Is a little bit of ionizing radiation good for you?

April 20, 2016 4:02 PM

Paracelsus, the father of modern toxicology, held that "The dose makes the poison.” Substances considered toxic are harmless in small doses, and conversely an ordinarily harmless substance can be deadly if over-consumed. Going a step beyond Paracelsus, hormesis is the idea that small doses of ordinarily harmful stressors actually improve a variety of outcomes by stimulating defense or repair mechanisms. Below a certain dose the stressor acts beneficially, there is a threshhold dose where the stressor has no net effect, and past that point the net effect is deleterious.

This general relationship shows up all over the place. Small amounts of cadmium reduce cases of testicular cancer in rats, large amounts virtually guarantee cancer. A little penicillin can encourage bacterial growth, a lot wipes bacteria out. A glass red wine a night is good for your heart, a magnum a night gives you cirrhosis of the liver.

The idea is straightforward, the complications are not. Identifying the ranges of hormetic doses versus deleterious doses is difficult, especially given potentially different end points. An exposure might reduce one negative outcome, but increase another, intercomparison may be difficult or impossible.

For all but the largest doses of ionizing radiation, the worry is an increased risk of cancer. There is solid evidence that doses of more than about 10,000 mrem result in measurable increases in cancer risk. Much below that the situation is more or less completely uncertain and it is not possible to pick out the signal from the noise. There is a huge scientific literature, much of which is linked from relevant wikipedia pages based on epidemiological studies of areas with high natural radiation dose, in-vitro cell studies in labs, studies of nuclear industry employees, survivors of Hiroshima and Nagasaki as well as animal data and other human cohorts, but the data is not conclusive. There is a reasonable amount of data finding for radiation hormesis, and a reasonable amount of data opposing it.

The 2005 National Research Council’s Biological Effects of Ionizing Radiation (BEIR) VII report concluded: “The possibility that low doses of radiation may have beneficial effects (a phenomenon often referred to as “hormesis”) has been the subject of considerable debate. Evidence for hormetic effects was reviewed, with emphasis on material published since the 1990 BEIR V study on the health effects of exposure to low levels of ionizing radiation. Although examples of apparent stimulatory or protective effects can be found in cellular and animal biology, the preponderance of available experimental information does not support the contention that low levels of ionizing radiation have a beneficial effect. The mechanism of any such possible effect remains obscure. At this time, the assumption that any stimulatory hormetic effects from low doses of ionizing radiation will have a significant health benefit to humans that exceeds potential detrimental effects from radiation exposure at the same dose is unwarranted.”

The response of regulatory bodies across the world has been the Linear No-Threshold Model, though France has recently said they don’t think it’s true, but is good enough for regulatory purposes. The LNT model asserts that small doses of radiation are as proportionately harmful as large ones. Even though those small doses could be proportionately more harmful, proportionately less harmful, not harmful at all, or even good for you.

To put some numbers around things: cancer radiotherapy delivers doses of perhaps 6 million mrem to tumors, in hopes of destroying them. A whole body dose of about 500,000 mrem is likely to kill a person in a matter of weeks. In Ramsar Iran the natural background dose is around 20,000 mrem per year, which is roughly where evidence of increases in human cancer risk can be seen, although studies in Ramsar have had mixed results. The regulatory limit for works in the US is 5,000 mrem per year and the average natural background exposure in the US is 300 mrem per year. We regulate at levels like 4 mrem/y for drinking water, 100 mrem/y as acceptable dose to the public from nuclear facilities, and 25 mrem/y for cleanup criteria of sites being released. Some medical numbers: a chest x-ray results in a dose of ~ 10 mrem, a full body spiral CT scan could result in a 10,000 mrem dose. A round trip NYC-London flight might result in a 10 mrem dose. The takeaway here is we regulate at low levels compared to where we can demonstrate radiation is harmful and compared to the background levels of radiation exposure due to primordial radionuclides in the earth's crust.

The Nuclear Regulatory Commission is currently taking comments on a petition to reject the Linear No-Threshhold model and the ALARA (keep doses As Low As Reasonably Achievable) model that goes along with it in favor of a hormesis model. Public comments are being accepted.

This would be a seismic change in the regulation of the nuclear power industry, the Department of Energy National Labs, and medical and industrial uses of radioactive material or radiation generating devices. Virtually all occupational radiation protection is done as levels that would be considered in or near the hormetic range. Some hormesis proponents even use terms like: “a subclinical deficiency of ionizing radiation,” implying that while it may not be required, it might be advisable for many people in areas of the country with less uranium ore in the ground to stand in front of an x-ray machine now and then.

Also, this isn’t really going to happen (probably).

