Good Calories, Bad Calories Part 15

George Cahil , a former professor at the Harvard Medical School, is a pedagogical example. Cahil had done some of the earliest research on the regulation of fat-cel metabolism by insulin in the late 1950s, and had coedited the 1965 Handbook of Physiology on adipose-tissue metabolism. In 1971, when Cahil gave the Banting Memorial Lecture at the annual meeting of the American Diabetes a.s.sociation, he described insulin as "the overal fuel control in mammals." "The concentration of circulating insulin," he explained, "serves to coordinate fuel storage and fuel mobilization into and out of the various depots with the needs of the organism, and with the availability or lack of availability of fuel in the environment." When I interviewed Cahil in 2005, he told me it was true that "carbohydrate is driving insulin is driving fat." But Cahil did not consider this chain of cause and effect to be a sufficient reason to speculate that carbohydrates drive obesity. Nor did he consider it a possibility that avoiding carbohydrates might reverse the process. Rather, he believed unconditional y that positive caloric balance was the critical factor. When it came to weight regulation, Cahil repeatedly told me, "a calorie is a calorie is a calorie." He acknowledged that the obese ate no more, on average, than the lean, and this is why he believed that the obese must be fundamental y lazy and this was the proximate cause of their obesity.*121 There was no reason to test competing hypotheses, Cahil said, because any competing hypothesis would contradict the laws of physics as he understood them.

When clinical investigators tried to unravel the connection between diet, insulin, and obesity in human subjects, as the University of Was.h.i.+ngton endocrinologist David Kipnis did in the early 1970s, the results were invariably a.n.a.lyzed in light of this same preconception. Kipnis had fed ten "grossly obese" women a series of three-and four-week diets that were either high or low in calories, and high or low in carbohydrates. The fat-rich diets lowered insulin levels, Kipnis reported in The New England Journal of Medicine in 1971, and the carbohydrate-rich diets raised them, regardless of how many calories were being consumed. Even when these women were semi-starved on fifteen hundred calories a day, a high-carbohydrate content (72 percent carbohydrates and only 1 percent fat) stil increased their insulin levels, even compared with the hyperinsulinemia of these obese women on their normal diets.

One interpretation of these results is that we could remove the carbohydrates from the diet and replace them with fat, and weight would be lost, perhaps without hunger, because insulin levels would drop, even if the total calories consumed did not. Kipnis's results, as the University of Heidelberg clinicians Gotthard Schettler and Guenter Schlierf wrote in 1974, underlined the "necessity of restricting carbohydrates in obesity in order to restore insulin levels to normal, thus hopeful y decreasing appet.i.te and fat deposition...."

Kipnis, however, refused to believe that carbohydrates might cause obesity, or that avoiding carbohydrates might ameliorate the problem. When I interviewed him over thirty years later, he described the findings of his research as "very obvious." "You manipulate the amount of carbohydrates you give a human," he said, "you can manipulate his or her basal insulin level." He also said that "insulin causes deposition of fat in fat cel s." But when it came to the cause of human obesity or weight gain, Kipnis rejected the relevance of these physiological phenomena. "Most people are obese because they eat more than they need to sustain the energy requirements that they have," he said. "They eat too d.a.m.n much."

For the past quarter century, Americans have become progressively heavier and more diabetic. By 2004, one in three Americans was considered clinical y obese; two in three were overweight. One in ten adult Americans had Type 2 diabetes-one in five over the age of sixty. It is now clear that the roots of this epidemic are evident even in infants and in the birth weights of newborns. Among middle-income families in Ma.s.sachusetts, for example, as a team of researchers led by Matthew Gil man of Harvard reported last year, the prevalence of excessively fat infants increased dramatical y between 1980 and 2001. This increase was most conspicuous among children younger than six months of age.

The probable explanation is that as women of childbearing age get heavier and more of them become diabetic, they pa.s.s the metabolic consequences on to their children through what is known technical y as the intrauterine environment. The nutrient supply from mother to developing child pa.s.ses across the placenta in proportion to the nutrient concentration in the mother's bloodstream. If the mother has high blood sugar, then the developing pancreas in the fetus wil respond to this stimulus by overproducing insulin-secreting cel s. "The baby is not diabetic," explains Boyd Metzger, who studies diabetes and pregnancy at Northwestern University, "but the insulin-producing cel s in the pancreas are stimulated to function and grow in size and number by the environment they're in. So they start over functioning. That in turn leads to a baby laying down more fat, which is why the baby of a diabetic mother is typified by being a fat baby."

This is also the most likely explanation for why children born to women who gain excessive weight during pregnancy also tend to be fatter. As Laura Riley, medical director of labor and delivery at Ma.s.sachusetts General Hospital, told the Boston Globe in response to the Harvard study, she now tel s her patients, "If you overdo it during pregnancy, you're setting yourself up for a bigger baby," and that, in turn, means "you are setting your baby up for potential y a lifetime of weight problems." Gil man and his col eagues described the problem this way: "Our observation of a trend of increasing weight among young infants may portend continued increase in childhood and adult obesity."

But if fatter mothers are more likely to make fatter babies, and fatter babies are more likely to make fatter mothers, which is also a wel -doc.u.mented observation, then this is another vicious cycle. It suggests that, once a generation of adolescents and adults start eating the highly refined carbohydrates and sugars now ubiquitous in our diets, even their children wil feel the effect, and perhaps their children's children as wel . The extreme instance of this phenomenon today is the Pima Indians, whose incidence of diabetes is among the highest of any population in the world. In 2000, NIH investigators reported that Pima born to mothers who were diabetic have a two-to threefold increased risk themselves of becoming diabetic as adults, and so have a two-to threefold increased risk of pa.s.sing diabetes on to their own children-of "perpetuating the cycle," as the NIH investigators explained. The "vicious cycle" of the "diabetic intrauterine environment," they wrote, can explain much of the postWorld War I increase in Type 2 diabetes among the Pima, and may also "be a factor in the alarming rise of this disease national y."

The question we now face is whether the same vicious cycle may also be a factor in the alarming rise of obesity national y, as wel as international y.

There's no reason to think that the hormonal and metabolic consequences of high blood sugar-from what James Neel in 1982 cal ed the "excessive glucose pulses that result from the refined carbohydrates/ over-alimentation of many civilized diets"-do not pa.s.s from mother to child through the intrauterine environment, whether the mother is clinical y diabetic or not. If so, the longer the obesity epidemic continues, and the longer we go without unambiguously identifying the causes of obesity, metabolic syndrome, and diabetes, the worse this vicious cycle is likely to get.

Chapter Twenty-three.

THE FATTENING CARBOHYDRATE DISAPPEARS.

We need the help of the psychosocial scientists in finding better ways of communicating with our patients, in explaining to them that obesity is dangerous, that weight is lost slowly, that carbohydrates make fat and so on.

