Leukemia and lymphoma study recently published in Lancet being strong challenged by SARI
A recent study published in Lancet Haematology claims to show that even extremely low doses of radiation increase the risk of leukemia and lymphoma.
The study includes several statistical flaws,
ignores the effects of medical exposures — which are of similar levels to occupational exposures — that change dramatically over the duration of the study, and avoids a proper accounting for uncertainties in both measured doses and in the expected level of subject diseases in a non-exposed population.
Update – (July 8, 2015) I have received a message from the SARI member who raised the concern about medical exposures. He has retracted that objection following a more detailed explanation of methodology received by direct communication from the study authors. End Update.
The study was conducted with funding from the following organizations:
Centers for Disease Control and Prevention, Ministry of Health, Labour and Welfare of Japan, Institut de Radioprotection et de Sûreté Nucléaire, AREVA, Electricité de France, National Institute for Occupational Safety and Health, US Department of Energy, US Department of Health and Human Services, University of North Carolina, Public Health England.
The list of authors for the study is lengthy and includes some names that Atomic Insights readers may recognize as “the usual suspects” of large scale radiation epidemiology studies.
Dr Klervi Leuraud, PhD,
David B Richardson, PhD,
Prof Elisabeth Cardis, PhD,
Robert D Daniels, PhD,
Michael Gillies, MSc,
Jacqueline A O’Hagan, HNC,
Ghassan B Hamra, PhD,
Richard Haylock, PhD,
Dominique Laurier, PhD,
Monika Moissonnier, BSc,
Mary K Schubauer-Berigan, PhD,
Isabelle Thierry-Chef, PhD,
Ausrele Kesminiene, MD
The challenges to the published study are coming from scientists in radiation health specialties and statisticians, some of whom are members of Scientists for Accurate Radiation Information (SARI). Though the Lancet publication is not one that accepts online comments, an article about the study titled Researchers pin down risks of low-dose radiation has been published in the online version of Nature. That publication provides a commenting capability for registered users who use their real names.
The subtitle of the Nature article provides a strong incentive for immediate action to respond to the well-financed study published in a respected journal by people with good credentials. Here is how the headline writers at Nature summarized the conclusion of the study, which included reconstructed dose histories and evaluation of medical records.
Large study of nuclear workers shows that even tiny doses slightly boost risk of leukaemia.
The study was large, the subjects were nuclear workers, but the conclusion is purposely scary and attention getting. The headline writer is good; the purpose is being served and attention is being paid. I’m not sure that the study authors are going to enjoy the specific kind of attention that they will be getting, but it is time to put modern communications tools to work.
We are not living in the same world as Mark Twain was when he quipped, “A lie can travel half way around the world while the truth is putting on its shoes.”
As of the time this article was published, there are 11 comments on the Nature article; nearly all of them offer solid challenges to the conclusions published in the article and the paper.
More later, but it’s time to pull on our shoes and go to work.
Immediate reactions to this study:
1. Even by their own statistics, there is no significant Excess Relative Risk of any disease, including chronic myeloid leukemia, any dose less than 50 mGy.
2. Two of the four disease categories studied (non-Hodgkins lymphoma and multiple myeloma) had ERRs of less than 1, indicating hormesis effect from low doses. Further, since non-Hodgkins lymphoma is the deadliest of the studied diseases, the overall ERR, from all studied diseases, is also less than 1! (ERR=0.78 from my computation).
So this large study indicates hormesis in the data, but the authors do an excellent job of burying the lede.
You can’t both claim, first with reason that it’s highly likely that finding one disease in the lot which has a significant ERR is just random, but also that finding two that have an ERR below one must be meaningful and demonstrate hormesis. If there’s no effect, then we expect half the results will be below 1.
The Lancet is an (or the most) highly respected peer reviewed scientific journal.
Suggest that you write a letter with this to The Lancet. They will certainly publish it, if your note is scientifically valid.
A publication in The Lancet, even if it’s only a letter, is great on you CV!
Journals such as The Lancet refuse normally more than half of the publications that are offered. Often with the argument that the results are not solid enough.
But they are less critical regarding critical letters as they want to stimulate open discussion, as everybody can learn from that.
But it may be smart to first find an answer yourself to the question:
Why did the peer reviewers (respected scientists in this field) not note it?
It is always good to see a discussion with arguments
What would have been (and could have been) really interesting was to determine the maximum dose rate of studied individuals as well as their total doses. That was a huge missed opportunity.
LNT is based on the (unproven) assumptions that all damage from radiation is (a) cumulative; and (b) irreparable. If that is the case, LNT must be true, and dose rate should have zero impact on radiation effects; only total dose would matter.
On the other hand, if radiation damage can be repaired (as is now known to be true), then there miust be a threshold below which radiation causes no damage. And in that case, there must also be a threshold of dose rate above which repair rates can no longer keep up with the rate of damage.
Thus the presence of a dose rate effect in the data would have been evidence of threshold and evidence against LNT. The authors, through ignorance (or more likely through the social blinders that a pre-conceived theory imposes upon experimental design) missed an opportunity to find something interesting. One wonders if their database is available for other researchers. One also wonders: how many studies exist that show dose-rate effects? How many studies are there that looked for dose-rate effects, and didn’t find any?
I’m not sure that necessarily follows. Repaired damage is damage nonetheless. You seem to assume that absolutely all damage will either be repaired or cleared out without a trace, with about as much a priori proof as those who hold otherwise. Its more complicated than that. I’d try to follow the epidem-illogical data, and try to misinterpret that correctly.
I’ve seen some of the hormesis evidence as well. Hormeis may well be a real effect for some pathologies, and not so much — or at all — for others. In which case we’re back to your first post, trying to assess net relative risk, and comparing that (if any) to the relative risks inherent in the alternatives to nuclear energy.
Prof Bernard Cohen made such an attempt at Understanding Risk. It didn’t really catch on, though one suspects its what Cheryl Rofer was trying to get at when she asked what we should do were LNT to ever be conclusively demonstrated, at least in some context. Not to put all our eggs in the “there’s-always-a-threshold basket”, as it were.
There is no 1:1 relationship between DNA damage and mutation. Oxidative damage is damage too, and activities like running exacerbate it, but the medical profession hasn’t recommended a ban on exercise to prevent cancer. That’s what they would do if LNT was applied uniformly and consistently, though.
