Tuesday, October 7, 2014

GMO junk science meets junk journalism

The debate about gmo safety is over! 

At the beginning of September, I met Kevin Folta, a plant scientist- who has taken an active role on the internet educating the public about  GMOs.. We talked about GMOS science over lunch and as the scientist was getting ready to leave  mentioned that a new study was about to be published on GMOs involving BILLIONS of animals. BILLIONS of animals!

-- I couldn't even imagine the reams of paper required to assess the health of BILLIONS.
 But....within a day the study popped up on Twitter and the charade began.

As it turned out the new data fit on just a few pages.

The author.
What happens when 100 billion animals, over 18 years, eat GMOs? http://t.co/GScfLlbhhc via @GeneticLiteracy

Looking under the hood of the billions of  animals  chickens  clunker  study

United States
Layer Hens

The fist thing to note is what the author actually means by "100 billion animals"

94.7% or  94,683,600,000 of the 100, 000,000,000  are broiler chickens, which the author doesn't disclose for 37 pages (Table 3) and neither does the media echo-chamber. 


This is poultry whose natural lifespan is about 5 years, slaughtered at less than 49 days of age. So, even if the study reported convincing health data, it would remain a very short term study. In other words, it is 19 years-worth of 49 day old chickens, which is quite different than 19 years of life-long animal studies.

The "study" is a hodgepodge of livestock production data  used to bolster the safety alleged from a literature review. The rest is  polemics on international harmonization, with a particular emphasis on Ukraine. The author is making a case for "harmonization" -- actually, code for take-over of Ukranian agriculture by international corporations- Monsanto, Dupont, Cargil --concomitant with  the relaxation of already outrageously unacceptable, crude and  inadequate safety testing standards.

The author recommends elimination of the 90-day rodent feeding trials- described  as "interference in risk assessment based on pseudoscience",  as well as  the long-term studies  " due to additional time, expense and animal experimentation." .

Health  Livestock production data on billions of chickens. 

What I expected to see

The whole data set for nearly 95% of the "animals"  is summarized in Table 4 and  Figure 2, from statistics reported by the Chicken Council- a poultry industry trade group- hardly a source of medical quality scholarship.

 The data is here: http://www.nationalchickencouncil.org/about-the-industry/statistics/u-s-broiler-performance/

Unsurprisingly, what I actually found

  • days to market
  • feed efficiency (feed to grain ratio)
  • percent mortality
  • carcass weight

Poultry statistics are used to prove the following hypothesis:  " It would be reasonable to hypothesize that if animal feed derived from GE crops had deleterious effects on animals consuming GE feed, then animal performance and health attributes in these commercial livestock populations would have been negatively impacted" 

  •  Firstly, livestock performance is not a marker of health- because the goal of livestock production is minimizing inputs and maximizing production of meat, eggs or milk regardless of costs to the animal's health or longevity.  For the rest of us, who aren't slaughtered at 49 days, the goals are completely different.
  • extrapolation of findings in livestock to non-livestock- either pets or humans- is absurd, scientifically indefensible and unreasonable! 

  •  It is well known in small animal medicine that rapid growth as indicated by feed efficiency, while desirable in livestock,  is deleterious to health thus, we restrict caloric intake to control growth in order to avoid joint and other diseases. So, feed efficiency is completely irrelevant to pet and human health. 

  • Interestingly, if one were to literally assume that GMOs lead to improved weight gain- as is seen  in poultry statistics- grossly extrapolating that hypothesis to humans and pets, we could attribute the obesity epidemic to GMOs.     

  • But since the traits aren't segregated, if  a particular crop reduced feed efficiency it would go undetected.  An example of this is high oleic GMO soybeans produced by Dupont  reported to  reduce feed efficiency  Pages 37-38 of the linked document summarizing two separate feeding studies on pigs and chickens. It may be important that in both studies Dupont supplemented the test diets with too little essential lysine. The implication is that these GM soybeans were also modified to produce higher levels of lysine so less lysine should have been necessary. Was it just an error, or was it to damp down the appearance of negative effects of the diet? Or were the higher levels of lysine toxic due to reactions at the processing temperatures? If you leave out an obvious important nutrient and the birds are struggling on the control diet, then the difference between the non-GM and GM may not be so obvious.

