Talk:Radiation hormesis

What to do about the Kerala study?

The section on the Nair et al. study regarding radiation and cancer risk in Kerala, India grossly misrepresented the study's results. Whoever wrote this section included the negative cancer risk statistic from the abstract, but neglected to include the confidence interval or the following sentence that clarifies that the study shows absolutely no statistically significant relationship, positive or negative, between cancer risk and cumulative radiation dose. I have attempted a quick fix, but the study, correctly interpreted, may not even belong on the page as it has nothing to do with radiation hormesis. Unless it's a good idea to keep it on there as a rebuttal to claims that the study does support it? I'm not sure about whether this would be appropriate as I'm new to WP editing. Any recommendations? Mjwood 26 (talk) 22:04, 23 October 2013 (UTC)

The section should really be removed, the study is flawed/biased to start with. Naturally occurring thorium only emits alpha particles, IE wont break through human skin. No damage is done unless you eat or breath this stuff. The study was also done over a time period of 10.5 years, any kind of exposure can take up to 3 decades to manifest. A very poorly done study — Preceding unsigned comment added by 75.97.126.239 (talk) 10:58, 6 December 2013 (UTC)

LNT model

The article states: "Therefore, the Linear no-threshold model (LNT) continues to be the model generally used by regulatory agencies for human radiation exposure."

Isn't that a bit strong? Denying benefits does not equal using a linear dose-toxicity curve. You could also say that lower doses have less/negligible impact, without claiming there's something like hormesis going on. tijmz (talk) 22:40, 15 October 2011 (UTC)

Seems to me like you're saying the same thing. The linear no-threhold model already does imply that lower doses are less damaging, in direct proportion to how much lower they are. But I agree with your change to the sentence that said, "There is no safe level of radiation exposure." "Safe" is a highly subjective term. Perhaps if it said, "There is no 100% safe level of radiation exposure," but even that is misleading, like saying, "You can never be sure you won't get struck by lightning within the next minute." Fnordware (talk) 06:00, 16 October 2011 (UTC)

For regulations to exposure, the LNT-model is the fail-safe approach. It means that you should not expose people needlessly to radiation. If the model is erroneous, and small doses of radiation are harmless, you have done nothing wrong. Contrarily, if you define a "safe threshold" higher than the dose where damage can occur, you may be liable for punitive damages when people are exposed to dangerous radiation below that threshold.

Even more so, if as a regulatory agent you define a radiation hormesis model ("a little radioactivity is good for you"). So far, there is no conclusive evidence for neither of these assumptions (LNT, hormesis, or NOEL) — Preceding unsigned comment added by Thalb2003 (talk • contribs) 21:13, 1 May 2013 (UTC)

Cohen section

I just added a POV tag to the Cohen section. This study uses an out of the mainstream research design. Even the author writes, "This is quite different from epidemiologists usually study" (p 158), and finds a result that goes against most of the literature. The national research council writes of the design that the problems associated with it make, "such studies essentially meaningless” (as cited in [1]). The claim is very strong and the source is very weak (a journal that accepts about 90% of submissions). I recommend either (a) removing the subsection or (b) finding a secondary source for the claims. 0¹⁸ (talk) 16:05, 17 March 2010 (UTC)

Cohen's gone, Sanders has gone. Soon people reading this will wonder why this this theory ever flew. --Old Moonraker (talk) 08:12, 24 March 2010 (UTC)

I think the point is the primary literature should not be the sole source for a controversial section. I would say these authors should be cited as prominent adherents but I think the French government is more prominent. 0¹⁸ (talk) 13:29, 24 March 2010 (UTC)

You link to one of the few mainstream scientific sources (BEIR VII) that uncritically supports LNT, and it is simply wrong. All other sources I've read either say that hormesis or threshold response is more probable than LNT, or that LNT is known to be overly conservative but the alternative hypotheses are not adequately supported yet.

