Processed Food and Hormones: The Stop Signal That Arrives Too Late

You’re standing at the counter and the packet is empty. You remember opening it. You do not remember eating most of what was inside.

That gap gets explained badly. Willpower, addiction, chemicals engineered in a lab to hijack your brain. The actual explanation is duller and considerably more useful, and a metabolic ward in Maryland worked it out with twenty people and a set of kitchen scales.

Think about a bad phone line. Satellite delay, two seconds of lag. Neither person is doing anything wrong and you still end up talking over each other constantly, because every reply lands after you’ve already moved on. Nobody is at fault. The timing is.

Your gut talks to your brain on a line like that.

What Processed Food and Hormones Actually Have to Do With Each Other

The claim you’ll usually see is that processed food disrupts your endocrine system, and it gets waved at testosterone, thyroid and insulin without much specificity.

The strongest evidence points somewhere narrower and more interesting. It’s about the hormones that manage appetite, the ones that tell you when to start and when to stop. Those signals are not damaged by processed food. They’re outrun by it.

Everything downstream, including the effects on testosterone that follow from carrying more body fat, runs through that one bottleneck.

The Study That Matched Everything

In 2019, Kevin Hall’s group at the National Institutes of Health ran the trial that settled the argument about whether processing itself does anything.

Twenty weight-stable adults, average age 31, moved into the NIH Clinical Center for four weeks. Two weeks on an ultra-processed diet, two weeks on an unprocessed one, in randomized order, each person acting as their own control. Three meals a day plus snacks, eat as much or as little as you want, every gram weighed.

The part that makes this study matter is what they controlled. The two diets were matched for presented calories, energy density, macronutrients, sugar, sodium and fiber. If processed food only causes problems because it contains more sugar and fat, matching those should erase the difference.

It didn’t. On the ultra-processed diet, people ate an extra 508 calories a day.

The extra came almost entirely from carbohydrate and fat, around 280 and 230 calories respectively. Protein intake was flat. Over two weeks, participants gained just under a kilogram on the ultra-processed diet and lost just under a kilogram on the unprocessed one, and how much weight each person moved tracked closely with how much extra they’d eaten.

Same calories on the plate. Same sugar. Same fiber on paper. Five hundred calories a day difference in what actually got eaten.

The Two Hormones That Moved

They took blood, which is where this stops being a story about calories and becomes one about signalling.

PYY is released by the gut as food arrives and it suppresses appetite. On the unprocessed diet, PYY was higher than on the ultra-processed diet and higher than at baseline. Ghrelin runs the other way, driving hunger, and on the unprocessed diet it dropped below baseline.

Both hormones moved in the direction of feeling satisfied sooner and staying satisfied longer, on the diet where people ate less without being asked to.

Which tells you the machinery was working. Give the same people real food and their appetite regulation performed exactly as designed. Nothing was broken.

Speed Is the Whole Problem

The other measurement they took explains the rest. Eating rate was significantly faster on the ultra-processed diet.

Gut hormone signalling is slow by design. Food has to arrive, be detected, trigger release, and reach the brain. That takes time measured in tens of minutes, which was fine for most of human history because food took time to chew.

The message got through. It just arrived after the plate was empty.

A Japanese trial published in 2024 tested this directly in adults with obesity and found an even bigger gap, around 813 extra calories a day on the ultra-processed diet. Their explanation was chewing. Participants chewed markedly less per calorie on the processed food, which is the mechanical version of the same point.

Processing is, at bottom, the removal of work. Fiber taken out, structure broken down, water added, texture softened. What gets removed is precisely the stuff that used to slow you down long enough for the reply to arrive.

The Thyroid Number That Looks Bad

Buried in the same results table is a finding that would make an alarming headline if you read it carelessly.

On the unprocessed diet, total T3 went down. That’s a thyroid hormone, and lower thyroid hormone sounds like damage. Real food lowered their thyroid function.

Except free T4 went up over the same period, and the people on that diet were in a mild energy deficit and losing weight. A drop in T3 during weight loss is a well-described adaptive response, the body trimming its metabolic rate to match reduced intake. Reading it as thyroid damage inverts what happened.

Worth holding onto whenever you see a hormone quoted in isolation. A number moving is not the same as a system failing, which is also why a single line on a blood panel rarely means what the headline says it means.

What This Does and Doesn’t Prove

Twenty people. That’s the first thing to say about a study this widely quoted.

The crossover design compensates for a lot, since each person served as their own control, and the Japanese trial found the same effect in a different population with a larger gap. But twenty is twenty, and four weeks is four weeks. Nobody has run this for a year.

The paper also carries a published correction. A documentation error in the diet order for one subject was found after publication. It didn’t touch the headline result on total daily intake, but it did change the reported breakdown across breakfast, lunch and dinner. That correction is on the record and almost nobody citing this study mentions it.

What the trial establishes is narrow and solid: holding the nutrition numbers constant, processing itself drives higher intake, and appetite hormones respond differently to the two diets. What it does not establish is any direct endocrine disruption from processed food, and it says nothing at all about testosterone.

Slowing the Line Down

If the problem is speed rather than chemistry, the fixes are about friction rather than restriction.

• Put something on the plate that needs chewing. Not as a virtue exercise. Chewing is the delay the signalling system was built around.
• Eat sitting down, off a plate. Eating from a packet while standing removes every natural stopping point, which is how the counter scene happens.
• Lead with the slowest thing. Whole vegetables, whole fruit, meat, eggs. Not because they’re moral, because they take longer.
• Change what’s within reach rather than what you intend. The trial worked because of what was available in the room, and changing your situation reliably beats changing your resolve.

None of that requires eliminating anything. The study didn’t test elimination and neither does this. It tested what happens when most of what’s in front of you takes longer to eat.

Fatigue is often tangled up in the same pattern, and if that’s the bigger complaint, the wider set of reasons energy stays low is a better place to start than a food rule. So is how much of the day you spend seated, which shapes body composition alongside anything on your plate.

When This Stops Being About Food

Some appetite changes belong with a doctor rather than a strategy. Unexplained weight loss or gain, appetite that has disappeared or become relentless, persistent thirst, or hunger alongside fatigue and frequent urination all warrant a conversation and some blood work.

And if eating has started to feel like something you’re at war with, if you’re tracking obsessively, eating in secret, or feeling shame afterwards, that’s a different problem entirely and it responds to proper support rather than better information. Speak to your doctor about it.

Two Seconds Behind

There’s a version of this topic that treats processed food as poison and eating it as a character failure. The metabolic ward found something less dramatic and harder to argue with. Same people, same appetite hormones, same willpower. Change what’s on the plate and they ate five hundred fewer calories a day without trying and without noticing.

Their bodies were never the problem. The real relationship between processed food and hormones is a timing one: the food was simply faster than the conversation.

You can’t speed up the line. You can put something in front of you that gives it time to answer.

References

1. Hall KD, Ayuketah A, Brychta R, et al. Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake. Cell Metabolism. 2019;30(1):67-77.e3. https://doi.org/10.1016/j.cmet.2019.05.008
2. Hall KD, Ayuketah A, Brychta R, et al. Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake (Erratum). Cell Metabolism. 2019;30(1):226. https://doi.org/10.1016/j.cmet.2019.05.020
3. Hamano S, Sawada M, Aihara M, et al. Ultra-processed foods cause weight gain and increased energy intake associated with reduced chewing frequency: A randomized, open-label, crossover study. Diabetes, Obesity and Metabolism. 2024;26(11):5431-5443. https://doi.org/10.1111/dom.15922

FAQ