Cannabis & Cannabinoids · Cannabinoids · One approved prescription medicine; limited evidence for other uses

CBD (Cannabidiol)

Q: What is CBD?

CBD (cannabidiol) is a compound made by the cannabis plant, formed from its acid form CBDA when heated or aged. It does not cause a high. It is sold in oils, gummies, vapes and creams, and one purified form is an approved prescription medicine for certain epilepsies.

Q: Is there an approved CBD medicine?

Yes, one. The FDA approved a prescription CBD solution (Epidiolex) for seizures with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex in people 1 year and older. The EU authorised Epidyolex from age 2. The FDA says it has approved no other CBD products.

Q: Does CBD work for epilepsy?

The purified prescription solution does, for those three syndromes, when added to usual medicines. In trials, seizures fell about 17 to 30 percentage points more than with placebo, though few patients became seizure-free. Shop products have not been tested this way.

Q: Does CBD help with anxiety?

Maybe a little in some settings, but the evidence is mixed. A 24-patient lab speaking test found less anxiety after a single CBD dose than placebo, and a 2024 review of 11 trials found contradictory results. It is not an approved treatment for anxiety.

Q: Does CBD help me sleep?

Not proven. A 2020 systematic review found insufficient evidence for any sleep disorder. A 2025 trial of a CBD and terpene blend found about 1% more deep and REM sleep, which is small and cannot be credited to CBD alone.

Q: Does CBD relieve pain?

The evidence is weak. The best controlled trial, in 136 people with hand osteoarthritis or psoriatic arthritis, found no difference from placebo. A 2026 Cochrane review found no clear evidence for CBD-dominant medicines in nerve pain.

Q: Can CBD harm the liver?

Yes. In trials of the prescription solution, liver enzymes rose above 3 times normal in 12% to 13% of treated patients versus 1% on placebo, and more with valproate. A 2025 trial in healthy adults found 5.6% on CBD versus 0% on placebo after 28 days. Ask a doctor about liver tests if you use CBD regularly.

Q: Can CBD interact with my medicines?

Yes. With clobazam, CBD raised its active breakdown product 3.4-fold and sleepiness was more common. With valproate, liver enzyme rises are more common. Case reports describe higher INR with warfarin. CBD also affects CYP2C19 and other liver enzymes, so blood levels of many drugs can change. Ask a pharmacist.

Q: Will CBD make me sleepy or affect driving?

It can cause sleepiness (32% versus 11% on placebo in the main trials, more with clobazam), and alcohol adds to it. In one on-road test, vaporized CBD-dominant cannabis did not increase lane weaving, but the authors say that does not rule out impairment. The label advises not driving until you know how it affects you.

Q: Do CBD products contain what the label says?

Often not. In a test of 80 hemp CBD oils, 37 were at least 10% off their label, and THC was found in 52 of 81 samples, including 5 of 21 labelled THC-free. The FDA says many products it tested did not contain the CBD claimed.

2-[(1R,6R)-3-methyl-6-(prop-1-en-2-yl)cyclohex-2-en-1-yl]-5-pentylbenzene-1,3-diol (C21H30O2; PubChem CID 644019) · also called CBD, cannabidiol, (-)-cannabidiol; the prescription form is sold as Epidiolex (US) and Epidyolex (EU)

CBD is a compound made by the cannabis plant that does not cause a high. One purified form is an approved prescription medicine for certain childhood-onset epilepsies. Shop products are different: their content often does not match the label, and CBD can injure the liver, cause sleepiness and change how other medicines work.

General information, not medical advice. Last substantive update October 25, 2026 · Not medically reviewed. A named clinician has not signed off on this page. How this page is made

About CBD (Cannabidiol)

CBD (cannabidiol) is a cannabis compound that does not cause a high. One purified prescription form is approved for seizures in Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex. Evidence for anxiety, sleep and pain is weak or mixed, and liver, interaction and product-quality concerns are real.

What CBD is

CBD (cannabidiol) is a compound made by the cannabis plant. Its chemical formula is C21H30O2. In the plant it starts as an acid form, CBDA, and becomes CBD when heated or aged. It can also be made in a lab.

Unlike THC, CBD does not cause a high, and WHO's 2018 review found no signs of abuse or dependence potential in humans. That does not make it harmless: it can injure the liver, make you sleepy and change how other medicines work.

