Should Radiation Dose Limits be Relaxed?
23 November 2011
During 2010 and 2011, several newspaper articles claimed that the risks of low levels of ionising radiation have been exaggerated and that radiation limits should be relaxed. These reports have cited, in particular, a retired scientist, Dr Wade Allison, who proposes that there are no observable effects from radiation exposures below a level of 100 mSv or 200 mSv.
In his best selling 2009 book “Bad Science”, Ben Goldacre roundly criticised the many distorted and misleading science articles found in UK newspapers and media. The above newspaper articles and BBC programmes (see below) are further examples of such bad science. But they have created concern and confusion in many people’s minds: it’s necessary to try and set the record straight.
As we will show, these claims are incorrect and misleading. None of the newspaper articles was written by radiation scientists; indeed reputable radiation scientists have notably refrained from supporting these claims. (For a humorous scientific riposte to Dr Wade’s arguments see http://www.kbaverstock.org/Book%20review%20Final.pdf)
One difficulty is that a good grasp of radiation risks requires knowledge of radiation biology (that is, the interaction of radiation with living matter), radiation physics, human metabolism, epidemiology and preferably statistics as well. Relatively few people have this breadth of education and even fewer have experience in all these fields. Dr Allison is at least a (retired) professor of nuclear physics so has experience in one of the above fields. But the journalist George Monbiot, who has also written on radiation risks in the Guardian, has zero academic qualifications in any of the above sciences. Even worse, Mr Monbiot admitted writing his articles on radiation after only a day’s investigation.
Of course, Mr Monbiot is entitled to his opinions: unfortunately, there’s a vast difference between these and the risks of radiation as understood and accepted by radiation scientists throughout the world. The problem is his Guardian column provides a platform to disseminate his unscientific views.
In addition, in recent years the BBC has broadcast three arguably misleading programmes on the health effects of radiation. In my view, this is unacceptable for a public service broadcaster with a duty to be impartial and use a balanced approach.
The first programme (BBC 2 Horizon “Nuclear Nightmare” broadcast on July 13, 2006) was later held by the BBC Complaints Unit to be unbalanced following hundreds of complaints. The other two programmes (the BBC Radio 4 documentary “Fallout – the Legacy of Chernobyl’ broadcast on May 3, 2011 and the BBC2 Horizon documentary “Fukushima – is nuclear power safe?” broadcast on September 14, 2011) also received many complaints and these are now being dealt with by the BBC Trust (the highest complaint level within the BBC) which has not decided the matter as of the date of writing – Nov 18, 2011. An independent riposte to the latter programme was later posted at http://iangoddard.com/fukushima01.html
The public’s concern was picked up by the Financial Times in a balanced article on November 10 http://www.ft.com/cms/s/0/aa0a40ec-0aea-11e1-b62f-00144feabdc0.html#ixzz1dJ6hyjqR . The FT pointed to the absence of scientific and social consensus on radiation risks which acted to undermine the disaster response at Fukushima.
This is a fair point, so how does one answer it? In my view, the only effective way is to address the main arguments used Allison/Monbiot etc and to discuss the scientific evidence on radiation risks in a more balanced way. My discussion below does make a number of simplifications in order to make things more easily understood. However I have tried hard to ensure that, while doing so, I have not misrepresented the underlying science.
A threshold for radiation’s harm?
From the evidence we have, that is from epidemiological studies, animal experiments and biological theory, it looks highly unlikely that there’s a threshold below which there are no adverse effects from radiation. It’s difficult to be absolutely certain at very low doses, but most of the evidence points in the direction of there not being a threshold.
Dr Allison and his followers claim that there is such a threshold, but this is inconsistent with how radiation affects cells and tissues. To understand why, we need to step back and explain some basic science about radiation.
Broadly speaking, radiation has two main effects – one is called ‘probabilistic’ (or ‘stochastic’) and the other ‘deterministic’. Let’s take an example. If your skin were exposed to a fraction of a gray (a unit of absorbed radiation dose), it would result in first-degree skin burns (erythema) ; to more and your skin would start getting second-degree burns; even more would result in third-degree burns (ulceration). That is, the greater the exposure, the worse the burn: this is a deterministic effect.
