Currently browsing posts about: Research

Julia Belluz and Kevin Hall.  Food Intelligence: The Science of how Food Both Nourishes and Harms Us.  Avery, 2025. ~340 pages

This is the long-awaited manifesto from the journalist Julia Belluz and scientist Kevin Hall.

As the press release puts it,  the book

digs deep into the fundamental, often overlooked, and always enthralling science of nutrition (the chemicals and energy we get from food) and metabolism (how our bodies use food)—covering what we know and the history of how we came to know it, up to the frontiers of research into the invisible forces that really shape our eating habits. The result is a sprawling tour of centuries of science into the wonders of food and the marvelous ways our bodies use it, for better and worse health.

As you can see, I wrote a blurb for the book (they edited it slightly).

If you want to understand how nutrition became so contentious and why we are still arguing about whether it’s better to eat more or less fat, carbohydrate, protein, or vitamins, you must read Food Intelligence. Well written, historically accurate, and scientifically rigorous, this book brings you up to the moment on contemporary dietary issues. 

Here are two excerpts, the first from a discussion of one of Kevin Hall’s studies comparing high fat to high carbohydrate diets:

[Kevin] predicted that the body would select fuels for metabolism in a way that caused body fat loss to vary only a little, regardless of the proportion of carbs or fat a person was eating.  Cutting carbs from a balanced diet caused the body to shift toward burning fewer carbs and more fat after several days.  But surprisingly, reducing dietary fat by the same number of calories didn’t seem to change the mixture of carbohydrate and fat the body burned.  The net result was that both diets led to similar body fat losses, but with a slight difference that contradicted the popular claims of low-carb acolytes like Atkins.  The reduced-fat diet, Kevin’s model predicted, led to a little more body fat loss compared to the reduced-carb diet.

Maybe a calorie wasn’t exactly a calorie, Kevin told his audience.  But the difference was in the opposite direction from the one claimed by the low-carb diet camp. P. 70

And here is one about why it’s useful to eat a variety of foods:

Food combinations matter—a complexity we’re only beginning to unravel.  Pairing foods rich in plant-based iron with foods rich in vitamin C increases the body’s ability to absorb the iron, while drinking alcohol with a meal hampers nutrient absorption.  Too many glasses of wine, and the ability to absorb vitamins and minerals, such as thiamine, vitamin B12, folate, and zinc, drops off.  pp 221-222

NIH, and agency of the Department of Health and Human Services under Robert F. Kennedy Jr, has issued its strategic plan for nutrition research for the next five years.

The Table of Contents states the goals and research objectives.

The goals:

1. Advance Science
2. Support the generation of evidence to address priority diet, nutrition, and health outcomes
3. Build Capacity and Strengthen the Field of Nutrition Science
4. Foster Stewardship, Collaboration, Transparency, and Accountability in Nutrition Science Research

I went right to #2.  Its impact objectives:

1. Improve the Approaches and the Precision of Methods to Assess the Determinants of Malnutrition
2. Support the Generation of Evidence to Enhance Nutrition Regulatory Science

Oops.  Nothing about chronic disease?  Where is the MAHA agenda in this?

On closer look, the report mentions chronic disease 8 times.  It recognizes the problem, stating that

food systems and the food environment…are critical factors affecting consumer choices; dietary patterns; and, ultimately, health. Using this framework, ONR [Office of Nutrition Research]…will address critical components of the nutritional ecology—such as the shaping and impact of consumer attitudes, beliefs, and behaviors regarding food systems—and consider key questions to identify knowledge gaps in nutrition science that have direct bearing on diet-related chronic diseases….topical areas may include:

• Food production
• Food distribution and marketing
• Food delivery
• Food Is Medicine interventions
• Brain–body interactions
• Cooking and nutrition education
• Personalized and precision nutrition interventions

Ah yes, precision nutrition (targeting diets to specific individual genetic factors).

Identifying factors that predict inter- and intra-individual variability will likely decrease the burden of diet-related
chronic diseases and conditions and will also offer ways to tailor interventions for individuals and populations. [Goal 1, research objective 2]

The plan is organized around a unifying vision of precision nutrition research and includes four strategic goals and five crosscutting research areas. These opportunities complement and enhance ongoing research efforts across NIH to improve health and to prevent or treat diseases and conditions affected by nutrition. [Box 1]

The strategic goals are organized around four questions:

• What do we eat, and how does it affect us?
• What and when should we eat?
• How does what we eat promote health across our lifespan?
• How can we improve the use of food as medicine?