^{52 46 38}

The one approved CBD medicine (prescription only)

The FDA has approved one CBD product: a prescription solution (Epidiolex) for seizures with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex in people 1 year and older. The EU authorised it (as Epidyolex) from age 2. Five studies in over 900 patients showed fewer seizures than placebo.

It is a medicine made to a fixed strength, started by a doctor, with liver blood tests before and during treatment. Its trial results do not apply to oils, gummies, vapes or creams from a shop.

^{50 47 49 13 40 41}

What the research says about other uses

Outside those epilepsies, the evidence is thin. For anxiety, results are mixed (one small lab speaking test was positive; a 2024 review of 11 trials called results contradictory). For sleep, a 2020 review found insufficient evidence. For arthritis pain, the best trial found no benefit, and a Cochrane review found no clear benefit for CBD-dominant medicines in nerve pain.

Some early trials are promising for schizophrenia (added to usual medicine) and for cutting down cannabis, but each rests on one study. The FDA has warned companies against selling CBD products claimed to treat serious diseases such as cancer.

^{6 10 37 42 2 27 17 48}

Shop products are not the medicine

In one test of 80 hemp CBD oils, 37 held at least 10% more or less CBD than the label said, and THC was found in 52 of 81 samples, including 5 of 21 labelled THC-free. The FDA says many products it tested did not contain the CBD claimed.

In the US, the FDA says CBD cannot lawfully be added to food or sold as a dietary supplement. Rules differ by state and country; this is not legal advice.

^{24 25 48 49}

What the lab numbers show

Our data holds 409 laboratory measurements of CBD in cannabis flower for 327 differently named strains. They come from state testing datasets and published papers. We did not test anything ourselves. A strain name is only a label: the same name can have very different chemistry, and these are real results from specific samples.

The biggest group is 232 values from Maryland, Hawaii and Nevada testing data (189 strain names). Each is the middle (median) result across many samples of that name, listed as percent by weight (the source does not label the unit column by column). Nearly all are zero: 195 of the 232 medians are 0.0, and only 37 are above zero. The highest are ACDC at 0.71 in Hawaii (3 samples) and 0.3 in Maryland (145 samples), and Cannatonic at 0.135 in Hawaii (6 samples). Well-known THC-type names such as Blue Dream, Wedding Cake and Gorilla Glue #4 show medians of 0.0, for example Wedding Cake in Nevada (915 samples).

CBD-rich hemp is a different story. A paper measured 'total CBD' as percent of dry matter (means of three test plots) for 45 hemp cultivars: two CBG-rich lines were 0.0, and the other 43 ranged from 5.7 to 17.0. Greenhouse tables list named CBD-rich plants at 4.71 (Juanita), 8.88 (Love Lace) and 9.84 (Thunderstruck) percent of flower dry weight, and other tables list Sweet Skunk CBD at 11.2 and Warlock CBD at 11.4 'as printed'. These units and methods differ from the state medians above, so do not compare them directly.

A further 72 values are single commercial batches printed in a paper's supplement with an unclear unit; we do not compare them. These numbers describe how much CBD the flower contained. They say nothing about how it will make anyone feel, and nothing about CBD oils or gummies.

At a glance

Traditional uses history, not proof

Cannabis has a long history of anecdotal medicinal use and limited licensed medicinal use, according to a pharmacology review. This page covers only the isolated compound.

Studied for research

Seizures in Lennox-Gastaut, Dravet and tuberous sclerosis complex (prescription medicine only): high
Anxiety: low
Schizophrenia added to usual medicines: low
Cutting down cannabis use: low
Sleep: very low
Pain (arthritis, nerve): low to very low
Crohn's disease, blood pressure, type 2 diabetes: very low
Parkinson's disease: mixed

Main compounds

CBD (cannabidiol) itself
CBDA, the acid form in the raw plant
7-OH-CBD, an active breakdown product
7-COOH-CBD, the main metabolite in blood

Common forms

Prescription CBD oral solution (the approved medicine)
Hemp-derived oils, gummies, vapes and creams (not approved medicines)
Capsules and under-the-tongue products used in trials
CBD plus terpene blends

Research links: FDA: CBD consumer update (March 2020) · FDA: Regulation of cannabis and CBD, questions and answers · EMA: Epidyolex · WHO: CBD critical review (2018) · DailyMed: Epidiolex prescribing information · Research Library · Genes & Biomarkers Library

