Scientists have discovered specific molecules in blood and urine that reliably indicate how much ultra-processed food (UPF) a person eats. Published on May 20th in PLOS Medicine, the study, led by Erikka Loftfield of the National Cancer Institute, USA, and colleagues, could provide an objective way to measure UPF intake, reducing reliance on self-reported dietary data.
The study analyzed blood and urine samples from 718 older adults alongside detailed dietary records. Researchers identified chemical fingerprints, called poly-metabolite scores, associated with UPF consumption. Using 28 blood metabolites or 33 urine metabolites, they created scores that accurately predicted UPF intake. These scores were further validated in a controlled feeding study with 20 inpatients at the NIH Clinical Center, where they distinguished between high-UPF and no-UPF diets.
Ultra-processed foods make up over half of the average American diet, but their health impacts remain unclear due to challenges in tracking consumption. The new method offers a more precise tool for large-scale studies.
“The identified poly-metabolite scores could serve as objective measures of UPF intake in large population studies,” the authors noted. They emphasized the need for further evaluation in diverse populations to refine the scores.
This breakthrough could enhance research on UPFs’ health effects and provide insights into their role in chronic diseases. Future studies will explore these markers in broader populations to improve accuracy and applicability.

Add comment