A team led by BBRF grantees and Scientific Council members has reported the first-ever randomized, controlled trial to assess the effects of a ketogenic diet on psychiatric symptoms, metabolism, and cognition in people with schizophrenia spectrum and bipolar I disorders. Although their trial was small, they found the diet “feasible” to administer over a 4-month period and “promising” in its impacts on psychosis-related symptoms, mood, cognition, and several indicators of metabolic health.

Ketogenic diets are high in fat, moderate in protein, and very low in carbohydrates. They are designed to induce a state of ketosis in the body. Ketosis is a metabolic state in which the body burns stored fat for fuel instead of glucose. Although it can be induced by diet, ketosis occurs naturally when carbohydrate intake is severely restricted, forcing the liver to convert stored dietary fats into fatty acids, which the liver converts into energy molecules called ketone bodies. These can travel through the bloodstream and cross the blood-brain barrier, fueling energy-demanding organs like the brain, heart, and muscles.

The team was led by senior members Judith M. Ford, Ph.D., president of BBRF’s Scientific Council and a 2003 BBRF Independent Investigator, and Daniel H. Mathalon, M.D., PhD., a Scientific Council member who has received BBRF Independent Investigator (2007) and Young Investigator (2001) grants. Co-first authors of the paper reporting results of the trial in Schizophrenia Bulletin were Samantha V. Abram, Ph.D., 2024 BBRF Young Investigator, and Juliette M. Kyner.

Second-generation antipsychotic medicines commonly induce weight gain, dysregulation of glucose, and hyperlipidemia, and contribute to metabolic syndrome. In addition to dysregulation of metabolism, metabolic syndrome can worsen psychosis symptoms, impair cognition, and reduce the lifespan by up to 20 years.

There is reason to believe that metabolic dysfunction is part of the pathophysiology of schizophrenia, as it was noted before the introduction of antipsychotic medicines in the 1950s, and in some cases, is seen before the first episode of psychosis. But post-diagnosis, there is also evidence that diet can play a role in the onset of metabolic syndrome, as poor diet and reduced exercise are common consequences of avolition—a lack of motivation that prevents individuals from initiating or completing everyday tasks, even when they want to—and functional impairment that attend schizophrenia. Antipsychotic medications themselves contribute to metabolic dysfunction, underscoring the need for non-pharmacologic treatment alternatives, the researchers note.

It has been suggested that a ketogenic diet could help address incomplete response to antipsychotic medications while also improving patients’ metabolic health. A number of small studies without controls have suggested this. The current trial was the first randomized controlled trial testing this idea in people with psychosis. Noting that weight loss, in itself, can improve metabolic health, a ketogenic diet might benefit both metabolism and brain function, including cognition, the team said.

For their trial, the team recruited 30 individuals with schizophrenia or schizoaffective disorder and 19 with bipolar I disorder with psychotic features. Of the 47 who completed the trial, the typical participant was in their late thirties and there were slightly more men than women. The 38 participants already being treated with medications continued taking them, including antipsychotics, mood stabilizers, anti-anxiety medications, and antidepressants. After randomization, 24 received a ketone diet for 1 month while 23 continued on their usual, non-ketogenic diet. A total of 25 participated in a 4-month extension of the trial, with those already on the ketogenic diet continuing on it for another 3 months, and those who had not been on it committing to a ketogenic diet for 4 full months. By the end of the trial, all 25 participants in the extension had been on the ketogenic diet for 4 months. The study was not blinded; participants knew if they were or were not on the diet.

Adherence to the diet was a concern addressed by the trial’s design. Those on the diet received 18 microwavable ketogenic meals each week via a meal delivery service. Participants were allowed to eat foods aside from those provided, but were instructed to eat at least 2 prepared meals daily and aim for a diet consisting of 70%-80% fat, 10%-20% protein, and 10% or less carbohydrates. The “diet-as-usual” participants were not provided with meals. A licensed dietician contacted all participants weekly, and a psychiatrist was on call. Although ketogenic diets can have side effects, there were no serious adverse effects reported. Some participants reported headache, which was resolved with increased intake of fluids and electrolytes. In the first 2 weeks of the initial month, those on the ketogenic diet attained ketosis 76% of the time, and 86% in the subsequent 2 weeks. That level was maintained through the 4-month extension by those who took part in it, all of whom were on the ketogenic diet.

“Compared to the diet-as-usual group, people who followed a 1-month ketogenic diet saw metabolic improvements,” the team reported. Among those who completed the 4-month extension study, the trend of “improvements in metabolic markers and depressive symptoms became significant. We also observed broader improvements in psychiatric symptoms and cognitive functioning after 4 months on a ketogenic diet.” The latter suggested to the team that a single month on the diet is not long enough to see the benefits that the trial demonstrated were possible after 4 months.

Regarding an important point that researchers have disputed, evidence from this trial indicated that it was the increased production of ketone bodies thanks to the diet, and the maintenance over time of ketosis, that correlated with improvements in depressive symptoms and metabolic health—and not merely the fact that participants lost weight. This suggests that differences in the degree to which people are able to adhere to the diet may influence any clinical gains associated with it. It also suggests that these benefits may not be achievable with the GLP1-receptor agonists, unless they are associated with an increase in ketone bodies, Dr. Ford noted.

Those who followed the ketogenic diet for 1 month “demonstrated numerical, but non-significant improvements in cognitive performance, depressive symptoms, and positive symptoms of psychosis, compared to those who followed their usual diet,” the team reported. It was at the conclusion of the 4-month extension of the ketogenic diet, however, that “significant improvements were found for all clinical measures of cognition and psychiatric symptoms of depression and psychosis.”

Improvement in depressive symptoms correlated with the proportion of days a participant spent in ketosis, another indication of the diet’s benefit but also that this particular improvement may be tied with ketosis, and not other metabolic or other factors affected by the diet. Also, those who adhered more faithfully to the diet tended to have greater benefits, both metabolic and psychiatric.

“We suspect the underlying mechanism driving depression symptom reductions is based on the presence of ketone bodies, and not simply due to weight loss,” the team wrote. “Unlike ketone levels, decreased body/mass index did not explain such improvements. This suggests that simply losing weight may not reap the same benefits as being in ketosis. Psychiatric improvements might therefore follow improved metabolic function.” Of the 80% on the ketogenic diet who demonstrated a decline in depressive symptoms after 1 month on the diet, 62% saw additional improvement after the completion of the 4th month on the diet. “The same trend of progressive improvement was seen in cognitive performance and negative symptoms, suggesting that the duration of a ketogenic diet may play an important part in clinical outcomes,” the team said.

They urged future studies with more patients, longer diet periods, more intense data collection, and consideration of whether “less intensive alternatives,” like ketogenic drinks, can augment benefits of the diet.