Ultra-processed food: is it really that bad?

This is general information, not medical advice. Always discuss your blood values with your doctor.

1. 1

   What counts as “ultra-processed”?

   Most studies use the NOVA system, which sorts food by degree of industrial processing into four groups: unprocessed or minimally processed, processed culinary ingredients, processed food and ultra-processed food. Examples in the studies include sugary breakfast cereals, ready meals, soft drinks, breakfast bars and a ready-made lasagne. The category is broad. The first author of the UPDATE trial notes that not every ultra-processed product is unhealthy based on its nutritional profile.

2. 2

   Observational research: a consistent link, low-quality evidence

   An umbrella review of 45 meta-analyses with about 9.9 million participants (Lane et al., BMJ 2024) found a link between more ultra-processed food and higher risk for 32 of 45 outcomes (71%), including mortality, cancer and cardiovascular, metabolic and mental health outcomes. According to the lead author, the strongest analyses showed about 50% higher risk of cardiovascular death and of common mental disorders. The quality of the evidence was low, though: by GRADE, 22 analyses were low quality, 19 very low and only 4 moderate. There were also no pooled analyses of randomised trials. So these are associations, not proven cause.

3. 3

   Experiments: you eat more of it

   In an inpatient RCT in 20 adults (Hall et al., Cell Metabolism 2019), participants ate an ultra-processed diet for two weeks and an unprocessed diet for two weeks, as much as they liked. The diets were matched for energy density, macronutrients, sugar, sodium and fibre, among other things. On the ultra-processed diet they ate 508 kcal a day more on average and gained 0.9 kg, versus losing 0.9 kg on the unprocessed diet. In the UPDATE trial (Dicken et al., Nature Medicine 2025), 55 adults with overweight ate a minimally processed or an ultra-processed diet following the UK Eatwell guidance for 8 weeks each, in random order. Both led to weight loss, but the minimally processed diet more (−2.06% versus −1.05%, P = 0.024), through less fat mass and body water, with no change in fat-free mass. Both studies are small and short: they mainly show that the effect exists, not how large it is in the long run.

4. 4

   What do you do with this?

   The UPDATE trial shows that even ultra-processed food that meets dietary guidelines can lead to weight loss, just less than minimally processed food. So the research does not say you must cut everything that is ultra-processed. It does suggest that more minimally processed food in your day helps with weight loss. Participants also reported better control of cravings, especially for savoury food, on the minimally processed diet.

Updated in October 2026