Ultra-processed food and psoriasis: what a hundred thousand people's shopping actually showed

The finding that made the headlines

In 2025 a group working with the UK Biobank published the largest prospective look at this question so far. They took 121,019 participants aged 40 to 69 who did not have psoriasis, classified their reported diets by the NOVA system that sorts foods by degree of processing, and followed them for a median of twelve years, over which 1043 people developed psoriasis. Higher ultra-processed food intake was associated with a higher rate of developing it, and the authors concluded that this provides dietary guidance for preventing psoriasis and its related conditions ^[9].

That is a substantial piece of work and it deserves to be read as one. It is prospective, so the diet was recorded before the disease appeared, which rules out the commonest trap in nutrition research, people changing what they eat because they are already unwell. It is large. And its authors tested whether inflammation score and body mass index sat on the path between the food and the disease ^[9], which is the right question to ask of an association like this. If you were going to build a case that diet influences who gets psoriasis, this is the kind of study you would want.

It is also, and this is where the headlines stop reading, a study of who develops psoriasis, in people who did not have it. It says nothing directly about whether the same foods make existing psoriasis worse, or whether removing them makes it better. Those are separate questions and they need separate evidence.

Sources for this section: [9] Ultra-Processed Food Consumption and the Risk of Psoriasis: A Large Prospective Cohort Study

What the category is, and what it swallows

Ultra-processed is a research classification, not a moral one. The NOVA system that defines it groups foods by the extent and purpose of industrial processing, and the practical test its authors give for the top category is the ingredient list: a product is ultra-processed if it contains a substance never or rarely used in kitchens, such as high-fructose corn syrup, hydrogenated oils or hydrolysed proteins, or an additive there to make it palatable or appealing, such as flavours, colours, emulsifiers, sweeteners or thickeners ^[10]. Soft drinks and reconstituted meat products pass that test in the way most people would expect. So does a supermarket loaf with an emulsifier in it, because the definition turns on formulation and additives, not on nutritional content.

That breadth is a strength for epidemiology and a problem for a shopping list. A cohort can show that people whose diets are heavy in the category do worse on average. It cannot tell you which items in the category are responsible, and it cannot tell you that a loaf with an emulsifier in it is doing what a two-litre bottle of cola is doing. Treat the finding as a pointer towards the parts of the category that other evidence also names, and that is what the next section does.

Follow the sugar

Line up the other recent cohort findings and one ingredient keeps appearing. A second UK Biobank analysis, this time of 210,474 people followed for a median of over thirteen years, found that those in the highest quarter of free-sugar intake had a higher rate of developing psoriasis than those in the lowest, while those eating the most fibre and starch had a lower rate ^[7]. A cross-sectional survey of 257 adults with psoriasis in the UK found free sugars over the reference value and fibre under it, and people in the top quarter for free sugars from drinks were more likely to report severe disease ^[5].

Then the finding that comes closest to a treatment question. In a prospective cohort of people receiving psoriasis treatment, those who drank sugary drinks four or more times a week had a markedly lower chance of reaching a meaningful improvement at eight weeks, with an adjusted odds ratio of 0.24 for a 50% reduction in severity. High-fat foods showed a weaker, borderline association in the same direction ^[8]. That is still observation, not a trial, and the people who drink the most soft drink may differ in other ways. But it is the one study in this set that looks at people who already have psoriasis, on treatment, and it points where the others point.

So the picture that emerges from the whole set is not really about processing. It is about free sugar, and about sugar in drinks more than in food, with fibre running the other way. That is narrower, more actionable and better supported than a war on packaging.

Sources for this section: [5] Macronutrient intakes and associations with psoriasis severity: a cross-sectional analysis of the asking people with psoriasis about lifestyle and eating (APPLE) study · [7] Association between carbohydrate intake and the risk of psoriasis: a prospective cohort study based on UK Biobank · [8] Sugary Drinks and High-Fat Foods are Associated with Poor Treatment Response in Psoriasis: A Prospective Study in Shanghai

Diet quality, and who has the least of it

One more UK Biobank study fills in what people with psoriasis actually eat. Comparing 2613 people with psoriasis against the rest of the cohort, diet quality on the national guideline score was about the same, and adherence to a Mediterranean pattern was slightly higher in the psoriasis group. But the psoriasis group also reported more red and processed meat, more sodium, more free sugars and more alcohol, and within it the people with the most severe disease were less likely to meet the fruit and vegetable guideline and less likely to score on nuts. People with a comorbidity had lower diet quality on both scores ^[6].

Read that against the usual avoid-lists, which tend to name individual vegetables and dairy, and the mismatch is striking. The measured difference in the psoriasis diet is more processed meat, more sugar, more alcohol and fewer vegetables. The ultra-processed finding does not say which of its contents matter; the diet-quality data say which items people with psoriasis actually eat more of, and none of them is a vegetable.

Sources for this section: [6] Are psoriasis severity and comorbidities associated with diet quality? A cross-sectional analysis using UK Biobank

Association is a reason to act, not a proof that acting works

Everything above is observational. None of it is a trial in which people with psoriasis were assigned to give up ultra-processed food and measured afterwards, and that trial has not been run. The systematic reviews are candid about this: the 2025 review of dietary recommendations describes the evidence on diet and psoriasis as lacking in quality even as it lists sugar among the reported aggravators ^[4], and the National Psoriasis Foundation's recommendations, built on 55 studies, could support only weight reduction strongly and gluten-free eating conditionally, with everything else filed under low-quality data ^[2].

There is, however, one trial that bears on this indirectly. The 16-week randomised trial of the Mediterranean diet in psoriasis, in which the diet arm improved by a between-group difference of -3.4 on the severity index and nine of 19 people reached a 75% reduction against none of the controls ^[1], was a trial of a pattern whose protocol told participants to eliminate or limit sugary drinks, pastries and ultra-processed foods ^[11]. It did not test processing as a variable, and it cannot say which part of the package did the work. But the diet that improved plaques under measurement had those limits built in, and the clinical-practice review calls that pattern a good option for people with psoriasis on the strength of its anti-inflammatory and cardiovascular effects ^[3].

Sources for this section: [1] Mediterranean Diet and Patients With Psoriasis: The MEDIPSO Randomized Clinical Trial · [2] Dietary Recommendations for Adults With Psoriasis or Psoriatic Arthritis From the Medical Board of the National Psoriasis Foundation: A Systematic Review · [3] Nutrition and Psoriasis: The Latest Evidence and How to Approach Nutrition in Clinical Practice · [4] Evidence-based dietary recommendations for patients with psoriasis: A systematic review

The one change in the trolley

If you came here wanting to know what the evidence points at in a trolley, here is the shape of it. The sugary drink is the item the cohorts name most consistently, the one linked with poorer treatment response in people already being treated, and the one the trial diet limited. Processed meat is the item the diet-quality data show people with psoriasis eating more of, and the trial diet steered away from it. The label on the bread is the thing no study in this set singles out.

The pattern that has a trial is vegetables, fruit, legumes, whole grains, fish, olive oil and nuts. The free seven-day psoriasis week on this site is that pattern cooked into twenty-one meals with a shopping list, and it contains no soft drink for reasons that have nothing to do with a category and everything to do with the studies above. If a week of it suits you, the cookbook is the same pattern with a hundred recipes and a 28-day plan.