Marion Nestle is a bona fide public expert on nutrition with an earned doctorate in molecular biology no less. She’s also an academic in good standing. She writes this review of a JAMA study in the Atlantic magazine:
The latest issue of JAMA has a paper on a “portfolio” of dietary means to reduce blood cholesterol levels. The paper is likely to get lots of press because it concludes that consuming the “portfolio”–a combination of plant sterols, soy protein, viscous fibers, and nuts–does a better job of lowering LDL-cholesterol (the “bad” kind) than does dietary advice to reduce saturated fat. The paper is unusually difficult to read. I interpret the study in part as a drug trial.
If you follow the link (pdf) Nestle provides, you can read this unusually difficult drug trial for yourself. The paper reports the outcomes from a randomized controlled trial where volunteers were assigned into one of three diet advice programs, followed for 6 months, then measured on a wide variety of indicators, but most particularly on LDL cholesterol. The main finding was that dietary recommendations to eat plant sterols, nuts, and soy proteins led to healthier LDL scores at a Small Windowpane effect size. Thus, making a change to one’s diet makes changes in LDL scores. That diet affects LDL is not news; the news here is that plant sterols, nuts, and soy proteins accomplish this and at about the same effect size as taking a pill. Some people have high LDL, but react badly to medication; some just don’t like pills, but want to lower LDL. Here’s a pill-free method that appears to work in the short term about as well as the pill.
The more interesting element of this rather commonplace event is Nestle’s description of the research as “unusually difficult” and as a “kind of drug trial.” If you are in good standing in the peer review literature there’s no way this report is Unusually Difficult or even Usually Difficult. The researchers do an excellent job describing both the methods and the results and I feel pretty confident that workers in this area could replicate this study without much difficulty. It’s all there. It actually over-reports by the usual JAMA standards, making it both easier to replicate, but also more detailed. Certainly, there’s a lot of stuff in there, but each piece is quite simple and straightforward.
And a “kind of drug trial?” Why would anyone, but most particularly a nutritionist with a doctorate in molecular biology call a food study a drug trial? There’s a huge scientific and legal difference between a Food and a Drug and to see them as similar is to note they are both 4 letter words. The active ingredient in this drug trial was dietary advice – what an expert told a volunteer participant to eat. Even the JAMA editors and reviewers saw this as a communication intervention for dietary change.
We can understand why Nestle might see this paper as an unusually difficult drug study when we note that she notes,
One look at the Abstract and I immediately suspected that this study must have been sponsored by a maker of plant sterol margarines. Bingo!
Another way to say, Bingo, is Ding-Dong! Either term denotes the process. Automatic. Thoughtless. When a for-profit group funds the science, you get a food study that is actually an unusually difficult kinda drug trial. Those lying liars at Big Food bought a team of scientists, then bought the reviewers and editors at JAMA, and snuck this unusually difficult, kinda drug trial in the peer review literature. Those biased, greedy, but effective bastards. (As we’ve studied before on “ghostwriting” scientific reports.)
You see the problem here. JAMA is a peer review research journal that for all its flaws is one of the stronger examples of this form of scientific communication. The scientific community has decided that full disclosure from authors is the best way to handle scientific and ethical challenges. The scientific community does not automatically Ding-Dong and exclude research submissions because of these concerns. Disclosure of bias, real or potential, is how science proceeds.
But, not the NonProfit Science of Nestle. She spots the raccoon in the Abstract. Thank goodness, we have NonProfit science to save us. Real scientists who don’t have financial incentives that bias their work . . . like writing for the Atlantic magazine. Nestle doesn’t get paid through those ads on her Atlantic page? And, writing in the Atlantic doesn’t have any other benefits, does it?
And because my book on calories is coming out next March, I must point out that the study groups reported losing small amounts of weight, which means they must also have reduced their calorie intake. Weight loss alone should help with blood cholesterol.
Hey, scientists, if you want the Truth, don’t read those biased peer review reports with Full Disclosure in JAMA, just wait until March for Nestle’s pop press book! You read about it in the Atlantic magazine! With all that advertising and self promotion. That affects no one’s judgment!
And, yes, as her last quote carefully observes, the research did report that all three groups ate less and lost weight. But only the diet advice for sterols, soy, and nuts led to weight loss and the lower LDL. I guess that’s what makes this report unusually difficult. When the lying liars who are Big Food lackeys say only one group lost weight and lowered LDL, that’s not what it really means. You can see that if you are reading the lines between the lines and not the lines themselves.
Nonprofit science. No bias. Bingo!