BMJ won’t retract controversial dietary guidelines article; issues lengthy correction

bmjThe BMJ has released a detailed correction to a much-debated article critiquing the expert report underlying the U.S. dietary guidelines.

After the article was published in 2015, the Center for Science in the Public Interest (CSPI) organized a letter signed by more than 100 researchers, urging the publication to retract the article. Today, the journal said it found “no grounds” to do so.

However, in a press release accompanying the announcement of the correction, the BMJ notes that some aspects of the CSPI’s criticisms were merited.

Editor in chief Fiona Godlee said in a statement:

We stand by [author Nina Teicholz’s] article with its important critique of the advisory committee’s processes for reviewing the evidence, and we echo her conclusion: ‘Given the ever increasing toll of obesity, diabetes, and heart disease, and the failure of existing strategies to make inroads in fighting these diseases, there is an urgent need to provide nutritional advice based on sound science.’

Teicholz — who already announced the article wouldn’t be retracted — said in a statement that she appreciates the journal’s support:

I am very grateful to The BMJ editors for their profound commitment to verifying the facts of my article and for their professionalism and integrity throughout this process. I am also grateful that they are providing a space for rigorous scientific debate, especially on a subject so important to public health. I hope the original intention of that article can now be fulfilled—to help improve nutritional advice, so that it is based on rigorous science. This will help us to better combat nutrition-related diseases that have caused so much human suffering around the world.

Bonnie Liebman, CSPI’s Director of Nutrition, told us she was not pleased with the decision:

The BMJ has stained its reputation by circling the wagons around Nina Teicholz’s discredited and opinionated attack on the science underpinning the Dietary Guidelines for Americans. The BMJ corrected or “clarified” 7 of the 11 errors cited by the letter from more than 180 scientists requesting a retraction, and failed to respond to the remaining four. (The clarifications are thinly veiled corrections.) It’s startling that despite this long list of corrections and clarifications—including several that undergirded the article’s attack on the Dietary Guidelines Advisory Committee report–the journal nevertheless stands by the article’s conclusions.

Liebman added:

As one reviewer noted, the article never should have been published as an “investigation by the BMJ” when it is “ better described as an opinion piece, editorial, or even an example of lobbying literature than an independent investigation.”

The scientific report guiding the US dietary guidelines: is it scientific?” criticized several aspects of the expert report, noting it used “weak scientific methods,” according to today’s statement; it has already received a “clarification” (issued soon after it was published), and a previous correction, which addressed the research behind the analysis about saturated fats. The 2015 article has been cited six times, according to Clarivate Analytics’ Web of Science, formerly part of Thomson Reuters.

After receiving heavy criticism, the journal asked two independent experts — Lisa Bero at the University of Sydney, and Mark Helfand at the Oregon Health & Science University — to review the article. As the BMJ statement notes:

Both reviewers found that the authors of the CSPI letter were correct in their contention that the [Dietary Guidelines Advisory Committee] report described methods for identifying, selecting, and evaluating evidence for its report.

But they also noted problems with the committee’s methods and rejected the letter’s contention that the article should be retracted.

Professor Bero concluded: “Teicholz’s criticisms of the methods used by DGAC are within the realm of scientific debate” and Professor Helfand said that “it is clear that further investigation of the composition of the committee, as well as its conflict of interest policies and work group structure, are warranted.”

The review culminated in a lengthy correction notice, published today:

This article (BMJ 2015;351:h4962, doi:10.1136/bmj.h4962) stated that the 2015 US Dietary Guidelines Advisory Committee (DGAC) “conducted an [ad] hoc examination of the scientific literature without well defined systematic criteria for how studies or outside review papers were identified, selected, or evaluated.” The article also stated that the committee “conducted ad hoc reviews of the literature, without defining criteria for identifying or evaluating studies.” These statements are incorrect. The DGAC defined the methods it used for identifying, selecting, and evaluating such evidence in its advisory report1 in Part C: Methodology and Appendix E2.2 However, it should be noted that the DGAC also used non-systematic approaches in its evidence selection.

The article also stated that the DGAC’s recommendations on saturated fats were based, in part, on “the 2015 committee’s ad hoc selection of seven review papers,” and that “papers on saturated fats published since 2010 were covered by the committee’s ad hoc review, which did not use a systematic method to select or evaluate studies.” These statements are incorrect. Appendix E2.43 of the DGAC report defines the search strategy and inclusion criteria for the committee’s selection of papers for this topic and gives the DGAC’s quality rating (using AMSTAR, the methods of which were also described in Part C: Methodology of the report).

The article’s data supplement Table D includes the phrases “DGAC ad hoc selection” and “DGAC ad hoc review” of the literature, under the sections “dietary patterns and heart disease” and “dietary patterns and obesity,” respectively, and notes that “no systematic methodology is given for the selection of these studies.” This is incorrect. The prespecified search strategy and inclusion criteria are described in Appendices E2.26 and E2.27, respectively.

The article also stated, in reference to evidence regarding DGAC’s recommended diets and heart disease: “The committee reviewed other, more recent studies but not using any systematic or predefined methods.” This is incorrect. The DGAC defined its methods in Appendix E2.26, including search strategy, inclusion criteria, and quality rating (using AMSTAR, the methods of which were also described in Part C: Methodology of the report).

