In Defense of Low Fat: A Call for Some Evolution of Thought (Part 1)

Not a typo.

Not April Fool’s Day.

Not a spontaneous and mystical possession by the spirit of George McGovern.

Not even a social experiment to see how many people I can get to unsubscribe from this blog in the span of a day (PLEASE STAY, I LOVE YOU).

Maybe a little bit of this, though:


Oh, Homer!

Over a year ago, I gave a presentation at the Ancestral Health Symposium called “Lessons From the Vegans: What the Paleo Movement Can Learn From the Success of Plant-Based Diets.” In retrospect, I probably should’ve called it “Lessons from the Low-Fatters: What the Paleo Movement Can Learn from the Success of People Who Eat Ridiculous Amounts of Carbs and Don’t Keel Over,” but that was too long for the conference brochure. And for my verbally dyslexic mouth. And also, I didn’t know it was really going to be about fat until I fell down an extended PubMed rabbit hole and, upon regaining consciousness two days later, realized I had found the Nerd Project to end all Nerd Projects.

In truth, though, this post started brewing long before my talk. Having witnessed some pretty impressive healing when I noshed among the low-fat raw vegans (and, after a decade of self-experimenting, concluding I do best on a lower fat diet myself), I just can’t get on board with the categorical “Fat rules, carbs drool!” trend infiltrating both mainstream and alternative nutrition. There are too many exceptions to the rule, too many gaps in the theory, and too many skinny fruitarians frolicking in the sun-dappled fig orchards.

But even beyond that, this post is born of a belief I hold dearly—one that guides my approach to research and underlies the very mission of this blog:

We can’t ignore evidence in order to preserve an ideology.

At least not under the guise of “science.”

When confronted with something that challenges our belief system, the worst possible thing we can do is clamp our hands over our eyes and say, “You do not fit into my understanding of reality; therefore, you do not exist. BE GONE, NON-EXISTING ANOMALY.” Yet that’s what so many of us do—often without even realizing it—when faced with outcomes our chosen philosophy can’t explain. On the flip side of cherry picking, we cherry-throw-out: selectively deleting data that threatens our version of the truth, nipping any cognitive dissonance in the bud before it has a chance to rattle our worldview. It’s easy to be “right” when we’ve shoved all competing evidence into the wood chipper!


For a long time in the nutrition world, our thrown-out cherries were the ones challenging the low fat ideology. We discarded the high fat Inuit cherries and the milky, bloody Masai cherries; the coconut-filled cherries of the Tokelau; the cherries of the traditional reindeer-herding Sami; even the smothered-in-butter French cherries—just to name a few.* It didn’t make sense that these populations could exist and be healthy with their fat-gorging ways, so we slapped them with a “paradox” sticker and deemed them weird exceptions to the Dietary Laws that govern the rest of us.

* For the record, all these examples come with some major caveats, and I don’t think they should be used as evidence to support the kind of high-fat diets many people are eating today (though they don’t necessarily stand as counter-evidence either). More on that in an upcoming post!

Only in more recent years have those cherries been rescued from the compost bin and plopped back into the world’s collective fruit bowl (please wash before consumption). Bestselling books like “Good Calories, Bad Calories” and “The Big Fat Surprise” carved new histories in which fat was an innocent bystander, dragged into the mud by bad science and even badder scientists. The phrase “healthy fat” moved from oxymoron status to popular catchphrase. People whir 80 grams of butter into their coffee and call it breakfast. Apparently Bob Dylan had it right in all but plurality: Time, it is a-changin’!


As awesome as the pro-fat movement has been for challenging outdated beliefs and reviving some truly nutritious foods, there’s been a dark side to the process as well. All of a sudden, the same rhetoric once leveled against high-fat diets is being slung against low-fat ones. Not only is low fat (and by consequence, high carb) not the dietary angel we once thought, the new story goes, but it’s actually the source of all edible evil: the driving force behind our obesity epidemic, a major contributor to heart disease, the puppet master pulling those blood-sugary strings of insulin resistance and diabetes. If only the USDA had recommended 6 to 11 servings of bacon instead of 6 to 11 servings of grains, we wouldn’t be in this mess!

See the problem here?

In the process of redeeming fat, we traded one form of oversimplified blame for another. And it’s led to a brand new wave of cherry genocide. We now dismiss (or paradox-sticker) high-carb populations in the same way we justified ignoring the high-fat ones. We snub decades of clinical success involving fat reduction (to the point where you might think such evidence doesn’t even exist—in which case, you’re in for a surprise with this post!). We deny the potential for low-fat diets to be anything other than a metabolic train wreck, ending in a smoking heap of shrapnel and insulin injections. “Surely those low-fatters are starving all the time,” we proclaim. “Surely they’re making themselves diabetic! They might feel okay right now, but won’t those carby diets go all Cujo on them as the years progress, eating their souls and whatnot?”


Let me be frank here.

If we’re really after the truth, we can’t keep throwing away perfectly good cherries. Seriously. It’s gotta stop. When we censor data we don’t like instead of revising our theories accordingly, we perpetuate the same problems we’ve been battling for decades: partial truths treated as gospel; public policies that do more harm than good; baffled consumers who can’t figure out if it’s the omelet that’s killing them or the OJ they wash it down with.

Do we really want to keep heading down that road? It probably goes somewhere awful! Like Stockton. (Sorry, Stockton.)

Hence why we’re gathered here today, around this massive compilation of pixels, delving into a decidedly hot topic. This post is my attempt to rescue some discarded cherries and return them to the Fruit Bowl of Our Lives. Which, if nothing else, will one day make a fantastic soap opera.

I do want to make one thing abundantly clear before we continue, though. The title “In Defense of Low Fat” doesn’t imply its inverse, “In Attack of High Fat.” Quite the opposite! My goal here is to create a space where two very different dietary approaches can sit down for tea, respectfully coexist, and interact without any subsequent homicide investigations. In fact, I’ll be arguing for a more panoramic view of nutrition where the success of both high-fat and low-fat diets are compatible, and maybe even make sense. It just requires zooming out farther than we’re used to looking, and acknowledging that our ever-rivaling communities could actually learn a lot from each other.

For the sake of reading ease, this sucker is divvied up into two parts: this one, which discusses the crazy-huge body of research behind truly low-fat diets (especially the really obscure stuff!), and the upcoming Part 2, which ties everything together with science, and whatnot. And because this post is long even by my standards, I’ve created a clickable Table of Contents to help you navigate the labyrinth. Good luck! (You’ll need it…)


1. Carbosis: The Magic of Truly Low-Fat Diets

2. The Low-Fat History You Probably Haven’t Heard

4. Modern Diet Doctor Squad: An update and apology for jumping the gun

5. Up Next…

Note: a lot of the papers discussed in this post are trapped behind paywalls. I have copies of the full text for all of the ones I discuss in depth, though, so shoot me an email if you’d like to read any!

1. Carbosis: The Magic of Truly Low Fat Diets

Let’s cut to the chase. My thesis, if all this can be boiled down into one, is that two unique metabolic states exist on either extreme of the fat-intake spectrum. One ends around 10% fat and the other begins around 65%. Both zones have their own benefits, and can elicit surprisingly similar effects.

This is a PowerPoint slide I made representing what seems to be going on.

Image rejected from the Lancet due to insufficient sparkles.

In sum:

  1. Something special happens at very high levels of fat intake (and very low levels of carbohydrate intake). That thing’s called ketosis. It’s where your body creates ketones to use when glucose is scarce, and where fat metabolism is optimized.
  2. Something equally special happens at very low levels of fat intake (and very high levels of carbohydrate intake). I’m not aware of a formal name for this, so I’m dubbing it carbosis until further notice. It’s a state where insulin sensitivity dramatically improves, and where carbohydrate metabolism is optimized.
  3. Between those two extremes, especially towards the very middle, is what I’m calling the Macronutrient Swampland. It’s not necessarily a bad place to be if you’re eating a high quality, non-energy-surplus diet and are at a healthy weight (or getting there!), but it’s hard to see the therapeutic effects of reducing fat intake while you’re in this zone. And it tends to be the most potent area for food reward, making it easy for people to overeat here. Any health improvements seen in the Swampland will typically be from losing weight, eating more protein (which has its own special metabolic effects), or boosting food quality (i.e., switching to less processed, “low reward,” nutrient-denser fare), rather than from moving laterally across the macronutrient spectrum.

Here’s the kicker: the Macronutrient Swampland is where our standard definition of “low fat” squarely lands—30% of calories. When we conduct those “low fat” studies with sucky results, they’re almost always using a fat intake of 30% of calories. When the USDA tells us to eat low fat, they mean 30% of calories. When the American Heart Association tells us to eat low fat, they mean 30% of calories. This number has been parroted far and wide across the Western world, drilled into our noggins, and slapped with a skull and crossbones in the “LOW FAT SCREWED UP AMERICA” narrative.

The only problem? It’s not actually low fat.

It’s not low fat relative to the many populations that eat (or ate) their traditional starchy diets: the Okinawans (12% of calories as fat), the Tarahumara Indians (12% of calories as fat), the pre-industrialized Thai (8.9% of calories as fat), the traditional Hawaiian (10% of calories as fat), the traditional Taiwanese (16% of calories as fat), the African Bantu (14 – 17% of calories fat), the traditional Pima (8 – 12% of calories as fat), and the highlanders of Papua New Guinea (3% of calories as fat), just to name a few. It’s not low fat relative to the carby diets that really do have clinical track records for treating modern diseases (which, as we’ll see in this post, hover almost universally at that 10% mark). And perhaps most importantly, it’s not low fat relative to what Americans already eat—which is about 34% of our calories these days, per the most recent available data.

That last point is what utterly handicaps our modern “low fat” research. Nudging 34% fat down to 30% in studies and then claiming nope didn’t work is absolutely face-palm worthy. It’d be like dismissing low-carbohydrate diets because switching from 50% carbohydrate to 46% doesn’t have a profound clinical affect. We’d unleash the hounds on a logical blunder like that, right?

This is a problem that tends to get the plant-based diet community a bit hot under the collar, and I have to agree with them. One of our biggest scientific bloopers was choosing a Swamplandy 30% of total calories as the benchmark for “low fat.” It’s resulted in a slew of unimpressive studies that make us think low-fatting is categorically worthless. I’ll probably never live this down, but I mostly agree with an article T. Colin Campbell wrote about the “low fat mythology” addressing this very topic. (Though in case you’re wondering, I stand by my criticisms of the China Study and Campbell’s rat studies. Nothing in this post supports the idea that animal protein is uniquely harmful, and acknowledging when some parts of the plant-based movement are legit doesn’t give a free pass to the ones that aren’t!)

And that’s just the tip of the iceberg lettuce. On top of our wonky definition of where “low fat” starts and ends, we’ve erroneously conflated “low fat” with the corn-syrup-injected, processed-up-the-wazoo, won’t-rot-for-200-years-because-woah-preservatives menu that emerged when the food industry found a new market to tap. As soon as the USDA released the Food Guide Pyramid in 1992, food manufacturers were all “Let’s low-fat ALL OF THE THINGS,” and accomplished that very feat. So now we hear “low fat” and remember the era of Fig Newtons and rice cakes and sadness, so much sadness. And also Susan Powter.

Once upon a time, though, low fat meant something different. Something hopeful. Something positive. Something that didn’t taste like regurgitated cardboard turd pellets. And the research that emerged from this era was a thing of great beauty! There were low-fat diet studies that actually studied low fat diets (imagine that!), and the notion that 30% fat was equivalent to 10% fat was just a twinkle in a future USDA employee’s eye. Even though our understanding of nutrition’s nitty gritty was less sophisticated back then, research designs were often better, testing “high fat” with things like fresh cream instead of the confounder-riddled carrot cake and milkshakes used in more recent trials.

Hence, much of this post will be a blast to the pre-low-fat-craze past—a romp through the research that artfully dodged the Macronutrient Swampland, and an introduction to the thought-leaders that have all but vanished from scientific memory. After that, in the next post, we’ll explore the mechanisms that tie it all together.


The Low-Fat History You Probably Never Heard

The popular version of events goes something like this:

America was happily eating its buttery, meaty, cholesteroley fare until Ancel Keys came along with the idea that fat causes heart disease. He cherry-picked data to make it seem like his theory was true, narrowed the culprit down to saturated fat, manipulated the Powers That Be into believing him, and then bull-horned the message far and wide until we all bought his myth. Low fat started with him!

