to give the Sievert. For alpha decays, 1 Gray becomes 20 Sieverts. The ICRP modelling of the dose from an internal exposure, and its problems, are shown in Table 1.
Table 1 Problems with the stages of modelling internal exposures used by ICRP to obtain doses.
The main failure of the system used to calculate dose is that the result is an average, in space and in time. The external dose calculation has just been applied to internal dose by averaging all energy of the decays which occur in a bag of water the same size and shape of the organ over its mass. Why is this wrong? Because it is individual cell doses which decide the magnitude of the biological effect, and for internal emitters which are point sources, some cells will receive very high doses whilst other cells receive none. Some clusters of cells will receive high doses whilst others will receive none. This is not the case with external irradiation where the source is effectively planar and thus all cells will receive the same dose. There are other problems also which must be addressed. And because the theoretical dosimetric model is wrong, it is not possible to use epidemiology of external irradiation to inform us of risks from internal exposure. Table 2 addresses the way in which doses have been correlated by ICRP and other risk agencies with effects like cancer and genetic damage. Here it is clear that scientific method has not been used properly. Cancer yield in the Hiroshima survivors who suffered almost lethal acute external doses has been linearly extrapolated to internal chronic doses at levels close to background radiation levels. Massive evidence of harm at low dose exposure from internal isotopes has been routinely dismissed by ICRP and its satellites on the basis of the deductive application of this external Hiroshima model.
Table 2 Problems with ICRP system of relating dose to health detriment The arguments which falsify the ICRP models may be reasonable divided into scientific, mechanistic and epidemiological, and I will now address these in turn.
3. Philosophical arguments The classical exposition of the scientific, or inductive method (originally due to William of Occam) is what is now called Mill’s Canons, the two most important of which are: · The Canon of Agreement which states that whatever there is in common between the antecedent conditions of a phenomenon can be supposed to be the cause, or related to the cause, of the phenomenon. · The Canon of Difference which states that the differences in the conditions under which an effect occurs and those under which it does not must be the cause or related to the cause of that effect. In addition, the method relies upon the Principle of Accumulation which states that scientific knowledge grows additively by the discovery of independent laws, and the Principle of Instance Confirmation, that the degree of belief in the truth of a law is proportional to the number of favourable instances of the law. To the methods of inductive reasoning we should add considerations of plausibility of mechanism. Science is also unique in the area of philosophy since theories may be experimentally falsified. It only requires one firm observation which cannot be predicted or explained in the accepted scientific model of a phenomenon to falsify this model. These are the basic methods of science [Mill, 1879; Harre, 1985; Papineau, 1996] The questions of interest here are: (1) What are the health consequences of exposure to external radiation doses at levels below 2mSv, the approximate dose received from natural background? (2) What are the health consequences of exposure to novel internal radioisotopes at whole organ dose levels below 2mSv? (3) Is the concept of dose applicable to internal radiation exposures? Although risks from exposure to high levels of ionising radiation are generally accepted, since they are fairly immediate and graphic, the situation with regard to low-level exposure is curious. There are now two mutually exclusive models describing the health consequences of such exposure. There is the ICRP one, which is that which is presently used to set legislation on exposures and argue that low level radiation is safe, and a radical one, which is espoused by the anti-nuclear movement and its associated scientists. I show these schematically in Figure 1. The two models arise from two different scientific methods. The conventional model is a physics-based one, developed by physicists prior to the discovery of DNA. Like all such models it is mathematical, reductionist and simplistic, but because of this is of powerful