
Crossover Trial Links 85% UPF Reduction to 3.15-Point PHQ-9 Drop in Depressed Adults
A small JAMA Psychiatry crossover trial found that sharply lowering ultra-processed food intake improved depression scores in adults with metabolic abnormalities. The within-subject design strengthens internal validity yet small size and self-report data constrain generalizability. Larger confirmatory trials are needed to establish mechanisms and clinical recommendations.
The study enrolled adults aged 18-80 with clinician-diagnosed depression, PHQ-9 scores of at least 5, high baseline UPF consumption, and at least one metabolic abnormality. Participants completed both their usual diet and a low-UPF diet in random order while keeping medications, therapy, exercise, and sleep unchanged; 19 of 20 completed both phases. Diets were assessed via self-report and the intervention achieved an average 85% reduction in UPF calories.
Depression scores fell significantly on three validated scales during the low-UPF phase, with PHQ-9 declining 3.15 points on average. Anxiety scores also decreased. The design controlled for individual confounders through within-subject comparison, yet the small sample, short duration, and reliance on self-reported intake limit causal inference about specific nutrients, additives, or energy density.
Observational data from the UK Biobank and a 2023 meta-analysis in Nutrients have previously associated highest UPF quartiles with 20-30% higher depression incidence, potentially via inflammation, gut microbiota shifts, and insulin resistance. The present trial adds interventional signal but cannot isolate whether benefits stem from removing additives, increasing whole foods, or weight change.
Larger parallel-group RCTs with objective dietary biomarkers and longer follow-up are required to confirm durability and identify active components before dietary guidelines can incorporate UPF reduction as an adjunct for depression.
Investigators: 12-week parallel RCT of 150 participants will show sustained PHQ-9 reduction of at least 2.5 points versus control diet by end of 2026
Sources (3)
- [1]Primary Source(https://jamanetwork.com/journals/jamapsychiatry/article-abstract/2823456)
- [2]Supporting Source(https://www.bmj.com/content/381/bmj-2022-073164)
- [3]Supporting Source(https://www.nature.com/articles/s41380-023-02017-7)