Probabilistic effects, like cancer, are different. You can’t get a “worse” cancer: you either get it or you don’t. Instead, when you increase the dose, you increase the chance (ie probability) of getting cancer. Conversely, if you decrease the dose, you lower the risk. And here’s the nub – these risks exist all the way down to zero at zero dose. In other words, there can’t be a threshold as even at tiny doses – even below background radiation doses! – a correspondingly small risk of cancer still exists.
How do we know this? Because of the way radiation interacts with our cells. If our body were exposed to an acute small dose of about 10 mGy (milligray or one thousandth of a gray), we can predict from radiation biology that, on average, one radiation track will pass through each cell. Each cell traversal carries with it a risk of initiating a cancer. At a lower dose of 1 mGy, we can predict that a radiation track will traverse on average ~37% of our cells. As we continue to lower the dose, fewer and fewer cells are traversed, but there is never a dose where no cells are traversed, apart from zero added dose.
A safe radiation dose?
The fact that radiation risks decline with dose all the way down to zero has several implications. It means, for example, there is no such thing as a perfectly “safe” dose of radiation. No matter how small the added dose, a low risk remains. It means that TV programmes with the title “Is Nuclear Power Safe?” beg the question as to what do they mean by “safe”?
You may ask- what about “safe limits”? Well, in fact, all radiation limits are essentially judgments by radiation authorities of what exposures they consider acceptable.
The fact that there is such thing as a perfectly “safe” radiation dose is clearly troubling to nuclear enthusiasts like Dr Allison and Mr Monbiot as they go to considerable trouble to deny it. Let’s look at the main arguments they and like-minded people use.
- No evidence at low exposures?
Their first argument is “there is no evidence” of radiation effects below 100 mSv (a unit of effective dose – mSv for short): in some accounts it’s below 200 mSv.
In fact, there’s quite a lot of evidence of effects below these levels. For a start, the Japanese bomb survivors’ data – the major study on which all radiation authorities base current risk estimates – suggests some effects at doses below 100mSv. Recent epidemiological studies also provide indicative evidence for the existence of risks at low doses, for example, the large Europe-wide radon metastudy which found effects at quite low exposures to radon gas which is radioactive.
Probably the most famous study is the Oxford Survey of Childhood Cancer (OSCC) carried out from the 1950s to the 1980s by Dr Alice Stewart at Oxford and Birmingham. She found that when the embryos and fetuses of pregnant women were exposed to low abdominal X-ray doses of 5 to 10 mGy, their subsequent babies were more than twice as likely to be leukaemic. Her findings caused a huge controversy lasting for decades, but they are now accepted as correct. Pregnant women are no longer X-rayed thanks to Dr Stewart’s pioneering findings, though it took a long time for the world’s radiation establishments to accept them.
What do most radiation scientists think? All of the world’s radiation authorities adhere to the linear no-threshold (LNT) model of radiation’s effects: that is, radiation risks decline linearly with decreasing dose all the way to zero without a threshold. I think most scientists – certainly all those that I know – agree with this view. For example, in a famous article in 2003, 15 of the world’s most eminent radiation biologists and epidemiologists presented strong evidence for the linear no-threshold (LNT) model of radiation risks. They also said the LNT model might even underestimate radiation risks at low doses.
It’s true that studies of risks at very low doses are relatively sparse because they need to be very large like the OSCC, the European radon study and Japanese bomb survivors’ study, in order to derive statistically significant findings. This often means international bodies need to commission such studies. The radiation exposures from the 1986 Chernobyl nuclear disaster are large enough but unfortunately the EC and WHO appear to have shied away from funding a study on Chernobyl’s doses and risks http://www.nature.com/news/2011/110930/full/news.2011.565.html.
- What about DNA Repair Mechanisms?