W.J.H. b.u.t.tERFIELD, later vice-chancel or of the University of Cambridge, introductory remarks to the first Symposium of the Obesity a.s.sociation of Great Britain, October 1968 It is incredible that in twentieth-century America a conscientious physician should have his hard-won professional reputation placed on the line for daring to suggest that an obesity victim might achieve some relief by cutting out sugars and starches.

ROBERT ATKINS, author of Dr. Atkins' Diet Revolution, testifying before Congress, April 12, 1973 THERE ARE TWO MOMENTS IN THE HISTORY OF George McGovern's Senate Select Committee on Nutrition and Human Needs when the competing paradigms of nutrition and obesity can be captured in the act of s.h.i.+fting-one coming, one going. The first was in April 1973, during a hearing that the committee held on the subject of obesity and fad diets. Appearing that day to testify were Robert Atkins-author of Dr. Atkins' Diet Revolution, a book that had already sold almost one mil ion copies in the six months since its publication-and three authorities in nutrition and health, who would testify that Atkins's severely carbohydrate-restricted diet was neither revolutionary, effective, nor safe. The tenor of the hearing was inquisitorial, and a pithy condemnation of Atkins and his diet by the Harvard nutritionist Fred Stare was read into the record by Senator Charles Percy of Il inois (Stare did not attend). "The Atkins diet is nonsense," Stare declared. "Any book that recommends unlimited amounts of meat, b.u.t.ter and eggs, as this does, in my opinion is dangerous. The author who makes the suggestion is guilty of malpractice."

A few weeks later, McGovern's committee hosted hearings on "Sugar in the Diet, Diabetes, and Heart Disease." Testimony came from an international panel of authorities, including Peter Cleave, Aharon Cohen of Hada.s.sah University in Jerusalem, George Campbel of the Durban Diabetes Study Program in South Africa, Peter Bennett of the NIH, and Walter Mertz of the U.S. Department of Agriculture. These investigators discussed the potential dangers of refined carbohydrates in the diet, and John Yudkin testified to the particular dangers of sugar. McGovern and his fel ow congressmen found the testimony compel ing, although difficult to reconcile with the growing acceptance, their own included, of the notion that it was fatty foods that caused heart disease, and carbohydrates that would prevent it.

Those on the committee saw no connection between the two sets of hearings. They believed Atkins was peddling dietary nonsense, whereas Cleave, Campbel , and the others were promoting reasonable science, albeit a minority viewpoint. The congressmen did not comprehend that both sets of hearings were about the role of refined and easily digestible carbohydrates and the damage they might cause. "We weren't thinking of those two things,"

said the committee staff director, Kenneth Schlossberg, looking back from a perspective of three decades, "which was not very bright."

Three years later, in July 1976, McGovern's committee returned to the subject of diet and disease in the hearings that would lead, a half year later stil , to the publication of Dietary Goals for the United States. The first witness was a.s.sistant Secretary of Health Theodore Cooper, who repeatedly emphasized the need for further research to establish reliable knowledge about the diet-disease connection. McGovern and his fel ow congressmen, however, wanted to tel the American public something more definitive, so McGovern asked Cooper if he could, at least, agree with the proposition that "overconsumption may be as serious a problem of nutrition as under-consumption."

"Particularly overconsumption of the wrong things," Cooper replied. "Very often in the poor we see people who are plump who might be cal ed obese, and people would then conclude that they do not have a deficiency because they look rotund, healthy in one sense of the word. But it is true that the consumption of high carbohydrate sources with the induction of obesity const.i.tutes a very serious public health problem in the underprivileged and economical y disadvantaged. I would agree with that."

This response seems clear enough: the overconsumption of "high carbohydrate sources"-a phrase used to describe carbohydrate-dense starches and refined carbohydrates rather than leafy green vegetables and fruits-was a.s.sociated with obesity in the poor, and perhaps even the cause.

McGovern then asked Cooper to provide a "general rule of thumb" about eating habits that would help prevent disease and lengthen our lives, and Cooper reluctantly agreed to do so.

"What kinds of foods in general should we be consuming less of and what should we be eating more of?" McGovern asked.

"I think what we need to consider doing is to reduce our total fat intake," Cooper replied. "Fat adds a caloric substance-almost twice as much-nine calories per gram-as compared to sugar. I think in order to have an effective reduction in weight and realignment of our composition we have to focus on reducing fat intake."

With that answer, Cooper had contradicted himself, and the conventional wisdom on diet and health in America had s.h.i.+fted. The problem was no longer overconsumption of high-carbohydrate sources, but the overconsumption of fatty foods. And if Cooper realized that reducing our total fat intake meant increasing our consumption of carbohydrates, he neglected to say so.

Between 1973 and the mid-1980s, the notion of the fattening carbohydrate, which had persisted in clinical and popular literature for wel over a century, was replaced with the belief that it is dietary fat, with its particularly dense calories, that is responsible for overweight and obesity. The prescription of reducing diets that restricted starches and sugars, and perhaps oils and b.u.t.ter as wel , was replaced with diets that targeted fat alone-restricting not just b.u.t.ter and oils but meat, eggs, and dairy products-thereby increasing carbohydrate consumption. Obesity was conceptual y transformed from a condition commonly a.s.sociated with the excessive consumption of carbohydrates and carbohydrate craving to one that would be described by prominent nutritionists as a "carbohydrate-deficiency syndrome," which in turn explained why "an increase in dietary carbohydrate content at the expense of fat is the appropriate dietary part of a therapeutical strategy."

What makes this s.h.i.+ft al the more perplexing is that it occurred immediately after the science of fat metabolism evolved to explain why carbohydrates were uniquely fattening, and it fol owed a six-year period in which carbohydrate-restricted diets achieved unprecedented credibility among clinicians. The latter coincided precisely with the genesis of obesity research as what would be considered a legitimate field of scientific study, a transformation marked by the increasingly frequent appearance of conferences and symposiums dedicated to reporting the latest findings in obesity research, al of which, through 1973, had been dominated by discussions of the peculiar efficacy of carbohydrate-restricted diets.

The first was hosted by the University of California, San Francisco, in December 1967. Among the dozen speakers was the veteran UC Berkeley nutritionist Samuel Lepkovsky, who used exactly the same logic as Alfred Pennington had in the 1950s to argue the biological rationale of carbohydrate restriction. "Positive caloric balance may be a result rather than a cause of the [obese] condition," Lepkovsky said. "It seems desirable in the treatment of obesity to direct efforts toward an increased utilization of fat. This effort can be made by restricting the intake of carbohydrates and increasing the ingestion of fat." The one presentation at the conference that was specifical y on the dietary treatment of obesity came from a team of U.S. Navy physicians, who had been prescribing an eight-hundred-to-one-thousand calorie "ketogenic" diet to overweight naval personnel. Their diet was 70 percent fat, 20 percent protein, and 10 percent carbohydrate, and it induced "significant weight loss" in al their patients. "Uniformly and without exception," they added, "patients who underwent dieting found that the satiety value of the ketogenic diet was far superior to that of a mixed or high-carbohydrate diet, even though the food selection was minimal...."

