Gary Taubes has been the most influential voice in nutrition since Ancel Keys, but since Keys is now universally excoriated, that may not be an unmitigated honor: how do we know that Taubes is not as flawed? It is certainly unprecedented for the most influential voice of a major scientific discipline, in this case nutrition, to be a journalist rather than a scientist, but since Keys was a scientist, perhaps nutrition does need to be led by a non-scientist. Let’s see.
Between 1953 and 2002, the world of nutrition was dominated by Ancel Keys’s hypothesis that heart attacks and strokes were caused by an excess of fat in the diet, but in 2002 Taubes wrote an article in the New York Times Magazine “What if It’s All Been a Big Fat Lie?” that almost overnight converted the nutritional avant garde from fat-phobia to carbohydrate-phobia. Taubes went on to expand his article into the best-selling 2007 book Good Calories, Bad Calories.
Taubes, a physicist who became a science journalist, has made a career of exposing poor research. To expose the brutal politics within the world of high-energy physics, he subtitled his first book Nobel Dreams (1986) as Power, Deceit and the Ultimate Experiment. In 1993 he published his second book, Bad Science: The Short Life and Weird Times of Cold Fusion, to analyze the failings of the Fleischmann-Pons experiment. Having eviscerated physics, Taubes moved on to nutrition because, in his words, “I had a lot of people in the physics community say to me that if I was interested in bad science I should look at this stuff in public health because that is really bad.”
So, what is Taubes’s difficulty with the Keys fat hypothesis? Keys (1904–2004) of the University of Minnesota faced the problem, as did all of America in the 1950s, of the sudden epidemic of heart attacks. Frighteningly, these had seemingly come out of nowhere. Absence of evidence is not evidence of absence, but in a study published in the British Heart Journal Leon Michaels, a Canadian physician, showed that the absence of evidence for heart attacks before the 20th century was indeed evidence for their absence. Cleverly, Michaels compared the characteristic chest pain of heart attacks with the characteristic pain and symptoms of migraine and gout, showing that whereas frequent descriptions of migraine and gout can be found in medical texts from all eras, stretching back to Greek and Roman times, angina and heart attacks started to be described with any frequency only in the 20th century: heart attacks really were a new disease, yet a plague of them had descended on the United States like avenging angels, and by midcentury they had become by far the commonest cause of death. Nor were they respecters of persons: President Eisenhower had a heart attack in 1955, while Franklin Roosevelt had died from the related condition of stroke.
Heart attacks and strokes are caused by atherosclerosis, which used to be known as “hardening” of the arteries but which is better described as “inflammation” of the arteries. By Keys’s day it seemed clear that atherosclerosis, in its turn, was caused by the deposition of cholesterol in the arteries, and since people with heart attacks had raised levels of cholesterol in the blood, and since eating an excessive amount of saturated fat raised the blood levels of cholesterol, it seemed obvious that cutting back on fat in the diet would help prevent heart attacks and their related condition of strokes.
And it did! Contrary to myth – a myth to which Taubes himself has contributed – the course of the heart attack epidemic in America followed the Keys fat paradigm almost faultlessly. So between 1960 and 2000 the per capita U.S. consumption of
- saturated fatty acids fell from 55 to 46 grams per day
- cholesterol fell from 465 to 410 milligrams per day.
Meanwhile the per capita U.S. consumption of
- carbohydrates rose from 380 to 510 grams per day
- fiber rose from 18 to 26 grams per day.
And at the same time, the incidences of heart attacks and strokes in America … collapsed! So between 1970 and 2010 the heart attack rate fell by 75 per cent (i.e., in 2010 it was only a quarter of the rate in 1970) while the stroke rate fell markedly too. Has Keys not been exonerated and Taubes disproved by these data? And with that extraordinary public benefit, need we worry about a bit of diabetes and a touch of obesity? These are, after all, apparently translating into a remission of the epidemic of heart disease.
I am of course writing that penultimate sentence ironically. The facts are unambiguous, but the truth – as Taubes himself acknowledges – is that we still don’t know what has caused the remission in heart disease nor the current epidemics in obesity and type 2 diabetes. Taubes blames the latter on sucrose, but I think the case is not yet proven – and indeed neither does Taubes, hence his wonderful initiative in creating with Dr. Peter Attia the Nutrition Science Initiative (NuSI) to study health to the highest standards of scientific objectivity.
As his essay demonstrates, Gary Taubes deservedly remains the most influential person in nutrition today, not only because he smashed the fat consensus, and not only because he has recognized the metabolic syndrome as the greatest health threat facing the western world today, but also because he is honest enough to know that even his own faith in the sugar hypothesis needs dispassionate testing. Moreover, in his defense of science as a process that must be freed from government demands for immediate results, he has shown he understands the scientific method better than many a practicing nutritional researcher.
 A Keys (1953) Atherosclerosis: A problem in newer public health. Journal of the Mount Sinai Hospital 2: 118-139.
 G Taubes (7th July 2002) “What if It’s All Been a Big Fat Lie?” New York Times Magazine.
 G Taubes (2007) Good Calories, Bad Calories Knopf NY.
 Quoted by B Appleyard (1 January 2017) How sugar got us in a sticky mess. Sunday Times Magazine, p 24-27.
