Michael Fumento's article "Big Fat Fake" in the March issue of Reason led Gary Taubes to make the following response. Taubes is the author of the New York Times Magazine story "What if It's All Been a Big Fat Lie?," which Fumento examined in his Reason story. To read Fumento's reply to Taubes, click here.
To the editors:
I am ambivalent about writing this response to Michael Fumento's article ("Big Fat Fake"). On the one hand, the article simply doesn't deserve a response. It is a noteworthy exercise in vitriol, and perhaps self-aggrandizement, but it falls far short of legitimate journalism. On the other hand, journalists and historians, not to mention the occasional lay reader, have a tendency to assume that if something makes it into publication it is somehow de facto true or justifiable. This is never necessarily the case. For that reason, which I find slightly more persuasive, a published response might mitigate that tendency toward excessive credulity, at least in this particular circumstance.
Fumento's article attacks my work and my credibility, and then tries to sell it as a commentary on the state of science and medical journalism. His attempt might have been compelling had he managed to get at least a small percentage of his facts right and to avoid journalistic sins of omission and commission worse than any of which he accuses me. To put it simply, even on those rare and splendid occasions when Fumento does get a fact right, he still manages to thoroughly misrepresent my article and mangle the interpretation of the relevant science. While it's effectively impossible, even in the copious space I've taken, to rectify all Fumento's excessive distortions, the following attempts to clarify the key issues and correct some of the more egregious errors.
For starters, in his second paragraph, Fumento characterizes my article as arguing "that the consumption of too little fat [Fumento's emphasis] could explain the explosion in obesity." He does not quote the article, which would have been easy to do had it included such a declaration anywhere in its nearly 8,000 words, but it doesn't. Rather my article challenged the accepted dogma that obesity and excess weight are caused by the excessive consumption of fat calories, and instead suggested that it was caused by the excessive consumption of calories from refined carbohydrates and starches. I referred to this proposition repeatedly as the "alternative hypothesis", using the word "hypothesis" to imply strongly that it is not a fact but a supposition that should be rigorously tested. The article discussed the possibility that refined carbohydrates and starches might have a unique effect on our metabolism that either causes excessive hunger or an unbalanced deposition of calories in fat tissue. If so, it suggested, such a metabolic effect could explain the 150-year-old popularity of low carbohydrate diets for weight loss.
Dr. Robert Atkins and his eponymous diet played a major role in the article because Atkins has been preaching the evils of carbohydrates for at least 30 years. Only recently, however, have mainstream medical researchers concluded that perhaps his very-low-carbohydrate diet is worth testing. These trials, as a result, might shed light on whether the alternative hypothesis is scientifically meaningful.
Among Fumento's primary criticisms are that I only cite individuals who support my thesis and disregard those who don't. He ignores the fact that because the article challenges the accepted dietary dogma, and indeed acknowledges that challenge in the opening paragraph, by definition it implies that considerably more than half of all researchers and administrators believe the accepted dogma. If they didn't, it wouldn't be dogma. The point of writing the article was precisely to note that some equally respectable scientists question this dogma, and then to explain why. I defined the state of the argument as having undergone "a subtle shift in the scientific consensus" over the past five years: "It used to be that even considering the possibility of the alternative hypothesis, let alone researching it, was tantamount to quackery by association. Now a small but growing minority of establishment researchers have come to take seriously what the low-carb diet doctors have said all along." The phrase "small but growing minority" implies a large, albeit perhaps shrinking majority that will disagree with what I say. The obvious point is that this majority has gotten plenty of space to air their views over the decades. They didn't need my help.
Moreover, I interviewed close to 100 researchers for The New York Times Magazine article, to go with 150 or so for its March 2001 predecessor in the journal Science ("The Soft Science of Dietary Fat"). I quoted or attributed information to two dozen of them. Once again, as in any good work of journalism, the opinions of the great majority of those interviewed were left out–up to 90 percent in my case, depending on how you want to do the calculation. One hopes this is true of Fumento's research as well. If he is implying otherwise, then he only interviewed a dozen people for his story, which is woefully insufficient for such a complex and controversial subject.
If Fumento did interview more, then it's conceivable he omitted the comments, for instance, of those who believed that my article had merit. And then he was undeniably selective about which opinions he would publicly embrace in support of his thesis from those researchers he did interview. For instance, he first cites Harvard's Walter Willett chastising me for neglecting his anxieties about red meat and colon and prostate cancer. But then Fumento characterizes me as "clipping the data" for saying "that the percentage [Fumento's emphasis] of fat in the American diet has been decreasing for two decades," when Fumento thinks it would be more relevant to discuss the absolute number of fat calories consumed. Yet Willett is perhaps the most outspoken proponent of the idea that total fat calories are irrelevant to the obesity epidemic, a point he makes in print several times a year. A recent example was an article in the American Journal of Medicine just last December, co-authored with Columbia University researcher Rudy Leibel, in which Willett and Leibel phrase the point almost exactly as I did. "Moreover, within the United States," they wrote, "a substantial decline in the percentage of energy from fat [my emphasis] during the last 2 decades has corresponded with a massive increase in the prevalence of obesity."
As for Willett's red meat/cancer anxieties, which he did indeed reiterate to me numerous times, Willett himself acknowledges that the data are ambiguous. Willett's own Nurses' Health Study revealed an elevated risk of colorectal cancer in women who ate red meat frequently, but the Nurses' Health Study has recently arrived at the wrong answer on several major health issues–most notably, the effects of post-menopausal hormone replacement therapy–and so its credibility is debatable. Moreover, Willett played a major role in preparation of a 1997 report published by the World Cancer Research Fund and the American Institute for Cancer Research. That report noted that of seven studies similar to Willett's, three, including Willett's, saw an association between red meat and colorectal cancer, while the other four did not. As for prostate cancer, the authors of the report could find neither "convincing" nor even "probable" reason to believe that diets high in red meat increase risk. I could have mentioned this but, like Fumento, I was working with limited space and chose to use what seemed most relevant.
