When I weighed 300 pounds, I counted calories like it was my job. I’d eat the “right” number, exercise an hour and a half a day, and I was still fat. It took me years to figure out the math was missing something, and it’s been fun watching science catch up.
A study that came out in Nature Metabolism on October 5th is going to make the “calories in, calories out” bros pretty mad.
A team at Virginia Tech put together two small meals of about 300 calories each. One was eggs, a banana, some dried cranberries, a bit of cheese, and sea salt. The other one was deli turkey, instant mashed potatoes, a chocolate chip cookie, frosted wheat cereal, veggie chips, and a tiny bit of white bread with peanut butter and jelly.
The impressive part is how carefully they matched the two plates. Calories, carbohydrates, fat, protein, fiber, sodium, and water all were within about 1% of each other. They matched glycemic index and glycemic load too, which is honestly kind of absurd when you think about trying to pull that off with a cookie and a banana. They even added 12 grams of water to the ultraprocessed plate to make the weights line up. These scientists were serious.
So on paper you have two identical meals. Any calorie app would score them the same.
They brought in 57 young, healthy, normal weight adults (average age 26) and had 32 of them eat both meals on different days, in random order, inside a sealed metabolic chamber that measures what you breathe out. That tells the researchers how many calories you’re burning minute by minute and whether you’re running on carbs or fat. They drew blood eight times over the next three hours.
The participants rated both meals about the same. Same liking, same craving, same hunger and fullness afterward. If you’d asked them, they would have told you the meals were interchangeable.
The blood work said otherwise. The glucose response looked about the same between meals across the three hours, which is exactly what you’d expect when glycemic load is matched. Insulin went its own way and was clearly higher after the ultraprocessed meal from about 40 minutes through two hours.
They had the same blood sugar, but it took more insulin to get there. The authors call that “a pattern consistent with a relative reduction in insulin sensitivity.” The pancreas had to work harder for the exact same number of carbs.
For the nerds: respiratory exchange ratio was lower after the processed meal, so people burned more fat and fewer carbs, and total energy expenditure was higher. You’d normally expect a clean shift toward carb oxidation after 37 grams of available carbohydrate. The processed meal blunted that shift. Same 300 calories, handled like a different meal.
Oh, and if more fat burning and more calories burned sound like a win, look at the timing. Right after a carb meal, a healthy metabolism switches over to burning those carbs. Staying on fat while the meal’s carbs go unburned and insulin stays high is the delayed fuel switch researchers see in insulin resistance and type 2 diabetes. And whatever extra calories people burned after one 300 calorie meal get swamped by what comes next, as you’ll see below.
This study is related to Dr. Kevin Hall’s famous 2019 NIH study, where people lived in a metabolic ward and ate as much as they wanted from either an ultraprocessed or an unprocessed menu matched for nutrients. On the processed menu they ate about 500 more calories a day and gained about two pounds in two weeks. Hall showed what happens to intake. This new study shows what happens inside you after a single meal, before you even get to overeating.
I love CGMs. They can tell you a lot about your metabolism. But in this case, a CGM would have drawn two nearly identical lines for these meals and told you both were fine. All the action happened in insulin, and a CGM only reads glucose. Plenty of people are making food decisions off a flat CGM curve while their insulin is doing other things.
You can measure insulin, and you can lower the insulin hit from almost any meal with a few simple changes, some of which take less than ten minutes.
Here’s what’s coming up:
How to test your own insulin, the exact numbers I aim for, and the advanced test that flags insulin resistance years before your fasting glucose changes
The eating trick that cut post-meal glucose by 37% in a Weill Cornell study, with no change in what people ate
The 10-minute habit after meals that lowers both glucose and insulin
The two-second grocery store test for spotting food your body will treat as ultraprocessed
My swap for every item on that processed plate