Bob Applebaum, please come argue against this. Pretty please.
This is in fact true.
Cell response to radiation damage is non-linear. Below a certain amount of DNA damage, repair mechanisms are not triggered, potentially increasing mutation probability. This is the origin of the sometimes heard antinuclear claim that low-dose radiation is actually *more* dangerous than higher dose radiation, because it could in principle cause DNA damage which remains undetected by cellular repair mechanisms, yet is serious enough to cause mutations. That theory falls down because at such low radiation doses, all DNA damage is caused almost completely by natural processes, so even if the additional damage due to radiation does go undetected by the cell, it does so *because* it is not significantly impacting the DNA any more than natural processes are already doing. Hence, the risk of such doses is effectively zero.
On the other hand, larger-than-low doses can overwhelm repair mechanisms to cause cell death, which precludes mutation obviously, invalidating the 1:1 relationship with mutation at that end of the spectrum as well.
Clearly then, the relationship between radiation dose, DNA damage, and mutation probability is highly non-linear and LNT is clearly false.
In any case., at sufficiently low radiation doses, the DNA damage/mutation probability relationship is unknown and unknowable because a host of other (natural) factors largely or completely unrelated to the dose magnitude determine mutation frequency.
I’m no radiation health specialist, but all of the above would appear to be self-evident. If I’m missing something, I’d love to know what.
Agree that at low doses other processes cause far more DNA damage, but that doesn’t change that low dose radiation also adds damage.
So while 1mSv/a may cause only 1% of the DNA damage in adults compared to the damage by other sources, still that extra 1% implies ~1% more repairs and ~1% more chance for non-repair or faulty repair generating cancer, etc.
For fetuses (high cell division rate), significant health damage increases were shown for radiation increases of ~0.2mSv/a (http://goo.gl/Sh2Wuq) =~10% of background radiation. Chance increase on health damage ~30%/mSv radiation.
LNT does not imply that all damage is cumulative, neither that it is irreparable.
The relation is more complex.
Repair can be partial or none if a radiation particle damages:
– the DNA repair mechanisms
– the energy producing Mitochondria (not enough energy for 100% repair)
– the cell is dividing (=grow). Then DNA is single stranded, so no repair possible.
The hell it doesn’t! That’s the entire point behind the LNT model: each exposure yields risk for cancer, mortality, or morbidity that is proportional to the amount of dose.
What part of “Linear” or “No-Threshold” do you not understand?
Oh … that’s right. You have no background in science, so you just throw around words that you don’t understand.
For the very young (fetuses = fast cell division) LNT is measured (roughly) down to ~0.2mSv/a (http://goo.gl/Sh2Wuq).
But for adults LNT isn’t measured fully below ~50mSv/a.
For adults the relation can also be a concave at low levels (e.g. RERF report 14). That would imply more damage than LNT predicts. But no significance, so we don’t know.
Note that the harm that a certain amount of radiation causes, depend very strongly on age. Much more harm if somebody is younger (=more grow = more frequent cell division).
LNT = damage irreparable? Not correct:
If eg 90% of damage is repaired correctly at all radiation levels, LNT will be measured. The threshold idea implies that repair % jumps to 100% below a radiation level. That contradicts the stochastic nature of biological processes.
Repair can be partial and/or repairs can be less/more stable, repair may take a long time, etc. People in high radiation environments may have improved repair mechanisms, etc (http://goo.gl/JO2jlb).More repair will exhaust the cell earlier. People who have to resist high levels of such attacks die earlier (miners, smokers, etc).
Rod, to me its unbelievable that in the post “anti vax” disaster world this “study” presented as is, would even be a thing. What a mess. There are so many issues here.
The Lancet published the original anti-vaccination paper. Also a paper claiming vast numbers of deaths in Iraq, based on a very sloppy “survey”.
It is not a reputable medical journal. It is a newspaper for doctors.
Please back up your claim that the Lancet report on Iraqi non-combatant deaths was a “sloppy survey”.
And please, don’t quote Limbaugh, Hannity, or some other lyin’ right wing propaganda drooling mouthpiece.
At a guess of 654,965 excess deaths it was certainly contested.
New study estimates 151 000 violent Iraqi deaths since 2003 invasion ( http://www.who.int/mediacentre/news/releases/2008/pr02/en/ ).
Of course beyond the political no one can include the might have beens. Its reasonable to assume, if no invasion Iraq could have been involved in other conflicts, internal and or external where the body count could have gone higher. The Syrian conflict has exceeded 200,000 casualties by some estimates now with half of them being civilians. We had the chance to intervene early on in that and didn’t. But then again if we had things could always also be worse.
Hmmmm, or we could believe GWB, who claimed the amount was 30,000.
Interesting how when someone else is doing the killing, the figures swell, as in Syria, where it is politically expedient for our “leaders” to state high numbers. Despite the fact that we are arming terrorists who will undoubtedly eventually turn to killing american soldiers or civilians when it suits their agenda to do so.
But hey, we did such a great job messing with Iraq, why not turn these Washington geniuses loose on Syria?
My suggestion? Talk to some Syrian refugees who have fled the violence. I have. And its a real eye opener. We are being fed a ration of CRAP about whats going on over there.
Or hey, just let Fox or MSNBC tell ya whats going on. Makes it so much more simple.Just open wide, and swallow.
Bush’s estimate was low ball but valid ( http://www.washingtonpost.com/wp-dyn/content/article/2005/12/17/AR2005121700017.html ) Saddam also did support international terror and kept materials for WMD’s.
But anyway back to the Lancet. I think Don made a valid point bringing that up. And in this study Im not so outraged as most probably, as these types of vague studies come and go, its the firm language they put their “conclusions” in that bothers me most.
No, he did not support international terrorism. In fact, he executed radicals. And had both women and Christians in his cabinet. One of the most despicable untruths that were advanced by the Bush lying machine was their attempt to tie Saddam to 9/11 in order to further justify an invasion based on false pretense.
“Baghdad actively sponsored terrorist groups, providing safe haven, training, arms, and logistical support, requiring in exchange that the groups carry out operations ordered by Baghdad for Saddam’s objectives. Terrorist groups were not permitted to have offices, recruitment, or training facilities or freely use territory under the regime’s direct control without explicit permission from Saddam.”
