This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Five Surprising Public Health Facts About Fukushima


Four years after the Fukushima nuclear accident, which catalyzed fears around cancer epidemics and large radioactive exclusion zones, it’s time to take stock of the scientific consensus on the health effects of the reactor meltdown. Increased scanning for thyroid cancer found that cancer rates were in fact lower for children in Fukushima than in other parts of Japan. Only a clearer perspective of such public health findings will allow us to understand future risks.

Journalist Will Boisvert Investigates and Finds Fewer-than-Average Thyroid Cancers and Seafood Safe to Eat

by Will Boisvert

Four years ago a large earthquake and tsunami devastated northeast Japan. More than 15,000 people were killed. A subsequent nuclear meltdown added fear to grief. 

As terrible as the meltdown was, the radiation did not have significant public health consequences, much less the catastrophic ones that many feared and some continue to claim.

On the fourth anniversary of the tsunami, earthquake, and meltdown, journalist Will Boisvert investigates and unearths five public health findings from Fukushima that you’ve probably never heard.

Thyroid Cancer Rates Lower in Fukushima Children Than Other Regions

Radioactive iodine from a nuclear accident can pose a serious risk of thyroid cancer in kids; thousands of cases emerged after the Chernobyl disaster. Thanks to bans on contaminated milk and other foods after the Fukushima accident, Japan expected to avoid a similar epidemic. Unfortunately, a new monitoring system for thyroid cancer seemed to reveal an immediate and drastic effect. Hundreds of thousands of children in Fukushima prefecture underwent sensitive ultrasound scans after the accident. The results showed that 44 percent of them had thyroid cysts or nodules, which are possible warning signs of cancer.13 So far more than a hundred confirmed and suspected thyroid malignancies have been diagnosed,14 a rate many times higher than normal for children. Critics of nuclear power took these findings as a portent of “The Nuclear Sacrifice of Our Children,” in the words of über-activist Helen Caldicott.15

Fortunately, the thyroid holocaust Caldicott foresees is just a mirage — a product of increased monitoring rather than any underlying spike in thyroid cancer incidence. It turns out that a 44 percent rate of thyroid “abnormalities” is actually quite normal. The Japanese government gave thyroid scans to 4,365 kids living in distant areas of Japan that received no fallout, using the same ultrasound equipment, protocols, and diagnostic criteria used in Fukushima. That baseline study found that the frequency of thyroid nodules and cysts in that uncontaminated population was about 57 percent — somewhat higher than among Fukushima kids.16 The spike in Fukushima thyroid anomalies isn’t caused by fallout — because there is no spike.

The apparent rise in diagnoses of thyroid cancer proper is also an artifact of the mass scannings, which are discovering small, non-aggressive, asymptomatic tumors that would normally not be found until adulthood, if ever. (Most thyroid tumors — more than 90 percent — are curable because thyroid cancer tends to be slow growing.) Indeed, the advent of super-sensitive ultrasound scanning has caused a surge in thyroid cancer diagnoses all over the world in recent decades, reports Gina Kolata in The New York Times. Thyroid cancer fatality rates have stayed the same, which means the new technology is unearthing innocuous tumors that would not have caused illness.17 The same dynamic is likely at work in Fukushima.

There may be a slight increase in thyroid cancers from radioactive iodine exposure in coming years, on the usual too-small-to-measure scale predicted by the linear no-threshold model. But the apparent surge in thyroid cancer will be huge given the zealousness of the monitoring program, which will scan the cohort of Fukushima kids every few years for the rest of their lives.19 Autopsy studies suggest that as many as 27 percent of us normally develop asymptomatic thyroid tumors that we never know about.18 So Fukushima is in for a tsunami of over-diagnosis and over-treatment of mainly harmless tumors, with serious side effects: another instance of alarmism that causes more harm than the risk it’s trying to avert.