 L Michaels (1966) Aetiology of coronary heart disease: An historical approach. Br Heart J 28: 258-264.
 S-S Zhou et al (2010) B-vitamin consumption and the prevalence of diabetes and obesity among US adults: population based ecological study. BMC Public Health 10: 746. www.biomedcentral.com/1471-2458/10/746
 A Statement from the American Heart Association/American Stroke Association (2013) Factors Influencing the Decline in Stroke Mortality. https://professional.heart.org/idc/groups/ahamah-public/@wcm/@sop/@smd/documents/downloadable/ucm_458896.pdf p8. Accessed December 2016.
If in 1977 Americans had actually followed McGovern’s 7 dietary recommendations, they’d likely be a lot better off today. Though McGovern’s recommendations did indeed call for increasing dietary carbohydrates, they recommended their increase come from fruits, vegetables, and whole grains, not Wonder Bread, Froot Loops, and Coca-Cola. They also recommended decreasing red meat consumption in favor of fish and poultry, that poly-unsaturated fats replace saturated, and with their admonishment to limit foods high in salt, sugar, and fat, they effectively recommended that the public shy away from ultra-processed foods – foods which tend to use one or more of those ingredients to create the “bet you can’t eat just one” phenomenon of hyper-palatability.
Pertinent to this discussion, the guidelines also expressly called for a drastic reduction in sugar, recommending it account for no more than 10% of Americans’ total daily energy intakes. (But they did likely get eggs wrong – the evidence suggesting that there’s real benefit to limiting dietary sources of cholesterol hasn’t stood time’s test.)
Taubes asserts that America’s dietary guidelines, starting with McGovern’s purported low-fat call to arms, led to fat’s unintended dietary replacement with sugar and simple carbohydrates. Whether it’s fair to call America’s current dietary woes the “unintended consequence” of not actually following recommendations that also explicitly called for a severe reduction in dietary sugar and ultra-processed foods, there’s no disagreement that the guidelines were not followed. And while a diet that provides 30% of calories from fat, as McGovern’s recommended diet did, isn’t fairly describable as a low-fat diet, Taubes’ point that over-simplified messages risk unintended consequences is an important one – and one we should be reminded of when we see headlines suggesting butter to be not just benign but rather a health food.
Truth be told, despite the inference that Americans replaced dietary fats with simple carbohydrates and sugars, the data would suggest that Americans consume just as much fat as they always did – it’s just that they’re consuming more of everything else. While the percentage of a person’s daily energy intake coming from dietary fat has indeed dropped, the absolute amount of dietary fat being consumed hasn’t changed since at least the very early 1970s.
Figure 1: Mean adjusted intakes of macronutrients among adults aged 20–74 y by NHANES study period. A: Results shown as absolute intake in grams per day. B: Results shown as percent of total intake.
Or hasn’t it?
Unfortunately, we don’t have great dietary consumption data to draw upon. Instead we’re left to expound on studies reliant on food frequency questionnaires, which in turn have been shown to be extremely unreliable. One recent scathing review of the use of dietary recall and food frequency questionnaires asserted that their use was “akin to creation science in that they fail to meet the basic requirements of scientific research.
So while it’s fair to point out the flaws in the science that built the case against dietary fats in the 70s and 80s, it’s also fair to point out that there are flaws as well in building the case against sugar, as those who are building it can’t help but also use much of that same challenged pool of dietary evidence. Worse, we may never have the means with which to determine truly specific causal factors when it comes to diet and health, and that’s not even considering the incredible number of confounders that freely living human populations possess. For example, even in captive populations of genetically identical mice fed tightly restricted and controlled diets, there are dramatic confounders, whereby a captive mouse’s living arrangements (eg. noise, bedding, light, the pH of their water, the microbiome prevalent from each particular vendor’s mice, fecal oral eating, and more) have been shown to markedly affect study outcomes, including their metabolic responses to identical diets.
This all helps to explain why Taubes’ assertion that “[t]he ‘best sense of the data,’ however, will always depend on who is asked” rings so true. With decades of food frequency questionnaires informing hundreds if not thousands of studies, and given the degree of confounding that exists from person to person let alone from nation to nation, there is something for everyone in the scientific literature. And given no one is free from confirmation and intellectual sunk cost biases, cherry picking to make the case against fat or sugar is neither difficult nor surprising to see.
What’s also not surprising is that there are scads of conflicts of interest in nutrition. Dietary recommendations are big business for the food industry, while recommending diets is big business for publishing. The weight loss industry has been reported as being a $60 billion per year juggernaut in North America alone, and that doesn’t include the dollars involved in the pursuit of different dietary strategies selling health (which would include such things as organics, GMOs, gluten, and other dietary recommendations and fads).
Taubes’ work compellingly cultivates concerns about the many conflicts of interest inherent to nutrition science. It’s important though to note that financial conflicts with the food industry have never been shown to be uniquely influential among all financial conflicts. Equally worthy of consideration are funding by nonprofits with established and specific points of view, as well as of course personal profit – where some have translated their take on nutrition science into massive book and speaking deals. So while Taubes justly notes that sugar science is speculative, unambiguous statements and cherry picked data are certainly not exclusive to the public discourses of nutrition researchers about dietary fats.