“Saddam used foreign terrorist groups as an instrument of foreign policy. Groups hosted by Saddam were denied protection if he wanted to improve relations with a neighboring country and encouraged to attack those Saddam wanted to pressure. If they refused Saddam’s “requests,” they were exiled from Iraq ” – 9/11 commission. ( http://www.9-11commission.gov/hearings/hearing3/witness_yaphe.htm )
You gotta be kidding me! You take the testimony of Judith S. Yaphe and present it as gospel??? Why not cite Feith, Cheney, or Judy Miller as the truthtellers if you are going to simply point us at the scriptwriters.
Egads, you have really swallowed the swill, haven’t you, John. Or do you simply peddle it?
Regardless of the merits of the Lancet study, the summary by Alison Abbot at Nature is very strange and would appear to contain numbers of her own calculation. Note this sentence, which is being discussed by one of the commenters there:
“Researchers expected that 134 of the workers (4.3 per 10,000 people) would die from leukaemia as a result of the average 27 years they spent in the industry; in fact, 531 people died from the disease.”
There is no way in hell that the Lancet authors attribute 400 excess leukaemia deaths to occupational radiation exposure.
They did estimate an ERR of 2.96 per Sv.
PER Sv, not per the mean study dosage of ~0.03 Sv (? someone please confirm).
Oddly, 531 versus 134 represents an ERR of 2.96, and I think that Alison Abbot must have constructed one of these numbers somehow. Certainly the study authors did NOT find 400 excess leukaemia deaths, “in fact”. Nor did they “pin down risks” or “scupper” any ideas.
Please see the 2 excellent comments by *Ruediger Meyer*, who seems to have access to the full paper, and who is I’m sure vastly more knowledgeable in this area than most of us here.
I wish I had more information, but I have to wonder if Alison Abbot’s reporting is both accurate and ethical, and I think that maybe someone at Nature should be aware of this glaring factual issue.
Thanks for your support, I thought I was going mad when no-one else picked up on this obvious mistake. Thankfully this has now been corrected. From my side, the lancet paper is open access, just click on the full text tab to see the full version.
I did not realize that the Nature summary article had been corrected.
Good work Ruediger.
I will have a quick look.
Rod, you wrote:
We are not living in the same world as Mark Twain was when he quipped,
“A lie can travel half way around the world while the truth is putting on its shoes.”
While I suspect a typo (“not” instead of “now”), your statement is still accurate because today a lie can travel ALL the way around the world before the truth even gets its shoes out from under the bed.
While you are correct that any utterance can travel around the world almost instantly, what I was trying to point out was that the lie doesn’t have a chance to get cemented into too many minds before challengers can respond. When communications were slower, lies could become accepted truths before a challenger could try to begin the process of dislodging it.
In an ironic twist, that quote actually isn’t attributed to Mark Twain.
Agree. We would think that. But may be wrongly.
Just consider the last Iraq war. Entirely based on lies.
The ultimate lie being the detailed graphics of Saddam’s mobile chemical plants presented by Colin Powells to the UN.
UK’s prime minister, Tony Blair, contributed also:
– “We know that he (Saddam Hussein) has stockpiles of major amounts of chemical and biological weapons…”
None existed. Etc.
And this was regarding an issue (going to war or not) which concerned the lives/deaths of many thousands of people (~100,000 Iraqis died).
May be progress is less than we thought.
I applaud your, and SARI’s, efforts to challenge LNT as a hypothesis. Every hypothesis should be challenged, so that science makes sure to carefully scrutinize it. As for me, I honestly don’t know what to think. I think you guys have provided good arguments that LNT *may* be a bad hypothesis, and that the studies supporting it may be flawed.
But, as a layman, I also recognize I’m not in a good position to judge the merits of either side of the argument.
In the meantime, however, I think it’s important to note that, even if LNT is correct, that doesn’t mean we shouldn’t use nuclear energy, or that it’s too dangerous. We routinely accept dangers. As a society, we use electricity. We use it because it has ENORMOUS benefits to us all. We try to make electricity as safe as possible. But, it’s unarguably true that as long as we use electricity throughout society, every year, there will be some number of adults and children who are electrocuted to death.
Likewise, even if nuclear power usage around the world leads to a very small number of cancers and leukemia, we should still do it, as horrible as that sounds, because the benefits are so profoundly positive.
What other stresses on biological systems or structure are optimized at Zero?
Check out Nassim Taleb, if not reading “AntiFragile”, then some of his Youtube or podcasts. You have a questioning mind. You’ll be fascinated…. Guaranteed.
Yes, “Truth” being potential verification of your true opinion on the matter.
I have read and participated here long enough to know that many of the “truths” expressed here are based in bias, and do not stand up to scrutiny and research. I speak not so much as to the matter of radiation exposures and its risks, or the scientific aspects of NE. Because I am not a scientist. But politically, I have seen some real doozies put forward here, with some of the most active debaters being also the most ignorant of historical and political realities. In criticizing Israel, I have been accused of being an anti-semite by individuals demonstrating a remarkable ignorance of international law, past events involving Israel, or the facts on the ground as they now stand in Gaza and the West Bank. I have even seen, without being challenged, one of the most vocal of the NE defenders here state that any scientist suspected of participating in a nuclear weapons program deserves asassination. Never mind proof or due process, I guess. In turn, I have seen the dishonest participation in polling processes advocated here, in the hopes it would skew the poll towards the desired conclusion.
So, do I come to the conclusion that this is a community of credible participants speaking the gospel about NE and the effects of radiation on the human body? Don’t get me wrong, there are some here I trust, even realizing their personal stake in the argument. But as a whole I have found very little, as a layman, in the way of information I would take to the bank. On subjects such as this particular thread, why, again as a layman, would I not be wise to err on the side of caution if it involves my or a loved one’s health?? If you’ll purposely skew poll results, or rabidly argue a geopolitical issue that you obviously know nothing about, why won’t you bend the truth on topics such as this one?
Scientist, layman, or whatever, there’s one thing that nobody will ever be able to accuse you of: meeting a soap box that you didn’t like.