What it is and its history

Name: Cannabidiol (CBD)
Chemical name: 2-[(1R,6R)-3-methyl-6-prop-1-en-2-ylcyclohex-2-en-1-yl]-5-pentylbenzene-1,3-diol
Other names: (-)-Cannabidiol; (-)-trans-Cannabidiol; cannabidiolum; CBD; CAS 13956-29-1; PubChem CID 644019; Epidiolex (brand name of the prescription solution)
Common names: CBD (English); Cannabidiol (English)
Formula: C21H30O2
Molecular weight: 314.5
CAS number: 13956-29-1
PubChem ID: 644019
Chemical class: Phytocannabinoid (a 21-carbon terpenophenolic compound; a cannabinoid made by the cannabis plant)
How the plant makes it: In the plant, cannabigerolic acid (CBGA) is turned by the enzyme CBDA synthase into cannabidiolic acid (CBDA). CBDA loses a carbon dioxide unit (decarboxylation) with heat, light or age and becomes CBD. The sister enzyme THCA synthase makes THCA, the precursor of THC.
Where it is found: Found in cannabis plants, in much larger amounts in high-CBD and fiber-type (hemp) plants than in typical THC-rich plants. WHO says cultivars used for recreation usually have more THC than CBD, and high-CBD cultivars are grown for medical purposes. It can also be made synthetically.
Related or similar compounds: Name: THC (delta-9-tetrahydrocannabinol); Note: Has the same formula (C21H30O2) but a different shape; THC causes a high and CBD does not.; Name: CBDA, the acid form; Note: CBDA is what the raw plant mostly holds; it becomes CBD when heated.; Name: Hemp seed oil; Note: Seeds do not naturally contain CBD or THC beyond traces picked up in processing (FDA); it is not the same as CBD oil.

Traditional use and history history, not clinical proof

Tradition or era	Documented use	Source
1963	The structure of cannabidiol was reported in a paper titled 'Hashish. I. The structure of cannabidiol' in the journal Tetrahedron.	p5879214
2007	The plant enzyme that makes CBDA from CBGA (CBDA synthase) was cloned and characterised, explaining why some cannabis plants make mostly CBD.	p17544411
2018	WHO's expert committee reviewed CBD, found no signs of abuse or dependence potential in humans, and noted a CBD product (Epidiolex) was awaiting approval in the US.	s-who-ecdd-2018
2018 (US) and 2019 (EU)	The first CBD medicine was approved: initial US approval 2018 (label), and EU marketing authorisation on 19 September 2019 (EMA).	s-epidiolex-label
2018 Farm Bill	US law removed hemp (cannabis with 0.3% delta-9 THC or less on a dry-weight basis) from the Controlled Substances Act while keeping FDA authority over products.	s-fda-regulation
2020	The FDA's list of announcements includes a July 2020 item on a new indication for the CBD medicine in a rare genetic disease (the title was read; the release itself was not). The current label includes tuberous sclerosis complex.	s-fda-regulation

History shows how a substance was used or discovered. It does not show that a product works or is safe today.

What is inside, and how it may work

What it contains and how those compounds act in cells and animals. Laboratory and animal mechanisms are not proven effects in people.

Main compounds

Compound	Family	Where it is concentrated	Activity studied	Level of evidence
Cannabidiolic acid (CBDA) PubChem 160570	Phytocannabinoid acid	Not read in the sources Not read; raw plant holds mostly the acid forms (WHO)	The plant's acid precursor of CBD; becomes CBD by decarboxylation with heat, light or age (WHO). ^{46 51}	Authoritative guidance
Cannabigerolic acid (CBGA) PubChem 6449999	Phytocannabinoid acid	Not read in the sources Not read	The shared precursor: CBDA synthase turns it into CBDA and THCA synthase into THCA. ^{46 38 51}	Authoritative guidance
Delta-9-THC PubChem 16078	Phytocannabinoid	Not read in the sources Not read	The intoxicating cannabinoid. In cells, CBD reduced the strength of THC signalling at CB1 receptors. Under lab acid conditions CBD can be converted to THC; WHO says this likely does not happen after oral CBD in humans. ^{26 46 51}	Laboratory research
7-Hydroxycannabidiol (7-OH-CBD) PubChem 11301963	Active metabolite	Made in the body; levels rose about 1.5-fold in AUC when CBD was given with clobazam	Active breakdown product of CBD (label and early human study). ^{39 30 51}	Human data
7-Carboxycannabidiol (7-COOH-CBD) PubChem 146592489	Main circulating metabolite	The main circulating metabolite in blood after oral CBD (early human study)	The major human metabolite; animal studies in the label of this metabolite showed developmental effects at higher exposures. ^{39 50 51}	Human data