Fumento next accuses me of tricking Stanford University researcher John Farquhar into seeming to support the Atkins diet and he quotes an infuriated Gerald Reaven, also of Stanford, calling my article "outrageous" and saying that I set him up.
For starters, Reaven was not quoted in the article, a fact that he and Fumento apparently consider irrelevant. Reaven's name and research were mentioned in the context of two paragraphs on the history of Syndrome X that have precisely zero to say about the Atkins's diet and sit over 3000 words and 24 paragraphs after one discussion of the Atkins diet and 1500 words and 18 paragraphs before the next. When I interviewed Reaven last year, however, he did say the following about Atkins's diet, on tape, on the record, and I trust he won't mind me repeating it: "I think it's a great way to lose weight. That's not the issue." The issue, he said, was whether it was safe for long-term weight maintenance, which he doubted. Reaven believes saturated fat should be avoided, as well as carbohydrates. Atkins only advocates avoiding the latter.
As for Farquhar, if Fumento's reporting is accurate, then he would like an apology for how I used him in the article. Fumento reports it this way:
"I was greatly offended by how Gary Taubes tricked us all into coming across as supporters of the Atkins diet," he wrote in an e-mail he broadcast to reporters and to colleagues who were stunned that Farquhar might actually hold the beliefs Taubes attributed to him. "We are against the Atkins Diet," he wrote, speaking for himself and Reaven. "I told him [Taubes] there is the minor degree of merit" to the idea that "people are getting fatter because too much emphasis is being placed on just cutting fats," Farquhar told me. But "once I gave him that opening—bingo—he was off and running, even though I said about six times that this is not the cause of the obesity epidemic."
It is conceivable, however, that Farquhar's memory on this issue is not up to snuff. The relevant interview, in this case, occurred through e-mail and so relying on his memory is unnecessary.
Here's the specific context: Farquhar is quoted only in the last paragraph of my story. It follows directly from a discussion of my own difficulty in accepting the seemingly counter-intuitive possibility that fat might be beneficial to one's health and weight, and carbohydrates detrimental. The story then ends with the Farquhar paragraph:
This is the state of mind I imagine that mainstream nutritionists, researchers and physicians must inevitably take to the fat-versus-carbohydrate controversy. They may come around, but the evidence will have to be exceptionally compelling. Although this kind of conversion may be happening at the moment to John Farquhar, who is a professor of health research and policy at Stanford University and has worked in this field for more than 40 years. When I interviewed Farquhar in April, he explained why low-fat diets might lead to weight gain and low-carbohydrate diets might lead to weight loss, but he made me promise not to say he believed they did. He attributed the cause of the obesity epidemic to the "force-feeding of a nation." Three weeks later, after reading an article on Endocrinology 101 by David Ludwig in the Journal of the American Medical Association, he sent me an e-mail message asking the not-entirely-rhetorical question, "Can we get the low-fat proponents to apologize?"
I had interviewed Farquhar over the telephone on April 25, 2002. Two and half weeks later, we had the following aforementioned e-mail exchange: Farquhar initiates the exchange with his e-mail using the "apology" line.
From: Dr. John Farquhar
Sent: Friday, May 10, 2002 7:55 PM
Subject: article on glycemic index
dear mr taubes
you may find a recent article in jama of interest in a search for blood sugar, a compensating increase in insulin, and an increase in
"hunger". the article in question is by ludwig— jama 2002;287:2414-2423 you recall that i believed, in contrast to reaven perhaps, that the blood sugar swings could contribute to post-prandial hunger and,
thus, could be a factor that would contribute to obesity. (see page 2417 of ludwig's article). he adds some fancy biochemistry that you may find interesting. can we get the low-fat proponents to apologize? hope the article is coming along well. regards, jack farquhar
I then replied to Farquhar, asking him if he meant what he said, and whether he minded if I used the line in question.
Thanks for the note. I'm featuring Ludwig prominently in the article. My idea is to portray the glycemic index and low carb diets docs as promoters of the alternative hypothesis to the low fat dogma. And, of course, one implication of the alternative hypothesis is that those low-fat proponents are at least part of the reason for the obesity epidemic. So were you just having fun with me when you wrote that maybe they should apologize, or do you, at least on occasions, wonder if it's true?
All the best,
Then Farquhar responded. Readers can judge for themselves whether Farquhar's take on the exchange and how I used it is justified:
From: Dr. John W. Farquhar
To: gary taubes
Sent: Sunday, May 12, 2002 8:17 PM
Subject: Re: article on glycemic index
dear gary–yes, i do think that they should apologize, but i don't expect it–so, that part is kidding. there is a long trail of nay-sayers in the CV medicine area. you might consider doing an article on sodium at some point.
i hope that i did make it clear that i believe the "low fat is good, therefore lower fat is better" crowd is dead wrong on many fronts–including that it is a good way to lose weight.
i know that low fat diets are really very bad for the overall type of lipid pattern that emerges. incidentally, ludwig didn't give that sufficient prominence. i believe that this more harmful lipid pattern accounts for the major disadvantage of low-fat diets (namely, that triglycerides rise, bringing HDL cholesterol down in the process and creating the "small dense" type of LDL that is more harmful than the larger varieties).