You are a strange fellow. You rail on about the middle east while hardly ever giving specific facts. Somehow we are to not trust major news sources but to trust you even though you don’t give specifics. Then you express doubt that if other’s do not agree with your statements about the middle east, you cannot trust their statements on the health effects of radiation.
Why should we be concerned about your concerns? You don’t seem to be worried about the actual issues involved in radiation science only in measuring the trustworthiness of the “sources” by how well they perform according to your ethical standards – especially as those relate to the middle east.
Depleted Uranium is a heavy metal with very little radioactivity left in it. Here is a link to the effects of Uranium in an educational format for health workers.
They are assuming all the current agreed science and it points out that the major effect of Uranium is on the kidneys. There is no evidence of cancer formation.
John Tucker quotes 654,965 excess deaths. That would be a very high number indeed. The population of Iraq has been climbing over the past 15 years from about 22 million to about 33 million today. If over 600,000 deaths came from depleted Uranium that means that a much much larger number were exposed to uranium, through eating it in large amounts over a long period of time, or by being shot and having the fragments remain in the body for a long period of time (months to years). A reasonable estimate would mean that about 5 to 6 million people were eating uranium every day and or had been shot with the fragments not being able to be removed. I find this remarkable even in the case of an all out war.
This simple analysis brings a reasonable doubt that over 600,000 people have died from Depleted Uranium in Iraq.
Expressing doubts over numbers that don’t make sense is not right or left wing. It is a simple case of checking.
I am also a “layman” in regards to the science. But I can read and find out if the basic science lines up on one side or the other.
“This simple analysis brings a reasonable doubt that over 600,000 people have died from Depleted Uranium in Iraq”
I rest my case. The Lancet study makes no such claim. Not even close.
You didn’t even bother to do a cursery search on what the Lancet Study claimed, its methodology, or its acceptance generally by geopolitical experts on either side of the issue. So, in short, you are commenting on something you know NOTHING about. Now theres a sure fire way to establish credibility, by golly!
@ POA, Well, I took the number in the above comment and applied some basic logic and simple research. You are very correct that I did not do an exhaustive study of the Lancet Study. No time. I was trusting that John Tucker was quoting a number from the study and reflecting on that comment.
If you are saying that the Lancet Study has different numbers would you mind posting those numbers for this layman who does not have the time to do that type of deep research?
Also, please note that you are trying to argue by undermining my credibility rather than actually discussing if depleted uranium causes illnesses for a large percentage of the population in Iraq.
Like most of us, I cannot research everything. If I limited my conversations to only those things that I understand completely I will never learn.
You remind me a bit of the old linux folks who when asked a question would react with frustration and shout “RTFM!!” Read the Manual!! This was quite a put off when I had read the manual but terms were not defined and examples were not given.
How many deaths does the study claim?
The amount is not the issue when criticizing your comment. The cause of the deaths is. Get back to me when you have armed yourself with a smidgeon of knowledge about it.
So, your comment is pointless. You have accused me of not studying the report when you did not study it. You have pointed out that the cause of death is irrelevant to you. You don’t care if they died from uranium poison. You only care in arguing about the Iraq war. You only care to make sure that everyone knows you think that the war was immoral. Brian is right, you want a soap box.
You have caused me to look this over and study a bit more about depleted uranium and the Lancet study, so I have learned a little. I did not read the whole study but the parts that related to DU.
The 654,965 deaths was the number of overall early than expected deaths from a wide range of causes some more direct than others.
DU is claimed to have poisoned the water supply in Basra. Note that the US military did not use uranium because it would poison people but because it would penetrate armor. Of course penetrating armor has the intention of killing people. The question is if the report is accurate in attributing excess deaths to that uranium contaminating the water supply.
There was one reference in the study to depleted uranium.
“Of particular concern is the cluster of cancers and genetic defects now being reported from Basra. There has been some suggestion that this is associated with depleted uranium (DU) usage in anti-tank weapons by the US Army during the 1991 war. It has subsequently contaminated the ground water in Basra, on which 40% of Basra’s population depends on for drinking water. 27” Page 20 appendix E. Note 27 says, 27 Hirschfield, R. An Arab-American Priest, Depleted Uranium, and Iraq. The Washington Report on Middle Eastern
Affairs. 2005. 24(8): 29-30.
I don’t have access to that report since you need to subscribe to the magazine.
Evidently they are attributing higher incidents of Lukemia and birth defects to depleted uranium.
This is NOT the illness expected from depleted uranium. The link I gave previously showed that kidney failure from heavy metal poisoning is the most expected result. There are no numbers associated with depleted uranium deaths in the Lancet study.
One has to wonder about the mechanism for moving the uranium from the weapons to the water supply. The uranium is alloyed with other metals which would change it’s ability to dissolve in water. If I was making a weapon, I would want to make that a weapon was not going to corrode in water. Thus, the alloy.
So, I question the idea that the depleted uranium weapons contaminated the water supply in Basra and led to excess civilian deaths from Lukemia or birth defect. This sounds a great deal like propaganda rather than science.
“You are a strange fellow. You rail on about the middle east while hardly ever giving specific facts.”
Actually, on many occassions I have provided specifics, particularly when queried about assertions I have made. On numerous occassions I have provided a link or two in order to provide the reader the ability to check my source, and the validity of my claim. Rod seems to favor allowing the query, while disallowing the answer. Is it a strategy Rod employs to make me seem unable to back my claims, or simply an effort to keep the discussion “on topic”? I don’t know, for I have some respect for Rod, and prefer to give him the benefit of the doubt.
I would comment that I see my original request for clarification about what was “sloppy” about the Lancet report has been ignored. Perhaps the individual making the claim had the same depth of knowledge you’ve just demonstrated.
I wonder, noting the daily carnage in Iraq, when we will be able to stop making the tally. We’ll never know the true number, will we? Should we start in ’91, and then include the terrible toll the sanctions wreaked on Iraq’s children and infants?
Hey, its all good, right? In Bush’s historical words; MISSION ACCOMPLISHED!
It wasn’t ignored. The numbers were kookily high. They didn’t even have accurate numbers there as to who was alive in the first place. It was based on a projection from a survey(s). ( https://www.iraqbodycount.org/analysis/beyond/exaggerated-orb/ ). “Even among the generally inexact survey results for deaths in Iraq the “Lancet estimate” was an extreme outlier”
Hey, thats a solid rebuttal! Kookily high! Gotta love it. And the UN report reporting 500,000 infant deaths due to the sanctions? Kookily high again?
