Is 'ultra-processed' a useful thing to avoid, or just a label?

Nina TeicholzPhD · investigative science journalist and author

A label. Judge the sugar and the flour, not the processing.

What it rests onNOVA classifies a home-baked cookie as merely processed and a store-bought one as ultra-processed, while the science says the cookie is harmful either way because of its sugar and flour. The formal definition uses "mostly" and "many" with no thresholds, which she argues invites lobbying and lets graders' preconceptions decide. The category carries an anti-meat bias: a hamburger counts as ultra-processed, per the system's author in a 2017 interview with her, because of the foods eaten alongside it, while a grilled cheese does not. Her policy prediction is that a rule built on it would remove school lunch meat and leave the chips.

Kevin D. HallPhD · metabolism researcher; formerly Section Chief, Integrative Physiology Section, NIDDK, National Institutes of Health

Useful. When you match everything the nutrition panel measures, people still eat 500 calories a day more.

What it rests onA randomised inpatient crossover trial in which meals were matched for presented calories, energy density, macronutrients, sugar, sodium and fibre still produced 508 kcal/day more spontaneous intake on the ultra-processed arm, with weight tracking intake. The nutrient panel therefore does not capture what drives the extra eating, and matching it is not enough. His own hedge is in the paper's final sentence: limiting ultra-processed foods "may be an effective strategy", n = 20, two weeks per arm, and the trial cannot speak to long-term outcomes.