Ever wondered where those calorie counts on the side of food boxes come from? Often times, the number is so specific, it seems like there must be some sort of science behind the numbers. In fact, the U.S. Department of Agriculture (USDA) uses a specific set of criteria, called the Atwater system; to figure out how many calories are in the foods people eat. But questions are circulating about whether this system really provides accurate calorie counts. A new study out this week has come up with a new way of calculating calories in food and applied it to walnuts. They found that past estimates might not have reflected the true caloric composition of this healthy nut.

What is the Atwater system?

The process of counting calories (which uses a device called a “bomb calorimeter” that in short, lights food on fire and measures the calories of energy that food gives off as a result) is lengthy and expensive to perform on every single food that comes onto the market. The Atwater system was developed to provide a shorthand way of calculating the likely number of calories based on what was already known about certain food groups. The projected amount of energy released by fats, carbohydrates, and proteins per amount in food was standardized based on the many experiments that had been done. In addition, the overall percentage of food that was used for energy versus food that couldn’t be digested and was excreted was standardized for different food groups. That meant that food producers could come up with a pretty good approximation of the number of calories in food just by using a bomb calorimeter, and the Atwater system could put together these standard numbers without having to test every single food on humans (to see how the body digests and processes different foods).

Why was this study needed?

Even though the Atwater system does a pretty good job of figuring out how many calories are in food, there are a lot of problems with the way it does, so that can lead to inaccurate results. For example, it’s well known that different types of fats, sugars, and proteins give off slightly different amounts of energy, but the Atwater system just uses one number for each category. We also know that the composition of food affects the way it’s digested and that each food is slightly different in how well absorbed its nutrients are. But in the Atwater system, large groups of similar foods were all lumped together. That meant that the calorie calculation for walnuts assumed that walnuts were digested in the same way as beans, even though their digestion is likely very different. While a few calories might not seem like a big deal, incorrectly high calorie counts on food can turn diet-conscious eaters away from foods that might actually be better for them than they think. This is why the USDA wanted to retest walnuts more rigorously to see how their current calorie counts matched up.

How did the researchers test the food?

The research team recruited 18 people to test exactly how many calories the body extracted from walnuts they ate. The study was a “cross-over” study, which means that participants spent half of the study in the “experimental” group that ate walnuts and the other half of the study in the “control” group eating regular food that didn’t contain walnuts. Their results in each part were then compared. In the walnut portion of the study, the participants ate premade meals that included 1.5 servings of walnuts per day. The control group ate meals that had the exact same number of calories, but didn’t include walnuts. Individuals had nine days to get used to the diet and then all of their feces and urine were collected for a week.

The team took the feces and urine and tested them with a bomb calorimeter. Since the calorie counts were technically the same in both the walnut and control group, the researchers could figure out whether walnuts were truly 185 calories per serving or some amount less than expected based on how the diets matched up. The researchers could also calculate how digestible walnuts were compared to the standard that is also used for beans.

What did the team find?

The team’s analysis showed that walnuts were much less digestible than the original calorie estimates assumed with the standardized Atwater system. More unused energy, often in the form of fat, was found in the urine and feces of participants who ate more walnuts than in those who didn’t. The new calorie count for walnuts was 146 calories, about one-fifth less than predicted. Interestingly, the team found that the digestibility of walnuts was also different than those found in recent studies of pistachios and almonds, indicating that how easy to digest a nut is varies from nut type to nut type. The team thinks that this lowered calorie absorption (mostly from lower fat absorption) may be part of the reason why nuts seem to help with weight loss.

What does this mean for me?

This study shows us just how hard it can be to come up with true estimates of how many calories your body is getting when you eat something. While scientists have decent systems in place to get a ballpark figure, we’re learning that counts can vary widely based on the type and composition of a food. If you had any doubts about walnuts based on their calorie counts, this should allay your fears. They’ve been shown to help with weight loss, pack a lot of omega-3 fats, and are good for your heart health. A lowered calorie count is just one reason to add a few to your next meal.