Uh, the predictions, made by the proponents of the invasion, of Iraqi oil paying for the adventure, and a conflict lasting mere months? Kookily low? Or hey, maybe “kookily PNAC bullshit”.
The numbers are staggering, no matter which “study” you choose to believe. Give it a few more years, and these lying pukes that launched this debacle will have it down to a few dozen. Gotta love ‘ol Rummie recently saying he never sold it as bringing democracy to Iraq. Guess he figures we’re all too stupid to to google his numerous sales pitches.
Ahh the infant deaths thing. Also a Lancet production:
“As many as 576,000 Iraqi children may have died since the end of the Persian Gulf war because of economic sanctions… ….The results of the survey will appear on Friday in The Lancet, the journal of the British Medical Association.” ( http://www.nytimes.com/1995/12/01/world/iraq-sanctions-kill-children-un-reports.html )
The research was even later retracted ( http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2805%2970470-0/fulltext ).
Not many people know I think because it didn’t fit the political narrative. The argument was being used for removing sanctions and for invasion satisfying multiple and even opposing agendas.
“Ahh the infant deaths thing. Also a Lancet production:”
That is disingenous. The report was not a “Lancet production”. It was a UN report, and Lancet was simply the first to publish it. And the report was not “retracted”. The numbers varied from 1995 to further investigation in 96 and 97. Bringing the original count into question. A researcher surmised the numbers were probably less, somewhere in the middle of the discrepency.
So, how many infant deaths are acceptable to you? Does 200,000 sound better? Is that a better figure, more fitting to your sense of morality, John?
Off topic. The point of this story is that radiation workers who receive a very low does have been found to have a slightly elevated rate of Leukemia and Lymphoma. The issue of depleted uranium in Iraq bears on the broader issue of the impact on health by uranium, thorium and plutonium. But the war in Iraq, it’s motives and results is really a totally different discussion. Interesting but not applicable here.
I would really like to hear your replies about the Lancet Study, that would be interesting and on topic and it sounds like you have read it carefully.
That is the portions of the Lancet Study that deal with deaths from depleted Uranium. 🙂
No, I have not “read it carefully”. Nor do I care to. In my opinion, one death would have been too many, being as how the invasion was marketed by deception. The framers of the sales pitch should be in prison, and would be if we were the nation we claim to be.
But I was not the one making the
assertion about the Lancet Report, and being aware of the report, and the numbers, I was curious about why the assertion was made, and how his conclusion was drawn.
“I would comment that I see my original request for clarification about what was “sloppy” about the Lancet report has been ignored. Perhaps the individual making the claim had the same depth of knowledge you’ve just demonstrated.”
Soap box again. You have not posted a single comment in response to weak science used to claim Lukemia or infant mortality. It is interesting that the population of Iraq has risen steadily on a fairly straight line.
I have no idea how many extra deaths occurred because of the war. Wars are always horrible, terrible and filled with pain. Many innocent die.
But I deeply doubt that there were and are extra deaths from depleted uranium. I am sure from the biology that there are no birth defects from depleted uranium.
All that has happened here is that Rod has shown his honesty and transparency, as he has done on countless occasions before.
Would we see such open and honest corrections on the Anti-Nuclear side?
I doubt it.
Do you really think that redox reactions are so dangerous that they need to be regulated?!!
Are you following the same story JG? There has been at least one MAJOR correction to the Nature article already. There are extremely strong and unanswered questions there. ( http://www.nature.com/news/researchers-pin-down-risks-of-low-dose-radiation-1.17876 )
Nature is at answering questions and following up. The same cant be said for the Lancet.
The Lancet is not a public discussion journal.
But if you have scientific valid arguments against (the results of) the study, than write a letter with those to The Lancet.
They will be happy to publish your letter in their expensive journal together with the response of the authors. I’ve seen numerous of those discussions.
I advice to communicate first with the authors, as they won’t publish a letter which contains arguments that are easily wiped out by the authors.
Anxiety itself is a cause of cancer, as are all things which stress the organism, causing the release of ‘fight or flight’ hormones which turbocharge the metabolism which is the prime cause of cancer in the first place.
Involving people with minute radiation exposure in research which is inspired by the belief that such minute exposures cause cancer would seem likely to increase the anxiety of those persons. It would seem as likely as not that this increased anxiety itself might increase cancer risk by a similarly minute amount as one might expect from LNT. It would therefore seem that ultra low dose cancer research might well be simply confirming that anxiety is a cause of cancer, something we already know to be true.
Especially the LSS of the A-Bomb survivors would seem to be a case in point. It is certainly true that atom bomb survivors have lived their entire lives under a heavy stigma in Japan. At the same time, the latest LSS study results show that there has appeared an astonishing ‘dent’ in the dose response in the low dose region. While this ‘dent’ does not by itself seem to push the dose response into a region where hormesis would become evident, when one supposes that the a-bomb survivors have an elevated cancer risk simply due to the anxiety of being ‘radioactive’ (The Japanese even have a special name for their countrymen who were affected by the explosions: “Hibakusha”), then the ‘dent’ in the dose response results might simply show that radiation hormesis is reducing the anxiety cancer risk.
I wonder if there is any radiation health research which was conducted using subjects who don’t know they have been exposed, and who therefore do not suffer from increased anxiety as a confounder?
Answer to your question:
The women and their fetuses in some areas were not aware that they got more fall-out than the women and their fetuses in other areas.
Still there was an highly significant difference in the frequency of serious malformations, Down, etc of new born between the areas: http://goo.gl/Sh2Wuq
You grossly overstate the conclusions of the study (and not for the first time!) In spite of having this repeatedly explained to you, you keep trotting it out like a broken record.
Let’s have another go.
For those who haven’t already plowed through this report, it performs various statistical gymnastics to show that an apparent hiatus in an overall-falling rate of stillbirths and other birth abnormalities in the years 1986 and 1987, in several European countries (Bavaria + GDR + West Berlin, Denmark, Hungary, Iceland, Latvia, Norway, Poland, and Sweden), is “highly significant” evidence of a link between the Chernobyl accident (in April 1986) and birth abnormality.
