The headlines have 'smoking gun' written all over them. "GMO feed turns pig stomachs to mush!" says one. And, right here at Huffington, "Damning new study demonstrates harm to animals raised on GMO feed."
Take a close look at the study and you will find that it is indeed a stinging indictment -- but not of GM feed. Read the paper, look at the data, and what you come away with is hard evidence that factory farming is a terrible way to raise pigs.
As for the correlation between inflammation and GM feed, that's a preliminary inkling, at best. Although this study appears to do a lot of things right, it has a glaring methodological problem: it has no hypothesis.
One hallmark of the kind of research that finds causal links is that it puts forward a hypothesis and looks for data to support or contradict; evaluation criteria are defined before the trial begins. But Carman et al. set out to do a "toxicology study," and planned to look at every outcome they could measure. I asked Dr. David Tovey, Editor-in-Chief of The Cochrane Library, which evaluates research and performs meta-analyses, about it. "Whenever you measure a large number of outcomes you increase the risk of chance findings," he said. "This study certainly is at risk of this."
The UK Science Media Centre concluded that "The paper does not support the claim that GM crops cause stomach inflammation or increased uterus weight," and quotes several statistics and animal health experts who are doubtful about the findings. Patrick Wolfe, professor of statistics at University College London, questions the statistical methods and concludes that "there is a higher-than-reported likelihood that the results are due purely to chance." Tom Sanders, of Kings College, London, concurs: "It seems unlikely that the effects observed were treatment-related."
All told, more of the the non-GM pigs (69) than the GM pigs (64) had stomach inflammation, and the researchers found that the GM pigs fared worse only once they parsed the groups into mild, moderate, and severe. The authors give short shrift to the many, many outcomes that were virtually identical, yet they do mention that two of the GM-fed pigs had a fluid-filled uterus, although that number is not statistically significant. What they don't mention is that the GM-fed pigs had about half the levels of heart, liver, and spleen abnormalities. Also not statistically significant, but why cherry-pick?
That said, if GM feed does cause stomach inflammation -- and this paper defintely raises that possibility -- we need to know about it. It's a good reason for another study, one that does have a hypothesis going in and does define its outcomes ahead of time.
Meantime, what the paper shows unequivocally is that factory farming is bad for pigs. The 145 animals in this study were raised "according to usual industry practices," and 133 had stomach inflammation. There were also many ulcers and erosions (a pre-ulcerous condition), and a good sprinkling of abnormalities in organs and lymph nodes. Eighty-five of the pigs had pneumonia. And one in eight died. Walter Jeffries, who raises pigs on pasture at Sugar Mountain Farm, says, "The general health of these hogs is a bit depressing, no matter which diet they ate."
I hope someone follows up on the stomach inflammation issue, and sooner rather than later. But the larger problem, the urgent problem, is those "usual industry practices." The way we treat our pigs - their lifestyle, their happiness, and their overall health - is much more important than the genes in their grains.