Unfortunately, that’s a big, pre-chewed wad of baloney.

Here’s the deal. People were already whipping out low-fat diets to treat diabetes, multiple sclerosis, high blood pressure, kidney failure, heart disease, and obesity when Keys was a mere young’un shoveling bat poop in Arizona. Heck, the ancient Egyptians prescribed a near-fat-free diet of wheat, grapes, honey, and berries for what was almost certainly diabetes (“too great emptying of urine”). And contrary to popular belief, the early-1900s evidence supporting low fat didn’t come exclusively from rabbits and test tube experiments and correlative scatterplots: it came from actual human people eating actual food with their actual mouths.

If anything, our pal Keys was a latecomer to the idea that fat could play a role in chronic disease. The notion that he single-handedly criminalized fat is complete and utter fiction. He might’ve been the loudest and most unflappably confident voice in the choir, but he certainly wasn’t singing solo.

Here’s a more realistic timeline of the past century, albeit still very incomplete. (Click to make ‘er big!)


Without further ado, let me introduce you to some of these disembodied heads.

Walter Kempner: Rice and sugar and diabetes reversal, oh my


Here’s one for the Paradox Files.

In the 1930s, a man by the name of Walter Kempner fled an increasingly Jew-hostile Germany and landed square in the halls of Duke University… where he proceeded to totally blow the medical community’s mind. His mission: treat kidney disease. His solution: put renal-failing folks on a special diet low in sodium, protein, and fat—a menu devised from in vitro experiments he’d done on kidney tissue.

At the time, very few researchers believed that food could have any effect on kidney disease. Or high blood pressure. Or diabetes. Or heart disease. Or most other chronically wrong-going things in the body. As with Ancel Keys, who was pretty much laughed out of the WHO conference where he presented his “fat causes heart disease” idea, Kempner spent the first chunk of his career swimming upstream in a river of skepticism.

But his colleagues’ dubiousness didn’t last long. After placing patient after so-called-hopeless patient on his unique regimen, it became clear that Kempner’s diet worked. Really ridiculously well. And it became equally clear that the kidney wasn’t the only body part made happy by the new cuisine. Obesity, diabetes, high blood pressure, heart failure, coronary artery disease, psoriasis, and arthritis often saw major improvement or total reversal as a result of the diet. During the course of his career, Kempner treated over 18,000 patients with the above conditions—all by changing what went on the stabby end of their forks.

So what was in this mystical diet of his? Brace yourself!

  • White rice
  • Fruit
  • Fruit juice
  • Refined table sugar
  • In some cases, vitamin supplements (A, D, thiamine, riboflavin, and niacin)

…And not a darned thing else. Kempner summed up the details himself in a 1974 article, readable here:

A patient takes an average of 250 to 350 gm. of rice (dry weight) daily; any kind of rice may be used provided no sodium, chloride, milk, etc. has been added during its processing. … All fruit juices and fruits are allowed, with the exception of nuts, dates, avocados and any dried or canned fruit or fruit derivatives to which substances other than white sugar have been added. Not more than one banana a day should be taken. White sugar and dextrose may be used ad libitum; on an average a patient takes about 100 grams daily, but, if necessary, as much as 500 grams daily should be used. Tomato and vegetable juices are not allowed.

In other words, it was the CARBPOCALYPSE. Along with feasting on impressive amounts of white rice, people were averaging 100 grams of pure sugar a day, and some ate over a pound of it. That’s up to 2,000 calories from refined sugar alone—the same amount deliciously packed into 25 Cadbury Creme Eggs.

(Wisely, Kempner knew his diet was at no risk of being crowned Dietary Homecoming Queen. He apparently described it as a “monotonous and tasteless diet which would never become popular,” and whose only saving grace was the fact that it worked. And as I mentioned in my AHS presentation, he apparently whipped some of his patients in order to help them comply, as—in his words—”the risk to their life was so great that it warranted harshness.” Ouch!)

Here’s a breakdown of how the diet panned out, macronutrient-wise. Image from Duke University files; red graffiti my own doing, to indicate percent of total calories:


What’s really noteworthy is that the diet wasn’t automatically calorie restricted. In fact, some patients had to increase their energy intake to help them gain weight, or to stabilize their weight if they were losing too much. That’s important, because it means we can’t write this off as a diet that improved biomarkers solely by inducing weight loss (Twinkie Diet, I bow in your general direction). It also means that many people spontaneously ate less than they needed when stuffing their faces with unlimited amounts of starch and sugar… as long as fat intake was super low.

If this seems totally baffling and Twilight-Zoney, that’s because it is. According to my calculations, there is an 84% chance that you are now Googling “rice diet Snopes” or contacting my mother to inquire about my recent psychotic break (joke’s on you; she thinks I’m great!). I urge you to keep reading, though, because we’re about to get to the ooey, gooey data at the center of this carb-filled Tootsie Pop.

Let’s start with something weighty: an obesity paper published in 1975 in the Archives of Internal Medicine, which should be of particular interest to anyone convinced refined carbs are inherently fattening:

Here, Kempner compiled data from 106 slimmed-down patients—a mere slice of the thousands he treated over the years—who all dropped at least 100 pounds on his program and achieved a normal weight. (The average loss amongst them was 140 pounds, and one man melted away over 300). These particular losers ranged from 16 to 65 years old, and featured a mix of women and men.

When it came to blasting obesity, Kempner employed what he called a “rice-reduction diet”—the same protocol he’d designed for renal failure and hypertension, but with lower calories:

In the unmodified initial diet, 90% to 95% of the caloric intake is carbohydrate, taken as rice and fruit. As in the original rice diet, salt intake is exceedingly low (less than 60 mg of sodium per day) and fluid intake is thus markedly reduced to prevent water intoxication. Thus, the initial diet is low-calorie, low-salt, low-protein, low-fat, and essentially free of cholesterol.

After getting into that sugary, starchy groove for a month, the dieters could start eating veggies again (which were initially nixed due to their sodium content—kept low, in part, to help tame high blood pressure and “reduce the stimulatory effect of salt on food intake”). A bit later on, lean meats could also make a triumphant gustatory return.

Thanks to Kempner’s hawkish monitoring and dietary tweaking, obedient dieters were greeted with a steady (and often plateau-free) slide towards a healthy weight, like so:


And since we’re all such visual creatures, here are some photos demonstrating the rice diet in action, first published in Kempner’s aforementioned obesity paper. This young woman lost 123 pounds in just shy of a year:



And here we have 278 pounds obliterated in a bit over a year, doing the same. His fasting triglycerides dropped from 187 mg/dL to 85 mg/dL:


And this lovely lady lost 115 pounds in 33 weeks. Her fasting blood sugar dropped from 315 mg/dL to 100 mg/dL, and her triglycerides plummeted from 516 mg/dL to a peachy keen 79 mg/dL—after eating a diet literally made of refined sugar and starch.


Looking at these sugar-fueled Incredible Shrinking People brings an important point to mind. If we assume weight loss is just a matter of calories in versus calories out, the rice diet isn’t any more remarkable than other low-calorie bootcamps: folks ate less and lost weight. Dur. But if you’ve hung around the internet for very long, you might’ve seen the theory—popular within some corners of the low-carb world—that successful weight-loss diets are invariably low carbohydrate diets (or at least low refined sugar and starch diets), regardless of what other rationale those diets masquerade under (low fat, high fiber, low calorie, food combining, eating only on Tuesdays in the presence of an ovulating jackalope, etc.).

How? Because folks inevitably slash their carb intake when they eat less food overall, the theory goes. This rests on the premise that insulin is the wizard behind the curtain of obesity, and that quelling its wrathful swings—triggered by carbohydrates and refined carbohydrates in particular—is necessary for losing weight. Gary Taubes explained this concept in detail in a 2010 blog post:

Simply put, anyone who tries to diet by any of the more accepted methods (i.e., Weight Watchers), and anyone who decides to “eat healthy” as its currently defined, will remove the carbohydrates from the diet that may be — if the carbohydrate/insulin hypothesis is correct — the most fattening. And if they’re trying to cut calories, they’ll be removing some number of total carbohydrates as well. And if these people lose fat on these diets, this is a very likely reason why.

The rice diet might be the most compelling hole-poker we have for that theory. While most carby programs—say, Pritikin or McDougall or Ornish—eschew refined grains and sugar (and thus could fit snugly into the insulin-centric concept above), Kempner’s program sure didn’t. He fed folks almost nothing but the “most fattening” carbohydrates and still managed to slim them down. Does your brain hurt yet?

Of course, losing weight is a far different beast than maintaining a 100-pounds-lighter frame after the losing’s been done—leading us to the question: what happened to these folks in the long run? Did they maintain their weight loss? Gain it all back? Develop a crippling phobia of small, white, oblong granules, requiring years of psychotherapy and Riki Lake guest appearances to overcome?

Alas, Kempner noted that long-term results were “not yet available for the patients analyzed in this report,” and I’ve yet to find any follow-up papers revealing their fate. The only clue I’ve seen comes from the first page of the “Rice Diet Renewal” book, which states that 43% of rice dieters had maintained their weight loss (or lost even more) six years after their stint in the program. (For comparison’s sake, an Annual Review of Nutrition paper estimates that on average, about 20% of folks who’ve lost significant weight are able to maintain that loss for at least a year.)

All that said, the rice diet was about far more than impressively svelte before-and-after shots. As alluded to earlier, it also had the uncanny side effect of improving diabetes and insulin resistance—even when weight loss wasn’t part of the equation. I warned you this was gonna get weird! Kempner published a whole paper on the topic in 1958, which you wouldn’t know by looking at its hauntingly empty PubMed entry:

(Email me if you want the full text!)

For starters, Kempner was just as perplexed as us modern-day health enthusiasts might be when it comes to the effect his diet had on diabetics. As he penned in the paper you cannot see:

We have for the past 15 years treated numerous diabetic patients with the rice diet. Since more than 90 percent of the calories in this diet are derived from carbohydrates, it was anticipated that increased amounts of insulin would be necessary to keep the blood sugar at its previous level. However, the opposite proved to be true. … Not only is the rice diet well tolerated but in many instances the blood sugar and the insulin requirements decrease.

In this report, Kempner analyzed 100 diabetics who’d entered the rice diet program between 1944 and 1955. All of them strictly followed the diet for at least three months (often much longer), and they were observed an average of nearly two years—with some folks monitored for up to eleven years after they’d first embarked on the carby cuisine.

The findings? Ladies and gents, place your bets…

More than half of those 100 diabetic ricers—63%—actually saw their fasting blood sugar drop by at least 20 mg/dL during the diet. Only 15% had their blood sugar go up significantly. The remaining 22 saw little to no change.

To get a visual sense of those numbers, here’s an aptly named pie graph (don’t worry; it’s fat and carb free!). “Increased” or “decreased” is defined as a change of at least 20 mg/dL:


Let’s repeat that: eating almost nothing but starch and sugar and fruit, the majority of diabetic patients lowered their blood sugar levels. In fact, when everyone’s results were pooled together, the average blood sugar change was a drop of 47 points.

‘Twas a similar story in Insulin Land. Of the study’s participants, 68 entered the scene already dependent on insulin. As the carbs raged on, 21 of those insulin-injecters didn’t have to change their dosage; nine needed an increase (including four people who initially weren’t on any insulin at all); and—again comes the cruel, cruel defiance of prediction—42 slashed their usage significantly. In fact, 18 folks were able to discontinue their insulin entirely. Feasting on white rice. And sugar. And fruit juice.

Here’s another delicious graph of pie, calculated for the 72 patients who needed insulin at some point during the study:insulin_usage_rice_diet

Once again: eating virtually nothing but this…


…the majority of diabetics ended up with better glucose control and insulin sensitivity, and in some cases freed themselves from diabetes entirely.

Let that sink in for a minute.

Or two, if you need to grab a glass of water and ward off the vapours.