(Other “significant” stillbirth-rate anomalies in 1981 and 1993 are hastily passed over without further discussion.)
“Because we apply a rather crude transformation from cesium fallout (density of contamination of 134Cs and 137Cs, in kBq/m2) to effective dose equivalent, in mSv/a, without considering more subtle aspects (for example, the different contribution of indoor/outdoor exposure and internal/external doses), the dose-based relative risk coefficients can be considered ‘preliminary’ “, say the authors.
“Preliminary” is putting it mildly. There’s no discussion at all of those “subtle” aspects, which are actually very important in determining whether and how much the increased fallout could have increased people’s personal exposures to any significant degree. Most people spend most of their time inside these days, after all. Was there any widespread consumption by humans of contaminated food in any of these countries? Not discussed. What about the multitude of other factors that also cause birth abnormalities? Not a mention. What about the possibility of more accurate record-keeping post-Chernobyl giving an apparent increase in the figures? Not discussed. What about stillbirth incidence in areas with high natural radioactivity, how does that compare? What about the increasing exposure to aromatic hydrocarbons, which are known toxins, with increasing car ownership? What about knock-on effects from the collapse of the Soviet Union around the same time? Without a thoroughgoing discussion of the above, the “highly significant” conclusions are completely fatuous.
The authors do admit in conclusion that “causal inference is of course difficult. However, opponents of our methods and findings should bear in mind that the mere possibility of confounding is not a proof of confounding and, even more so, it is not a proof of no effect.”
This is a classic “correlation, therefore causation” fallacy. Yes, the authors manage to calculate impressive-looking confidence values for their chosen variables. Did you know that there is a highly significant correlation (94%) between:
– U.S. per-capita cheese consumption per annum, and
– the number of people who died by becoming tangled in their bedsheets
for the years 2000 – 2009? But without some demonstrated causal mechanism between cheese and bedsheet entanglement, it means nothing at all!
(Other hilarious examples can be found at http://www.tylervigen.com/spurious-correlations ).
And, while we’re on the subject of Chernobyl, take a look at the documentary “Radioactive Wolves”, available on documentaryheaven.com. Thorough investigations of the exclusion zone by a team of eco-scientists a few years ago was unable to discover an ecosystem that was in anything other than in robust good health.
turnages, thank you!
Sorry, you have to wait for my response. I tried to upload it several times. Copied it from windows scratch-pad into the comment field, checked but it still vanishes. Your comment has similar length, you had no such problems?
Another try in (parts) by just typing.
Your points are mainly directed towards the other studies referred in the publication.
As I referred to the study itself, as that is rock-solid, I respond likewise.
The study concerns the frequency of serious congenital malformations, Down, etc (abnormalities) in newborn. in 20 Bavarian districts. Although ~1000miles from Chernobyl, a passing cloud dropped radio-active rain in 10 districts (av. 37 kBq/m2 Cs = ~0.5mSv/a) and no rain in 10 other nearby districts.
In the 10 low-dose districts no changes, while in the 10 ‘high-dose’ districts highly significant increase of abnormalities after Chernobyl
– conversion factor from kBq/m2 to mSv/a
The factor doesn’t change the huge difference between low- and ‘high’-dose districts. Neither the high upward jump of abnormalities after Chernobyl in only the ‘high-dose’ districts.
Your other points in part 2
Your other points:
– more/less indoor
– other toxics
– more accurate record keeping after Chernobyl
apply for ‘high-‘ and low-dose districts the same. And low dose districts showed no increase!
– More stillbirth with high natural radiation? no comparison
More stillbirth in those areas but that implies little as other factors may also cause it. Such as lower quality medical care, etc.
– Collapse USSR
Those districts were in rich West-Germany. No influence at all.
– Consumption contaminated food?
People bought food in the super-market. Only very few eat sometimes local food.
– the precaution statement in the conclusion
A demand of the publisher (asked the author). I assume due to the outstanding results of the study which may have caused remarks by the peer reviewers.
– classic correlation, therefore causation fallacy
Your examples are found by applying many correlation trials. Statistic science deliver rules for those situations. The p value should be lowered with each trial. So then a p<0.01 is no longer significant.
This study is not that type of research. Still the p value was in some relevant test extremely low P<0.00005.
– Exclusion zone
No people so no hunting. Hence many wild animals despite early death due to radiation (mankind also increased while av.age was only 40years).
Studies found typical radiation caused anomalies in the Exclusion zone: goo.gl/pRzC0b
“The (Scherb & Weigelt) study is rock-solid”
No, the study has grave defects. But leaving these aside for now, by the author’s own words, the findings are “preliminary” and conclusions are “postulated”. The most that could be said for it, is it indicates an interesting correlation. You say that the cautious language in the conclusion was demanded by the publisher. Quite right and proper, it’s the job of scientific publishers and reviewers to see that the natural bias of the authors in favour of their own work does not cause them to draw exaggerated and specious conclusions, leaving the publishers with egg on their faces later on. As I said, you overstate the case it makes.
The Nature article was ‘corrected’ in that as the ‘531 excess deaths’ statement was revised down to ‘about 30’ – this study is now well into the statistical noise territory.
Even given that the medical radiation source objection was withdrawn, the downward adjustment of excess deaths – compared with the size of the investigated population – is a definitive explanation that there is no ‘there there’
But I’m sure scientists and researchers will continue to torture data to prove what is basically unprovable…and we’ll be right back where we started: acting as if the LNT is a fully fleshed out scientific fact.
new findings contradict the LNT model http://www.sciencedaily.com/releases/2015/04/150430113159.htm
Thanks, interesting article. This is from the reputable side of the Helmholz Zentrum (Institute of Radiation Biology). There is also the charlatan side of the Helmholz Zentrum (Institute of Computational Biology), where grandees like Hagen Scherb and Christina Voigt look for random anomalies in regional sex ratios, then look for conveniently located local nuclear installations of any kind (no matter how small any emissions from these installations might be) and then “prove” that the evil radiation from the evil installation selective kills unborn girls or boys dependent on the direction of the observed “effect”.
Needless to say, the effect is always “significant”. Bas finds their work highly convincing. Unfortunately some German polititians also do.
The study involved 308,297 nuclear workers from France, the United Kingdom and the United States whose radiation exposure was monitored.All had worked for at least a year. The workers were followed for an average of 27 years. As the cumulative dose increased the risk of dying from some forms of leukemia increased. For each Gray of total radiation the worker’s risk of leukemia increased 3-fold acoording to Reuters -“Long-term low-dose radiation exposure may increase leukemia risk” By kathryn Doyle. July 10, 2015..
Thank you Susanne. There is an error in Reuters’ otherwise good reporting on the Lancet Haematology study:
“In the U.S., the average person’s yearly medical exposure to ionizing radiation in 1982 was 0.5 mGy, but by 2006 had risen to 3 mGy.”
I haven’t yet read the Lancet article, but as average U.S. natural background is 2.5 mGy/yr (3.5 mGy where I live), its hard to imagine Leraud et al. having written anything else.
The study cites 16mGy as the average total cumulative dose across the study’s subjects, 16 mGy x 3 fold risk increase per Gy yields 5% increase in leukemia incidence in the study.
Drawing conclusions for general population is fraught with statistical peril. For example, taking data from NCI/CDC State Cancer Profiles for “leukemia”, “all races”, “both sexes”, “50+” age group, one finds
However, there are extremely wide county-by-county variances within each state, and a USGS Background Gamma-Ray Map shows no obvious correlation of leukemia in the general population vs. background radiation, which trends higher in the western states. Similarly, while a CDC state-by-state map of solid-cancer incidence clearly shows a decreased incidence in the higher-background-radiation western states, it also shows low incidence in some low-background eastern states e.g. Florida, South Carolina, Virginia.
The US population is pretty varied, and there are probably substantial background variations in more than just radiation. Studies in far more uniform populations experiencing much larger background radiation variations in and near Ramsar, Iran (up to 260 mSv/yr), and in Kerala State, India (70 mSv/yr), have not found significant epidemiological differences between the “affected” populations, and controls.
For beta and gamma radiation a Sievert is very roughly equivalent to a Gray. For ingested or inhaled alpha emitters 1 Gray is approximately 40 Sievert. Either Sievert or Gray are pretty large units of exposure, one of either over short period of time will likely be fatal. So we use milli (one thousandth) and micro (one millionth) to maintain a healthy perspective.
Very High Background Radiation Areas of Ramsar Iran: Preliminary Biological Studies (pdf).
What can we Learn From Kerala? Comment by ksparthasarathy, a contributor to the current Nature discussion thread (which I’ve been unable to access the past few days.)
Increased Gamma radiation is one of the factors causing increased chance for cancer.
Others: the smog in LA compared to the clean air in Denver, the other life-style (more movement), in house radon (dependent on building material used, etc), more/less smoking, etc.
The av. background radiation in the high radiation districts in Ramsar is ~9mSv/a:
http://goo.gl/98FPGr The 260mSv/a is an anomaly.
While people living there have increased levels of aberrations in their DNA:
As only a few thousand people live there, present (statistical) tools cannot show increased levels of abnormalities in newborn (and population registes are not that detailed & accurate as in Germany.
As usual people develop resistance (also by selection), so a new high radiation dose delivers less abnormalities in the DNA of these people.
The same way people develop resistance against high levels of arsenic. Those adapted have no problem with a deadly dose of arsenic. The common idea is that this extra resistance shortens your life (if it would prolong life; then let me take arsenic).
Bas are you aware that there is more radiation released from a coal plant than from a NPP? Much more?
If that would be true, then studies should find DNA damage via the increased male/female sex-ratio of newborn. As far as I know that is the most sensitive measurement tool we have.
Increased m/f sex ratio of new born for:
– fathers working at the coal plant. Similar as found for fathers working at nuclear Sellafield (up to 30% more boys):
– fathers living in the surroundings of coal plants. Similar as with nuclear power plants: http://goo.gl/yOJ1kp
I’m not aware of such study results. You?
It’s backed up by thousands of measurements from external (plant fence) radiation detectors at NPPs, triggered by above-warning-level readings… on the upwind sides of the plants, when the plumes from coal-fired plants happened to hit them.
Yes, the coal plants were emitting more radioactivity in their normal operation than the NPPs are allowed to emit period.
Look at TMI on Google Maps (Just south of Harrisburg PA.) about 3 miles south of TMI is a Pulverized Coal Plant, three units. Both the NPP and the coal station are clearly visible in the satellite view. TMI is directly east of Goldsboro, Brunner Island Station is directly west of Bainbridge. When the wind is blowing from the south south east (straight line from coal plant to TMI) towards TMI, the alarms on TMI and some times some of the area monitors (Those inside the protected area) go off. Caused quite a bit od concern during the accident as many of the HP technicians assisting were not aware of the coal plant and many false alarms were raised when the took surface wipes of vehicles, buildings, homes, etc. in the prevailing wind direction and got particulate contamination fro the coal plant. There are several “studies” attributing cancers in the area to the TMI release that completely ignore the fact that the coal station releases more monthly, and has for more than 50 years, than TMI did over the entire course of the accident and recovery. Thus, what they really did was prove Coal plants cause cancer.
Indeed, at the start of the Fukushima incident, radiation detectors on a US Navy vessel could not even show clearly that the increased radiation detected in the air was not simply coming from Chinese coal air pollution.
If you were paying attention, you would realize that the data is so widespread that it has been getting into the popular media for quite a few years.
Not of arsenic, lead, mercury, sulfur oxides, nitrogen oxides, or fine particles. The doses from coal plants are not “small”.
Not when “reasonable” is defined as “well below the region in which hormesis occurs”. It’s like keeping kids suffering from rickets indoors to fight melanoma.
From the Nature Article: “Researchers found that 531 of the workers died from leukaemia during the average 27 years they spent in the industry; the data suggest that 30 of those deaths could be attributed to the radiation. ”
This study purports a definite “strong” linear trend in low radiation dose exposure cancers that supports LNT (?). All I can find is discussions about general numbers – like cancer incidence and cancer at high doses. I wish I could find a breakdown of the groups and things like exposure location, age (increasing age), race, gender and risk factors and/verses higher exposures. Also you would expect cumulative cancer totals to be higher than the general population (when risk normalized). I dont even know if that occurred. It doesnt seem like medical exposures were factored in at all. (?) The excess cancer/exposure rates alone don’t seem linear or all that significant with respect to the large population.