These are good questions, but to me they seem like public health questions.  It’s hard for me to imagine how they could be answered through precision nutrition.

I look forward to finding out how NIH plans to do this.

 

[Sorry for my error: This post did not get sent out yesterday to subscribers so I am re-posting it.  Apologies if you are getting it twice.]

Let me start by confessing that I did not review the references in the MAHA Comission report I wrote about last week—except for mine.

The reference to my book, Food Politics, is a bit garbled (In Food Politics?  No.  This is Food Politics), but these are basically OK.  It’s easy to make mistakes like that one and I rely on the help of many proofreaders and factcheckers to try to avoid such errors in my published books and articles.  I checked a couple of the other references related to food topics and they seemed basically OK too.

So I was surprised by the report from NOTUS that The MAHA Report Cites Studies That Don’t Exist,

This finding was immediately attributed by the New York Times and other sources to the report’s having been referenced by Artificial Intelligence (AI), a tool well known to be scientifically inaccurate and to make things up.

To immediately plagiarize (well, quote) Ted Kyle at ConscienHealth: The MAHA Report: Make America Hallucinate Again.

I was also surprised—no, dismayed—by the administration’s response to these discoveries.

According to FoodFix,

White House Press Secretary, Karoline Leavitt told reporters…“I understand there were some formatting issues with the MAHA report that are being addressed,” Leavitt said. “But it does not negate the substance of the report, which, as you know, is one of the most transformative health reports that’s ever been released by the federal government, and is backed on good science that has never been recognized by the federal government.”

FoodFix also quotes the HHS Press Secretary:

Minor citation and formatting errors have been corrected, but the substance of the MAHA report remains the same — a historic and transformative assessment by the federal government to understand the chronic disease epidemic afflicting our nation’s children…“It’s time for the media to also focus on what matters.”

Formatting issues?  Oh come on.

Calley Means, the top advisor to RFK Jr, posted “The least surprising thing about the MAHA Report is that the media and failed medical leaders are talking about footnotes instead of its actual content.”

Sorry.  Footnotes matter.  Everything in a report making policy recommendations depends on where its information comes from.  Hallucinating references implies hallucinating data.

The MAHA Report is now being continually updated to fix the citation problem.

Some of the updates are introducing other errors. 

Yikes.

The Washington Post has published details: The MAHA Report’s AI fingerprints, annotated.

I was interviewed by Reuters about all this:

Nobody has ever accused RFK Jr. of academic rigor…The speed (of the MAHA report) suggests that it could not have been vetted carefully and must have been whisked through standard clearance procedures. The citation problem suggests a reliance on AI.”

Science magazine headlined the downplaying of the fake citations and pointed out the irony:

Problems with the MAHA report’s integrity came to light even as Kennedy has threatened to prevent government scientists from publishing in leading medical journals like The Lancet, The New England Journal of Medicine, and JAMA, which he claims are “corrupt” and controlled by pharmaceutical companies. Kennedy has instead proposed a state-run alternative.

Discovery of the fake citations also came just days after President Donald Trump unveiled an executive order that called for “Restoring Gold Science Standards” to government activities…One goal, Trump wrote, is to ensure that “Federal decisions are informed by the most credible, reliable, and impartial scientific evidence available.”

Yeah, right.  The MAHA report cites articles—26—from those “corrupt” journals as sources for its statements.

All of this has led cartoonists like Clay Bennett to ridicule the report.

Here’s another good one from Carlos Muñoz.

Ridicule—or lack of credibility if you prefer—is one reason why this matters.

What I had drilled into me as a graduate student in molecular biology was the importance of reading references, and never under any circumstances citing a reference I hadn’t read.

Why?  Because the credibility of my work depends on where I got my information—how I know what I claim to know.

When I managed the editorial process for the 1988 Surgeon General’s Report on Nutrition and Health, checking references was crucial to supporting the report’s recommendations.  It took years to get the report out, not least because of the enormous amount of vetting involved—from scientists, but also government agencies.

This report, unfortunately, was a rush job.  It astonished me that it got done in only three months (I really want to know who wrote it).

It’s one thing to make editorial errors in citing references (try as hard as I can to get them right, errors invariably get overlooked).

But this report had references that were made up.  Hallucinated.  This means nobody looked at them.

If its references are not reliable, nothing else in the report can be trusted either.

And that’s a shame.  It said a lot of things that badly needed to be said.

Too many corners were cut in throwing this together at the last minute.  I know this was a rush job because I have four versions of the report.