Just to be clear, the point here isn’t that the rice diet is the Best Thing Ever for diabetics and everyone should trade their insulin pumps for a metric ton of Skittles (nobody needs to taste that many rainbows). After all, 15% of the rice-dieting diabetics actually got worse, many of the improvers still had above-normal blood sugar (despite huge drops from baseline), and we could probably hack Kempner’s protocol to make it more nutritionally sound without ruining its therapeutic effects. Clearly, it ain’t perfect. All I’m saying is that these results totally fly in the face of what most of us consider possible. Sugar and white rice improving diabetes? Blasphemy!

All that said, an important critical-thinky question remains: was this all just a byproduct of weight loss? We know that restricting calories and dropping pounds can definitely boost insulin sensitivity and glucose control, regardless of whether the diet used is particularly healthy. It’s one thing to not be diabetic because you’re eating kale and grass-fed buffalo whose ancestors were blessed by Sacagawea, and another to not be diabetic because you’re living on napkins and crack. Can we at least say that the successful diabetics were the ones who lost weight throughout the program, spontaneously eating less, unamused by a diet that had exactly one-and-a-half flavors?

NOPE. As Kempner pointed out, any obese patients were indeed encouraged to lose weight—but the improvements in blood sugar levels and insulin requirements occurred “both in patients who lost weight and in those who did not have a significant weight change” (his words). Kempner’s data, both in this paper and in the massive collection of his work filed away at Duke University, showed that the diet could benefit diabetics even when their weight and energy intake didn’t budge. Even supplements such as Kratom (you can always buy kratom from here) showed no increased benefits in diabetics.

And it didn’t end there. The rice diet also proved helpful for heart failure. It rapidly healed psoriasis. It excelled at its original goal of treating high blood pressure. The “good for” list stretched on nearly as far as those endless bowls of rice! As early as 1949, Kempner had observed that the rice diet was healing more than 70% of his seriously ill, not-responding-to-other-treatments patients from a wide spectrum of disease backgrounds. That figure stayed pretty stable as the decades rolled on.


Psoriasis obliteration on the rice diet.

Just last year, the Journal of Electrocardiology published something of a Rice Diet Resurrection, dredging up Kempner’s key findings and blasting open his oft-forgotten legacy: “An archaeologic dig: A rice-fruit diet reverses ECG changes in hypertension.” In it, the authors pointed out something pretty important regarding the ultimate success of the diet. A band-aid treatment it was not; the rice diet actually seemed to permanently reverse the conditions it set out to treat, at least for many adherents:

A poorly known but important observation was that patients who were able to follow the regime, and who were slowly guided through a gradual modification of the diet over many months, were able to transition into a very tolerable low fat, largely vegetarian diet, while leading a normal, active life, without medications, indicating that the disease state had been permanently modified.

“Permanently modified” probably needs a qualifier, since those folks couldn’t make a total return to their former gustatory ways. But over time, they could start eating a more diverse diet with (lean) animal foods, all manner of vegetables, a moderate level of salt, and the magnificent return of tastiness. Not too shabby, considering many of those folks were initially riding a bullet train towards death. (In another very recent article, “Who and what drove Walter Kempner?“, the authors noted that in Kempner’s day, life expectancy for anyone with malignant hypertension—one of Kempner’s main patient demographics—was only six months. The fact that he gave most of them decades of recouped earth time was pretty fantastic.)

So whatever became of the rice diet? Like most things in life, it lost out to stuff that was newer, prettier, shinier, and easier to squeeze inside an FDA-approved pill. Kempner relinquished the Rice Diet throne in 1992 (and in case you’re wondering, died of a heart attack five years later, at the age of 94—though it’s unclear what his own diet and lifestyle actually were). After his departure, the rice diet predictably loosened up: the program later allowed “a wider selection of largely vegetarian food choices,” though still with low sodium and protein intake (and ostensibly less whipping).

In his 1983 article “Kempner Revisited,” Eugene Stead—who’d worked at Duke alongside the Rice Man himself—summed up Kempner’s unorthodox legacy in a way that captures my own thoughts:

Who in his right mind would have ever thought that rice and fruit could modify vascular disease appreciably? Who would have fed a protein-deficient patient, losing large quantities of protein in his urine, a protein-poor diet? Who would have dared to give a more than 90% carbohydrate diet to a diabetic? Every expert knew that cholesterol levels were not influenced by diet. Nevertheless, all these leads have paid off richly.


Kempner walkin’ the talk. From Duke University files.

Ultimately, I find Kempner’s work both important and wildly uncomfortable.

Important, because it exposes some cracks in our current view of carbohydrates and sugar—areas where our thinking has room to grow and our assumptions have room to crumble.

Uncomfortable, because it demolishes my long-held rationale for why low fat, veganish diets can be successful: that they work because they involve switching to whole foods (Kempner’s sure didn’t!); that they work because they reduce refined sugar and starch intake (the very lifeblood of the rice diet!); that they work because they increase disease-fighting plant compounds (nary a vegetable to be seen under Kempner’s strictest watch). Let me go on record here: I TOTALLY stand corrected!

Perhaps the only areas of overlap with an ancestral framework are that all that rice would’ve provided a decent source of resistant starch, gluten was nowhere to be seen, and the uber-lean diet would’ve smashed polyunsaturated fat (PUFA) intake to smithereens. (If you haven’t caught wind of the PUFA-hatin’ yet, these fats are garnering quite a bad rap due to their unstable, oxidation-prone structure—especially omega-6 PUFAs, the pro-inflammatory Evil Cousins of omega-3s. But even PUFAs as a whole have taken a clobbering in some spheres (hello Ray Peat!), and higher-than-trivial intakes have been indicted as a cause of many terrible things.)

Indeed, after my AHS talk, a few people contacted me suggesting it may be the PUFA reduction that improved sugar metabolism and other aspects of health, thus allowing rice dieters to thrive on an otherwise nonsensical diet. And more broadly, that rock-bottom PUFA intake might be the biggest reason low fat, plant-based diets have any positive effect on chronic diseases to begin with. That would leave saturated and monounsaturated fat in the clear, needlessly nixed in the quest for better health.

I might be inclined to guess the same thing, if not for one human-sized monkey wrench by the name of Roy Swank.

Roy Swank: Kickin’ some multiple sclerosis butt… by nixing saturated fat


You might be familiar with our current diet-wielding, multiple-sclerosis-blasting warrior Terry Wahls, but less known is the fellow who preceded her by over half a century: Roy Swank. And what a story he had!

Swank exited the womb practically destined for greatness. Along with growing up near my beloved home city of Portland, young Swank was musically gifted, athletically inclined, and devastatingly suave. He spent his formative years wooing his future wife, Eulalia, in his dad’s mortuary hearse (chicks dig that stuff), and started driving a local doctor around town to see patients when he was just 13. It was then that his interest in medicine roared to life! By the time he was 26, Swank had racked up a Bachelor of Science, a PhD, a medical degree, and a fitting destiny for his surname.

But our relevant saga began in 1948, when Swank got invited to Montreal to research multiple sclerosis (MS)—a devastating autoimmune disease affecting the brain and spinal chord. He soon discovered a peculiar trend: across the globe, MS was rare in areas where saturated fat intake was low, but much more common where meat and dairy were a mainstay. And as he scoured disease patterns from the previous two decades, Swank noticed that World War II heralded a drop in MS rates wherever meat and dairy were rationed. The clincher came with a Norwegian survey he helped conduct, which showed MS was clustering around the country’s rural, mountainous, dairy-noshing farming regions, while coastal fishing communities—whose fat intake was lower and mostly polyunsaturated—were pretty much spared.

And you might get a kick out of this: when plotting MS prevalence against national intake of animal fat, Swank arrived at a perfect upward curve that was nearly indistinguishable from the one we vilified Ancel Keys for. Here’s Keys’ six-country graph overlaid with Swank’s MS data. Eerie, no?

From page 13 of

From page 13 of “The Multiple Sclerosis Diet Book” by Roy Swank.

We can (and should!) fly our “correlation isn’t causation” flag here, but keep in mind that MS research was in its infancy back then. Population trends were the only thing researchers had to work with. So, armed with the scant clues of his time, Swank put his thinking cap on and devised the first-ever dietary experiment for multiple sclerosis. And it went a little somethin’ like this:

  • Low total fat
  • Very low saturated fat (10 – 15 grams per day, maximum)
  • Small amounts of polyunsaturated vegetable oils and fish oil (10 – 40 grams per day)
  • Grains
  • Vegetables
  • Fruit
  • Skim milk
  • Fish and other seafood
  • Extremely lean land-animal products (skinless chicken and turkey; egg whites; trimmed meats)
  • 60 to 90 grams of protein daily, mostly from fat-free animal foods

To put his hopeful menu to the test, Swank rounded up 150 MS patients—70 men and 80 women—to guinea-pig his diet in real life. And we’re not just talking a breezy month or two of experimenting; starting in 1948, he meticulously followed these folks for 50 years and beyond, keeping track of what each person ate and how much their disease progressed after they entered the study. One of his most detailed papers, “Multiple sclerosis: fat-oil relationship,” documented the 34-year results of that massive undertaking.

As Swank explained in his paper, the patients had to keep detailed food logs of their daily gustatory adventures. They also trekked to a Montreal clinic once every two weeks to get their MS assessed and endure a pop quiz about their eating habits. (Due to humans being sucky remember-ers (and notorious “historical revisionists” of our own mistakes), dietary recalls are often unreliable—but Swank collected so many of them, over such a vast span of years, that he could paint a pretty accurate picture for each patient.) Once their new, Swank-approved eating habits had stabilized, those patients were weaned onto less frequent clinic checkups and did much of their reporting by snail mail.

To gauge how each person’s disease changed throughout the study, Swank used a neurological grading system ranging from 0 to 6, abridged here for reading ease:

0. Remission, more or less

1. Some neurological symptoms, with fatigue and periodic exhaustion

2. Mild physical impairments

3. Severe physical impairments

4. Wheelchair-bound, with memory impairment

5. Confined to bed and chair

6. Dead

From the get-go, the goal was to slash participants’ animal fat intake down from an average of 125 grams per day (pre-Swankification) to a maximum of 10 to 15 grams per day (post-Swankification). Starting in 1951, they could also add 5 grams of cod liver oil and eat between 10 and 40 grams of “fluid vegetable oils” daily, though saturated fat was supposed to stay as low as possible.

Of course, we humans are wont to err, and not all of the patients colored within the low-fat lines. Swank noted that some people “doubled or even tripled the amount of [saturated] fat* recommended,” and that in lieu of a control group, those variations made it possible to see how different fat intakes correlated with MS progression over the decades. Yay math!

* A Swanky disclaimer: although I’m employing more familiar terminology for the sake of this blog post, note that Swank uses the words “fat” and “oil” a bit differently in his papers. “Fat” in his work refers only to saturated fat, while “oil” refers only to unsaturated fats. So if you take a gander at any of his publications (which I recommend you do!), just keep in mind that his “fat” translates to our definition of “saturated fat” rather than “total fat.”

By the 34-year mark, a bit over half of the original group—81 people—had passed away. That’s actually pretty promising: Swank noted that when he first started studying MS, most patients were expected to end up bedridden or wheelchair-bound within a decade or two, and the assumption was that “all would be dead within 35 years.”

But where it gets really interesting is when the Swankers were divided up based on their average saturated fat intake. Of those who didn’t exceed 20 grams per day (70 people), only 31% died over the course of the study (20% specifically from MS). But of those who ate more than 20 grams of saturated fat per day (74 people), a whoppin’ 80% perished in that same time frame (62% specifically from MS). Voila:swank_diet_mortality_ms_non_msAnd when we break it down a bit further, something even more interesting emerges: the saturated fat/mortality trend wasn’t linear. Like, at all. Folks eating between 20.1 and 30 grams per day had ridiculously similar mortality rates to those eating 30.1 to 50 grams and beyond. Basically, once people crossed the 20-grams-per-day threshold, it didn’t seem to matter how much saturated fat they were eating: mortality rates rapidly maxed out and then stayed relatively constant. It’s like a saturated-fatty version of the Macronutrient Swampland, where everything outside the “magic” zone is super samey!


FYI: average saturated fat intake for the lowest bracket was 17 grams per day; for the middle bracket, 25 grams per day; and for the highest bracket, 42 grams per day.

