Cutting Junk Food Improved Mood in New Trial

The first trial to test this simple dietary intervention

STUDY: Chagwedera DN et al, JAMA Psychiatry 2026

STUDY TYPE: Randomized crossover pilot trial (open label)

FUNDING: National Institutes of Health; UCSF Weill Institute for Neurosciences

Background

Drew Ramsey practices nutritional psychiatry, guiding patients toward a Mediterranean-style diet for depression, anxiety, and cognition. When I interviewed him for the Carlat Report, he shared that many patients see the biggest effect just by cutting junk food.

Past trials proved that this approach treats depression, but they required patients to increase healthier food while cutting the junk. This is the first to test cutting out junk by itself.

The Study

• 20 adults with depression and at least one metabolic problem who ate large amounts of ultra processed food (score ≥ 8 on sq-HPF). 19 completed the trial.
• With a cross-over design, each patient spent four weeks on a low ultra processed diet and another four weeks on their usual diet, in random order.
• At the start of the intervention period, a physician gave 45 minutes of education on reducing ultra processed foods (from Group 4 of the NOVA system) and checked in by phone on progress.
• Medications, therapy, exercise, and sleep were unchanged during the trial.
• Main outcomes: feasibility and acceptability. Mood scales were exploratory.

Results

• When ultraprocessed food was lowered, depression fell significantly, by 11 points on a clinician rated scale (the 60-point MADRS), compared to a 4 point drop during their usual diet.
• Patient rated scales for anxiety and depression fell 3 pointseach (the 27 point PHQ-9 and 21 point GAD-7).
• Ultraprocessed food intake fell 85% during the intervention.
• Patients rated the diet useful but only moderately easy.

Limitations

• Tiny sample, no blinding, and self-reported diet.
• Depression scores fell during the usual diet period too, from 13 to 9 on the PHQ-9.
• The 11-point MADRS gap is about four times what antidepressants show over placebo, which suggests expectation effects.
• The trial can’t tell whether sugar, fat, calories, or additives drove the change.

Practice Implications

1. This is a simple intervention for depression and physical health.
2. Start by assessing the diet with the sq-HPF scale below. This identifies patients likely to benefit, and engages them in nutritional awareness.
3. Use the NOVA classification to guide your intervention, focusing on the ultra processed Group 4. When buying packaged foods, the Open Food Facts app rates level of processing through the bar codes.
4. As junk food decreases, there’s more room for brain-healthy foods.

—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report

What’s Your Take? Share in Comments