• Version 1: May 22 early morning.  It has no references at all.  My first reaction: What?  No references?
• Version 2: Also May 22, but later in the morning. It has references but not much in the way of formatting.
• Version 3: May 22. This is the one released to the public.
• Version 4: This is the one online on May 31.  Other versions, perhaps multiple, came in between and may come after.

None of this bodes well for the future of MAHA initiatiatives.  Sad.

 

Let me start by confessing that I did not review the references in the MAHA Comission report I wrote about last week—except for mine.

The reference to my book, Food Politics, is a bit garbled (In Food Politics?  No.  This is Food Politics), but these are basically OK.  It’s easy to make mistakes like that one and I rely on the help of many proofreaders and factcheckers to try to avoid such errors in my published work.  I checked a couple of the other references related to food topics and they seemed basically OK too.

So I was surprised by the report from NOTUS that The MAHA Report Cites Studies That Don’t Exist,

This finding was immediately attributed by the New York Times and other sources to the report’s having been referenced by Artificial Intelligence (AI), a tool well known to be scientifically inaccurate and to make things up.

To immediately plagiarize (well, quote) Ted Kyle at ConscienHealth: The MAHA Report: Make America Hallucinate Again.

I was also surprised—no, dismayed—by the administration’s response to these discoveries.

According to FoodFix,

White House Press Secretary, Karoline Leavitt told reporters…“I understand there were some formatting issues with the MAHA report that are being addressed,” Leavitt said. “But it does not negate the substance of the report, which, as you know, is one of the most transformative health reports that’s ever been released by the federal government, and is backed on good science that has never been recognized by the federal government.”

FoodFix also quotes the HHS Press Secretary:

Minor citation and formatting errors have been corrected, but the substance of the MAHA report remains the same — a historic and transformative assessment by the federal government to understand the chronic disease epidemic afflicting our nation’s children…“It’s time for the media to also focus on what matters.”

Formatting issues?  Oh come on.

Calley Means, the top advisor to RFK Jr, posted “The least surprising thing about the MAHA Report is that the media and failed medical leaders are talking about footnotes instead of its actual content.”

Sorry.  Footnotes matter.  Everything in a report making policy recommendations depends on where its information comes from.  Hallucinating references implies hallucinating data.

The MAHA Report is now being continually updated to fix the citation problem.

Some of the updates are introducing other errors. 

Yikes.

The Washington Post has published details: The MAHA Report’s AI fingerprints, annotated.

I was interviewed by Reuters about all this:

Nobody has ever accused RFK Jr. of academic rigor…The speed (of the MAHA report) suggests that it could not have been vetted carefully and must have been whisked through standard clearance procedures. The citation problem suggests a reliance on AI.”

Science magazine headlined the downplaying of the fake citations and pointed out the irony:

Problems with the MAHA report’s integrity came to light even as Kennedy has threatened to prevent government scientists from publishing in leading medical journals like The Lancet, The New England Journal of Medicine, and JAMA, which he claims are “corrupt” and controlled by pharmaceutical companies. Kennedy has instead proposed a state-run alternative.

Discovery of the fake citations also came just days after President Donald Trump unveiled an executive order that called for “Restoring Gold Science Standards” to government activities…One goal, Trump wrote, is to ensure that “Federal decisions are informed by the most credible, reliable, and impartial scientific evidence available.”

Yeah, right.

All of this has led cartoonists like Clay Bennett to ridicule the report.

Here’s another good one from Carlos Muñoz.

Ridicule—or lack of credibility if you prefer—is one reason why this matters.

What I had drilled into me as a graduate student in molecular biology was the importance of reading references, and never under any circumstances citing a reference I hadn’t read.

Why?  Because the credibility of my work depends on where I got my information—how I know what I claim to know.

When I managed the editorial process for the 1988 Surgeon General’s Report on Nutrition and Health, checking references was crucial to supporting the report’s recommendations.  It took years to get the report out, not least because of the enormous amount of vetting involved—from scientists, but also government agencies.

This report, unfortunately, was a rush job.  It astonished me that it got done in only three months (I really want to know who wrote it).

It’s one thing to make editorial errors in citing references (try as hard as I can to get them right, errors invariably get overlooked).

But this report had references that were made up.  Hallucinated.  This means nobody looked at them.

If its references are not reliable, nothing else in the report can be trusted either.

And that’s a shame.  It said a lot of things that badly needed to be said.

