AtkinsFacts.org Ignores the 34 studies “Supporting Atkins”
In fact, as documented on the Atkins website, there are currently no fewer than thirty-four studies demonstrating the weight loss and other health benefits – and absence of adverse health effects – of a low-carbohydrate diet.
The “no less than some number of studies support the Atkins Diet” line seems to be the Atkins corporate mantra.[629-632] That’s how you responded when the American Heart Association condemned your diet. That’s how you responded you were sued for allegedly lying to the public about the cardiac risks associated with your diet. That’s how you responded to evidence that the Atkins Diet may make women infertile. That’s how you responded to the recent formation of a coalition of leading non-profit consumer, nutrition and public health organizations formed to combat the “low-carb hype.”
In citing studies you claim support your diet, you cast your position as scientific, though no major governmental or nonprofit medical, nutrition, or science-based organization in the world agrees. A 2004 review in the Journal of the American College of Cardiology concluded the Atkins Diet “runs counter to all the current evidence-based dietary recommendations.”
So even if there were 34 studies published in peer-reviewed scientific journals supporting the Atkins Diet–just like the 34 studies showing that tobacco can be good for you in certain circumstances–independent systematic reviews of the entirety of scientific evidence support neither smoking nor the Atkins Diet for one’s health. Still, since these studies seem to be the basis of your defense, let us to look at these studies in greater detail.
Over a Quarter of the “Research Studies” Are Not Even Published
First of all, 9 of the 34 cited studies “supporting Atkins” are not published studies at all,[643-652] but merely abstracts (brief paragraph(s) written about an unpublished study) presented at meetings, which in general, wrote an editor of the New England Journal of Medicine, “must be presumed to be unreliable sources of public information.” This is another tried and true tobacco industry strategy.
Information provided in abstracts is considered “of little use to the general medical community” because the information presented is “limited and insufficient to allow critical appraisal of the work.” “Abstracts of papers presented at meeting,” concluded a review in the Journal of the American Medical Association, “have been seen as an untrustworthy basis for scientific communication.”
Another review states: “the validity of material presented in abstract form but not published is difficult to evaluate; abstracts usually have not undergone rigorous peer review, and citation of data found only in abstract form may be misleading or inappropriate. Last, several studies have shown that abstracts are less likely to be published if their results are considered negative rather than positive, leading to potential problems with publication bias when meta-analyses of the existing literature are performed.” Abstracts “cannot and should not be considered equivalent to peer-reviewed articles.”
Meeting abstracts cannot be examined for their validity nor reviewed for their credibility. “Because abstracts have not undergone the same stringent peer review as have full-length articles, one cannot be sure of the validity of the abstracts’ methods, results, or conclusions.” As such, data presented in meeting abstracts should be viewed with caution. As one medical journal editor concluded, the guiding principle when considering abstract credibility is “caveat emptor.” “It is hard to criticize a formal manuscript,” one medical journal editorial reads, “when none has been offered.”
Less than half of studies presented as abstracts may ever be published at all. As the New England Journal of Medicine editorialized, “Presumably, most of the other papers were so flawed or so unimportant that either they were never submitted for publication or they were rejected as a result of peer review.” If and when the studies are published, one finds that the abstracts may routinely overstate the case.
The purpose of meeting abstracts is to stimulate dialogue, not to be used to support a position or product, as you seem to be doing. A past editor of the New England Journal of Medicine considered abstracts so unreliable that he felt that news media shouldn’t even report research findings presented only in abstract form. Conference abstracts “were never meant to be viewed in the same light as full-length, peer-reviewed articles.”
The “Uniform Requirements for Manuscripts Submitted to Biomedical Journals” drafted by the International Committee of Medical Journal Editors and in use by over 500 journals specifically admonishes researchers to “Avoid using abstracts as references.” The validity of abstracts is so questionable that some medical journals entirely prohibit authors from even citing them. Yet abstracts make up over a quarter of the “research studies” that you use as references to “support” the Atkins Diet.
