New Nutrition Study Changes Nothing

Why the science of healthy eating appears confusing—but isn’t

A hand holds a backlit egg.
Hannibal Hanschke / Reutuers

If you’ve ever been on the internet, you’ve noticed that some things are popular, and other things aren’t. The popular ones have something in common. It’s not quality, or importance, or accuracy, but novelty.

An example of this is Moby-Dick. It’s a timeless novel by an acclaimed writer, and most people haven’t read it. The complete text is free on the internet. You could be reading it right now. But are you? And how many of your Facebook friends shared Moby-Dick today? Probably not more than one or two.

Obsession with novelty has been called “neophilia.” The term was used as early as 1965 in a story by J.D. Salinger, who was apparently wary of it. But being attracted to novelty doesn’t necessarily mean readers are dumb or attention-deficient. To some degree, novelty-seeking is evolutionarily adaptive and associated with good health.

Readers are trained to expect and value newness in what they read. I can’t blame people who see in the term news an implicit promise to deliver something new. When a story fails in that, comments flood in, like “This is not news!” and “This writer looks like he’s 12!” and “The eclipse is real!”

Neophilia is a problem for science, though. And especially the sort of science pertaining to nutrition. Demand for newness leads writers and publishers to focus on narratives that upend conventional wisdom. If new research doesn’t change or challenge the way readers think about the world, why is it a story worth publishing? Eggs are in, and now they’re gone. Butter? It’s back. Every six weeks, The New York Times is legally obligated to tell us either that breakfast isn't important or that skipping it causes death.

The effect of all this, day after day, year after year, is a perception that all kinds of contradictory evidence is coming up every day—and that each bit is roughly equally valid.

Of course, it’s not. Eating in ways that are good for our bodies isn’t conceptually complicated. It’s complicated by money and time and access—but eating based on scientific findings is not. Though recently you might have heard otherwise. There is new news about dietary health, and that news is more important than the typical weekly nutrition news. But it doesn’t upend conventional wisdom.

This week, the world learned the results of an enormous study of food and health. The study included 135,335 people from 18 different countries across five continents who were followed for seven years. It’s known as the Prospective Urban Rural Epidemiology study, or PURE, and the results were presented at the European Society of Cardiology meeting in Barcelona and published in The Lancet.

The practically important findings were that the healthiest people in the world had diets that are full of fruits, beans, seeds, vegetables, and whole grains, and low in refined carbohydrates and sugar.

Melissa d'Arabian / AP

As a writer and a reader, though, this is very boring. If I pitched that to my editor, he would laugh at me. What is new here? Why is this interesting? You know what would be novel? You getting fired! Now get out there and find me a story, dammit!

Let’s look at the internet. The coverage of the study was mostly fine, but the headlines alone promise novelty. Stat ran “Huge New Study Casts Doubt on Conventional Wisdom About Fat and Carbs.” Reuters ran “Study Challenges Conventional Wisdom on Fats, Fruits, and Vegetables.” Medscape ran “PURE Shakes Up Nutritional Field: Finds High Fat Intake Beneficial.” Even the editors at The Lancet used the headline “PURE Study Challenges the Definition of a Healthy Diet.”

Does it?

These assertions are all predicated on the idea that most people’s definition of “healthy diet” is “low fat.” That was the recommendation from the USDA in the 1990s. But given that there have been a lot of nutrition studies and books and stories since then, that idea has been disproved again and again. There is still some disagreement over exactly how much of various types of fats are optimal for human health in which populations, but the differences are academic.

In a press release, the study’s lead author, Mahshid Dehghan of McMaster University, described the study’s novelty relative to the World Health Organization’s dietary recommendations: “The current focus on promoting low-fat diets ignores the fact that most people’s diets in low- and middle-income countries are very high in carbohydrates, which seem to be linked to worse health outcomes,” she said, noting that in these countries, “guidelines should refocus their attention toward reducing carbohydrate intake, instead of focusing on reducing fats.”