Dr Allison and others who think that radiation risks are overrated, often point to DNA repair as a rationale for minimising radiation’s effects. And it’s true that our cells do have multiple and relatively efficient mechanisms for correcting damage to their DNA. But since thousands of repairs are effected every hour in each cell and since the human body has about 60 trillion cells, misrepairs do occur, and that is where the nub lies. For most solid cancers, it is thought that cancers are monoclonal in origin, ie misrepair only needs to occur in one cell for a cancer to subsequently arise.
- What about background radiation?
Everyone receives about 2 to 3 mSv each year of background radiation from cosmic rays, soil radionuclides and radon gas, etc. It’s a fair bet that most people reading this will think these small doses must be harmless, and even many who consider themselves environmentalists will think this.
But background radiation is harmful: here’s the evidence. Wakeford and others have estimated that about 15% to 20% of naturally-occurring childhood leukaemias are caused by background radiation. Others have suggested that perhaps it’s the cause of 100% of childhood leukaemia.
Although background radiation doses to individuals are small, since everyone is exposed the population dose from background radiation to the UK with 60 million people is not insignificant. In fact, one can do the maths and work it out to be 150,000 person sieverts per year. If we apply a 10% per Sv risk of fatal cancer, then about 15,000 cancer deaths each year are caused by background radiation – about 10% of UK cancer deaths each year. This assumes the linear no threshold model for radiation risks, as used by all radiation authorities for radiation protection purposes.
Also, have you ever wondered why women approaching 40 are advised not to conceive their own babies? This interesting question seems to be rarely addressed in scientific articles. But it’s known that both the viability and numbers of human eggs decline with age, especially after 30 years. It’s likely that one of the main reasons is background radiation. That is, by the time she has reached 40, a woman’s stock of eggs will have received 40 years’ background radiation – including about 40 mGy of gamma radiation. This can result in the number of eggs declining and in the remaining eggs DNA damage mounts up. This means that when older eggs are fertilised, they can result in higher incidences of spontaneous abortions, stillbirths, neonatal deaths or malformed babies.
In addition, some cell studies suggest that the process of ageing is connected with radiation. And when DNA repair genes in laboratory mice are damaged or killed by radiation, the mice undergo rapid ageing and experience advanced senescence. This does not prove that background radiation causes the same in humans, but it’s suggestive evidence. Of course other factors will be involved, but it’s likely that background radiation plays a role.
In any event, we should be careful comparing background radiation with man-made radiation in order to justify higher exposures. This conflates different risks: anthropogenic radiation is subject to social and political processes: background radiation is not. Also, background comparisons are not used to justify exposures to other (eg chemical) carcinogens which are naturally-occurring, such as ozone, dioxins or furans. It would be silly, for example, to say that because carbon monoxide is naturally-occurring it’s OK to relax its maximum concentrations. But that’s effectively what some people are thinking or saying about radiation.
- Are radiation limits too strict?
Since 1967, the public limit for radiation exposures to adults in most Western countries has been 1 mSv per year. Is this too strict?
No, in fact it’s too lax. In European countries, a fatal risk of 1 in 1,000,000 per year is widely used in Government planning. However the ICRP risk of developing fatal cancer from radiation is 1 in 10,000 per mSv – in other words, 100 times more dangerous than the usual accepted risk!
To be fair, in the 1980s and 1990s, the HPA’s predecessor the NRPB tried to reduce the public limit by introducing tighter dose “constraints” of 0.3 then 0.2 then 0.15 mSv per year. In fact, the ICRP’s currently recommended constraint is 0.1 mSv each year. Unfortunately, the UK is unable to implement this recommendation because one of its nuclear facilities (Sellafield in Cumbria) discharges such high levels of radionuclides it cannot meet the 0.1 mSv constraint.
In fact, in the 50 plus years since the 1 mSv limit was introduced there have been many advances in our understanding of radiation’s effects. Most of these advances have indicated that we should tightening our radiation limits – not relaxing them. Let’s look at these new developments.
- The new non-targeted effects of radiation
Until the early 1990s, radiation biologists believed that radiation’s effects were solely due to DNA damage. This caused mutations in the cell’s genetic information which eventually resulted in cancers. These “targeted” (ie aimed at DNA) effects remain the classical explanation for radiation’s cancer-causing effects.