descriptive utility. Its quantities, dose, are average energy per unit mass or dE/dM and in its application, the masses used are greater than 1kg. Thus it would not distinguish between the average energy transferred to a person warming themselves in front of a fire and a person eating a red hot coal. In its application to the problem at hand, the internal, low-level, isotopic or particulate exposure, it has been used entirely deductively. The basis of this application is that the cancer and leukaemia yield has been determined following the external acute high-dose irradiation by gamma rays of a large number of Japanese inhabitants of the town of Hiroshima. Together with this, other arguments based on averaging have been used to maintain that there is a simple linear relationship (in the low-dose region) between dose and cancer yield. This Linear No Threshold (LNT) assumption enables easy calculations to be made of the cancer yield of any given external irradiation. By comparison, the radical model arises from an inductive process. There have been many observations of anomalously high levels of cancer and leukaemia in populations living near nuclear sites, especially those where the measurements show that there is contamination from man-made radioisotopes, e.g. reprocessing plants. In addition there are populations who have been exposed to man-made radioisotopes from global weapons tests, downwinders living near nuclear weapon test sites and those exposed to these materials because of accidents (like the Chernobyl infant leukaemia cohort, which I return to) or because of work in the nuclear industry or military. In contrast to the averaging approach of the conventional model, the biological model considers each type of exposure according to its cellular radiation track structure in space and in time. It is not easily possible to employ such a model to predict risks from ‘radiation dose’ to ‘populations’ but only from microscopically described doses from specific isotopes or particles whose decay fractionations are considered to interact with cells which themselves respond biologically to the insults and may be in various stages of their biological development. The dose-response relationship following from this kind of analysis might be expected to be quite complex. In examining radiation risk, we see that these models are mutually exclusive and we must decide which one is correct. In making such a decision we should employ the basic philosophical rules of scientific method. We might argue that the Linear No Threshold ICRP model is scientifically sound in its application to acute, high dose, external irradiation and its extension to acute, external, low level radiation may be justified on the basis of theory, since the plausibility of the model rests on the idea of uniform density of radiation track events in microscopic tissue volumes. However, with regard to internal radiation doses, there has been a serious misuse of scientific method in the extension and application of the ICRP external model. The basis for this conclusion is that such a process involves deductive reasoning. It falsely uses data from one set of conditions, high-level, acute, external exposure to model low-level, chronic, internal exposure. The procedure is scientifically bankrupt, and were it not for political considerations, would have been rejected long ago. On the other hand, it should be clear that the model which begins with the epidemiological observations (which suggest high risk) conforms to all the requirements of the scientific method listed above. Man-made radioisotopes, often in the form of ‘hot particles’ are common contaminants to the areas near nuclear sites where there are cancer and leukaemia clusters, and to nuclear site and test site downwinders, and to fallout-exposed populations. This satisfies the Canon of Agreement. The contingency analysis tables with control populations for such studies show that the Canon of Difference is also satisfied: people living in more remote regions than the downwinders show lower levels of illness. The Principle of Instance Confirmation is fulfilled since so many studies have shown that increases in cancer and leukaemia follow exposure regimes at low dose. Poppers description (Popper 1962) of scientific models being open to falsification is addressed by a number of studies, but most plausibly by the two sets of unequivocal findings described later in this paper which falsify the ICRP model in this strict philosophical sense. We are left only with ‘Plausibility of Mechanism’, which will be addressed next.
