Download a PDF version of this blog post
By Nina Teicholz
In March, at the last and final meeting of the Dietary Guidelines Advisory Committee (DGAC), members suggested that they’d like to lower the caps on saturated fats even further: from the current 10% of calories down to 7%—or even zero. These ideas are entirely contrary to a steadily growing consensus among top researchers over the past decade that the caps on saturated fats were never based on strong science and ought to be reconsidered.
Indeed, as a recent panel of senior scientists from the U.S., Canada, and Denmark stated, in a letter to the Secretaries of the U.S. Department of Agriculture and Health and Human Services (USDA-HHS), “we respectfully request….that [you] give serious and immediate consideration to lifting the limits placed on saturated fat intake for the upcoming 2020 Dietary Guidelines for Americans (DGA). This request is based on a review of the most rigorous scientific data available.”
“…'[M]aking recommendations based on anything but the most rigorous science available is likely to have unintended or even potentially harmful consequences to health,” they added.
This group of prominent scientists, which included three former members of the DGAC, including the 2005 committee chair, said that they are working towards publishing their findings in an academic journal.
Members of the 2020 Subcommittee on Dietary Fats and Seafood, responsible for the reviews on saturated fats, however, seemed unaware of the recent evolution in the science on saturated fats. The committee relied on DGA reviews of the science from 2010 and 2015 and appeared simply to bring forward these previous recommendations, which date back to the launch of the DGA in 1980. Indeed, members of the subcommittee sought to push the cap on saturated fats even lower.
Jamy Ard
“Why should it not be zero?” wondered DGAC member Jamy Ard, “since saturated fat is not an essential nutrient.” Dr. Ard, who is a medical director of Nestle, Inc., one of the world’s biggest junk-food manufacturers, was echoed by Linda Van Horn, a longtime saturated-fat foe who has argued against these fats for decades. Neither seemed to realize that saturated fats, in addition to contributing to metabolic health directly by raising the “good” cholesterol (HDL-C), are an integral part of many whole, natural foods, such as meat and dairy, which contain, in their most bio-available form, the essential nutrients that are vitally important for growth and disease prevention.
Linda van Horn
Van Horn, a member of the Subcommittee, said she preferred the 7% limit on saturated fats promoted by the American Heart Association (AHA), but this group addresses only those people at risk for heart disease, not the entire population. Moreover, the AHA has long been questioned for the 20%-plus funding it receives from food and drug companies, many with a direct financial interest in maintaining caps on saturated fats.
Big Pharma, for interest, is heavily invested in drugs that lower LDL-C. Thus, it’s extremely important to note that the cardiovascular benefits reported from cholesterol-lowering via drugs have never been replicated in trials that lower cholesterol via diet. In fact, in nearly a dozen large trials[1] on diet, researchers found that lowering one’s cholesterol by restricting saturated fats had a minimal effect on cardiovascular deaths--while at the same time reliably raising the risk of death from cancer.
Despite this, the 2020 Subcommittee stated that it found the link between saturated fat and heart disease to be “strong,” for both adults and children.
In addition to the issues noted above, we find these reviews by the Subcommittee to be flawed in numerous ways:
The analysis on adults used, as its starting point, the review on saturated fats from 2015, which has been criticized by the National Academies of Sciences, Engineering and Medicine, for being non-systematic. A separate peer-reviewed analysis I authored in The BMJ found that this review relied heavily on outside organizations that receive major contributions from food-and-drug companies and also that the conclusions from this review did not accurately reflect the evidence.
The review on children relied only on two trials, “DISC” and “STRIP,” both from the 1990s, which are deeply flawed as a basis for population-wide recommendations, because (1) DISC subjects had been selected for their unusually high cholesterol levels, suggesting that many of them had a genetic disorder that could not be extrapolated to a general population and (2) STRIP subjects were all Finnish, under the age of three. These unusual study groups cannot be used as a basis for population-wide recommendations to all American children.
Moreover, the DISC children consuming a low-fat diet, failed to meet basic nutritional standards—and this has also been found in other studies on children when fed diets low in fat.
Finally, there has never been evidence to show that cholesterol-lowering in children translates into better heart-disease outcomes later in life. Indeed, there is ample evidence to show that children need more fat and more naturally occurring proteins then do adults and that their growth and development is harmed when these foods are reduced during childhood[2].
(Note: it is problematic that DGAC member Van Horn, who was a lead author of the DISC study, did not recuse herself from discussions of the study’s merits in the committee proceedings. On the contrary, she was the most vocal proponent of this data and used it to push for further lowering of caps on saturated fats.)
Finally, the subcommittee noted “limitations” of the studies used to support its conclusions, namely that “[t]he majority of studies did not control for all key confounders, and many studies did not have CVD [cardiovascular disease]as a primary outcome.” These limitations are significant enough to raise serious questions about the legitimacy of the entire reviews.
In the discussion period that followed the Subcommittee’s presentation, DGAC Vice Chair, Ronald Kleinman, brought up the findings from the international expert panel, mentioned above, noting that it had reached the “opposite conclusion, that saturated fat in the diet doesn’t impact heart disease or stroke.” (starting at approx. 2:06)
Van Horn and others on the subcommittee rejected these findings, based mostly on vague terms.
Subcommittee member Joan Sabaté said that the clinical trials on saturated fats were complicated by refined carbohydrates, but all the major clinical trials on these fats kept carbohydrates constant, swapping out only saturated fats for polyunsaturated vegetable oils—implying that Sabaté’s comment was in error.
DGAC Committee chair Barbara Schneeman also emphasized that these trials lowered LDL-cholesterol (an “intermediary outcome”). However, the far more definitive “hard outcome” data—heart attacks, strokes and death—show that saturated-fat restriction had no positive benefit.
Focusing on intermediary outcomes rather than long-term “hard” outcomes is like judging the performance of a marathoner at the half-way mark rather than the finish line. As mentioned, any intermediary LDL-lowering effects do not matter when it comes to the more definitive outcomes of heart attacks and death. The latter data is more reliable—and ultimately what one wants to know. Does reducing saturated fats limit one’s chance of having a heart attack or dying from heart disease? The clinical trial data does not support this idea.
It is deeply concerning that not a single member of the Subcommittee on Dietary Fats raised any objection to its “strong” conclusions, despite the obvious, widespread scientific debate in the field. This suggests that the committee is unbalanced and one-sided. The National Academies of Sciences, Engineering, and Medicine, in its critique of the DGA process, suggested that biases be managed within the committee, but on the subject of saturated fats, this kind of management has clearly not taken place.
Transcripts and videos of this meeting are available here. (See Meeting Five, Friday morning)
[1] Feinleib, M., “On a Possible Inverse Relationship Between Serum Cholesterol and Cancer Mortality,” American Journal of Epidemiology 114, no. 1 (1981): Feinleib, M., “Summary of a Workshop on Cholesterol and Non- cardiovascular Disease Mortality,” Preventive Medicine 11, no. 3 (1982).
[2] An overview of these studies can be found in Teicholz, N., The Big Fat Surprise, pp 152-155, with all studies to support the above statements cited in the footnotes and bibliography.