One of the things that I find most disappointing in any debate is the realisation that somewhere along the line somebody knows that what they are saying is not true. It’s the reason that I get angry at comments from the Catholic church about the ineffectiveness of using condoms against HIV, and it’s the reason that I get pretty frustrated by large parts of the anti-GM lobby too.
We’ve all heard claims that GM foods aren’t safe because they aren’t properly tested yet, or they haven’t been independently validated by scientists with nothing to be gained from their success. In 2013 there are around 600 peer-reviewed journal articles documenting the safety of genetically modified groups. Of these, around a third were funded by independent organisations. Around 3 billion GM meals have been eaten (since the vast majority of American soy and maize is now GM) without a single human health law suit. This is not to say that the case is closed and there’s nothing left to be learned, just that the public perception about these things is remarkably skewed.
Rats with cancer
One of the most well known studies claiming that GM foods are dangerous was a paper last year by Séralini et al, looking at the impact of feeding rats GM glyphosate-resistant corn (NK603). While the study showed a concerning development of tumours in the rats, it failed to take into account that the strain of rats being used (Sprague-Dawley) were specifically bred to develop mammary tumours! (For use in cancer-treatment studies). The rates of tumour development in both the GM group and the control group are normal for that strain of rat, and interestingly the highest rate of tumours were seen with the lowest ‘dose’ of GM corn. The incidence of tumours in this rat strain is known to be strongly affected by feeding regime: which was not documented in the paper. If that wasn’t bad enough, the stats in the paper are fairly scarily skewed: There were 9 times as many GM rats as control group rats (so it’s interesting that the study reported that a GM rat died first…) Finally there’s an awful lot of missing data in the paper. A ll of these things were picked up by regulators (from the EFSA; FSANZ and Health Canada), but when Russia and Kenya banned the import of transgenic foods they mentioned the paper as having influenced their decisions.
All of this worries me, not because of the lay-people talking about the study, but because it really does sound like the scientists, and potentially some of those promoting the paper know how fundamentally flawed it is and don’t care, because it’s getting the results that they want!
Anyway, it seems we have another similar story to the rat one on our hands.
Pigs with tummy upset
A long-term toxicology study on pigs fed a combined genetically modified (GM) soy and GM maize diet is a new paper by Judy Carman et al in which pigs were fed either GM corn and soya or non GM corn and soya. Once again, there’s either some untidy paper writing going on or some rather dodgy science because the authors make no mention of the cultivars of grain used to feed the animals – or whether they were consistent between the two halves of the trial. In fact, getting equivalent seed for each type is fairly challenging, so we’d expect the authors to produce a detailed description of where the grains were grown and/or their nutritional quality, which they don’t. The fact that the paper is published in a journal so minor (Journal of Organic Systems) that it doesn’t appear on Web of Science (the scientist’s Google) sets my ears pricking up too. And then there’s the small issue about the journal being sponsored by the Organic Federation of Australia… Umm impartial?
Anyway the headline of the paper comes from the fact that 32% of the GM-pigs (c.f. 12% of the non-GM) had severe stomach inflammation. So far, so worrying. But then we take a closer look at the categories into which stomach inflammation are divided: Of the non-GM pigs, 31 have ‘mild’ inflammation, and 29 have ‘moderate’ inflammation. Looking at the GM group… 23 and 18 pigs. It’s only in the severe stomach inflammation that the scales tip, so what we should be looking at is some kind of multi-factor statistics, which we just don’t get.
So the stats are crazy and all over the place, but actually I’m kinda concerned anyway because around 52-56% of both sets of pigs have stomach inflammation. These clearly aren’t healthy, happy pigs! It is also recorded that almost 60% of the animals had pneumonia at the time of death. And I wouldn’t pretend to know what a normal pig mortality rate looks like, but 13-14% sounds pretty high too.
I hate that we have a scientifically uninformed public, and I hate that we have an anti-GM lobby with such sway. But I also hate that we (well, Australia) have scientists who are so certain that they are right that they are willing to do inappropriate stats on poorly designed experiments in order to generate prophecy-fulfilling data.