The Sugar Study That Blamed Fat: The Review the Sugar Industry Bought
In 1965, two days after new research tied sucrose to heart disease, the Sugar Research Foundation paid two Harvard scientists to review the evidence. The foundation set the objective, supplied the articles, received the drafts, and approved the conclusion in writing — and the 1967 New England Journal of Medicine review that taught America to blame fat never mentioned any of it. Built from the industry's own internal letters, published in JAMA Internal Medicine in 2016, with the strongest case for the defense argued at full strength.
By MyAudioBooks.ai ·
Listen free: The Sugar Study That Blamed Fat: The Review the Sugar Industry Bought
On July eleventh, nineteen sixty-five, the New York Herald Tribune gave a full page to new medical research suggesting that sugar — sugar, not fat — might be the dietary driver of heart disease. Two days later, in Washington, the executive committee of the Sugar Research Foundation approved a line item called Project two-two-six: five hundred dollars for one Harvard professor and one thousand dollars for another, half paid when the work started and half, in the foundation's own words, "when you inform me that the article has been accepted for publication." We have read the industry's internal letters about the review that money bought, and our verdict is blunt: the most consequential nutrition paper of the twentieth century was purchased before it was written. The receipt survives. This is the story of the study that blamed fat.
The work was a review of the evidence linking sugar to heart disease, and it would run in the New England Journal of Medicine in nineteen sixty-seven, under three Harvard names, with no mention of the Sugar Research Foundation anywhere in it. What that money bought, and what the country got instead of the truth, is on file.
At My Audio Books dot A I, you can create your own audiobooks from prompts, turn your documents into audio, all with one subscription, and store your items in your own personal library.
Section One. The Two Theories
Start with the fear that made the market. In the nineteen fifties, American men were dying of coronary heart disease at rates that turned diet into a national question, and by the early nineteen sixties two prominent physiologists had staked out rival answers. Ancel Keys blamed total fat, saturated fat, and dietary cholesterol. John Yudkin, a British physiologist, pointed at added sugars and argued the case against fat was, at best, unproven. Both men had data. Both had followers. What neither had yet was a referee, and into that vacuum stepped an industry with everything at stake: the Sugar Research Foundation, the sucrose trade's own research arm, founded in nineteen forty-three.
The foundation had seen the future early, and it liked what it saw. In nineteen fifty-four its president, Henry Hass, told the American Society of Sugar Beet Technologists that leading nutritionists were connecting the high-fat American diet to cholesterol and heart trouble — and then he did the arithmetic out loud. Fat carried forty percent of the calories in the national diet, and in his view it ought to carry twenty. If the carbohydrate industries recaptured that twenty percent, and sugar merely held its share of the carbohydrate market, the shift would mean, in his words, "an increase in the per capita consumption of sugar more than a third." Note what that sentence is. It is not a hypothesis about health. It is a sales projection that depends on a particular scientific answer, written down eleven years before the science was supposedly settled. The foundation would go on to spend six hundred thousand dollars — five point three million in twenty sixteen dollars — teaching, as Hass put it, people who had never had a course in biochemistry "that sugar is what keeps every human being alive."
By nineteen sixty-two the foundation's own scientific advisory board was nervous. It reviewed an American Medical Association report suggesting that, in low-fat diets, the kind of carbohydrate eaten might influence the formation of serum cholesterol, and it concluded that research developments in the coronary field "should be watched carefully." Watching was the foundation's vice president and director of research, John Hickson. And in December nineteen sixty-four Hickson wrote the memo that sets everything else in motion. From a number of laboratories of greater or lesser repute, he reported, there are flowing reports that sugar is a less desirable dietary source of calories than other carbohydrates — and he named the threat: Yudkin. His recommendation was that the foundation "could embark on a major program" to counter the negative attitudes toward sugar. Fund a committee of nutrition specialists. Find the weak points in the detractors' experiments. Replicate the studies with appropriate corrections. Then, in his words, "we can publish the data and refute our detractors."
Section Two. The Purchase Order
In nineteen sixty-five the foundation brought Frederick Stare, chairman of the nutrition department at the Harvard School of Public Health, onto its advisory board as an ad hoc member. Stare was the establishment: consulted by the National Academy of Sciences, the National Heart Institute, and the American Heart Association. Then, in June of that year, the Annals of Internal Medicine published a set of articles that made the threat concrete. One demonstrated that sucrose, more than starch, aggravated a condition called hypertriglyceridemia — an excess of fatty particles called triglycerides circulating in the blood. An accompanying editorial drew the conclusion the industry dreaded: if elevated triglycerides were a risk factor for coronary heart disease, then "sucrose must be atherogenic" — artery-clogging. On July first, Hickson traveled to Harvard to see a young faculty member of Stare's department, Mark Hegsted. On July eleventh, the Herald Tribune told the public that the new research threatened to tie the whole business of diet and heart disease in a knot. On July thirteenth, two days later, Project two-two-six was approved.