Fukushima Seafood Safe to Eat

Stories about California tuna contaminated with radioactive cesium from the Fukushima spew are a staple of antinuclear websites. Most press accounts dismiss the contamination as harmless. But in Japan radioactive fish are a major issue. Fishermen near the plant are barred from harvesting many species that are prone to concentrating radio-cesium, and South Korea has banned seafood imports from the region.20 Concerns about the fishing industry are the primary obstacle to dumping the accumulating tons of decontaminated water stored at the Fukushima Daiichi plant. The crippled coastal fishery has become a symbol of nuclear energy’s power to poison the environment on a grand scale.

But as so often in the Fukushima saga, claims of inordinate danger turn out to be exaggerated. The Japanese government’s safety limit for radioactive cesium in fish is 100 becquerels per kilogram (Bq/kg), an amount that’s roughly equal to the natural radioactivity of potassium and carbon isotopes present in all food. The Fukushima fishery easily meets that standard. Out of 95 fish caught within 20 km of the plant in a census conducted in January, a grand total of one specimen came in above the cesium limit at 113 Bq/kg. Most fish were far below the limit, with 43 showing no detectable contamination at all.21

And that 100 Bq/kg standard is extraordinarily strict. The US EPA estimates that eating a becquerel of radioactive cesium gives you a one-in-a-billion chance of contracting cancer.22 If you were to eat a kilogram of fish contaminated with 100 becquerels of radio-cesium every day, for eighty years, your odds of getting cancer from the cesium would be about 0.3 percent, compared to the typical American’s 40 percent chance of getting cancer. (That’s not counting any anti-cancer benefits from eating fish instead of cheeseburgers.)

There’s little question that the Fukushima fishery is safe enough to reopen. On the other hand, its ongoing closure is a boon to hard-pressed fish stocks — which may make the nuclear accident a net win for marine life in the area.

Fukushima Evacuation Zone Is Mostly Habitable

The chaotic evacuations from the area around the Fukushima Daiichi plant, and the resulting scenes of terrified refugees, abandoned towns, and ruined houses, figure prominently in the apocalyptic aura surrounding the accident. Nightmarish evacuation scenarios were floated during the crisis: NRC chairman Gregory Jaczko advised Americans to withdraw from a 50-mile zone around the stricken plant,3 while Japanese Prime Minister Naoto Kan contemplated evacuating Tokyo itself.4 Cooler heads — and computer models5 — prevailed. But the perception remains that evacuations from the 20-kilometer zone around the plant (with a further bulge to the northwest) prevented massive loss of life in a region of lethal radiation that will remain uninhabitable for decades to come. Yet data buried in UNSCEAR’s Fukushima report, and little noticed by the media, show that the evacuation zone is not — and never was — the unsurvivable wasteland it’s made out to be.

Let’s crunch the numbers. UNSCEAR estimated the average radiation doses that would have been incurred inside the 20-kilometer evacuation zone in the first year after the accident, had there been no evacuation: the highest was Tomioka township’s 51 millisieverts.8 The Committee also reckoned that 80-year lifetime doses in contaminated areas will be two to three times the first-year dose. (Radiation levels drop quickly because of radioactive decay and weathering.)9From there we can reckon the dose people would have received from fallout had they lived their whole lives in the evacuation zone: about 100-150 mSv in the most contaminated townships, substantially less elsewhere in the zone. Natural background radiation in the United States averages about 2.4 mSv per year, so 150 mSv is about equal to the lifetime background dose of a typical American.

So how unhealthy is that extra radiation? Not very. Again, radiation is a weak carcinogen: applying the LNT theory and standard risk factors from the National Academy of Sciences,10 a lifetime dose of 150 mSv confers a fatal cancer risk of about 0.9 percent — the same odds an American runs of dying in a car crash.11 Those are average risks; there are hotspots with higher radiation levels, and children would have faced somewhat higher risks, especially from thyroid doses in the first three months after the spew. Still, these numbers give a good ballpark sense of the health risks from fallout in the Fukushima evacuation zone: about the same as the risk of having a driver’s license.