Conflicts aside, there’s no denying the rise of chronic, non-communicable diseases (NCDs) since the McGovern era. So where to lay blame? Well, we’re definitely eating more. Portion sizes have grown, both in our homes and when eating out. We’re spending 18 times as much money as we once did in restaurants. We’re purchasing twice as much processed food and sweets as we did in the 80s. The Don Drapers of the real world command multi-billion dollar junk food marketing budgets, and are spending – this is not a typo – 150 times more than in the 80s, to specifically target children. Our teachers, coaches, and camp counselors use junk food to reward, pacify, and entertain our kids at nearly every turn. And there seems to be no occasion too small to not warrant celebrating it with junk food. And it’s not just that we’re eating too much, what we’re eating is a mess.
Studies conducted in developing nations show the rise of NCDs to proceed in near lockstep with the introduction of a so-called westernized food supply. Our best evidence currently points to ultra-processed foods as the causal culprits, with a basic definition of ultra-processed being:
Ready to consume (or heat) formulations manufactured from cheap ingredients either directly extracted from whole foods or processed from components extracted from whole foods (e.g. high-fructose corn syrup). These products typically contain the addition of preservatives and “cosmetic additives” and are usually energy dense with high fat, sugar, and salt contents and with little or no water, fibre, micronutrients, or other “protective bioactive compounds” which exist in whole foods.
We don’t fully know what it is about ultra-processed foods that leads them to be so tightly linked to NCDs including obesity and diabetes. While the prospect of a single, causal, Occam’s razor-esque nutrient to blame is both seductive and hopeful, science has not clearly identified one. Loss-adjusted food availability data, bolstered by documented portion size and spending pattern changes, offers more objective data than dietary recall surveys, but at this point such studies only clearly finger our eating much more than we ever did. That said, there’s no doubt sugar is a major raindrop in our flood of excess calories, serving as one of the primary drivers of hyper-palatability, as well as a key ingredient in a large percentage of likely less sating, ultra-processed foods, and especially of the sugar-sweetened beverages of which we’re drinking a great deal more.
Discussions of which nutrient to blame may further confuse an issue that doesn’t benefit from the confusion, especially when viewed from the clinical trenches. While Taubes feels very strongly about knowing “beyond reasonable doubt” the triggers of obesity and diabetes, and the specific role of sugar in human health and disease, and putting aside the possibility that sometimes multiple roads lead to the same destination, as someone whose full time daily job for the past 13 years has been to work with individuals on improving their lifestyles, I think that looking only to sugar misses the forest for the trees when it comes to helping people to improve their health.
As a clinician with a full time behavioral weight management practice, and as a fierce public health policy critic, I believe that pragmatism and realism are more important than idealism and rigidity in cultivating behavior change. While I would readily agree with Taubes that sugar reduction is a very worthy goal, as a goal it’s simultaneously not enough, and too much. It’s not enough as there are many other actionable areas ripe for behavior changes which if enacted would markedly reduce a person’s risk of developing NCDs, and it’s too much in that with food certain to forever serve all of us in roles of both comfort and celebration, sugar’s singular demonization, if embraced, may lead to calls for extremes of action that even if enacted, aren’t likely to be enjoyable enough to be sustainable for most.
History also informs us of the risks of single-nutrient battle cries. The excessive amplification of “low-fat” messages enabled the sale, and likely encouraged the consumption, of ultra-processed low-fat foods. With this as our backdrop, let’s hope that public health policy makers heed Taubes’ warning that,
Whatever actions we take, though, are likely to have unintended consequences. One of them is further ingraining the precedent for taking other actions that are likely to be wrong and damaging as well.
We should offer diet guidance far more broadly than just on sugar so as not to unintentionally fuel a surge in the manufacture and consumption of new low-sugar ultra-processed foods. There are incredible health benefits to be had from looking at broad based behaviors; an emphasis on sugar alone might overlook them. Behaviors such as:
- Cooking meals from fresh whole ingredients while minimizing restaurant meals and ultra-processed foods; eating those meals free from distractions, and ideally with friends or family.
- Exercising as often and as much as is enjoyable, and appreciating that short bouts also provide benefit.
- Not smoking.
- Cultivating good night sleeps.
- When considering any dietary indulgence, asking first if it’s worth it, and second, if it is, what’s the smallest amount needed to be happy?
- Drinking alcohol only in moderation.
- Nurturing friendships and relationships.
This sort of back to basics approach is what underlies Brazil’s recently published and highly praised dietary guidelines, where rather than try to distill divergent, challenged, and often conflicted studies into a set of highly specific recommendations, their aim was to encourage much broader patterns of healthy eating and living. Were Americans as a whole to adopt even just the aforementioned 7 behaviors, not only would their sugar intake be reduced, but so too would the burden of many NCDs.
Lastly, while Taubes believes that “Effective public health policy requires that the public be convinced to change their behavior for their own good,” it’s important to remember that behavior change for one’s own good may be an unrealistic luxury for many. Instead effective public health policy requires that the public be guided to behavior change by way of an environment designed to make health enriching choices easier, cheaper, and/or unavoidable.