The GM soybeans in the pig study had a lower nutritive value at normal processing temperatures (soybeans are always processed into a cake for animal feed). It is unknown whether the effects were nutritive or toxic.  And it is known that some toxins only appear at some temperatures. The first table (13)  reports F/G ratio, which is an “indication of how much food (in pounds) it takes to put on 1 lb of body weight in the animal”. At the indicated temperature, animals fed on heated high oleic acid soybeans often were less able to convert food energy into body mass. This only occurred in the GM soybean. An explanation wasn't offered.

A similar effect was seen with chickens. When pigs or chickens were fed the GM high oleic acid soybeans, the efficiency of feed conversion fell in comparison to control diets. At the time, FSANZ attributed the effect on chickens to lower amino acid levels in the test diets. The tested varieties of high oleic acid soybeans were also modified to produce higher levels of lysine. At temperatures used to cook soybeans and corn,  lysine will react with sugars to both reduce the amount of available lysine and also create potentially toxic compounds. The key point here is that the cause of the processing effect on high oleic acid soybeans was never determined. It could have been an effect of lysine at high temperature. It may have been caused by other intended or unintended effects of silencing the fad2-1 gene, high oleic acid or high linoleic acid isomer 9,15 alone or in combination with lysine.  This particular soybean was withdrawn from the market soon after approval.  ( J. Heinemann, personal communication) Was it included in this data set?  -Who knows!

    So we can't say that GMOs improve feed efficiency as a category, because some traits actually reduce it, Each trait  needs to be evaluated individually- rather than be lumped into a giant category of GMOs as was done in this study which presumably compares livestock production data for the entire herd "before GMOs"  to "after GMOs", as articulated in the "hypothesis" below. 

Beyond wishful thinking, the author doesn't convincingly explain how  these exceptionally limited parameters encompass  health. Health is measurable with medical tests (bloodwork, microbiology, imaging such as X-rays,  biopsies/ histopathology), but no medical tests are cited in this "study".

It should be  intuitively obvious by now that  even if  medical tests were reported--wholly absent from this study they are entirely meaningless to any animal other than a 49-day old  broiler chicken, never mind a human.

And there are reasons chickens aren't a used in preliminary testing of effects of substances on humans--they are a very poor model for human health.

The remaining 5% of the animals in the data set are dairy cattle and beef cattle. The authors cobbled  together data from  disparate sources,  and when data wasn't available-simply made it up.

Page 11 states " data on cattle condemnation rates were available for 1999-2002.

Data for 2003-2007 is based on FOIA.

Data for 1994 (before GMOs)  was collected  from National Non-Fed Beef Quality Audit--but these animals do not spend  time  in  feedlots!
Since there was no data for cattle for 1983-1994 or 2000-2011, claims that this study examined parameters for 19 years is a lie. Data was unavailable for cattle before GMOs were introduced, so the authors fills in a 22 year gap with  regression analysis  -a projection- for 11 years (1983-1994) before GMOs were introduced, as well as for 11 years  (2000-2011) after.

The only definitive data was for 1999-2002, even if condemnation rates were associated with animal health, but they aren't. Speed of slaughter lines and the number of assigned federal meat inspectors whose numbers have been dwindling. affect condemnation rates which do not substitute for proper post-mortem analysis with histopathology.

The only parameter found for the duration  was  milk yield.

Somatic Cell Counts  were reported for a 1995-2011, thus, there is no data "before GMOs" and condemnation rates for 1998-2006, after the introduction of GMOs.
SCC is used to report mastitis ( breast infection), but it can't differentiate mastitis due to good/ poor animal health resulting from impacts of GMOs  on health from  hygiene,  milking equipment sanitation or antibiotic use. These confounding factors make the association of this test with GMOs spurious.

  •  Milk yield and SCC certainly  do not reflect the animal's overall health  indirectly indicated by longevity, since dairy is culled at 4-5years of the age of a 15-year normal lifespan. One most certainly can't use milk parameters for  males who don't produce milk at all.