And it wouldn’t be radiation related if there weren’t an assortment of people off the deep end. On one hand you’ve got the usual anti-nuclear groups and on the other you’ve got hormesis advocates attributing the lack of acceptance to radiation professionals conspiring to keep their jobs.

Finally we have the most powerful arguments for radiation hormesis.

posted by Across the pale parabola of joy (30 comments total)

There was a king reigned in the East:
There, when kings will sit to feast,
They get their fill before they think
With poisoned meat and poisoned drink.
He gathered all that springs to birth
From the many-venomed earth;
First a little, thence to more,
He sampled all her killing store;
And easy, smiling, seasoned sound,
Sate the king when healths went round.
They put arsenic in his meat
And stared aghast to watch him eat;
They poured strychnine in his cup
And shook to see him drink it up:
They shook, they stared as white’s their shirt:
Them it was their poison hurt.
—I tell the tale that I heard told.
Mithridates, he died old.
posted by leotrotsky at 4:08 PM on April 20, 2016 [10 favorites]

I've spent years building up a resistance.
posted by Greg_Ace at 4:33 PM on April 20, 2016 [5 favorites]

Damn, I got the quote wrong. It's "I've spent the last few years building up an immunity."

Pop culture fail.
posted by Greg_Ace at 4:37 PM on April 20, 2016 [8 favorites]

Inconceivable!
posted by leotrotsky at 4:39 PM on April 20, 2016 [5 favorites]

We do have some knowledge of a mechanism that would cause this effect for lower-energy UV radiation: UV radiation causes DNA damage, while the violet light it's often found with promotes production of the particularly efficient DNA repair photolyases. James Watson talked about DNA damage experiments that his colleagues did which were giving seemingly crazy results until someone figured out that the petri dishes on the top of the pile were getting more light, and this was kicking in their (up to that time unknown) light-activated DNA repair mechanisms.

Placental mammals, though, including humans, don't have photolyases.

One intriguing thing we do have, though, is increased vitamin D production in response to UV radiation, along with a possible link to lower cancer rates for those with higher vitamin D production.

But that is all somewhat beside the point, because - so far as I know - no-one has found a mechanism by which high energy ionizing radiation kicks a DNA repair mechanism into gear the way that lower energy UV radiation does.
posted by clawsoon at 4:42 PM on April 20, 2016 [4 favorites]

*picks up vial, sniffs*
Alpha Particles... I'd bet my life on it.
posted by Mr.Encyclopedia at 4:56 PM on April 20, 2016 [3 favorites]

This is great news for my thermal neutron spa / spent fuel storage pool business plan.
posted by [expletive deleted] at 5:17 PM on April 20, 2016 [7 favorites]

Radium springs are a thing, though.
posted by mccarty.tim at 5:55 PM on April 20, 2016 [1 favorite]

Previously, from the hormesis tag: The Radium Water Worked Fine Until His Jaw Came Off
posted by XMLicious at 6:08 PM on April 20, 2016 [1 favorite]

DNA damage caused by ionizing radiation can trigger various repair mechanisms in human cells, but some of these might actually contribute to cancer:

It has long been known that radiation can induce a broad spectrum of DNA lesions including damage to nucleotide bases, cross-linking, and DNA single- and double-strand breaks (DSBs). DSBs were originally assumed to be the critical cytotoxic lesions, whereas base damage particularly thymine glycols were implicated in mutagenesis. It is now accepted, however, that misrepaired DSBs are the principle lesions of importance in the induction of both chromosomal abnormalities and gene mutations (6,7).
...
While DSBs can be repaired by homologous recombination in an error-free manner (13), this mechanism is apparently a fairly uncommon one in mammalian cells. Most DSBs appear to be repaired by an illegitimate recombination process that is error-prone and thus likely accounts for many of the potentially mutagenic DNA lesions induced by radiation.

In other words, ionizing radiation causes double-strand breaks in DNA, which in turn triggers a repair mechanism (homologous recombination), but the process is sloppy enough that it occasionally turns damaged cells into cancerous cells.

So yeah, probably best to avoid unnecessary exposure to ionizing radiation for now.
posted by dephlogisticated at 6:21 PM on April 20, 2016 [3 favorites]

I've spent years building up a resistance.
posted by Greg_Ace

Damn, I got the quote wrong. It's "I've spent the last few years building up an immunity."
Pop culture fail.
posted by Greg_Ace

UNIMAGINABLE!
posted by DoctorFedora at 7:08 PM on April 20, 2016 [27 favorites]

Ra-di-a-tion. Yes, indeed. You hear the most outrageous lies about it. Half-baked goggle-box do-gooders telling everybody it's bad for you. Pernicious nonsense. Everybody could stand a hundred chest X-rays a year. They ought to have them, too.
posted by ZenMasterThis at 7:20 PM on April 20, 2016 [8 favorites]

Clawsoon, here is a link to a LLNL guy's pretty good presentation on adaptive response, about half way in it gets to the details of the repair mechanism. Adaptive response is a general search term to pull up information on the subject. Bystander effects and abscopal effects are other avenues for hormetic effects. Cells are weird and complicated.

