Does ultra-processed food harm your health? What the studies show

Trials find people eat extra calories and gain weight on ultra-processed diets. Large studies link high intake to heart disease, diabetes and depression.

Published October 2026

Short answer

Ultra-processed diets led people to eat several hundred extra calories a day and gain weight in controlled trials. Cohort studies link high intake to heart disease, type 2 diabetes, depression and earlier death, but can't show cause. Evidence grade B (Moderate). Several small, short controlled feeding trials agree on higher calorie intake and weight gain (downgraded for sample size and no blinding); long-term disease links come from cohort studies only.

Key takeaways

In an inpatient NIH trial, people ate about 508 extra calories a day on a matched ultra-processed diet and gained about 0.9 kg in two weeks [1].
A 2025 English trial found less weight loss over eight weeks on an ultra-processed version of the same healthy-eating plan: about 1 kg less than on the minimally processed version [2].
Umbrella reviews of cohort studies link high ultra-processed food intake to heart disease, type 2 diabetes, depression and earlier death, but these studies can't show cause [3][4].
An analysis of 18 cohorts covering over a million people found the highest intake linked to a 15% higher risk of death from any cause [5].
The NOVA ultra-processed category is broad, including foods such as mass-produced bread and breakfast cereal, and researchers are still debating how to define it [2][6].

The short version

Controlled feeding trials, in which people eat only study meals, find that ultra-processed diets lead to several hundred extra calories a day and more weight gain over a few weeks [1][2]. Large observational studies link high ultra-processed food intake to heart disease, type 2 diabetes, depression and earlier death, but they can't show cause and effect [3][4]. The definition of ultra-processed is broad and debated [6], and the trials so far have lasted weeks, not years, and measured weight and blood markers rather than disease [7].

What the studies show

The feeding trials. In the best-known experiment, 20 adults stayed at the NIH Clinical Center and ate an ultra-processed diet for two weeks and an unprocessed one for two weeks, in random order, with the meals matched for calories offered, energy density, macronutrients, sugar, sodium and fibre [1]. Eating freely, they took in 508 kcal a day more on the ultra-processed diet and gained 0.9 kg, then lost 0.9 kg on the unprocessed diet [1]. A 2024 crossover trial in nine overweight Japanese men found 813.5 extra kcal a day and 1.1 kg more weight gain over one-week periods [8]. In the 2025 UPDATE trial, 55 adults in England ate ultra-processed or minimally processed versions of the UK's Eatwell Guide for eight weeks each; weight fell on both diets, but by 1.01 percentage points less on ultra-processed, a difference of about 0.96 kg, with about 1 kg less fat-mass loss [2]. A 54-hour whole-room calorimeter trial in 21 young adults found a protein-enriched ultra-processed diet cut intake by about 196 kcal a day versus a standard ultra-processed diet, yet it did not prevent overeating [9]. One 2025 crossover trial found higher body weight and a higher LDL:HDL ratio on ultra-processed food regardless of calorie load [10].

The cohort evidence. A 2024 BMJ umbrella review, a study of meta-analyses, pooled 45 analyses covering 9,888,373 people; higher ultra-processed food intake was linked to 32 of 45 health outcomes, with the strongest links to cardiovascular disease mortality, 50% higher, and type 2 diabetes, 12% higher per 10% increase in dietary share [3]. Other suggestive links included 21% higher all-cause mortality, 66% higher heart-disease mortality, 55% higher obesity and 22% higher depression [3]. A second 2024 umbrella review, updating 39 meta-analyses with 122 studies, found convincing evidence for kidney-function decline and wheezing in children and adolescents, suggestive evidence for diabetes, overweight, obesity, depression and common mental disorders, and no study reporting a beneficial association [4]. A third umbrella review found moderate-certainty evidence linking each 50 g a day of ultra-processed food to slightly higher risks of death, cardiovascular disease and colorectal cancer [11]. Individual meta-analyses put numbers on several of those links: across 18 cohorts with 1,148,387 people and 173,107 deaths, the highest intake was linked to 15% higher all-cause mortality, while each 10% step up in intake was linked to about 10% higher risk [5]. Across 12 cohorts, the highest intake was linked to 48% higher type 2 diabetes risk [12], and pooled Korean cohorts plus 17 prospective studies found 10-11% higher risk per 10% increment [13]. For depression, six prospective cohorts gave 32% higher risk with high intake [14], and a study emulating a trial in 11,192 older adults linked four or more servings a day to a 10% higher risk of depressive symptoms [15]. For cardiovascular events, an 11-paper analysis found 35% higher risk at the highest intake [16].

The nuances

The ultra-processed label is broad: NOVA's category includes breakfast cereals, sweets and mass-produced bread [2]. A review of six classification systems found that none clearly specifies which processing techniques to look for, and researchers have proposed alternatives, including one with quantitative cut-offs for sugar, fat and salt [6]. In one meta-analysis, ultra-processed cereals and breads and packaged savoury snacks were linked to slightly lower diabetes risk, while processed meats were linked to 34% higher risk [12].

The cohort findings can't show that processing itself causes disease: the 2025 umbrella review said it remains unclear whether processing raises health risks or whether ultra-processed intake is a marker of overall poor diet quality [11], and only four of the 45 pooled analyses in the BMJ review were rated moderate quality, with 22 low and 19 very low [3]. The trials have their own limits: they last one to eight weeks, cannot be blinded, and measured weight, calories and blood markers rather than disease [1][7]. Not every marker moved the same way: in the UPDATE trial, LDL cholesterol fell slightly more on the ultra-processed diet, while weight, body fat and triglycerides favoured the minimally processed diet [2]. Within one type of food, how it is eaten also seems to matter: in a trial of 69 adults, a slow-eating, low-energy-density ultra-processed lunch led to 573 fewer calories than a fast, high-energy-density one [17].

What this means for you

The most direct evidence is about weight and calories: in controlled settings, ultra-processed diets led people to eat more and gain more weight over weeks [1][2]. The long-term disease links come from observational studies, so they can't show that eating any particular product causes harm [11]. When the same healthy-eating guidance was used to build both diets, the minimally processed version produced more weight loss [2], and people who cut back on ultra-processed food ate fewer calories in the trial evidence reviewed here [7]. No trial has yet measured whether cutting back prevents heart disease or diabetes; the studies so far ran for weeks and measured weight and blood markers [7]. The Ultra-processed food topic page has the full breakdown of the evidence and its limits. If you manage a condition such as diabetes or heart disease, it is worth talking to a doctor or dietitian before making big changes to how you eat [3].

Go deeper

Ultra-processed food — Controlled feeding trials find people eat several hundred extra calories a day on ultra-processed diets and lose less weight than on minimally processed ones.