regarding the possibility that low fat diets contribute to the obesity epidemic–i am remaining a bit cagey on that point. i think that it might contribute, given the swings of glucose and insulin–and the possibility that an increase in hunger occurs on the tail end (when the insulin overcompensates and the blood sugar falls—or, at least, starts to fall and a compensating drive to snack keeps it from falling, assisted by a bit of catecholamine release). i thought that ludwig gave some support for that notion. in your interview with me i believe i emphasized that the portion size issue was staring us in the face and that the passive response of eating what is in front of us certainly gives the average person more calories than they would choose on their own. all the stuff about soft drink sizes going up, etc, etc.
of potential interest to you is that the famous dean ornish was recently not as dogmatic as i expected. i had to miss the session, held at stanford last saturday, because of my medical school class reunion. the session had reaven and ornish debating (as well as other speakers on other topics). i was told that ornish agreed that low fat diets make triglycerides rise, (a remarkable admission!!!)–he said his purpose of advocating such diets was to assist in weight loss (i need to confirm this point with jerry reaven), and to find an easy way to cut down on saturated fats. jerry and i would challenge the assertion that low fat diets work on weight loss better than any type of lowered calorie diets—this of course begs the question of whether these low fat diets are less successful than others.
Moving on, we get to Fumento's disagreement with my use of percentage of fat as a legitimate variable in the obesity epidemic. He says it's true that the percentage is decreasing, as I reported, but irrelevant: "The amount of fat consumed has been steadily climbing, as has consumption of all calories. Individual caloric consumption jumped from 3,300 calories per day in 1970-79 to 3,900 in 1997, an 18 percent increase. Per-person consumption of fat grams increased from 149 to 156, a 4.5 percent increase."
Fumento's source for these numbers, although he doesn't cite it, is the U.S. Department of Agriculture. The USDA has a variety of mechanisms for estimating macronutrient intake–i.e., protein, carbohydrates and fat–and has published a variety of reports on the subject. For instance, in April 1998, the USDA published an article entitled "Is Total Fat Consumption Really Decreasing?" This article reported that average total fat consumption for men aged 19 to 50, for instance, dropped from 113 grams per day in 1977-78 to 96 in 1989, the period that encompasses the beginning of the obesity epidemic. In 19- to 50-year-old women, the relevant numbers are 73 grams of fat per day in 1977-1978 and 62 grams in 1989.
Fumento prefers to use what are known as food availability data, as do I in most circumstances, although not this one. The USDA calculates how much food is provided by industry, and then adds imports, subtracts exports, and tries to adjust for waste–i.e., how much food is eventually thrown out. Fumento's data could have come from any number of these USDA reports but one that provides the same numbers is "Nutrient Content of the U.S. Food Supply, 1909-1997," by S. Gerrior and L. Bente of the Center for Nutrition Policy and Promotion (Home Economic Research Report No. 54). On page 26, Gerrior and Bente report that fat consumption from red meat, butter, lard and milk products all declined during the 1970s and 1980s. Thus the 7 gram per day, 63 calorie increase in total fat consumption noted by Fumento, was due to "the greatly expanded use of fried foods by the fast-food industry and in food service outlets as well as the increased use of salad oils on salads consumed both at home and away."
While the increase in fried foods might seem like a bad thing, the increase in oils on salad dressing would seem to be a good thing. The point, however, is that this particular data measures the availability of these fats in the food supply and, as Gerrior and Bente then emphasize, this particular category–cooking and salad oils–is almost impossible to estimate accurately:
While food supply estimates reflect trends in the availability of fats and oils for human food, they have never accurately measured the amount of food eaten because the portion of food wasted or discarded is difficult to determine. With the growth of the fast-food industry in the past three decades, it has become even more difficult to estimate the waste portion or discard of deep-frying fats. Since this discard is not available for human consumption, these estimates are limited as indicators of actual intake. A 1993 study estimated that about 50 percent or more of deep-frying fat used in food service operations is discarded after use and is not available for consumption. Reliable estimates of total fats and oils are difficult to determine partly because the actual amount of frying fat discarded by food service operations, particularly fast-food restaurants, varies with the type of the establishment.
As a result, the numbers Fumento uses to bolster this point are the least accurate available. It is the primary reason that I use percentage of fat calories, and may be why Willet does as well. It's safe to say that percentage of fat calories has been decreasing; it's a dubious proposition to make that claim about the total fat consumption.
More interesting, however, is what Fumento skates over in his quickie analysis: those extra 600 calories per day. Using the same report and this time taking Fumento's 1978 to 1997 as the period of interest, the American food supply, minus estimated wastage, offered up 448 more calories of carbohydrates, 48 more calories of protein and 63 more calories of fat. Fumento might want to blame the obesity epidemic on the extra 63 fat calories of fat or even the extra 48 protein calories, but he'll have to ignore the 448 calories from carbohydrates to do it. Under the circumstances, what I said in my article seems like a reasonable assessment: "If these trends are correct, then the obesity epidemic can certainly be explained by Americans' eating more calories than ever—excess calories, after all, are what causes us to gain weight—and, specifically, more carbohydrates. The question is why?"
Fumento then goes on to accuse me of shoving aside "decades of published, controlled randomized clinical trials comparing nutrient intake and weight loss." Regrettably, Fumento seems unclear on the concept of a "controlled randomized clinical trial," an unfortunate failing for a wannabe medical journalist. The relevant clue and the salient point is that there were no such trials. There are decades of trials looking at the effects of diet on weight loss, but they are for the most part, neither randomized nor controlled, and there are decades of observational studies trying to compare what people say they eat to how much they weigh. The latter are meaningless in this context and would require more time than I prefer to spend to explain why.