  • Clearly, milk tests do not evaluate the function of the metabolic organs-kidneys, liver, pancreas, the endocrine system, GI tract, bone marrow, the musculoskeletal system  or the immune system any organ outside of mammary glands.

  • Evaluation of milk parameters omits half the population--the entire male population that doesn't produce any milk.

Meanwhile, veterinary health data has been collected by the FSIS and it shows remarkably high rates of malignant lymphoma in livestock

But appropriate data was only collected for five years:  1998-2002. So there is no long-term data useful in analyzing long- term impacts of GMOs on livestock health. 


The study doesn't  follow the rigidly defined science journal format. Indeed, most of the text focus is on  international trade rather than science- more appropriate for The Wall Street Journal or Forbes, rather than a peer-reviewed journal.

The Hypothesis is conventionally articulated in the abstract, expanded in the introduction, and the remainder of a journal article usually describes materials and methods  used to prove the hypothesis. Then, we expect to see results and an explanation of results in the discussion section.

None of these are found in this study- the Methods and Materials section doesn't exist, while the hypothesis itself is hidden on page 10 of the introduction: " It would be reasonable to hypothesize that if animal feed derived from GE crops had deleterious effects on animals consuming GE feed, then animal performance and health attributes in these commercial livestock populations would have been negatively impacted" 

There isn't a single citation  showing a precedent for using  livestock production as a proxy for animal or human health--an outrageous claim!

  Irrelevant, indirect data is missing for meager livestock parameters reported which are not parameters of overall health-- and there is no medical testing or medical data of any kind permitting assessment of overall animal health.

To recap, joint disease is common in poultry, for example, because of selection for growth--but would be devastating for humans.  The hypothesis that milk production encompasses health is unreasonable and  is not backed up by scientific evidence. Additionally, confounding factors weren't segregated out, so it's hard to attribute these limited livestock parameters to GMOs.

Given the glaring gross science  ineptitude,  promotion of  exceptionally poor scientific standards, the obvious bias and  explicit agenda to force GMOs into war-torn Ukraine & European Union, I felt reading the author's lengthy science illiterate opinion on GMO safety literature to be a  complete waste of  time.

Quite unsurprisingly- the body of solid work by Manuella Malatesta  demonstrative of hepatic and pancreatic harm by GMO soybeans, which were reversible on cross-over analysis,  were dismissed- while  flawed studies and a flawed review (Snell et al) that purport not to show harm were accepted.

 The Snell et al review cites  Monsanto's own  short-term studies  actually suggestive of  kidney and liver harm induced in rats in under 90 days, as well as other studies reporting kidney damage and other forms of medical harm caused by GMOs.

Tweeterverse and the echo chamber explode in a convulsion of GMO celebrations.

Livestock health has improved since the introduction of genetically engineered crops...TRILLIONS of meals show no harm to animals or humans!

Livestock health has improved since the introduction of genetically engineered crops. http://t.co/zfGjzFcoXa #GMO

The hype circled the globe reaching Russia within ten days.

Ukraine has historically been Russia's bread basket. Russia criminalized the sale of  unlabeled GMOs imposing a ban on their cultivation. Meanwhile, the new Western-imposed junta government in Kiev hurriedly passed laws to ease entry of GMOs into Ukraine, as well as using it as a backdoor into Europe. While Ukraine is in the midst of a civil war, it is a perfect target for deregulation of GMOs, which the author expresses cloaked in the language of "harmonization".


This study purported to show lack of harm from GMOs to billions of animals-  majority of whom were 49 day old chickens-- while the GMO propaganda machine wildly exaggerated this substandard pile of bovine manure to also mean lack of harm to people- even though he study doesn't cite any traditionally accepted medical tests employed to evaluate animal or human health, and its unstated goal is to argue for "harmonization" of Ukraine with our embarrassing junk science standards.

The author bound arbitrary, incongruent  livestock production data with spit, bailing wire and zip-ties,  to arrive at a predetermined conclusion that GMOs do not cause harm.

Finally, some good news-- the public portrayed as science illiterate, has a good bullshit detector, as this  image illustrates.