RFK Jr Gives the NY Times Reason to Write About Keto and Schizophrenia. Good News?

Not initially, but sometimes even bad press can become good (enough) press.

“We now know that the things that you eat are driving mental illness in this country,” Mr. Kennedy told a crowd at the Tennessee State Capitol, adding that a doctor at Harvard had “cured schizophrenia using keto diets.”

Sheryl Gay Stolberg, The New York Times, February 5, 2026

I was hard at work this week on a post about NIH director Jay Bhattacharya publicly describing the state of modern medical science as a “disaster,” when I got distracted. The New York Times had finally reported on the budding field of metabolic psychiatry. In a nutshell, that’s the use of diet therapy for cognitive disorders, keto specifically.

But the Times wasn’t reporting on the subject because its health reporters were willing to take the research seriously, but because Robert F. Kennedy Jr did. And if RFK Jr did, then the Times could do what it typically does and assume he was being irresponsible. Hence, the headline:

Sheryl Gay Stolberg, who covered the story, is not a health reporter. Her beat is health policy and politics. Here’s how she describes her interests in her NYT bio:

My stories focus on policy debates over issues including the coronavirus pandemic, reproductive rights, gun violence prevention, the ethics of virology research and the cost of prescription drugs. I’m interested in how the partisanship that divides America is affecting the nation’s health.

From this perspective, RFK Jr is the living embodiment of political partisanship in medical science. The popularity of keto diets, which tend to be meat-rich, may also be influenced by political affiliation, but I can find no evidence that Stolberg has ever reported on the subject. (The Times search function is notoriously difficult when it comes to these kinds of searches, but Google suggests not.)

Hence, Stolberg’s article is not about the efficacy of keto therapy for schizophrenia. It’s about what Kennedy said, as stated in the epigraph at the top. Then, of course, the Times can imply that RFK Jr was doing what RFK Jr often does, which is to speak irresponsibly and promote quackery.

Mr. Kennedy has a history of promoting ideas with little to no scientific evidence to back them up. He has rejected established evidence that H.I.V. is the cause of AIDS, pushed the idea that Covid-19 was “ethnically targeted” to spare Jews and Chinese people, and repeatedly insisted that vaccines are a possible cause of autism despite a lack of proof.

This is where the adjective knee-jerk comes to mind for Stolberg’s reporting, but I can’t blame Stolberg, or at least not entirely. She’s working on a deadline. She knows RFK Jr has a history of overinterpreting or misinterpreting evidence. No, we do not “know that the things that you eat are driving mental illness in this country,” as RFK Jr is quoted as saying; we know that they might be, but that’s the difference between a hypothesis and a fact. Now, keto and the claim of a “cure” become more ammo to make this case.¹

Because RFK Jr was discussing the work of Chris Palmer at Harvard, whose name he mentions, Stolberg discusses the easily discoverable source for RFK Jr's claim:

Mr. Kennedy was apparently referring to Dr. Christopher Palmer, who in 2019 wrote about “two patients with longstanding schizophrenia who experienced complete remission of symptoms” with the keto diet. He said both patients “were able to stop antipsychotic medications and have remained in remission for years now.”

More recently, Dr. Palmer and his colleagues described the diet as a “promising therapeutic approach for schizophrenia.” Dr. Palmer did not immediately reply to a request for comment.

Remember that last sentence, because we’ll get back to it. It’s important.

Stolberg also does her due diligence and quotes a pair of Columbia University psychiatrists who are what Times reporters would consider legitimate authorities in this field. That means they have impressive credentials: they’re at a university that has a very low acceptance rate, and they can represent what stands for conventional wisdom in the discipline. Not surprisingly, the authorities are skeptical.

Stolberg uses these authorities to describe the state of the evidence from the position of mainstream psychiatry:

Some small short-term studies, including one at Stanford University, “offer very preliminary evidence” that the diet “might be helpful” in patients with schizophrenia, said Dr. Paul S. Appelbaum, a professor of psychiatry at Columbia University and past president of the American Psychiatric Association.

But it is “simply misleading to suggest that we know that ketogenic diets can improve schizophrenia symptoms, much less that they can ‘cure’ the condition,” he said.

