USDA Ignoring the Science on Low-Carb Diets

Since 2014, the US Dept. of Agriculture has a record of "burying" the science

“I was wondering if we should have a separate section on low-carb diets rather than burying it,” wrote Harvard nutrition professor Frank Hu about the official expert report for the 2015 U.S. Dietary Guidelines for Americans. Over the past decade, thousands of scientists, doctors and others have urged officials overseeing our nation’s nutrition policy to pay attention to studies on low-carbohydrate diets, yet even as the science has advanced rapidly, the federal agencies in charge of the guidelines seem ever-more intent on ignoring and yes, burying the science.

The literature on low-carb is now vast. A quick search for “low-carbohydrate” on pubmed.gov, a government database of scientific studies, today turns up 7,821 publications, including 858 clinical trials (the most rigorous kind of evidence).¹² Compare this to the 401 trials on the “vegetarian diet,” which has been formally recommended by the U.S. guidelines since 2015.

Of course quantity is not quality, but the American Diabetes Association judged low-carb to have “most evidence for improving glycemia” [blood sugar control], the key issue for people with with diabetes. And the diet produces “more weight loss” than a higher-carb option, according to the American Heart Association.

Yet here is the recent history on how the U.S. Departments of Agriculture and Health and Human Services (USDA-HHS), the federal agencies that co-issue the guidelines every five years, have treated this large body of scientific literature:

• For the 2015 guidelines, the USDA³ conducted a formal review of this literature and found 40 studies but unlike its other reviews, kept this one hidden and did not even reveal the fact that it had been conducted. The review’s existence was only discovered in emails obtained via the Freedom of Information Act (FOIA);

• Five years later, for the 2020 guidelines, the USDA publicly reviewed the low-carbohydrate diet but could find no studies meeting its inclusion criteria on the topic;

• For the upcoming 2025 guidelines, USDA has proposed ignoring low-carbohydrate diets altogether.

How could this happen?

A New Cycle Begins; Still No Recognition of Low-carb

In March, the USDA launched its process for the 2025 guidelines by publishing a draft list of scientific questions (no longer at its original URL), which did not include a single query on low-carb diets. If no question is asked, no review gets done, and this would mean that the guidelines would have no chance of including a low-carbohydrate option until the 2030 guidelines or later.

The public, given a month to comment, was clearly upset. Nearly 80% of the total comments, or some 1,200 individuals, urged the USDA-HHS to consider the extensive scientific literature on low-carb. The public comment period closed in mid-May, and the agencies have not yet responded.

Many commenters expressed frustration that they had suffered from obesity and other diet-related conditions for years without being told about the availability of a low-carbohydrate approach—which subsequently resolved their health problems. As one commenter wrote:

“I am a retired physician who personally struggled with overweight or obesity since childhood. I counseled patients on the same issues, using nutritional & exercise guidelines from government and medical experts. Those guidelines did not work for my patients or me. I would lose weight on calorie counting & exercise but could not sustain this when eating carbohydrates, including so-called healthy whole grains. Only when I discovered information about low carb/ketogenic way of eating at age 69 did my struggles end. I have been following low carb eating since then, feel wonderful, have excellent exercise endurance, saw my HDL increase to 104, and suffer no more joint pains.”

It All Goes Back to the Guidelines…

The inability to get low-carb information from most doctors, dietitians or other clinicians can be traced directly to the guidelines, which are considered the ‘gold standard’ by healthcare practitioners and are taught at every level of schooling. By law, the guidelines also determine the food served in all federal nutrition programs, such as school lunches, feeding programs for the elderly, and more.

Grains, bread, and other starches have been the foundation of the U.S. ‘food pyramid’ since the guidelines began in 1980. Ten servings of bread per day was the original recommendation. Today, the guidelines recommend six servings of grains daily (including 3 refined), plus up to 10% of calories as sugar. Altogether, the 2020 guidelines recommend eating 52%-56% of calories as carbohydrates, up a couple of percentage points from the last guidelines, in 2015.

Contrast that with the <25-30% of calories as carbs on a low-carb diet—clearly a very different idea. This amount of carbohydrates may seem quite low to the average person, but remember that for the most part, people turn to this way of eating because they’ve been diagnosed with a serious health problem, such as diabetes, which usually means that they can no longer tolerate high levels of sugars or starches.

A low-carbohydrate approach clearly differs sharply from the government’s guidelines—and is likely one reason why the UDSA-HHS have been unresponsive to the relevant science.

Congress is Concerned

Congress first aired its awareness of lapses by the USDA-HHS regarding the science on low-carbohydrate diets at a rare hearing on the guidelines, in October, 2015. Quite a few members of Congress expressed concern that the guidelines were too high in carbohydrates for many people.