What Fumento does is turn for support to an April 2002 article in the Journal of the American Dietetic Association <(JADA) that was, he says, "`a review of all studies identified' that looked at diet nutrient composition and weight loss. It found over 200." It is this article he cites as one leg of a three-legged "crushing" mass of evidence rejecting the efficacy of Atkins-like low carbohydrate diets. On his own Web site, Fumento refers to this article repeatedly as evidence that he has done his library research, as though he himself personally read all 200 articles.
If Fumento even read this particular review article, however, he does a poor job of demonstrating that fact. For starters, it was not published in April 2002, but in April 2001. Secondly, while the authors of the study refer to the "more than 200 individual studies" that they allegedly included, they reference only 58, thus begging the question of what happened to the other 142+. Thirdly, the study was written by 4 employees and former acting undersecretary of the USDA, an organization that has been, bar none, the foremost advocate of high-carb, low-fat diets in America for 25 years.
This raises the issue of whether the authors might be tempted to bias their review to support the USDA's long-standing public position on the health-benefits of fat-reduced diets. It doesn't suggest they did, but it does beg the question of why Fumento readily intimates that a researcher who received funding for a diet trial from Atkins's Foundation would assuredly slant his results to please Atkins, while Fumento treats these government employees as somehow inherently beyond suspicion of the same kind of bias. Who knows. Maybe they might be motivated to please the bureaucrats who pay their salaries? Maybe their goal, consciously or subconsciously, was to get an article into print that appeared to support the agency's ubiquitous low-fat diet advice? Either way, it seems only fair that Fumento hold government employees up to the same level of malicious insinuation to which he holds private citizens. Why he doesn't escapes me.
Finally, biased or not, the USDA authors do happen to come to conclusions, which Fumento promptly quotes, that are not supported by the data from the studies they reference. In some points they simply err on the side of low-fat diets and against the low-carb diets. For instance, the authors state that of the 22 low carbohydrate diet trials they included, "there is a pattern of weight loss ranging from -2.8 to -12.0 kg." This happens to be incorrect. The range reported in the trials referenced runs from -2.8 to -16.8 kilograms. Compare this, in any case, to the range of weight loss in the low-fat studies referenced: +0.4 kg to -11.8 kilograms.
More to the point, it's possible to compare the efficacy of low fat and low carb diets from the data the USDA authors provide. Although the USDA authors chose not to engage in this exercise, and although the two groups are not strictly comparable, it's still relatively easy with a simple calculator to come up with an average rate of weight loss for the two different types of diet. Even Fumento could have done it. The average weight loss for the 28 low fat trials listed–at an average fat intake 25 percent and energy intake of 1665 calories per day–is a little over four kilograms in nearly 23 weeks, or less than 1/2 pound per week. The 22 low carbohydrate diets referenced led to an average weight loss of 7.4 kg over 48 days, or more than two pounds a week at an average intake of 1300 calories each day. This is over four times the rate of weight loss from the low-fat diets. The number speak for themselves, despite the author's conclusions, repeated faithfully by Fumento, that low carbohydrate diets seemed to offer no advantages over low fat diets.
What's more, the USDA authors say that "the results of several of the [low-carb] studies actually refute the contention that low-carbohydrate diets, in the absence of energy restriction, provide a metabolic advantage for weight loss." They cite four studies supporting this proposition. I happen to have two of the four in my files, and neither support their point.
One was an uncontrolled Atkins's diet trial, published in 1980 by Larosa et al., that reported an average weight loss from the Atkins diet of .9 kg (2 lbs) per week. The researchers report they would have predicted only half that from the apparent reduction in calories. They then say they can account for another quarter of a kilogram by taking into account water weight, but still fall nearly .2 kg short per week. This sounds trivial, but if sustainable, it would amount to some 10 kg or 20+ pounds of weight lost in a year beyond that explicable by the reduction in calories. There may be a simple answer to this discrepancy, but Larosa et al. don't offer any. To say this study refutes the contention that low carbohydrate diets provide a metabolic advantage is patently untrue.
The other study is even more interesting, if for no other reason than the fact that one of the co-authors is the recently infuriated Gerald Reaven. His collaborator and first author was Alain Golay, a Swiss researcher. The study was published in 1996. In this study, Golay and company randomly assigned obese individuals to receive one of two 1000-calorie diets. One was a low carb diet–32 percent protein, 15 percent carbohydrates and 53 percent fat–and the other was a high carb diet–29 percent protein, 45 percent carbohydrate and 26 percent fat. After six weeks, the low carb dieters lost an average of 8.9 kg plus or minus .6 kg, while the high carb dieters lost 7.5 kg plus or minus .5 kg. While the authors concluded that there was "no significant difference in the amount of weight loss in response to the diets," the difference between these two diets is 1.4 kg or 3 pounds in six weeks, and the error bars–the plus or minus–do not overlap, which is an elementary measure of statistical significance. If this 3 pounds in six weeks difference were to be real and sustainable, which Golay's study cannot establish, it would translate to an extra 25 pounds of weight loss over a year, without eating a single calorie less. It may or may not be real, but it is certainly evidence that this study does not refute the aforementioned contention. (It's worth noting that Golay did a second study, without Reaven, that was the identical story. When I asked Golay why he didn't follow-up on this suggestion that low-carb diets do offer a metabolic advantage, he said longer term studies were prohibitively expensive.)
Fumento then moves on to tout the merits of a type of review called a meta-analysis and cites a December 2000 study in the International Journal of Obesity and Related Metabolic Disorders co-authored by one of Fumento's favorite sources, James Hill of Colorado. This is the second leg of Fumento's three-legged "crushing" mass of evidence. It is a quasi-systematic review of the literature that concludes that low-fat diets lead to reduction in calories and weight loss.
