KSJ Tracker September 22, 2012

"A rancid, corrupt way to report about science"

Corn/Wikimedia commons

The quote in the headline on this post comes from Carl Zimmer's blog, The Loom, in a commentary on coverage of recent European study on possible health effects of eating GM crops. To give you the short version, the study authors appear to have practiced some very questionable science and some - and cynical - manipulation of the science media.

Or as Ivan Oransky put it on Embargo Watch:  "A study of the effect of genetically modified corn on rats that you may have read about earlier this week doesn’t seem to have said much about whether GMOs are safe. But it sure said a lot about how the scientists who did the work used a crafty embargo to control their message."

And Zimmer and Oransky weren't the only notable science writers to express real dismay over the way the French researchers worked to control the media in this case (you'll find in Oransky's post the fact that the lead researcher has a book coming out on the subject of GM crops, by the way).

The story began in fairly standard way, publication of a study on the effects of feeding the weedkiller, Roundup, and Roundup resistant corn to rats.  The study, published in the Journal of Food and Chemical Toxicology, which was embargoed for release this week.

Here was the catch: The researchers agreed to provide advance copies of the study, which purported to find that a large increase in tumors in the GMO-crop and herbicide exposed rodents, only  journalists would sign an agreement not to show the paper to any other scientists for comment. In other words, the initial stories would be required to take the study at face value. Although Oransky notes that news outlets later updated their story, the first accounts are still enjoying an uncritical ride in activist circles, such as this one from France's Le Monde, where I assume the agreement was signed and which reads almost like a press release.

As the researcher and science blogger Scicurious pointed out in a post at Discover's The Crux, this meant that such stories played right into the anti-GMO narrative rather than putting the story in full context or calling attention to the concerns expressed by other researchers about the validity of the results. 

To quote: Scicurious: "Following the release of the study, numerous scientists questioned the findings, citing 'anomalies throughout the paper that normally should have been corrected or resolved through the peer-review process.' In particular, there are problems with the statistics performed on the data, the way the data were presented, and the numbers and types of animals used in the study."

Among the problems, for instance, the researchers chose a strain of rats known to spontaneously develop tumors without exposure to any of the products under study. Given a 30 percent mortality rate in their control animals, they also failed to provide ncessary data about which those animals had developed tumors. In fact, so many scientists criticized the findings that the British-based Science Media Centre ended up putting a resource list of such comments on its website.

At The New York Times, Andy Revkin noted that anti-GMO advocates were already promoting the uncritical versions of the study. The piece is also worth reading for the excellent point about the "single-study syndrome" - the tendency of advocacy groups to use these kinds of reports, often out of the context of the greater body of work, to argue their points.

Which is exactly what Rosie Mestel, at The Los Angeles Times, reported - that  the study - despite the methodological criticisms -  "was embraced by opponents of genetically altered foods, including backers of Proposition 37, which if approved by California voters in November would require most foods with genetically modified ingredients to bear a label."

Still, outside of activist circles, it's worth noting that many (good) journalists responded with skepticism to the report:

 At The Washington Post, Tim Carman wrote a thoroughly researched story, describing the study as controversial in the very first graf. (The headline - "French Scientists Question Safety of GM Corn" - oversimplified but that's another risk inherent in single study syndrome).

At Boing-Boing, Maggie Koerth-Baker wrote a savvy and sharp-edged analysis with the excellent headline: Authors of study linking GM corn with rat tumors mainpulated media to prevent criticism of their work."

At Discovery News, we find Emily Sohn's excellent "GM Corn-Tumor Link Based on Poor Science".

At Geekosystem, Ian Chant goes farther, calling lead researcher, Gilles-Eric Séralini of the University of Caen,  sketchy for refusing to turn over his data to the EU agency responsible for food safety. This follows on a report from Martin Enserink at Science that agencies with both the EU and the French government are planning to investigate the results more thoroughly. And UPDATE: Over at Slate, science writer and blogger Keith Kloor points out numerous other aspects in this researcher's background (from longtime anti-GM) advocacy to upcoing book on the subject) that should offered up an alert.

In other words, that non-disclosure demand from Séralini and his colleagues should have been a red flag from the beginning, a warning that this could be some very suspect research. And the appropriate response from any journalist should have been a flat no - it's far better to be second-day on a suspect story than to be suckered on the first day. Or, to quote Zimmer again, as he calls out organizations that did sign the agreement: "For shame."


comments powered by Disqus