A new method will show how much junk food is in your diet - by urine analysis

American scientists have made a big step in studying how much we depend on ultra-processed food. They have discovered that it is possible to track how much of such food a person has eaten in their blood and urine. And this isn't science fiction - it's the result of a major scientific study published in the journal PLOS Medicine.

Details: Abar L, et al. Identification and validation of poly-metabolite scores for diets high in ultra-processed foods: An observational study and post-hoc randomised controlled crossover-feeding trial.PLOS Medicine (2025). DOI: 10.1371/journal.pmed.1004560

🍟What are ultra-processed foods?

They are foods that have been industrially processed and contain ingredients you won't find in your home kitchen: flavour enhancers, colourings, preservatives and flavourings. Chips, soda, sugary cereals, frozen pizzas and ready-to-eat dinners are classics of the genre. In the US, such products make up almost 60% of the adult diet.

🧪What did the researchers find?

The team, led by Erica Loftfield from the US National Cancer Institute, analysed blood and urine samples from more than 700 people over the age of 50. All participants also filled out detailed dietary questionnaires over the course of a year.

It turned out that in biological fluids can be found hundreds of molecules - metabolites - which are directly related to the consumption of ultra-processed foods. Based on this data, scientists have compiled a "metabolic score" - a set of 28 indicators in the blood and 33 in the urine, by which you can fairly accurately predict the proportion of such food in the diet.

🔍 It's not just "had crisps yesterday"

It's not about a single product, but a whole food pattern. The study found that two amino acid compounds and one type of carbohydrate were particularly common among the metabolites. One of the markers even appeared to be potentially linked to the risk of developing type 2 diabetes.

To confirm their findings, the team tested the markers on data from another experiment: in 2019, 20 people lived at an NIH centre for a month, receiving an alternating diet of ultra-processed and "clean" foods. Researchers pinpointed exactly when participants ate "junk" food - simply by biomarkers.

🧬What does this do?

If this technique is refined, it could be a real breakthrough. It could be used to conduct more accurate studies of the impact of food on health - including in projects that have already been completed, where blood and urine samples have been left behind. This will help find out how much ultra-processed food increases the risk of cancer, diabetes and other diseases.

Scientists say there is not yet enough funding for large-scale research. But scientific and public attention to the topic is growing.

"We're seeing tremendous interest in the question: is ultra-processed food harmful and how exactly does it affect health?" - says Loftfield.