Fumento then gets something right. He says the only meta-analysis (of which I am aware) on this issue that was properly conducted was done by the Cochrane Collaboration. Fumento doesn't bother to describe this organization, which is what made this statement relevant, so I will. The Cochrane Collaboration is a world-wide network of researchers dedicated to providing unbiased reviews of the scientific literature. The collaboration was created specifically because the 77 scientists from 11 countries who founded it a decade ago believed that meta-analyses could be so easily biased by researchers' prejudices that their disciplines needed a standardized methodology to minimize the influence of such prejudice and a venue that would allow for the publication of unbiased reviews.
The Cochrane methodology makes it effectively impossible for researchers to pick and choose which studies they would like to include in their analyses on the basis of which studies are likely to give them the result they want, a common practice in this field. Cochrane Collaboration reviews must include all studies that fit a pre-specified set of criteria, and they must exclude all that don't. Indeed, the Cochrane review to which Fumento refers started out with 3,000 citations of potential fat-restricted dietary trials and ended up with only a dozen trials that were performed to the Collaboration's standards for what constitutes good science.
Fumento's statement that the review found "no advantage to low-carbohydrate diets" is true but slightly farcical and more than slightly dishonest. The review was only a review of "fat-restricted diets" and so the authors made zero attempt to review the efficacy of carbohydrate-restricted diets. The Cochrane Collaboration review can be found on-line at http://www.update-software.com/abstracts/ab003640.htm. Its conclusion is worth noting:
The review suggests that fat-restricted diets are no better than calorie restricted diets in achieving long term weight loss in overweight or obese people. Overall, participants lost slightly more weight on the control diets but this was not significantly different from the weight loss achieved through dietary fat restriction and was so small as to be clinically insignificant." [My emphasis.]
Fumento then goes on to communicate in his idiosyncratic style my explanation of the why these dietary trials fail to live up to reasonable criteria of good science. Regrettably, Fumento doesn't seem to understand the critical point, and perhaps even the less critical points, and so makes us both look like idiots.
The point is this: to establish the effect of a diet on weight (or a pill on some physical or mental disorder), researchers have to do what's called a clinical controlled trial. The key word is "controlled". If they do such a study, they can be relatively certain that the effect they witnessed was due to the diet (or the pill) and not some other mysterious uncontrolled variable. The way to do this is by setting up the trial in advance so as to rule out these other variables–control for them, in the lingo, hence a controlledtrial.
One crucial factor in any controlled trial is to assure that the two groups being compared are legitimately comparable, which means among other factors that they must get equal treatment throughout the trial. In drug trials, for instance, placebos are used (hence the phrase "placebo-controlled" diet) to avoid any distortion that might occur when comparing individuals who are taking a pill every day (and, with it, the belief that their condition might improve) to individuals who are not. Drug trials are also done double-blind, which means neither the subjects nor the clinicians know which pills are the placebo and which are not. This controls for any possible effects that might occur if either patient or clinician knows the truth. Thus the common cliché in medical research is that double-blind, placebo-controlled clinical trials are the "gold-standard" for doing research. If a drug trial is not double-blind and placebo-controlled, it can't be trusted to get the right answer.
Diet trials, however, are problematic. It's effectively impossible to do them with placebos or double-blind. It's hard to fool subjects, for instance, into thinking that there are carbohydrates in their diets when there are not. High fat diets don't look and taste like low fat diets. Diets with pasta, bread, potatoes and sugar do not look and taste like diets without them. As a result, most of the trials used to demonstrate the effect of low-fat diets (or low-carb diets, for that fact) put a group of subjects on the diet and either measure their weight loss and nothing more, or compare the results to individuals who never went on the diet. Such trials, though, are uncontrolled. Often, for instance, the individuals on the low fat diet are given counseling, classes, and extensive follow-up to assure they stick to the low-fat diet. This intervention is not matched among the individuals who are simply told to eat their habitual diet. Thus there is a potential intervention effect, in which those getting the counseling and classes may be more motivated to live up to the researchers expectations than those who aren't. And that motivation, rather than the diet, might explain whatever weight they lose. For this reason, the Cochrane Review rejected all studies that were not randomized and did not compare low-fat diets at least to other weight-reducing diets, with the hopes that this would minimize any intervention effect.
Studies in which the level of intervention is unequal are uncontrolled trials and can't be interpreted. As medical statistician Stuart Pocock explained in his classic textbook Clinical Trials: A Practical Approach (John Wiley & Sons, 1983), "Uncontrolled trials have the potential to provide a very distorted view of therapy especially in the hands of slipshod, over-enthusiastic or unscrupulous investigators." The emphasis is Pocock's.
Fumento should know this and he shouldn't need me to explain it to him, and he should at least get it right when he tries to explain it to his readers. Indeed, Hill and his co-authors of the December 2000 International Journal of Obesity and Related Metabolic Disorders meta-analysis should have known this, too. Yet they included in their analysis studies that compared low-fat diets to medium-fat diets or habitual diets. These are uncontrolled trials. Walt Willett has made precisely the same criticism of this meta-analysis: it is a review of uncontrolled trials and so meaningless.
One last note from Pocock, which I happen to agree with whole-heartedly. Once again, the emphasis is Pocock's: "One basic premise is that it is unethical to conduct research which is badly planned or poorly executed. That is, if a trial is of sufficiently poor quality that it cannot make a meaningful contribution to medical knowledge then it should be declared unethical." To do a meta-analysis of such studies and conclude that from such scientific dreck, gold can be spun is of dubious benefit and ethics as well.
Next Fumento accuses me of having "circumvented this mass of peer-reviewed literature readily open to public scrutiny in libraries and often on-line" and basing my entire thesis on five unpublished studies.
