In a 20-person pilot, cutting ultra-processed food tracked lower depression scores

What the research found

Study (pilot crossover). Chagwedera, Krystal, and colleagues report in JAMA Psychiatry (23 September 2026; doi 10.1001/jamapsychiatry.2026.2320; UCSF release the same day). Twenty adults with depression, high ultra-processed food intake, high BMI, and at least one of hypertension, diabetes, or elevated LDL cholesterol — mean age 43.7 years, range 18 to 80, about 70% women — spent four weeks on their usual food and four weeks on a low ultra-processed food pattern they shopped for themselves. Nineteen finished. Intake scores fell from about 10.1 to about 1.4, described as more than an 85% cut, and depression scores were moderately to largely lower on three measures during the low ultra-processed phase. Anxiety scores improved modestly. Activity, sleep, and usual care were held steady. Self-reported intake, no objective food assay, and no adverse events reported. A feasibility pilot, not proof that this food pattern treats depression. Educational nutrition notes.

Evidence level: Human study. A study in people, but an association isn't proof of cause.

Read the original at EurekAlert / UC San Francisco