Gary Taubes paints a lively picture of lawmaker overzealousness, industry subterfuge, and researcher bias to argue that fat may have been unfairly blamed for the ravages of sugar, and as a result, misguided government dietary advice drove Americans to eat more sugar, ultimately contributing to obesity, diabetes, and coronary heart disease. Yet the key questions Taubes raises cannot be evaluated via historical narrative. It’s time to hang up our tweed blazers and slip into our white lab coats, because Taubes has made specific and testable assertions, which I will evaluate in three parts. First, I will examine the 1980 Dietary Guidelines to determine if they condemn fat and take a weak stance on sugar as suggested. Second, I will evaluate the hypothesis that the Guidelines contributed to obesity, diabetes, and coronary heart disease. And third, I will evaluate the hypothesis that sugar “may be the primary cause” of the three aforementioned conditions.
Inside the 1980 Dietary Guidelines for Americans
In the opening sentence of his essay, Taubes states that “forty years ago this month, January 1977, the federal government entered the business of giving dietary advice,” referring to a congressional report that gave rise to the 1980 Dietary Guidelines for Americans. In fact, the federal government has been issuing dietary advice regularly for over a century. The 1980 Guidelines were, however, notable for the fact that they abandoned the traditional “eat more” approach of diversifying and balancing the diet, in favor of an “eat less” approach of limiting foods perceived as unhealthy.
Taubes argues that this document condemned fat and took a weak stance on sugar. Let’s have a look for ourselves. At the core of the 1980 Guidelines were seven pieces of dietary advice, shown below:
As you can see, “avoid too much fat” and “avoid too much sugar” are equally prominent and identically worded recommendations. Taubes suggests that the statement “avoid too much sugar” is vague and lacks conviction—it is “a tautological statement that could apply to any food.” Yet we are intended to believe that the same language applied to fat altered the course of the American diet.
The Guidelines went further, providing explicit guidance on how to “avoid too much sugar”:
- Use less of all sugars, including white sugar, brown sugar, raw sugar, honey, and syrups.
- Eat less of foods containing these sugars, such as candy, soft drinks, ice cream, cakes, cookies.
In the section titled “Maintain Ideal Weight,” the document lays out its four-part plan for weight loss: Increase physical activity, eat less fat and fatty foods, eat less sugar and sweets, and avoid too much alcohol. Again, the advice to limit fat and sugar intake receive equal attention, and the Guidelines clearly implicate excess sugar intake in obesity.
A person who faithfully adhered to the Guidelines would end up eating a diverse, largely unprocessed diet composed of whole grains, beans, nuts, lean meats, seafood, dairy, eggs, vegetables, and fruits, with little added sugar, added fat, or highly processed foods or beverages. Are we really to believe that this advice—which clearly and repeatedly recommended eating less sugar—made Americans eat more sugar, leading to obesity, diabetes, and coronary heart disease? Having reviewed the Guidelines, the notion strains credulity, but let’s test it anyway.
Wrath of the Guidelines
Taubes argues that the Guidelines may have unintentionally contributed to obesity, diabetes, and coronary heart disease. This rests on the assumption that the Guidelines themselves substantially influenced American eating behavior, either directly or indirectly via the food industry, and therefore that changes in the American diet over time can be attributed to it rather than to the numerous other shifts in the American food and cultural landscape that were happening at the time. No evidence is presented to support this assumption, so we will examine it here.
If the Guidelines substantially influenced the American diet, then we should observe that total fat, added fat, and added sugar intake declined in its wake. In perfect contrast to this prediction, our total intake of fat remained approximately the same after the Guidelines, our intake of added fat increased, and our intake of added sugar increased.,
Furthermore, we should observe a decline in the consumption of highly processed foods rich in fat and sugar that the Guidelines clearly advised against. To test this prediction, here are the top six sources of calories in the U.S. diet, in descending order of importance, as of 2006:
- Grain-based desserts (cakes, cookies, donuts, pies, and related items)
- Yeast breads
- Chicken and chicken-mixed dishes
- Soda, energy drinks, and sports drinks
- Alcoholic beverages
This list is both disturbing and informative. Cake and related desserts are the number one source of calories in the American diet. Soda, pizza, and alcohol are three of the remaining five items. The average American diet is not even remotely inspired by the Guidelines.
The truth is that we don’t wash down our pizza and cake with beer and soda because we think it’s healthy or because we believe the government recommended it. We do it because we are human beings who are driven by considerations of pleasure, cost, and convenience—precisely the qualities the food industry has been optimizing for the last 40 years.
Taubes assumes that we suffer from increasing rates of obesity and diabetes because we have failed to identify their true cause. Yet the evidence suggests a simpler and more compelling explanation: We eat too much food that is obviously unhealthy, and it’s not because researchers or the government told us to, but because we like it.
A Slow-Acting Toxin
According to Taubes, sugar may be a “toxin” and “the primary cause of diabetes, independent of its calories, and perhaps of obesity as well.” Elsewhere in the essay, coronary heart disease is added to the list. Yet Taubes asserts that this speculative hypothesis cannot currently be tested because there is so little existing research on sugar, and so little interest in conducting such research, that “the research necessary to nail it down would take years to decades to complete and is not even on the radar screen of the funding agencies.”
This belief is remarkable in light of the fact that a Google Scholar search returns hundreds of scientific papers on the health impacts of sugar, many of them human randomized controlled trials, and many funded by the U.S. National Institutes of Health. In reality, the health impacts of sugar are of considerable interest to the scientific community, and as such, they have been studied extensively. Having established that this research exists, let’s take a look at it.
