Fighting Obesity? The New England Journal of Medicine Has a Calorie Fixation

If the journal editors can't accept uncertainty in the science of weight regulation, the physicians never will.

Quiz of the week.

Question 1 (with apologies for evoking any unpleasant memories from high school biology courses): Imagine you’re feeding a mouse a very low fat diet. Four percent or even 1 percent of the calories are from fat and the rest mostly carbs. And you’re underfeeding it, so the animal is losing weight. In technical terms it’s in negative energy balance.

What happens to (most of) the fat it’s consuming, the dietary fat?

A. Well, the animal is losing weight, so it’s losing fat. Must be burning it up.

B. This question has to be a trick question or it wouldn’t be on the quiz. So…. it stores the fat, but I have no idea why or how.

Question 2, courtesy of The New England Journal of Medicine and last week’s “Clinical Decisions” feature: Imagine you’re a physician (in the event you’re not), and your patient is a 42-year-old man concerned about his weight. He’s recently had a heart attack, and the surgeons have attended to his clogged arteries. He’s on a statin for his cholesterol and a drug for high blood pressure.

He’s obese—5’11” and 258 pounds—and he smokes.

Over the past 7 months, he has been trying to follow a diet of 1500 kcal per day and has walked his dog daily for 30 minutes. However, he finds it difficult to adhere to the diet, primarily because of frequent cravings for sweet snacks; he has lost only 2 kg of weight. He has reduced his smoking to three cigarettes per day and consumes a glass of wine about once a week.

His labs are normal, so, apparently, is his blood sugar. The clinical problem is his weight. He’s asking about surgery or drug therapy.

What do you tell him?

A. Let’s try Ozempic? It’s “been rigorously proven to lower cardiovascular risk while producing clinically meaningful reductions in adiposity.”

B. Bariatric surgery? After all, “[m]etabolic and bariatric surgical procedures are safe and achieve meaningful weight loss in most patients and result in abatement of the obesity-associated conditions.”

C. Suck it up, and do a better job sticking to your diet. There’s no way he’s really eating 1500 kcal/day because if he was, he’d have lost more than five pounds in 7 months.

D. We’ve known for half a century or more that eating less and exercising has little meaningful effect on body weight, and we apologize for the physician who gave that advice and has had you semi-starving yourself for the past six months. Evidence accumulating since the early 2000s suggests a low-carb, high-fat ketogenic diet might put his obesity into remission and lower his heart disease risk in the process. (It should also lower his blood pressure.) He won’t have to count his calories, and it’s worth a concerted effort. If it doesn’t work for him, then we can discuss options one or two.

Short answers.

Question 1: Yes, it was a trick question, so B is the correct guess.

Question 2: Well, clearly I think option 4 is the correct answer, but the NEJM didn’t offer either that or option 3 (to its credit).

In August 2025, the two options were drugs or surgery and nothing more.

The two quiz questions are related, as we’ll discuss. Ultimately, and once again, these questions are asking whether a calorie is a calorie when it comes to regulating body fat and what the therapeutic implications are. Should we expect calorie restriction—1500 calories daily, for a man who weighs almost 260 pounds?—to have the same effect as carbohydrate restriction? The editors at the NEJM and the physicians asked to defend the options certainly seem to think so.

Are they right to do so? Not only do I very much doubt it (unsurprisingly), but I’ll argue the evidence to refute it was clear 80 years ago, which is when two Columbia University biochemists reported the results of that mouse experiment in question 1.

Let’s start, though, with question two because it’s the context we’re dealing with today.

Will doctors make the effort to understand weight regulation (if they’re not obese themselves)?

The problem with a conventional wisdom of any sort in medicine, right or wrong, is that it is almost impossible to dislodge. Physicians will believe it, since this is what they would have been taught in medical school (even as they might have been told by their professors that half of what they were being taught is wrong). And the longer they hold onto the belief, the more they have to doubt their own judgement should they accept it’s incorrect.