In 1968, the newly founded Obesity a.s.sociation of Great Britain hosted in London its first symposium on obesity. The presentations were dominated by investigators who believed in the fattening nature of carbohydrates and the efficacy of carbohydrate-restricted diets. These included John Yudkin and his col eague Stephen Szanto; W.J.H. b.u.t.terfield, who would later become vice-chancel or of the University of Cambridge; Alan Kekwick and Gaston Pawan of the University of London, who were primarily responsible for reviving the concept of Banting's diet in the U.K.; and Denis Craddock, a general pract.i.tioner and author of Obesity and Its Management, which would be published in 1969 and was one of at most two or three clinical guides to obesity treatment published in the U.K. in the 1960s or 1970s. As Craddock reported at the conference, he had recently completed a survey of a hundred pregnant patients, sixty of whom had begun to fatten excessively during the early months of their pregnancy. "This weight gain was control ed in most cases"-fifty-seven of the sixty-"simply by restricting carbohydrates in the diet," he said.

The conference had been organized by Alan Howard and his col eague Ian McLean Baird. Howard was a biochemist and pathologist at the University of Cambridge who would later become the founding editor, with George Bray, of the International Journal of Obesity. Howard had become interested in carbohydrate restriction because he had been twenty pounds overweight, had unsuccessful y dieted for years, then final y lost the weight and kept it off by avoiding flour, starches, and sweets. At the London conference, Howard reviewed the literature on carbohydrate restriction dating back to Banting and concluded that this was the only effective method to induce and maintain weight loss. "A common feature of al who have written on the subject," he said, is "that the patient's hunger is satisfied whilst on a diet high in carbohydrate of the same caloric value, patients complain of hunger."

After the London meeting, obesity conferences evolved from local to international affairs. The first was in Paris in 1971, hosted by European nutrition and dietetics a.s.sociations. Here the sole presentation on the dietary treatment of obesity was by a col aboration from the French National Inst.i.tute on Health and Medical Research (INSERM), which is the local counterpart of the NIH in the United States and the Medical Research Council in the United Kingdom. These INSERM investigators had prescribed diets of twelve to eighteen hundred calories to over a hundred obese patients, in either three or seven meals a day, and with varying amounts of carbohydrates. Weight loss increased, they reported, when the subjects divided their calories among seven meals, which served to moderate the insulin response. Moreover, "lowering the carbohydrate content of the diet increased the weight loss at both meal frequencies."

The next conference was hosted by the NIH in Bethesda, Maryland, in October 1973. Six of the presentations at this meeting discussed the treatment of obesity by methods other than drugs or surgery. Two were on physical activity, and neither reported any significant effect of exercise on body weight. Two addressed the benefits of behavioral modification on weight loss, and neither reported any significant benefit. Of the two remaining presentations, one was by Ernst Drenick of UCLA on prolonged fasting to treat obesity-"our experiences are disappointing," said Drenick-and the other was by Charlotte Young of Cornel on dietary treatments.

As Howard had in London, Young reviewed the hundred-year history of carbohydrate-restricted diets, including the research of Pennington and that of Margaret Ohlson and her own trials in the 1950s. Young then discussed her recent studies, in which she had put obese young men on eighteen-hundred-calorie diets with the protein content fixed at 460 calories (26 percent), but with varying proportions of fat and carbohydrates. Over the course of nine weeks, she reported, "weight loss, fat loss, and percent weight loss as fat appeared to be inversely related to the level of carbohydrate in the diets"-in other words, the fewer carbohydrates and the more fat in the diet, the greater the weight loss and the greater the fat loss. "No adequate explanation could be given for the differences in weight losses," she said. Al of the carbohydrate-restricted diets, she said, "gave excel ent clinical results as measured by freedom from hunger, al aying of excessive fatigue, satisfactory weight loss, suitability for long term weight reduction and subsequent weight control."

The last of these conferences to be held before the nutritional wisdom began to s.h.i.+ft definitively was in London in December 1973, just two months after the NIH meeting. This one was organized by Yudkin, and many of those giving presentations had also attended the NIH conference. Their presentations were similar, but here there was more of a tendency to implicate carbohydrates specifical y as the cause of obesity. Lester Salans and Edward Horton, both col aborators of Ethan Sims on his experimental obesity studies, discussed the effect of carbohydrates on hyperinsulinemia and the role of hyperinsulinemia in obesity. "It is clear that in both lean and obese subjects the carbohydrate content of the diet influences...insulin and glucose concentrations," Horton reported. He added that it was probably hyperinsulinemia that induced both obesity and insulin resistance. Yudkin then gave the only talk on dietary therapy, ent.i.tled "The Low-Carbohydrate Diet," noting that these diets are higher in vitamins and minerals than calorie-restricted diets, simply because the foods restricted-starches and sugars-have few or no vitamins and minerals. The diet wil "reduce superfluous adiposity," Yudkin said, "but it wil not need to be changed when this has been done.... The diet is intended as anew but permanent pattern of eating and not simply as a cure for obesity, to be abandoned when an acceptable loss of weight is achieved." Harry Keen, who was then at Guy's Hospital Medical School and would become one of the most influential diabetologists in the U.K.,*122 said the critical issue wasn't just obesity, but the chronic diseases that accompanied it. "With the chronical y failed case of obesity we are dealing with the wreckage of the situation," he said, so it was necessary to set "new patterns of body weight and body size, if we are going to make a serious attempt to reduce the frequency, for example, of atherosclerosis, of diabetes mel itus and of a number of other conditions." Keen and his col eagues had tested the viability of this goal, he reported, on a group of "ostensibly normal men in whom obesity is represented no more frequently than in the population at large." These men were instructed to restrict their carbohydrate intake to less than five hundred calories a day, but to continue eating protein and fat as desired. The result was an average weight loss of fourteen pounds, impressive because these individuals were not necessarily overweight to begin with. That weight loss had been maintained for almost five years. To those who might be pessimistic about the prevention of obesity and overweight in the public at large, Keen said, this result should be taken as "a word of rea.s.surance and optimism."

By 1972, The New York Times Natural Foods Dieting Book was offering both a low-calorie weight-loss plan, at a thousand calories a day, and a low-carbohydrate method. "You strictly curtail the amount of carbohydrates you eat daily," the book explained. "You eat, instead, foods in which the carbohydrate content is very low or nonexistent. Meat...fish, poultry, fats, b.u.t.ter, most cheeses and eggs are equal y low in that fattening substance, and these are the foods that form the basis for your diet...for without carbohydrates you cannot gain weight!"

