We Already Knew Ultra-Processed Food Was Bad. This Study Shows How Bad.

So what's actually new?

Not the headline. In 2024, a BMJ umbrella review of nearly 9.9 million people found direct links between ultra-processed foods and 32 health outcomes, spanning mortality, cancer, and mental, cardiovascular, metabolic, and digestive health.

What this study adds is a dose. Instead of comparing heavy eaters with light eaters, it measures risk per gram of daily intake. That turns "eat less junk" into something you can actually track.

Why it deserves attention

It's big and global. Nearly 9 million people across four continents is hard to dismiss as a fluke.

It's prospective. Diets were recorded before people got sick, which cuts down on the bias of asking sick people to remember what they used to eat.

The pattern scales. Risk climbs steadily as intake climbs. A dose-response pattern like that is harder to explain away than a single comparison.

It lines up with a controlled trial. In a 2019 NIH experiment, people on an ultra-processed diet ate about 500 more calories a day than on a minimally processed one and gained 2 pounds on average, even though the meals were matched for calories, sugars, fiber, fat and carbohydrates.

Where it falls short

It can't prove cause and effect. People who eat the most ultra-processed food also tend to differ in income, sleep, stress and exercise. Researchers adjust for what they can, but no statistical model catches everything.

The diet data is self-reported. Intake was assessed by food frequency questionnaires, 24-hour dietary recall or dietary history, and people are famously bad at remembering what they ate.

"Ultra-processed" is a very big tent. Packaged whole-grain bread and a can of soda can land in the same category. The study tells you about the category, not about the specific foods in your cart.

Grams are a strange yardstick. A 12-ounce can of soda weighs roughly 365 grams, so 100 grams is less than a third of a can. A one-ounce bag of chips is about 28 grams. Drinks rack up grams fast; snacks don't.

Relative risk isn't absolute risk. A 24% higher risk sounds alarming, but what it means for you depends on your starting risk, which headlines rarely mention.

Some scientists want a pause. After the 2024 review, one expert argued that the quality of the evidence linking ultra-processed food to poor health is weak by standard grading criteria. A 2026 paper in the European Journal of Clinical Nutrition went further, calling for systematic reviews and meta-analyses of observational ultra-processed food studies to be suspended until methods improve.

What to do with this

Start small. Choose two or three ultra-processed foods you eat regularly and swap them for whole or minimally processed versions, dietitian Maya Feller told ABC News. No pantry purge required.

Look at your drinks first. Because the study measures by weight, sugary drinks are the fastest way to rack up grams, and one of the easiest swaps to make.
Flip the package over. Feller's rule: if the front says blueberries, the ingredient list should say blueberries.

Use the freezer aisle. Frozen vegetables, fruits and grains with nothing added, lower-sodium canned foods, and affordable staples like pasta, rice and beans keep costs down without adding more ultra-processed food.

Don't fear all processing. Canned vegetables, cheese and freshly made bread count as processed, not ultra-processed. Slicing, freezing and canning aren't the problem.

The bottom line

This study doesn't overturn anything. It sharpens it. The evidence still can't say exactly why ultra-processed food is linked to worse health, but after millions of participants and a controlled trial pointing in the same direction, cutting back is a low-risk bet. You don't need perfection. You need fewer grams.

More health news, minus the hype, at Several.com lifestyle.