Dr. Appelbaum said most of the patients in those studies continued to require antipsychotic medication. He and another Columbia psychiatry professor, Dr. Mark Olfson, both said that while the early research was promising, more thorough long-term studies were needed.

“There is currently no credible evidence that ketogenic diets cure schizophrenia,” Dr. Olfson said. Most of the studies that tested the ketogenic diet as a treatment for mental health disorders, including the Stanford study, did not include a control group that followed a regular diet

“No credible evidence” is a strong statement. That’s the damning indictment.

When I got into this business of reporting on nutrition and chronic disease science, it was that kind of description of evidence that made me wonder whether researchers whose training is in medicine, not science, are all too likely to be bad scientists when they attempt to do it.

If there is a controversy, then there is credible evidence. It may not be compelling enough to believe the minority view, but to say “there is no credible evidence” is always to either misinterpret the existing evidence or to implicitly define the word credible as meaning that which agrees with my preconceptions. Both are indicative of someone, no matter how well-credentialed, who has no deep understanding of what science is and how it’s done. I wouldn’t expect that from a psychiatrist (as I’ll discuss in the post that has been delayed because of this one). It’s always disappointing, though.

Either way, the no credible evidence claim is the takeaway from Stolberg’s article. Nutrition therapy for cognitive disorders and the entire emerging field of metabolic psychiatry finally make it into the paper of record, but only in the context of RFK Jr and his questionable wisdom on medical science. Damned by association and by insinuation, at the very best. And, of course, from the NY Times, it spreads…

Now the good news: The Times doesn’t stop there

Remember, as I requested above, Stolberg said that Chris Palmer “did not immediately respond to a request for comment.” But he did eventually respond.

And Stolberg talked to him, or maybe Ellen Barry did, who would co-write the follow-up article and whose beat is mental health:

I write about people living with mental illness, the big ideas defining the field, and systems set up to provide care.

Now we get story number two, published the next day, a follow-up. This one is based on the interview with Palmer, so we get his perspective. Of course, the Times being the Times (or maybe its headline writers being headline writers), the article was still couched in the context of RFK Jr’s credibility.

But the Palmer interview shifts perspective. Instead of those legitimate medical authorities being used to counter RFK Jr’s characterization of Palmer’s work, Palmer gets to speak for himself and for the field, and to respond to the authorities. Now we get Appelbaum first, and then we get Palmer, the inverse of the first article:

Dr. Paul S. Appelbaum, a Columbia University psychiatry professor and past president of the association, described the evidence in favor of the diet as “very, very preliminary.” He called for studies comparing patients both on and off the diet, who were followed for six to 12 months and evaluated by people who did not know which patient was eating what.

“Until then, it’s simply misleading to suggest that we know that ketogenic diets can improve schizophrenia symptoms, much less that they can cure the condition,” he said.

Still, there is intense enthusiasm among patients around Dr. Palmer and his work.

“This has been my heartbreaking journey for the last several years,” Dr. Palmer said, “because I literally have thousands of people reaching out to me begging for this treatment, and the reality is there are not many clinicians” who offer it.

Dr. Palmer’s patients include the son of Jan Ellison Baszucki and her husband, David Baszucki, the chief executive of the online gaming platform Roblox and a major donor to President Trump’s White House ballroom project. Ms. Baszucki, a former Silicon Valley marketing executive, founded a nonprofit, Metabolic Mind, to fund and promote research on diet and mental health.

Dr. Palmer said there were around 20 trials underway using ketogenic therapies for a range of psychiatric disorders, including two randomized control [sic] trials in patients with schizophrenia, whose results would be published within the next year. In addition, he said, Wellcome, a global health foundation, is funding a large trial in Britain.

And Palmer makes all this palatable to the Times by pointing out that while his research may have been embraced by RFK Jr and the Make America Healthy Again (MAHA) Initiative, that does not mean his politics are more aligned with the current administration. Hence the Times ends the story (nicely) by suggesting that we can take Palmer seriously because he’s not one of those people (and we know who those people are).