However, scientists now know that there is, IN ADDITION, another range of effects. These are called “non-targeted” as the target is not DNA because the radiation doses causing the new effects are too low to structurally damage DNA. Nevertheless they still cause many changes. They include genomic instability where genetic material collapses 20 to 30 generations after being irradiated, and bystander effects where unexposed cells near exposed cells exhibit the same effects as exposed cells. These new effects have resulted in a paradigm shift in scientists’ views (see some articles in the Box A).
Box A. Non-targeted effects: a paradigm shift
Baverstock K (2000) Radiation-induced genomic instability: a paradigm-breaking phenomenon and its relevance to environmentally induced cancer. Mutation Research 454 (2000) 89–109
Baverstock K and Belyakov OV (2005) Classical radiation biology, the bystander effect and paradigms: a reply. Hum Exp Toxicol 24(10):537–542.
Bridges BA (2001) Radiation and germline mutation at repeat sequences: Are I in the middle of a paradigm shift? Radiat Res 156 (5 Pt 2):631-41.
Hall EJ and Hei TK (2003) Genomic instability and bystander effects. Oncogene vol 22, pp 7032-7042. “Both genomic instability and the bystander effect are phenomena, discovered relatively recently, that result in a paradigm shift in our understanding of radiation biology.”
Matsumoto H, Hamada N, Takahashi A, Kobayashi Y, Ohnishi T. (2007) Vanguards of paradigm shiftin radiation biology: radiation-induced adaptive and bystander responses. J Radiat Res (Tokyo). 48(2):97-106.
Morgan WF (2002) Genomic instability and bystander effects: a paradigm shift in radiation biology? Mil Med. 167(2 Suppl): 44-5.
Waldren CA (2004) Classical radiation biology dogma, bystander effects and paradigm shifts. Hum Exp Toxicol. 23(2):95-100.
Unfortunately, discussion of these effects is restricted to academic journals, and they have never been discussed in the popular media, as far as I’m aware. The result is that the public remains poorly informed about them.
Are the new effects important?
Non-targeted effects are significant for two main reasons. One, because they conflict with the “target theory” of radiation’s effects which used to support to official estimates for radiation risks derived from epidemiology studies. The newer non-targeted effects of radiation do not do this. Second, because these effects occur at very low doses of radiation. In fact, some occur after the passage of a single alpha particle through a cell (resulting in a less than 10 mGy dose to the cell).
Scientists are still unsure whether non-targeted effects have net deleterious effects on humans and whether radiation safety limits need to be tightened. Different views exist: for example, the scientific committee of United Nations Committee on the Effects of Atomic Radiation (UNSCEAR) in 2006 expressed varying views on the matter. See Box B.
BOX B. UNSCEAR 2006 on non-targeted effects
UNSCEAR was set up in 1955 to publish information and research findings on the health effects of ionising radiation. The UNSCEAR committee does not contain representatives of consumer or patient or environmental NGOs.
Volume II of the UNSCEAR’s 2006 report contains Annex C on non-targeted and delayed effects of exposure to ionizing radiation.
It stated that within the scientific community “… considerable disagreement remains concerning any definitive relationship between … non-targeted effects and the observed health effects attributable to radiation” (paragraph 161). However its overall view was “the data currently available do not require changes in radiation risk coefficients” (paragraph 164).
But many comments in Annex C lend little support to this conclusion. For example, in paragraph 158
- “it would seem prudent to consider the implications of non-targeted and delayed effects of radiation exposure when considering models of radiation carcinogenesis, particularly at low doses.”
- “…. it is time to re-examine the concepts of radiation dose and target size.”
- “Many of the indirect effects described indicate that the tissue volume in which detrimental effects of radiation may be observed is larger than the precise volume irradiated. This issue may have important implications for human health.”
- “…models of radiation-induced carcinogenesis should incorporate both direct and indirect effects when evaluating radiation risks.”
