Scientists who called for the retraction of an article by author of The Big Fat Surprise Nina Teicholz take credit for end to low-fat diet
A scientific study examining the relationship between dietary fats and coronary heart disease (CHD) that featured in the Journal of the American College of Cardiology (JACC) in October has been described as a “game changer” by a host of scientists and nutritionists, according to an article published Monday by MedPage Today.
The “game changing” JACC study built on the recommendations of a scientific report by the 2015 Dietary Guidelines Advisory Committee (DGAC), about which Harvard Professor Meir Stampfer, MD, said “for the first time, and based on strong scientific evidence … abandoned any explicit ceiling on total fat intake, focusing instead on the types of fat.”
Other quotable notables from the MedPage Today article include Stampfer’s Harvard colleague Frank Hu, MD, co-author of the JACC study and member of the DGAC, who said “If saturated fat is replaced by refined carbs, there is no reduction in CHD risk.”
If this sounds familiar, it should. Journalist/author Nina Teicholz heralded an end to the low-fat diet with the release of her book on the history of nutrition science, The Big Fat Surprise, in May 2014.
In November, Teicholz was the subject of a letter that was sent to the BMJ (formerly the British Medical Journal) requesting the retraction of her September 23 article, The scientific report guiding the US Dietary Guidelines: is it scientific?, which criticised the methodology and findings of the 2015 dietary guidelines report.
The letter was organised by Bonnie Liebman MS at the Center for Science in the Public Interest (CSPI), a Washington, DC-based advocacy nonprofit, and was signed by over 180 credentialed professionals,¹ including several of those mentioned in the MedPage Today article.
After receiving the November 5 retraction request, the BMJ published this November 19 post by Executive Editor Theodora Bloom asking the authors of the letter to submit it as a “rapid response” on thebmj.com, requiring all signatories to declare any competing interests.
As reported in an item I co-authored with US (Atlanta, GA) investigative reporter Peter M. Heimlich, on December 17 the BMJ posted an updated version of the CSPI letter, absent the names of 18 scientists and graduate students.
¹Including these three co-authors of the JACC study: Frank Hu, MD; Eric Rimm, ScD; and Walter Willet, MD.
This item has been updated to include the names of three signatories to the CSPI letter.
Shooting the Messenger 
In the MedPage today press release you linked, the Harvard coterie cite their own paper “Saturated Fats Compared With Unsaturated Fats and Sources of Carbohydrates in Relation to Risk of Coronary Heart Disease: A Prospective Cohort Study” (Yanping et al 2015) as their sole reference.
Yanping et al 2015 is a bizarre thing – a meta-analysis (so to speak) of TWO other studies (NHS and HPFS). Its conclusions regarding carbohydrates are contradicted by at least one proper meta-analysis, “Intake of carbohydrates compared with intake of saturated fatty acids and risk of myocardial infarction: importance of the glycemic index.” Jakobsen et al 2010, which found no benefit from replacing saturated fats with low-GI carbohydrates, but harm from replacing SFA with low-GI carbs.
Furthermore, the 2 studies in Yanping et al 2015 used the least reliable FFQ method of data collection and SFA intakes changed significantly over the study period. The criteria for plausibility of dietary intake reporting seem to have been over-permissive, with 600 kcal/day being the lower cut-off for plausibility.
MUFA is given a positive value which is not supported by the isolated odds ratios for MUFA, which are identical to that from SFA. Even a stage magician shows that he has “”nothing up his sleeve” before performing a trick, but in regard to MUFA in Yanping et al 2015 this disclosure is lacking.
The conclusions of Yanping et al support replacing 5% energy from SFA or refined carbohydrates with 5% energy from polyunsaturated fats. The authors nowhere mention a benefit or lack of harm from increasing fat beyond this. Yet in the NHS highest intake of total fat was associated with a significant reduction in heart attacks – Yanping et al 2015 does not discuss this correlation.
In “Dietary fat intake and early mortality patterns–data from The Malmö Diet and Cancer Study.” (Leosdottir 2005) we have a large population study (n=28,098) in which dietary intake is determined by FFQ, interview, and 7-day food diary – (the most reliable method short of biomarker studies) and in which subjects who changed their diets could be excluded.
“In men, a significant trend towards lower cardiovascular mortality in upper quartiles of total fat
intake was observed (P ¼ 0.028) with the RR for men in the fourth quartile being 0.65 (CI 0.45–
0.94, P ¼ 0.023).” There was no association between SFA or PUFA and CVD mortality.
(In another paper from this population, a combination of high SFA intake and high fibre intake, mostly from fruit and vege, was associated with the lowest risk of CVD events)
Women had increased cancer risk with highest fat intake, with this being linked to MUFA (other cancer-specific studies in this population link this risk to low-quality liquid margarine used in cooking and to vegetable oils in highly processed foods).
In other words, Yanping et al 2015 ignored wider and better quality evidence when concocting the results that they used to stake their claim to being high-fat visionaries.
These proposed changes in fat limits have come about because of the investigative journalism of Gary Taubes, Nina Teicholz, and the activism of countless others, which has put this question and the data concerning it in the public eye, so that the DGAC could no longer ignore it. Otherwise, it is very unlikely that anything would have changed during what would have still been a largely secret process.
After reading Nina’s comment, I better understand yours. Undoubtedly you’ve made any casual readers (that would include me) smarter having read it, so thank you.
Thanks to Dean for this post! I think what is particularly crazy about crediting the Harvard study (which the analysis above shows to be confused and contradictory) is that it is an epidemiological study–a weak form of evidence that has a poor track record of confirmation when clinical trials are conducted (clinical trials, or RCTs, are the only kind of evidence that can demonstrate actual cause and effect, whereas epidemiological studies can only show correlation but not causation). There are several large, NIH-funded, multi-year RCTs on the low-fat diet, including the largest-ever RCT in the history of nutrition science (the Women’s Health Initiative), and taken together, these have found that the low-fat diet is ineffective for combatting obesity, diabetes (as measured via blood glucose control), cardiovascular mortality (as well as certain CVD risk factors) or any type of cancer. These trials, not conducted by Harvard, are the definitive evidence that condemned the low-fat diet. Harvard had nothing to do with it.
You’re welcome, Nina.
“epidemiological studies can only show correlation but not causation”
Ah, I see. So it’s unlikely that the Harvard study would have yielded reliable results regarding the dietary fats-CHD link, yet its findings were talked-up as if the evidence was conclusive. After the hysterical response to your BMJ article being flawed,* kind of puts an ironic twist on things, doesn’t it? Bandwagon science.
*Their opinion, not mine.