Low-income countries eat diets that are high in starches like white rice and low in protein and fats because they lack access, not because they’re meticulously adhering to bad guidelines. This isn’t a finding that’s accurately extrapolated to anyone reading this article and deciding between a salad and a burger. Still, Dehghan’s conclusion, lumping together all people, for what it’s worth: “The best diets will include a balance of carbohydrates and fats—approximately 50 to 55 percent carbohydrates and around 35 percent total fat, including both saturated and unsaturated fats.”

“My hope is that our results will stop the whole population from feeling guilty if they eat fat in moderation,” said Salim Yusuf, another author on the study. “While very high fat intake—when it accounts for 40 percent or more of your dietary intake—may be bad, the average fat intake is about 30 percent, and that’s okay. We’re all afraid of saturated fat, but actually we shouldn't be.”

Are we all afraid of saturated fat? The evidence on that particular type of lipid has shifted over the past half-century. The story of the public perception of saturated fat is long and embattled, but essentially there was initially an argument over how harmful it was, and as more evidence has come to light, it became clear that saturated fat is like other types of macronutrients—good in moderation, bad in scarcity, and probably bad in overexposure. Some experts argue that it’s impossible to eat too much saturated fat, and a few outliers believe it should be significantly limited, but mostly there is agreement that it’s unwise to either chase or fear it.

The U.S. Dietary Guidelines still recommend that people limit saturated fat intake, and many experts find this inappropriate. These guidelines are issued by the Department of Agriculture, though, which is not a scientific entity. Its supporters argue that telling people to go wild with saturated fat leads to bad decisions: People take it to mean ice cream and fettuccine alfredo and bacon, not so much the saturated fats from nuts and avocados.

While it's likely that recommendations to limit saturated fat will be lifted entirely in the future, there is less chaos to the debate than it often seems. Even some doctors contribute to that perception. Last year, the chair of cardiology at Cleveland Clinic, Steven Nissen, wrote in the Annals of Internal Medicine that there is a “nearly complete absence of high-quality randomized, controlled clinical trials (RCTs) studying meaningful clinical outcomes for dietary interventions.” This 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.”

That sort of obsession with randomized, controlled trials is common and misguided, a case well made this week in The New England Journal of Medicine by the former CDC director Tom Frieden. When measuring diet, for example, lifelong randomized, controlled trials are impossible. Even if people would volunteer to change their diets for a decade or so—a period long enough that rates of death and cancer and heart attacks could be meaningful—it would be impossible to keep the research subjects blinded. Our perceptions of how well we’re eating change how we behave in a lot of other ways.

Insofar as randomized, controlled trials for diet can be done, many have. One, also published in The New England Journal of Medicine, compared food with medication for treating type 2 diabetes. For many people, the lifestyle intervention was more effective than the common drug metformin. And their diet was high in vegetables and fruits and nuts—more or less the same diet that proved effective in this week’s study and many other observational studies.

This should all be reassuring. There is a tremendous amount that’s unknown about how food affects our health, and there are subtle differences in every study, but on the whole it’s generally consistent. There is some disagreement over precise numbers about which macronutrients are best for which populations with which diseases, and discovery is happening constantly, and this is all part of science.

But almost never is the news upending conventional wisdom. And to regularly pretend that it does invites readers to lose faith in the ability of science to discern truth, or the idea that truth exists at all.

All that said, I should end with something novel. There’s always something new and interesting in any study, and there were a few things in this one. For instance, eating vegetables in any state was associated with good health, but as compared to cooked vegetables, eating raw vegetables was more strongly linked to a lower risk of death (during the study) compared to cooked vegetable intake.

As researcher Victoria Miller of McMaster University put it, “Our results indicate that recommendations should emphasize raw vegetable intake over cooked.”

There is a novel idea. Dietary guidelines usually don’t encourage people to prioritize raw vegetables over cooked. Maybe they should. That could be a headline. “Cooking Your Vegetables? Welcome to Early Death.”

... Except that raw vegetables often don’t taste as good as cooked vegetables, and cooking is the basis for human social interaction and culture.

People are complex, and the ways we perceive and communicate and relate to one another are complex. But the basic agreement on what to eat for the health of people and the planet is not: diverse, naturally high-fiber, minimally processed foods, mostly plants. Eating based on macronutrient numbers isn’t likely useful and is easily distracting. Sufficient amounts of protein and fats come to most people in wealthy countries without thinking.