Saying that LNT cannot be disproven using ecological studies is bogus, because the ecological fallacy is the basic assumption of LNT. Confounders are also not the answer, because Cohen analyzed 320 different confounders and combinations of them, and none of them caused the discrepancy with LNT to go away. The removed text (and the article itself) made it clear that an ecological study cannot be used to determine the actual dose-response relationship, but *can* be used to disprove LNT, which is something entirely different. Screaming "confounders!!" and "ecological fallacy!!" at Cohen's papers is just ignorance. --Tweenk (talk) 19:47, 4 August 2010 (UTC)

Tweenk, you have correctly identified the argument uniquely advanced by Cohen. Have you read the NAS report? If so, what did you make of their response to studies of cells? 0¹⁸ (talk) 02:20, 6 August 2010 (UTC)

no such thing?

Here is an interesting point about tracks per cell [2]. Several papers on low dose response, including around 1 track per cell dose rates. 0¹⁸ (talk) 18:36, 23 November 2010 (UTC)

Journal of American Physicians and Surgeons

The Association_of_American_Physicians_and_Surgeons#Journal_of_American_Physicians_and_Surgeons is not a reliable source. Click the link. Because of this, I removed the link to it. 0¹⁸ (talk) 18:11, 17 July 2011 (UTC)

BIER too technical

I don't think the BIER report quotes are too technical. They are about as technical as the French study's results that are reported. If you can find less technical quotes that say the same thing, or want to summarize them, that might be okay, but outright removing them does not make sense. 0¹⁸ (talk) 18:14, 17 July 2011 (UTC)

DuPort, University of Ottawa

This addition notes that Philippe DuPort, University of Ottawa, received research grants from the electricity generating industry (not specifically "the nuclear industry", although that's possibly no more than a quibble) but I couldn't find any confirmation that the Académie des Sciences also received such money. Have two references been confused or, equally possibly, am I missing something? Furthermore, it could be seen as a bit selective in noting only the grants from electricity generators, while excluding any mention of the sponsorship the US Department of Energy gave to the project.

WRT this conversation: this is another page with rather a ragged history that would benefit from a broad, fresh overview if anybody had the time and expertise. --Old Moonraker (talk) 08:51, 4 December 2011 (UTC)

I am sorry if the way I formulated this sentence was misleading, and I will obviously correct my choice of words to avoid such amibuity.

You are correct in pointing out that refering to the "Nuclear Industry" is an oversimplification. As you have probably read already, Duport's paper acknowledges his funding as follows: This work was supported by grants from Electricite de France, the Central Research Institute of Electric Power Industry, the US Department of Energy, COGEMA Resources Inc and MDS Nordion. Although most of them are strongly linked to nuclear power generation (e.g. Cogema), I agree that referring to them as the "electricity generating industry" is far more neutral (article rewording has been done).

I never meant to imply that the Academies ever received financial sponsorship from the nuclear lobbies:

• The Academies' joint report was written by a comittee that comprised the following authors: André Aurengo (reporter), Dietrich Averbeck, André Bonnin, Bernard Le Guen, Roland Masse, Roger Monier, Maurice Tubiana (chairman), Alain-Jacques Valleron, Florent de Vathaire.
• A number of these individuals have links with the electricity generating and/or nuclear industries in France. For example, among the two main authors, Professor Aurengo is connected with Electricité de France (Executive profile and Biography, Bloomberg Businessweek) whereas Maurice Tubiana (chairman) is connected with Areva (AREVA Science and Ethics Committee, AREVA Launches Construction of the "Maurice Tubiana Laboratory” to Produce Lead-212). Howver, those are individual links and do not involve the Academies as a whole.
• There are a few other similar connections. The only writer whose choice of inclusion seems somewhat dubious is Bernard Le Guen: although he is a valued scientist and belongs to the French Society for Radioprotection, he does not belong to any of the Academies and is directly employed by Electricite de France. Still, even this slightly borderline case only encompases a single individual and does not by any means involve the whole of the Academies.

At the end of the day, I do not believe that the Academies are subsidized by the nuclear industry, and it's only by a poor choice of words that I mistakingly implied such a thing. I have just changed the notice into "Quoting results from a litterature database research sponsored by the electricity generating industry". Please feel free to correct me if you feel there is yet some ambiguity or non-neutral implications.