Two years later, when the nonprofit organization Consumer Guide published its first edition of Rating the Diets, a 380-page compendium of the pros and cons of popular diets, carbohydrate restriction seemed firmly established in the canon. Rating the Diets, which obesity authorities would repeatedly recommend as a valuable review of the evidence, concluded that a diet including less than sixty grams of carbohydrates each day had "much to recommend it" and so was "helpful and beneficial" for weight loss. It also quoted a medical textbook to the effect that "the difficult-to-treat obese patient,"

which effectively means every obese patient, "appears to suffer from some defect in dealing with carbohydrate which leads to an unnatural conversion of it to fat and to storage of the fat. Avoidance of too much dietary carbohydrate reduces this tendency." The only caveat with these diets, according to Rating the Diets, was that they "pay little attention to the kinds of fats you eat" and so might increase heart-disease risk.

The s.h.i.+ft in the nutritional wisdom was now taking place, driven by the contagious effect of Ancel Keys's dietary-fat/heart-disease hypothesis on the closely related field of obesity. Any diet that al owed liberal fat consumption was to be considered unhealthy. Clinical investigators working on the problem of human obesity concurred.

Through the 1950s, the carbohydrate-restricted diet had chal enged only the positive-caloric-balance hypothesis of obesity. Yudkin had managed to reconcile carbohydrate restriction with this conventional wisdom by insisting that low-carbohydrate diets were low-calorie diets in disguise. By doing so, Yudkin made the diets political y acceptable, although he also directed attention away from the underlying science. In the same 1960 Lancet article in which Yudkin proclaimed what he cal ed "the inevitability of calories," he had made the point that if the diet was indeed low in calories, then its fat content would also be comparatively low, reconciling his diet with Keys's dietary-fat hypothesis. This was Yudkin's "no bread, no b.u.t.ter" argument. If carbohydrate calories are restricted, fat calories are, too. Though the proportion of fat in the diet increases if carbohydrates are avoided, the absolute quant.i.ty of fat may actual y decrease. This is why Yudkin insisted that the correct terminology for these diets should be "low-carbohydrate" rather than "high-fat." "It is highly implausible," Yudkin wrote in 1974, "that a given amount of fat that is harmless when energy intake is excessive becomes harmful when this excess is corrected by a reduction in the intake of sugar and starch."

As a result of Yudkin's conciliatory efforts, the only carbohydrate-restricted diets that elicited a backlash from nutritionists were those promoted by clinicians whose interpretation of the science disagreed with Yudkin's. This situation was exacerbated by the fact that it was these physicians, without university affiliations, who adopted the diet quickly and then wrote books for the lay public that sold exceptional y wel . Because their claims sounded like quackery-The High-Calorie Way to Stay Thin Forever , as Dr. Atkins' Diet Revolution was subt.i.tled-they were treated as such, and particularly so after the medical and public-health authorities decided that dietary fat might cause heart disease.

The smal contingent of influential nutritionists from Fred Stare's department at Harvard provide an example of how this process of entrenchment evolved. In 1952, when Alfred Pennington lectured at Harvard on the benefits of carbohydrate restriction and Keys was only beginning his crusade against dietary fat, Mark Hegsted had suggested, "Dr. Pennington may be on the right track in the practical treatment of obesity." A decade later, and a year after the American Heart a.s.sociation had official y sided with Keys, the Brooklyn obstetrician Herman Tal er published his best-sel er, Calories Don't Count, based on Pennington's work and Tal er's clinical experiences with the diet. Stare cal ed the book "trash," and Jean Mayer described the high-fat aspect of the diet as "potential y dangerous." Philip White, who received his doctorate in nutrition from Stare's department, then wrote a review of Calories Don't Count for JAMA, accusing Tal er of perpetrating "nutrition nonsense and food quackery." In 1973, in response to the publication of Dr. Atkins' Diet Revolution, based on Atkins's clinical experience with overweight patients and another decade of science, White edited a critique of carbohydrate-restricted diets in JAMA-the first draft of which was written by Ted Van Ital ie, another veteran of Stare's nutrition department-that now dismissed the diets as "bizarre concepts of nutrition and dieting [that] should not be promoted to the public as if they were established scientific principles."

Meanwhile, these nutritionists would readily admit that they didn't know what caused obesity (why some people ate too much and others didn't) and that calorie restriction conspicuously failed to cure it. After nearly twenty years in the field, as Jean Mayer wrote in the introduction to his 1968 monograph, Overweight, he was "as aware as any man of the gigantic gaps in our knowledge-and of the likelihood that many of our present concepts may be erroneous." He also noted, in his discussion of hormonal influences on obesity, that insulin "favors fat synthesis" and that someone who over-secretes insulin could "tend to become hungry as a result." But when a physician suggested publicly, as Atkins did, that carbohydrates raised insulin levels, that insulin favors fat synthesis, and that a diet lacking carbohydrates might reverse this process, these nutritionists would denounce it, as Mayer himself did in 1973, as "biochemical mumbo-jumbo."

With the publication of Dr. Atkins' Diet Revolution and its subsequent censure by the American Medical a.s.sociation, the nature of the professional discussions on carbohydrate-restricted diets turned from their clinical utility to the reasons to avoid them. The actual science suddenly mattered less than ever.

Atkins was a Cornel -trained cardiologist. Between 1959 and 1963, coinciding with the early years of his practice in Manhattan, he gained fifty pounds.

He eventual y decided to try carbohydrate restriction, he said, "because that's what was being taught at the time." His attempt coincided with the 1963 publication in JAMA of a lengthy article by the University of Wisconsin endocrinologist Edgar Gordon, ent.i.tled "A New Concept in the Treatment of Obesity." Gordon was one of the few clinicians of that era who studied fat metabolism and then designed a diet based specifical y on that science.

Gordon's diet, as described in JAMA, began with a forty-eight-hour fast-"not to produce a spectacular loss of weight, but rather to break a metabolic pattern of augmented lipogenesis"*123 -and then al owed protein and fat as desired but limited carbohydrates to minimal fruits, green vegetables, and a half-slice of bread every day. "The total caloric value is quite high in terms of reducing diets," wrote Gordon. Atkins later said his attention was caught by Gordon's observation that his subjects lost weight without ever complaining of hunger.

In his diet, Atkins replaced the two-day initiatory fast with a week or more of complete carbohydrate restriction, under the a.s.sumption, as the Atlanta physician Walter Bloom had noted, that the two states were physiological y identical. Atkins said he lost twenty-eight pounds in a month and felt energized in the process. In 1964, while Atkins was personal y reaping the benefits of his diet, he was also working part-time as a company physician with AT&T.

The junior executives noticed his weight loss, so he told them about the diet. Sixty-five of them eventual y tried it, as Atkins told it, and al but one reduced to their desired weight. The sole exception wanted to lose eighty pounds but lost only fifty.