(In case you’re wondering, it didn’t seem to be a matter of having a longer but more miserable, enfeebled life for the low-fat adherents. As Swank noted, “Patients on our low-fat diet have been remarkably free of bacterial and viral infections, ‘colds and flu’ occur rarely, and recovering from urinary and other bacterial infections has been rapid when appropriately treated.” Indeed, quite a few measures of their health improved.)

The Swankers’ disease progression—as tracked by that neurological scale mentioned earlier—followed a similar trend. For the folks eating 20 or fewer grams of saturated fat per day, the disease remained relatively stable, with just over half a grade of progression (worsening) on average. But for the subset of folks eating between 20.1 and 30 grams per day, the disease progressed an average of 2.6 grades. And for those chowing down on more than 30 grams per day, the disease worsened by an average of 3.0 grades. (To help conceptualize that, a three-grade change is the equivalent of going from “mild physical impairments” to “confined to bed and chair.”)

Here are more Smurf-colored bars for the visually inclined:


And because I love me some graphs, here’s another one straight from Swank’s paper, showing the dramatic drop in exacerbation (flare-up) rates before and after folks got all Swanky:


Pretty impressive, no?

In addition to all this fun stuff, Swank commented (again, in his 1991 paper “Multiple sclerosis: fat-oil relationship“), that some patients needed an even steeper slash in their saturated fat intake—down to 10 grams per day, maximum—in order to “gain an improvement in energy and freedom from severe fluctuations of disease.” Swank wrote:

In recent years, this experience induced us to eliminate all meats and other sources of fat from the diet of many patients when first seen. We estimate that saturated animal fat was reduced to ~5 g/day plus the fractional or trace amounts of fat contained in many foods. It is our impression that these patients improved faster than others in whom this was not done.

In the vein of Devil’s Advocation, though, we should give our critical-thinking wheels a whirl and see if there are any alternative explanations for these findings. Could there be some suspicious lifestyle confounders like we see in studies today? Were saturated fats lumped in with hydrogenated ones and mutually given the boot, while only the latter was a true health-harmer? Was the supplementation of cod liver oil, chock full of omega-3 and vitamin-A-rich goodness, enough to explain the mortality patterns? Was the culprit saturated fat itself, or was it just guilty by association—maybe serving as a proxy for generalized I-Don’t-Give-A-Hoot-About-My-Health-itis, as the rebels who ignored Swank’s “eat less saturated fat!” message were more likely to ignore the “sugar is bad” or “vegetables are good” or “don’t run with scissors while chewing arsenic candy” messages as well?

As much as I love wailing my confounder alarms, I don’t think the usual suspects apply in this case. Here’s why!

  • Hydrogenated fats were indeed banned from the menu, but not until 1951—two or three years after folks started the diet, at which point they’d already seen profound improvements.
  • Likewise, cod liver oil (and other omega-3-rich fluid oils) weren’t allowed during the first couple years of the diet, and again, folks saw an 80% decrease in their MS flare-up rates during that time.
  • As for potential bad-guy carbs skewing the results? Remember, Swank developed his diet in the late 1940s—decades before refined grains and sugar joined the Dietary Villain ranks and were widely considered harmful. In fact, his diet didn’t restrict refined grains or sugar to begin with, so it wouldn’t make sense that a “screw it all” mentality would result in folks rebelliously downing processed carbs. (Heck, in Swank’s Multiple Sclerosis Diet book, there are plenty of recipes calling for white flour and sugar, and no indication that his patients—at least during the first few decades—were told to stick to whole-food sources of carbohydrate.)

And lest we wonder if the folks eating more saturated fat were also eating more of those gnarly, high-PUFA vegetable oils we so love to hate (thus obscuring the real relationship between saturated fat and MS progression): no dice! Swank’s data showed pretty clearly that the more saturated fat people ate, the less vegetable oil they were eating—an inverse correlation of -0.62 by the study’s 22nd year. Likewise, as Swank explained in the paper we’ve been discussing, higher unsaturated oil intake went hand-in-hand with better outcomes in terms of disability and mortality. The inclusion of vegetable oils “seemed to make patients feel better, increased their energy, and improved the condition of hair and skin,” even though Swank maintained they weren’t necessary for treating MS itself (since his patients took a fantastic turn for the better before those oils were added back to the diet).

Hence why I called Swank a “monkey wrench” a bit earlier on. In contrast to the theory that rock-bottom PUFA levels are the real reason very low fat diets work (while saturated fat would be an innocent, needlessly reduced bystander), Swank found the opposite to be true—that saturated fat had the strongest relationship with MS progression and mortality, while polyunsaturated fat levels could be scaled up or down without negatively impacting health. At least within that 10 to 40 gram range. And at least for multiple sclerosis (although mortality from other chronic diseases dropped during Swank’s diet as well).

That said, there is another possibility that could maybe, maybe, MAYBE be obscuring the saturated fat trend: Neu5Gc. This is a little sialic acid molecule that humans, rather uniquely, can’t synthesize. We produce a similar molecule called Neu5Ac, but lost the ability to make Neu5Gc after splitting from our last common ancestor with the great apes. Because gene mutations ‘n stuff. Other primates (and almost all mammals) can still synthesize Neu5Gc just fine!

Here’s why it matters: even though we can’t make Neu5Gc on our own, we can incorporate it into our tissues when we ingest it from food. And the prime sources of Neu5Gc happen to be red meat and dairy, the same food-vehicles that deliver most of our saturated fat. Because Neu5Gc looks like a foreign substance to our paranoid human innards, we’re capable of producing antibodies against it (although levels of those antibodies vary widely from person to person), which, in turn, can stir up all sorts of trouble. So far, we’ve got theories about a potential role for anti-Neu5Gc’s antibodies in systemic inflammation, cancer, heart disease, hypothyroidism, and also… WAIT FOR IT… multiple sclerosis. (See: “Why does multiple sclerosis only affect human primates?” and “Missing links in multiple sclerosis etiology. A working connecting hypothesis.“)


From “Quantitative analysis of sialic acids in Chinese conventional foods by HPLC-FLD,” 2014.

I waffled over whether to even include this speculation here, since Neu5Gc research is just a wobbly-kneed babe in the woods right now, and we’d need a lot more data before we could definitively link it to any disease. But, considering that 1) humans are the only primate that develops multiple sclerosis, 2) humans are the only mammal that can’t make Neu5Gc (and that produce antibodies against it), 3) multiple sclerosis tends to cluster around areas with a higher intake of Neu5Gc-containing foods (land meat and dairy), and 4) there’s a plausible mechanism linking Neu5Gc with the development of multiple sclerosis (through the effects of antibodies on the blood-brain barrier and axon-myelin unit)… well, all I’m sayin’ is it seems quite intriguing!

That said, Swank actually had some compelling, research-backed ideas about why saturated fat per se could trigger MS, which we’ll discuss in Part 2 of this post. It’s worth noting that he didn’t believe saturated fat causes multiple sclerosis, just that it “precipitates or accelerates it in susceptible individuals,” as he wrote in 1954. It’s not outside the realm of possibility that Neu5Gc somehow plays the role in the development of MS, while saturated fat, under specific conditions, aggravates it. So, even if this Neu5Gc stuff pans out, it doesn’t necessarily pardon saturated fat!

At any rate, we’re not quite at the end of Swank’s research rope. In 2003, he published the 50-year results of his never-ending study: “Review of MS patient survival on a Swank low saturated fat diet.” By that time, after 16 years of letting them do their own thang, he’d managed to track down 15 of his surviving participants and pull them in for interviews and in-person visits. They were all between 72 and 84 years old. (Swank himself was 94, and in case you’re wondering, was still goin’ strong until he passed away in 2008 at the age of 99). Amazingly, only two survivors needed help walking and had any sign of their disease; as for the others, Swank described it thusly:

The remaining 13 patients were remarkably well. They were very active, could care for themselves, could walk as necessary, and were normal mentally. … [They] stood and were active and unusually youthful looking, with very smooth facial skin devoid of wrinkles due to good subcutaneous circulation. They were all in friendly, good spirits, had joyful laughter, and generally quite youthful behavior. This study also indicated that patients with MS, if they rigorously follow the extremely low-fat diet proposed by Swank, which contains no more than 10 to 15 g/d of saturated fat, can expect to survive and be ambulant and otherwise normal to an advanced age.

Take note of that “good subcutaneous circulation” comment, because it’s at the crux of Swank’s theory about why saturated fat is harmful for MS patients (hint: it has to do with blood cell aggregation and oxygenation). More on that in the next blog-post installment! We’ll be coming back to some of Swank’s research and ideas when we discuss the science behind why low-fatting works, because he’s just that awesome.

In the meantime, we’re hardly done with our historical saga. Next up is another obscure figure from the dusty, carb-encrusted pages of science’s past: Lester Morrison.

Lester Morrison: ???

Lester Morrison is a man of mystery. A mystery, in part, because he was using low-fat diets to treat heart disease years before Ancel Keys supposedly introduced the idea to the world. A mystery also because for a guy who did some neat stuff, he’s totally not Google-stalkable. Nary a pixel of his face could be found in my sleuthing efforts, and the only bio-esque document seems to be his obituary.

Here’s what we do know. A rather precocious and multitalented individual (along with being a physician, he was an accomplished violinist, historian, symphony concertmaster, novelist, and maybe also Spiderman, and won his first research award when he was a sweet 16), Morrison plunged into the research field with an initial focus on gastroenterology. His first taste of Fat Suspiciousness came in the 1940s, after noticing heart disease and stroke mortality dropped hand-in-hand with wartime food rationing—which curbed, among other things, dietary fat. As he explained in his 1955 article, “A nutritional program for prolongation of life in coronary atherosclerosis,” World War I saw the first parallel dance between food rationing and heart disease, with the British blockade of Germany; World War II was fat-jà-vu all over again:

During World War II … the various Scandinavian governments supplied the information that, when fat in the diets of the population had to be reduced to a minimum because of the scarcity of dietary fats, the death and sickness rate from atherosclerosis was likewise reduced to a minimum. As soon as the war ended and dietary fat once again became plentiful, the mortality rate and morbidity rates from coronary and cerebrovascular disease promptly soared to prewar levels and even began to surpass them.

If you’ve read my critique of Forks Over Knives—a documentary that cited the same wartime stats as evidence for low-fat-plant-based superiority—then you already know I have some bones to pick with those correlations. As I explained in my critique, the rationing-induced changges involved far more than tanking total dietary fat: during World War II, seafood consumption doubled; sugar intake halved; vitamin K2 intake rose; trans-fat-containing-margarines all but exited the table. Not to mention, any situation of food restriction tends to boost vascular health, at least initially.

Nonetheless, the link between war rationing and mortality seemed to catch a lot of eyes, and Morrison’s were two of ’em. Cardiovascular disease had slayed a hefty portion of family (including his mom and dad), so the issue was quite near to his heart (literally!). Thus, he was inspired to investigate the matter in a more controlled setting.

In 1946—years, again, before Keys was hot on fat’s trail—Morrison launched a dietary study involving 100 people who’d recently survived a heart attack. For 50 patients, he left their menu as-is: fairly high in fat (80 – 160 grams per day) and high in cholesterol (200 – 1800 mg per day). For the other 50, he prescribed a war-rationing-inspired diet limited to 20 – 25 grams of fat per day—with continual supervision for both the patients and their families to ensure they stuck with it (and the boot given to anyone who couldn’t promise total compliance). He then followed everyone for eight years to see how many folks perished from each group.

In his papers “Arteriosclerosis: Recent advances in the dietary and medicinal treatment” (1951) and “A nutritional program for prolongation of life in coronary atherosclerosis” (1955), Morrison gave a detailed outline of his experimental diet, which was designed specifically to lower folks’ cholesterol. (Worth noting: the diet didn’t limit sugar or other sweeteners, and didn’t reduce animal protein intake, as a variety of lean meats and dairy were allowed. In fact, it prescribed 60 – 100 grams of mostly animal-based protein per day, and folks ended up eating closer to 120 grams on average—well above the 5% threshold deemed toxic by T. Colin Campbell. Just sayin’!)

The encouraged edibles:

foods_permitted_morrison And the no-nos: foods_avoided_morrison(Patients were also encouraged to take a multivitamin containing vitamin A.)

