August 25, 2004
Pop Goes The Bias
It'd be hard to miss the blaring headlines that trumpet the latest anti-soda scare study -- this time supposedly linking increased consumption with type 2 diabetes. Once again, this is a case of biased researchers pushing frenzy over facts. As we pointed out in June (when the results were discussed before publication), five of the study's seven coauthors are genuine obesity alarmists, and some have close ties to the self-described "food police" at the Center for Science in the Public Interest (CSPI). One of those coauthors and food-cop friends is Walter Willett, who's using this questionable study to advocate a truly radical agenda.
Here's what Willett is telling the press about soda: "The message is: Anyone who cares about their health or the health of their family would not consume these beverages." That's just plain nuts. As we explained in the Houston Chronicle today:
This is fizzy science to promote an anti-soda agenda that tries to scare Americans without the data to back it up ... It's the typical hysterical rhetoric, without solid evidence, that indicates a biased agenda by nutritional Puritans.While the study's authors are spinning the media like tops in an effort to pump up fear of soft drinks, a key finding was buried in the article's data tables: For non-obese women (those who are "normal" weight and "overweight"), there was no statistically significant relationship between soda consumption and type 2 diabetes. In other words, the study's own data shows that soda consumption has nothing to do with diabetes in the vast majority of women. And even in obese women, the authors concede that their observations about type 2 diabetes may "reflect dietary and lifestyle changes accompanying changes in soft drink consumption" -- rather than soda consumption itself.
But no matter. Willett and company boast an extensive history of anti-soda activism, and they don't seem interested in pointing out the gaps in their own data. As far back as 1998 Willett signed a CSPI letter condemning sugar consumption, and he played a starring role in one of CSPI's recent press conferences. Along with CSPI, he endorsed a petition from Ralph Nader's Commercial Alert in support of global restrictions on food marketing. (To learn about potential biases of the other coauthors, click here.)
Michigan State University professor Jon Robison harshly condemned the soda-bashing study for adding to "the feeding frenzy that drives our nation's love affair with epidemiological risk factorology." He writes today:
This article is a textbook case study in the misuse of epidemiological research for the development of health recommendations for the public. The article is strewn with misleading and sometimes inaccurate statements and enough statistical hocus pocus to make all but the most adept junk-science sleuth dizzy.