Too many corners were cut in throwing this together at the last minute.  I know this was a rush job because I have four versions of the report.

• Version 1: May 22 early morning.  It has no references at all.  My first reaction: What?  No references?
• Version 2: Also May 22, but later in the morning. It has references but not much in the way of formatting.
• Version 3: May 22. This is the one released to the public.
• Version 4: This is the one online on May 31.  Other versions, perhaps multiple, came in between and may come after.

None of this bodes well for the future of MAHA initiatiatives.  Sad.

 

I’ve written previously about the list of words that automatically disqualify applicants for federal grants, but the New York Times has now published a more complete list of about 200 forbidden terms, along with examples of editing of federal websites.

These are perceived as “woke,” which the Trump administration especially opposes.

But many of these terms describe reality.

As a long time public health advocate, I take this quite personally.

Now, you can’t research or write about anything having to do with underrepresented minorities or inequality, but you also can’t do anything related to terms like these.

• activists
• advocates
• at risk
• barrier
• climate science
• confirmation bias
• female
• health disparity
• institutional
• mental health
• political
• privilege
• sociocultural
• systemic
• undervalued
• women

This is right out of George Orwell’s 1984.  It would be funny, if it weren’t having an effect.  I know of at least one instance of a federal scientist had to remove his name from a paper because it dealt with inequity—a topic very much on this list.

This kind of group-think deserves ridicule and firm pushback.

I don’t know about anyone else, but I intend to keep writing about these issues and flaunt the scarlet A—for Activist and Advocate. (along with a W for woman).

A member of the Academy of Nutrition and Dietetics forwarded this email sent to academy members.

From: PepsiCo Health & Nutrition Sciences <pepsiconutritionscience@pepsico.com>
Subject: Help combat nutrition misinformation 📢

We’re sure you’ve seen this firsthand with your patients and clients: Nutrition research has become increasingly complex for the general public to understand – and the volume of contradictory headlines and misinformation in the media doesn’t help. As a healthcare professional, you have the power to inspire trust and deepen the general public’s understanding through credible communication of balanced, high-quality, evidence-based nutrition science.

That’s why we hosted our most recent Lab & Learn webinar, Communicating Evidence-Based Nutrition Science Effectively, on the topic. Whether you attended live or viewing on demand, we wanted to share an additional resource with you on this topic to enrich your practice even further.

Didn’t get a chance to tune in to the webinar live last week?
Watch on demand here and earn 1.25 free CPEUs!
Communicating Evidence-Based Nutrition Science Effectively awards 1.25 CPEUs in accordance with the Commission on Dietetic Registration’s CPEU Prior Approval Program.

Download the handout here.

Comment

Who better than PepsiCo to counter nutrition misinformation?  The handout gives standard information about how to interpret scientific studies, and useful for that purpose.  Perhaps it is an oversight but it omits any mention of biases introduced by funding by food companies.

More important, it implies that science alone will be enough to counter misinformation.  It would be nice if erroneous beliefs about nutrition could be corrected by presenting facts, but beliefs, especially those that are deeply held, are not necessarily fact-dependent.  They often have more to do with faith in what trusted people say.

PepsiCo wants dietitians to trust PepsiCo and avoid advising clients to cut back on sugary beverages or salty snacks.

The dietitian who sent this to me was skeptical, as dietitians should be in situations like these.

An anonymous reader sent me this list, purportedly from the National Science Foundation, of words that disqualify scientists from submitting papers for publication, and applicants from getting grants.  The reader also sent the NSF decision tree for use of these words.  I cannot trace the original source of this material.  If you have any information about whether this is for real, please let me know.  My favorite words on this list?  Advocate and women.

The disqualifying words:

• activism
• activists
• advocacy
• advocate
• advocates
• antiracist
• barrier
• barriers
• biased
• biased toward
• biases
• biases towards
• bipoc
• black and latinx
• community diversity
• community equity
• cultural differences
• cultural heritage
• culturally responsive
• disabilities
• disability
• discriminated
• discrimination
• discriminatory
• diverse backgrounds
• diverse communities
• diverse community
• diverse group
• diverse groups
• diversified
• diversify
• diversifying
• diversity and inclusion
• diversity equity
• enhance the diversity
• enhancing diversity
• equal opportunity
• equality
• equitable
• equity
• ethnicity
• excluded
• female
• females
• fostering inclusivity
• gender
• gender diversity
• genders
• hate speech
• hispanic minority
• historically
• implicit bias
• implicit biases
• inclusion
• inclusive
• inclusiveness
• inclusivity
• increase diversity
• increase the diversity
• indigenous community
• inequalities
• inequality
• inequitable
• inequities
• institutional
• lgbt
• marginalize
• marginalized
• minorities
• minority
• multicultural
• polarization
• political
• prejudice
• privileges
• promoting diversity
• race and ethnicity
• racial
• racial diversity
• racial inequality
• racial justice
• racially
• racism
• sense of belonging
• sexual preferences
• social justice
• socio cultural
• socio economic
• sociocultural
• socioeconomic status
• stereotypes
• systemic
• trauma
• under appreciated
• under represented
• under served
• underrepresentation
• underrepresented
• underserved
• undervalued
• victim
• women
• women and underrepresented

