UCSF's 20-person pilot shows a larger trial of cutting ultra-processed food for depression is workable

The design is the strongest thing in this study. Researchers split the participants into two groups. One kept its usual diet for four weeks while the other switched to a diet low in ultra-processed food, and then the groups traded places ^[5]. That put each person in the study for eight weeks ^[1], and each was compared against themselves. This removes the simplest alternative explanation, that the people who ate better happened to start out less depressed. The authors believe it is the first study to compare depression symptoms in the same participants before and after they moved off their usual ultra-processed diet ^[6].

The weakness is that participants knew what they were eating. They chose and bought their own food, with guidance from the research team ^[7]. Someone who knows she is in the healthier-diet phase may report feeling better for reasons unrelated to emulsifiers. A crossover with no blinding cannot separate that from an effect of the food.

The improvements were moderate to large across three depression measures ^[3], and anxiety scores improved modestly ^[4]. The release does not give the scores or their confidence intervals. With 20 participants ^[16], those intervals are likely to be wide. Andrew Krystal, the senior author, put feasibility first. The pilot "successfully established the viability of carrying out this research in a larger-scale trial," he said ^[8].

The pilot makes a stronger case on that question. A larger trial depends on people changing what they eat without the study supplying their meals. These participants cut their intake of ultra-processed food by more than 85% while buying their own groceries ^[2][7]. Nyasha Chagwedera, the corresponding author, explained why that is hard. "Reducing the amount of ultra-processed food in our diet can be challenging, because it's highly palatable, cheap, and convenient," she said ^[9].

The sample was narrow. Participants averaged 44 years old and 70% were female ^[10]. Fourteen of the 20 also had anxiety ^[11], or 70% ^[2]. Every participant had a high BMI and at least one metabolic disorder, such as hypertension, diabetes or elevated LDL cholesterol ^[12]. The result concerns depression in people with metabolic disease. Depressed people with normal metabolic health were outside the study.

Chagwedera's proposed explanation is what she called a "leaky, out-of-balance gut," in which ultra-processed foods may damage the gut's protective lining ^[13]. "This gut damage can create low-grade, chronic inflammation throughout the body that either directly causes depression or makes people more vulnerable to it," she said ^[14]. The release offers this as one possible explanation for the link ^[13].

The team plans to test the effect in a larger group, with phase 2 enrollment expected to begin in 2028 ^[15]. The pilot appeared in JAMA Psychiatry on Sept. 23, 2026 ^[1], so the gap between publication and enrollment is about two years ^[3].