Moreover, group think among physicians is a powerful a force. Physicians will believe what their colleagues believe and their colleagues will believe what they were taught in medical school. The path of least resistance in their careers and in their clinics will always be aligning with the group.

While the results of major clinical trials can change thinking on treatment options (new pharmaceuticals, medical devices or surgical techniques because those are relatively easy to test) but only whether the new options tested should replace old ones. Shifts in therapeutic thinking will face far less inertia and far less resistance than shifts in thinking about the pathophysiology or etiology of the disorders themselves—i.e., the hard science—because they don’t require any acknowledgement that mistakes in judgement have been made.¹

For physicians to fundamentally change how they think about the cause and the mechanism underlying a particular common disease. they have to accept that the way they used to think about it was wrong. And only did they misconceive the underlying science (which they will typically take pride in understanding), but so did their colleagues and their group. In theory it can happen. In practice it is exceedingly rare. And the more mature the particular medical discipline—the longer it’s been around, the greater the number of researchers—the greater the resistance to what scientists would call alternative hypotheses and the more powerful the existing group think.

If physicians have problems controlling their own weight, they might find themselves experimenting with unconventional diets (as many of us in this world did), and if those work, if they help significantly, they might explore the thinking behind those diets. If not, the easiest belief system to embrace is the one shared by your colleagues. Confronted with an obese patient, in other words, a calorie is a calorie and any diet that restricts calories works the same as any other (even those that don’t).

In this world, the influential medical journals can play a role in getting physicians to understand that the underlying science, at the very least, is open to question. But then the journal editors themselves have to care and, well, they’re just doctors, too.

Clinical Decisions and the inevitability of calories

When the NEJM launched its monthly Clinical Decision feature back in May 2007, the editors described it as “an experiment in understanding the therapeutic decisions that physicians make every day at the bedside and in the office.”

Presented with the results of a new clinical trial or a particularly newsworthy medical controversy, the editors create a fictitious case vignette and then choose physicians with the necessary expertise to present different clinical options for treatment: 3 in the early days, now down to two. “We realize that by writing the case vignette and outlining the management options,” the editors explained in their introductory editorial, “we are limiting the options of both our expert commentators and our readers. But only by focusing the question can we expect a clear answer to emerge.”

Regrettably, though, a clear answer can still be a wrong answer or, at least, an inadequate one.

In the 18 years since, the Clinical Decision have covered everything from the kind of specific clinical decisions of interest, it would seem, only to physicians (anticoagulation for secondary atrial fibrillation?) to the kinds of discussions that show up in the newspaper and are of far broader interest: recommending mammography screening, for instance, or how to manage prostate cancer. How about medical marijuana for chronic pain? Statin therapy? How about for children? Physician-assisted suicide? Even the regulation of sugar-sweetened beverages?

Along the way, we’ve also gotten sporadic glimpses of how the NEJM editors think about weight reduction and it’s naive—eat less, move more—and how their thinking, even in these seemingly benign clinical decision vignettes, then reinforces the group think in the field.²

In January 2008, for instance, the fictitious case study was a 55-year-old woman with type 2 diabetes, obesity and hypertension. She had “successfully lost weight (from 5 to 12 kg) on various diets but each time has regained all of the weight she lost.” Hence, all three options presented were pharmaceutical, all aimed at controlling her blood sugar. The underlying assumption: a patient who had failed to maintain weight on various diets could be expected to fail on all of them.

In 2014, the fictitious vignette was about Martin, a 61-year-old mildly obese man. Martin, we are told, is married, which is relevant because he’s asking about testosterone replacement therapy (TRT). He had gone to an online website and answered “yes” to whether he had “decreased energy, decreased ability to play sports, and decreased sexual pleasure and erectile function.” The site suggested he speak to his physician about TRT and so here he is. The physician arguing against the TRT did so on the basis that the longterm risks and benefits were still unknown, and, well, Martin should just lose weight:

To address Martin's symptoms, I would recommend daily exercise (at a more vigorous level than recreational tennis) and reduced caloric intake. If he loses weight, his serum testosterone level might increase. These changes are more likely than “a brew of T” to improve his cardiovascular fitness, sense of well-being, and overall sexual function.