As an assistant professor of psychiatry at Harvard Medical School, Dr. Palmer is also affiliated with Mass General Brigham. He said Mr. Trump’s cuts to biomedical research, and the continuing legal battles between Harvard and the Trump administration over the president’s demands that the university pay hundreds of millions of dollars to the government, were affecting his work.

“I think in some camps, they do affiliate me somehow with MAHA and I want to just remind you, I’m at Harvard Medical School,” Dr. Palmer said. He went on: “Trump was not very kind to Harvard Medical School, and I’m working at MGB. Both institutions are in dire financial problems, and that is absolutely affecting me.”

By day two, for those who are reading, we have a reasonably good article on ketogenic diet therapy for schizophrenia, although buried in the RFK Jr MAHA context. We also have at least two Times reporters who might follow-up in the future, as they now have the minimum necessary understanding of the research and it’s importance.

The need for precision when speaking about evidence

Even here, I have issues with how the evidence is discussed on all sides, from RFK Jr on down. For starters, as I noted, RFK Jr oversells the hypothesis that cognitive disorders might be influenced by diet, if not driven by diet. He could certainly be right, but he doesn’t know.

Maybe if we lived in a world without sugar and sugary beverages, as I suggest in my book The Case Against Sugar, we’d live in a world in which obesity and diabetes are the uncommon disorders they used to be. Maybe the same would be true of cognitive disorders like schizophrenia and depression. I wouldn’t be surprised, but I’m biased. Do we know this to be true? No.

But what about the cure statement? Here’s Stolberg’s first article, quoting an article that Palmer co-authored in 2019 with his Harvard colleague Javier Gilbert-Jaramillo, and Eric Westman at Duke:

“two patients with longstanding schizophrenia who experienced complete remission of symptoms” with the keto diet. He said both patients “were able to stop antipsychotic medications and have remained in remission for years now.”

Here’s the paper, and one of the patients was Eric Westman’s. (Westman did a fair share of the early research on keto diets, back in the 00s when they were still known as Atkins.)

The point is that stopping antipsychotic medications and remaining in remission looks very much like a cure, although it’s likely to be dependent on the patients remaining on the diet. It’s also possible, though, that remission in these patients was a coincidence, spontaneous remissions unrelated to diet, which is why randomized controlled trials are necessary to establish what’s really happening.

Still, the correct statement of the evidence is that a cure may be possible, even if we might personally consider that to be an unlikely possibility.

That also raises the issue of whether a patient can be considered cured if falling off the diet will result in a return of the symptoms. That’s also a likely scenario. If that’s the case than remission is the far better term.

What is it with medical doctors? Is it them or is it me?

This brings me to my pet peeve, which is (I regret to admit) why I wrote this post. This is the tendency of physicians on all sides of these controversies to be clueless about how to discuss evidence, particularly on subjects about which they seem to know little or nothing.

If they were lawyers, we could hope they’d be disbarred. In medical science, and particularly on social media, it seems to be standard operating procedure. Do physicians consider it way to establish their bona fides, which is to say, why they’re a better source of medical knowledge than their competition?

The Columbia psychiatrists were doing this to some extent in the Times articles, but they weren’t on X where they might be liberated from any social restraints to speak thoughtfully.

The offender who caught my attention on X is Neil Stone, a British physician who studies infectious diseases. His X bio says he is “Officially the 1st person to use word #Covid,” which might be reason enough to worry about his ego. More importantly, he seems to have no history of studying either mental health or chronic diseases.

I’d like to assume he’s a reasonably smart man. His accomplishments suggest as much. So why does Stone dip his toe into this pond at all, particularly after Stolberg’s second article was published? And why does he do it as though he knows with certainty what he is talking about?

Now of those (then) 175 comments (and the innumerable comments on the comments), most are disagreeing with him. Many are providing the evidence to make their case that Stone is the one is who is uninformed. If nothing else, it should be clear to Stone that his certainty on this subject is unwarranted.

How does he respond? He reposts his original post. He doubles down (to borrow from the gambling jargon).

This is a case study in miniature of what’s happened to medical science over the past quarter-century, the reason why so many of us have lost faith in the endeavor. Instead of discussing the uncertainty surrounding these vitally important issues, as scientists are (supposedly) trained to do, these medical authorities pretend there is none. Instead of acknowledging what they don’t know, they try to convince us that the evidence is unambiguous.