So far, however, the mass of peer-reviewed literature that I have circumvented is one error-ridden, potentially biased USDA study, that it's doubtful Fumento read, and one dubious meta-analysis. As for the five studies, the salient point is that they are randomized, controlled trials in which the investigators actually compared a low-fat, calorie-restricted diet of the kind recommended by the American Heart Association, to a low-carbohydrate diet in which the subjects are allowed to eat as much as they want. They are not ideal, but they are better than most of the studies discussed by Hill et al. and about the best that can be done in diet studies. And if there is any intervention effect, it is likely to be biased in favor of the low-fat, low calorie diets, because those require the greater intervention.
Next along in Fumento's not-so-critical analysis, is a quote from Hill saying in reference to my book deal that I "sold out". At the risk of sounding catty and invoking the kind of standards that Fumento applies to these issues (for academicians, at least), Hill might be expected to know about such conflicts. After all, Hill has been funded by the Sugar Association to write an article for their website exonerating sugar or sugary foods as causes of obesity. It can be read on-line at www.sugar.org/science/carbohydrates.html.
Fumento also quotes Hill saying "I haven't seen any data anywhere saying Atkins is better than these other diets for weight loss." In the very next paragraph, Fumento then quotes Hill's collaborator, Gary Foster of the University of Pennsylvania, saying that the recent diet trial he did with Hill resulted in greater weight loss for those on the Atkins diet. Twice the weight loss to be precise. The statements of Hill and Foster are contradictory, which could be more a figment of Fumento's reporting than the actual data or their opinions. It would be nice, however, if the three got together and worked this one out. Maybe they could get back to us.
Fumento also mentions a University of Cincinnati study in which the Atkins group also lost twice as much weight as the low-fat diet group. This begs the question of what I misrepresented, considering my statement on the weight loss in these diets was that in all five studies those on Atkins "lost twice the weight as the subjects on the low-fat, low-calorie diets."
What's more, Fumento neglects to mention one interesting result from the University of Cincinnati study: the Atkins diet group not only lost twice the weight and twice the body fat of those on the low-fat, calorie-restricted American Heart Association-type diet, but they did it eating the same amount of calories. The researchers estimated that both groups consumed 1200 calories a day. This is mildly inexplicable without evoking some metabolic benefit gained by restricting carbohydrates. Fumento might have mentioned it, but it would have run contrary to his thesis, which is that I was irreprehensible in how I selected only those facts and opinions that agreed with my case.
The point I made in my article–not to be confused with Fumento's parody of my article–was that the existing low-carb dietary trials all suggest that when subjects are told to eat freely on a low carb diet, they voluntarily chose to consume considerably less calories. The pertinent question is why? The fact that such a diet is "a low-calorie diet in disguise" is not an answer. Naively, it seems that if a diet that encourages "pigging out", to use Fumento's words, results in a considerable reduction in calories, along with considerably more weight loss then a diet that actively restricts calories and certainly does not advocate "pigging out", it raises interesting questions that shouldn't be dismissed quite so cavalierly.
Fumento then goes on to describe "the kicker" regarding these five studies as the fact that they were "intervention studies [Fumento's emphasis], conducted using the same methodology that Taubes cites to dismiss the mountain of published material that undercuts his position."
Now Fumento confuses intervention studies with intervention effects, demonstrating a degree of what might be willful ignorance that stretches the boundaries of the imagination. The point is all clinical trials are intervention studies. You intervene in someone's life with a diet or a drug and you see what effect you have. The key to a controlled trial is to make the two interventions as similar as humanly possible so as to minimize the intervention effect. The five Atkins studies compared low-fat, low-calorie diets with low-carb diets and so the type and level of intervention were roughly equivalent.
Moreover, Fumento points out that none of the five low-carb/low-fat trials lasted more than a year, but neglects to mention that only one of the 28 low fat trials reviewed in JADA by those USDA researchers of unimpeachable integrity lasted more than a year. The subjects in that study, if our friends from the USDA can be trusted to get this one right, actually reduced their daily calorie intake to 1300 calories for 18 months and lost on average .4 kg, or less than a single pound. I pity those people.
Fumento then says that University of Cincinnati researchers Randy Seeley and David D'Alessio, were "upset that Taubes made use of their material." He neglects to mention that the reason I knew about the results of the trial was because D'Alessio e-mailed me the abstract of results presented at a 2001 meeting of the North American Association for the Study of Obesity (NAASO), with a note that said in part "I think it is fair to use the material we have already presented at meetings (we also had abstracts at the American Diabetes Association and American Dietetics Association last year, but the NAASO abstract is the most complete) in your research." Fumento also neglects to mention that the only reason he knew about the Cincinnati study was because I forwarded D'Alessio's e-mail to him, along with the abstract and the note. Fumento then moves on to the National Weight Control Registry and its 3000 successful dieters, most of whom claim that they lost weight and kept it off using low-fat, calorie-restricted diets and exercise. First he states that since NWCR dieters lost weight on low fat diets that somehow this implies in the world of Fumento-esque logical deductions that "what doesn't [Fumento's emphasis] work is a high-fat diet." This kind of logical sleight-of-hand is a Fumento specialty. The fact that these people say they lost weight by restricting fat calories means only that these people say they lost weight by restricting fat calories. It implies nothing about whether they would have lost more or less weight by restricting carbohydrate calories. That most of them exercised is an interesting fact but irrelevant to the carbohydrate/fat issue.
Fumento then states that the NWCR is meaningful to the debate because it has been "written-up in peer-reviewed medical publications." Fumento's faith in peer review is charming and considerably greater than mine. Here I will quote Pocock again:
There is a tendency for students, and indeed many clinicians, to treat the medical literature with undue respect. Major journals such as the Lancet and the New England Journal of Medicine are presumed to present new medical facts which are not to be disputed. Such a naïve faith in the "clinical gospels" is perhaps encouraged by the dogmatic style that many authors adopt, so that the uncertainties inherent in any research project often receive inadequate emphasis.