Atkins then started treating obese patients out of his cardiology clinic and developed the diet as he came to prescribe it in his book. He instructed his patients to start off with an initiation period, eating no carbohydrates other than a smal green salad twice a day. Once they were losing weight at a suitably rapid rate, they could begin adding smal amounts of carbohydrates back into their diet until they reached what he cal ed the critical carbohydrate level, when their weight loss either leveled off or could no longer be maintained. Then they would have to back off again on the carbohydrates to experience further benefit from the diet. He also had them check their urine for ketone bodies-with the same ketosticks used commonly by diabetics-to ensure that they remained in ketosis and were stil burning body fat. The reliance on ketosis to initiate and maintain weight loss, and the progressive addition of carbohydrates to the diet, are what Atkins considered his contributions to the clinical science of carbohydrate restriction.*124 His career as a diet doctor grew slowly until 1966, when the women's fas.h.i.+on magazines began recommending his diet, and his business boomed. After Vogue popularized the diet in 1970, Atkins set out to write Diet Revolution, which was then advertised as "the famous Vogue superdiet explained in ful ."

The gist of Dr. Atkins' Diet Revolution can be distil ed down to three a.s.sertions. The first is that weight could be lost on his diet without hunger, and perhaps without even restricting calories. Atkins said that his patients regularly lost weight eating three thousand calories a day, and that he had one three-hundred-pounder who reduced significantly while eating five thousand. His only explanation was that obesity is caused by the kind of calories we consume and not the quant.i.ty, and so if we avoid carbohydrates our bodies function correctly and shed any excess weight. He attributed the absence of hunger to the copious calories, the ketosis (which is probably not the case), the effect of insulin on blood sugar-al overweight people "produce too much insulin," he wrote, and that lowers blood sugar and makes people hungry-and the secretion of what the British clinicians Alan Kekwick and Gaston Pawan had cal ed fat-mobilizing substance. (Virtual y al hormones, with the exception of insulin, wil mobilize fat from adipose tissue, but none of them wil do so effectively when insulin is elevated.) Atkins's second claim was that his diet was inherently healthy, much more so than a low-fat diet, because refined carbohydrates and starches, not saturated fat, caused heart disease and diabetes. Atkins later said that Peter Cleave's Saccharine Disease had been a revelation to him. In Diet Revolution he discussed the research from Yudkin, Margaret Albrink, Robert Stout, and Peter Kuo implicating triglycerides as a more significant risk factor for heart disease than cholesterol. He also claimed, on the basis of his experience with "ten thousand" overweight patients, that cholesterol "usual y goes down" on his diet, despite the high saturated-fat content, and that triglycerides invariably decrease.

His third claim was what he cal ed the "cruel hoax" of calorie-restricted diets: "The balanced low-calorie diet has been the medical fas.h.i.+on for so long that to suggest any alternative invites professional excommunication," Atkins wrote. "Yet even most doctors admit (at least privately!) the ineffectiveness of low-calorie diets-balanced or unbalanced." Atkins supported his accusation by invoking Albert Stunkard's 1959 "comprehensive review of the thirty years of medical literature," and offering three reasons why calorie-restricted diets inevitably fail. First, they "don't touch the primary cause of most overweight," which is a "disturbed carbohydrate metabolism." They also fail because they reduce energy expenditure. "Dr. George Bray," he wrote, "has demonstrated that people on low-calorie diets actually develop lower total body energy requirements and thus burn fewer calories." (Although Atkins didn't say so, this research had led Bray himself to publish an article ent.i.tled "The Myth of Diet in the Management of Obesity.") And, final y, Atkins wrote, "The main reason low-calorie diets fail in the long run is because you go hungry on them.... And while you may tolerate hunger for a short time, you can't tolerate hunger al your life."

Had Atkins wanted to avoid professional excommunication, he might have published something other than a polemic couched as a diet book. But he was feeling "resentment," he wrote, "that [he] had been duped so long by misinformation given me in the medical literature." The Diet Revolution was not just advocating a way to lose weight, which Atkins credited, in any case, to Banting, Pennington, Kekwick, and Pawan, but overthrowing the current nutritional wisdom entirely. Unlike Irwin Stil man, whose 1967 megabest-sel er The Doctor's Quick Weight Loss Diet was also based on carbohydrate restriction, Atkins wanted "a revolution, not just a diet." "Martin Luther King had a dream," Atkins wrote. "I, too, have one. I dream of a world where no one has to diet. A world where the fattening refined carbohydrates have been excluded from the diet." Atkins deliberately portrayed his diet as diametrical y opposed to the growing orthodoxy on the nature of a healthy diet. Whereas Keys had insisted that the solution to obesity was to convince fat people that overeating was a sin and overeating fat would kil them, Atkins said his patients lost "thirty, forty, 100 pounds" eating "lobster with b.u.t.ter sauce, steak with Bearnaise sauce...bacon cheeseburgers...." "As long as you don't take in carbohydrates," Atkins wrote, "you can eat any amount of this 'fattening'

food and it won't put a single ounce of fat on you."

Diet Revolution may have been, as its publisher claimed, the fastest-sel ing book in history. Nonetheless, its "chief consequence," as John Yudkin noted in 1974, may have been "to antagonize the medical and nutritional establishment." In fact, Atkins had to antagonize only a very smal and select group of men to have a profound and lasting effect on how we think about obesity and weight regulation. In obesity research, particularly in the United States in the 1970s, the established wisdom was determined not by any testing of hypotheses or even establis.h.i.+ng of consensus but by the judgment of fewer than a dozen men who dominated the field: Jean Mayer, Fred Stare, Jules Hirsch, George Bray, Theodore Van Ital ie, Albert Stunkard, George Cahil , Philip White, and perhaps a few others. (And when these men began to retire from the scene in the 1980s, their younger col eagues-Johanna Dwyer, who received her Ph.D. with Mayer; Francis Xavier Pi-Sunyer, who col aborated with Van Ital ie; Kel y Brownel , who worked and studied with Stunkard -a.s.sumed the leaders.h.i.+p and perpetuated their beliefs.) When these men came of age in their careers, in the 1950s and early 1960s, obesity research was a new and expanding field of science. It had been reinvented in the United States after World War I , and the National Inst.i.tutes of Health was just beginning to provide money for research. These men fil ed the expanding vacuum. They al came out of the Northeastern academic corridor-Harvard, Yale, Columbia, Rockefel er, the University of Pennsylvania -and they al knew each other. Van Ital ie befriended his cla.s.smate Stunkard on their first day of medical school at Columbia; he then went to work with Mayer at Harvard and enlisted Stunkard's help to test Mayer's theory of hunger, and so Stunkard got to know Mayer as wel . Philip White received his doctorate with Mayer at Harvard and remained in Stare's department until 1956, when he became secretary of the American Medical a.s.sociation's Councils on Foods and Nutrition and wrote an influential nutrition column for JAMA. Van Ital ie then became a member of White's council and initiated its 1973 public condemnation of Atkins and al similar carbohydrate-restricted diets. White edited the article. If you weren't in the club, you had little influence.