The hypothesis that sugar is the primary cause of coronary heart disease is easily refuted. In the United States, coronary heart disease mortality has plummeted by more than 60 percent over the last half century, despite a 16 percent increase in added sugar intake. Roughly half of this decline can be attributed to better medical care, while the other half is attributed to underlying drivers of disease such as lower cholesterol and blood pressure levels and an impressive drop in cigarette use.This striking inverse relationship is incompatible with the hypothesis that sugar is the primary cause of coronary heart disease, although it doesn’t exonerate sugar.
Is sugar the primary cause of diabetes, “independent of its calories”? Research suggests that a high intake of refined sugar may increase diabetes risk, in large part via its ability to increase calorie intake and body fatness, but it is unlikely to be the primary cause. An immense amount of research, including several large multi-year randomized controlled trials, demonstrates beyond reasonable doubt that the primary causes of common (type 2) diabetes are excess body fat, insufficient physical activity, and genetic susceptibility factors.
The ultimate test of the hypothesis that sugar is the primary cause of obesity and diabetes would be to recruit a large number of people—perhaps even an entire country—and cut their sugar intake for a long time, ideally more than a decade. If the hypothesis is correct, rates of obesity and diabetes should start to decline, or at the very least stop increasing. Yet this experiment is far too ambitious to conduct.
Or is it? In fact, this experiment has already been conducted—in our very own country. Between 1999 and 2013, intake of added sugar declined by 18 percent, taking us back to our 1987 level of intake. Total carbohydrate intake declined as well. Over that same period of time, the prevalence of adult obesity surged from 31 percent to 38 percent, and the prevalence of diabetes also increased.
U.S. sugar intake and adult obesity prevalence, 1980-2013. Data are from Centers for Disease Control and Prevention NHANES surveys and USDA Economic Research Service food disappearance records (5,6). Axes are bounded to illustrate correlation, or lack thereof.
Americans have been reining in our sugar intake for more than fourteen years, and not only has it failed to slim us down, it hasn’t even stopped us from gaining additional weight. This suggests that sugar is highly unlikely to be the primary cause of obesity or diabetes in the United States, although again it doesn’t exonerate sugar. Furthermore, it suggests that the laser-like focus on sugar is a distraction from the true, more complex nature of the problem.
I have presented a small piece of a large body of evidence that is more than sufficient to refute the assertion that sugar may be the primary cause of obesity, diabetes, and coronary heart disease. At the same time, this evidence does suggest that added sugar is part of a broader diet and lifestyle landscape that contributes to these three conditions, a conclusion that is not especially controversial within today’s scientific, medical, and public health communities.
 Davis C, Saltos E. Dietary Recommendations and How They Have Changed Over Time. USDA Economic Research Service; 1999.
 Nestle M. Food Politics: How the Food Industry Influences Nutrition, and Health, Revised and Expanded Edition. Berkeley: University of California Press; 2007. 510 p.
 Nutrition and Your Health. Dietary Guidelines for Americans. US Department of Agriculture and US Department of Health and Human Services; 1980.
 For example, the declining cost of sweeteners and processed food in general, increasing ubiquity of food and advertising, advances in food sensory engineering, and the replacement of home cooking with restaurant and pre-prepared food. I discuss these trends in my book The Hungry Brain.
 Taubes has previously argued that the percentage of fat in the U.S. diet declined in the wake of the Guidelines, which is correct. However, this percentage change was due to an absolute increase of carbohydrate intake, not an absolute decline of fat intake. Since our absolute intake of fat remained unchanged, this cannot be interpreted as a decline of fat intake. USDA Economic Research Service - Food Availability (Per Capita) Data System [Internet]. 2016 [cited 2016 Dec 22]. Available from: https://www.ers.usda.gov/data-products/food-availability-per-capita-data-system/. Austin GL, Ogden LG, Hill JO. Trends in carbohydrate, fat, and protein intakes and association with energy intake in normal-weight, overweight, and obese individuals: 1971-2006. Am J Clin Nutr. 2011 Apr;93(4):836–43.
 Throughout this piece, the term “added sugar” refers to table sugar, high-fructose corn syrup, honey, and all other calorie-containing sweeteners, including both home use and consumption via prepared foods and beverages. It does not include the sugar naturally found in fruit, vegetables, and milk.
 U.S. Department of Agriculture and U.S. Department of Health and Human Services. Dietary Guidelines for Americans, 2010. U.S. Government Printing Office; 2010.
 Moss M. Salt Sugar Fat: how the food giants hooked us. New York: Random House; 2013.
 Ibid. See also USDA, above, and Fox CS, Evans JC, Larson MG, Kannel WB, Levy D. Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study. Circulation. 2004 Aug 3;110(5):522–7. And Ford ES, Roger VL, Dunlay SM, Go AS, Rosamond WD. Challenges of ascertaining national trends in the incidence of coronary heart disease in the United States. J Am Heart Assoc. 2014 Dec 3;3(6):e001097.
 Ford ES, Ajani UA, Croft JB, Critchley JA, Labarthe DR, Kottke TE, et al. Explaining the decrease in U.S. deaths from coronary disease, 1980-2000. N Engl J Med. 2007 Jun 7;356(23):2388–98.
 Horst KW ter, Schene MR, Holman R, Romijn JA, Serlie MJ. Effect of fructose consumption on insulin sensitivity in nondiabetic subjects: a systematic review and meta-analysis of diet-intervention trials. Am J Clin Nutr. 2016 Dec 1;104(6):1562–76. Black RNA, Spence M, McMahon RO, Cuskelly GJ, Ennis CN, McCance DR, et al. Effect of eucaloric high- and low-sucrose diets with identical macronutrient profile on insulin resistance and vascular risk: a randomized controlled trial. Diabetes. 2006 Dec;55(12):3566–72. Stanhope KL, Schwarz JM, Keim NL, Griffen SC, Bremer AA, Graham JL, et al. Consuming fructose-sweetened, not glucose-sweetened, beverages increases visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese humans. J Clin Invest. 2009 May;119(5):1322–34. Lewis AS, McCourt HJ, Ennis CN, Bell PM, Courtney CH, McKinley MC, et al. Comparison of 5% versus 15% sucrose intakes as part of a eucaloric diet in overweight and obese subjects: effects on insulin sensitivity, glucose metabolism, vascular compliance, body composition and lipid profile. A randomised controlled trial. Metabolism. 2013 May;62(5):694–702.
 Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002 Feb 7;346(6):393–403. Hamman RF, Wing RR, Edelstein SL, Lachin JM, Bray GA, Delahanty L, et al. Effect of weight loss with lifestyle intervention on risk of diabetes. Diabetes Care. 2006 Sep;29(9):2102–7. Tuomilehto J, Lindström J, Eriksson JG, Valle TT, Hämäläinen H, Ilanne-Parikka P, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001 May 3;344(18):1343–50. Ramachandran A, Snehalatha C, Mary S, Mukesh B, Bhaskar AD, Vijay V, et al. The Indian Diabetes Prevention Programme shows that lifestyle modification and metformin prevent type 2 diabetes in Asian Indian subjects with impaired glucose tolerance (IDPP-1). Diabetologia. 2006 Feb;49(2):289–97. Kosaka K, Noda M, Kuzuya T. Prevention of type 2 diabetes by lifestyle intervention: a Japanese trial in IGT males. Diabetes Res Clin Pract. 2005 Feb;67(2):152–62. Pan XR, Li GW, Hu YH, Wang JX, Yang WY, An ZX, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. The Da Qing IGT and Diabetes Study. Diabetes Care. 1997 Apr;20(4):537–44. Chan JM, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men. Diabetes Care. 1994 Sep;17(9):961–9. Poulsen P, Kyvik KO, Vaag A, Beck-Nielsen H. Heritability of type II (non-insulin-dependent) diabetes mellitus and abnormal glucose tolerance–a population-based twin study. Diabetologia. 1999 Feb;42(2):139–45.
 See Taubes, above, at 5.
 Fryar CD, Carroll MD, Ogden CL. Prevalence of Overweight, Obesity, and Extreme Obesity Among Adults Aged20 and Over: United States, 1960–1962 Through 2013–2014. Natl Cent Health Stat Health E-Stats. 2016 Jul; Menke A, Casagrande S, Geiss L, Cowie CC. Prevalence of and Trends in Diabetes Among Adults in the United States, 1988-2012. JAMA. 2015 Sep 8;314(10):1021–9.
Forty years ago this month, January 1977, the federal government entered the business of giving dietary advice, first with a Congressional report released by George McGovern’s bipartisan Select Committee on Hunger and Human Needs, which in turn prompted the first Dietary Guidelines from the U.S. Department of Agriculture.
With the perfect acuity of hindsight, two stories from those initiatives bear recalling. The first involves McGovern’s report, Dietary Goals for Americans, which was primarily controversial for its recommendation that Americans should reduce the fat content of their diets and compensate with more carbohydrates. After releasing the report, McGovern’s committee held a series of hearings to deal with the backlash, both from scientists and industry. While McGovern, his colleagues, and their committee staff may have all hoped the guidelines would lead to a nation eating, perhaps, as Michael Pollan would later memorably put it, (whole) food, “mostly plants and not too much,” that would not be the case. Hence story number one is about the dangers of unintended consequences.
The Committee’s staff director later told me a story of how he got an inkling of what would eventually happen. As he recalled it, an economist pulled him aside after one of the hearings and gave him a lesson on market disincentives to healthy eating: ”He said if you create a new market with a brand-new manufactured food, give it a brand-new fancy name, put a big advertising budget behind it, you can have a market all to yourself and force your competitors to catch up. You can’t do that with fruits and vegetables. It’s harder to differentiate an apple from an apple.”
And this is indeed what happened. Supermarkets became full of low-fat food products—“foodlike substances,” to borrow again from Pollan—all with implicit or explicit health claims that they would lower cholesterol and bestow a longer and healthier life. Many of these products replaced fat with some variant of the 50-50 combination of glucose and fructose that makes up what we think of as sugar—most notably, high fructose corn syrup, which had just recently entered the food supply. While this transition was occurring, our nation and the world also experienced unprecedented increases in the prevalence of obesity and diabetes, raising the question of whether these two phenomena are causally related.
