Skeptoid PodcastSkeptoid on Facebook   Skeptoid on Twitter   Skeptoid on Spotify   iTunes   Google Play

Members Portal

Support Us Store


Get a Free Book



MercolaWatch:Poor research on Genetically Engineered foods

by Josh DeWald

March 1, 2013

Share Tweet Reddit

Donate This month published an article where he claims to dispel 10 myths about "mainstream nutrition". In this article I will just address his "Lie #9" regarding Genetically Engineered foods and whether he has cited sufficient evidence to demonstrate harm versus the current body of evidence. The list also discusses various nutrients, grains, milk, etc that I may investigate in later articles as well.

In the article, "10 Lies and Misconceptions Spread by Mainstream Nutrition", "Lie #9" is that "'Genetically Engineered Foods are Safe and Comparable to Conventional Foods'."* What I want to focus on is his supposed evidence to counter this "lie". He starts with a rather strong statement that "genetically engineered (GE) foods may be one of the absolute most dangerous aspects of our food supply today". He then discusses ways of avoiding GE foods, which seem like reasonable tips if that were your goal.

But the important part is where he claims, via links to two other articles, that GE foods are linked to "increased disease, infertility and birth defects" and that a "first-ever lifetime feeding study" showed an increase in damage to organs and reduced lifespans, and that these can therefore be considered "likely side effects". Certainly makes GE food sound pretty bad. But let's take a closer look at his sources.

Are GE foods linked to increased disease, infertility and birth defects?

His link about an increase in disease goes to a May 2010 article about a video with Jeffrey Smith that discusses a Russian hamster study wherein the group fed GE soy apparently experienced infertility after 3 generations and had other nasty side effects. In what is certainly a responsible way of reporting on health research, the study in question had not yet been published in a peer-reviewed journal, but was "anticipated" for that summer. Naturally, since it was not even a published study, Mercola does not reference what the title of the study might be, only saying it was performed by "Dr. Surov" of the "Russian equivalent of the US National Association for Gene Security" (This organization appears to have no web presence, but exists only in reference to articles about how bad GMO foods are. So it's not clear why Mercola and others feel the need to mention the group as if its a well-known prestigious research institution.)

Regardless of what Mercola says, I always prefer to track down the actual study being referenced and, more importantly, what the scientific community has to say about the research. Google Scholar searches were unfortunately fruitless using "Surov" with "hamster soy" and various other combinations. However using the "National Association for Gene Security" with some of the terms I stumbled on an article on the "Vegan Skeptic" blogthat had managed to track down the study using some of the same searches before finally resorting to searching directly in the Russian language. What the author identified was that the study design effectively had asingle samplefor each of their groups (non-GE soy, two different GE soys, and no soy at all), such that statistically significant differences couldn't actually be detected. And a showstopper for even trying to discuss the study as adding to our scientific knowledge of "most likely side effects", the study was never published in a peer-reviewed journal. A commenter on the vegan skeptic site provided some Russian-language responses to the (unpublished) study, some of which where themselves actually published in journals. One of the responses is rather thorough(written by a researcher at the Institute of Genetics of Cultivated Plants [translation from Russian]) and points out a good number of flaws in the design based on what they could gather from the rather vague description in the paper.

So basically what we have here is an unpublished "study" performed by an anti-GE lobby group... hardly a smoking gun and definitely not something that should be cited as scientific evidence. It's no wonder none of the previous and articles cited the actual study -- it would have become very clear how much the research did not fit into anything like the normal scientific process.

What about the "first-ever lifetime feeding study"?

The second link in support of his anti-GE claims goes to a September 2012 article which references a French study titled "Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize" published in late 2012 in the journalFood and Chemical Toxicology. The study followed rats for 2 years (versus the standard 90 days or so). What's interesting is the sheer number of (mostly negative) responses to the study from within the scientific community. If you click on the link above, the journal lists them prominently above the article, mostly in the form of Letters to the Editor (some asking for retraction). The most commonly cited issues are their lack of balanced controls, usage of rats prone to tumors, no dose-response effect, and atypical statistical analysis.Steven Novella of Science-Based Medicinealso discussed the studyon his less formal "Neurologica" blog. Dr. Novella mentions the strangely restrictive embargo placed on journalists before publication of the study --Arstechnica has more details --which was very much against the norms (normally journalists are simply prohibited from publishing their articles until the study is "live", but this embargo prohibited them from discussing it with critics to get a different side). Arstechnica also contains links to more criticisms of the study.

Perhaps more damning is thejoint statement(English) from the six French scientific Academies dismissing the study, as well as a similar statement from theEuropean Food Safety Authority (equivalent of the US FDA). EFSA found "[c]onsidering that the study as reported in the Séraliniet al.(2012) publication is of inadequate design, analysis and reporting, EFSA finds that it is of insufficient scientific quality for safety assessment"

An additional bit of "newer" info is that the authors do attempt mount a response to much of the criticism in the March 2013 print edition, which will hopefully have more responses in later editions of the journal. So time will tell how things play out in the arena of peer-reviewed journals separate from the informal criticisms being carried out online. In any case, as with the previous research, it seems irresponsible to cite this single study, which goesagainst thecurrent weight of evidence and has a massive amount of criticism currently going against it.


Mercola makes the claim that GE foods "may beone of the absolute most dangerous aspects of our food supply today" but he does a poor job of backing up this statement. One of the studies is -- as is the norm for -- a non peer-reviewed fringe study that is difficult to access and severely criticized. The second is peer-reviewed but also heavilycriticized (time will tell whether or not the criticisms hold or not), but is definitely too new to be considered a counter to all the existing studies which show no adverse effects from consumption of GE foods. Echoing EFSA, the studies that Mercola (sort of) found shouldn't be used to assist in the choice of whether or not to avoid GE/GMO foods; which many people are trying to do, but it seems to currently be a value rather than a scientific judgement.

*For some reason Mercola puts most of the "lies" in quotes, but doesn't cite anybody who makes the specific quote, so they are really just his own interpretation of what the "mainstream" says


Séralini, Gilles-Eric, Emilie Clair, Robin Mesnage, Steeve Gress, Nicolas Defarge, Manuela Malatesta, Didier Hennequin, and Joël Spiroux de Vendômois. "Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize."Food and Chemical Toxicology(2012).
European Food Safety Authority. "Review of the Séralini et al. (2012) publication on a 2-year rodent feeding study with glyphosate formulations and GM maize NK603 as published online on 19 September 2012 in Food and Chemical Toxicology". Published October 2012. Visited February 2013.

by Josh DeWald

Share Tweet Reddit

@Skeptoid Media, a 501(c)(3) nonprofit








Want more great stuff like this?

Let us email you a link to each week's new episode. Cancel at any time: