Entine’s new “big list of studies” tells us very little about GMO safety, writes Claire Robinson
Jon Entine, PR man for pesticides and GMOs, has come out with another GMO promotional in Forbes, called “The debate about GMO safety is over, thanks to a new trillion-meal study”.
Entine is referring to a new study on the impacts of GMOs on livestock by a former Monsanto employee, Alison Van Eenennaam (Van Eenennaam and Young, 2014). The study, Entine writes, “reviewed 29 years of livestock productivity and health data from both before and after the introduction” of GMOs. These data, Entine adds, represent “more than 100 billion animals covering a period before 1996 when animal feed was 100% non-GMO, and after its introduction when it jumped to 90% and more.”
The findings, according to Entine, show that GM feed is safe and nutritionally equivalent to non-GMO feed, on the grounds that “There was no indication of any unusual trends in the health of animals since 1996 when GMO crops were first harvested. Considering the size of the dataset, it can reasonably be said that the debate over the impact of GE feed on animal health is closed: there is zero extraordinary impact.”
The truth is, however, that the new study tells us almost nothing about GMO safety. Entine’s claim that it shows GMOs are safe is false. The new study should be treated with the same sceptical scrutiny that we should apply to all “big lists of studies” claimed to show GMO safety: the Snell review, the Nicolia review, and the GENERA database.
Generalized claims about what these lists and reviews of studies show are not accompanied by analysis of the individual studies that make up the “big lists”. Such claims rely on readers not examining the studies and seeing for themselves what they contain. If they did, they would see that the studies, taken as a whole, do not support a “one size fits all” conclusion that GMOs are safe.
1. Key data presented by Van Eenennaam and Young are uncontrolled.
The authors show a table of livestock production statistics in the US before and after the introduction of GM feed in 1996. At first glance, everything looks fine. From 1983 to 2011, the weight of cows, pigs and chickens at slaughter has climbed steadily. Milk yield of cows has almost doubled; and the percentage of animals condemned at slaughter on the grounds that they are too unhealthy to be allowed into the food supply has decreased.
This may be reassuring to some, but scientifically it cannot tell us much. Science tests one variable at a time. If you want to know if a GMO is safe for livestock to eat, you feed it to one group of animals and feed another group of animals a similar but non-GMO diet. All other conditions of the experiment – animal housing and husbandry, animal genetics and starting weight, must be the same between the GMO-fed and non-GMO fed groups. Antibiotic use has to be avoided in both groups, since it suppresses inflammation and can hide treatment-related toxic effects.
This is what’s known as a controlled study. If you see differences in the GMO-fed group, you know it’s due to the GMO component of the diet and not to other factors.
Van Eenennaam and Young’s data (Table 4) do not represent a controlled study. The way livestock are bred, kept, and managed has changed massively since the 1983 start date of the study, as explained in point 2) below. These changes mean there are so many uncontrolled variables that the data are almost worthless in assessing GMO safety.
2. The study is blind to important trends in the livestock industry.
US livestock are relatively unhealthy because they are raised in CAFOs (confined animal feed operations). To compensate, they are fed huge amounts of antibiotics – 80% of antibiotics in the US are fed to livestock. One US farmer told GMWatch, “Livestock producers and veterinarians know that as the proportion of GMOs in feed has increased, so has antibiotic use. We don’t know for sure that GMOs are the cause, but we can’t rule them out, either.”
The experience of Danish pig farmer Ib Borup Pedersen suggests that the possibility is real. Pedersen reports reducing antibiotic use by 66% after he changed his herd’s feed to non-GMO, saving enough money to more than cover the increased cost of the non-GMO feed.
Proper toxicological analyses are not carried out on slaughtered animals at abattoirs, so even if they were sick, this would likely be missed. For example, the GMO-fed pigs in Dr Judy Carman’s feeding study had more severe stomach inflammation and heavier uteri at slaughter than the controls. But all the pigs passed inspection at the abattoir and were judged acceptable to enter the meat market. These unnoticed sick animals would simply not show up in Van Eenennaam and Young’s data, since they were not “condemned” at the abattoir. Similarly, some of the main effects of a GMO diet reported by Ib Pedersen were diarrhea and bloat. As long as these pigs could be kept alive with antibiotics up to slaughter, their sicknesses would not show up in the data presented by Van Eenennaam and Young.
Neither does Van Eenennaam and Young’s review take account of animals that farmers don’t bother sending to the abattoir because they know they wouldn’t pass. For example, Ib Pedersen, who reported increased malformations in his pig herd when they were fed GMO feed, did not send his malformed piglets to market. Pedersen also found that sows fed a GMO diet had fewer live-born piglets. Malformed piglets and loss of fertility in sows are potentially catastrophic trends for a pig farmer. Yet Van Eenennaam and Young’s study is effectively blind to them.
3. The study misses long-term health effects of GMOs.
Many of the studies reviewed by Van Eenennaam and Young are not toxicological but animal production studies. Studies of this type do not look at health effects in detail but at issues of interest to the agricultural industry, such as whether a farm animal fed on the GMO will survive to marketing age, gain enough weight, and produce an acceptable milk or meat product. The commercial lifespan of a livestock animal is very short compared with its natural lifespan. These studies are usually short-term in relation to the animal’s natural lifespan and are not able to reveal long-term ill health effects, which take time to show up. They cannot tell us whether the GMO is safe for the animal to eat over the long term, let alone whether it’s safe for humans.
Van Eenennaam and Young seem unaware of this issue. For example, they categorize a 25-month GMO Bt maize feeding study in dairy cows as long-term, even though a cow’s natural lifespan is 17–20 years. They are not alone in this oversight. The review of animal feeding studies on GMOs by Snell and colleagues, which is cited as proof of GMO safety by Van Eenennaam and Young and by Entine in his article, wrongly describes many studies that run for a fraction of the animal’s natural lifespan as “long-term”.
With or without GMOs, it’s clear that the short commercial lifespan of livestock animals hide a multitude of problems. In the US, beef cattle are raised to slaughter weight over several months in feedlots, with rapid weight gain (about 1.5 kg/day). They are fed on "concentrate" – corn, soy and supplements – and antibiotics, for the most part with no clear ill effects. But if they had been kept on those same diets for a prolonged period of time, they would become very ill, even fatally so, because cattle need forage to be healthy. The symptoms of ill health do not usually show up because the cattle only are in the feedlots eating grain for several months. This illustrates how major illness can be masked in short-term animal production studies.
If Entine is reassured by data such as this, he probably believes that smoking and asbestos are safe too. In fact, evidence elsewhere in the article suggests that Entine may not understand the difference between acute toxicity and chronic and low-dose toxicity. He says of Séralini’s study, which found toxic effects in rats from GM maize and tiny amounts of Roundup herbicide fed over a long-term period, that if the “data were real and 80% of food was poison, animals and people would be dropping like flies.”
But this is way off the mark. We know that commercialized GMOs are not acutely toxic. Just as with smoking and asbestos, you don’t keel over dead on a single exposure. Some studies indicate that some GMOs and their associated pesticides might be low-level toxins. If that turns out to be true, then in livestock animals fed GMOs, we’d probably see the kinds of deterioration in health remarked on by the US farmer I interviewed and documented by Ib Pedersen in Denmark. Contrary to Entine’s claim, there are anecdotal reports of such problems. Other farmers may not notice, due to the problems creeping in gradually and the rapid animal turnover of commercial livestock farming.
If GMOs were causing problems in humans, we wouldn’t see a pile of bodies with a label stuck on their foreheads saying, “Killed by GMOs.” We’d see a rise in already common chronic illnesses, such as cancer, intestinal problems, or liver and kidney damage, which would, however, be untraceable to any single cause. Americans have indeed got sicker since the introduction of GMOs, as US government data shows. But there is no way of knowing if there is a connection with GM foods, as they are not labeled in the US. Scientists cannot track consumption and identify possible correlations with rising ill health patterns.
Against this background, it is not clear that the data presented by Van Eenennaam and Young have any value. Probably the only thing that would be picked up by this study is if the GM feed made up a very high proportion of the diet and was acutely toxic. Lower-grade toxicity and long-term toxicity would likely not be detected.
4. The study appears to have no quality control.
A man who reports that there is no spider in the room but has his eyes glued shut is not a reliable witness. A study that concludes a GMO is safe but fails to look for possible harm is not useful. Van Eenennaam and Young rely in part on a study that is not only failing to look, but was carried out in such a way that it would be incapable of finding harm from GMOs. This is the study mentioned above (Steinke and colleagues, 2010), in which dairy cows were fed GM Bt maize. This study is approvingly cited by Van Eenennaam and Young in their review of the literature as “comprehensive” and “thorough”, and as using “appropriate controls” and “sophisticated techniques” to look for the presence of GM material in tissues.
Unfortunately, however, Van Eenennaam and Young appear to miss a cavernous breach of scientific standards in the study. During the course of this experiment, nine cows from the treatment group and nine from the control group – half of the 18 animals in each group – fell ill or proved infertile. These cows were simply removed from the study and replaced with other cows. We don’t know whether the problems that the cows suffered had anything to do with either of the two diets tested, since no analysis is presented.
It is never acceptable to replace animals in a feeding experiment. For this reason alone, no conclusions can be drawn from this study and Van Eenennaam and Young should not have included it in their analysis. This blunder wipes out all claims of comprehensiveness, thoroughness, and use of appropriate controls.
While this is just one study among many in Van Eenennaam and Young’s review, the fact that they included it and singled it out for praise raises questions about their quality control – and throws into question all the other studies in their review. It also lends a whiff of unscientific double standards to Table 5 in their review, which lists alleged shortcomings in studies that have found toxic effects from GMOs. Incidentally, this table contains incorrect information, claiming that Ewen and Pusztai failed to use a non-GMO control crop of “similar genetic background” to the GMO crop in their GMO feeding study on rats, when their paper clearly states that the control crop was the “parent” line – of similar genetic background to the GMO crop.
Van Eenennaam and Young are not the only ones to include Steinke’s flawed dairy cow study in their analysis. They are joined by Snell and colleagues and Nicolia and colleagues, authors of two other reviews of “big lists of studies” that are quoted by pro-GMO lobbyists, including Entine in his article, to support their claims that GMOs are safe. The authors of the Snell and Nicolia reviews, like Van Eenennaam and Young, appear not to notice the gaping hole in its methodology.
Incidents such as this (and many others would come to light if more studies in the “big lists” were analyzed) illustrate the perils of generalizing about science. Honest scientists resist the practice, though as the political and economic imperative to push GM foods comes to dominate our scientific institutions, we are seeing it more and more from pro-GMO scientists-turned-advocates.
5. The study appears to sacrifice scientific rigour to political lobbying.
The speed with which scientific standards are vanishing down the drain in the name of GMO promotion is matched by the synchronous rise of lobbying messages in peer-reviewed papers. The timing of the Van Eenennaam and Young paper is convenient for GMO lobbyists, coming as it does in the middle of the TTIP US-EU trade deal negotiations. One of the US’s main priorities in this deal is to force Europe to drop its GMO regulations, which are far more stringent than the US system. What better message to slip into the discussion than the conclusion of this paper, that GMO animal feed is as safe and nutritional as non-GMO feed?
Just in case the message is overlooked, the authors spell it out, criticizing the “asynchronous regulatory approvals” of GMOs across the globe that result in “trade disruptions”. In simple terms, that refers to incidents in which, for example, the EU or China turn away American GMOs that their governments have not approved for safety.
The paper also takes up arms on behalf of the GMO lobby to remove another obstacle in its path: the prospect that GMOs developed through “targeted genome modifications”, or genome editing, might be regulated as GMOs. Given that the regulatory process for GMOs, in Van Eenennaam and Young’s words, “takes years and costs millions of dollars”, the GMO industry does not want to see these new-type GMOs regulated or labeled as GMOs. But contrary to the paper’s claim, the process of creating a GMO through new techniques like ZFN, CRISPR and TALEN is not “precise” and according to the accepted definitions, the products are certainly GMOs.
The lively discussion of these points in the peer-reviewed literature, however, seems to have gone unnoticed by Van Eenennaam and Young. Perhaps this is because their eyes are fixed on an economic goal: the “urgent need for international harmonization of both regulatory frameworks for GE crops and governance of advanced breeding techniques to prevent widespread disruptions in international trade of livestock feedstuffs in the future”. It’s a conclusion that will be welcomed by the US’s negotiators in the TTIP talks. For the rest of us “eaters”, locked out of the secret negotiations between corporate lobbyists and lawyers who will decide the future of our food, there is Entine’s message that the GMO safety debate is “over”.
The scientific answer to Van Eenennaam and Young, and to Entine, is the same. You must justify your claims about GMO safety with specific data points from specific studies. That’s not an extraordinary expectation: it’s nothing other than basic scientific standards.
NOTE: In his article for Forbes, Entine also relies on the “big lists of studies” in the Snell review, the Nicolia review, and the GENERA database to back up his claims of GMO safety. Detailed analyses of these three sources are as follows:
GMO Myths and Truths on Snell:
GMO Myths and Truths on Nicolia:
Download entire GMO Myths and Truths report:
GMWatch on GENERA:
1. Entine J. The debate about GMO safety Is over, thanks to a new trillion-meal study. Forbes 2014. Available at: http://www.forbes.com/sites/jonentine/2014/09/17/the-debate-about-gmo-safety-is-over-thanks-to-a-new-trillion-meal-study/. Accessed September 22, 2014.
2. Sourcewatch. Alison L. Van Eenennaam. Sourcewatch 2010. Available at: http://www.sourcewatch.org/index.php/Alison_L._Van_Eenennaam.
3. Eenennaam ALV, Young AE. Prevalence and impacts of genetically engineered feedstuffs on livestock populations. J. Anim. Sci. 2014:jas.2014-8124. doi:10.2527/jas.2014-8124.
4. Snell C, Aude B, Bergé J, et al. Assessment of the health impact of GM plant diets in long-term and multigenerational animal feeding trials: A literature review. Food Chem. Toxicol. 2012;50:1134-48.
5. Nicolia A, Manzo A, Veronesi F, Rosellini D. An overview of the last 10 years of genetically engineered crop safety research. Crit. Rev. Biotechnol. 2013:1-12. doi:10.3109/07388551.2013.823595.
6. Biology Fortified, Inc. Welcome to GENERA. GENERA 2014. Available at: http://genera.biofortified.org/.
7. Pedersen IB. Changing from GMO to non-GMO natural soy, experiences from Denmark. Sci. Soc. 2014.
8. Carman JA, Vlieger HR, Ver Steeg LJ, et al. A long-term toxicology study on pigs fed a combined genetically modified (GM) soy and GM maize diet. J. Org. Syst. 2013;8:38–54.
9. Steinke K, Guertler P, Paul V, et al. Effects of long-term feeding of genetically modified corn (event MON810) on the performance of lactating dairy cows. J Anim Physiol Anim Nutr Berl 2010;94:e185-93. doi:10.1111/j.1439-0396.2010.01003.x.
10. Séralini G-E, Clair E, Mesnage R, et al. Republished study: long-term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize. Environ. Sci. Eur. 2014;26(1):14. doi:10.1186/s12302-014-0014-5.
11. Samsel A, Seneff S. Glyphosate, pathways to modern diseases II: Celiac sprue and gluten intolerance. Interdiscip. Toxicol. 2013;6(4):159-184. doi:10.2478/intox-2013-0026.
12. Ewen SW, Pusztai A. Effect of diets containing genetically modified potatoes expressing Galanthus nivalis lectin on rat small intestine. Lancet 1999;354:1353-4. doi:10.1016/S0140-6736(98)05860-7.
13. Harvey F, Correspondent E. EU under pressure to allow GM food imports from US and Canada. The Guardian 2014. Available at: http://www.theguardian.com/environment/2014/sep/05/eu-gm-food-imports-us-canada. Accessed September 26, 2014.
14. Fagan J, Antoniou M, Robinson C. 1.2 Myth: Genetic engineering is precise and the results are predictable; Truth: Genetic engineering is crude and imprecise, and the results are unpredictable. In: GMO Myths and Truths. 2nd ed. London, UK; 2014:38–41. Available at: http://bit.ly/Yjc96j.