“The finding implies that some cases of leukaemia will even be caused by a high level of natural background radiation” – thats not been shown to be true, even in more complete and rigorous studies.
In the supplemental materials there are general tables that seem fitted to a linear model and a vague line graph with no points plotted. Seriously? Is there more complete link(s) available? What am I missing?
You can ask the authors for the original measurement data. Normally they deliver that to any other scientific researcher.
Realize that it is a soft obligation; they may ask what you want to do with the data, and can refuse if you intent to do unscientific manipulations.
Reason; such manipulations may create a lot of fuzz, and create a lot of work for them to clear things up. Still it may give their research results wrongful a filthy name.
Yes Bas Im sure they will email it all to me. It should have been released with the study. This has profound implications in medicine, biology and space exploration at the very least. Part of everyday decision making and part of fundamental issues in the fields. It shouldnt have been thrown out there skewed and without backup.
Thanks Bas. I do not know what the norm is in your country. But in US and Britain it is not a soft obligation. If a member of the public requests the raw data from a published scientific investigation, the study authors are obliged to oblige under our Freedom of Information Acts.
We went through all this during the (alleged) Climategate scandal six years ago. At that time it was indeed common for climate researchers to publish their filtered and normalized temperature trends and retain their raw data for themselves. The one positive outcome of that exercise was an encouragement from Parliament that raw temperature records be made readily accessible, and that has been done. The bottom line is these were publicly funded studies; they are either defensible or they aren’t. They have very broad public policy implications, and however onerous, a study’s authors must be prepared to defend their results. Please see Climategate and Berkeley Earth Surface Temperature Study for references and a brief summary.
I think we agree. It’s unacceptable that our tax-payers money is spent for research whose results are hided for us.
The National Institutes of Health has released an excellent report on LNT and Hormesis “Radiation Hormesis: The Good, the Bad, and the Ugly” it is available at
Another report condoning LNT is “Atomic bomb health benefits.”
There are others. Also there is a link for a PDF in the top of the right hand column. for these documents. Many more, enough to even convince “Bob.”
The paper is a sales promotion for hormesis. What is that National Institutes?
The papers abstract suggests that Japan accepts hormesis.
But ~3years ago Japan made the limits for radiation in food 5times more strict than those in USA or Europe: http://goo.gl/wnoxGW
As far as I know it is the only country which brought its limits in line with the recommendations of the UN Codex Alimentarius Commission (linked to the UN WHO and FAO).
So it seems to me that Japan does not accept hormesis.
Not strange since the famous RERF studies regarding the atomic bombs, involving also Japanese leading universities, concluded in their most recent report (no.14 from 2012) that analysis delivered no indication of any threshold. So also not for hormesis:
“So also not for hormesis” does not necessarily follow. It does if you believe LNT is already proven under all circumstance and is to be taken literally. But such reasoning is circular.
It is apparent that the Atomic Bomb Survivors Studies and some “threshold or hormesis studies” measure two different things: the first’s subjects were irradiated, at various levels, over a very short period of time, a few seconds to a few minutes. The second, over many years of very low doses. In the latter case one might imagine how early up-regulation of protective mechanisms might protect against later doses.
But such is not the case here, as various claims have been made to have found hormetic effects in Atomic Bomb Survivors Studies, albeit over shorter lifetimes than the RERF Report 14 you link.
Mohan Doss, Brian Egleston, and Samuel Litwin have published Comments on “Studies of the mortality of atomic bomb survivors, report 14, 1950–2003: an overview of cancer and noncancer diseases” Radiat Res. 2012 Sep; 178(3): 244–245, wherein they tactfully suggest the RERF authors might have perhaps a bit too optimistically assumed they had identified and corrected for all possible confounding factors, then show an alternate result wherein such assumption was not made.
I wouldn’t know, but suspect this was not the end of the conversation. I’m personally most concerned with long-term effects of chronic low-level exposure, and how they may affect safe operation of nuclear power plants and public perception thereof. Atomic Bomb Survivor Studies do shed some insight on that issue, and establish some upper bounds. But they do not address long-term chronic exposure directly. Globally, coal is making a disconcertingly robust rebound, so this is likely the defining issue of our time.
You are right that ‘Also not for hormesis’ does not necessary follow. Though it’s likely.
I’m not sure yet how health damaging in the long range, the up regulation of protecting mechanisms is.
You find it with toxics, such as arsenic. People can adapt so much that they can take lethal doses without visible health consequences. Doses that kill non-adapted persons within hours. The issue is whether taking those arsenic doses on a regular schedule will cost that person years (months) of his life?
Similar with mountaineering. If adapted, you stay alive at 5500m (despite air/oxygen pressure being only half). Will living there take years of your life?
Being a mountaineer, the ideas of my friends are that it will indeed take years off (which doesn’t bother much as you get so much back).
Thank you for the info.
It explains a lot.
One should keep in mind that medical exposure consists of x- rays which can vary in energy but are weaker than most gammas. Radon exposure is associated with ingestion of alpha particles. Worker exposure largely measures betas and gamma rays.
Actually inhalation is a better term than ingestion in talking about alpha particles associated with radon.
It’s a pity that the research didn’t also measure the m/f sex-ratio of new born concept ed in the period the father worked at the nuclear facility.
That sex-ratio should be enhanced, implying increased level of DNA damage in those newborn.
As that has shown to be a far more sensitive measurement tool (http://goo.gl/TJWAoy), the found significance levels should be much higher (assuming there is DNA damage).
Yes, all things considered, Im sure that would totally be Lancet material Bas. Perhaps next issue.
Comments are closed.
Recent Comments from our Readers
The Clinton Nuclear Plant also in Illinois was shutdown essentially for almost 2 years before it was taken over by…
Good Podcast – Very informative One thing that was not discussed is how to deal with a particular fear that…
Renewables people are masters in marketing. Unreliable intermittent generators whose output is all over the place, and usually badly correlated…
Looking at their lineup, Westinghouse seems bound and determined to keep Gen IV in its “place” which is apparently the…
So they are developing a scaled down version of the AP1000, which is a scaled up version of the AP600,…