("The Mississippi River is very deep, or at least it used to be," is how the biochemist and diabetologist Gerald Grodsky of the University of California, San Francisco, described the inability of West Coast investigators to influence medical wisdom in the 1960s and 1970s.) These individuals became the field's "leading authorities," as the newspapers would cal them. They hosted the conferences, edited the textbooks, chaired the committees, and determined research priorities. By the end of the 1970s, they had determined what clinicians and researchers in the field would believe, at least in the United States, and what they stil believe overwhelmingly. When McGovern's committee held its post-facto hearings in February 1977 to address the Dietary Goals for Americans, only members of this club testified on obesity*125 (Mayer had been the committee's consultant), and they al embraced the committee's recommendation of a national diet richer in carbohydrates and poorer in fat. Although Van Ital ie also testified that he was unaware of any research to support their opinions: "Thus, what I am saying is an a.s.sumption rather than a statement of established fact," he acknowledged.

None of these authorities actual y specialized in the clinical treatment of obesity except Stunkard, who did so as a psychiatrist treating an eating disorder. Nor were they necessarily the best scientists in the field. Fred Stare and Philip White never studied obesity at al . Cahil 's research on fat metabolism and fuel part.i.tioning was seminal, but he didn't see why it should be relevant to human obesity. Stunkard's primary contribution to obesity research through the 1970s was his observation that the obese rarely lose weight on diets, and if they do, they don't keep it off. But he never noted, and as a result neither did anyone else, that the only dietary studies he addressed in his seminal a.n.a.lysis were of semi-starvation, so what he had confirmed was that semi-starvation failed, not that al diets did.

Van Ital ie and Bray deserve a disproportionate share of the responsibility for effectively removing the concept of the fattening carbohydrate from the nutritional canon, and thus the carbohydrate-restricted diet as wel . Virtual y everything we believe about what const.i.tutes an effective weight-loss diet can be traced back to the 1970s and the efforts of these two men.

Before Van Ital ie decided to write what he cal ed the AMA-sponsored "denunciation" of Atkins in 1973, his only substantive involvement in the science of obesity, as either a researcher or a clinician, was his work with Mayer twenty years earlier. In the intervening years, he had worked on intravenous feeding of hospital patients and dietary influences on cholesterol, among other subjects, but he returned to the subject of weight in 1971, only when one of his post-docs developed an interest in the subject. This led to what Van Ital ie considered his primary contribution to obesity research, the development of a feeding machine to study food intake: "You could basical y feed yourself by pus.h.i.+ng a b.u.t.ton," Van Ital ie explained. "The machine would deliver a measured quant.i.ty of formula diet into your mouth, and then keep a record of how much you took."

Van Ital ie felt that Diet Revolution was ful of what he cal ed "gross inaccuracies," and that there were far too many reasons to believe that the diet could be dangerous to disseminate it so widely. There may have been some personal enmity as wel : Van Ital ie had been chief of medicine at St. Luke's Hospital in New York when Atkins had served under him as a cardiology resident in the late 1950s. Van Ital ie said he didn't work with Atkins closely enough to know him personal y, but he did not find him "an appealing personality" nonetheless. Stunkard, talking about al his col eagues in the field, said, "We just despised [Atkins]. We thought he was a jerk, an idiot, who just wants to make money."

The critique that Van Ital ie drafted and White edited, which was then published as the official AMA declaration on carbohydrate-restricted diets, was not a balanced a.s.sessment of the science, nor was it absent its own gross inaccuracies. It was akin to the diatribes that had been aimed at Banting in the 1860s, Pennington in the 1950s, and Tal er, by White himself, in the early 1960s. Atkins, like Banting, Pennington, and Tal er, was censured for advocating a diet that was "neither new nor revolutionary." The article accused Diet Revolution of lacking "scientific merit," primarily by implying that here was a "way of circ.u.mventing the first law of thermodynamics." The diet itself was denounced as "grossly unbalanced," because it "interdicts the 45 percent of calories that is usual y consumed as carbohydrates," and so cannot "provide a practicable basis for long-term weight reduction or maintenance, i.e., a lifetime change in eating and exercise habits." That this was the opinion of a nutritionist and a physician, neither of whom had worked clinical y with obese patients, was lost in the publication of the critique under the auspices of the AMA itself. Mayer dedicated one of his syndicated newspaper columns to condemning Atkins, based on the AMA critique, which he cited repeatedly as though it were the considered opinion of the entire American Medical a.s.sociation and not just his former col aborator, edited by his former student. "The American Medical a.s.sociation," wrote Mayer, has "taken the unusual step of warning the U.S. public against the latest do-it-yourself diet as propounded in 'Dr. Atkins's Diet Revolution.'" The AMA report, Mayer wrote, "explains why the 'diet revolution' cannot work."

After Van Ital ie drafted the AMA attack on Atkins, he spent the next decade as the princ.i.p.al arbitrator of the risks and benefits of weight-reduction diets.

This coincided with his rise to prominence in the field, after receiving a 1974 NIH grant-"a few hundred thousand dol ars," he says, "which was a lot of money at the time"-to start the first federal y funded clinical obesity center in the United States, now known as the Theodore Van Ital ie Center for Weight Control at St. Luke's Roosevelt Hospital Center. That same year, Van Ital ie gave the review presentation on dietary approaches to obesity at the First International Congress on Obesity, although he had yet to do any research personal y on the dietary treatment of obesity, or to treat obese patients. In 1975, Van Ital ie cowrote (with Pi-Sunyer) the review chapter on obesity and diabetes in the textbook Diabetes Mellitus, having now been in the field, in a part-time capacity, for at most four years. He then wrote the chapters on dietary therapy for obesity in the 1978 symposium report Obesity: Basic Mechanisms and Treatment, edited by Stunkard; in the 1979 NIH report Obesity in America, edited by Bray; in Bray's 1980 textbook, Obesity: Comparative Methods of Weight Control; and in Stunkard's 1980 textbook, Obesity. In 1983, Van Ital ie cochaired the Fourth International Congress on Obesity. In 1984, he coauth.o.r.ed the obesity chapter in the fifth edition of Present Knowledge in Nutrition, which had been a standard nutrition reference since its first edition was published thirty years earlier. Because Van Ital ie was also engaged as chair of the medical department at Columbia University's Presbyterian Col ege of Physicians and Surgeons, he says, he had no time to do research himself, and relied almost entirely on his col aborators for the few studies he did publish.