Former Congressman Dan Benishek, a physician from Michigan, commented,

“Fifty-two percent of Americans are pre-diabetic or diabetic, and yet, the dietary recommendations that, as I understand it, are not really appropriate for that….These people have a carbohydrate intolerance and — and there [are] more carbohydrates in the diet that you’re recommending than is really appropriate for that.”

Vicki Hartzler from Missouri said,

“I have heard concerns that the [USDA] has ignored a large, credible and growing body of peer-reviewed science on low-carb diets, as it contradicts the evidence from previous guidelines.”

The National Academies’ Report on the Dietary Guidelines

Out of this awareness came a Congressional mandate—and a $1 million appropriation—for the first-ever peer-review of the guidelines, by the National Academies of Sciences, Engineering, and Medicine (NASEM).⁴  The resulting report, in 2017, was published in two parts, with a number of recommendations for the USDA to adopt. Just recently, the NASEM issued another report, to evaluate how the USDA had done in complying with its 2017 recommendations thus far.

The 2022 evaluation is not rosy. The guidelines process still has a ways to go to “be trusted by the scientific community, health professionals, and the public,” it says.

Not one of the seven recommendations made by the NASEM in 2017 has been fully adopted (or, in one case, evidence is lacking to make an assessment), says the report.

Worse, the guidelines process continues to raise “serious concerns about transparency and conflict of interest.” Among other things, it does not undergo outside peer-review (as recommended by the 2017 NASEM report) and its study-selection process is lacking to the point that it “compromise[s] the integrity, transparency, deliberative process, and leave[s] open the possibility for the introduction of bias.”

These are pretty fundamental methodological problems and probably help explain how the USDA has somehow managed to avoid the entire scientific literature on low-carbohydrate diets since at least 2014.

The 2015 guidelines: Formal Review of Low-carb Is Hidden

The day before Thanksgiving, 2014, USDA nutritionist Eve Essery brought up the subject⁵ of low-carb diets with members of the outside expert group, called the Dietary Guidelines Advisory Committee (DGAC), which by law is intended to direct the scientific reviews. Essery writes:

As you all know, there has been a continued, and perhaps growing, interest in low-carbohydrate diets, and we have been asked on numerous occasions why the Dietary Guidelines do not recommend low-carbohydrate diets as the dietary pattern of choice to the general public.⁶

Essery mentions in passing that her USDA colleague, Julie Obbagy, had conducted a review of this science,⁷ the results of which appear to have been shared with a single member of the DGAC, Anna Maria Siega-Riz, a professor of nutrition at the University of Massachusetts Amherst. Essery says Siega-Riz had written up a statement on the topic, to be included in the methodology section of the scientific report.

DGAC member and Harvard professor Frank Hu pipes up to object:

I wish that the committee has more time to summarize and update the… evidence on low-carb DIETS [sic]… In any event, we should not simply dismiss the role of low-carb diets in weight control and improving overall health outcomes. [emphases added here and below]⁸

Relegating the low-carb discussion to the methodology section would not allow it to be considered alongside other diets, says Hu:

“Given the popularity of [a low-carb] pattern and enormous amount of research that has been generated in the past several years, I was wondering if we should have a separate section on low-carb diets rather than burying it in the Methodology section…”

He added:

“People who are familiar with the field may complain that we gloss over recent evidence and don’t give low-carb diets…sufficient attention that they deserve.”

For several days, Hu and other DGAC members continue to misunderstand that a USDA formal review on low-carb had actually been done—and that they’d just been left out of the loop.

Siega-Riz eventually drives home the point, saying plainly that the USDA “did conduct a review for us….” End of conversation—and the review was stuffed.

The 2020 Guidelines: 40 Studies Forgotten

The 40 studies in this review undoubtedly undercounted the number available at the time. Still, 40 is worth contrasting to the number of low-carb studies found worthy of review by the USDA five years later, which was zero.⁹

Again, one has to ask, how was this was possible? Eve Essery, now Eve Essery Stoody, was still involved. In fact, she had risen to be the “designated federal officer” for the entire USDA on the guidelines (Stoody declined requests to comment for this article).

Even if she had suffered a memory lapse, it’s hard to understand how every one of the 18 members of the USDA systematic review team, plus all 20 members of the 2020 DGAC could not have found a single acceptable study on low-carbohydrate diets.

By then, at least 100 systematic reviews on low-carb could be found on pubmed.gov, according to an advocacy group, which also formally submitted a list of 52 relevant clinical trials to the DGAC for consideration. The American Diabetes Association in 2018 had adopted a “very low-carbohydrate diet” as a “standard of care” for managing type 2 diabetes.

To be unaware of any of these publications implied that all the experts involved in the 2020 guidelines process were either afraid to speak up or completely out of touch with the peer-reviewed literature in the field of their apparent expertise.