The first time a clinical decision addressed weight loss specifically, it was September 2016, and a vignette about Ms. Chatham, a relatively healthy 29-year-old woman who weighed over 200 pounds. (She was 5/’7”, so a BMI of 32).

She says that she has been told indirectly, by her friends and family, that she is “overweight.” She has tried several popular diets without success; each time, she has lost 4.5 to 6.8 kg (10 to 15 lb) but has been unable to maintain the weight loss for more than a few months.

This was the early days of the clinical trials using GLP-1 drugs for weight loss—the NEJM had recently published the results of a trial on Liraglutide—but it was still unknown what effects the drug would have, for better or worse, on longterm chronic disease risk.³

Hence, the clinical decision options circa 2016:

1. Start (or, rather, restart) lifestyle modification (calorie restriction and exercise) and add an FDA-approved drug,

2. Maximize lifestyle modification (a more concerted effort of eating less and exercising) and put off the drugs.

Here’s the argument to add a drug immediately (with my emphasis):

The primary rationale for the use of weight-loss medications is to improve dietary adherence, which is one of the most important predictors of successful weight loss. Since the primary biologic effect of most weight-loss medications is reduced hunger, increased satiety, or both, the use of pharmacotherapy enables patients to adhere to a dietary plan that includes a reduction in calories and to do so with better control and a reduced sense of deprivation.

And the argument against (also with my emphasis):

No drug has been shown to improve any clinically meaningful health outcomes (i.e., beyond a surrogate measure such as weight or glycated hemoglobin level), and patients almost invariably regain the weight they lost once the medications are stopped. Are we seriously proposing indefinite medication of up to a third of the population of young women (and half the black women) who, like Ms. Chatham, are obese, at a cost of up to $1,000 per month per person? Are we sending a false message that we can address this public health problem with drugs?

… The makers of weight-loss drugs cynically exploit the vulnerabilities of stigmatized patients while failing to help them create the more healthful food and physical environments that are necessary to genuinely address the global obesity epidemic. Approaches to obesity that direct us away from working with our patients and from harnessing multipronged community-wide efforts to target the actual causes and best solutions to obesity are misguided prescriptions.

Ah, those were the days.

The catch to making the argument against drugs is that you have to believe that some kind of lifestyle modification can be as effective. In this case, the physician arguing against believed this was true of “a realistic, supervised, intensive, supportive plan to moderate caloric intake and exercise regularly.” I would argue we know it’s true for carbohydrate restriction—thanks to the Virta Health research on type 2 diabetes. We do not know it’s true for eating less and exercise.

Now in August 2025, with clinical trials suggesting that the GLP-1 drugs do improve clinically meaningful health outcomes (for the first half a dozen years of use, at least), that talk of “seriously proposing indefinite medication” for a large proportion of the population has mostly dropped by the clinical wayside.

Now the NEJM has taken lifestyle modification off the table. If a patient spends 7 months trying to eat 1500 calories a day (with 30 minutes of dog walking), then the options are drugs or surgery. To be fair, the fictitious patient did not ask his doctor about a different diet or a “supervised, intensive, supportive plan”. He asked about drugs or surgery.⁴ So that’s the clinical decision to be made.

When I asked the NEJM editor about this limited choice, his response was that they know drugs and surgery work and that they induce greater weight loss more consistently than diet. They seem safe enough and the benefits are clear.

But the implication, nonetheless, is that if calorie-restriction failed this patient—trying to eat only 1500 calories day—we can assume all diets will. And we can assume all diets will, because we should believe that all diets induce weight loss merely by getting people to eat less.

But what if that’s not true, as I and others have been arguing for longer than the NEJM has been running its Clinical Decisions feature?