The mandatory relocations from the Fukushima evacuation zone, which are responsible for much of the accident’s cost and all of its cataclysm, owe more to apocalyptic expectations built into regulatory standards than to objective health hazards from fallout. Those hazards are well within the range of risks we negotiate in ordinary life. It might be time to reconsider policies that require precipitate or long-term relocations, which carry their own risks. Hundreds of people died from the stress of the Fukushima evacuation, and thousands more were uprooted from their homes over radiation doses that would almost certainly never affect their health. Instead of requiring people to leave, it could make more sense to give them the information they need on radiation exposures and likely health risks, and let them make their own decisions.

Cancer Rates in USS Reagan Crewmembers Lower Than Control Group

It sounds almost like a Hollywood thriller. On March 12, 2011, dispatched on a humanitarian mission to help victims of the Tohoku tsunami, the aircraft carrier USS Ronald Reagan cruised straight into the fallout plume from the Fukushima Daiichi reactors. Sinister portents were noticed during the weeks the vessel stayed off the Japanese coast: jumpy radiation meters; a strange metallic taste to the air; cryptic warnings from commanders about drinking water; a massive decontamination scrubdown. In the years after the mission ended, a plague of cancers and other mysterious diseases erupted among the ship’s mainly young and formerly healthy crew.23 The result is a lawsuit in the US District Court in San Diego, seeking $1 billion for medical bills along with other damages from the plant’s owner, TEPCO, on behalf of some 239 (and counting) stricken plaintiffs.24

The case so far has turned on tort law, jurisdiction, and lawyerly histrionics. (TEPCO lawyers argue the case belongs in a Japanese court; plaintiffs’ attorneys told the judge that sending their clients to Tokyo to testify would be a “death sentence” because of the lingering radiation there.)25 But there’s an underlying scientific issue here that’s crucial to our understanding of the Fukushima accident. Does the Reagan’s experience reflect the underestimated peril of nuclear radiation — or the human penchant for finding meaningful patterns in random misfortune?

Since cancer is the main long-term illness that has been linked to low-level radiation, press accounts of the lawsuit emphasize the anomalous number of cancers among Reagan sailors. But there may be less anomaly than meets the eye. For one thing, many Reagan plaintiffs don’t have cancer. Among 22 plaintiff statements excerpted on the lawsuit’s website, half say nothing about cancer. Their complaints run the gamut: ulcers; “migraines and sinus problems”; irregular menstrual cycles; “declining sex drive;” anxiety; “my back, neck, and entire right side of my body were cramping up and hurting from sitting in a chair too long.” None of these ailments are recognized chronic effects of radiation.26

Paul Garner, an attorney for the Reagan plaintiffs, told me that about three quarters of his clients do have cancer, a claim I did not try to independently verify.27 That works out to about 180 cancers, which might well be an anomalously high number for the Reagan’s crew of 4,843. But there’s another problem: not all the Reagan plaintiffs served on the Reagan. The lawsuit includes sailors from at least six other warships, including the aircraft carrier USS George Washington with its crew of 5,000. It also includes people who served at naval shore installations and two children of about 9 and 12 years old.28 The right baseline for the Reagan cancers may not be the number normally expected among the ship’s crew, but the number expected among tens of thousands of US service members and their dependents who were at sea and on shore during the accident. Assessing a possible Fukushima link to illnesses in that large, disparate population requires sophisticated epidemiology that isn’t captured in the lawsuit’s heartbreaking testimonies.

The Pentagon undertook just such a study. It compared illness rates among the Reagan crewmembers who were on board for Operation Tomodachi, the two-month humanitarian mission during the Fukushima release, with those in a control group of 65,000 other sailors. Its 2014 report to Congress found mildly elevated levels of digestive, respiratory, genitourinary, obstetric, and male infertility disorders among Reagan crewmembers, along with heightened rates of vague “symptoms, signs, and ill-defined conditions.” The report also found normal or decreased rates of 24 other disease categories. In particular, rates of cancer, the one disease category plausibly linked to low-level radiation, were 45 percent lower among Reagan crewmembers than among controls. That haphazard mix of higher and lower disease rates likely had nothing to do with Fukushima radiation, the report concluded, because cumulative radiation doses on the ship during Tomodachi averaged just 0.08 millisieverts, far too small to produce any noticeable illness.29 Indeed, the Reagan was one of the less irradiated berths in Operation Tomodachi. The George Washington got a dose four times higher just sitting in a Japanese port.30