The results? You can probably guess where this is headed! As intended, the diet tanked people’s total cholesterol—by an average of almost 100 mg/dL, in fact. But that’s just the beginning. At the three year mark, 15 people from the control group had died of heart disease, compared to only seven from the low-fat group. (That’s a difference of 70% versus 84% survival.) By the eight year mark, the survival rates had whittled down to 24% of the control group versus 56% of the low-fat group (again, most deaths being from heart disease). And by the study’s 12th year? A whoppin’ nobody was alive from the control group, whereas 34% of the low-fat group was still roaming this lovely green earth.


12-year data taken from Morrison’s “Diet in Coronary Atherosclerosis,” JAMA, June 25, 1960.

(Along with the whole “dying less” thing, Morrison noted that his fat-restricted patients frequently reported “a sense of optimism, well-being, and good spirits” after adopting the diet—a sentiment Swank echoed as well, though without either of their studies being blinded, it’s hard to say how much was just a placebo effect.)

As with the other old research we’re resurrecting in this post, the beauty of Morrison’s study is that it gracefully dodged some confounders that muck up science today. Back when he conducted his fat-slashing experiment, nobody was telling the public that sugar and refined carbs were bad along with fat. No one had an inbox full of Meat’s gonna kill ya news headlines. No one had nutrition labels on their food, or a reason to fear egg yolks, or retinas permanently emblazoned with the image of the Food Pyramid. Heck, even smoking wasn’t widely considered a health hazard yet (the Surgeon General didn’t officially declare a causal link between smoking and lung cancer until 1957). Americans’ minds were clean slates, and prescribing a shiny new dietary change (like reducing fat) didn’t invoke the same confounder cascade we see today. How grand!

All that said, in Morrison’s case, there were two potential caveats we should take note of. One, the low-fat group saw some initial weight loss—an average of 21 pounds for men and 17 for women during the first three years, but with no additional changes during the rest of the study—whereas the control group remained weight-stable the whole time. And two, the low-fatters appeared to spontaneously increase their protein intake, making it hard to know what macronutrient change—the uppage of protein or downage of fat—was really driving the results. Could these be the true wizards behind the better-survival curtain?

It’s certainly possible, though I’m hesitant to say they could fully account for 12 years of dramatically lower death rates. Protein doesn’t seem uniquely beneficial for heart disease patients, and there’s a dearth of research looking at intentional weight loss on post-heart-attack survival—so we really have no way of knowing how big a factor that was for Morrison’s patients. (Oddly enough, population studies show that being overweight rather than normal-weight actually predicts better long-term survival for people who have heart disease—and what’s more, losing weight after having a heart attack is counter-intuitively associated with higher mortality rates. Researchers call this the “obesity paradox“: carrying extra weight tends to increase people’s risk of getting heart disease, but for folks who already have the disease, it seems to prolong life. Of course, there could be quite a few alternative explanations for that trend, which are thoroughly discussed here. Interesting, nonetheless!)

At any rate, Morrison was excited (but wisely cautious) about his study’s results, and ended up penning a book called “The Low-Fat Way to Health and Longer Life,” which he published in 1958—the same year Ancel Keys launched the Seven Countries Study. But, refreshingly non-gun-jumping scientist that he was, he also called for “similar surveys utilizing larger numbers of cases for statistical evaluation” in order to deepen and replicate his findings.

And as luck would have it, those surveys eventually came from one of his very own patients: Nathan Pritikin!

Nathan Pritikin: heart un-breaker extraordinaire

pritikin(NOTE: unless otherwise referenced, the background info in this section comes from the memoir, “Pritikin: The Man Who Healed America’s Heart,” which is actually one of the most interesting biographies I’ve ever read!)

Small world gettin’ smaller, eh?

Nathan Pritikin is probably best known for two things: 1) promoting an uber-low-fat diet (and running a longevity center dedicated to such); and 2) being downright chummy with George McGovern—the senator in charge of the 1977 Dietary Goals for the United States. (As I detailed in “Death by Food Pyramid,” McGovern ate a quasi-Pritikin diet for many years, delivered Pritikin’s eulogy, and drew a fair bit of inspiration from the man while crafting America’s new low-fat food recs.)

Pritikin’s dietary saga started with the same World War II trends that inspired so many of his fat-slashing contemporaries. Thanks to some work he’d done on bombsights for the Air Force (he was a prolific inventor and engineer by trade), Pritikin had free-for-all access to classified military documents—including the mortality data being churned out for civilians and prisoners. His puzzle-loving mind was intrigued that heart disease rates were dropping in areas plagued by intense stress and low food availability, when most health authorities expected the opposite to occur.

In 1955, an increasingly health-interested Pritikin made a trek to visit Mystery Man Morrison. The two nerded out about heart disease theories, and at the good doctor’s urging, Pritikin had his cholesterol tested for the first time—revealing a borderline-high level of 280 mg/dL. Although he started making some minor dietary tweaks after that, it wasn’t until an official heart disease diagnosis in 1958 that Pritikin kicked his nightly-pint-of-ice-cream habit and got serious about healing himself.

And we’re talking serious serious. For the next ten years, Pritikin guinea-pigged his body with every dietary permutation imaginable. He’d go for weeks eating almost nothing but lentils, or brown rice, or brown rice plus beef; he’d make slight or dramatic adjustments to his vegetable-grain-meat ratios; he’d experiment with eating ten dates after dinner (fruit, not women; this was before Tinder). And in true scientist fashion, he got his blood tested after each new food stint—meticulously documenting changes in his cholesterol levels, triglycerides, fasting and non-fasting glucose, red blood cell count, white blood cell count, hemoglobin, platelets, carbon dioxide, electrolytes, free thyroxine, and pretty much everything else he could cajole his docs into measuring. Be still, my nerd heart!

Although he managed to whittle his cholesterol down to 155 mg/dL with a lowish-fat menu (including a daily helping of nuts, tiny amounts of oil, and some fish and meat), his next EKG didn’t show a lick of improvement. But the man was not deterred! Pritikin low-fatted harder. He nixed the nuts and oil and meat. He pounded his total cholesterol down to 120 mg/dL. And at his next EKG six months later, the report was music to his ears (and a shock to his diet-skeptical doctors): “Definite improvement since the [last] tracing … Normal electrocardiogram.”

Ensuing stress tests over the next few years confirmed that his once-diseased ticker was, for all intents and purposes, healed.

By the 1970s, Pritikin had not only dialed in his own diet to his strict standards of perfection; he’d also amassed hundreds of low-fat devotees—family, friends, friends of friends, and eventually word-of-mouthers—whom he counseled over the phone for free. And in 1976, the real fun began: Pritikin opened the doors of his first Longevity Center in California. Riiiiight here:


The Casa Del Mar building, later transformed into Pritikin’s healing grounds. From Santa Monica Library Archive.

It was in this magical palace, wringing patients through a 26-day diet overhaul and hawkishly watching their food intake, that Pritikin could document the effects of his program on a large-scale basis.

And document he did! Although his reputation was that of a heart-healer, Pritikin’s diet did far more than assuage troubled arteries. Just as Kempner saw, the uber-fatlessness had the fortunate side effect of rapidly—and often permanently—healing diabetes. The earliest PubMed-able record of Pritikin’s success with diabetics came in 1976, with this little gem:

(FYI: Pritikin’s name isn’t on this paper, but it darn well should be. Scandalous tidbit alert! In 1974, Pritikin divulged the details of his diet to a Dr. James W. Anderson—an internationally acclaimed diabetes researcher at the time—and proposed an official study to test the uber-low-fatness on diabetics. Pritikin designed every detail of the study, drummed up $10,000 to fund it, created meal plans, and then received a grand total of zero credit for any of his contributions once the study was completed and published. In fact, Anderson proceeded to repackage Pritikin’s diet under a different name (the “HFC Diet,” standing for the “High Fiber, High Carbohydrate Diet”) and claimed it as his own invention—conducting a number of additional studies so successful that, in 1979, the American Diabetes Association was inspired to downsize its recommended fat intake. Despite years of haranguing Pritikin for more funding money (and thanking him profusely in their private correspondences), Anderson never publicly mentioned his collaboration with Pritikin or credited him for originating the diet. Buuuurn.)

So what was this study all about? For one week, 13 diabetic men—all who needed either insulin or oral drugs to control their blood sugar—ate the standard American Diabetes Association diet of the time: 34% fat and 43% carbohydrate, the rest made up of protein and wishful thinking. After that, they spent at least two weeks in Carbsville, eating a diet of only 9% fat and 75% carbohydrate. (The two diets were isocaloric, meaning they had the same number of calories, and were designed to help folks maintain their weight rather than lose or gain any.)

Surely that carb palooza sent their blood sugar into a frenzy! …Except it didn’t. After switching from the ADA diet to Pritikin-hijacked-by-Anderson one, nine patients had their insulin and oral drugs completely discontinued—at which point their fasting blood sugar was actually lower than it had been when they were still on medication. (The cutoff for drug discontinuation was a fasting blood sugar of 120 mg/dL.) It only took nine days of low-fatting to make that happen! Another patient’s insulin needs dropped from 28 to 15 units per day. As a further head-scratcher, fasting triglycerides, contrary to what we might expect, dropped significantly for ten men. The only folks who didn’t see any benefit from the 9% fat diet were the three who had the most advanced diabetes at the study’s onset—needing 40 to 55 units of insulin per day.

A tabular and graph-ular representation of changes in fasting blood sugar, in case you’re curious:



Even more intriguing, those results weren’t just a result of weight loss. Only five patients dropped more than three pounds, and there was no difference in fasting blood sugar and triglyceride levels between them and the weight-stable folk. And most importantly, the results seemed to stick: after the study, the nine most successful patients were weaned onto more flexible diets—60 to 65% carbohydrate instead of 75%. And after at least four months of followup, the researchers noted their blood markers hadn’t changed and the “control of the diabetes has been satisfactory without any drug therapy.” Pretty neat, huh?

Okay, okay: this study was tiny and used only men and we could find plenty of things to complain about (more womenfolk! Longer time frame! More Instagram pictures of the researchers’ feet overlooking exotic and envy-inducing locales!). Those are fair criticisms, to be sure. But this study does offer something we hardly ever see: a direct comparison between a Swampland menu (34% fat) and a Carbosis menu (9%), with the subsequent revelation that Oh hey, they actually have totally different results!

Again, as I’m trying to hammer home in this post, most of our “low fat” studies are actually only comparing different shades of Swamplandness—without ever hitting that fantastical 10% that brings a dramatic metabolic shift. And that makes us think that low fat is just a sham that does nothing except make our food taste like rabbit chow. So finding a study-gem like this, where fat intake actually does dip into the magic zone, and where the impressive results challenge our “low fat is bunk” narrative, is a rare and valuable find—even if it could be better designed.

That said, we’re in luck because Pritikin actually published some of his own studies on diabetics, and they were bigger and longer. Two of those papers came out in 1982 and 1983, respectively:

  1. Response of non-insulin-dependent diabetic patients to an intensive program of diet and exercise.
  2. Long-term use of a high-complex-carbohydrate, high-fiber, low-fat diet and exercise in the treatment of NIDDM patients.

The first paper looks at how 60 diabetics fared during the 26-day program; the second paper looks at how they did once released back into the scary, high-fat-food-filled ‘real world.’ In the latter, Pritikin described the in-house program as such:

During the 26-day session, the patients were served and taught to prepare the Pritikin high-complex-carbohydrate, high-fiber, low-fat diet. The diet consisted of unprocessed natural food with no supplements, e.g. guar. Less than 10% of the total calories were obtained from fat, … 13% from protein, and the remainder from carbohydrate (90% complex—whole wheat grain, rice and bread, beans, peas and other vegetables, and fresh fruit). … Protein was derived primarily from vegetable sources, except for nonfat milk, which was served daily, and small amounts of fish or fowl, of which 85 g/wk were provided.

Right off the bat, we can see the study wasn’t free from animal protein, as folks were allowed to drink a fair amount of skim milk, chock full of that awful casein vilified in The China Study. It also contained very small (but not irrelevant) amounts of poultry and fish—about this much per week:


Image from HealthFinders Collaborative.

On top of that, the Pritikinites who started out overweight had their calories restricted, but everyone else could eat as much as they wanted (ad libitum). And all the participants were encouraged to go on short walks each day.