Throughout this period, Van Ital ie's reviews of dietary therapy for obesity were singularly dedicated to dismissing any evidence that favored the use of carbohydrate-restricted diets. They would invariably begin with the declaration that carbohydrate-restricted diets were just another way to restrict calories, and they would proceed to refute claims made about the diets on the basis that these claims (not to be confused with observations of the diets' efficacy) had not been established beyond reasonable doubt. By the end of these reviews, Van Ital ie would promote the continued treatment of obesity by balanced, calorie-restricted diets, while acknowledging that there was "increasing recognition of [their] ineffectiveness."*126 He would reject any suggestion that carbohydrate-restricted diets should be tried instead, while simultaneously acknowledging that these diets were "quite popular and have been fol owed with varying degrees of success by many dieters."

George Bray's influence in removing the fattening carbohydrate and carbohydrate-restricted diets from the nutritional wisdom was more subtle than Van Ital ie's, but may have been ultimately more significant. Bray was a graduate of Harvard Medical School. In the late 1960s, he studied animal models of obesity at UCLA's Harbor General Hospital in Torrance, California. He also col aborated peripheral y with Ethan Sims on his experimental obesity studies (Bray had been a medical-school cla.s.smate of Sims's col eague Ed Horton) and had notable disagreements with Sims about how this research should be interpreted. In 1973, Bray cochaired the NIH's first obesity conference; he then edited and drafted the subsequent NIH report, Obesity in Perspective.

In 1977, he chaired the Second International Congress on Obesity and a second NIH conference on obesity. He then edited the NIH report Obesity in America, which was published in 1979. Meanwhile, he edited or wrote three of the half-dozen textbooks or clinical manuals on obesity that were published in the United States during the decade-Treatment and Management of Obesity (1974), The Obese Patient (1976), and Obesity: Comparative Methods of Weight Control (1980)-which means effectively al of those not edited or written by Stunkard.127 Bray believed that al diets worked by restricting calories, and since restricting calories eventual y failed, nothing else need be discussed. He dismissed as irrelevant the work of those investigators who did actively study the dietary treatment of obesity, like Charlotte Young, who gave the presentation on dietary therapy at the NIH conference on obesity that Bray organized and chaired in 1973. Young specialized in the study of body composition, and she had been studying diets and obesity at Cornel since 1950. In the official NIH report on the conference, Obesity in Perspective.

Bray treated her discussion of carbohydrate-restricted diets as naive and of no consequence. In the book he coedited the year after the conference, Treatment and Management of Obesity, Young's observations on carbohydrate-restricted diets are described as stil requiring further "confirmation before they can be ful y accepted.... The question of the value of a low carbohydrate diet and its effectiveness in weight loss is stil unresolved." In The Obese Patient, published three years after the NIH conference, Bray wrote of Young's studies, "The data are suggestive and require careful replication with larger groups of individuals." Yet nowhere in the NIH report on the conference, including a lengthy list of research priorities and "gaps in our current knowledge," did Bray raise the possibility that further research was needed on any dietary therapy for obesity, let alone, as Bray's own textbooks had suggested, the unresolved question of the value of carbohydrate restriction. Bray then proceeded to become the leading proponent of the hypothesis that obesity, like heart disease, was caused primarily by the dense calories of dietary fat, and thus could be cured or prevented by replacing the fat in the diet with carbohydrates.

The dissociation of the science of fat metabolism from any discussions of the cause or treatment of obesity was particularly conspicuous throughout this era and could be considered its legacy. When Bray, Van Ital ie, Cahil , and Hirsch gave review talks at these conferences, as they did throughout this period, they would raise the issue of carbohydrate-restricted diets only to refute the claims that such diets offered a metabolic advantage over low-calorie diets. They would omit any mention of research that might explain the reported efficacy of the diets, even when that research was discussed at the same conferences and by investigators they knew personal y. In 1977, for instance, Donald Novin, director of the Brain Research Inst.i.tute at UCLA, discussed what he cal ed the "carbohydrate hypothesis of ingestive behavior" at Bray's Second International Congress on Obesity. Novin suggested that the "widespread popularity of the low carbohydrate diets" could be explained by the effect of carbohydrates on insulin, and then of insulin on fat deposition and thus hunger. Bray, who had worked closely with Novin at UCLA, gave the summary talk at the conference on obesity therapies and omitted mention of Novin's hypothesis.*128 When M. R. C. Greenwood discussed the effect of insulin on the enzyme lipoprotein lipase, LPL, the "gatekeeper" for fat acc.u.mulation in cel s, at the Fourth International Congress on Obesity, Hirsch ignored the implications in his review of dietary therapy, even though Greenwood had received her doctoral degree with Hirsch.

In retrospect, the influential figures in the clinical investigation of human obesity in the 1970s can be divided into two groups. There were those who believed carbohydrate-restricted diets were the only efficacious means of weight control-Denis Craddock, Robert Kemp, John Yudkin, Alan Howard, and Ian McLean Baird in England, and Bruce Bistrian and George Blackburn in the U.S.-and wrote books to that effect, or developed variations on these diets with which they could treat patients. These men invariably struggled to maintain credibility. Then there were those who refused to accept that carbohydrate restriction offered anything more than calorie restriction in disguise-Bray, Van Ital ie, Cahil , Hirsch, and their fel ow club members. These men rarely if ever treated obese patients themselves, and they repeatedly suggested that since no diet worked nothing was to be learned by studying diets.

Bray would routinely equate the carbohydrate-restricted diet to every fad diet that came along-the grapefruit diet, the banana diet, the ice-cream diet.

But when he testified before McGovern's subcommittee in 1977 and described McGovern's Dietary Goals of a carbohydrate-rich diet for the entire nation as "highly commendable," he also submitted as part of his testimony a two-hundred-page report by the British Medical Research Council ent.i.tled Research on Obesity, apparently ignoring the fact that the report contradicted his own testimony. Published the same year, it referred to carbohydrate restriction as the diet "commonly prescribed by general pract.i.tioners" and considered it more effective and certainly more worthy of discussion than the prescription of diets that depended on restricting calories. The report also noted that the best weight-reduction results on record were those reported by Robert Kemp and Denis Craddock, both British pract.i.tioners who prescribed carbohydrate-restricted diets to their patients and published their results, Kemp in the medical journals and Craddock in Obesity and Its Management.

When a new diet book was published every few years touting yet another physician's variation on carbohydrate restriction, it was treated by Bray and his col eagues as the ultimate evidence that the diet itself didn't work. "If such diets are truly successful," asked Van Ital ie in his AMA denunciation of Atkins, "why then, do they fade into obscurity within a relatively short period of time only to be resurrected some years later in slightly different guise and under new sponsors.h.i.+p. Moreover, despite the claims of universal and painless success for such diets, no nationwide decrease in obesity has been reported." Of course, the efficacy of the diet could explain the continued popularity of such books. The diet had survived more or less continuously for over a century and had certainly thrived since the end of World War I . That the medical and nutrition establishments refused to take it seriously, and had even taken to advocating carbohydrate-rich diets instead, could explain the continued high prevalence of obesity.