Reagan attorney Paul Garner thinks the military’s estimate of Reagan doses is too low, but declined to offer his own. (The Reagan is a nuclear carrier with expertise in radiation measurement.) He also told me that none of his clients’ ailments appeared in the Pentagon’s registry of illnesses of Tomodachi sailors. When I asked how that was possible, since the Pentagon did an exhaustive electronic search of medical records, he suggested that Veterans Administration doctors had “covered up” his clients’ illnesses, and further suggested that I was in the pay of the Defense Department myself.31 (The US military cannot be sued by the Reagan plaintiffs and has no obvious stake in their suit against TEPCO.)

There’s no clear evidence of a surge of illness among Reagan sailors, let alone one caused by the Fukushima accident. Maybe that’s because of a conspiracy to hide the evidence. But, perhaps, it’s due to the gap between scientific epidemiology and folk epidemiology. We’re mentally primed to search for causes for the ills that befall us, and to locate them in the scariest things we can think of — like a nuclear accident. Factor in the way a dense social network, like the one among a ship’s company, can amplify rumors and theories, and a set of coincidental ailments can suddenly take on an ominous patterning. That dynamic explains much of the alarm surrounding the Fukushima spew.

Fukushima Death Toll Is Too Small to Measure

Reports warned of vast casualties stemming from the Fukushima accident. (Activist Arnie Gundersen predicted up to 1 million cancers in the next 30 years.)1 But radiation scientists looking at the data on radioactive contamination saw a different picture. For them, the question is whether anyone at all will die from the Fukushima fallout.

No one in Fukushima, neither plant workers nor civilians, suffered immediate harm from the radiation; long-term effects, mainly cancer, are the chief concern. But radiation is such a weak carcinogen that it’s often hard to tell how much risk it poses. That problem has led to bitter disputes over the “linear no-threshold” (LNT) theory of radiation and cancer — the “no safe dose” model that predicts that any exposure to radiation, no matter how small, will causes a proportionate increase in cancer risk. Mainstream radiology largely accepts LNT, but some scientists contend that radiation doses below about 100 millisieverts (mSv) pose no cancer risk.

The debate grinds on because it’s almost impossible to settle with empirical data. The doses Japanese civilians will receive from Fukushima fallout are very low — 18 mSv over a lifetime on average for the most exposed kids and less for others, according to the authoritative report on the accident by the UN Scientific Committee on the Effects of Atomic Radiation (UNSCEAR).6 At those doses the additional cancers predicted by LNT will be too few to cause any discernible uptick in cancer rates. So forecasts of Fukushima casualties hinge on the epistemological conundrum of whether effects that are too small to observe really exist.

Because of that uncertainty, few epidemiologists have bothered to quantify the prospective death toll. Still, LNT can give us a rough upper bound. Using it, a few researchers — most of them critics of nuclear power — have predicted from 1,0002 to 3,00012 eventual cancer fatalities from the fallout, spread over many decades in which millions of Japanese will die from other cancers. To put that number in perspective, air pollution from coal-fired power plants kills hundreds of thousands of people every year. Fukushima doesn’t even rate on the scale of common radiation hazards. The EPA estimates that the natural radiation from household radon gas causes 21,000 yearly cancer fatalities in the United States — the equivalent of seven Fukushima spews each year.7

So even if you feel that risks that are too small to measure are still worth worrying about, one important conclusion we can draw from the Fukushima accident is that nuclear disasters are not all that disastrous.


So what are the real “lessons of Fukushima?” The Fukushima accident did spotlight safety issues, including the need to pay more attention to tsunami records, stash backup generators beyond floodwaters, and install better vents. But its modest consequences should reassure us that the apocalyptic anxieties surrounding nuclear energy are fundamentally off base. The second lesson is that the first lesson isn’t getting through to people. Visceral fears stoked by the release continue to cloud perceptions and shape politics.

