The Minimally vs Ultra Processed Food trial – Diet and Health Today

In Summary

* The results of a diet clinical trial have just been published by authors from University College London and University College London Hospital (UCLH).

* The trial aimed to compare two diets both meeting the UK ‘Eatwell’ Guide (EWG) recommendations – one would be based on minimally processed food (MPF); the other would be based on ultra processed food. Both diets would be consumed for eight weeks.

* The trial was based on a flawed assumption – that the UK ‘Eatwell’ Guide is healthy. It is not. It is deficient in several nutrients. It was designed by the UPF industry.

* The trial recruited 55 staff from UCLH. The typical participant was a female nurse with an average BMI of 32.7 who was consuming 67% of her diet in the form of UPF at baseline.

* The trial was a crossover design. Participants were randomised to do the MPF diet or the UPF diet first and then to switch to the other diet after a four week washout period in between.

* To meet some of the EWG recommendations, both diets were designed to provide 2,000 calories per day, <35% fat, c. 50% carbohydrate, c. 15% protein, fewer than 6 grams of salt per day, at least 30g of fibre per day and 5 fruit and veg a day. The amount of MPF and UPF in each diet was intended to be the only difference between the diets. All food was delivered to the participants at home to ensure adherence to the diets.

* The trial was “ad libitum”, meaning that people could eat as much as they liked. To achieve this, 4,000 calories were provided each day – double what should have been needed. This lost control over what was actually consumed.

* The outcome was that the diets actually consumed differed from each other in calorie intake, fat intake, carbohydrate intake, protein intake, salt intake, free sugar intake, fibre intake and fruit and veg intake and likely more factors of which we’re not aware.

* The results claimed that the MPF diet generated double the weight loss of the UPF diet. Both weight losses were small – 1.84kg vs 0.88kg – and the UPF numbers could have been confounded by water retention and other factors in favour of the MPF diet.

* Sweeping claims were made in the discussion section of the paper. None of them withstood scrutiny. The specific inclusion criteria for the participants made the results non-generalisable. Providing all food made the results non real-world. The fact that every aspect of the diet changed, not just the one being measured, made the claims invalid for MPF vs UPF.

* I predict that this paper will be much cited as evidence that i) dietary guidelines should be followed and ii) that a new recommendation should be added to these guidelines telling people to avoid UPF. It has not made the case for this.

Introduction

Clinical trials in the nutrition world are rare and so they are usually worth reviewing. This week’s note is about a clinical trial that made news headlines. It’s worth reviewing for what didn’t make the headlines. Up front, I should say that this was an impressive project, which would have taken time, effort, money and commitment. I applaud the researchers for their endeavour but there were many limitations and issues. I fear it will be much cited in the future to make claims that it shouldn’t.

The trial was called UPDATE, which was a loose acronym for “investigating the effects of Ultra-Processed versus minimally processed Diets following UK dietAry guidance on healTh outcomEs.” The paper about the trial was published on 4th August 2025 in Nature Medicine. It was called “Ultraprocessed or minimally processed diets following healthy dietary guidelines on weight and cardiometabolic health: a randomized, crossover trial” (Ref 1). The lead author was Dicken. There were 21 authors named on the paper. All but two were affiliated to University College London (UCL) and/or University College London Hospital. One of the non-UCL authors was Dr Kevin Hall. He is worth noting as he has conducted a number of crossover clinical trials in the field of diet, weight and health and four of his papers were referenced in this paper. Another author was Dr Chris van Tulleken who has written a book called Ultra-processed PEOPLE (Ref 2).

UCL published a press-release about the trial on its website, entitled “Less processed diet may be more beneficial for weight loss” (Ref 3). The widespread media reporting of the paper matched the UCL statement closely. I saw a number of articles that ran with the same headline about “double weight loss.” The CNN headline was “Eating minimally processed meals doubles weight loss even when ultraprocessed foods are healthy, study finds” (Ref 4). A medical news site reported “Weight loss doubles when eating minimally processed foods over ultra-processed alternatives” (Ref 5). The New York Times ran with “Avoiding Ultraprocessed Foods Might Double Weight Loss” (Ref 6). We’ll see how UNimpressive even double weight loss was in a moment.

We need to define processed food and what was meant by minimally processed and ultra processed food in this study. Then we’ll cover the basics of the study – the aim, what it did and who it involved. We’ll cover the diets that were intended to be consumed and the diets that were actually consumed. We’ll move on to the results and then I’ll share my concerns about the trial and the legacy that I forecast it will leave.

Processed food

There have been a number of attempts to define processed food and to categorise different degrees of food processing. The NOVA system is commonly used and it was used by Dicken et al for this trial. The NOVA system was first presented in a 2010 paper by C. A. Monteiro et al (Ref 7) and further papers have been published by Monteiro and colleagues since. The Dicken et al paper referenced the Monteiro et al paper from 2019 (Ref 8). Monteiro et al identified four categories of food processing. Supplemental Table 1 in the 2019 paper listed all foods in the four categories (Ref 9). I’ll share the title, description and a couple of examples in each group below and then I’ll share a diagram from this 2022 paper to illustrate the four groups (Ref 10).

Group 1) Unprocessed or minimally processed foods.

Unprocessed foods are edible parts of plants (fruits, seeds, leaves, stems, roots, tubers) or of animals (muscle, offal, eggs, milk), and also fungi, algae and water, after separation from nature. E.g., an apple, a steak, a mushroom.

Minimally processed foods are unprocessed foods altered by industrial processes such as removal of inedible or unwanted parts, drying, crushing, grinding, fractioning, roasting, boiling, pasteurisation, refrigeration, freezing, placing in containers, vacuum packaging, non-alcoholic fermentation, and other methods that do not add salt, sugar, oils or fats or other food substances to the original food. E.g., white rice, pasteurised milk.

Group 2) Processed culinary ingredients.

Substances obtained directly from group 1 foods or from nature by industrial processes such as pressing, centrifuging, refining, extracting or mining. E.g., oil from olives, starch from corn.

Group 3) Processed foods.

Products made by adding salt, oil, sugar or other group 2 ingredients to group 1 foods. E.g., canned vegetables in brine, salted nuts, smoked meat or fish, fruits in syrup.

Group 4) Ultra-processed foods.

Formulations of ingredients, mostly of exclusive industrial use, that result from a series of industrial processes (hence ‘ultra-processed’), many requiring sophisticated equipment and technology. E.g., confectionery, cakes, biscuits, ready meals, instant soups etc.