This nihilistic argument became a mantra. "The evergrowing list of diets are an affirmation of the fact that no diet yet described is by itself a solution to the problem of obesity," Bray said in his 1977 testimony to McGovern's Senate committee. When Hirsch gave the review talk on obesity treatments at the Fourth International Congress on Obesity in 1981, he said: "The proliferation and seemingly endless concern with diets for the treatment of obesity suggests that this search is more motivated by financial rewards for the promoters rather than by an earnest desire to provide healthy and safe diets."

This theme of financial rewards for the promoters of these diets would also be echoed repeatedly. A "common factor of reducing regimens is their commercialism-someone stands to make money from their promotion," wrote George Mann, another veteran of Stare's nutrition department, in The New England Journal of Medicine in 1974. This didn't explain those like Pennington, Ohlson, Young, Gordon, or Kekwick and Pawan, who never wrote popular diet books and advocated similar advice to their obese patients, but it was an easy way to dismiss those like Atkins and Tal er who did.*129 They were "instant monetary nutritionists," wrote Stare, who liked to point out that Atkins made over $1 mil ion in one year from Diet Revolution, while simultaneously treating five hundred patients weekly in his "very lucrative private medical practice."

But this conflict-of-interest accusation, as we've discussed, often cuts both ways. Stare and his Harvard col eagues played the decisive role in ensuring that anyone who claimed that carbohydrates were uniquely fattening would carry the taint of quackery. When White, Mayer, and Stare publicly condemned Herman Tal er's Calories Don't Count it was a year after the Harvard nutrition department broke ground on a new $5 mil ion building that was paid for largely through private donations. What Stare cal ed the "lead gift" of $1,026,000 came from the General Foods Corporation, the maker of the very carbohydrate-rich Post cereals, Kool-Aid, and Tang breakfast drink. Over the next decade, Stare became the most public defender of sugar*130 and additives in modern diets, while his department continued to receive significant funding from the sugar industry; from Oscar Mayer, the maker of hot dogs; from Coca-Cola and the National Soft Drinks a.s.sociation. Would the resident nutritionists in Stare's department have been more accepting of the efficacy of a diet that restricted refined carbohydrates and sugars if the money had come from another source? If so, would this have effected how other clinical investigators in the field came to interpret the controversy?

The funding of research projects, laboratories, and entire academic centers by the food and pharmaceutical industries is now a fact of life in modern medical research, which is why many journals require that their authors declare potential conflicts of interest. But it raises important questions, just the same. When Science dedicated special issues to obesity research in 1998 and again in 2003, James Hil from the University of Colorado was selected both times to write the review article on diet and lifestyle factors that influence weight gain. In those articles Hil argued that pa.s.sive overeating and sedentary behavior were the causes of obesity, and he recommended reducing fat in the diet. Hil had long been a defender of the role of carbohydrates and particularly sugar in weight regulation. He even wrote an article, paid for by the Sugar a.s.sociation, promoting the use of sugar in weight-loss diets, under the a.s.sumption that a high-carbohydrate diet, even if loaded with sugar, would "reduce the likelihood of overeating, rather than increasing it, as some popular diet theories purport." ("The theory that dietary sugar equals high insulin levels equals excess fat deposits is unproven and makes little biological sense," Hil wrote.) Over the years, as Hil has acknowledged in his conflict-of-interest statements, he has also received consulting fees from Coca-Cola, Kraft Foods, and Mars (makers of Snickers, M&M's and Mars Bars), companies that would stand to suffer significant setbacks if the notion of the fattening carbohydrate was inst.i.tutionalized as a fact of science. He has also received over $2 mil ion in what are technical y termed "gifts" to his laboratory from Procter & Gamble, the maker of the fat subst.i.tute olestra, which has been described in the press as potential y a "dieter's dream."

Olestra's only reason for existence is that it wil al egedly help us manage our weight by replacing fat in the diet and making it easier for us to consume a low-fat, low-calorie diet. If carbohydrates are the fattening nutrients in human diets rather than fat or al calories, as Atkins suggested, then these diets have no role in weight loss or weight regulation, and olestra's rationale vanishes.

If the study of weight regulation were a legal issue, rather than a medical and scientific one, the support from Procter & Gamble would have been considered reason enough for Hil to recuse himself from any discussions of the dietary treatment of obesity or partic.i.p.ation in any dietary trials that might directly influence Procter & Gamble's profitability, and thus perhaps Hil 's interests.

In 2002 and 2003, Hil also received over $300,000 a year from the NIH to do a clinical trial testing the Atkins diet against a low-calorie, low-fat diet and, by implication, the justification for olestra as a fat subst.i.tute in a weight-reduction diet. And Hil was one of three princ.i.p.al investigators in the fol ow-up trial of the Atkins diet, for which the NIH provided $5 mil ion. The salient question is whether Hil and the other academics in this pursuit are any less open to having their interpretation of the evidence influenced by financial considerations than Atkins or Tal er or any of the other diet-book authors.

"A resolution of the very controversial question of the efficacy of low carbohydrate diets has great practical and theoretical significance," wrote Donald Novin of UCLA in 1978. Because a generation of obesity authorities were determined to dismiss the practical significance of carbohydrate-restricted diets, they dismissed the potential theoretical significance at the same time. Obesity researchers today say they stil have no hypothesis of weight regulation that can explain obesity and leanness, let alone account for a century of paradoxical observations. They insist that obesity is inevitably caused by overeating and thus consuming more calories than we expend, but when asked what causes someone to overeat, they have no answer. Yet the research on insulin and fat metabolism offers one, and it has for several decades.

Chapter Twenty-four.

THE CARBOHYDRATE HYPOTHESIS, III: HUNGER AND SATIETY.

There is only one way to lose weight, and that is to grow accustomed to feeling hungry. This simple fact, known to most people in affluent countries, seems somehow lost on the authors of the diet, weight-loss, and exercise books that find their lucrative way through the drugstore book racks. Two questions, then: Why do they fail to mention it? And why is it so?

Emory University anthropologist Melvin Konner, The Tangled Wing, 2003 IN 1975, THE DUKE UNIVERSITY PEDIATRICIAN James Sidbury, Jr., described a "rational basis" for the dietary treatment of childhood obesity, one that would neither torment his young patients with hunger nor rely on pharmaceutical means to prevent it. Such a diet, he wrote, would induce weight loss with a "minimum of anguish and struggle." Sidbury had an advantage over other investigators treating obese patients in that he had spent his career studying disorders of carbohydrate metabolism and, indeed, had already earned international renown for his development of a diet, stil used today, to treat what are cal ed glycogen storage diseases. The same year Sidbury published his description of a "Program for Weight Reduction in Children," however, he left his clinic at Duke