As for going into some in-depth rewriting, I am sorry I can't commit to anything right now. For one thing, I don't think I have that kind of competence yet: I still need to learn more about the wikipedian writing style and guidelines, and I definitely need to regain some of my lost English writing skills. Above all, I still have a wagonload of litterature to read before I can feel confortable doing a full review of an article on this topic. But I will sure go for it (and probably ask for your help) when I have gained sufficient mastership of the area.

In any case, thanks for having spotted these issues, and for your patience with regards to my beginner's errings. Regards, ConradMayhew (talk) 11:39, 4 December 2011 (UTC)

Edit of financial sponsorship of the Duport's report

Regarding the reports' claim that 40% of laboratory studies show radiation hormesis, I had first written that it was "Based on research sponsored by the nuclear industry", then, following comments by another contributor (see previous section in this talk page) I subsequently edited it into to "Quoting results from a litterature database research sponsored by the electricity generating industry". This sentence has now been edited into "Quoting results from a literature research", the sponsorship by the industry being considered as "weasel words" by a contributor.

Well, the quoted Duport's paper clearly acknowledge the grants that allowed for the creation of this litterature database:

This work was supported by grants from Electricite de France, the Central Research Institute of Electric Power Industry, the US Department of Energy, COGEMA Resources Inc and MDS Nordion.

So does the project page from the International Centre for Low Dose Radiation Research at University of Ottawa where this database was established[3]:

The project is jointly funded by the U.S. Department of Energy (USA); Electricité de France (France); Cogema Resources, Inc. (Canada); Central Research Institute of the Electric Power Industry (CRIEPI) (Japan); Candu Owners Group (COG) (Canada); the Canadian Nuclear Society (Canada); and MDS Nordion (Canada).

This is known as full disclosure, and actually rather shows great probity from Duport. I thus have no will to use "weasel words". Still, Duport discloses these information for a reason: they show potential conflicts of interest, and it is now considered standard practice within the scientific litterature that the readers should be made aware of the potential spin such conflicts of interest can induce. I don't see why the same would not hold true for Wikipedia. Wikipedia readers don't need to know about potential conflicts of interest? Why? ConradMayhew (talk) 07:32, 6 December 2011 (UTC)

Edit of reports quoted by the French Academies

A paper by Calabrese EJ has replaced Duport's report as the basis for the following sentence regarding the French academies' report:

Quoting results from a literature research, they furthermore point out that approximately 40% of laboratory studies on cell cultures and animals report some degree of radiobiological hormesis

In fact, the French academies report actually uses both sources, as seen in the following quotations:

Indeed, a meta-analysis of experimental animal data shows that in 40% of these studies there is a decrease in the incidence of spontaneous cancers in animals after low doses.

This has been observed in approx. 40% of toxicological studies [Calabrese 2004], i.e. a proportion similar to that observed in Duport’s meta-analysis [Duport 2003] concerning experimental radiocarcinogenesis.

Among the experimental studies in which the incidence of cancer was sufficiently high in control animals, a reduction of this incidence was observed following low dose irradiation in 40% of them, an observation which is consistent with the concept of hormesis.

In animal, not only does a threshold seem to exist, but also in 40% of experiments, there is even a hormesis [Duport 2003].

Hormesis has been reported in 40% of the animal experiments [Duport 2003]

However, Calabrese's paper is directly used only for chemical hormesis (which he studied) : he was not used as a direct basis for any claim regarding radiological hormesis since he did not study it (or very little) as far as I know.

Still, the report indeed mixes both sources so much that it becomes hard to distinguish them (it's actually a methodological flaw, but that's their problem): I'll include both references together. ConradMayhew (talk) 08:01, 6 December 2011 (UTC)

Duport's work is clearly relevant and should remain as a source. As his "declaration of interest" is standard in academic papers the "sponsored by the electricity industry" perhaps had too much prominence in the article text. While factual, the phrase certainly had a weasely effect, "softening the force of a potentially loaded or otherwise controversial statement", as the article has it.--Old Moonraker (talk) 08:54, 6 December 2011 (UTC)

