Former NHS director Kevin Hall became a household name after in 2019 he published a groundbreaking study showing that participants on a diet of ‘ultra-processed’ foods subconsciously consume an extra 500 calories a day, compared to participants who selected from unprocessed foods. His paper garned efusive praise by the media and podcasts as being an ‘obvious’ or self-evident explanation for the obesity crisis. The praise has continued to come in, with a 2026 headline in Yahoo Health proclaiming, “Kevin Hall sparked a cultural revolution against ultra-processed food.” But if only it were that simple though.
There are several issues with blaming obesity on ultra-processed foods. For one, there’s no universally accepted definition of what counts as “ultra-processed,” as opposed to simply processed. While foods like salami or crackers are technically processed, they don’t belong in the same category as, for example, a candy bar. Another issue is the tendency to confuse correlation with causation. If people who are obese tend to eat more candy bars, it’s easy to jump to the conclusion that candy bars cause obesity and are therefore uniquely ultra-processed. But that reasoning may be backwards–or at least incomplete. Just because two things occur together doesn’t mean one directly causes the other.
Second, obese people can eat a lot of ‘healthy’ food and can easily gain weight with that, such as cheeses, dairy, or meats. These are often considered healthy but are very calorie dense and easy to consume to excess when you’re hungry from dieting. If people are wired due to brain chemistry to eat too much, or their bodies are not burning enough calories due to metabolic dysfunction, avoiding ultra-processed foods will not be enough. Dr. Hall at least to his credit acknowledges the shortcomings of diet modification to treat obesity.
The problem also is food is too calorie dense in general, ultra-processed or not. For example, I was at the food store recently and a tiny package of salami has 300 calories. I could easily eat three of these in a single hour, having done so in the past. That is almost half the standard ‘2000 calories/day’ recommended guidelines eaten in just an hour. I’m sure a 300-pound obese man could easily eat more than three packages. On those obesity reality TV shows (e.g. My 600 Pound Life), obese participants eat too much of everything–not just desserts. They eat tons of pasta, meats, bread–everything.
Some may be surprised that the following foods have a greater calorie density than even Skittles:
-Salad dressing
-Butter
-Milk
-Cereal
-Oats, granola
-Red meat
-Pork
-Cheeses
-Olive oil
-Avocado
-Beef jerky
I’m not saying that calorie density is at the root of obesity, or that eating candy will lead to weight loss, but it’s little surprise that so many people fail at dieting when trying to ‘eat healthy’.
Fourth, Hall overlooks the major role of metabolism. I agree that increased appetite is a contributing factor for weight regain during dieting, but metabolism is possibly more important, as the latter determines where one will ultimately stall during dieting, relative to calories consumed and bodyweight. For many unfortunate individuals, this is set rather low, precluding sufficient weight loss despite diet compliance and ‘doing a lot of cardio’. I discuss such cases, and metabolism in more detail, in my 5,000-word post “Fast or slow metabolisms are not a “lie”. It’s time to put this myth to rest.”
There are also studies that show that obese people, especially women, are not consuming that many calories to maintain their obese weight, typically only 1,800-2,300 calories/day under supervised laboratory conditions, in order to remain ‘weight stable’ at an obese BMI. From my post, “The obesity trap, and common misconceptions about obesity:”
As I discuss earlier, a famous ’92 study which tracked obese dieters who were unable to lose weight found they were eating only 2,081 kcal/day. People cite the above study as evidence of obese people failing to track their calories when stalling, yet ignore or overlook the part that these people are not eating that much food to begin with, yet have still stalled and are obese.
Undercounting of calories is often blamed, but even when counting, these obese study participants were not eating much, only around 2,081 kcal/day–the same as the recommended FDA guidelines. This points to metabolism more so than food quality being the culprit.
This is quite disappointing because hunger and discomfort ramps up, but there is none of the enviable leanness to show for it. These people are flabby despite eating little, and just not losing weight anymore. Or any hope of leanness, or even just 15% bodyfat (which is not even that lean), requires cutting to absurdly few calories–way fewer than predicted by calculators. As it’s said, ‘many such cases’. If people are not eating much yet are still overweight or obese by BMI, blaming ultra-processed foods seems like a red herring, when the underlying problem is slow metabolism.
Moreover, no one afik, has ever reproduced Hall’s original study despite choruses of praise by the media and being featured on podcasts, books, and other media. It has been taken as literal fact or truth by credulous people looking for a scapegoat for obesity, and what can be more convenient than creating a boogeyman ‘Big Food’ to follow in the footsteps of ‘Big Tobacco’.
Overall, I think the main problem is his findings are just not that actionable. Sure, maybe ultra-processed food plays some role in obesity, but it’s dwarfed by other factors, or just not that helpful. Otherwise, we’d see better results applying this advice, and yet we don’t. Anecdotally on communities such as Reddit, and also according to studies, people universally struggle at weight loss no matter what diet they try, even when abstaining from junk food altogether. Thankfully, we now have GLP-1 drugs, which have finally brought sought weight loss for many of these people.