While we may not be right, the possibility that we might be is certainly now a mainstream proposition, per this Nature Metabolism review published last year? Meanwhile, the use of ketogenic diets for a host of chronic conditions, including diabetes and obesity, has become mainstream, even if the CW on those is still generally to worry about negative effects (LDL!?) or assume that they’re unsustainable, that its somehow easier to walk away from the table still hungry, for the rest of your life, than to avoid sugars, starches, grains and legumes.

But wouldn’t that be a clinical option worth discussing in a journal like the NEJM?

If the NEJM editors are going to avoid passing their (naive) biases onto their physician readers, they have to do what scientists are expected to do, which is entertain all reasonable hypotheses as viable working hypotheses, not just the one they prefer.

The 42-year-old patient in this latest Clinical Decisions feature clearly has the willpower to follow a restricted diet. If he can cut his smoking to three cigarettes a day, I’m assuming (as a former smoker), that he can do virtually anything if he sets his mind to it.

We have evidence now that ketogenic diets can be as effective as the GLP-1 drugs (as I discussed here), and we have that consensus of influential obesity researchers that there is a legitimate question about whether calories or carbohydrates are what need to be restricted to achieve and sustain a healthy weight. For the NEJM in 2025 to assume that obesity is all about calories is understandable from the perspective that the medical establishment still mostly does. But it still mostly will, so long as the leading journals don’t revise their perspective. Our job is to get them to do that.

That 80-year-old mouse experiment. It is all about calories, isn’t it?

One of the fascinating aspects of obesity medicine is that precious few, if any, of the doctors who think of themselves as knowledgable are aware of the history of their thinking. While physicists, for instance, learn as undergraduates why they should believe in relativity or the quantum physics (and even who was responsible for the work and the experiments that tested these concepts) this kind of education is lacking in medicine.

This absence is particularly problematic with obesity, because there were always, in effect, two ways to think about the accumulation of excess fat. Physicians come out of medical school aware of only one.

1. It’s caused by the imbalance between how much we eat and how much energy we expend: hence a disorder of energy balance.

2. It’s caused by what our bodies do with what we eat: hence a disorder of fuel-partitioning. People who store excess fat (who have obesity) do so because their bodies prioritize storing the calories consumed in fat tissue, and do so even to excess.

Through the early years of the 20th Century, physicians writing and thinking about obesity assumed the energy balance explanation because it resonated with their preconceptions (gluttony and sloth) and because energy intake and expenditure and excretion was all they could measure.

No experimental techniques existed to establish what our bodies do with what we eat—i.e., whether dietary fats and carbohydrates, specifically, are stored or used for fuel. What our bodies do with what we eat is captured in the term intermediary metabolism. This is everything that happens between consumption and expenditure or excretion.

In 1940, two Columbia University biochemists, Rudolf Schoenheimer and David Rittenberg, described the challenge this way:

The constituents of normal diets (proteins, fats, carbohydrates, etc.) are in principle the same as those which compose the animal body. The experimenter loses track of them as soon as they pass the intestinal wall and mix with the same substances in blood and organs. Very few experiments, therefore, on the intermediary metabolism of physiological compounds have been carried out with adult animals on normal diets, as in most instances the tracing of the food constituents was impossible.

By 1940, though, Schoenheimer and Rittenberg had already solved the problem, which is what that review article was discussing. Five years earlier, in 1935, they had published their first results paper on this research, reporting what they had learned about dietary fat metabolism and storage in the Journal of Biological Chemistry. That very first paper should have gone a long way to killing the idea that our body fat is regulated by how much we eat or even how much fat we eat.

It fundamentally changed how biochemists and physiologists thought about fat metabolism and storage, but it had essentially zero effect on the doctors who were thinking about what to do with their obese patients. This may be a problem with medical practice, more so than medical education. Physicians are busy enough reading the medical journals. The specialty scientific journals are left to the scientists.

Schoenheimer and Rittenber’s breakthrough was the use of deuterium—a heavy form of hydrogen—as a tracer to study intermediate metabolism. Schoenheimer was a German Jew who had been forced out of his academic position in Frieberg in 1933 and emigrated to the U.S., where he gone to work in Harold Urey's department at Columbia University.