Old Moonraker, you might want to reread your weasel words link, it pretty clearly doesn't apply here. I think the question is if this fact should be reported, and I'd look for a guideline on reporting author sponsorships. 0¹⁸ (talk) 14:20, 6 December 2011 (UTC)

Sorry, is the link wrong, or are you suggesting that the text doesn't apply? I've checked the link... --Old Moonraker (talk) 15:08, 6 December 2011 (UTC)

What text, exactly, did you think used weasel words? 0¹⁸ (talk) 18:11, 6 December 2011 (UTC)

The article has now moved on: it was here. My point is that DuPort should (a) be in and (b) shouldn't necessarily be caveated by weaselly qualifiers, although I'm sure that could be fixed in a way acceptable to contributors. --Old Moonraker (talk) 18:30, 6 December 2011 (UTC)

Okay, I don't see anything like, "some suggest" or "it has been argued" there... no weasel words. 0¹⁸ (talk) 19:44, 6 December 2011 (UTC)

Thanks to you both for your comments. Following today's discussions:

• I've reintroduced the two references ([Calabrese 2004] and [Duport 2003]) on which the French Academies' report based their 40% figure.
• The [Calabrese and Baldwyn 2001] paper does not appear whatsoever among the 306 citations of the Academies' report; this erroneous citation has been removed.
• I still don't know whether to and how to properly mention Duport's sponsorship. I'll browse through the Wikipedia help pages and forums to see if there are some guidelines on the topic (any help appreciated!). In the meanwhile, I've removed any such mention that could be considered as inappropriate.

ConradMayhew (talk) 22:28, 6 December 2011 (UTC)

WP:RSN may help re sponsorship. Their specific interest is the reliability of sources per se, but they may welcome a change in diet!--Old Moonraker (talk) 07:45, 7 December 2011 (UTC)

Thanks for the tip, I'll post a question there, ConradMayhew (talk) 21:37, 7 December 2011 (UTC)

I'm happy with the included references. I just changed it so say the 40% figure refers to chemical and radiation hormesis and reduced the certainty by changing a few words; "stated" to "claimed" and "report" to "indicated". The inclusion of Calabrese may go some way to ally worries over the veracity of Duport, given both studies came to similar conclusions. I think any remaining uncertainty can be expressed by careful wording. If the Duport reference is lost following WP:RSN feedback, it may well alter the article quite a bit, I would be unhappy to see that happen. --Diamonddavej (talk) 06:17, 8 December 2011 (UTC)

These changes are fine by for me just as well. No problem regarding the citation of [Calabrese 2004], since the authors effectively refered very often to this Calabreses' work (probably for the very same reasons as you).

I don't think you need to worry about the feeback from WP:RSN. The only questions are "Should Duport's funding be mentioned or not? If yes, how should it be done?" That's about it. Whatever funding Duport had, it doesn't change the fact that the Academies' report 40% figure is based on his study, so just as Old Moonraker and you, I can't see any valid reason for removing this reference. ConradMayhew (talk) 20:19, 8 December 2011 (UTC)

2011 United States Department of Energy Lawrence Berkeley National Laboratory

If I understand correctly, the study by Bissell and others shows non-linearity in harm, but does not specifically support any benefits of radiation. In context readers will misunderstand it to be in support of hormesis, which it is not. I think therefore that it does not belong in the article (though it certainly fits in the LNT article.) sbump (talk) 14:02, 23 December 2011 (UTC)

I second the comment. The inclusion of this work here wrongly makes believe that the authors support the hormesis hypothesis.ConradMayhew (talk) 10:31, 24 December 2011 (UTC)

Consider these two statements (bold is my addition) from the hormesis article - In the intro: "Radiation hormesis (also called radiation homeostasis) is the hypothesis that low doses of ionizing radiation (within the region and just above natural background levels) are beneficial, stimulating the activation of repair mechanisms that protect against disease, that are not activated in absence of ionizing radiation." From the USDOE Berkeley lab study (quoting Dr. Bissell): "Our data show that at lower doses of ionizing radiation, DNA repair mechanisms work much better than at higher doses. This non-linear DNA damage response casts doubt on the general assumption that any amount of ionizing radiation is harmful and additive." This study certainly has a relationship to hormesis, though it may not be direct. It is well sourced and comes from a reputable peer reviewed paper. PRONIZ (talk) 11:49, 24 December 2011 (UTC)

There is no doubt that the paper is well-sourced. However, I disagree with the connection you're putting there, as you are taking words out of their context. Although definitions vary between authors, the core lemmas of the radiation hormesis hypothesis are that:

• Radiation protection mechanisms may overshoot at low dose and/or low dose rates,
• Low dose and/or low dose rates may thus prove be beneficial to health.