In this case, the inadequate emphasis is communicated by ignoring the fact that the NWCR is completely uncontrolled. Fumento does point this out, but then he decides it's irrelevant. He quotes Suzanne Phelan, a Brown University NWCR co-investigator saying, "you cannot get around the problem" that people who sign up for the NWCR are self-selected. Phelan is right. You can't. As a result, the NWCR is no more than an uncontrolled exercise in data collection. The only reliable statement that can be made from the NWCR data is that some 3000 individuals out of the tens of millions each year who try to lose weight, said they succeeded by reducing the fat and calories in their diet and maybe by exercising, as well. That's a nice factoid, but it adds excruciatingly little to the relevant science.
Fumento then rightfully asks why low-carb dieters do not appear in the NWCR. That's a good question and one worth investigating. This is of particular interest considering, for instance, the May 2002 issue of Consumer Reports. CR queried their readers and came up with 8000 who reported that they lost ten percent of their body weight and kept it off for at least a year –including 4000 "super losers" who lost an average of 37 pounds. According to CR, the number one lesson learned from these successful dieters was the need to "tame your blood sugar" and to do so by eating less carbohydrates and particularly less refined carbohydrates and starches. Now this is no scientific survey, but it is no less scientific, regrettably, than the National Weight Control Registry.
Fumento then raises the question of whether low carb diets might suppress hunger. He invokes as evidence that there is no "empirical support for this" an April 2002 review in the Journal of the American College of Nutrition. This is leg number three of his tripodal "crushing" mass of evidence. The JACN article, reports Fumento, reviewed high and low fat treatments when subjects were allowed to eat as much as they wanted, and found that "energy intake on the low-fat diets ranged from 16 percent to 24 percent less than those on high fat diets."
This time Fumento gets the issue date of the article correct, but he incomprehensibly butchers the quote. The relevant quote actually reads "energy intakes on the low fat diets averaged 71 percent (10 days to 2 months) to 84 percent (1-9 days) of intakes on the control higher fat regimes."
Either way, this is an interesting finding but irrelevant. It says nothing about why individuals on low carbohydrate diets or very low carbohydrate diets like Atkins's, as even Fumento reports to be the case, lose considerable weight, and why they apparently find it relatively easy to restrict their calories to do so. Fumento turns to Penn State nutritionist Barbara Rolls on this subject and describes her as "widely considered the nation's top authority on satiety", which is a lovely compliment but a bit of a stretch. Rolls invokes studies in which she and her colleagues infused pure fat and pure carbs into their subjects and found very little difference in subsequent short-term satiety. These experiments, however, say precious little about whether the macronutrient content of the diet would have an impact on the kind of weight loss or gain that takes place in real life and over periods of months or years, not hours. Both Fumento and Rolls, if her opinions are represented accurately, confuse evidence with proof.
It is worth remarking, which Fumento did not, that the sentence in the JACN review that followed his misquote made this point: "it is interesting to note that the changes in body weight observed in the low fat intervention studies are small in absolute terms (0.7 kg-1.0 kg in short-term and long-term studies on average) and also appear small compared to the changes in energy intake." For those, perhaps like Fumento, who might not be expected to read these articles carefully, this was included as one of five "teaching points" of the JACN review: "low fat dietary intervention studies have resulted in small weight loss–less than 1 kg on average in studies of up to one year's duration."
Fumento's next assault on my reporting is to accuse me of not being able to extract a single useful line from five other "top obesity researchers." One of these, however, Xavier Pi-Sunyer, I did not interview for this story. One of them, Marion Nestle, is a nutritionist and administrator with a background in molecular biology. She is not and never has been an obesity researcher, nor has she ever treated obese patients. One of them, Arne Astrup, co-authored with Hill the December 2000 International Journal of Obesity and Related Metabolic Disorders meta-analysis. My earlier comments about meta-analyses and controlled trials would suggest why I might have shied away from quoting Astrup for the enlightenment of my readers. And one of them, Jules Hirsch of Rockefeller University, I did quote in early drafts of my article, but the relevant paragraphs, regrettably, were among the last to be cut for space reasons. First it quoted Hirsch saying "Of all the damn unsuccessful treatments, the treatment of weight reduction by diet for obese people just doesn't seem to work. " It then continued:
This has led Hirsch, for example, into such a state of frustration that when we spoke last March, he said he could no more explain how obese individuals could lose weight, then he could explain how they gained it to begin with. "I've been working on this since 1960," he said. "That's a hell of a long time. You think I would have gotten a little farther along with it." For the last 20 years, Hirsch has worked with Rudy Leibel on some of the seminal experiments in obesity research. When I interviewed Leibel, who is now at Columbia University, he capped our conversation this way: "if you do feel you understand this," he said, "it will probably indicate that you've lost your mind."
I hated to see it go.
Fumento moves next to the subject of glycemic index, which is a measure of the effect of carbohydrates on blood sugar and insulin secretion. My article suggested that the glycemic index concept might be relevant to the question of why we gain weight so easily and have such trouble losing it. Fumento first mangles his explanation of the concept, and then dispenses with it as thoroughly irrelevant to the scientific discussion at hand.
The gist of the glycemic index idea, is that the more easily digested the carbohydrates, the quicker and more dramatic their effect on blood sugar and the greater the resulting secretion of insulin. These effects, so the hypothesis goes, might then have some long-term effect on hunger and fat deposition and eventually on weight. Carbohydrates–such as sugar, white bread or potatoes–have a high glycemic index and are absorbed into the blood stream quickly. Those with a low glycemic index, such as whole grains, are absorbed more slowly.
