What if everything we think about obesity is wrong?

The U.S. government is approaching the study of obesity all wrong.

According to leadership at the U.S. Department of Health and Human Services, food companies are to blame — they’re engineering their snacks, fast food and sweets to addict us the same way cocaine and nicotine do. Others assessing the epidemic say people overeat because they lack willpower and could stop if they really wanted to.

While the blame game has been good for stirring public outrage, it hasn’t led to anything useful for improving our health. Meanwhile, some health officials have been undermining potentially useful research on novel causes of overeating. A leader in diet and obesity research quit his NIH job after he says officials interfered with his ability to discuss his results — which he said didn’t align with the agency’s talking points.

Those blaming Americans’ obesity on their lack of willpower also get it wrong. That viewpoint came up on a recent episode of Bill Maher’s show, during a discussion with Casey Means, U.S. President Donald Trump’s nominee for surgeon general, who also argued that junk food was sometimes engineered to hook people. Maher scoffed, countering that McDonald’s is "delicious” and that’s why people choose to eat it — no fancy science needed to explain that.

There is credible evidence that our obesity rate — about 40% — is related to our heavy consumption of fast food, packaged and instant foods, sweets and sweetened drinks — all of which were recently lumped into the giant category of ultraprocessed food. These so-called UPFs make up about 53% of the average American diet, according to the Centers for Disease Control and Prevention.

Research by Kevin Hall — the scientist who quit his NIH job — linked ultraprocessed foods to weight gain, but the reasons were more complicated than originally thought.

In Hall’s study, volunteers confined to a food lab were fed one of two diets. One consisted mostly of ultraprocessed foods — turkey bacon, fries, instant noodles, chicken nuggets and similar items. The other was primarily minimally processed foods. Participants were allowed to eat as much as they wanted for two weeks, then they switched to the other diet.

During the ultraprocessed phase, participants consumed an average of 500 more calories a day and gained a couple of pounds on average. The diets contained the same amounts of protein, carbohydrates, fats, sodium and even fiber — so those components alone couldn’t be blamed for weight gain.

Hall said the volunteers were asked whether the ultraprocessed diet was tastier or more enjoyable to eat and they reported that both diets were equally appealing. That undercuts the simplistic notion that junk food causes overeating simply because it’s delicious. "There’s something more interesting going on,” he told me.

The language of addiction is common in NIH-sponsored science and political rhetoric. At a Senate hearing last year, Robert Califf, former commissioner of the Food and Drug Administration, told senators that food companies were creating addictive products, and HHS Secretary Robert F. Kennedy Jr. has repeatedly blamed the industry for knowingly "poisoning” Americans with its products.

Since publishing the results of his study in 2019, Hall has been working to uncover what the "something more interesting going on” might be. In 2023, he and a team of collaborators conducted additional analysis of the original data and found an important clue. The researchers found the foods that people tended to overconsume followed a pattern: They were typically high in sodium and starch, sweet and fatty or fatty and high in sodium, according to Tera Fazzino, a psychologist and addiction expert at the University of Kansas who was part of the team. She defined these foods as "hyperpalatable.”

I asked Fazzino whether "hyperpalatable” was the same as delicious. Not quite.

"I’m in Italy right now,” she said, where she described the food as "so fresh and so good and so rewarding.” Hyperpalatable ultraprocessed food might be addictive, she explained, or it might disrupt the body’s signaling mechanisms that tell people they’re full.

Hall tried to get at the addiction question in another experiment. Volunteers were given an ultraprocessed milkshake, after which their brains were scanned to detect any dopamine surges — similar to those sometimes triggered by addictive drugs. The participants varied in body size, ranging from thin to obese. According to Hall, some of the individuals exhibited an increase in dopamine. However, the response didn’t appear to correlate with their level of obesity.

He said the NIH initially planned to issue a press release on the study, but then pulled back. Later, he got a call from the New York Times requesting an interview — and press office officials forbade him from speaking with the reporter. When he issued a written statement in response to the Times, he said it was altered by the agency to play down the study’s size and significance.

Hall said leaders at HHS, which oversees the NIH, didn’t want him to talk about his work, "because it went against the narrative that ultraprocessed foods are addictive.” HHS spokesman Andrew Nixon said via email that "at no point was Dr. Hall’s ability to share or advocate for his ideas restricted — his academic freedom and free speech were fully respected throughout,” and he called Hall’s concerns "fake news.”

But Hall’s story is consistent with what he’s told other news outlets. He’s also spoken to reporters about previous work — as opposed to communicating via written statement as in this case.

For some experts, there’s no question that food addiction is real. "I’ve treated people who have this compulsive relationship with these ultraprocessed foods,” said Ashley Gearhardt, an addiction specialist at the University of Michigan. Sometimes people eat until they get sick, she said, and she sees the same cravings and compulsive behavior with junk food that drives alcoholics. And people don’t binge on apples or avocados or chicken thighs. It’s almost always things like pizza, candy bars, ice cream or donuts.

But that still doesn’t explain why the participants in Hall’s study ate those extra 500 calories. In ongoing research, Hall and his collaborators are comparing a diet of ordinary ultraprocessed foods with one that consists of foods that have been reformulated so that they aren’t "hyperpalatable.” He’s also examining caloric density, a characteristic of ultraprocessed foods that people tend to overeat.

Preliminary results suggest there are ways to make inexpensive, convenient and shelf-stable food that’s no less tasty than conventional ultraprocessed fare but won’t pack on pounds. That’s a result that doesn’t cast blame — and might lead to real changes that actually make us healthier.