Dephlogisticated, genomic instability is a whole different thing can have a negative impact. The single versus double strand break makes sense when you picture DNA. If there is one break the DNA is still being held together by the other strand. A double strand break can cut the DNA in two, leading to many more opportunities for misrepair.

Alpha radiation deposits energy in a way more likely to cause double strand breaks per unit energy deposited than beta or gamma radiation.

Absorbed dose (unit=rad, joules per kilogram) is a raw measure of energy deposited. A fudge factor of between 1 and 20 depending on the type of radiation is applied to the absorbed dose to measure biological effect, effective dose (unit=rem, joules per kilogram). The fudge factor for beta and gamma is 1 (so no effect), the fudge factor for alpha is 20. So 1 rad of beta radiation is 1 rem, but 1 rad of alpha radiation is 20 rem. This is saying the same amount of energy deposited is twenty times more harmful when deposited by alpha vs beta radiation.
posted by Across the pale parabola of joy at 7:24 PM on April 20, 2016 [1 favorite]

UNIMAGINABLE!

Anybody want a cashew?
posted by Greg_Ace at 8:08 PM on April 20, 2016 [9 favorites]

That's awful convenient for the medical industry, whose contribution to average radiation dose has exploded from about 15% in the 80s to nearly 50% in the mid-00s. (NCRP 160) They've been getting a lot of pushback on that lately such as the "image gently" campaign. Would I be too cynical for my first reaction to this to be "oh, someone bought themselves a study"?

I can pretty much guarantee we won't change from LNT in the program I'm in unless the hormesis theory can be conclusively proven, and the data is just too buried in the noise for low doses to do that. LNT is most likely conservative, and that's a good thing where worker safety is at stake. Plus, ALARA is easy to train and enforce without sending mixed messages.
posted by ctmf at 8:10 PM on April 20, 2016 [3 favorites]

So all the hand-wringing over a little secondhand smoke is just whinging?
posted by arnoldsnarb at 8:55 PM on April 20, 2016

As you want.
posted by Greg_Ace at 9:17 PM on April 20, 2016 [2 favorites]

Reddit had a thread on the work cited in the Petition for Rulemaking.
posted by ctmf at 9:25 PM on April 20, 2016

Never get a plantar wart in Asia!
posted by XMLicious at 9:31 PM on April 20, 2016 [5 favorites]

The "Linear no threshold model" is applied to chemicals, too. IMHO the best artificial sweetener is cyclamate, but it was banned in the US because it caused cancer in lab rats. (Everything causes cancer in lab rats. They die of cancer if you look at them wrong, I sometimes think.)

But the dose of cyclamate they gave to the rats was preposterously high, completely unrealistic compared to how the stuff was actually being used.

The released study was showing that eight out of 240 rats fed a mixture of saccharin and cyclamates, at levels of humans ingesting 350 cans of diet soda per day, developed bladder tumors.

If you drank that much water per day it would kill you.

In 1970 the FDA banned Cyclamate. It's always been legal in Canada, though, and in a lot of other countries who don't seem to be having a massive flood of cancer cases as a result.

Cyclamate could become legal again in the US if someone would pay for all the testing the FDA would insist on. The problem is that the patent on Cyclamate ran out a long time ago, so as soon as one company spent all that money and time and got it reapproved for the US, everyone could start making it again, so the company that made the investment wouldn't get any return. So no one has ever done it.

And there's a lot of reason to believe that the "Linear no threshold model" is just as bogus when applied to food additives as it is when applied to radiation exposure.
posted by Chocolate Pickle at 9:54 PM on April 20, 2016 [3 favorites]

Dephlogisticated, genomic instability is the name of my next band.
posted by five fresh fish at 11:05 PM on April 20, 2016

And there's a lot of reason to believe that the "Linear no threshold model" is just as bogus when applied to food additives as it is when applied to radiation exposure.

But somehow I don't think by "just as bogus" you don't mean "no particular reason to think it's bogus?"

The idea that it's bogus on radiation is popular with some people but is hardly consensus. I assume buried in some of the OP links are, for example, references to the long term study of French nuclear workers with minimal exposure to radiation and slightly higher cancer incidence. (Not the study in question, but similar results.) Increases in cancer get harder to measure as they get smaller, but they seem to be found when we have the ability to do a sufficiently powered to study to see them.

