Cannabidiol
A substance from the cannabis plant that does not make people high. Small trials in psychosis gave mixed results; a large European trial began in 2026.
At a glance
- Type
- Medicine: cannabinoid
- Used for
- Tested as an add-on to antipsychotics, and once on its own, in psychosis
- Status
- In trials
- Evidence
- EarlyHow evidence is rated
- Administration
- By mouth, as an oral solution or capsules, twice a day in the trials
- Common side effects
- Sleepiness
- In Europe
- Being tested in Austria, Germany, Greece, Italy, the Netherlands, Spain and the UK, in the STEP-ENHANCE trial
Clinical summaryFor health professionals
- Does not activate CB1 receptors; inhibits the breakdown of anandamide, an endocannabinoid. In Leweke 2012, rising anandamide tracked clinical improvement.
- Leweke 2012 (Cologne): 42 acute inpatients, CBD or amisulpride up to 800 mg a day for 4 weeks; similar improvement, with fewer extrapyramidal symptoms, less weight gain and lower prolactin on CBD.
- McGuire 2018: 88 patients, 1,000 mg a day added to antipsychotics for 6 weeks; positive symptoms 1.4 PANSS points better than placebo. Boggs 2018: 36 outpatients, 600 mg a day; no benefit, more sedation.
- Meta-analysis, 8 trials, 288 patients: no significant effect on psychosis; tolerability similar to controls (Marchi 2026).
- STEP-ENHANCE: 250 patients with poorly responding first-episode psychosis, 1,000 mg a day or placebo for 6 weeks.
What it is
Cannabidiol (CBD) is one of the main substances in the cannabis plant. Unlike THC, the substance that makes people high and can bring on psychosis, it does not activate the brain’s main cannabinoid receptor. It slows the breakdown of anandamide, a cannabinoid the body makes itself, and in the first trial in schizophrenia, people whose anandamide rose more improved more.1
How well it works
In that first trial, in Cologne, 42 people in hospital with acute schizophrenia took either cannabidiol or the antipsychotic amisulpride for four weeks. Both groups improved to a similar degree, and cannabidiol caused fewer movement side effects, less weight gain and no rise in the hormone prolactin.1 Added to antipsychotics in 88 people for six weeks, it eased hallucinations and delusions slightly more than placebo.2 In 36 outpatients with long-standing illness, a lower dose did not help and caused more sleepiness.3 Pooled across eight trials with 288 people, cannabidiol had no clear effect on psychotic symptoms.4
In people with schizophrenia who also used cannabis heavily, a dose of cannabidiol taken before cannabis did not protect against its effects; memory and psychotic symptoms were worse.5
Side effects
In the trials, side effects were about as common as on placebo or the comparison medicine; sleepiness was the most frequent.4,3
Trials in Europe
STEP-ENHANCE, led from Oxford, is testing cannabidiol in 250 young people whose first episode of psychosis has not responded well to an antipsychotic.6 In July 2026, it was authorised in Austria, Germany, Greece, Italy, the Netherlands and Spain.7
What we don’t know yet
- Whether cannabidiol helps people with a first episode of psychosis, as STEP-ENHANCE is testing.
- Which dose, if any, is effective.