Even if this could be believed, it’s an unavoidable fact that every member of the DGAC and all senior members of the USDA’s guidelines staff witnessed at least 15 doctors and other practitioners present oral testimony—three minutes each, standing at a microphone—at two public meetings over the course of the 2020 process. These clinicians spoke about the abundance of science on low-carbohydrate diets and recounted their own personal breakthroughs in recommending this diet for their patients.

Among those testifying was the late Dr. Sarah Hallberg, the principal investigator of the largest and longest clinical trial using a low-carbohydrate approach to successfully reverse a diagnosis of type two diabetes. Given that diabetes costs $350 billion per year to treat, the possibility of rolling back this epidemic should have been tantalizing to experts entrusted with the nation’s health, yet no one from USDA-HHS followed up with Hallberg to learn more.

One DGAC member listening from the stage was Lydia Bazzano, a Tulane University professor who had been the lead investigator on a year-long clinical trial on diet, funded by the National Institutes of Health (NIH), with three publications, that found a low-carb diet to be superior to one low in fat for weight loss, appetite control, and cardiovascular risk. However, USDA excluded even Bazzano’s publications from its 2020 reviews on obesity and cardiovascular disease. Bazzano said nothing about this publicly and coincidentally, received $7,102,897 in grants from the NIH (part of HHS, co-issuer of the guidelines), in her two years while serving on the DGAC, including two supplemental funding grants totaling $679,525 in 2020.

Time to Review the Low-carb Science

The public is currently waiting to hear if the USDA-HHS will reverse course and decide to address the substantial body of science on low-carb for the 2025 guidelines.

Meanwhile, the 5-year clinical-trial results from the trial overseen by Hallberg were just published. These data show that a low-carb diet, plus support via a mobile app, can put a diagnosis of type 2 diabetes into remission—safely and sustainably—while eliminating the need for most medications. Only one other nutritional approach, the Mediterranean diet, is supported by 5-year clinical-trial data, and that diet, while moderately reducing cardiovascular risk, has never been shown to put type 2 diabetes into remission or cause significant weight loss. (Also, the diet in this 5-year trial, conducted in Spain, had 40% of calories as fat, not the 32% in the USDA’s “Mediterranean-style” version.)

Carbohydrate restriction is currently the only whole-foods approach that can reverse a diagnosis of type 2 diabetes. It also improves the vast majority of cardiovascular risk factors. In a nation where a least 88% of the public has the features of metabolic disease, it seems self-evident that the scientific literature on this nutritional option should at least be considered. Disadvantaged populations are especially in need of more choices, as they suffer disproportionately from diet-related diseases and also rely more heavily on food from federally funded programs. But the only way to create change is if the Dietary Guidelines considers low-carb as an option. Not a prescription, just an option.

The USDA is at a crossroads. No doubt many of us would like to have greater faith in federal agencies that provide food, healthcare, and dietary guidance to so many people in need. To build confidence, the public needs to see that policy makers are responsive to the people—and the science.

1. Search terms used were “low-carbohydrate” or “ketogenic.” Note that the results likely include double-counting of trials, since many result in multiple publications. Some results may also be trial protocols rather than completed trials. Searches conducted on 6/24/22.

2. For the purposes of this article, the term “low-carbohydrate diet(s)” includes the ketogenic diet. Low-carb is generally defined by researchers in the field as allowing a maximum of 25-30% of calories as carbohydrates, although the limit is often instead defined as grams (<130 grams) of carbohydrates per day, while a ketogenic diet is lower, around <50 grams. See: https://www.mdpi.com/2072-6643/13/10/3299/htm

3. The USDA administers the guidelines. Buried inside the USDA is an office, called the Nutrition Evidence Systematic Review, or NESR (formerly the Nutrition Evidence Library, or NEL), which conducts all the scientific reviews

4. The Nutrition Coalition, a non-profit group I founded in 2015, was instrumental in generating Congressional interest in the NASEM report. The aim was not to promote any particular diet but simply to have an outside peer-review of the guidelines’ process.

5. The discussion can be found on pp. 206-216.

6. Essery is writing not to the full DGAC committee but to one of its subcommittees, “SC2,” which was responsible for looking at the science on recommended diets, or “Dietary Patterns,” as they came to be called.

7. For this review, low-carbohydrate was defined as <45% of calories as carbs; people diagnosed with diseases were excluded; the term “ketogenic” was not included in the search string. The review lists 43 articles, but one article is listed 4 times, yielding 40 distinct articles.

8. This information was first published in an op-ed I wrote for the Wall Street Journal, May 30, 2020. Available at: https://www.nutritioncoalition.us/news/low-carb-strategy-for-fighting-the-pandemics-toll

9. At a public meeting on March 12, 2020, DGAC member Carol Boushey reports that no studies on diets with carbohydrates below 25.3% could be found on cardiovascular disease (p. 159) obesity (p.165), or diabetes, (p.169). Available at: https://www.dietaryguidelines.gov/sites/default/files/2020-06/2020DGACMtg5TranscriptDay1FINAL.pdf