So what happened? At the end of their 26-day bootcamp, their pancreases exploded! Just kidding. They did super well. Of the 23 patients who’d entered the program needing to take oral hypoglycemic drugs, all but two had ditched them by the end of the program:

hypoglycemic_drug_usage_pritikin_dietAnd of the 17 folks who’d been taking insulin (with dosages ranging from 14 to 75 units per day), all but four were released from its needly shackles:

insulin_usage_pritikin_diet(After crunching the numbers, it turned out people’s reduction in fasting blood sugar was not correlated with weight loss, or with the amount of walking they did, or other changes that we might suspect played a role. Dietary adherence reigned supreme!)

Okay, I know what you’re thinking. Those results are nice and all, but 26 days is nothing in the span of a person’s disease history. Heck, I’ve had hold-sessions with Comcast last longer than that! What happened afterwards? Did Carbjo cometh?

Luckily, Pritikin wanted to know the answer to those questions too. Between two and three years after their romp at the Longevity Center, the patients got a phone call quizzing them on how well they’d stuck with the diet, how much they were exercising, and whether their medical status had changed. They also had to answer dietary recall surveys and food-frequency questionnaires, and got a fasting blood-sample kit in the mail, which is kind of weird but probably a lot more exciting than bills and MasterCard offers.

The results? Compared to when they were freshly released from their low-fat boot camp, seven more people were taking oral hypoglycemic drugs, and four more people were taking insulin. Does that mean the diet failed?! Can we sling a big, fat “I told you so!” to the low-fat warlords and reassure ourselves that the diet is bogus? Actually:

… the main difference between those patients who went back on medication at follow-up compared with those remaining off medication was the percent of calories derived from fat.

Basically, those who stuck with the diet kept reaping the initial rewards, but those who gallivanted back into the Swampland paid dearly!

Of course, all that’s a mere smidgen of the data that got churned out from the Longevity Center over the years. R. James Barnard, who’s served as Research Director at the Pritikin Center (among quite a few other professional feats), published over 100 studies on the Pritikin Program, looking at everything from cancer to diabetes to heart disease.


Hi, James!

And while part of me would love to sit here and write about EVERY SINGLE ONE of those studies in attempt to explode the internet with Literally the Longest Blog Post Ever, another part of me—the part that realizes even the most patient of my readers have sanity limits and day jobs—has vetoed that plan in favor of a briefer sampling. Behold!

ESCAPE HATCH FOR THE FATIGUED OF BRAIN: what follows is a pretty long and sciency scroll of scientific scienceness. I know this blog post is big. I know your brain is turning into soup. CLICK HERE if you want to bypass the study summaries and get back to Pritikin’s life narrative!

Note: the Pritikin Program involves both the Pritikin diet and an hour of daily walking. Although we could definitely consider exercise a confounder (since it can independently improve metabolic and hormonal markers), studies of walking alone haven’t demonstrated anything nearly as dramatic as what’s been achieved with the Pritikin Program. At best, we could wager that walking probably boosts the results of the program, but only contributes to a fraction of its overall effects.

Another note! Unless otherwise mentioned, all the studies below allowed an ad libitum (non-calorie-restricted) energy intake. The participants could eat as much as their hearts desired out of any Pritikin-friendly item except for fish and poultry, which were capped at three servings per week (combined).

Prostate Cancer Protection

  • Long-term adherence to the Pritikin Program was a major boon for squashing out insulin resistance and reducing prostate cancer risk. Folks who followed the diet between 1 and 28 years reduced fasting insulin by 52%, fasting glucose by 20%, HOMA IR (a measure of insulin resistance) by 62%, and fasting triglycerides by 45%. And in a group of overweight men, even a quick jaunt on the program (two weeks) reduced fasting insulin by 30%, fasting glucose by 14%, HOMA IR by 40%, and fasting triglycerides by 33%. Not impressed yet? Blood from the Pritikin-fed fellows (both two-week and long-term) drastically slowed the growth of prostate epithelial cells, suggesting benefit for the prevention of prostate cancer.

From “Effect of diet and exercise intervention on the growth of prostate epithelial cells.” Barnard, 2008.


  • In a similar study of both short-term (11 days) and long-term (average 14.2 years) Pritikin followers, more evidence of prostate cancer protection emerged. The short-termers saw a 20% drop in insulin-like growth factor 1 (IGF-1) levels and a 53% rise (that’s a good thing!) in insulin-like growth factor binding protein 1 (IGFBP-1); the long-termers saw a solid 55% drop in IGF-1 and 150% rise in IGFBP-1. (IGF-1 has been strongly associated with cancer due to its role in regulating cell growth and death, and and IGFBPs are proteins that bind insulin-like growth factors and neutralize some of their unsavory activities.) Likewise, the short-term group saw a 25% drop in fasting insulin levels and the long-term group saw a 68% drop. As with the previous study, both groups’ Pritikin-ified blood was particularly awesome at fighting cancer: the serum from the short-termers caused a 30% decrease in the growth of LNCaP cells (a line of human prostate cancer cells) and serum from the long-termers caused a 44% decrease, compared to baseline. Both groups’ blood also increased the rate of death for those cancer cells.
  • Yet another study of Pritikin dieters (27 obese men) showed the program may be an excellent defense against the disease: the men saw a 28% increase of sex hormone-binding globulin (which binds androgens and appears to help protect against prostate cancer) and a 43% drop in fasting insulin; for three men who started out with slightly elevated PSA (prostate-specific antigen) levels, the diet also reduced that. According to the researchers, “The increase in SHBG would result in more testosterone being bound and, therefore, less of the androgen available to act on the prostate. The decrease in insulin might also decrease mitogenic activity in the prostate.”
  • One more on the prostate front! An analysis of post-Pritikin-Program blood shed light on the mechanisms behind the aforementioned anti-cancer effects: apparently, the Pritikin Program reduced tumor cell inflammation and resulted in lower NFκB activation in LNCaP cells. (NFκB, or “nuclear factor kappa-light-chain-enhancer of activated B cells” if you want to be fancy about it, is a protein complex involved in DNA transcription, cell death, and cytokine production.) The program’s results also seemed to hinge on reducing IGF-1 levels, because when the researchers added back IGF-1 back to the participants’ squeaky clean, post-Pritikin-Program serum, the results were completely reversed:

From “Analyzing serum-stimulated prostate cancer cell lines after low-fat, high-fiber diet and exercise intervention.” Barnard, et al., 2011.

Reduced LDL Oxidation and Spiffed-Up HDL

  • Three weeks on the Pritikin Program resulted in LDL particles that, even by paleo- and low-carb-advocate standards, were much less likely to oxidize and promote heart disease. In a group of 80 Pritikinning men and women, average LDL particle size increased (with larger LDL considered less atherogenic), and 27% of the folks who’d started out with LDL pattern B switched to LDL pattern A (again, considered less atherogenic). In vitro, the participants’ LDL also became more resistant to copper-induced oxidation—with a 21% drop in initial oxidation, 13% increase in lag time (the delay before oxidation happens, as the LDL’s antioxidants become depleted), a 20% reduction in peak diene formation (a way to measure oxidation), and a 17% reduction in maximal rate of diene formation. All of that jargon basically means the LDL had become far more resistant to oxidation—at least based on this particular assay (there’s always some uncertainty about how well in vitro (outside the body) studies translate to in vivo (inside the body)).

From “Effects of diet and exercise on qualitative and quantitative measures of LDL and its susceptibility to oxidation.” Barnard, 1996.

  • Similarly, a study in postmenopausal women found that the Pritikin Program led to more oxidation-resistant LDL particles.
  • Even though the Pritikin Program often reduces HDL (the so-called “good cholesterol” lipoprotein), it makes the existing HDL that much awesomer. A study of obese men found that after three weeks in Pritikin Land, the HDL inflammatory index (AKA the ability of HDL to prevent LDL from oxidizing) changed from “pro-inflammatory” to “anti-inflammatory.” The activity of platelet activating factor acetylhydrolase also increased, LDL decreased by 26%, and triglycerides decreased by 29%. Basically, even though HDL was lower than at baseline, its function was mucho improved—”suggesting increased turnover of proinflammatory HDL,” according to the researchers.

From “Effect of a short-term diet and exercise intervention on inflammatory/anti-inflammatory properties of HDL in overweight/obese men with cardiovascular risk factors.” Barnard, et al., 2006.

Breast Cancer Protection

  • In postmenopausal women, the Pritikin Program fantastically improved the major metabolic and hormonal risk factors for breast cancer: on average, insulin dropped 29%, insulin-like growth factor 1 (IGF-1) dropped 19%, insulin-like growth factor binding protein 1 (IGFBP-1) increased 32%, estradiol (among women who were on hormone therapy) dropped 34%, and estradiol (among women who weren’t on hormone therapy) dropped 37%. In vitro, serum from those Pritikinettes dramatically slowed the growth (and induced apoptosis, or programmed cell death) in three breast cancer cell lines (MCF-7, T-47D and ZR-75-1) compared to blood from the same ladies before they’d started the diet.
  • In a study of pre-menopausal women, two months of the Pritikin diet led to some fun breast-cancer-protective hormonal changes: serum estrone and estradiol (types of estrogens) fell during the women’s early folicular and late luteal phases (with decreases ranging from 18% to 26%), with no negative impact on ovulation. (Exposure to ovarian hormones, especially estrogen, appears to increase breast cancer risk, so these changes could be expected to reduce it.)

From “Effects of a very low fat, high fiber diet on serum hormones and menstrual function. Implications for breast cancer prevention.” Barnard, et al., 1995.

Colon Cancer Protection

Heart Disease, and Related Adventures

  • Here’s a goodie called, “Effects of an intensive exercise and nutrition program on patients with coronary heart disease: Five-year follow-up.” (There’s no link because for some awfully frustrating reason, this paper is nonexistent online, and I was only able to track it down via the library. Email me if you want a PDF copy!) This study followed the progress of 64 Pritikin adherents with coronary heart disease, five years after they left the sheltered, intensive clutches of the residential program. Prior to entering the program, 59% of the patients had experienced a heart attack and 80% had angina (chest pain). During that five-year follow-up, four people died (one of cancer, two of a heart attack, and one of heart failure during a mitral valve replacement), and two additional folks had non-fatal heart attacks. Reports of angina dropped from the initial 80% to only 32%. Fascinatingly, despite the fact that this was a highly diseased group with the odds stacked against them in terms of morbidity and mortality, their annual death rates ended up being close to that of the “regular” population after they Pritikinized their lives! (At the time the paper was written, post-heart-attack mortality was around 10 to 15% for the first year following the event, and as high as 50% by the third year.)



From “Effects of an intensive exercise and nutrition program on patients with coronary heart disease: Five-year follow-up.” Barnard et al. J Cardiac Rehab 1983;3:183-190.

  • In 31 obese men, the Pritikin Program brought awesome improvements in a huge range of risk factors for heart disease—including blood lipids, oxidative stress, inflammation, and cell adhesion. After just three weeks, there were significant drops in LDL (25% lower), triglycerides (28% lower), fasting glucose (12% lower), insulin (30% lower), HOMA IR (33% lower), C-reactive protein (39% lower), soluble cell adhesion molecules, inflammatory cytokines, and monocyte adhesion. And to sweeten the deal, nine out of the 15 men who entered the program with an official diagnosis of metabolic syndrome no longer met diagnostic criteria by the time their three weeks of low-fatting was up.
  • In 15 hyperlipidemic men, two weeks of the Pritikin bootcamp improved a number of risk factors for heart disease: it lowered the men’s blood pressure, slashed triglycerides, reduced the aggregation velocity and maximum aggregation of platelets, and reduced the formation of thromboxane (which is made by platelets and promotes blood clotting and constriction of blood vessels).
  • Combined with daily aerobic exercise, the Pritikin diet facilitated major improvements in myocardial blood flow. After six weeks, resting blood flow decreased by 12%, and hyperemic blood flow increased by 9%—leading to an improved myocardial flow. (It’s impossible to know how much diet versus exercise played a role in these results, though!)
  • In postmenopausal women on hormone replacement therapy, the Pritikin Program heralded some promising changes on the heart-disease-protection front. After just two weeks, 20 women saw significant drops in their BMI, glucose levels, insulin levels, all serum lipids, and HOMA IR. The researchers conclude that these “rapid improvements may reduce the risk of acute myocardial infarction (MI), and if sustained, these changes may mitigate the risk for atherosclerosis progression and its clinical consequences.” Not too shabby.
  • In a small study of 11 men, the Pritikin Program appeared to be uber protective against atherosclerosis progression. In the span of 21 days, the program dropped LDL by 23%, triglycerides by 41%, fasting insulin by 46%, and fasting blood sugar by 7%. Serum 8-iso-PGF 2 alpha (a product of lipid perodixation) also sank, whereas nitric oxide availability rose. In Barnard, et al.’s words: “The present study is the first to show that unrestricted consumption of a low-fat, high-fiber diet and daily exercise can mitigate oxidative stress, improve NO availability, and normalize BP in obese men within 3 weeks.”

