U.S. Dietary Guidelines: An Evidence-Free Zone.

O 7 January 2016, the U.S. Department of Health and Human Services and Department of Agriculture released the Dietary Guidelines for Americans 2015–2020 (1). The preliminary report (released in February 2015) had generated considerable media attention by reversing decades of dogma with the statement that “cholesterol is not a nutrient of concern for overconsumption.” Incredibly, in the final 2015 report, this statement has been removed, instead suggesting that “individuals should eat as little dietary cholesterol as possible.” Which version should we believe? How can the same committee arrive at diametrically opposite conclusions? Now that the final report is available, it is prudent to examine how, for decades, the U.S. medical establishment has erroneously advised the population to severely limit cholesterol intake and to consider whether other conventional dietary advice will eventually prove faulty. How strong is the scientific evidence supporting the current guidelines? The 2015 advisory committee was charged to “provide science-based advice” on nutrition and physical activity to “promote health across the lifespan and reduce the risk for major chronic diseases in the U.S. population.” Most of the recommendations are similar to prior guidelines, advising the population to limit intake of sodium, saturated fat (substituting unsaturated fats), and simple sugars and to increase consumption of fruits, vegetables, and nuts. However, a detailed review of the new guidelines confirms a disturbing reality: the nearly complete absence of high-quality randomized, controlled clinical trials (RCTs) studying meaningful clinical outcomes for dietary interventions. The report repeatedly makes recommendations based on observational studies and surrogate end points, failing to distinguish between recommendations based on expert consensus rather than high-quality RCTs. Unfortunately, the current and past U.S. dietary guidelines represent a nearly evidence-free zone. The lack of high-quality RCTs has left dietary advice to cult-like advocates, often with opposite recommendations. One group advises virtually complete elimination of carbohydrates from the diet, whereas others promote a virtually fat-free diet. A search of online bookstores and Web sites reveals an unlimited choice of diets, all with extraordinary claims for incredible weight loss and health benefits. The peer-reviewed medical literature also disappoints. One observational study, the Nurses' Health Study (NHS), has generated a plethora of questionable dietary claims. One NHS report claims that eating 1 ounce of nuts twice per week reduces the risk for pancreatic cancer by 35% (2) and another claims a 33% reduction in the risk for chronic obstructive pulmonary disease for the top quintile of consumption of whole grains, polyunsaturated fatty acids, nuts, and longchain -3 fats and low intakes of red processed meats, refined grains, and sugar-sweetened drinks (3). Yet another NHS report claims that daily consumption of more than 2 servings of artificially sweetened soda is independently associated with doubling the risk for a 30% or greater decline in renal function (4). These types of poorly controlled observational studies would be difficult to publish in the peerreviewed literature in any other field, but they are often reported with dramatic headlines by respected news organizations. Findings that suggest harm are particularly attractive to the media, such as a published study that claimed aspartame consumption doubles the risk for multiple myeloma (5). Such outrageous claims strain the credulity of thoughtful scientists but have little difficulty finding a journal that will publish them and less difficulty finding media outlets that will bring this “science” to public attention. Typically, dietary studies rely on a similar and flawed method, use of periodic dietary questionnaires to ascertain the eating patterns of participants. Recall bias and residual confounding plague such methods. There would be less interest in cult diets and poor-quality studies if nutritional research included properly designed and executed RCTs, but few exist. How did the American medical establishment embark on a decades-long misadventure about dietary fat and cholesterol? Many observers and a popular author (6) trace the current state of confusion to the renowned Seven Countries Study directed by Ancel Keys. Begun in 1956 and funded by a grant from the U.S. Public Health Service, the study was first published in 1970 and linked intake of saturated fat and cholesterol to the risk for coronary disease (7). Before the study, Keys had already aggressively promoted the concept that dietary fat and cholesterol were closely related to the development of heart disease. He even appeared on the cover of Time magazine in 1961, advocating a low-fat diet as the solution to the coronary heart disease epidemic. Critics have suggested that the Seven Countries Study was biased in favor of the hypothesis that dietary fat and cholesterol were critical factors in coronary disease (6). The study examined heart disease rates in Italy, Greece, Yugoslavia, Finland, the Netherlands, Japan, and the United States. Yet data were available for 22 countries. The researchers omitted countries, such France, where consumption of total and saturated fat are very high but the risk for heart disease remains low. Even before the publication of the Seven Countries Study, the American Heart Association (AHA) took up the cause, recommending that Americans reduce dietary fat intake and substitute corn or soybean oil for butter. Soon, margarine (with large amounts of trans