4. Mechanistic arguments The target for radiation action is the living cell. Thus it is the impact on the cell and its supporting environment that should be the measure of radiation stress. The ICRP model uses units of dose, Sieverts and Grays, which remove the calculation of cell dose from the cell to the organ or the whole body. Radiation stress should properly be assessed in terms of tracks or hits to each cell. For low LET radiation, such as X-rays or gamma rays, there are approximately 70 ionizations across a 8m diameter cell nucleus, equivalent to about 1mGy dose, although individual tracks may exceed this value because of track stochastics. However, within the cell nucleus, low LET radiations can produce dense regions of ionization, and these may be important in the case of isotopes, particularly Auger emitters boiund to DNA. For high LET alpha particle tracks of 4MeV, there are about 23000 ionizations and .an absorbed dose of about 400mGy
. In one year, natural background external radiation levels, measured as 1mGy to the whole body, ensure that each cell in the body receives one track or 'hit'. The health consequence of this to the whole organism is purely a function of the overall probability that the cell will acquire a harmful fixed mutation. If all cells were equal in their sensitivity to radiation this would be merely a function of the number of cells hit, and it is this argument that is the basis of the linear, no threshold dose response relationship assumed by ICRP. Double the dose and twice as many cells are hit; treble the dose and three times as many. The problem is that living cells do not conform to the kind of stress/strain relationships exhibited by steel wires. The damage from a single track of ionizing radiation is either lethal or else causes the cell to enter a DNA repair and cell replication sequence which lasts about 10 hours. This repair replication cycle, once begun, is irreversible, and the daughter cells may be exact copies of the parent or, as a result of inefficient DNA repair, may have acquired a fixed mutation. This fixed mutation may be harmless, lethal to the cell, or harmful, in that it may be a critical part of the process that leads to unchecked replication or cancer. Therefore the probability of acquiring either the initial or some later critical step in the path to clinical expression of cancer following a radiation dose is seen as both a binomial and quantitative probability calculation involving a group of very low probabilities. Assessing the mechanistic basis of such a risk comes down to examining each stage in this process. Clearly any factor which alters each of the probabilities in the chain of events, particularly those affecting the sensitivity of the cell to radiation, will increase the level of risk from the exposure being considered. The ICRP concede part of this argument: for example, they argue that at very high doses there is a sparing effect for cancer induction owing to the preponderance of cell killing over cell mutation. At low dose and protracted exposures they introduce a dose rate reduction factor, which allows for this. Some pro-nuclear authorities argue more radically that at low doses, most cells are able to repair damage, and that there is a threshold below which exposure has no harmful effects whatever. Others go further and argue that low doses of radiation effect a priming of the repair systems and actually protect against subsequent radiation damage: this is called ‘hormesis’. I will address threshold and hormesis below. All contemporary theories of low dose action accept is that cells repair radiation damage and are not passive agents. This has a number of serious consequences for attempts to average dose over tissue. It is of interest that the ICRP and others, by assuming a no-threshold linear response, implicitly exclude consideration of cell repair responses.
4.1 The double hit in space For external whole body irradiation from gamma rays and X-rays, both human epidemiology at high dose and animal studies support a linear quadratic dose response for cell transformation and certain cancers e.g.leukemia in the intermediate dose range i.e.50-5000mGy. Outcome can be written:
Response = A(dose) + B(Dose)2
Where A and B are constants. Kinetic theory suggests, and most now agree, that this result is best interpreted as showing that at low doses, single events dominate the risk but that at higher doses, the quadratic term demonstrates that two correlated events can occur (resulting in enhancement of effect). Because of the 2-stranded complementary nature of DNA, and for other reasons, it is suggested that the two correlated events are two breakages of separate DNA strands opposite each other, called a ‘double strand break’. Such a process leads inevitably to high risk of a fixed mutation since, unlike a single strand break, there is no complementary copy available to use for a repair. The two correlated hits must occur on the same section of the chromosome, and must occur either together or within the 10-hour repair replication period. The quadratic portion of the dose response curve occurs at high dose, when the radiation fluence or consequent ionization density is sufficently high to confer a high probability of two correlated hits occurring in the same cell and same section of chromosome. At natural background radiation levels of 2mSv, a double hit to a cell within the 10 hour repair replication period is very unlikely. It can easily be calculated from Poisson cumulative probability and based on 500keV events is about 4 x 10-4, some 40 times higher than the natural heritable genetic damage frequency in human populations. So we can say that, for external irradiation, correlated double hits do not take place in the low dose range and on this basis a linear dose response is mechanistically justified. However, this is not true for internal irradiation from an immobilised point source where there are sequential decays. Micron and sub-micron diameter ‘hot particles’ represent such a risk. Tables 3 and 4 below show calculations of hits and dose to local tissues of two common alpha emitting particles that have been introduced into the environment, plutonium oxide (Sellafield, weapons fallout) and ceramic depleted uranium (military use in the Persian Gulf, Bosnia and Kosovo). In both these cases, inhalation and translocation to lymphatics results in high and repetitive local doses to cells. These exposures are modelled by ICRP, after dilution of the decay energy into large masses of tissue, as vanishingly small doses. The response to these small doses is then modelled as a linear response. But strictly, since there are two hits inside the repair replication period, the response will be proportional to the square of the dose. We might more accurately approximate it to the response to a very large dose, one which results in a mean two-hit flux to every cell in the tissue (and proportional to dose squared), divided by the cell density per kilogram. It is clear that this is an error in the ICRP model as applied to internal irradiation.