What people need, but rarely get, is information that lets them understand the continuum of risk posed by different radiation levels and assess for themselves how acceptable the risks are. Once we do that we can situate nuclear accidents on a familiar spectrum of risk. Radioactive contamination works like other mild toxins — alcohol, car exhaust, trans fat — that we live with every day: a huge dose can kill us; a modest dose poses modest risks; a tiny dose is innocuous. 

Will Boisvert writes on energy, environmental, and urban policy for The New York Observer, Dissent Magazine, and other publications. He lives in New York.

Note: Footnotes are not in chronological order within the text. 






6., Table 7, p. 59.


8., Table C-11, column 7, “Projected”, p. 190.

9. Ibid., p. 209 paragraph C 154.










19., p.8



22., p. 49



25.  Official transcript of Court hearing emailed to me by Paul Garner.


27.  Telephone interview.



30.  p. 208

31. Statements by Paul Garner in telephone interview and personal emails sent to me.

32. Japan’s resettlement standard is 20 mSv,–Radiation-Exposure/


Breakthrough Institute's picture

Thank Breakthrough for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.


Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Bob Meinetz's picture
Bob Meinetz on Apr 8, 2015

Will, thanks for needed and well-referenced perspective on the situation in Fukushima.

I have one bone to pick: since thyroid cancers resulting from ionizing radiation can take decades to develop, with average onset of symptoms ~10 years after exposure, the picture is far from complete. Most researchers agree that any increase will be minimal. Nonetheless, crowing about “fewer-than-average” thyroid cancers at this stage of the game is premature, and resorts to the tactics of the Helen Caldicott/Arnie Gunderson camp.

Accuracy is our friend.

Keith Pickering's picture
Keith Pickering on Apr 8, 2015

A couple of other things worth mentioning:

1. “53 families continue to live inside the Fukushima exclusion (no-go) zone. Some are located inside the most highly-contaminated areas. None of the families live in areas where even overnight stays are sanctioned. The law does not stipulate punishment for ignoring the evacuation or overnight stay orders. The families have been repeatedly told to leave by government officials, but they have ignored the requests. While living in the no-go zone is difficult, it is not impossible. Some residents say they have stayed to tend to livestock.

2. The LNT theory was developed from, and depends entirely upon, evidence from exposure to high doses of radiation, such as from atomic bomb survivors. LNT has never been scientifically validated for low-dose radiation, and indeed a growing body of evidence suggests that both LNT and the assumptions on which it is based are invalid at low doses. Thus even the very low risks you cite are very likely greatly exaggerated. 

The key assumptions underlying LNT are that all radiation damage is (a) cumulative, and (b) irreparable. If these assumptions are true, then there should be no difference in observed damage depending on the dose rate at which a person or animal is exposed; only the total dose should matter. Yet study after study continues to show that dose rate makes a huge difference in the amount of damage observed, even when the total dose is the same. Thus the assumptions underlying LNT are known to be wrong for low doses, and the theory itself should be viewed with skepticism if not outright disbelief.

Joris van Dorp's picture
Joris van Dorp on Apr 9, 2015

– those risks do not directly damage DNA of offspring, increasing their health risks.”

Yes they do. There are a great number of different toxins and activities which damage sperm DNA, including toxins routinely released by burning fossil fuels or digging-up rare earth metals to build wind turbines and solar panels. All energy sources cause some amount of pollution, which causes some amount of damage, including sperm damage. What matters is *how much*! Nuclear is the cleanest energy source, when compared with all others.

And damaged sperm DNA is a natural fact of life. The study you linked admits that more than 10% of normal male sperm contain severely damaged DNA, whether or not you are exposed to radiation!

So you are fear mongering again, Bas Gresnigt. Why are you still here anyway? I thought you were banned.

By the way, the study you linked makes no attempt to show how much radiation the ‘exposed group’ got compared to the ‘non-exposed’ control group. It appears to be implied in the text that the ‘exposed group’ got a total cumulative radiation dose of about 2.5 mSv. That is nothing! If I would spend a year living in Denver, Colorado instead of The Netherlands, then I would already get this dose. So what your ‘study’ is doing is trying to scare me away from visiting Denver Colorado? Or a thousand other places humans have lived and thrived for millenia with high background radiation?