The idea that there is a safe dose for radiation is just that--an idea. It's not theoretically more compelling than the idea that there is no safe dose, and there's no empirical evidence that it's true.
posted by mark k at 12:11 AM on April 21, 2016 [2 favorites]

cancer radiotherapy delivers doses of perhaps 6 million mrem to tumors, in hopes of destroying them

Radiation therapy is a very high exposure targeted very specifically and deliberately at a tumor centroid. Cancer-targeted therapy is not at all the same as "hormesis" or a deliberate increase in background radiation dosage.

I would probably avoid taking radiation safety advice from the NRC anyway. They have a long record of interfering with local governments to keep unsafe reactors open and running, and their budget and staff are currently being slashed.

You can't downsize or "right-size" a regulatory agency without making deliberately calculated tradeoffs about public safety. As NASA and its astronauts were forced to find out with the Challenger and Columbia disasters, you can't fool nature (but you can probably figure out who votes in sufficient numbers to hold politicians to account for funding decisions, and be comfortable with putting the rest of the populace at risk).
posted by a lungful of dragon at 12:34 AM on April 21, 2016 [2 favorites]

One of the side effects of hormesis can be serious crankery. This guy has moved on from "CO2 couldn't possibly be a greenhouse gas" to "hormesis exists, therefore nuclear power plants are totally safe!" (I couldn't remember his name, but a search for "scottish hormesis" returns this article in first place on Google.)
posted by sneebler at 6:43 AM on April 21, 2016 [3 favorites]

My name is Nino Mantegna, you murdered my grandfather, let's get ready to RUMMMMBLLLLE!
posted by Splunge at 9:16 AM on April 21, 2016 [3 favorites]

I remember reading in the book Merchants of Doubt (recommended) that hormesis--and relatedly the concept of a threshold below which stressors are not harmful--got a big push in the 70s by polluting companies, against the newly growing envirnomental movement. IIRC it was used to argue against the regulation of a bunch of different chemicals, often with "in very low doses, X is not harmful" elided into "X is not really harmful".

I first heard of hormesis years ago, and without doing a lot of research was (regrettably) left believing it mostly because of the elegance of the concept, the way it neatly turns the world on its head. Learning about this politicization made me much more skeptical of what is, after all, a small and tricky effect to tease out, subject to all the usual difficulties of statistics (I'd expect that most studies demonstrating it are spurious).
posted by gold-in-green at 10:27 AM on April 21, 2016 [4 favorites]

Xkcd has a nice chart of radiation exposure levels from "sleeping next to someone" to "Chernobyl just after meltdown", though be careful with its constant switches between doses per year, per day, and per hour.

One thing that stands out from the chart is that the average person already receives lots of low background doses of radiation, most of it from natural-ish causes (e.g. "yearly dose from natural potassium in the body"). The only benefit of knowing whether microdoses of radiation help or hurt would be if we found out that it's harmful and we could stop building with concrete and eating bananas. If it helps, we don't need to do anything, since we're already getting dosed.
posted by clawsoon at 12:38 PM on April 21, 2016 [2 favorites]

Before this effect was established and accepted, my uncle, a biologist at Los Alamos in the 1950s - 1970s, studied cobalt irradiation of 83 successive generations of mice, attempting to determine if it would create mutagenic effects that could be passed to offspring. Incidental to the intent of the study, the statistics showed that while highly dosed mice had greater mortality, as would be expected, the low-dosed mice lived longer than the control group with no dose. Colleagues at Oak Ridge didn't beieve it, but his data stood scrutiny.

More commonly, the same effect is noted with alcohol consumption.
posted by Pressed Rat at 7:31 PM on April 21, 2016

More commonly, the same effect is noted with alcohol consumption.

Unfortunately, closer looks at the alcohol data have shown that's not true. The studies which found a health effect for moderate alcohol consumption were lumping together people who had stopped drinking with lifelong abstainers as "non-drinkers". Once you separate those two groups, you find that the lifelong abstainers live as long as people who have one or two drinks a day.
posted by clawsoon at 4:16 AM on April 22, 2016

So I have been idly putting together links for a similar Nuclear Hormesis post, as this one got up first I hope it is ok to share some of the links I put together ( I don't think there are any repeats):

Blog entries or other short form posts of the anecdotal variety:
A blog article on Nuclear Hormesis from a blog focusing on scientific anti-FUD approach (you may disagree with them, but that is their stated goal). They have a bunch of global warning denialism also that is unrelated to this issue.

Contarian Forbes article on Nuclear radiation fears post Fukushima


UN radiation advice

then some long form or scientific studies:

Slate Article on animals and plant in the Chernobyl exclusion zone

UN report on radiation effects

NIH study on radiation effects in mice

And an alternative lifestyle choice:
Living in a radon mine
posted by bartonlong at 1:24 PM on April 22, 2016 [1 favorite]