The ketogenic diet has had an unusual journey through psychiatry. Originally designed in the 1920s as a treatment for childhood epilepsy that no longer responded to medication, it has resurfaced over the last decade as a candidate intervention for serious mental illness — including schizophrenia and bipolar disorder. The story is genuinely interesting. It is also early, and it is easy to oversell.
Pilot trials suggest a medically supervised ketogenic diet may improve symptoms and metabolic health for some people with schizophrenia, but the evidence base is small and it is not yet a standard treatment.
What "keto" actually is
A ketogenic diet shifts the body's primary fuel from glucose to ketone bodies (mainly beta-hydroxybutyrate) by sharply restricting carbohydrate intake — typically below 30–50 grams per day — while increasing fat and keeping protein moderate. Within a few days, the liver begins producing ketones, which the brain can use efficiently. Clinical ketogenic diets used for epilepsy are highly structured and supervised by dietitians; the casual versions sold online are not the same thing.
Why anyone thought this might help schizophrenia
Three threads converged. First, schizophrenia has long been linked to brain energy metabolism abnormalities — reduced glucose uptake in certain regions, mitochondrial dysfunction, and oxidative stress. Second, ketones are a more efficient brain fuel under those conditions. Third, anecdotal reports — most famously from Harvard psychiatrist Chris Palmer — described patients with treatment-resistant psychosis whose symptoms substantially improved on medical ketogenic therapy.
The hypothesis, increasingly called "metabolic psychiatry," is that some psychiatric conditions are partly disorders of cellular energy and that nutritional ketosis may stabilise the underlying biology. It is plausible. It is not yet proven.
What the trials actually show
The most cited recent study is a 2024 Stanford pilot trial led by Shebani Sethi (published in Psychiatry Research; available via PubMed). It enrolled 21 adults with schizophrenia or bipolar disorder and existing metabolic problems on antipsychotics. Participants who adhered to the diet for four months saw reductions in psychiatric symptom scores, improvements in body weight and insulin resistance, and better self-reported function. Importantly, the trial was small, uncontrolled, and unblinded — it tells us the intervention is feasible and worth larger study, not that it works.
Older case series, including reports from the 1960s and modern case studies aggregated in reviews on PubMed Central, describe individuals with treatment-resistant schizophrenia whose voices, paranoia, or affective symptoms eased after weeks to months of ketogenic therapy. Larger randomised controlled trials are now underway in several countries.
What the evidence does not yet show
- That ketogenic therapy works for the average person with schizophrenia
- That it can replace antipsychotic medication
- That benefits persist if the diet is stopped
- That it is safe long-term across all metabolic profiles
Anyone telling you keto is a "cure" for schizophrenia is moving well beyond the data.
The metabolic angle is independently important
Even setting aside psychiatric symptoms, people with schizophrenia have roughly two to three times the rate of cardiovascular death compared to the general population, much of it driven by the metabolic side effects of antipsychotics (see our guide on metabolic syndrome and antipsychotics). Any approach that meaningfully improves weight, insulin sensitivity, and lipid profile is worth taking seriously — and ketogenic diets reliably do those things in motivated patients.
Risks and trade-offs
Ketogenic therapy can change how some psychiatric medications are absorbed and metabolised, can interact with diabetes medications dangerously, and is not appropriate for people with certain medical conditions (pancreatitis, gallbladder disease, some genetic disorders). Do not start without medical supervision.
- "Keto flu" — fatigue, headache, irritability in the first 1–2 weeks
- Constipation — common, particularly relevant for people on clozapine
- Lipid changes — usually improvements in triglycerides and HDL but sometimes large rises in LDL
- Adherence — sustaining a strict ketogenic diet is genuinely hard, especially with cognitive symptoms or limited cooking capacity
- Cost and food access — fresh fats and proteins are expensive in many neighbourhoods
How a serious trial of this would look
If you and your prescriber decide to explore ketogenic therapy, a credible plan usually includes:
- Baseline labs (lipids, fasting glucose, HbA1c, liver and kidney function, electrolytes)
- Working with a registered dietitian familiar with clinical ketogenic protocols
- Daily measurement of urine or blood ketones in the first weeks to confirm ketosis
- Symptom tracking on a structured scale (PANSS-derived items, sleep, function)
- Repeat labs at 6–8 weeks and again at 3–4 months
- A plan for what counts as "success" and what counts as "stop"
Apps like Frida can help with the symptom-tracking side of that picture, building the kind of week-by-week record that lets you and your prescriber tell whether anything is actually changing.
Where this fits
Ketogenic therapy is not a replacement for evidence-based treatment of schizophrenia. Antipsychotic medication, psychosocial support, and tools like CBT for psychosis remain the foundation. What the keto literature is starting to suggest is that, for a subset of patients, a serious nutritional intervention may be a meaningful add-on — both for symptoms and for the metabolic risks that quietly shorten lives. The honest position is cautious optimism: take the science seriously, do not get ahead of it, and do not try this alone.
This article is for educational purposes only and is not medical advice, diagnosis, or treatment. Always consult a qualified mental health professional. If you or someone you know is in crisis, call or text 988 in the US, or your local emergency number.