Part of our weekly "In Focus" series—stepping back, looking closer.
For the first time in two decades, the U.S. Food and Drug Administration has proposed sweeping changes to the familiar nutrition labels on food packages. If the changes are approved, labels would display calories per serving in very large, hard-to-ignore numbers. Nutritionists applaud the change, convinced that prominent calorie counts will help people lose weight and keep it off. But are those calorie counts accurate?
Maybe not, says Harvard University scientist Richard Wrangham, who believes that many of the official numbers are wrong. "Where foods are highly processed, such as white bread or a Twinkie, the calories on the package are probably reliable," he says. "But for less processed foods, you're probably getting fewer calories than the official caloric value." Wrangham thinks some numbers may be off by 30 percent or more.
That's because the calories listed on labels are determined by a method that ignores whether the food has been processed, cooked, or otherwise made more easily digested. Wrangham says the physical structure of food influences how much of it the body absorbs. Foods that are harder to digest, like chewy whole grains or raw kale, ultimately provide fewer calories than processed foods, such as wheat bread made from pulverized flour or a kale smoothie that's been liquified in a blender.
By misleading consumers into thinking they're getting the same number of calories from a given weight of nutrients, regardless of whether they've been processed, Wrangham says the calorie counts on food labels may be steering us toward energy-dense foods instead of the whole foods that could fill us up without adding on pounds.
Lighting a Fire
Wrangham's contention is controversial, and that's partly because he's not a nutritionist. He's a professor of biological anthropology—not a field that often weighs in with diet advice. The journey that took him from studying human evolution to critiquing food labels began in the 1970s, when he was studying the feeding behavior of wild chimpanzees. Out of curiosity, he tried to eat like a chimpanzee for a day at a time—a task that proved all but impossible, because the available raw foods were so difficult to chew and digest.
That set him to wondering whether cooking, which makes foods more digestible, played a breakthrough role in human evolution. In a 2009 book entitled Catching Fire: How Cooking Made Us Human, Wrangham argues that the invention of cooking—by breaking down nutrients and making foods easier to eat and metabolize—made it possible for early humans to get enough calories to meet the growing energy demands of increasingly big brains. (Read National Geographic Fellow Paul Salopek's interview with Wrangham on the original paleo diet.
Several recent findings support Wrangham's idea. At the Federal University of Rio de Janeiro in Brazil, neurologist Suzana Herculano-Houzel and her colleagues analyzed the number of neurons in primate versus human brains, and the amount of energy required to fuel them. On a raw-food diet, chimps spend 7.3 hours a day eating. Gorillas spend 8.8 hours. If humans, with our energy-hungry brains, needed to depend on a raw-food diet, we'd have to spend 9.3 hours eating.
"When we looked at the energy intake, we realized that great apes could not afford a larger brain. Neither could humans," she explains. "The difference is that, at least 1.5 million years ago, our ancestors learned to cook."
The Raw Truth
Further proof comes from studies of "raw-foodists"—people who try to eat most of their food uncooked and unprocessed. Those who eat a purely raw-food diet quickly show signs of chronic energy deficiency and weight loss. "They are starving themselves in terms of calories," says Wrangham. One German study of 513 people who shifted from a cooked to mostly raw-food diet lost an average of 26.5 pounds for women and 21.8 pounds for men. Over time, women stopped menstruating because of the energy shortfall.
Researchers have long known that cooking makes the calories in starchy foods more available. In a 2011 study of mice, one of Wrangham's colleagues, Harvard researcher Rachel Carmody, showed that cooked meat also provides more calories than raw meat. Mice fed cooked meat gained weight compared to mice given the same amount of raw meat. That makes sense, says Wrangham, because cooking in effect "predigests" food so the body requires fewer calories to metabolize it, making the energy available for other functions.
What's on the Label
Wrangham's beef with food labels is that the official calorie counts don't take into account the energy required to digest those foods. Indeed, the method that's still used to measure calories—and many of the official numbers—dates back to research done in the late 19th and early 20th century by chemist Wilbur Olin Atwater. Over the years, researchers have refined the so-called Atwater convention to make it more accurate. But nutritionists acknowledge that the numbers are estimates at best.
And some are proving to be wrong. Several years ago, scientist David Baer and a research team at the USDA's Agricultural Research Service began to look at nuts. "Using a new experimental design, we found that a serving of pistachios provides 5 percent fewer calories than the food label indicates," says Baer. "Almonds provide 20 percent less."
The reason, he suspects, is that fat in nuts is contained in cell walls that the human digestive tract can't easily break down. Some of it passes right through. His team has just launched a new study to compare raw almonds, chopped almonds, and almond butter. "What we expect is that with almond butter, because it's physically ground up, the fat will be more readily available for digestion."
Baer thinks that more accurate calorie counts on nut labels would encourage more people to eat them and reap their nutritional benefits. (Recent studies show that nuts protect against heart disease and some forms of cancer and may even help people maintain a healthier weight.)
But raw nuts may be a special case, he says, because they have unusually large amounts of fat, much more than in most plant foods. He doubts that discrepancies as large as 20 percent are likely to show up with other foods. Still, most plant foods are also very high in fiber, and researchers don't fully understand how fiber affects the number of available calories. Baer hopes to get funding for studies that will look at the calories provided by beans and other legumes, which should help to answer that question.
"Good Enough" to be Helpful
For now, most nutritionists say the difference between calories contained in a food and the amount absorbed by the body shouldn't make a huge different to anyone hoping to maintain a healthy weight. "Though cooking foods does make some nutrients more available, in terms of weight management the difference between raw and cooked foods isn't going to be very important," says Baer.
Marion Nestle, professor of nutrition and food studies at New York University and co-author of Why Calories Count: From Science to Politics, agrees. "Calorie measurements are approximations," she says. "They're good enough because one bite more or less will have a bigger effect on the number of calories than mistakes in measurement."
Despite quibbles over calories, the experts are in agreement when it comes to dietary advice. Cooking and processing foods may have made us human, but today the proliferation of processed and highly refined food is making us fat and unhealthy. Public health experts hope that prominently posted calorie counts will steer people toward smarter choices. Wrangham advocates for more accurate calorie numbers for raw and unprocessed foods—vegetables, whole grains, fruit, and nuts—because he thinks lower numbers would encourage people to eat more of them.
That's something every mainstream nutritionist would applaud.