From “Effect of diet and exercise intervention on blood pressure, insulin, oxidative stress, and nitric oxide availability.” Barnard, et al. 2002.


From “Effect of diet and exercise intervention on blood pressure, insulin, oxidative stress, and nitric oxide availability.” Barnard, et al. 2002.

Diabetes Damage Control!

  • For 70 diabetics, a 26-day jaunt in the world of Pritikin led to all sorts of disease-blasting perks: fasting blood sugar fell by 24%, blood pressure dropped by 10%—and even better, 71% of the folks who’d been taking oral hypoglycemic drugs pre-Pritikin were able to ditch their prescription (ditto for 44% of the folks taking insulin and 61% of the folks who’d been taking blood pressure medication)! Also, VO2max (“functional maximum oxygen uptake) improved by 41%.
  • In diabetic men, a three-week Pritikin Program reduced heart-disease risk factors associated with diabetes: the program whipped down fasting glucose by 20%, dropped fasting insulin by 30%, and totally clobbered markers of oxidative stress, inflammation, and monocyte-endothelial interaction (a major part of the heart disease process). Boom!
  • Among 652 non-insulin-dependent diabetics, three weeks of the Pritikin Program significantly improved their condition: fasting glucose dropped by 16%, and 71% of folks taking oral hypoglycemics (and 39% of the folks taking insulin) were able to discontinue those medications. Triglycerides and blood pressure also fell to delightful new lows.
  • In diabetics, the Pritikin Program reduced fasting insulin in diabetics by an average of 33% in just three weeks, and restored normal fasting insulin levels in 59% of insulin-resistant folks in the same time frame. (Triglycerides also dropped across the board!)

In the Elderly

  • Among 70 folks aged 70 and older (average age of the group being 78.7), 26 days of the Pritikin Program totally spiffed up their health: they lost an average of five pounds, total cholesterol and triglycerides fell significantly, their treadmill performance increased by 49%, the type II diabetics of the group reduced their fasting blood sugar by 27%, and half of the patients taking blood pressure medication were able to discontinue it.

But What About the Children?!

  • In overweight and normal-weight children, two weeks of the Pritikin Program yanked down multiple cardiometabolic risk factors: insulin levels fell (by 28.1% in overweight kiddos and 52.5% in normal-weight kiddos, respectively), HOMA-IR fell (28.4% and 53.1%, respectively), leptin levels fell (44.1% and 69.3%, respectively), and a variety of proinflammatory cytokines were reduced (molecules that increase inflammation in the body). LDL went down by 24.5% and 29.4%, while HDL hardly budged. Those changes didn’t correlate with weight loss, either!
Effect of Pritikin Program on inflammatory cytokines

Effect of Pritikin Program on inflammatory cytokines. From “Effects of an intensive short-term diet and exercise intervention: comparison between normal-weight and obese children.” Barnard, 2013.

  • Also in children, two weeks on the Pritikin Program (for 19 overweight kiddos) massively improved factors associated with heart disease: the program tanked LDL by 25.3% and triglycerides by 39.5% (HDL didn’t significantly change); 8-isoprostaglandin F2 alpha (a marker of lipid peroxidation) fell by 81%, CRP fell by 41%, and MMP-9 (a measure of plaque stability and progression) fell by 49%. Markers of endothelial cell activation (ICAM-1 and sE-selectin) also went kerplunk! As the researchers summed it up:

    “The primary findings of this study provide evidence that even a short-term lifestyle modification program may (1) improve the lipid profile; (2) decrease production of the reactive oxygen species superoxide and hydrogen peroxide and increase NO production; (3) decrease endothelial cell activation and adhesion; (4) decrease inflammation; (5) decrease monocyte chemoattraction; and (6) decrease MMP-9, a marker of plaque destabilization, all of which may contribute to a reduction in atherosclerosis progression.”


From “Effect of a short-term diet and exercise intervention in youth on atherosclerotic risk factors.” Barnard, et al., 2007.


  • WE’RE NOT DONE WITH YOU YET CHILDRUNS. Yet another study on overweight youth found, again, a massive drop in insulin (33%), HOMA-IR (29%), triglycerides (40%), systolic blood pressure (10%), diastolic blood pressure (10%), and LDL (27%), with no change in HDL. All seven of the kiddos who were classified with metabolic syndrome at the start of the study had reversed their diagnosis by the end of it!

And that’s just a freakin’ handful of the research out there. Seriously. A handful. If you want to venture into the full-on Pritikin Research Safari, I recommend bringing snacks and water, ’cause you’ll be gone awhile!

Just to be clear, I’m not trying to say that the Pritikin Program is wildly successful for everyone who jumps aboard, or that it’s unequivocally the best treatment for the diseases it’s been used for. The point I want to make is more conceptual than anything. Contrary to the belief that low-fat, high-carb diets raise insulin levels, muck up glucose control, encourage obesity, and promote inflammation, the overwhelming majority of studies published on the Pritikin diet have shown the polar opposite. That doesn’t mean there isn’t a range of outcomes within each study; indeed, some participants do well, some do really well, and some don’t do well at all, even when the collective effect is positive. But seriously, the Pritikin Program yields impressive results, and it’d take a whole lotta’ somersaults of logic to conclude otherwise.

I know your brain is probably maxed out on scienceness now, so let’s resume our narrative of Mr. Pritikin’s life! First up, an interesting parallel between his way of thinking and that of the modern Paleo community.

In the early ’70s, Pritikin wrote a three-volume opus compiling his theories—dotted with hundreds of scientific references—about every major condition of the time: atherosclerosis, angina, high blood pressure, gout, arthritis, gallstones, kidney stones, diabetes, lung cancer, colon cancer, prostate cancer, breast cancer, as well as hearing and vision diseases. He was convinced they were all, essentially, manifestations of a huge mistake called “What Americans Typically Eat.” In fact, Pritikin had a fascinatingly ‘ancestral’ approach to nutrition (back at a time when linking diet to chronic disease period was considered cuckoo for Cocoa Puffs), which he discussed in the introduction to his three-volume work:

Such a postulation linking diet and the scourge of degenerative diseases may seem far-fetched, until one reflects upon some basic biological facts. All animals, man included, have a diet that emerged from the long-term experience of the species in nature over many thousands of years. Now, man no longer eats foods his body was designed to eat, but has created a synthetic diet—the penalty of which is to endure the adverse effects of short-term experience. The more man’s diet departs from foods to which he is biologically suited, the more the adverse effects.

Whereas the modern incarnation of Paleo has drawn heavily from higher-fat-eating societies (the Inuit and Masaai are particularly cherished), Pritikin found dietary consistency among the starch-based populations: the Tarahumara Indians, the Bantu, and the natives of New Guinea, whose traditional low fat intake and virtual freedom from heart disease fed into his overarching philosophy about food. Just keep that in mind if you find the idea of low-fat eating to be ancestrally preposterous!

For what it’s worth, Pritikin certainly walked his talk. Before his mortal departure, he’d requested an autopsy be performed so the world could have an honest look at his once-diseased heart. The results were published in a New England Journal of Medicine article called “Nathan Pritikin’s heart,” where the state of his ticker blew the pathologist away:

The epicardium was smooth, and no scars were visible. The endocardium and all valves were normal. … The coronary arteries were soft and pliable … there were no raised plaques and no compromise of the lumens. No clots were present. … No infarcts of any size, or other finding referable to vascular disease, were present in any organ.

The report concluded, “In a man 69 years old, the near absence of atherosclerosis and the complete absence of its effects are remarkable.”

Important note on Pritikin’s death: Critics of Pritikin (Priticritics?) often point to his departure from the world—death by suicide, amidst a painful battle with leukemia—as evidence that his low-fat diet either pulverized his mental health (suicide) or gave him cancer (leukemia). Those are both pretty bold claims, so let’s take a moment to assess them.

In the 1950s, Pritikin went through a series of high-dose radiation treatments for a skin condition, pruritus ani, that was stubbornly resisting all the pills and ointments he tried. Over the course of two months, he got blasted with a total of 220 rads (“radiation absorbed doses”) of unfiltered x-rays—the equivalent of getting 3700 chest x-rays or 22 million dental x-rays. Not surprisingly, it wreaked havoc: he soon ended up with a blood condition he’d battle for the rest of his life, later diagnosed as a rare form of leukemia. That all happened before he embarked on a low-fat diet.

Pritikin actually managed to keep his disease in remission for almost three decades, but in late 1984, saw an unfortunate resurgence of symptoms—including leg swelling so severe that he had to stop going on his beloved daily runs. His docs ushered him onto some experimental chemotherapy, telling him it was the only way he could ever resume jogging. Not only did the chemo not help, it also left Pritikin with crippling anemia, kidney failure, diabetes, all-consuming pain, and 30 pounds stripped off his already slender body. In February of 1985, he flew out to New York for a second opinion, was told his case was hopeless and he would certainly die within the next six months, and proceeded to take his own life in a hospital bed in Albany.

As George McGovern reflected, “He was always in charge of his life. It rather followed he’d want to be in charge of his death.”

With all that in mind, can we implicate Pritikin’s diet in his sad demise? With enough mental gymnastics and anti-low-fat bias, sure—but I don’t think it’s particularly logical or fair. Did his diet shorten his life, or lengthen what was originally a grim prognosis? And if we make an automatic diet-death jump, should we do the same for the recent passings we’ve seen in the high-fat community (low-carb guru Barry Groves, saturated-fat-redeemer Mary Enig, or “Carbohydrates Can Kill” author Robert Su)?

As drawn as our human brains are to stories and anecdotes, I think it’s wiser to look at bigger, statistically stable data instead of case studies. Everyone has to die of something, and food isn’t the only tool of the grim reaper.

Ancel Who?

keysOh hey! Remember this guy?

The one who allegedly pulled the low-fat theory out of thin air, manipulated the American Heart Association into believing it, and took the world by storm with his Machiavellian wiles?

The one who cherry-picked his way into a groundless theory?

The one whose very countenance makes us tremble with rage, as we thrust our butter-encrusted fists in the air in defiance?

Yeah… him.

Let’s get real here for a minute. When it comes to evidence supporting low-fat recommendations, Keys was more like the ending caboose than the engineer driving the train. He was far from the first person to think that fat could play a role in chronic disease. We didn’t need his suspiciously curvilinear six-countries graph, we didn’t need his Seven Countries Study, and we didn’t need his can’t-prove-causation epidemiological utterings to form our “low fat movement.” Those things certainly played a role in pushing low-fat theory into the realm of public policy, but holy massive research pile Batman, there was plenty of evidence already there before he made his own contributions! Heck, Keys was clearly aware of his low-fat predecessors by the time he hit the scene: he even referenced the Rice Diet in his now-infamous 1953 paper “Atherosclerosis: A Problem in Newer Public Health” (PDF), where his six-country graph made its grand debut.

Even if you exit this blog post still believing “low fat” is an awful sham, we can’t keep spreading the myth that Keys was its originator. Every time we place the low-fat movement squarely on his shoulders, a kitten gets ejected from a machine gun and blasts a hole through the ozone. Do you really want to destroy the planet? Do you? I didn’t think so! Let’s just agree right now to put this whole myth to rest, and fill our alotted “hatin’ on Keys” time with more important matters, such as how to invent prosthetic goosebumps for people who cannot feel fear and/or coldness.

Modern Diet Doctor Squad: An update and apology for jumping the gun

Although this post isn’t diving into research by the modern “Plant-Based Diet Doctor Squad” (Caldwell Esselstyn, John McDougall, Dean Ornish, Neal Barnard, and by some definitions Joel Fuhrman), I do want to comment on the first gentleman in that lineup—the heart-disease-bustin’ Esselstyn of the renowned Cleveland Clinic.

If you aren’t already familiar with him, Esselstyn is a super cool cat who I want to adopt as my grandpa, and who’s been treating death-bed-borderline heart disease patients for decades. In my Forks Over Knives critique, I criticized his fat-shunning program on account of the high triglycerides and low HDL it produced in some of his patients, while also pointing out the inconclusiveness of his published work due to its tiny sample size. Allow me to awkwardly quote my self of four years ago:

Holy triglycerides, Batman! Although Esselstyn’s diet helped lower most of his patients’ triglycerides, a couple still have values in the major danger zone (362?). Some of those HDL numbers are looking pretty sorry as well.

All in all, Esselstyn’s study shows that a whole-foods, plant-based diet is probably infinitely better for cardiovascular health than the junky cuisine many folks eat. But it’s far from conclusive evidence that this diet is the best we can do for reversing heart disease, or that it would generally be effective in a population beyond his 11 self-selected subjects. A diet that reduces triglycerides and increases HDL more than his did, for instance, might have an even better outcome.

I’ve written a lot of things in my life. Some of those things have been wrong. This is one of them.

While I still suspect Esselstyn’s diet could be unveganized without harming its therapeutic effect (perhaps through the addition of gelatin, seafood, and whole-grain unicorn flour), I’m no longer convinced that low HDL and high(ish) triglycerides are bad in the context of unprocessed low-fat diets. I’ll explain why that’s the case in Part 2! But in a sneak-peek nutshell, it has to do with the improved quality of HDL (which we saw with one of the Pritikin Program papers) and the fact that higher trigs are typically a marker for insulin resistance, and when that’s absent, a heftier number probably isn’t pathological (which you can read more about in this paper yonder). We also have plenty examples of non-Western cultures that remain virtually free from heart disease, despite having triglycerides we’d consider too high and HDL we’d consider too low (hellooooo there Kitavans!)

As for the issue of Esselstyn’s statistically wobbly sample size (only 11 people who stuck with the diet through its 10-year follow-up): it was a fair criticism at the time, and one that’s been echoed by plenty of others. But it’s time to put that quibble to bed, preferably with enough sleeping pills so that it never again rouses. Why? It just so happens that in July of 2014, Esselstyn published a bigger, badder, bodacious-er study of 198 people instead of his original handful. The prescribed diet was the same oil-free, animal-food-free, whole foods regimen he’d been using for decades—with no confounding elements from exercise, meditation, psychosocial support, or yoga to be had. And the results were just as impressive—if not more so—than the ones he achieved in his earlier work:

  • Among the 177 folks who stuck with the diet (for an average length of 3.7 years), there was only one cardiovascular event related to disease progression: a stroke. That’s a recurrent event rate of 0.6%.
  • Among the 21 folks who strayed from the program, 13 had cardiac events. That’s a recurrent event rate of 62%.
  • (The five other deaths in the adherent group included three cancers, one pulmonary embolus, and one case of pneumonia.)

Image from “A way to reverse CAD?” Esselstyn, 2014.

Image from “A way to reverse CAD?” Esselstyn, 2014.

For sure, there’s still plenty of stuff we could nitpick about this study (and his previous ones): there was no control group; the patients were abnormally motivated and self-selected (rather than reflective of the general heart-disease population, who might not be so inclined to give up steak forever); the study wasn’t randomized; and some level of self-fulfilling prophecy could’ve been at play, since the patients were told initially how awesome the diet was and how stupendously it was going to scrub out their arteries.

But even with those shortcomings, I’ll say it loud ‘n clear: I’m impressed. The study documents true heart disease arrest, and actual reversal for some. We don’t yet have any published studies of that sort on ketogenic, low-carb, or paleolithic diets (as far as I’ve seen!). And in my old age, I’ve come to appreciate the fact that real-world outcomes (i.e., whether or not someone keels over and dies) are more valuable than intermediate risk markers (like specific blood lipid ratios). It doesn’t matter that your health looked good on paper if you still end up in the ER!

And now, for an apology.

For the past few years (five? six? seven ate nine?), I’ve asserted that the success of plant-based diets is due to their whole-foodsness (eliminating processed junk, refined sugar, and refined flour), their low PUFA intake regardless of total fat (with the implication that higher non-PUFA fat consumption would be hunky dory), and the increase in other health-promoting behaviors that come with making a big change in the foods you eat (more exercise, less drinking and smoking, less couch-potatoism, etc.). I still think those things are relevant. But I now believe I dismissed the role of low total fat intake before I gave the data a fair and thorough analysis. This is a breach of the standards I hold myself to as a science blogger, which involve impartially examining all evidence before drawing conclusions.

So, I’d like to take this opportunity to say, HEY GUYS: I’m sorry.

To the aforementioned doctors: I’m sorry for jumping the gun with your research and being snarky when I reference you. I don’t always agree with the way you interpret your own work; I’m often unsettled by your debate tactics; I worry that the co-mingling of animal rights activism and nutritional research is toxic for remaining scientifically objective. But good heavens, I sure dismissed y’alls results before digging as deep as I should have—and in the process, missed out on some amazing opportunities to broaden my understanding of diet (and communicate those findings to my readers). Thanks for being on this planet and helping broken people heal.

To my readers: You guy are the best, seriously. Somehow you put up with these sporadic and unreasonably long blog posts and say nice things to me and wish me a happy birthday, sometimes with the inclusion of cat pictures, which is very wonderful. I don’t take your readership, your trust, or my quasi-position in the public eye lightly. And while I can’t guarantee that every word I write in this blog will be totally accurate, or that the ideas I present now and in the future will stand the test of time, I can promise that I strive to question my own biases just as rigorously as I question others’. That process of questioning will result in errors coming to light, which I see as a very good thing indeed: science (and its interpretation) should be self-correcting! Thanks for hanging in there with me. I’m honored for your readership.

Up Next…

Hopefully by now, some things have become clear:

  • We can’t categorically blame low-fat, high-carb diets for heart disease and diabetes and obesity. We just can’t. NOT EVEN A LITTLE BIT. This needs to go away.
  • We probably can’t even blame refined sugar for that stuff (at least not in isolation). Oh, the pain of shattering assumptions!
  • We can’t ascribe the effects of low-fat, plant-based diets to their lack of animal products. Quite a few of the uber-low-fat studies here still allowed a fairly high animal protein intake, and still managed to whip people into diseaseless shape. Sorry, vegans!
  • Ancel Keys did not invent low-fat.
  • Ancel Keys did not invent low-fat.
  • Once more, with feeling! Ancel Keys did not invent low-fat.
  • I need to employ the “delete” button a little more rigorously. (Can you imagine how computer-crashy this blog post would be if I hadn’t split it into two parts?)

Admittedly, I’ve gone pretty easy on some studies in this post and—had I decided to devil’s-advocate from the other side—could’ve critiqued certain ones far more aggressively. But there’s a reason I didn’t go into BAD SCIENCE INCINERATOR MODE, and it’s not because I got bought off by the garbanzo bean industry (they’re gonna have to try a lot harder than 43 cents). For one, the sheer volume and consistency of the research here points to something very real, something totally non-under-the-rug-sweepable—and analyzing each individual study in search of holes and inadequacies wouldn’t change that fact. And secondly, there’s a whole universe of scientific mechanisms explaining why the diets discussed here work: the ability of dietary fat to reduce insulin sensitivity, the effect of different fats on tissue oxygenation and blood flow, the little-known ways that saturated fat can make LDL more likely to get incorporated into plaque, and all sorts of other fun stuff. That’s what the next post is about, so stay tuned!

In sum, there comes a point where it’s more of an intellectual stretch to rationalize something away than to accept that it may have merit. My friends, we’ve reached this point on the issue of low-fat diets. Let’s face it: they can actually do some good. And it’ll make a whole lot more sense why after you read Part 2!

Guess what? You made it to the end! Please remove of your pixel-dizzy eyes from the computer screen and get some fresh air. You freakin’ earned it!

ONE LAST THING. I want to give the stickiest, gooiest hug and eardrum-shattering shout-out to the amazing folks who donated to this blog over the course of the last year. You guys single-handedly made this post (and the one that’s coming next!) possible to finish, and your support allowed everything in here to reach a much wider audience than just the inanimate objects in my office-room. Thank you, thank you, thank you. If we meet in person, I’ve got a hug with your name on it; if we don’t, here’s a whole tray of Internet Bonus Biscuits! Mwah!

I fear that a picture of actual biscuits may be offensive to some readers, so I’ve chosen a picture of a cat “making biscuits,” which is the next best thing.

I fear that a picture of actual biscuits may be offensive to some readers, so I’ve chosen a picture of a cat “making biscuits,” which is the next best thing.



  1. I read this entire page and loved all this fantastic information. I did Atkins and paleo for abt 5 yrs total years and 3 years ago quit to do a Matt Stone rest and refeed because I just could not do it any longer. Recently I came acrocc Kempner’s rice diet, McDougall and now your lovely looong post. I will never be vegan and probably never vegetarian but will be following low of a very low fat high starch approach now that I realize I am starting to feel better on it and that even moderate fat always makes me feel like I am never digesting my meals.

  2. I’m sure you’ll never read this, but I just wanted to thank you for changing my mind about fat. I’ve been buying into the healthy fats propaganda for almost a decade, and while I think it can be good for some, it’s clearly not the best thing for my body. The worst blood sugar readings I ever had were on a low carb diet (my fasting glucose was always in the mid to high 90’s no matter how low I took my carbs). I ate low-carb paleo for almost two years, and while I did lose weight and get off my meds, my body was falling apart and I could not get my blood sugar to go down (and I couldn’t sleep, and I became agoraphobic, had internal bleeding, couldn’t think straight—yeah, LC was pretty destructive for me). I decided to go off the diet and just eat real food, but I could never drop the “good fat” mentality. As soon as I finished your article, I cut out all the extra fat I was eating, which really wasn’t hard, and within days my fasting blood sugar went down into the mid to low 80’s, where it has stayed as long as I keep the fat at about 10-12% (not counting the few times I tried to add in a supplement that made it go up). I’ve lost over 20 lbs and am now the lowest weight I’ve ever been in my adult life (126), and amazingly that happened without me even trying. I don’t count calories or fat grams or anything; I just eat healthy food and make sure I pick low-fat protein (except in the morning when I often eat fatty fish) and it all just falls into place. So thank you again for presenting this in such a logical way, and I am looking forward to part two whenever you get it up ❤

  3. Great post. I am sorry I’ve only found it now. I recently came to a somewhat similar conclusion myself recently when I took on 14 day raw vegan diet challenge.

    Being a 14 year low carb veteran I was not expecting the results I got. It causes me to look deeper.

    While I still prefer my low carb life I now know that carbs don’t make you fat. Too much fat combined with too much fat make you fat.

    Your chart on the magic points is right on the money as far as weight loss goes. I thought I was the only low carb advocate to see this. I guess you beat me. 🙂

    That being said, I still wouldn’t say gorging in white flour and sugar are a great plan especially after reading the work of Weston A. Price. Physical degeneration linked with addition of sugar, white flour and sweetened jams alone was real and significant.

  4. I wonder if the “squirrel” hypothesis written about here is at all relevant to the research and evidence that you have examined here, and of your working hypothesis.

    The blog’s fascinating finding is that the only two natural foods that have similar macronutrient profiles (high in BOTH fat and carb) to pretty much all modern industrial “junk food” are breastmilk (for infants who must gain weight to be healthy) and acorns (eaten only in autumn, by animals fattening up for winter). Isn’t that interesting? And a reason to reflect upon “autumn” style foods (which mix high fat ratios with high carb ratios) being consumed all year round, and the likely consequences…

    1. I wouldn’t hold my breath. Considering Denise hasn’t touched this web site in forever and she has disappeared from Youtube I don’t hold out much hope we’ll see part 2 or for that matter anything else from her. It could be that by publishing part one she has lost her following in the Primal/Ancestral group and the vegans have always hated her since she published he thoughts on the China Study. No one is interested in the truth. They just want confirmation of their own current program.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s