Home / Environmental Disasters / Fukushima’s Impact in the United States

Fukushima’s Impact in the United States


This seems like a small sample size to me, but the scientists say it is statistically significant:

The recent CDC Morbidity and Mortality Weekly Report indicates that eight cities in the northwest U.S. (Boise ID, Seattle WA, Portland OR, plus the northern California cities of Santa Cruz, Sacramento, San Francisco, San Jose, and Berkeley) reported the following data on deaths among those younger than one year of age:

4 weeks ending March 19, 2011 – 37 deaths (avg. 9.25 per week)
10 weeks ending May 28, 2011 – 125 deaths (avg.12.50 per week)

This amounts to an increase of 35% (the total for the entire U.S. rose about 2.3%), and is statistically significant. Of further significance is that those dates include the four weeks before and the ten weeks after the Fukushima Nuclear Power Plant disaster. In 2001 the infant mortality was 6.834 per 1000 live births, increasing to 6.845 in 2007. All years from 2002 to 2007 were higher than the 2001 rate.

A story worth following.

  • Facebook
  • Twitter
  • Google+
  • Linkedin
  • Pinterest
  • Why compare 4 weeks to 10? I smell cherry-picking: http://xkcd.com/882/

  • SP

    No, the data appear to have been pretty badly cherry picked. First clue, the original report was on counterpunch, nothing remotely close to peer reviewed.

    • Mike Schilling

      You’re sure it was on CounterPunch? I don’t see where it’s the Israeli’s fault.

  • McWyrm

    For my money, this study was pretty nicely debunked here.

    It’s a fairly straight forward case of cherry picking – why those eight cities? And why four weeks instead of six or eight or ten weeks? If you expand the study temporally or spatially the conclusions don’t hold up.

  • wengler

    That evidence isn’t compelling at all. The two sample sizes are different, the cities are too random, and the best comparisons are to similar times in different years.

    What would be good is to have transparent monitoring of radiation levels by the EPA.

  • elm

    This story didn’t pass the smell test for me when I read Erik’s excerpt and I thank McWyrm for linking to that excellent article discussing the problems with it. (In comments there, someone links to a story that shows that looking at a longer pre-Fukishima time fram (or comparing to the same time frame the year prior) and including all cities in the region, the increase in the Pac NW is no different from the national increase.)

    But above and beyond the potential statistical cherry-picking, what’s the causal mechanism the authors are offering? I’m not a doctor, but from what I understand, radiation has 2 different health effect: massive amounts can lead to radiation poisoning and quick death; small amounts can lead to long-term increases in cancer rates.

    I don’t think enough radiation has made it to the U.S. to cause the former and not enough time has passed to cause the latter. So how has Fukishima led to an increase in deaths in the Pac. NW?

    • DocAmazing

      Without bothering to actually, y’know, read the studies in question, I can say that infant mortality differs from overall mortality because the first few weeks of life are fraught times, and the effects of radiation on the fetus and the immediate newborn don’t fall neatly into those two categories. There are a lot of complex processes at work in fetuses and newborns–lots of rapid cell growth and differentiation, lots of systems that haven’t quite come online yet–that can be interrupted and screwed up in ways that older babies and adults wouldn’t fall prey to.

      Measurement of radiation was done: too lazy to find the link, but rainwater in Berkeley was quite radioactive–with the levels being thought of as too low to be significant by most epidemiologists and oncologists. Still, that which is well-tolerated by an adult may not be well-tolerated by a newborn.

      • elm

        Thanks. It’s a good point that newborns will be affected in ways adults wouldn’t be. But I also vaguely remember the reports of the radiation measurement and I don’t remember any of the doctors then saying the levels were high enough to influence newborns in this way. I would think that if levels were safe adults but not children, that would have been blared in all headlines. Or do we not actually know what the safe level is newborns?

        • DocAmazing

          Very hard to test what that safe level might be. No one wants to volunteer their newborn for radiation testing, mirabile dictu. Also, corporate news outlets (many of them owned by outfits like GE) are famous for soft-pedaling the dangers of radiation and of nuclear power.

          I’d be careful about dismissing the idea that Fukushima fallout wasn’t harmful to US West Coast newborns just yet. We still have a lot to study and a lot to evaluate.