Table 3 Doses to sphere of tissue 30 micron radius by one particle of U3O8 of various diameters
Assumptions: Uranium Oxide (U238) is in the U3O8 form (density = 8.6); specific activity of U238 = 12.43 MBq/Kg: Alpha decay energy = 4.45MeV; Alpha range = 30 microns. Relative Biological Effectiveness factor for Alphas = 20 (from ICRP) has been used to convert dose in Grays to effective dose in Sieverts
4.2 The double hit in time The response of cells to exposure in the low dose range has been referred to. Measurements of variation of transformation sensitivity to radiation in different phases of the cell cycle demonstrate that cells in repair replication are hundreds of times more sensitive to radiation than cells in quiescent phase. (e.g. Sinclair and Morton, 1966). This has been known for most of the radiation age, and is the basis of radiotherapy for cancer, where the rapidly dividing cells are preferentially killed. The first consequence of this is that this variation of sensitivity of cells between the two phases should lead to dose-response relations which reflect two populations. It was pointed out by Elkind [Elkind 1991] that in the living animal, there are always cells engaged in repair-replication which are therefore highly sensitive to mutation and killing by radiation. It follows that living systems should theoretically exhibit a biphasic dose response resulting from the sequence: high sensitivity mutation-death followed by low sensitivity mutation. Such a dose response is commonly seen in experimental and epidemiological systems e.g. Burlakova’s meta-analysis of leukemia studies [Burlakova 1996] the Chernobyl infants [Gibson et al 1988, Petridou et al 1996 , Michaelis et al 1997, Mangano 1997, Busby and Scott Cato 2000, 2001, 2001a) ,Weinberg’s minisatellite studies of Chernobyl liquidator children (Weinberg et al 2001) the nuclear industry workers and their children (Draper et al 1997, Roman et al 1999). This has implications for epidemiology since, if this is so, then if a dose-response which is not continuously increasing is found in a radiation health study, this is not evidence that the radiation is not causing the health problem.Clearly if it is possible to move quiescent (G0) cells into repair-replication by a sub lethal hit and then subsequently hit the cell in repair-replication this two hit pattern represents an enhancement of hazard over the same dose given at once. That this type of exposure represented an enhanced hazard for mutagenesis was first suggested by Busby and Busby in 1987 in what was termed the ‘Second Event Theory’. There are two main types of such hazard. The first is immobilized sequential decay isotopes like Sr90/Y90 or Tl-132/I-132 [Busby 1995, 1996, 1998,1998b,2000a, 2000c) Cox and Edwards of NRPB recently attempted to show that for the case of Sr90, there is almost but not quite the equivalent probability of effecting a double hit from external radiation at background doses. Despite some erroneous assumptions in their paper, they have nevertheless implicitly shown that for more effective Second Event isotopes like Tl-132, or other more efficient second event sequences than Sr-90/Y-90 the probability is very much higher. Tl-132 was a significant Chernobyl isotope and may be the cause of the anomalous thyroid cancer increases.[Cox and Edwards 2000, Busby 2000c] The second type of Second Event system of interest is the hot particle e.g the immobilized plutonium oxide or uranium oxide (DU) particle [Busby 2000d, 2001b] Tables 3 and 4 indicate the sizes of particles most likely to effect second event enhancements. It is of interest that the most common size of plutonium oxide particle found in the environment, of diameter 1 m, is also the size that can most efficiently deliver two hits to its immediate environment in ten hours.