Now, I, too, disagree with how RFK Jr phrased his discussion of diet and mental health, and I wish he, too, understood how best to communicate the state of science when we do not know the truth. If so, the NY Times would not have had the opening to smack him again and to damn keto therapy by association.

But here’s what we do know.

Per Palmer’s publication and other sources, we know of anecdotal cases in which schizophrenia appears to have been put into remission with a dietary intervention.

These are well-documented cases: Physicians reporting their clinical observations, and we can trust that they reported them honestly. In some cases, these are patients themselves reporting, and we should believe their accounts, even as we can question the interpretation from both sources.

Does that mean schizophrenia has been cured by diet, or can be cured by diet?

No, but it certainly suggests that this is a possibility, or what researchers in a functional science might think of as a viable working hypothesis.

We also know that this hypothesis—keto cures schizophrenia—is not, on the face of it, absurd.

We have a precedent for an intractable cognitive disorder that can be cured by diet, essentially the same diet at issue here. This is intractable pediatric epilepsy. A significant proportion (10-15%) of these otherwise intractable cases—those in which drug therapy is incapable of controlling the seizures—can be cured by ketogenic diets. This means that the children afflicted can return to eating much like other children, and their seizures do not return. Perhaps a third can be put into remission, which means they have to stay on the diet if they want to remain seizure-free.

So this is a viable hypothesis, and a very testable hypothesis. Preliminary tests have found evidence to support it, as Palmer said and Stolberg reported. And researchers who are as well-credentialed as Stone (if not more so) and as smart as I would like to think he is (albeit far more informed on this subject) are conducting further trials.

It may be true that we can cure schizophrenia by diet, or put it into remission in many or even most cases, but we don’t yet know.

That’s what you communicate.

If you insist, as Stone did, that schizophrenia cannot be cured by diet, then you dig yourself into a hole that is hard to escape. You reject even the possibility of a viable working hypothesis, suggesting you will be biased in the future to any evidence that suggests otherwise.

The reason good scientists have always counseled humility and precision in describing the evidence is precisely this: Should the evidence emerge to establish that your preferred hypothesis is wrong, you are more likely to accept it and even admit it publicly. It’s always easier to say, “Here’s how the evidence has evolved to support the hypothesis and overcome my initial skepticism,” than to say, “Oops, I guess I was dead wrong about that, but you can trust my opinion about everything else.”

As a general word of caution, the learning experience here is a bipartisan one. Anyone who makes these kinds of unambiguous, declarative statements about any controversial area of medicine—whether on the side of the establishment or the side of the heretics (aka the quacks, depending on your perspective)—cannot be trusted.

If there’s a controversy, it means there’s evidence on both sides. Just because we can’t trust one side, doesn’t mean we can trust the other.

Had Stone written, “I do not believe that schizophrenia can be cured by diet,” I would have had no issue with him. Go ahead and repost it. One thing Stone can know with reasonable certainty is what he believes. He cannot know, though, whether what he believes is really so. Maybe Stone assumes that his belief is implicitly communicated in a post. But he’s a medical doctor, a voice of authority. He should know better.

Good scientists will discuss the evidence, pro and con, and explain why they find one side compelling and the other not. Bad scientists, and all those physicians and influencers who don’t understand the nature of scientific evidence, will make declarative statements as Stone did.

It doesn’t mean they’re not right, but I wouldn’t take their word for it.

1. The Times notes at the end of Stolberg’s article that she received assistance in reporting it from Alice Callahan, a nutrition reporter who consistently takes the position that authority figures are correct. I understand why she does, and have written about why this is a problem when reporting about nutrition and chronic disease (here). Still, I wonder if the knee-jerk aspects of the reporting on keto in Stolberg’s article—hard to adhere to and linked to increased heart disease risk—are due to her influence, despite the fact that she is the only reporter assisting on the article who should know better. Other assistance came from Emily Cochrane, who covers southern politics for the Times, and Ellen Barry, who covers mental health. Barry, as I note, co-authored the second article.