For the millions of people living with schizophrenia-spectrum or bipolar-1 disorders, antipsychotic medications often trade psychiatric stability for metabolic harm — accelerating diabetes risk, obesity, and cardiovascular disease. Evidence that a dietary intervention could address both problems simultaneously would represent a meaningful shift in how clinicians approach adjunctive care for these populations.
This first randomized controlled trial of a ketogenic diet in psychosis enrolled 58 participants, randomizing them to either a ketogenic diet (KETO, n=28) or diet-as-usual (DAU, n=30) for one month, with an optional four-month extension completed by 25 participants. The KETO group achieved clinically meaningful ketosis, and compared to DAU showed statistically significant reductions in weight, HbA1c, and insulin resistance. More notably, those who sustained the diet for four months demonstrated improvements across positive symptoms, negative symptoms, depression, and cognitive performance — all reaching p-values below 0.001. Blood ketone levels directly correlated with improvements in pre-diabetic markers and depressive symptoms, while weight loss itself was not the mediating factor, pointing toward ketone-specific metabolic or neuromodulatory mechanisms.
The mechanistic implications here are scientifically interesting. Ketone bodies, particularly beta-hydroxybutyrate, are known to influence neuroinflammation, GABA/glutamate balance, and mitochondrial function — pathways implicated in both epilepsy and psychosis. The anticonvulsant parallel is well-established; the psychiatric application remains far less explored. That said, several limitations demand caution. The trial is small, the four-month extension was non-randomized and subject to self-selection bias, and one month may be insufficient to characterize true psychiatric trajectories. Adherence to ketogenic diets in community settings — especially among individuals with severe mental illness — presents a substantial real-world challenge not fully captured here. This is incremental but genuinely promising work that justifies larger, longer trials with active dietary control conditions.