The purchase was specific. Hegsted and his colleague Robert McGandy, overseen by Stare, would write, in the foundation's words, "a review article of the several papers which find some special metabolic peril in sucrose and, in particular, fructose." Five hundred dollars for Hegsted, one thousand for McGandy, half on starting, half on acceptance. In the end the foundation would pay six thousand five hundred dollars — forty-eight thousand nine hundred in twenty sixteen dollars. On July twenty-third, Hegsted wrote to Hickson asking him to supply the articles the review should consider. Read that again, because it is the first hinge: the funder chose much of the evidence base. Most of what Hickson sent contained findings that could threaten sugar sales — which tells you the articles were selected to be answered, not weighed. On July thirtieth, Hickson stated the objective in writing, and this sentence deserves to be read slowly. Our particular interest, he told Hegsted, had to do with claims that sucrose makes an inordinate contribution to the metabolic condition hitherto ascribed to aberrations called fat metabolism, and, quote, "I will be disappointed if this aspect is drowned out in a cascade of review and general interpretation." On August tenth, Hegsted answered: "We are well aware of your particular interest in carbohydrate and will cover this as well as we can."
The project took more than a year, and the letters from that year are the second hinge. In April nineteen sixty-six, Hegsted explained the delay to the foundation with a sentence that should be carved into the wall of every journalism school: "Every time the Iowa group publishes a paper we have to rework a section in rebuttal." The Iowa group — Lopez, Hodges, and Krehl — kept finding links between sugar and serum cholesterol, and each new paper meant the review had to be rebuilt against it. This was not a survey of the evidence; it was a rebuttal with a publication schedule. Through the summer and fall, the drafts flowed to the funder. On September sixth, Hickson asked, "Am I going to get another copy of the draft shortly?" On September twenty-ninth, Hegsted promised it within a week or two. On October twenty-fifth, Hickson received the final draft, days before submission. And on November second, nineteen sixty-six, the foundation's director of research wrote back his review of the review: "Let me assure you this is quite what we had in mind and we look forward to its appearance in print."
Section Three. What the Review Did
In the summer of nineteen sixty-seven, the New England Journal of Medicine carried the result in two installments: Dietary Fats, Carbohydrates and Atherosclerotic Vascular Disease, by McGandy, Hegsted, and Stare. Notice the title. A project commissioned to answer papers about the special metabolic peril in sucrose had become a paper about dietary fats. Its conclusion was that there was "no doubt" that the only dietary intervention required to prevent coronary heart disease was to reduce dietary cholesterol and substitute polyunsaturated fat for saturated fat in the American diet. No doubt. In a field where two Nobel-grade reputations were still trading data, the most prestigious medical journal in America printed certainty — certainty that happened to be the exact sentence a sugar trade association had paid six thousand five hundred dollars to hear.
How did the review get there? Not by fabricating anything — the documents show something more elegant. The authors took each class of evidence implicating sucrose and discounted the class. Epidemiological studies showing sugar-heart associations across populations were set aside because of confounding — the technical term for causes so entangled that no single one can be isolated. Experimental studies showing sucrose raising cholesterol and triglycerides in volunteers were set aside because the doses were large compared with typical consumption. Mechanistic studies were set aside because they used fructose or glucose rather than sucrose itself. Animal studies were set aside because of species differences. Every one of those objections, taken alone, is a thing a careful scientist might honestly say. The review's craft was in saying all of them, in a row, about one side of the question — and none of them about the other.
Here our read turns from method to asymmetry, because the other side got the velvet treatment. When the review summarized the trials supporting its fat conclusion, it reported few study details and no quantitative results at all — and reading the underlying trials, the twenty sixteen analysis found the consistency had been overstated. Exactly one randomized trial cleanly supported the review's conclusion that cutting cholesterol and swapping in polyunsaturated fat substantially improved serum cholesterol, and that trial was conducted by Hegsted himself, with McGandy and Stare, in nineteen sixty-five. Then the review built the fence that settled everything: it implied that only randomized trials using serum cholesterol as the biomarker — the measurable stand-in for disease risk — could definitively answer how Americans should eat. Triglycerides, the very signal that sucrose actually moved, had already been waved out of bounds. Set the acceptable evidence to cholesterol trials, dismiss the cholesterol trials that used sucrose-sized doses, and the verdict writes itself. You do not have to fix a result if you are allowed to choose the measuring stick.
And one more thing was missing from the printed page: the Sugar Research Foundation. The authors' other research funding was disclosed. The foundation's role — the objective, the supplied articles, the draft copies, the written approval — was not mentioned anywhere. Here the defense gets its first fair point, so we will give it early: the New England Journal of Medicine did not require authors to disclose conflicts of interest until nineteen eighty-four, seventeen years later. No rule was broken, because the rule did not exist. But that is precisely why the non-disclosure was the product. A reader in nineteen sixty-seven could not discount what a reader in nineteen sixty-seven could not see. The most prestigious journal in the country had carried a marketing document dressed as a referee's report, and nobody holding the journal could tell the difference.
At My Audio Books dot A I, you can listen to this story and thousands of others that explore the hidden science and mechanics behind the headlines.