Others Were Published in a Journal Founded by an Atkins Spokesman
The remaining studies were published in peer-reviewed journals. Four of these “peer-reviewed” studies, however, were published in a journal we had never heard of called “Metabolic Syndrome and Related Disorders.” When we couldn’t find it in Harvard’s medical library, which boasts 26,000 serial titles, or indeed anywhere in the medical mecca of Boston, we became curious.
We called the publisher. We asked them if in fact a single medical library in the country carried their journal. They confirmed that Atkins Nutritionals, Inc. had an active subscription, but that the only medical library that seemed to carry their journal, they said, was one in Alabama. We checked. They don’t.
In fact, the journal isn’t even indexed by the U.S. National Library of Medicine’s Index Medicus or any of the other major medical databases, which contain over 12,000 titles. We clearly had more detective work to do.
It turns out the journal was founded by Dr. Eric Freedland, a member and self-described representative of THINCS, THe International Network of Cholesterol Skeptics. The Cholesterol Skeptics deny that cholesterol and saturated fat cause heart disease.
One medical journal editorial entitled “Cholesterol Myth Club on Par with Flat Earth Society” reads “as mixed up as Flat Earth Society members obviously are, at least you can laugh their dumb idea off, and if you want to believe the Earth is flat, this view is not going to cause serious problems like… coronary artery disease.”
Dr. Eric Freedland, founder and still Associate Editor of Metabolic Syndrome and Related Disorders, has not only been a featured speaker at conferences sponsored by the Atkins Foundation but reportedly was the president of your Atkins At Home program, where followers spend a thousand dollars a month to get Atkins meals delivered to their door. Dr. Freedland has also been reportedly listed as one of your official spokesmen.
Is it a coincidence that the journal “dedicated its premiere issue to the life’s work of Dr. Atkins?” The Atkins issue was organized by Dr. Freedland, who wrote the issue’s “Tribute to Robert C. Atkins, M.D.” and dedicated it to Atkins’ memory. “I have had the privilege of personally knowing Dr. Atkins,” Freedman wrote, “and it is indeed sad to lose a loved one…” He wrote about the responsibility to “carry the torch and continue his vision.”
This is the “prestigious” “authoritative peer-reviewed journal” that you boast published research supporting Atkins? A journal reportedly founded by one of your paid spokesmen? The journal even featured a Commentary by Atkins’ widow. Veronica Atkins admitted that her late husband never published anything in any peer reviewed medical journal but that she was “delighted to be able to say that I have now accomplished this on his behalf!”
Most of the Studies Were Published by Atkins-Funded Researchers
Most of the 34 cited articles were published by Atkins-funded researchers–those given money by you directly or through the Dr. Robert C. Atkins Foundation.[691-710] Another 6 of the studies did not reveal the source of their funding.[711-716] (less than half of major scientific and medical journals require disclosure of conflicts of interest).
As you know, the Atkins Foundation was started by Dr. Atkins and has given millions of dollars to researchers to, in the words of co-founder Veronica Atkins, “prove Dr. Atkins right.”
Asbestos corporations fund, publish and cite studies that downplay the risks of asbestos. Chemical companies fund, publish and cite studies that downplay the risks of their products. Tobacco corporations fund, publish and cite studies that downplay the risks of tobacco. It is not surprising that you, the billion-dollar Atkins Corporation, fund, publish and cite studies that downplay the risks of your product, the Atkins Diet.
Corporations have had a long history of misrepresenting science. Speaking at a trial at which DuPont was fined over $100 million for a “clear pattern of concealment and misrepresentation,” DuPont’s CEO said “When DuPont says what the science means, that is what the science is.”
DuPont was actually one of the founders of the Formaldehyde Institute, which collaborated with other industry organizations to fund over 30 studies that downplayed the risks of formaldehyde. General Motors was able to pay for research showing that leaded gasoline was safe. The asbestos industry funded about a dozen studies that unanimously denied that asbestos caused lung cancer. The non-industry funded studies, of course, found just the opposite.