Table 4 Doses to sphere of tissue 30 micron radius by one particle of PuO2 of various diameters
Assumptions: Plutonium Oxide (Pu239) is in the PuO2 form (density = 11.6); Alpha decay energy = 5.2 MeV; Alpha range = 30 microns. Relative Biological Effectiveness factor for Alphas = 20 (from ICRP) has been used to convert dose in Grays to effective dose in Sieverts
Early formulations of the second event theory assumed that a specific cell, or even cell nucleus had to be sequentially hit. However, recent research has revealed the existence of a cell communication field, whereby a hit to one cell causes predisposition to genetic mutation in nearby cells, the ‘bystander effect’. This has implications for the Second Event theory since the target for the second hit now may be any cell in the field and this target is thus is made very much larger. Interestingly, Sonnenschein and Sato [1999] have drawn attention to the fact that cancer cells do not develop when transplanted into a normal cell matrix whereas normal cells are transformed into cancer cells when transferred into cancer tissue. They argue that this suggests a field effect necessary for cancer promotion. Such a field would be more easily disrupted by high local dose from internal immobilised particles and this would mean that such exposure would be more hazardous than the averaging model predicts.[Zhou et al, 2001]
4.4 TransmutationOne mechanism which is entirely absent from ICRP deliberations results from the effect of the radioactive decay process changing one atom into another. There are three common radioisotopic pollutants where this effect is likely to have serious consequences. They are Carbon-14, Tritium and Sulphur-35. All three are major components of enzyme systems and critical to the processes which are fundamental to living systems.
4.5 Threshold and hormesis The idea of a threshold below which radiation damage is repaired seems plausible and there is evidence for the existence of mechanisms whereby repair systems may be induced by 'priming doses' of radiation. Since repair mechanisms are now well accepted, clearly there must be a point where they become overwhelmed, and above this point, in dose terms, the effect will begin to increase at a greater rate than below it. Furthermore, if, as in all other physiological systems, the repair efficiencies may be increased by acclimatization or induction, it could be argued that low doses of radiation might effect such an increase in repair efficiency, and there is evidence from animal studies that there are such processes. The question therefore is, why do these systems not remain on the highest level of efficiency all the time? The probable answer is that if the cell were on high alert and repair efficiency was always high, there would be a much higher rate of repair replication all the time, and this on its own would result in greater erosion rates and a shorter overall life span of the DNA, with more periods of high sensitivity available for DNA damage from other endogenous or oxidative sources. There is no such thing as a free lunch, and in this case, continuous rates of high repair efficiency (hormesis) would put the organism itself at risk of accelerated aging. It is of interest that the two population biphasic dose response appears to show hormesis, so long as the origin point of no dose, no effect is ignored (Fig3). Population studies, claiming to show hormesis are usually based on ecological comparisons of people living in high Radon areas are unpersuasive since they do not take account of social class confounding. If thresholds exist, then the question becomes, what types of exposure are likely to bypass the systems which give rise to the threshold? And it is this question that results in a more accurate picture of the mechanisms of risk from radiation exposure.
5. Epidemiology 5.1 Epidemiology, fallout and the present cancer epidemic Evidence from Wales suggests that the present cancer epidemic is a consequence of the 1960s exposure to weapons fallout. [Busby 1995, 2000a ] Similar temporal trends exists in other European countries, e.g. Finland. Cancer in Wales increased in a temporal pattern which correlated with the cumulative exposure to Sr90 twenty years earlier. The correlation coefficient is high and the relationship suggests an error in the risk models of 300 times (Fig 2). In addition, breast cancer incidence and mortality exhibits a cohort effect which suggests Strontium-90 as a cause [Busby 1997] (Fig3). The chemical affinity of Sr90 and its high levels in milk in the 1960s are additional factors. There is an intuitively persuasive argument also. Considerable evidence points to cancer originating in somatic genetic damage due to environmental carcinogens. It is mainly an environmental disease. Twins studies indicate that the heritable genetic component is less than 15% (Lichtenstein 2000) without asking about the origin of this damage. The only mutagen which was both released into the global environment twenty years before the present cancer epidemic and which also follows rainfall in its dispersion is anthropogenic radioactivity. I have been able to show that there is evidence that the children born to those who were themselves born over the peak years of global weapons fallout have a higher risk of developing leukemia. The numbers used for this study were small and obtained from a leukemia charity [unpublished] but I show it here in Fig 4. Attempts to obtain the data from ONS have failed.
