One-month trial found rapid metabolic improvements
A ketogenic diet trial in people with schizophrenia-spectrum disorders and bipolar-1 disorder found statistically significant metabolic improvements after 1 month. Participants who stayed on the diet for 4 months also showed gains in depression, schizophrenia symptoms and cognitive performance.
Researchers at the University of California, San Francisco led the randomized controlled trial. Schizophrenia Bulletin published the study on 18 June 2026, and the National Institute of Mental Health funded it in part.
The trial enrolled 58 participants, and 47 completed the first comparison between a 1-month ketogenic diet group and a diet-as-usual control group. After that phase, 25 participants chose an optional 4-month single-arm extension.
During the 1-month randomized phase, 83% of daily-tested participants maintained ketosis. In the 4-month extension, that figure rose to 94%.
The study recorded no significant side effects from the diet. That safety result covered participants with schizophrenia-spectrum disorders and bipolar-1 disorder.
In the controlled phase, higher ketone levels tracked with lower blood glucose and lower depression scores on the PHQ-9. Those links remained after the researchers accounted for weight loss.
As a result, the findings suggest ketosis itself may play a part in the metabolic and mood changes, not only weight reduction. The randomized data add to research on ketogenic diets in psychotic disorders.
UCSF four-month extension
The 4-month extension showed that metabolic improvements continued over time. It also found significant reductions in depression and schizophrenia symptoms, alongside better cognitive performance.
However, that extension included 25 participants and did not use a control group. UCSF described the findings as an encouraging signal rather than a final answer.
Study lead Judith M. Ford, PhD, is a Professor of Psychiatry at UCSF and works at the Weill Institute for Neurosciences. She argued that the cognitive and psychological improvements stood out because standard psychosis medicines often do not improve broader mental wellbeing.
“The improvement we saw in cognitive and psychological symptoms is particularly important in people with psychotic disorders, because current medications that address their psychosis don’t address their overall mental wellbeing, including cognitive or depressive symptoms, which can be debilitating,” Ford said.
Ford also cautioned that the results need larger, longer and fully controlled trials. She added that a 1-month intervention may have been too short to capture the diet’s full mental health effects.
Meanwhile, the 4-month psychiatric and cognitive findings still need replication under controlled conditions. The trial report also described correlations between ketone levels and cognitive and psychiatric symptoms in the study group.