Corporate sponsorship can overtly or covertly influence the conduct and publication of research in a variety of ways. In particular, investigators may fear that future funding may be denied if they publish data unfavorable to the hand that feeds them. A study showed that the odds that researchers with tobacco industry affiliations would conclude that second-hand smoke was harmless, for example, are almost a hundred times higher than the odds non-affiliated researchers would come to such a conclusion.
Just because a corporation provides funding for research about its product, though, does that mean that the research is necessarily tainted? A historic trial in 1998 offered an unprecedented opportunity to answer that question. On May 8th, 1998, the tobacco industry “surrendered” to the State of Minnesota. As part of the settlement, more than 30 million pages of previously secret, internal corporate documents were released to the public. For the first time this allowed consumer advocates to find out just how much sway corporate interests had over academic research.
The documents detail a strategy of deceit and cover-up. To combat the wealth of evidence showing that cigarette smoking caused lung cancer, U.S. tobacco companies established entities like the Tobacco Institute. Exposed confidential reports showed that one of the industry’s key strategies would be to fund and publish its own scientific research to downplay the risks, “enhance its credibility,” and “improve its public image.”
U.S. tobacco companies formed the Council for Tobacco Research ostensibly to “fund independent scientific research on the health effects of smoking.” Not surprisingly, though, internal documents showed that their true aim was to create studies specifically designed to produce data to support their arguments. The Atkins Corporation seems to be taking tips from the tobacco industry.
You claim that you are relying on “sound science.” This is the same name Philip Morris’ public relations firm gave to a campaign to downplay the risks of cigarette smoke and “manipulate the scientific standards of proof for the corporate interests of their clients.” Your attempt to recruit doctors into your “Atkins Physicians Council” to downplay the risks of the Atkins Diet is reminiscent of Philip Morris’s secret campaign code named “Project Whitecoat.”
Though the products of both industries have been condemned by the American Cancer Society, Heart Association, and Medical Association, you and the tobacco corporations both built up networks of scientists sympathetic to your position, funded independent organizations to give an impression of legitimate, unbiased science, and organized symposiums to publish non-peer reviewed research and give your funded research “a deep fragrance of academia.”
Evidence shows that tobacco corporation-affiliated researchers routinely ignored data that didn’t support their position. Might not Atkins-affiliated researchers be guilt of the same?
Atkins Responds to Criticism “Who Pays the Piper Calls the Tune”
You deny that your funding influences the results. “Speaking of funding,” Dr. Atkins himself wrote, “the media jumped on the fact that the Dr. Robert C. Atkins Foundation funded the study, implying that the results were therefore suspect. Get real! Who do they think is funding the vast majority of funding for drug research? Pharmaceutical companies, of course. Does that mean that all research on prescription drugs is equally suspect?”
In choosing the pharmaceutical industry, Dr. Atkins seems to have picked the wrong business to exemplify lack of funding-source influence. According to World Health Organization director Jonathan Quick, “researchers who publish or communicate results unfriendly to the [drug company] sponsors have faced intimidation, attempts to discredit them professionally, and legal threats to recover ‘lost sales.’”
Citing drug manufacturers’ profit motive for skewing reports of their drugs’ safety, an editorial in The Lancet reads “Tobacco is not the only aspect of medicine open to twisted corporate communications strategies… All policymakers must be vigilant to the possibility of research data being manipulated by corporate bodies and of scientific colleagues being seduced by the material charms of industry. Trust is no defense against an aggressively deceptive corporate sector.”
Highly statistically significant associations have consistently been found between sources of funding and outcome of drug studies. One medical professor looked at over a hundred studies of new drugs sponsored by the manufacturer. Not a single one found their drug inferior to the competitor’s. Other studies have found the same uncanny “coincidence.”
Corporations manipulate funded research in a variety of ways. Companies may be selective in publishing results, for example, and “may delay or not publish unfavorable results at all.” If the researchers find negative results or adverse effects, the corporation can just suppress the truth. “Withholding the publication of unfavorable results,” according to the Journal of the American Medical Association, “is not uncommon.”