Included with the correction is a three-paragraph-long “Clarification,” which states:

The article stated that the committee did not systematically review certain studies on saturated fats from the 1960s and 1970s. This statement was insufficiently clear. It should have stated that the committee did not “directly” review these studies. The committee did consider two systematic reviews45 that themselves included five6 7 8 9 10 of the six trials mentioned in the article.6 7 8 9 10 11

The article also stated, “There have been at a minimum, three National Institutes of Health funded trials on some 50 000 people showing that a diet low in fat and saturated fat is ineffective for fighting heart disease, obesity, diabetes, or cancer. Two of these trials are omitted from the NEL review.” This statement was insufficiently clear. The two trials referred to12 13 were indeed omitted from the NEL’s 2010 review, as stated in the article, but it should be noted that they were evaluated in a Cochrane review,4 and that this review was considered by the committee in its 2015 advisory report.1 In addition, the two trials evaluated the effects of diet modification on serum lipids as a proxy for heart disease.12 13

The article also stated, regarding the Nutrition Evidence Library (NEL), that “a systematic review on health and red meat has not been done. Although several analyses look at ‘animal protein products,’ these reviews include eggs, fish, and dairy and therefore do not isolate the health effects of red meat, or meat of any kind.” This statement requires clarification. Several of the NEL reviews used “meat” as a search term and presented and discussed results of individual studies of red meat and total meat.14 15 1617 18 19

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

9 thoughts on “BMJ won’t retract controversial dietary guidelines article; issues lengthy correction”

  1. Do you see an evolution to a publishing environment where the active electronic links to citations, revisions, supplemental information, corrections, retractions, and ongoing peer critique should be more important than the tick mark in the self-feeding algorithm used to fund/drive research?

  2. Alas, Retraction Watch, I will forever read your coverage with enormous skepticism. The way you have treated this issue has been consistently on the side of CSPI. And it’s no different for this piece, which is patterned almost exactly on the CSPI press release: https://cspinet.org/news/bmj-stains-its-reputation-not-retracting-teicholz-article-20161202
    I had wondered how you could spin this incredibly supportive press release and comment by The BMJ to look negative, but I knew you would find a way.

  3. Via “The sugar conspiracy” by Ian Leslie, The Guardian, April 7, 2016: http://tinyurl.com/gplxq74

    In September last year (Teicholz) wrote an article for the BMJ (formerly the British Medical Journal), which makes the case for the inadequacy of the scientific advice that underpins the Dietary Guidelines. The response of the nutrition establishment was ferocious: 173 scientists – some of whom were on the advisory panel, and many of whose work had been critiqued in Teicholz’s book – signed a letter to the BMJ, demanding it retract the piece.

    Publishing a rejoinder to an article is one thing; requesting its erasure is another, conventionally reserved for cases involving fraudulent data. As a consultant oncologist for the NHS, Santhanam Sundar, pointed out in a response to the letter on the BMJ website: “Scientific discussion helps to advance science. Calls for retraction, particularly from those in eminent positions, are unscientific and frankly disturbing.”

    The letter lists “11 errors”, which on close reading turn out to range from the trivial to the entirely specious. I spoke to several of the scientists who signed the letter. They were happy to condemn the article in general terms, but when I asked them to name just one of the supposed errors in it, not one of them was able to. One admitted he had not read it.

  4. Via “NFL, take note: Calls for retraction are usually a bad idea” by Ivan Oransky & Adam Marcus, Stat, March 29, 2016: http://tinyurl.com/h7lo8ff

    (When) it comes to calls for retraction, we think advocates are better off exercising restraint.

    That may sound odd coming from the cofounders of Retraction Watch. But we rarely see anything good — including the sought-after retraction — coming from these petitions and public demands.

    …The sense one gets is that the people calling for these retractions — often followed by the word “immediately” — are less interested in correcting the scientific record than they are in punishing those whose views they don’t share.

  5. I am a nutrition scientist, but havent signed the rettraction letter. However, I do wonder what Teicholz’ sentence in the 1st paragraph of her 2015 means ‘ an investigation by the BMJ has found’.
    Does she report on this investigation, if so, who carried this out? Or she did it herself? And why would the BMJ investigate the science behind DGAC anyway? And note that if the purpose was to reveil the lack of science behind the DGAC the errors or omissions which are now revised are not trivial at all. Strange.

  6. I am disappointed to read this entry. Retraction Watch has let itself be used as a blunt instrument in the crude attempt by advocacy groups to silence debate at a time where debate is more important than ever. I’ve followed Retraction Watch for some time and feel like this behavior is uncharacteristically partisan against someone trying to shed light on one of the most critical issues facing our health and in favor of an organization that has used bullying and intimidation as a tool to silence debate. It is not the BMJ’s reputation that was stained by these doings, but Retraction Watch’s.

    1. Why? No, seriously, why does this ‘entry’, in your view, show any partisanship by Retraction Watch? Does it say the BMJ’s reputation was stained? Does it say anything that is not true? Does it only show one side of the story?

      It is almost if some people _wanted_ this to be a partisan story, and therefore are disappointed it isn’t!

  7. Via “My press release on BMJ decision not to retract,” Nina Teicholz, December 3, 2016: http://tinyurl.com/zupq9z9

    “(Teicholz’s) paper raises serious questions about how the dietary guidelines conducts its reviews, especially its systematic reviews, and whether they are based on a comprehensive review of the science,” said Cheryl Achterberg, Dean of the College of Education and Human Ecology at The Ohio State University and a former DGAC (Dietary Guidelines Advisory Committee) member. “These criticisms have now been confirmed by outside reviewers. For future guidelines, as well as for the health of the nation, we need to consider these points and move towards reforming the dietary guidelines process.”

    “I congratulate The BMJ on withstanding pressure to retract this article,” added Richard Smith, the former editor-in-chief of The BMJ. “There is an ugly tendency these days for powerful groups to call for retraction of any article they don’t like. The BMJ has stood up for science on a crucial subject that affects everybody.”

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.