This brings us to story number two. McGovern’s Dietary Goals prompted Carol Foreman, an assistant secretary of agriculture appointed by President Jimmy Carter, to begin the process of codifying the goals as official government policy. As Foreman later told me, she was aware of scientific controversy on the nature of a healthy diet, but she believed that sufficient consensus existed to provide at least provisional recommendations. “Tell us what you know and tell us it’s not the final answer,” Foreman says would tell nutrition researchers at the time; “tell [us] what your best sense of the data is right now.”
The “best sense of the data,” however, will always depend on who is asked. And therein lies what may be the second problem with the last forty years of dietary guidance in the United States: The existence of a controversy in science implies that uncertainty exists about whether the consensus is correct. Clearly in the 1970s, the majority opinion among nutrition authorities was that dietary fat, particularly saturated fat, caused heart disease, and so probably diabetes and obesity as well. The three disorders are highly associated, suggesting a common cause. Since then, the dietary fat hypothesis has repeatedly failed confirmation in clinical trials, but the strong suspicion remains that fat consumption should be limited; it’s still a foundational assumption about the nature of a healthy diet, if not among public health authorities then among the public whose diet they have been trying to hard to influence.
Another unintended consequence emerged from the official embrace of the dietary fat consensus in the 1970s, however, and this may have had far more profoundly damaging implications. The dietary fat consensus negated the viability of a competing hypothesis that highly processed carbohydrates, and particularly sugars, are the macronutrients to worry about in the American diet. By this thinking, the modern diet damages our health not through cholesterol-related mechanisms (or the presence or absence of vitamins and minerals) but through disruptions of insulin signaling, in which diabetes and so premature death is the end state. Coronary heart disease is a consequence as well.
This hypothesis was actively studied and debated through the mid-1970s, but then it was disassociated from the likely causal link that pointed back to refined carbohydrates and sugar. This disassociation was a casualty of the consensus building of the 1970s, aided and abetted by the sugar industry itself, which was the party with the most to lose in the controversy.
The sugar industry worked diligently beginning in the mid-1960s to help shape the anti-fat consensus and assure that sugar was never perceived as anything more than a source of excess, if empty, calories. As influential nutritionists began to argue in the United States (Jean Mayer at the Harvard School of Public Health) and in the United Kingdom (John Yudkin, founder of the first dedicated nutrition department in Europe) that sugar was a very likely cause of heart disease and diabetes, they were in direct conflict with the great majority who believed fat was the problem. The sugar industry used that conflict to its advantage. It launched a methodical public relations campaign with the goal of assuring that no “confirmed scientific evidence link[ed] sugar” to chronic disease. To do so, it recruited and compensated influential researchers in the anti-fat camp (from Harvard, the University of Minnesota, the University of Washington, and elsewhere) to publish authoritative reviews and a 90-page journal supplement, Sugar in the Diet of Man, that reinforced the message that dietary fat was the problem and sugar was not. These documents played a pivotal role in the FDA’s 1976 preliminary assessment that sugar should retain its status as Generally Recognized as Safe (GRAS). And that decision, along with sugar-industry-funded research and industry-funded researchers, then played the critical role in forming the consensus that sugar was benign.
The first USDA Dietary Guidelines, released in 1980, cautioned against consuming “too much” sugar—a tautological statement that could apply to any food. They also stated that “too much sugar does not seem to cause diabetes,” acknowledging the possibility and some controversy. In 1985, the second edition of the Guidelines dropped the conditional and proclaimed unambiguously “too much sugar in your diet does not cause diabetes.” Worth noting is that much of the relevant research done on this question between 1980 and 1985 was carried out at the USDA’s own Carbohydrate Nutrition Laboratory, and the researchers involved vehemently disagreed with this conclusion. By then, though, researchers studying the question of sugar and health were already tarred with the perception of having crossed the line from legitimate science to quackery.
In 1986, the FDA officially concluded that sugar should retain its GRAS status on the basis that “no conclusive evidence demonstrates a hazard to the general public when sugars are consumed at the levels that are now current.” Seminal government reports that followed – from the Surgeon General’s Office (1988), the National Academy of Sciences (1989) and the British Committee on the Medical Aspects of Food Policy (1989) – all echoed this thinking. Dietary fat is a killer; sugar is a benign source of empty calories. To come to the latter conclusion, though, required that the authors of these reports equate “no conclusive evidence” with no evidence at all.
By doing so, they successfully transformed a consensus of opinion into dogma. Among the consequences may have been the worldwide epidemics of obesity and diabetes. Had the nutrition research community and the public health authorities in the 1970s and 1980s understood that scientific controversy can only be resolved by careful experiment, not the assignment of truth to a consensus, then both our medical and public health approaches to obesity and diabetes might have had some effect. We cannot say. Should they understand it now, as I would hope to be the case, we’re still looking at perhaps 20 years of concerted research before definitive answers can be obtained. As such, we’re still in the position of the USDA in the late 1970s, having to act on incomplete and uncertain information, and still left having to make difficult decisions about who has the best sense of the data.
At the core of this kind of nutrition controversy is a fundamental philosophical conflict between the requirements of an effective public health policy and the requirements of science. The belief that imminent action must be taken, that it would in fact be irresponsible not to act, may be incompatible with the institutionalized skepticism and rigor required to establish reliable knowledge. The urgency required to prevent imminent deaths can mean that public health authorities do not have the time to gather what they would often describe to me in my research as “definitive scientific evidence,” sometimes perceived naively as dotting the i’s and crossing the t’s.
