Lipstick on a Pig Study
|

This week, an Australian group published a study that claimed that pigs that were fed a GM diet developed inflamed stomachs and larger uteri. Does this mean genetically modified foods bad for you?

Without even attempting to answer this, it is clear that some people believe that they are. But what is this belief based on?

Unfortunately, the "answer" is a combination of agenda-driven science and reporting--not real science. And sometimes money.

The study appeared in the obscure Journal of Organic Systems--a journal heavily funded by the giant organic food industry. But even a cursory look at the actual study data revealed that, despite the flashy headlines, there is nothing there. Nothing.

Nonetheless, it still generated quite a bit of press coverage, which was undoubtedly the point of the whole exercise. There were thousands of online references to this study, and virtually all of them said the same thing: "Pigs fed GMO feed found to have severely inflamed intestines."

Too bad it's all garbage. By selective reporting of data, the authors managed to fool pretty much everyone. Here's how:

A group of 168 pigs was divided into two groups--half ate a "normal" diet and half ate the identical diet, except the corn and soy in their diet were genetically modified. After 23 weeks, the pigs were sacrificed and examined.

The conclusion: Pigs that ate the GM diet were more than twice as likely to develop severe inflammation of the stomach and also slightly more likely to have a heavier uterus.

Can this really be true? Does GM food really damage pig stomachs or cause enlarged uteri? Do the data match the headlines? Not even close. The devil is in the details, and upon closer inspection, the conclusion goes to hell. Here's why:

As part of the autopsy, the group examined the weights of 8 pig organs. There was a statistically significant difference in only one organ--the uterus--and just barely-- 0.12 percent of the GM-fed animal's body weight compared to 0.10 percent from the other group. Statistical significance is a mathematical method of teasing out real data from numbers that arise by chance. It is helpful, but not foolproof.

Worse still, 18 categories of pathological abnormalities were measured, but the authors found only one category with a significant difference--severe gastric inflammation.

True--GM-fed pigs were more than twice as likely as non-GM fed animals to develop "severe" stomach inflammation. But the GM-fed pigs also had a lower (although not statistically significant) incidence of both more severe stomach conditions (erosions, ulcers) and less severe conditions (mild and moderate inflammation). And twice as many pigs fed GM diets showed no inflammation.

Does this make sense? No--and it shouldn't. This is because the authors based their conclusion on two pieces of statistically significant data, and pretty much ignored 95 percent of the rest of the study. It's a common trick.

If you perform enough measurements in any study, it is virtually certain that something statistically significant will show up, just by chance. But when the remaining data fail to support (or even contradict) these findings, the conclusions become meaningless. Which the authors clearly knew, yet still presented it as fact.

Sadly, this is the typical way that agenda-driven science is done these days, and it works. Sloppy or ideology-driven journalists report these "findings" as facts and they become just that in the collective consciousness of non-scientists around the world.

As such, complex and important medical and scientific issues do not get a place at the table, leaving behind nothing but confusion and bad information.

No wonder no one knows what to believe anymore.


keep in touch     Follow Us on Twitter  Facebook  Facebook


Our Research

Rhetoric and Reality—The Obamacare Evaluation Project: Cost
by Paul Howard, Yevgeniy Feyman, March 2013


Warning: mysql_connect(): Unknown MySQL server host 'tmiweb52.vwh.net' (2) in /home/medicalp/public_html/incs/reports_home.php on line 17
Unknown MySQL server host 'tmiweb52.vwh.net' (2)
Archives

Blogroll

American Council on Science and Health
in the Pipeline
Drugwonks
Pharmalot
Reason – Peter Suderman
WSJ Health Blog
The Hill’s Healthwatch
Forbes ScienceBiz
The Apothecary
EyeOnFDA
KevinMD
Marginal Revolution
Megan McArdle
LifeSci VC
Critical Condition
EconLog
In Vivo Blog
PharmaGossip
Pharma Strategy Blog
Drug Discovery Opinion