How do they figure out the number of calories in a recipe?

You may have noticed a lot of restaurants suddenly listing calorie counts on their menus or signage. In November 2014, the U.S. Food and Drug Administration (FDA) required chain restaurants, grocery stores serving prepared foods, vending machines with 20 or more locations and other food providers to post calorie counts on their menus and menu boards by Dec. 1, 2016 [source: Federal Register]. But how are these places coming up with their caloric information?

There are a number of private companies that analyze food samples to determine their caloric count. One, QC Laboratories in Pennsylvania, charged $700 per sample in 2014 if customers wanted the item independently tested to determine its calorie count and other nutritional information. If clients only wanted a simple calorie count via the use of a standardized database, the price was $150 to $200. In the latter instance, food providers may find it easiest to do the work themselves. All they need to do is add together a menu item's calories, then divide by the number of servings [source: Samuelson].

A quick example is a simple PBJ sandwich. According to the USDA's National Nutrient Database for Standard Reference, the calorie counts are:

2 slices of white bread: 154 calories

2 tablespoons peanut butter: 191 calories

1 tablespoon jelly: 56 calories

Total calories: 401

If the above sandwich is served in a half-sized portion, it would contain 200.5 calories. The only hitch to this system is that databases use average calorie counts for their foods — for instance averaging the calories in Skippy, Smuckers and Jif creamy peanut butters to come up with a calorie count for "creamy peanut butter" — so they're not 100 percent accurate. Because of this, food providers sometimes prefer to pay for the pricier independent testing [sources: Samuelson, Schechner].

Yet even if a restaurant has every menu item tested, calorie counts can still be a little off. Researchers from Tufts University tested foods from 42 restaurants in a 2011 study, and found 20 percent actually contained 100 more calories than advertised. Most of the undercounting was on the items labelled as "diet" or "light." Other foods actually contained fewer calories than advertised [source: Avila and Marshall].

In general, the researchers found that fast-food places have more accurate calorie counts, as much of their food is prepackaged and created in a formulaic style. Sit-down restaurants, where dishes are prepared individually, have more problems with accurate calorie counts. It's easier for portion size to differ slightly when each dish is made by hand, and often by different people.

Recipe developers, cookbook companies and the like use databases to help determine the calorie counts in their dishes, says Jessica Cox, a registered dietician and recipe developer who has consulted on cookbooks. But since cooking a food can affect both the number of available portions and their caloric content, you have to make sure you know whether to look at a food's raw calorie count or cooked calorie count.

When you're preparing a fatty, 3-pound (1.4-kilogram) pork roast, for example, some of the fat will cook off, reducing the weight of the roast. So if Cox had to analyze your pork roast recipe, she would first find the calories for a 3-pound (1.4-kilogram) raw roast, using the USDA's National Nutrient Database for Standard Reference. Then she would prepare the roast to determine its yield, or final weight, which in turn determines the number of portions available. The standard serving size for meat is 2-3 ounces (57-85 grams) [source: American Heart Association]. Let's say your roast cooks down to 2.5 pounds (1.1 kilograms). That would mean you now have fewer portions available to eat — about 13 servings in the cooked version versus 16 servings in the raw roast.

Interestingly, each 3-ounce serving of your cooked roast will actually have more calories than a 3-ounce serving of its raw cousin. That's because the raw meat contains both water and fat, says Cox, and more water is lost during the cooking process than fat. This means the resulting portions will contain a higher concentration of fat, and thus calories. Here's the comparison using the USDA's National Nutrient Database for Standard Reference:

Cox says recipe developers and testers also make sure they're only looking at the calories in the portions of a food available for consumption. So they'll only count the calories in the trimmed asparagus, for example, not those in the woody portions tossed away before cooking. They will even determine the percentage of a marinade that's left in the bowl after the meat is removed, subtracting out those calories from the dish's total count.

One caveat for consumers: Not all cookbook publishers or recipe websites hire experts like Cox to ensure their recipes' nutritional information is as accurate as possible. "It is impossible to know how different publications are calculating their calories," she says, adding the subject of calorie counts "is a kind of gray thing."

A lot of time and effort have gone into determining calorie counts over the years. But new research suggests they may be wrong. More precisely, the research, done by Richard Wrangham of Harvard University, indicates cooking food increases its calorie count, possibly by as much as 25 to 50 percent. And he's not talking about the water-versus-fat issue dietician Jessica Cox mentioned in the previous section.

Wrangham asserts past scientists never considered the caloric cost involved in the digestive process. Your body spends a lot of energy digesting, say, a raw carrot or raw piece of beef. But if you cook these foods, they soften, and your body expends fewer calories digesting them. Thus, the net effect is that you've taken in more calories from a cooked meal of beef and carrots than from the raw version [source: Wrangham].

Another aspect of cooking food that past researchers overlooked concerns digestive enzymes. Our body can only digest what it can get its hands on. A raw potato, for example, has tightly packed starches that our digestive enzymes can't access. When cooked, however, those starches are gelatinized, releasing sugar-based molecules that our bodies can now feast upon, adding to our caloric intake. Similarly, cooking beef opens up its muscle proteins to our digestive enzymes. In addition, preparing food for cooking by chopping, grinding and the like also makes them more accessible to digestive enzymes [sources: Wrangham, You Beauty].

Wrangham and his associates also say their studies show the Atwater system for determining calories is flawed. For one thing, it ignores the bacteria in our stomach. When we eat food, our stomach's bacteria steal some of it to essentially feed themselves. Thus, less is left over for our body as a whole to take in, meaning we're getting fewer calories. The Atwater system also doesn't look at the effect of processing, assigning the same caloric value to, say, whole-grain wheat and highly milled flour. But since our bodies will use more energy digesting the whole-grain wheat, it should be assigned a lower caloric value.

Don't think this means a healthy diet should be composed of all raw foods, though. Yes, our bodies work harder to digest them, so eating a lot of raw foods means you're taking in fewer calories — definitely a bonus if you're trying to lose weight. But if you only eat raw foods, that can be problematic. One study Wrangham looked at showed those who ate a 100 percent raw-food diet were unhealthy. The average woman didn't have a functioning menstrual cycle, and half of the women stopped getting their period at all, suggesting they were taking in far too few calories. Moderation is the key, he says; a diet of both raw and processed foods is best.