Reports one insider, “Companies can play hardball, and many investigators can’t play hardball back. You send the paper to the company for comments, and that’s the danger. Can you handle the changes the company wants? Will you give in a little, a little more, then capitulate? It’s tricky for those who need money for more studies.” Or sometimes, corporations have been known to just ghost-write the entire articles themselves.
The bottom-line is that whether you’re talking about pharmaceuticals or tobacco, recent studies “have all found that physicians with financial ties to manufacturers were significantly less likely to criticize the safety or efficacy of these agents.” The bottomline… is the bottomline.
“Researchers with financial relationships with for-profit funders of research not surprisingly tend to overstate the favorableness of their own products and tend to be less likely to report unfavorable outcomes, especially issues surrounding safety or efficacy.” Is there any reason to suspect that your corporation is any different?
Most of the Supporting Studies Were Inadequately Controlled
One of the favorite ways drug manufacturers design studies to skew results in favor of their product is to choose inappropriate controls. Just like a drug company might choose an inadequate dosing of the comparison drug to artificially inflate the results of their own product, many of the “supporting” studies of the Atkins Diet were compared to diets that were “low fat” in name only, and yet the Atkins Diet still failed to outperform them long-term.
In what was to become the single largest and longest controlled study of the Atkins Diet to date, researchers published “A Low-Carbohydrate as Compared to a Low Fat Diet in Severe Obesity” in the New England Journal of Medicine. Both the press and the Atkins Corporation heralded the findings as proof that low-carb diets were in some ways superior to low fat diets. The problem is that the control group was never actually on a low fat diet. They started out eating 33% calories from fat like the rest of America, and at the end of the 6 month study period in which they were supposedly switched to a “low-fat” diet, they were eating… 33% calories from fat.
At the end of a year, those that remained in the “low-fat” group were eating even more fat than average–34% calories from fat, yet you continued to erroneously describe this as a low fat diet. On your website “Atkins professionals” claim that the “low fat” control group “followed a calorie-restricted diet with less than 30% of calories from fat per day.” This is false. And even though the study was inadequately controlled, the low-carb diet, according to the researchers, “provided no weight loss advantage” compared to the group that hardly seemed to be on a diet at all.
This is the same story we see in every single year-long controlled trial of low-carb diets. Not a single one showed significantly more weight lost at the end of the year on the low-carb diets than on the control “low fat” diets. And in the two year-long studies that compared low-carb diets to actual low fat diets, the low fat diets seemed to win.[765,767]
In one year-long study, those who ate as much as they wanted of a low fat vegan diet lost an average of 52 pounds–60% more than those reportedly on Atkins lost.[765,766] And in the year-long comparison of the Atkins Diet to Ornish’s diet, Weight Watchers, and The Zone Diet, the Atkins Diet came in dead last in terms of weight lost at the end of the year. Ornish’s low fat vegetarian diet seemed to show the most weight loss. Yet you continue to list this study as “supporting Atkins.”
Most of the studies cited as “supporting Atkins” lasted at most a few months,[769-786] half had 15 or fewer people on the diet,[787-803] and a third lacked any control group at all.[804-814] And the majority of those that did randomize people into a control group compared low-carb diets to “low fat” diets containing between 29% and 35% calories from fat.[815-822] That is not a low fat diet.
Calorie-restricted, portion-controlled, moderate fat (30%) diets have a consistent history of failure to maintain long-term weight loss. What better diet to use as a comparison group, then, to increase the odds that low-carb diets will show more weight loss. And still, even using inadequate controls, low-carb diets failed to outperform control “low fat” diets in every single long-term study ever done.[826-829]
A truly low-fat diet (<25%) and exercise is the only approach found to provide long-term successful weight loss, based on a study of 5000 Americans who lost an average of more than 70 pounds each and kept it off for more than 6 years. According to one of the chief investigators of that study, “Almost nobody’s on a low-carbohydrate diet.”