The NSF decision tree for disqualifying papers or grant applications using those words:

Tomorrow: How to comment on all of this.

Resources sent by readers

• A Washington Post story on the list
• The likely effects of the list on researchers at UC San Diego
• What caused all this, according to the Senate Commerce Committee
• The full report of the Senate Commerce Committee (see Appendix B for much longer list of trigger words)

PCAST, the President’s Council of Advisors on Science and Technology, has released its REPORT TO THE PRESIDENT A Vision for Advancing Nutrition Science in the United States.

I learned about the report from a Tweet (X): 

I wrote about an earlier draft of the report in a previous post: The federal vision for chronic disease prevention: individual behavior, not the environment.  I called for the report to take on the need for fundamental improvements in the food environment aimed at preventing obesity-influenced chronic diseases.

If you read the fine print, the report has indeed done some of that.  It now mentions ultra-processed foods, for example,

people’s food selections are complex, influenced by various factors in a multifaceted U.S. (and global) food ecosystem, with many of these factors beyond an individual’s control, e.g., increased production and availability of ultraprocessed foods which are associated with overconsumption and obesity. In addition to widespread availability of inexpensive ultra-processed foods, the U.S. food environment has undergone huge changes in recent decades, including easy access to low-cost fast food and eating away-from-home becoming much more common…in the era of widespread internet and digital technology access and use, people’s food habits increasingly are influenced by advertising and social media, which are sources of both facts and misinformation. Acknowledging and understanding these factors and their intersections is critical to addressing nutrition-related health disparities.

It also says useful things like these:

• new emphasis must be placed on nutrition research that can equitably and effectively help all Americans achieve better health.
• [needed is an] equity focus that particularly considers those who are disproportionately affected—racially, ethnically, and socially minoritized groups—due to long-standing and structural inequities which make it hard for many people to eat healthy and be physically active.
• For such a highly developed nation, the U.S. has distressingly high rates of food insecurity, imbalanced nutrition, and inequities in food access, all further exacerbated by the pandemic.  With diet-related disease rates increasing, we have responded by focusing resources on costly medical treatments, further widening disparities and directing efforts away from prevention or addressing social determinants of health and a food environment that for too many Americans does not provide or promote good nutrition. The only way to reverse these trends and achieve robust health for our nation is to focus on prevention, which will require significant modifications of our overall food environment and must be informed by improved nutrition research.
• Preventing diet-related chronic diseases is among the most urgent public health challenges facing the nation.

Despite these statements, its two recommendations say nothing beyond the need for coordination aimed at addressing that challenge.

1. The Administration should implement a coordinated and sustained federal interagency effort, co-led by HHS and USDA, to strengthen the nutrition science base for current and future public and private sector actions to reduce the burden of diet-related chronic disease and maintain momentum toward the President’s 2030 goal.

2. To ensure equitable access to the benefits of nutrition research, federal agencies should prioritize equity in nutrition research, focus research on improving program delivery, continue efforts to diversify the nutrition science and dietetics workforce and engage the academic and private sectors in multisector research and intervention initiatives.

Yes, coordination would be a big help.  Nutrition research is all over the place at the federal level.  So would increased funding for nutrition research aimed at improving the food environment to prevent chronic disease.  Only a tiny fraction of the NIH budget goes for this purpose.  NIH’s main nutrition focus is “precision nutrition” aimed at individuals, not public health.  And much of the USDA’s nutrition funding goes to the kinds of industry-funded studies I post here on Mondays.

The report mentions what’s needed in theory; it ducks dealing with the tough politics of chronic disease prevention.

And alas, it did not cite my suggestions for what is needed (which I had sent to the committee).

So where is leadership for chronic disease prevention at the federal level?  It’s in an odd place at the moment, as I will discuss tomorrow.