As to its relevance, it was this glycemic index concept that Consumer Reports had in mind when it advocated "taming your blood sugar" to lose weight. Curiously enough the same April 2002 Journal of the American College of Nutrition review that Fumento misquoted, also discussed the evidence that the glycemic index of carbohydrates could have relevance to hunger, food intake and weight, although Fumento missed this, as well, or ignored it. This point, too, made it into one of the five teaching points; ""short-term studies suggest that low-glycemic index carbohydrates suppress hunger more effectively than high glycemic index carbohydrates, but there are no long-term intervention studies to examine the effects of lowering the glycemic index on body weight."
Moving toward his finale, Fumento returns back to the Atkins diet and the question of whether it is safe for the long term. He cites the five unpublished trials and says they are evidence that the diet might not be "as harmful as was once generally believed" but then says the natures of the fats involved in the diet are "a distinction Taubes decided to lose." Wrong again. To be precise I discussed the effects of the fats on the various cholesterol and fatty acid particles in the blood, which, short of an actual heart attack, is the end result of interest.
I said "In all five studies, cholesterol levels improved similarly with both diets, but triglyceride levels were considerably lower with the Atkins diet. Though researchers are hesitant to agree with this, it does suggest that heart-disease risk could actually be reduced when fat is added back into the diet and starches and refined carbohydrates are removed." I then went on to describe the studies that were in the works to extend this result and see if it would be sustained over longer time periods, and I discussed my own anxiety eating such a diet that included, in my case, eggs and sausage every morning. "I can look down at my eggs and sausage" I wrote, "and still imagine the imminent onset of heart disease and obesity, the latter assuredly to be caused by some bizarre rebound phenomena the likes of which science has not yet begun to describe. The fact that Atkins himself has had heart trouble recently does not ease my anxiety, despite his assurance that it is not diet-related."
Fumento's finale is his insistence that I misrepresented an American Medical Association critique of the Atkins diet that was released publicly in March 1973 by the AMA and published the following June. Fumento agrees with my characterization of the critique as scathing, but he takes exception to my claim that the AMA's anonymous author "acknowledged that the diet probably worked but expressed little interest why." In his own press release touting his article, Fumento accuses me of having "grossly misrepresented" the AMA's position. Fumento then quotes many of the scathing comments included in the critique, but he neglects to mention, as is his wont, one key sentence: "The fact remains, however, that some patients have lost weight on the low-carbohydrate diet `unrestricted in calories'." This seems to me–and perhaps, once again, I'm being naïve here–to be an admission that the diet probably works.
The next sentence, which Fumento does quote, read, "When obese patients reduce their carbohydrate intake drastically, they are apparently unable to make up the ensuing deficit by means of an appreciable increase in protein and fat." To put it in plain English, they consumed less calories. They ate less. The AMA then left it at that. The article did not try to explain–hence my characterization of the position as expressing "little interest"– why an obese man, for instance, who would, according to the scientific literature of balanced calorie-restricted diets, be ravenous if he lost ten pounds, suddenly find it difficult to eat enough steak or chicken or eggs or cheese, despite his weight loss, to get, say, 2000 calories a day and maintain his weight. Two half-pound burgers, naked, four boiled eggs and a tin of tuna fish (canned in oil), hold the mayo, will do the job nicely and, for many obese men, would constitute little more than hors d'oeuvres.
Fumento then tries to back up his own rough treatment of my reporting with mention of a "fatlash" in response to my article. The substance of that fatlash, however, constituted a single page in Newsweek, by a writer who was having a book published two months later claiming that fatty foods and indolence were the cause of the obesity epidemic; an article in The Washington Post by a diet writer who has been pushing low-fat diets since the mid-1980s, and which happened to be no more accurate nor in command of the relevant science than Fumento's; and an article in the Nutrition Action Healthletter, a publication of the Center for Science in the Public Interest (CSPI). The CSPI is an advocacy group that has been pushing low-fat diets since the 1970s. In January Reason's science correspondent, Ronald Bailey, described CSPI as "a Naderite spin-off that has not been above a bit of sensationalism in trying to get its nutrition message across either. Famous as the self-styled "food police," CSPI launches highly publicized jihads against foods that it feels are not up to snuff nutritionally. That's their right, of course, but others feel that CSPI exaggerates its claims and is misreporting scientific results." The CSPI philosophy on dietary fat and carbohydrates was summed up nicely by its director Michael Jacobson in a 1979 article in the journal Science as "Eat less sugar. Eat less fat. Bread and potatoes are where it's at." It's understandable that the food police might object to an article suggesting that bread and potatoes are not where it's at.
And this is the point: when an article such as mine suggests that three decades of dietary dogma might be both wrong and hazardous to the health, it will elicit public and perhaps angry responses from purveyors of that dogma. These responses will assuredly be exacerbated if the editors of The New York Times Magazine choose to run the article on the cover, as they did in this case. There seems little way to avoid that fact. If Fumento is dedicated to defending the dogma–and with it, the arguments he made six years ago in his book The Fat of the Land–neither malice, vitriol, nor his consistently remarkable ability to screw up both the facts and the science, will take the place of good, solid journalism.
With that note, I'd like to make two last small corrections. One is that Fumento reports that I am one of two writers to ever win the science-in-society award of the National Association of Science Writers three times, which is the maximum the NASW allows. He is half right. I did win it three times, but the only other three-time winner is a documentary film-maker, not a writer. And finally, Fumento refers to my editor at Knopf as "Scott Segal." His name is Jonathan.