Urey had just discovered deuterium, for which he would share the 1934 Nobel Prize in physics. Rittenberg was a student of Urey’s and had done his PhD thesis on deuterium. Working on a “brilliant idea”⁵ that Schoenheimer had come up with in Germany studying cholesterol, he and Rittenberg realized that they could replace the hydrogen atoms in carbohydrates and fats with deuterium, feed those modified macronutrients to mice and see where the deuterium ended up. They could trace its path, and so the path of the carbohydrates and fats, through the intermediary metabolism.

Question #1 in today’s quiz is about the results of their first experiments. The observation: The fat tissue in their mice took up and stored significant amounts of the fat they ate, even when they were being underfed and so losing weight.

This is how I summarized the results in my book Rethinking Diabetes:

Among the revelations that emerged from Schoenheimer and Rittenberg’s research was that the fat tissue is a site of remarkable physiological activity. Some fats consumed in the diet, they reported, are transported directly to the fat cells and stored as fat, and some aren’t. Moreover, the types of fats being stored are being continuously converted, one into another. Carbohydrates, too, are clearly being converted into fat, at least in the animals studied, because these animals would accumulate fat in their fat tissue; they would get fatter, even when fed only carbohydrates in the diet.

In 1948, another German-Jewish emigre, Ernst Wertheimer, and his graduate student Benyamin Shapiro, published what would become the seminal review paper in the science of fat metabolism. Wertheimer and Shapiro put it this way (with my emphasis):

Mobilization and deposition of fat go on continuously, without regard to the nutritional state of the animal.

And they added, making the point that decades of physicians would resolutely ignore:

The classical theory, that fat is deposited in the adipose tissue only when given in excess of the caloric requirement has been finally disproved…

As Wertheimer and Shapiro explained, fat accumulates in the adipose tissue when the forces of deposition exceed those of mobilization, and “the lowering of the fat content of the tissue during hunger is the result of mobilization exceeding deposition.”

By the mid-1960s, these physiologists and biochemists had demonstrated, unequivocally, that the forces of deposition are dominated by hormones and particularly insulin, while the forces of mobilization are dominated by other hormones—glucagon, for instance, growth hormone and adrenalin—and the central nervous system stimulating the fat cells directly.⁶

By that time, though, physicians had fully embraced the idea that energy imbalance (gluttony and sloth, in effect) caused fat storage and were ignoring almost entirely this physiological/biochemical perspective on fat storage.

This research simply evolved independent of the medical thinking on obesity itself, as though the actual science of fat storage and its therapeutic implications for the obese patient (i.e., the patient with excess fat storage) had nothing in common.⁷

Every time physicians assume that weight loss diets (or drugs) work merely by getting people to eat less (or tolerate eating less), they are perpetuating the same mistake.

1. The passive construction here is purposeful, echoing my favorite book title (ever) and an excellent analysis of cognitive dissonance: Mistakes Were Made (But Not By Me).

2. When the NEJM ran a review article last year on diets specifically, the authors merely ignored the possibility that different diets induce weight loss via different mechanisms.

3. I’ve argued that it still is, but mine is very much a minority opinion.

4. I’d bet in real life, he’d be asking only about drugs. Who wouldn’t with all the media attention?

5. Quoting the biochemist Erwin Chargaff.

6. This is how Wertheimer described it in a 1965 review (very much worth the read) of the history and science of fat metabolism:

“The deposition of fat by adipose tissue is considered today to be one of the most important aspects of energy metabolism. This process is regulated by and localized within the adipose tissue itself and not by the liver as was previously believed. This lipogenesis from glucose is regulated by the state of nutrition; it is decreased to a minimum in carbohydrate deficiency and accelerated considerably during carbohydrate availability. It is further regulated by insulin or, more precisely, by the ratio of insulin to symathetic-adrenal tonus.”

7. And, yes, this is not the first time I’ve made that point.