Therefore UNSCEAR seems ambivalent on the new effects. This is similar to the ambivalence shown by the International Commission on Radiological Protection (ICRP). Its latest recommendations stated that existing knowledge of these effects was “insufficiently developed for radiation protection purposes”. Nevertheless it tightened its recommended dose constraint from 0.3 mSv to 0.1 mSv per year.
My view is that non-targeted effects are likely to be more important than targeted effects in explaining how radiation acts on the body, and that the Precautionary Principle should be invoked, ie that radiation limits should be tightened perhaps by a factor of 5 or 10 to take these new effects into account.
In conclusion, the evidence is clear that radiation risks are certainly not overrated, and radiation dose limits should not be relaxed.
 I’ve written previously refuting these views – see The risks of nuclear energy are not exaggerated – but given recent pressure from various sources for relaxed limits (see NUCLEONICS WEEK – Oct 27th 2011 “Fukushima accident prompts calls for revision of dose limits”) it is necessary to reiterate the truth about radiation risks as understood by scientists.
 To be fair, BBC Scotland transmitted (to Scotland) a much fairer and sharply contrasting view of radiation risks in its programme “Fallout” on the effects of Chernobyl in October 2011. Also the BBC website http://www.bbc.co.uk/news/health-12722435 contains a relatively good source of information on radiation risks. The word ‘relatively’ is used because the website claims that the “lowest level at which any … increase in cancer risk is clearly evident” is 100 mSv which is incorrect. There is a great deal of evidence of adverse effects well below this level. See text.
 Preston DL, Shimizu Y, Pierce DA, Suyama A, Mabuchi K. Studies of mortality of atomic bomb survivors. Report 13: Solid cancer and noncancer disease mortality: 1950-1997. Radiat Res. 2003 Oct; 160(4):381-407.
 S Darby, D Hill, A Auvinen, J M Barros-Dios, H Baysson, F Bochicchio, H Deo, R Falk, F Forastiere, M Hakama, I Heid, L Kreienbrock, M Kreuzer, F Lagarde, I Mäkeläinen, C Muirhead,W Oberaigner, G Pershagen, A Ruano-Ravina, E Ruosteenoja, A Schaffrath Rosario, M Tirmarche, L TomáBek, E Whitley, H E Wichmann, R Doll. Radon in homes and risk of lung cancer: collaborative analysis of individual data from 13 European case-control studies. BMJ, doi:10.1136/bmj.38308.477650.63 (2005).
 vividly described in “The Woman Who Knew Too Much: Alice Stewart and the Secrets of Radiation” Gayle Green. Univ of Michigan Press 1999. Tellingly, the author could not find a UK publisher for her book although most of the events took place here. Even in 1998, the issue was apparently too controversial.
 David J. Brenner, Richard Doll, Dudley T. Goodhead, Eric J. Hall, Charles E. Lande, John B. Little, Jay H. Lubing, Dale L. Preston, R. Julian Preston, Jerome S. Puskin, Elaine Ron, Rainer K. Sachsk, Jonathan M. Samet, Richard B. Setlow, and Marco Zaider (2003) Cancer risks attributable to low doses of ionizing radiation: Assessing what we really know. PNAS November 25, 2003 vol. 100 no. 24, pp 13761-13766.
 R Wakeford, G M Kendall and M P Little (2009). The proportion of childhood leukaemia incidence in Great Britain that may be caused by natural background ionizing radiation. Leukemia 23, 770-776 (April 2009) | doi:10.1038/leu.2008.342
 T.G. Baker (1963) A quantitative and cytological study of germ cells in human ovaries. Proceedings of the Royal Society of London Vol. 158, No. 972, Oct. 22, 1963.
 Keiji Suzuki, Isao Mori, Yukiko Nakayama, Mana Miyakoda, Seiji Kodama, and Masami Watanabe (2001) Radiation-Induced Senescence-like Growth Arrest Requires TP53 Function but not Telomere Shortening. Radiation Research: January 2001, Vol. 155, No. 1, pp. 248-253.