There are some interesting sentences in that ‘study’, such as:

only one technically competent individual was involved in analyzing the semen characteristics to reduce the potential impact by inter-observer variation.”

And there I am thinking that it would be better to include multiple individuals to do this work, to reduce the potential impact of anti-nuclear bias! Did they even do this work double blind? We can’t know because this is not explained anywhere in the ‘study’.

“future studies are certainly needed in large population to address the reproductive fitness of the exposed individuals and also the health status of the children born to radiation exposed health workers.”

Ah, so THAT’s what this study is all about: Securing long-term employment for these researchers, who have nothing to gain from admitting that their ‘findings’ are most likely nothing but noise.


Bruce McFarling's picture
Bruce McFarling on Apr 9, 2015

“The big differences:
– people can choose, weight the benefits and harm
– those risks are visible
– those risks are temporarily
– those risks do not directly damage DNA of offspring, increasing their health risks.”

It would help your argument if you included a big difference that stood out as a clear difference. The claim that people can choose what exposure they have to increased mortality due to exposure to coal mining and power generation pollution of the air and water seems to be false. The claim that the risks are visible is quite obviously false. The claim that the risks are temporary is quite obviously false.

The claim that the risks of coal air pollution do not directly damage DNA of offspring is rather a case of “not proven to” than “proven not to” (for instance, DNA damage from oxidizing particulates is hypothesized to be part of the air pollution to lung cancer pathway, and fine particulates can enter the bloodstream), and your citation is referring to people working in higher radiation environments than the environment being discussed here.

Joris van Dorp's picture
Joris van Dorp on Apr 9, 2015

Bas, you are back to your old tricks of pointing to discredited studies in order to create radiophobia (again). You need to be banned (again).

Bruce McFarling's picture
Bruce McFarling on Apr 11, 2015

Per kWh, those costs are much bigger for coal, and, at least in the US under current conditions, nuclear power plants shut down coal power plants. So if those are the critical issues, the net damage is caused by coal, not by nuclear.

As far as whether the damage of coal or car pollution is greater, it likely depends on whether you live in an area with air or water heavily polluted by coal. Where I live at present, it is not hard to see how bad the pollution from coal can be.


“We agree …”

If you are the spokesperson for a group, you ought to identify which group you are speaking for. Otherwise “we agree” sounds either like the Queen of England, Australian, Canada & etc., or else the mentally unhinged.

Jeffrey Miller's picture
Jeffrey Miller on Apr 10, 2015

I agree with the main two messages of this article, both of which are well supported by evidence given in the text. 

These messages are that the health effects of radiation from the disaster have been and will continue to be extremely (undetectably) small and that, despite this reality, people and governments have a wildly exaggerated fear of radiation. 

That said, some of the comparisons used in the article are misleading and should be modified. 

For example, the author writes:

“Still, LNT can give us a rough upper bound. Using it, a few researchers — most of them critics of nuclear power — have predicted from 1,0002 to 3,00012 eventual cancer fatalities from the fallout, spread over many decades in which millions of Japanese will die from other cancers.”

This is an invalid comparison. The relevant question is not how many Japanese will die from cancer, but whether the rate of cancer amongst those exposed to radiation from the meltdown is higher (by a statistically significant and meaningful amount) than those who were not. 

Likewise, the statements 

“To put that number in perspective, air pollution from coal-fired power plants kills hundreds of thousands of people every year. Fukushima doesn’t even rate on the scale of common radiation hazards. The EPA estimates that the natural radiation from household radon gas causes 21,000 yearly cancer fatalities in the United States — the equivalent of seven Fukushima spews each year.”7

are totally irrelevant. The number of deaths from coal (worldwide) or from radon (in the US) depends linearly on the size these populations. The relevant comparison is the rate of cancer from exposure to typical levels of coal power plant emissions in an industrial country like Japan, or the rate of cancer from typical radon exposures, compared to the rate of cancer from a Fukushima like exposure. 

But overall a good and necessary article. 

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »