A randomized, parallel-group controlled clinical trial conducted in Queensland, Australia, has demonstrated the real-world feasibility and safety of ketogenic metabolic therapy among adults diagnosed with schizophrenia, schizoaffective disorder, and bipolar disorder. The investigation, named the North Queensland Dietary Intervention Trial and published in Frontiers in Nutrition, evaluated whether individuals managing severe, enduring psychiatric conditions could prepare their own meals and sustain nutritional ketosis over an extended period. Rather than relying on commercially manufactured or delivered meal plans, participants purchased their own groceries and cooked meals independently, supported by regular consultations with a clinical dietitian. Across twelve weeks of active intervention following a two-week lead-in, participants randomized to the ketogenic therapy achieved high rates of attendance and sustained biochemical adherence, demonstrating that structured therapeutic dietary change is viable within an outpatient community cohort managing complex psychiatric conditions.

Schizophrenia, schizoaffective disorder, and bipolar disorder present profound clinical, cognitive, and physical burdens, often compounded by a reduction in life expectancy of thirteen to twenty-five years compared to the general population. In response to post-mortem and functional brain imaging research identifying bioenergetic abnormalities, compromised glucose metabolism, mitochondrial dysfunction, and oxidative stress in psychiatric neuropathology, researchers have proposed providing alternative energy substrates. Prior evidence included animal models, case reports, uncontrolled trials, and an exploratory study using microwavable pre-prepared meals, leaving open the question of whether patients could execute self-prepared dietary regimens.

The study protocol recruited seventy-one clinically stable community-dwelling adult outpatients in Townsville between January 2024 and October 2025 who possessed psychiatrist-confirmed diagnoses of schizophrenia, schizoaffective disorder, or bipolar disorder lasting at least six months. Following screening of ninety-six prospective candidates, enrolled participants were randomized using a computer-generated, stratified design accounting for sex, diagnostic category, and duration of illness. The trial compared ketogenic metabolic therapy directly against the Australian Guide to Healthy Eating, an active, evidence-based dietary comparator based on national nutritional recommendations. Eligible individuals had access to cooking facilities, had not followed a ketogenic diet in the preceding two months, had no active substance misuse or eating disorder history, and showed baseline body mass index readings of at least 18.5 kilograms per square meter. Assessing consultant psychiatrists were blinded to dietary allocations, and participants were instructed not to disclose their allocated intervention group during psychiatric and metabolic monitoring.

The active control arm receiving guidance based on the Australian Guide to Healthy Eating was prescribed individualized, nutrient-dense meal plans emphasizing core food groups: vegetables, fruits, whole grains, lean proteins, and reduced-fat dairy while restricting saturated fats, added sugars, and alcohol. Conversely, the ketogenic metabolic therapy protocol prescribed a macronutrient distribution of approximately seventy percent fat, twenty-three percent protein, and seven percent carbohydrate, reflecting a two-to-one modified ketogenic ratio. Clinical dietitians provided personalized grocery shopping guides, visual portion estimation tools, carbohydrate-free multivitamin supplementation recommendations, and instruction on electrolyte management to reduce transitional adaptation symptoms. To protect cardiovascular profiles, dietitians emphasized mono- and polyunsaturated fat sources alongside low-carbohydrate vegetables for dietary fiber.

A total of fifty-two participants completed the twelve-week intervention, divided equally with twenty-six completers in each arm, representing an overall completion rate of 73.2 percent. Among the seventy-one enrolled individuals, thirty-one were randomized to the Australian Guide to Healthy Eating, where three were excluded by investigators due to an alcohol relapse, an involuntary psychiatric admission, and unexplained elevated ketones, while two withdrew voluntarily. In the ketogenic therapy arm, forty participants were randomized, with three not commencing due to relocation or unwillingness to commit. The demographic profile of completers revealed a mean age of 46.14 years, 62 percent female representation, and eighty-five percent having illness durations exceeding five years, with diagnoses comprising bipolar disorder, schizoaffective disorder, and schizophrenia.

Both arms demonstrated strong engagement with weekly dietetic support, consisting of an initial one-hour education session and weekly thirty-minute check-ins delivered either in person or via telephone. Overall attendance reached 92 percent in the ketogenic group, with 78 percent attended in person and 14 percent by phone, while the active control group achieved 85 percent attendance, comprising 75 percent in person and 10 percent by phone. Participants tracked metabolic adherence through daily afternoon finger-prick capillary blood testing using a dedicated glucose and ketone monitor, transmitting timestamped photographs of test results by text message to the trial coordinator daily at sixteen hundred hours. Reminders were sent whenever readings were delayed. This structured contact ensured continuous oversight, allowing dietitians to reinforce adherence, address cooking barriers, manage transient adverse symptoms, and conduct regular twenty-four-hour nutritional recalls to evaluate food selection and pattern compliance without requiring continuous laboratory visits.

Objective biochemical tracking revealed that participants assigned to ketogenic metabolic therapy maintained nutritional ketosis, defined as blood ketone levels of at least 0.5 millimoles per liter, for 70.06 percent of the total intervention period. Among the twenty-six ketogenic completers, every individual achieved daily blood ketone concentrations above 0.5 millimoles per liter during the trial. Specifically, participants spent 30.07 percent of the trial in light ketosis, 33.96 percent in moderate ketosis, and 6.03 percent in high ketosis above 3.0 millimoles per liter. At the participant level, 73.1 percent achieved full adherence by maintaining ketosis across at least eighty percent of monitored days, 23.1 percent achieved semi-adherence between fifty and seventy-nine percent of days, and only one completer exhibited non-adherence. Daily capillary blood glucose measurements remained stable, with over ninety-six percent of readings falling within normal reference parameters, while mild hypoglycemia accounted for less than one percent and hyperglycemia for three percent of readings.

In the control arm adhering to national dietary guidelines, 61.5 percent achieved moderate-to-high dietary adherence, with 53.9 percent semi-adherent and 7.7 percent fully adherent, while no control participant fulfilled all dietary guidelines across one hundred percent of assessments. Body mass index reductions served as supportive physiological evidence of adherence in both arms over the twelve-week intervention. In the Australian Guide to Healthy Eating group, where ninety-two percent began with a baseline body mass index exceeding twenty-five, average index values fell from 32.74 to 31.77, reflecting a mean reduction of 2.70 percent, with seventy-three percent experiencing weight loss. In the ketogenic cohort, average body mass index dropped from 33.37 to 31.16, representing a mean reduction of 6.50 percent, with eighty-eight percent losing weight. These shifts align with recognized metabolic pathways of ketogenic diets, including early water excretion from glycogen depletion, enhanced lipolysis, and appetite-suppressing hormone shifts that curb overall caloric intake.

Both dietary regimens proved tolerable, with no serious adverse events recorded throughout the trial. Reported side effects were mild and predominantly involved gastrointestinal disturbances, such as constipation, nausea, abdominal discomfort, and diarrhea, which typically resolved within the first six weeks as digestive systems adjusted to modified fiber and fat intake. In the ketogenic group, headache occurred in 34.6 percent of completers, nausea in 30.8 percent, while muscle cramps and minor cutaneous bruising occurred in 11.5 percent each, potentially reflecting known shifts in platelet behavior associated with ketogenic nutrition. Control participants experienced higher frequencies of hunger at 38.5 percent and fatigue at 34.6 percent, consistent with caloric restriction without the appetite-blunting effects of sustained ketosis. No participant withdrew from the study due to adverse dietary side effects, and differences in side effect rates between groups were not statistically significant.

The study population presented substantial baseline medication burdens, with 94.2 percent taking psychiatric pharmacotherapy and 38.5 percent meeting criteria for polypharmacy with three or more psychotropic agents. Antipsychotics were prescribed to 67.3 percent of completers, including clozapine in 23.1 percent and long-acting injectables in 13.5 percent, while mood stabilizers were used by 59.6 percent and antidepressants by 44.2 percent. Baseline medication counts were comparable between completers and non-completers, indicating that complex pharmacological regimens did not impede retention. However, authors noted that clozapine was prescribed to 23.1 percent of completers compared to 10.5 percent of dropouts, suggesting that established routines of mandatory hematological monitoring may have reinforced appointment adherence.

Subject of Research: Feasibility of dietitian-guided ketogenic metabolic therapy in schizophrenia spectrum and bipolar disorders

Article Title: Ketogenic metabolic therapy in schizophrenia spectrum and bipolar disorders: population descriptors and feasibility outcomes from a randomized, controlled clinical trial

Article References: Longhitano, C., Swift, J.-L., Finlay, S., Suvarna, B., Gordon, S., Shareef, O., Peachey, I., Rudd, D., Campbell, I., Sethi, S., Palmer, C., & Sarnyai, Z. (2026). Ketogenic metabolic therapy in schizophrenia spectrum and bipolar disorders: population descriptors and feasibility outcomes from a randomized, controlled clinical trial. Frontiers in Nutrition, 13, Article 1962072. https://doi.org/10.3389/fnut.2026.1962072

Image Credits: AI Generated

DOI: 10.3389/fnut.2026.1962072

Keywords: ketogenic metabolic therapy, schizophrenia, bipolar disorder, schizoaffective disorder, dietary intervention, clinical trial, nutritional psychiatry, metabolic health, Ketogenic, metabolic, therapy, spectrum