Pigs in the real world – feed them different diets, measure many health parameters, some which show differences – but what does it all mean?

12 June 2013. Source:

A long awaited study by Australian Dr Judy Carman has appeared in the open access journal Journal of Organic Systems. It is an American-Australian collaboration, including Carman’s IHER, Howard Vlieger’s non-GMO marketing operation Verity Farms based in Iowa in the United States. It seems to be the same study for which a preliminary report has been presented at a South American scientific conference several years back.

It is an investigation in which pigs were fed a so-called real world diets for ~23 weeks, and analysed for about 35 health related parameters.

In most of the parameters measured on these pigs there is no apparent difference between animals fed a diet that included genetically modified corn and genetically modified soy beans compared to pigs fed conventional mixture of the same grains, but two out of about 35 measured parameters showed a difference.

These are presence of inflammation in the animal gut at autopsy gut and average size of female animal’s ovaries.

The question raised by the study is what are the reasons for these differences. Are they due to chance, because of the random distribution of differences between individual animals: are they caused by the diet, and if they are caused by the diet, or is the indeed transgenic components of the diet that has possible causal effect.

The paper by Carman and colleagues avoids rigourous analysis of whether the differences are attributable to chance.

In the study there is no clear-cut hypothesis about what component(s) of the diet is different and what effect the component might have specifically on the animal.

Instead of a well formulated prior hypothesis the investigation consists of a survey of a fairly large number of parameters -18 are mentioned in one table, 17 in another, and there is no necessary statistical analysis to check for false discovery of effects because of repeated searching for differences.

It’s what some call a fishing expedition in search of a finding, and a known pitfall of animal feeding trials on whole foods.

The individual statistical tests actually done in this study in each of the individual parameters measured do not provide for this false discovery rate effect due to multiple testing testing.

Using the standard criteria of a one in 20 chance that observed differences are randomly generated, about one or two apparent effects in this study might be a false discovery.

The observed differences might also be caused by compositional differences in the variety of soybeans or corn used in the study, and the crucial difficulty with such a complicated study is that there are many components in these animals diets.

Unfortunately there is relatively little information in the paper about nutritional formulation, methods used for producing the pig diets, storage time for the grain and which particular varieties of grain were used in the diets.

A crucial missing piece of information is analysis for soybean isoflavone content. Soybean isoflavones are known compounds with female animal hormone activity, and as some differences were seen in ovary size in these animals, whether or not they have been exposed to different levels of isoflavones in formulating the two test diets is a most obvious question that does not appear to be considered by these investigators.

Because of the complicated way the experiments have been designed for this investigation we don’t know the answer to this question.

The study claims to be an investigation of the real world effects on hog health health. The real world is full of complications when it comes of physiological effects of diets and we still don’t know whether the observed differences between the test group –“GMO diet” as compared with “Conventional diet “– are explained by chance, due to the high number of different types of tests carried out in the animals, or whether some of these differences are caused by grain composition variation — especially soy isoflavone variation.

The most particular difference is a claim about ovary size variation — and the papers authors do not seem to be aware of the very plausible effects of differences between the test and control diets in phyto-oestrogen compounds in the different soybean varieties used in the diets.

Perhaps the most newsworthy nonscientific aspect of the report is the statement about conflicts of interest in which none are claimed.

The Verity Farms non-GMO grain marketing venture linkage is not seen to be a conflict of interest, and a previous revelation on Australian television channel SBS by Dr Carman that the Institute of Responsible Technology associated with Fairfield, Iowa based author Jeffrey Smith was funding such studies is also not mentioned in the paper.

A long-term toxicology study on pigs fed a mixed GM diet. Adverse effects of GM crops found.

11 June 2013. Source: Dr. Judy Carman

This is a briefing about the contents of a new, peer-reviewed scientific paper titled: A long-term toxicology study on pigs fed a combined genetically modified (GM) soy and GM corn maize diet, by Dr Judy Carman, Howard Vlieger, Dr Larry Ver Steeg, Veryln Sneller, Dr Garth Robinson, Dr Kate Clinch- Jones, Dr Julie Haynes and Dr John Edwards.

At a commercial piggery in the US, we took 168 just-weaned pigs and fed them a typical diet for the piggery, containing soy and corn, for 22.7 weeks (over 5 months) until the pigs were slaughtered at their usual slaughter age. Half of the pigs were fed widely-used varieties of GM soy and GM corn (the GM-fed group) for this whole period and the other half of the pigs were fed an equivalent non-GM diet (the control group). The GM diet contained three GM genes and therefore three GM proteins. One protein made the plant resistant to a herbicide and two proteins were insecticides. We chose a mixed diet instead of a single crop because this is usually what pigs and people eat. Regulators do not require animal feeding studies on mixtures of GM genes and their proteins, regardless of whether the genes are all “stacked” into the one plant or spread across several plants that are eaten in the same meal. We chose pigs because they have a similar digestive system to humans, and because some of the investigators had been observing reproductive and digestive problems in pigs fed GM crops. We took blood from the pigs a few days before they were slaughtered to do standard biochemistry tests. Autopsies were done by qualified veterinarians who didn’t know if a given pig was fed the GM diet or not, so their observations were completely unbiased.

Some of the investigators had previously seen a reduced ability to conceive and higher rates of miscarriage in piggeries where sows were fed a GM diet, and a reduction in the number of piglets born if boars were used for conception rather than artificial insemination. Artificial insemination guarantees the presence of a certain number of viable sperm. Because male pigs were neutered at 3 days of age in order to provide meat free of boar-taint, we were only able to look at the female reproductive system in these pigs. We found that, on average, the weight of the uterus of pigs fed the GM diet, as a proportion of the weight of the pig, was 25% higher than the control pigs. We found that this biologically significant finding was also statistically significant. We list some of the pathologies that could be occurring in these uteri in the paper.

Some of the investigators had also previously seen higher rates of intestinal problems in pigs fed a GM diet, including inflammation of the stomach and small intestine, stomach ulcers, a thinning of intestinal walls and an increase in haemorrhagic bowel disease, where a pig can rapidly “bleed-out” from their bowel and die. We weren’t able to look inside the intestines, due to the amount of food in them, but we were able to look inside the stomach. We found that the level of severe inflammation in stomachs was markedly higher in pigs fed the GM diet. Pigs on the GM diet were 2.6 times more likely to get severe stomach inflammation than control pigs. Males were more strongly affected. While female pigs were 2.2 times more likely to get severe stomach inflammation when on the GM diet, males were 4 times more likely. These findings are both biologically significant and statistically significant.

We found that these key findings were not reflected in the standard biochemistry tests that are done in GM feeding studies, probably because standard biochemistry tests provide a poor measure of inflammation and matters associated with uterus size. We did however find a marginally significant change on a measure of liver health in the blood of GM-fed pigs.