Good scientists will argue that the absence of definitive scientific evidence means that the truth remains unknown and, therefore, so does any clearly beneficial path of action, one with minimal and acceptable risks. Science is about acknowledging uncertainty. Effective public health policy requires that the public be convinced to change their behavior for their own good, which requires that we believe such a change is based on the strongest possible evidence. These perspectives may be right in their particular fields of application, but they can also be mutually exclusive. In 1999, when I first began investigating these controversies for the journal Science, the then director of the NIH’s office of disease prevention, William Harlan, captured the conflict this way: “We’re all being pushed by people who say, ‘Give me the answer. Is it or isn’t it?’ They don’t want the answer after we finish a study in five years. They want it now. No equivocation … [And so] we constantly get pushed into positions we may not want to be in and cannot justify scientifically.” A former director of the Center for Food Safety and Applied Nutrition at the FDA put it even more simply: “science,” he said, “is a destabilizing force in public policy.”
While the obesity and diabetes epidemics have finally motivated authorities, most notably the WHO, to advocate limited consumption of sugary beverages, they still do so based on the thinking that the sugar in these beverages provides only empty calories consumed in excess. As such, we’re back to the USDA’s admonishment from 1980: Avoid “too much sugar” and all will be well enough.
The minority opinion is much the same as it was forty years ago: sugar is far worse than a benign source of empty calories. It’s a toxin, albeit not a quick-acting toxin, but one that does its damage over years and decades. It may be the primary cause of diabetes, independent of its calories, and perhaps of obesity as well. If we didn’t consume sugar, by this thinking, both obesity and diabetes would be relatively rare conditions, just as lung cancer would be in the absence of cigarettes.
Another fact that has to be considered when institutions transform a consensus into a dogma is that reliable science not only should move slowly, but inevitably it does. The notion that sugar is a toxin is based on a scientific understanding that was evolving from the 1960s onward. It did so without the attention given to the also evolving research on dietary fat, cholesterol, and heart diseases. When the FDA, USDA, and other organizations embraced the dietary fat consensus, this research and its implications was still virtually unknown to those involved. Its transition to reliable textbook science only began afterward. Only with the full awareness of the obesity epidemic that emerged in the late 1990s have the broader public health implications of this research come to be partially embraced, but by then these new implications had to be reconciled with the assumption that dietary fat is the real problem.
The sugar hypothesis is based on the fact that the fructose component of sugar (and high fructose corn syrup) is metabolized primarily in the liver, resulting in a chain of effects that is speculative but still compelling: first the liver converts the fructose into fat, and that fat accumulates in liver cells (hence a condition, now also epidemic, known as non-alcoholic fatty liver disease). This in turn results in a condition called insulin resistance, which is the fundamental defect in type 2 diabetes. And that insulin resistance is accompanied by a cluster of metabolic abnormalities (high blood pressure, low HDL cholesterol, glucose intolerance) that is now known as metabolic syndrome. It can be thought of as a pre-diabetic condition and is so strongly associated with obesity that it may play a causal role there as well. (The first symptom that a physician is instructed to look for in diagnosing metabolic syndrome is increasing central adiposity – i.e., the patient is getting fatter) The CDC estimates that 75 million Americans have metabolic syndrome. Insulin resistance and metabolic syndromes are powerful predictors of heart disease. So the assumption is that what causes one, causes all. Most authorities believe that metabolic syndrome, like obesity, is caused by excess calories – eating too much and sedentary behavior. Experiments in both animals and humans suggest that sugar-rich diets should be at least a prime suspect.
If sugar consumption does cause insulin resistance and metabolic syndrome, then the dietary prescriptions of the past forty years are tragically incorrect. The question is what we can do about it now, when the evidence is still ambiguous and when the sugar-is-toxic camp is still very much a minority. The research necessary to nail it down would take years to decades to complete and is not even on the radar screen of the funding agencies.
Once again we’re in the position of having to take action based on uncertain evidence amid what is now a public health crisis. The twin epidemics of obesity and diabetes worldwide are a “slow-motion disaster,” in the words of Margaret Chan, director general of the World Health Organization. Chan gave a keynote address at an October meeting of the National Academy of Medicine and predicted with virtual certainty that the situation would continue getting worse for the foreseeable future.
Whatever actions we take, though, are like to have unintended consequences. One of them is further ingraining the precedent for taking other actions that are likely to be wrong and damaging as well. Clearly rigorous research is necessary to clarify the inherent uncertainty in the nature of a healthy diet and specifically the role of sugar in human health and disease. We have to know beyond reasonable doubt the dietary triggers of obesity and diabetes. If sugar plays a unique role, then we have to act. But that research will take time and a concerted effort that is not even on the radar screen of the National Institutes of Health. If that research is guided by the same influential authorities who have overseen the last forty years of dietary guidance and the epidemics of obesity and diabetes coincident with them, and if that thinking is indeed wrong, further errors are likely to be propagated.
If this were any other crisis, if this were any other public health failure, we would be assembling task forces of unbiased, independent researchers, perhaps from outside the field if necessary, to rigorously assess the evidence, determine what errors have indeed been made, and decide what to do next. It’s the only solution I can see to an intractable problem.