Thirty-three percent of study participants said they almost always look at calorie content on labels, but only 9 percent of them actually looked at the calorie count.
People who meticulously check the calorie counts on nutrition labels and restaurant menus are in for some bad news: the tallies may be wrong, experts say.
Recent studies show that the amount of pounding, slicing, mashing and perhaps even chewing that goes into preparing and eating food affects the number of calories people get. For some foods, a proportion of the calories in them remains "locked up" during digestion, and isn't used by the body. People also expend some of the energy from food just digesting it; and even the bacteria in people's guts steal a fraction of food's calories. None of these factors are accounted for in our current system for calculating calories, which dates back more than 100 years.
Scientists have always known that calorie counts are just estimates. And over the years, some scientists have called for changes to the system. Now, researchers are again shining a spotlight on the issue, saying an overhaul of the calorie count system is needed so consumers have a better idea of exactly how many calories they get from the food they eat.
"If we're going to put the information out there on the food label, it would be nice that it's accurate," said David Baer, a research physiologist at the U.S. Department of Agriculture's Human Nutrition Research Center in Beltsville, Md. In a study last year, Baer and colleagues showed that almonds have 20 percent fewer calories than previously estimated. Now, the researchers are considering retesting other foods, including some types of whole grains and legumes.
For the most part, the inaccuracies are small, but some foods may have actual caloric values that differ from the estimated values by as much as 50 percent, experts say. [See 9 Snack Foods: Healthy or Not?]
One way to measure a food's energy, or caloric content, is by burning it in a device called a bomb calorimeter. However, this method doesn't take into account the fact that humans lose some calories through urine and feces and as heat. Over the years, researchers have tried to figure out ways to account for these losses.
In the late 1800s and early 1900s, a man named Wilbur Atwater conducted experiments in which he calculated the number of calories in various diets, and collected people's feces to determine how many calories were wasted. Based on these experiments, Atwater concluded that proteins and carbohydrates have about 4 calories per gram, fats have 9 calories per gram, and alcohol has 7 calories per gram.
These values are still used today. Their existence means food manufacturers and restaurants can use a simple formula to calculate the calories in their foods.
However, these values are rough estimates. Certain foods, such as those high in fiber, are not digested as well, meaning the calories we get from them would be lower than those calculated using the formula. In the 1970s, researchers introduced modified Atwater values that were intended for specific foods, such as fruits, vegetables and beans.
More changes needed
While these changes are a good start, some experts say we should do more.
Research by Rachel Carmody, a postdoctoral fellow at Harvard University's FAS Center for Systems Biology in Cambridge, Mass., and colleagues, shows that food processing — eating a carrot that's pureed rather than whole, for example — changes the calories we get from it.
Food processing takes some of the work out of digestion, Carmody said, meaning that generally, a processed food will have more calories than an unprocessed food.
Calories in processed foods are likely close to the values that the Atwater system estimates.For example, if you eat a mashed potato that's been calculated by the Atwater system to contain 300 calories, you're likelygetting most of those calories, Carmody said. But if you eat a whole, unprocessed potato of the same size, you'll take in around 200 calories, she said.
The difference is biggest for starchy foods, like potatoes, and is lowest for meats, Carmody said. (The calories from unprocessed versus processed meats only differ by 5 to 10 percent, she said.)
The Atwater system also fails to account for structural differences in food that make some calories inaccessible to our bodies. For example, the almond study, which also accounted for calories lost in feces, suggested that some of the fat in whole almonds is locked away in a structure our bodies can't digest. While the Atwater system says a serving of whole almonds has about 170 calories, the almond study found it actually has about 130.
"Given that the Atwater system is treating essentially all foods the same, we aren’t getting a good perspective when it come times to make dietary choices," Carmody said.
When we digest food, we also give off energy as heat. The amount of heat we radiate depends upon the exact components of the food. For proteins, it's about 20 to 30 percent of the food's calories — so if we eat 100 calories worth of protein, we get about 80 calories from it, Carmody said. For fats, it's much less, about 0 to 3 percent, she said. (So if we eat 100 calories worth of fat, we'd get 97 of those calories.)
This month, Carmody and colleagues will give a presentation at the annual meeting of the American Association for the Advancement of Science in Boston that will discuss ways in which to improve the system for calculating calories.
Does it really matter for waistlines?
Some researchers say that, on the whole, the inaccuracies in calorie estimates don't make a big difference. "For most uses, I think they're good enough," said Malden Nesheim, professor of nutrition emeritus at Cornell University, in Ithaca, N.Y., and co-author of the book "Why Calories Count" (University of California Press, 2012).
People tend to eat a variety of foods, not just almonds or starches. So overestimating or underestimating the calories in one particular food will likely not have a huge impact on a person's daily calorie intake, Nesheim said.
And generally, the omissions in the Atwater system tend to result in overestimates, meaning they likely wouldn't interfere with weight loss.
"It would only be a problem for people who want to gain weight," said Mary Ellen Camire, a professor at the University of Maine's Department of Food Science & Human Nutrition in Orono.
But other researchers say the goal of a revision would be to give people as much accurate information as possible to help them make informed choices about food, Carmody said. Such a process could result in broad changes, such as new numbers for the total calories people need in a day.
"By getting a better understating of the effective calories in food, we'll get a better sense of human energy requirement," Carmody said.
A change to the calorie system would not be easy, Carmody said. And because of differences between individuals, it would be impossible to create a system that would work for everyone.
But researchers may be able to fill in some of the system's biggest gaps, such as the effects of food processing and heat loss, Carmody said.
"We can start to think of simple ways to improve [the system] that will be better for the average consumer," Carmody said.
Pass it on: An overhaul of the system we use to calculate calories is needed so people can better gauge how many calories they get from food.