According to your own quotation from Bissel's paper, she only claims to have observed better repair efficiency at low dose rates. So low-dose rates are less harmful that elevated dose rates. I don't see any response overshoot or health benefits there.ConradMayhew (talk) 14:13, 24 December 2011 (UTC)

Kerala's monazite sand

Thorium occurs all over the earth's surface in common rocks such as granite, so it is incorrect to say that Kerala has 1/3 of the world's thorium. Rather it is generally considered to have around 1/3 of the world's economically recoverable thorium reserves. See Thorium#Reserve_estimates.CharlesHBennett (talk) 20:25, 7 March 2013 (UTC)

1.5 times smaller, 20 times lower....

I never understand these terms. How can something be 20x smaller? I mean a box can be 20x larger. A box could be 20% smaller. but nothing can be 20 times lower. I read the article that was cited the words used were "3.5 deaths per 100,000 person-years–only 3 percent of the rate (i.e., 116) expected for the general population". from that we could say "the death rate was 3% of the expected" or "there was a 97% decrease in cancer related mortality. Can we correct these confusing terms so useful information can be conveyed.Mantion (talk) 06:52, 3 July 2014 (UTC)

I corrected the "20x" comment for the reasons given in the edit summary. However, in everyday English the phrase "20 times lower than expected" would be readily understood as meaning one twentieth of what was expected. But it would not have been wrong for you to change it from the former to the latter for clarity, if you'd wanted to. I left the "1.5 times smaller/less" language, changing it only to use the actual language of the study cited. In ordinary English, that will normally be understood to mean two thirds as much. I agree that might confuse some, but that's the language of the study. Unless you can examine the actual data of the study to confirm exactly what was meant, you shouldn't change it. - Embram (talk) 14:36, 3 July 2014 (UTC)

"The BEIR-VII report argued that, "the presence of a true dose threshold demands totally error-free DNA damage response and repair.""

The above quote doesn't make sense in the context of neither LNT nor hormesis. On LNT, the DNA repair doesn't have to be perfect for LNT to be invalid; if it varies but a little with dose rate, LNT is invalid. On hormesis, if DNA repair is not perfect, there can still be a definate real threshold, namely the point where hormetic effects are equal to damage effects. That's the whole premise of hormesis. Other than DNA repair there is also meiosis and apotosis which highly likely also varies in effectiveness with dose rate (as any bodily process does).

The BEIR argument seems to miss the entire point about hormesis, and it doesn't help defend LNT either.

Siphon06 (talk) 20:51, 24 August 2014 (UTC)

low dose rate vs high dose rate

The quote on the dogs study mentions 75 mGy/d as a low dose rate.

Let us be clear on this. 75 mGy/d is NOT a low dose rate, by any standard. It is over 27 Sieverts per year even if only external gamma radiation.

Low dose rates, per hormesis science, is generally regarded as below 2 mSv/day.

Siphon06 (talk) 20:56, 24 August 2014 (UTC)

Another source

The conclusion of https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2477708/ may be worth adding to the text, citing the article of course.

The article is already well reverenced but one more won't hurt. Andrewa (talk) 02:43, 21 August 2022 (UTC)

• I know this isn't saying much since I have not actively edited this article (so feel free to say "SOFIXIT") but with a topic like this we ought to defer to reviews and meta-analyses. For instance the above article is cited in the course of this review ( don't know if it is paywalled. if it is I can send you a copy) which better contextualizes it. This topic isn't under MEDRS but it is so complex and so defined by uncertainty and difficulty in assessment that we want to be careful about using a single study for claims in text. Protonk (talk) 21:13, 22 August 2022 (UTC)