        • Just for reference, the DOE allowable occupational dose for a pregnant woman is 500 millirem (5 millisieverts) over the course of gestation. My facility additionally restricts the allowable dose to 50 millirem (.5 millisieverts) in any one month.

          Nothing I’ve seen leads me to believe that anyone in the US has received anything close to either of these limits from the Fukushima accident.

  • John

    Yeah this reeked of cherry-picking.

    Doesn’t reflect well on Eric’s BS detector.

    Reading Counterpunch and taking it at face value is asking to be lied to.

    • To be fair to me, I did say that it seemed like an awfully small sample size.

      • (the other) Davis

        The bigger red flag, as Dr. Drang noted above, should have been the differing sample periods (4 weeks before vs. 10 weeks after). That kind of inexplicable asymmetry is a hallmark of ex post cherry-picking.

      • John

        Perhaps a brief update is in order then…

        • Yes, clearly I owe it to you…..

          • elm

            Well, you don’t owe it to anyone, but if you now think the story is false, you probably should update the post to say that.

          • John

            Yes, because I clearly said you owed me an update.

            Maybe we should take a step back then. I had assumed you recognized the foolishness of this story when it was repeatedly pointed out to you above. Is that so, or do you still think this exercise in fun-with-numbers is a “story worth following?”

            If you now realize that it isn’t, in fact, worth following, it would be logical to update the post saying so.

            I know that if I wrote something stupid I’d want to correct the record given the opportunity…

          • Ed Marshall

            Y U such a dick to your commentors?

            • Ed Marshall

              I probably should have prefaced that with a Erik Loomis!

          • Ed Marshall

            and thirdly, Charli Carpenter made some piss poor predictions about the safety of nuclear power when it was all falling apart right after the tsunami. I think I asked for an update and it was granted.

            She had the good grace to acknowledge facts as they unfolded. It’s not asking a shitload out of you to do something similar. Do whatever you want, you don’t owe anyone anything, but you don’t need to be surly about it.

          • Sorry, I seemed to have missed what the problem is with 1) updating the article to point to various debunkings and 2) suggesting so in comments.

            Even if the latter were annoying I don’t see that it makes the former a bad idea. The post would be stronger if one didn’t have to delve into comments (or Google around) to find the debunkings (which seem pretty good to me, with lots of nice discussion of sample selection and stats).

            • On further consideration, clearly you owe it to Matthew Yglesias to update this post!

              He did give you a blated, hostile acknowledgement!

  • jon

    That’s just bullshit, and it serves to diminish thorough consideration of Fukushima.

    All radiation poses a certain danger for illness and mortality, but the amounts recorded to have reached the US are quite small. Moreover, unless exposed to many rems over a short period, illness resulting from low level radiation exposure generally takes one or several decades to manifest.

    It would be better if this attention were focused on the continuing accident at Fukushima, continuing unrestrained releases of radioactivity, the continuing irradiation of nearby residents, and the confused and halting efforts to bring the accident under control and contend with the aftermath.

  • This thread makes me proud.

    Is there even the slightest chance of a similar thread appearing below a post on a right-wing web site that purports to debunk global warming?

    • FHD

      This thread makes me proud.

      Of the comment section, anyway.

  • John#2

    Not statistically significant, unless you lower the threshold of “significance” to levels which are almost never used, because of the relatively high chances of false positives. Even then, you have to make some pretty strong assumptions to get those results, like that deaths are independent of each other (e.g., no weather effects, no illnesses sweeping through the area…) Either that or make a mistake in how you analyze the data…

  • snarkout

    I can’t come up with an explanation for the 4-weeks vs. 10-weeks discrepancy that doesn’t involve the authors cherry-picking pre-meltdown period with anomalously low number of infant deaths in the cities in question (and, indeed, that Scientific American blog post suggests that if you do an apples to apples comparison the death rate has actually gone down slightly). It strikes me as very much like hollering “global cooling” because 1998 (and then 2007) was the hottest year on record.

    You don’t owe any of us individually anything, but I’d also like to see an update on this post.

    • John

      Erik, please allow me to save you some time and respond to snarkout for you.


  • Chicken Little is out with a new survey…

  • Sophia

    A story worth following.

    This is borderline “it would be irresponsible not to.”

  • ajay

    A story worth following.

    I for one am keenly looking forward to Erik’s next post on this important subject.

It is main inner container footer text