5.2 Epidemiology and nuclear sites There are statistically significant childhood leukemia excesses at Sellafield, Dounreay and La Hague. Childhood cancer is high near Sellafield [review by Beral et al 1994, Viel and Poubel, 1997]. There are many other nuclear site childhood leukemia clusters e.g. Aldermaston, Kruemmel [Beral et al 1994, Busby and Scott Cato 1997, Hoffman and Greiser 1998] These collectively suggest an error in the Hiroshima risk model of about 300-1000 times based on an aggregate of external and internal dose as calculated by NRPB and COMARE [NRPB 1984,1986,1988,1995, COMARE 1996] The most studied of these sites is Sellafield, which has come to be a representative case for the discussion. Here, there is a large question mark over internal doses from inhalation of radioactive material in the air. The calculations of internal doses made by NRPB and used by COMARE to exonerate the radiation from the leukemia in COMARE IV have not been published nor peer-reviewed [NRPB 1984,1986,1988,1995, COMARE 1996].
5.3 Recent epidemiology of populations living near the Irish SeaA three year study of cancer in Irish Sea coastal populations in Wales and Ireland by Green Audit supported by the Irish government revealed significant excess risk associated with living close to the sea in areas where there was measured radioactive contamination from Sellafield. Using data from Wales Cancer Registry Cancer 1974-89 Standardised Incidence Ratios (SIR) were calculated for 100 small areas in Wales and related to covariates which included proximity to the coast, rainfall, radon, plutonium in soil, plutonium in grass, plutonium in air modelled on Harwell measurements, Carstairs deprivation, Welsh Index of Deprivation and altitude. Temporal and spatial development of variation in risk followed releases from Sellafield and proximity to coastal areas where they concentrated. For all the different cancers studied (with one or two exceptions) there was a sharp and significant excess risk very close to the coast, i.e.within a kilometre or so, particularly in children and particularly in towns in north Wales close to known offshore sediment banks contaminated with radioisotopes. Figs 5 to 8 show the trend for all cancers, breast, lung and colon cancer. For children, brain tumour and leukemia risks in some small coastal areas were extremely high. Fig 9 shows the trend for childhood cancer 0-4 and Fig 10 for brain tumours in children aged 0-9. The trend with distance from the coast was the same as that for sea to land transfer of plutonium particles and also the penetration of salt spray particles inland [Busby 2000b]In a study of Irish National Cancer Registry 1994-1996, small area data was supplied to Green Audit. Standardised Incidence Ratios were calculated for these small areas and aggregates of areas were made into bands by distance from the sea with transects from east to west, west to east and south to north. Results showed significant sea-coast effect on cancer in women of all ages for the east coast but not the south or west coasts. For men, the effect existed in young men but not older men. The trend with distance from the sea on the east coast was similar to that found in Wales. [Busby, Kaleta et al 2000] In order to investigate the effect at high spatial resolution, Green Audit and the Irish group STAD undertook a questionnaire study of Carlingford, Co Louth, cancer cases were mapped in an area where Sellafield pollution was concentrated in offshore sediment. Results showed that the trend in cancer was similar to the trend in sea to land transfer of plutonium. The effect was extremely local, and showed itself in a 100metre band near the contaminated intertidal sediment [Busby and Rowe 2000] The authors concluded that sea to land transfer of plutonium or other material produced airborne particulate radioisotopic pollution which was inhaled or ingested by those living near coasts wher |