Skip to main content

Researchers Trumpet Another Flawed Fukushima Death Study

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


In June I wrote about a claim that babies in the U.S. were dying as a direct result of Fukushima radiation. A close look at the accusation revealed that the data used by the authors to make the argument showed no such thing. "That data is publicly available," I wrote, "and a check reveals that the authors’ statistical claims are critically flawed—if not deliberate mistruths." The authors appeared to start from a conclusion—babies are dying because of Fukushima radiation—and work backwards, torturing the data to fit their claims.

Now the authors have published a revised study (PDF) in theInternational Journal of Health Services. A press release published to herald the article warns, "14,000 U.S. Deaths Tied to Fukushima Fallout." This is an alarming accusation. Let's see how the authors defend it.

First, the authors assert: "In the United States, Fukushima fallout arrived just six days after the earthquake, tsunami, and meltdowns." They provide no evidence for this assertion, no citation to back up their facts. The authors then note that the U.S. Environmental Protection Agency monitored radioactivity in milk, water and air in the weeks and months following the disaster. Ah, here must be the data, the careful reader hopes. Alas, "the number of samples for which the EPA was able to detect measurable concentrations of radioactivity is relatively few," the authors write. They then conclude, with evident disappointment, that "clearly, the 2011 EPA reports cannot be used with confidence for any comprehensive assessment of temporal trends and spatial patterns of U.S. environmental radiation levels originating in Japan." In other words, the EPA didn't find evidence for the plume that our entire argument depends on, so "clearly" we can't trust the agency's data.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Yet even if there isn't evidence for a plume, where do all the dead people come from? Here, from the abstract, is the chain of reasoning: "U.S. health officials report weekly deaths by age in 122 cities, about 25 to 35 percent of the national total. Deaths rose 4.46 percent from 2010 to 2011 in the 14 weeks after the arrival of Japanese fallout, compared with a 2.34 percent increase in the prior 14 weeks....Projecting these figures for the entire United States yields 13,983 total deaths." In sum: Sloppy statistics killed 14,000 people.

To unpack a little more, the authors take mortality figures from the Centers for Disease Control and Prevention (CDC) Morbidity and Mortality Weekly Reports. I talk a little about these reports in my original piece. Suffice it to say that they are an incomplete record of deaths in the U.S. (as the authors acknowledge). The authors draw a hard line at the week of March 20, 2011, the 12th week of the year. They sum up all deaths around the country for both the 14 weeks preceding and the 14 weeks following March 20, 2011. They do the same for 2010. They find the CDC reports include 4.46 percent more dead people in the 14 weeks after March 20, 2011, than the reports did in the 14 weeks after March 20, 2010. The 14 weeks preceding March 20, 2011 (presumably before the radiation plume arrived and spread across the land) include only 2.34 percent more dead people than the 14 weeks preceding March 20, 2010. Since the CDC only reports on about 23.5 percent of all deaths, the authors claim, they helpfully multiply the supposed "excess" by 1/0.235 to arrive at the final number of 13,893 deaths.

No attempt is made at providing systematic error estimates, or error estimates of any kind. No attempt is made to catalog any biases that may have crept into the analysis, though a cursory look finds biases a-plenty (the authors are anti-nuclear activists unaffiliated with any research institution). The analysis assumes that the plume arrived on U.S. shores, spread everywhere, instantly, and started killing people immediately. It assumes that the "excess" deaths after March 20 are a real signal, not just a statistical aberration, and that every one of them is due to Fukushima radiation.

The publication of such sloppy, agenda-driven work is a shame. Certainly radiation from Fukushima is dangerous, and could very well lead to negative health effects—even across the Pacific. The world needs to have a serious discussion about what role nuclear power should play in a power-hungry post-Fukushima world. But serious, informed, fact-based debate is a difficult enough goal to achieve without having to shout above noise like this.

(Image by Li-sung at Wikipedia Commons.)

Michael Moyer is the editor in charge of physics and space coverage at Scientific American. Previously he spent eight years at Popular Science magazine, where he was the articles editor. He was awarded the 2005 American Institute of Physics Science Writing Award for his article "Journey to the 10th Dimension," and has appeared on CBS, ABC, CNN, Fox and the Discovery Channel. He studied physics at the University of California at Berkeley and at Columbia University.

More by Michael Moyer