Section Four. The Strongest Defense
Now the other side at full strength, because a verdict that cannot survive its countercase is not a verdict. The strongest defense begins inside the twenty sixteen paper itself, in its own limitations section, and we will not soften it: there is, in the authors' words, "no direct evidence that the sugar industry wrote or changed" the review manuscript, and the evidence that the industry shaped its conclusions is "circumstantial." The documents are also, by the authors' own accounting, a narrow window — a few hundred letters from one trade association, assembled by researchers with a thesis. And the people who could explain themselves are dead. That is all true, and it matters.
The second line of defense is that the review's individual arguments were not absurd. Large-dose sucrose experiments really did not resemble the American diet. Population studies really were confounded. And triglycerides were genuinely not an accepted marker of heart-disease risk in nineteen sixty-seven — the American Heart Association was still issuing scientific statements on triglycerides and cardiovascular disease in twenty eleven, forty-four years later. Hegsted was a serious scientist, and Keys's fat hypothesis had real institutional momentum that owed nothing to sugar money. If the review had appeared with a disclosure line, a reasonable reader might still have found its science persuasive. That reader just would have known whose thumb had been near the scale.
Here is why the defense loses anyway. Circumstantial evidence is a weak word for a strong chain, and this chain is written in the buyer's own hand. The objective was dictated in advance: do not let the sucrose question be "drowned out in a cascade of review." The evidence was supplied by the buyer. The schedule was driven by the need to rebut new anti-sugar findings as they appeared — rebuttal, the author's own word. The final draft went to the buyer before the journal. And the buyer's written approval arrived before publication: quite what we had in mind. In business-school language this is a principal-agent failure — the public delegated its judgment to experts whose incentives were invisible — but we'd argue you need only one question: if this process had been printed on page one of the review, would anyone have read page two the same way? The industry's own behavior answers it. They did not disclose because disclosure was the one part of the product that would not have sold.
Section Five. What the Money Kept Buying
The review was not a one-time purchase; it was a working asset. The sugar industry kept funding research on heart disease and other chronic conditions, in the words of its own later account, "as a main prop of the industry's defense." In nineteen seventy-one the pattern repeated at the National Institute of Dental Research, which was building a federal program to fight tooth decay. The twenty fifteen analysis of that episode — same method, internal documents — shows the industry submitting its research priorities to the federal task force and the program's eventual research contracts tracking the industry's submission: steered toward enzymes, vaccines, and additives, and away from the one intervention that costs nothing and works, which is eating less sugar. The comparison table between the industry's wish list and the government's contract announcement is published with the paper, and the alignment is not subtle.
Then came the capstone. The industry commissioned a massive review called Sugar in the Diet of Man, and its own internal history credits that review, among other tactics, with favorably influencing the Food and Drug Administration's nineteen seventy-six evaluation of sugar's safety — the assessment that kept sucrose in the category regulators call G R A S, generally recognized as safe. And the cultural result? By the nineteen eighties, few scientists believed added sugars played a significant role in coronary heart disease, and when the first Dietary Guidelines for Americans appeared in nineteen eighty, their heart-disease advice centered on total fat, saturated fat, and cholesterol. Yudkin was a footnote. The study that blamed fat had won so completely that most people never learned there had been a contest.
Now the honest unwind, because this is where our verdict has to earn itself. In twenty ten, a meta-analysis pooling twenty-one prospective studies — three hundred forty-seven thousand, seven hundred forty-seven participants followed for five to twenty-three years, eleven thousand and six of whom developed coronary heart disease or stroke — found that saturated fat intake was not significantly associated with either. The pooled relative risk for coronary disease was one point zero seven, inside a ninety-five percent confidence interval of zero point nine six to one point one nine, which is statistical language for: no measurable signal. So was the sugar industry accidentally right that fat was oversold? Perhaps — the modern evidence cuts both ways, and we will not pretend otherwise. But note the detail that makes the twenty ten paper the mirror image of the nineteen sixty-seven one: it was supported in part by the National Dairy Council and a Unilever fellowship, and it says so, in print, on its first page. By twenty ten, you disclose. That norm — the receipt printed next to the result — is the real legacy of the exposés, and it is exactly what Project two-two-six was built to avoid.
The rest of the record is catch-up. By twenty sixteen, the World Health Organization had issued a guideline on sugar intake, the federal dietary guidelines had been revised, and the F D A was rewriting the nutrition label — the twenty sixteen analysis lists all of it as policies finally promulgated around sugar. And in September of that year, forty-nine years after the review, JAMA Internal Medicine published the letters. We think the lesson is sharper than sugar. A literature review is not a verdict from above the fight; it is a weapon inside the fight, and whoever funds it gets to choose the measuring stick. The study that blamed fat taught two generations what to fear on their plates. Its other legacy is quieter and more useful: it taught the rest of us to ask, of every study, every guideline, every no doubt — who paid for the question. Six thousand five hundred dollars bought fifty years of an answer. The receipt, at least, is finally ours.
At My Audio Books dot A I, you can create fiction, non-fiction, and turn your documents into audio, all stored in one place with a single subscription — plus get instant access to thousands of audiobooks and deep-dive investigations. Learn more today at My Audio Books dot A I.