A pilot trial by researchers at the University of California, San Francisco (UCSF) found that people diagnosed with depression reported fewer symptoms during a four-week period of reduced ultra-processed food intake than during a period of normal eating. The trial involved just 20 participants.
The study was led by psychiatry researcher Nyasha Chagwedera. It enrolled 20 adults who ate large amounts of ultra-processed food, with an average age of about 44 and an average of roughly 13 years living with depression.
Participants were randomly split into two groups: one continued its usual diet for four weeks while the other cut back on ultra-processed food over the same period, after which the groups switched. They bought their own food but followed guidance from the researchers. Medication, therapy, exercise and sleep routines were kept unchanged throughout.
Nineteen participants completed the trial. A survey measuring consumption of highly processed food showed an average score of 10.10 at the outset, which fell to 1.42 during the intervention period.
Results on three separate depression scales were lower during the period of reduced ultra-processed food intake. On the PHQ-9 test, the average score was 8.68 during normal eating and 5.53 during the intervention; an analysis accounting for baseline symptoms, the order of the diets and the study phase found a difference of about three points between the two diets. On the MADRS scale, scores were 18.9 and 8.05, respectively. A third depression test and an anxiety questionnaire showed similar results.
Chagwedera said cutting ultra-processed food can be difficult because such products are “tasty, cheap and convenient.” Participants rated the ease of the intervention at an average of 5.74 and its usefulness at 8.42, and no side effects were recorded.
The researchers described the approach as feasible and acceptable, saying it could lay the groundwork for larger trials. They also noted that symptoms were lower than baseline even during the normal-eating period — the PHQ-9 baseline score was 12.8 — meaning the main effect should be attributed only to the difference between the two diet periods.
The authors acknowledged that the trial was small and not blinded, that food intake relied on self-reporting, and that all participants had a high body weight and at least one metabolic condition. Depression and anxiety scores were treated as secondary outcomes, and the findings are considered hypothesis-generating.
The researchers recommended larger, blinded trials with objective measures of ultra-processed food intake, as well as further study of the mechanisms behind the results. The study was published in the journal JAMA Psychiatry.