Mechanisms, receptors and pathways

Target or pathway	Gene or receptor	What was found	Level
Cannabinoid CB1 receptor (negative allosteric modulation)	CB1 receptor (CNR1)	In cells that carry CB1 receptors, CBD reduced the strength and potency of 2-AG and THC signalling and stopped receptor internalization. This is a laboratory finding. A systematic review said CBD is very unlikely to act on neurological disease through the endocannabinoid system. ^{26 21}	Laboratory research
Intracellular calcium-related targets	VDAC1, GPR55, CaV3.x calcium channels	Of more than 65 reported molecular targets, the reviewers judged these plausible for neurological effects; no causal proof exists yet. Many reported effects occurred only at concentrations too high to reach in the body. ²¹	Lab animal
Calcium movement and adenosine (seizure control)	Not named	The European regulator says CBD is thought to act on targets controlling calcium movement in nerve cells and on adenosine, a brain chemical that helps suppress seizures; how it works is not clearly understood. ⁴⁷	Authoritative guidance
Serotonin 5-HT1A receptor	5-HT1A (HTR1A)	A 2024 anxiety review states CBD may act as a partial agonist at this receptor, which could help explain calming effects. The review presents this as a proposed mechanism, not shown in the trials. ¹⁰	Lab animal
Liver drug-metabolism enzymes and transporters	CYP2C19, CYP2C8, CYP1A2, CYP2B6, UGT1A9, P-glycoprotein	In people CBD acts as a moderate inhibitor of CYP2C19, an inhibitor of UGT1A9, a weak inhibitor of CYP1A2, and raised everolimus exposure about 2.5-fold (a P-gp substrate). It did not affect CYP3A4 activity in six healthy-volunteer and patient trials. ^{50 33}	Human data

Absorption, metabolism and elimination

Absorption: By mouth, absorption is low and varies between people. After single oral doses of the prescription solution, peak blood level came at about 4 to 5 hours. A review estimated absolute oral bioavailability of about 6% when fasting, rising about fourfold with a high-fat meal.
Distribution: The apparent volume of distribution was very large (about 21,000 to 43,000 L) in healthy volunteers.
Metabolism: Mostly by the liver. 7-OH-CBD is an active metabolite and 7-COOH-CBD is the main circulating metabolite. Strong CYP3A4/CYP2C19 inducers such as rifampin lowered CBD by about 32% and 7-OH-CBD by about 63% (label). CBD itself inhibits several liver enzymes (see Mechanisms).
Elimination: Terminal half-life was about 60 hours after twice-daily dosing, but the effective half-life was 10 to 17 hours. Steady state was reached in about 2 days with moderate build-up (1.8- to 2.6-fold).
Bioavailability: About 6% fasting, estimated; about four times higher with a high-fat meal. Whole milk and alcohol also raised exposure.

^{39 12 34 30 50}

Isolated compounds versus whole extracts

'CBD' on a shop label can mean an isolate, a full-spectrum hemp extract (which also holds THC and other cannabinoids), or a blend with terpenes. The strongest evidence is for one purified, plant-derived solution at a fixed strength (the prescription medicine). Other trials used capsules, sublingual products, CBD-rich cannabis with some THC, or a CBD plus terpene blend, and results should not be carried across. Shop products vary: in one test 37 of 80 held at least 10% more or less CBD than the label, and most samples held some THC.

Nutritional composition

CBD is not a nutrient and has no nutritional claim. In the US, the FDA says CBD cannot lawfully be added to foods or sold as a dietary supplement.

12benefits and uses looked at

52studies and sources checked

High certaintybest evidence rating found

13lab markers tracked in trials

1 high certainty1 mixed findings4 low certainty6 very low certainty

What was looked at

The 12 benefits and uses checked for CBD (Cannabidiol), each with how strong the evidence is. Tap one to open it.

Seizures in Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex (prescription CBD medicine)High certainty
AnxietyLow certainty
Sleep and insomniaVery low certainty
Schizophrenia and psychosis (added to usual medicines)Low certainty
Cannabis use disorder (during a quit attempt)Low certainty
Cravings and anxiety in people with heroin use disorder (lab study)Very low certainty
Pain from hand osteoarthritis and psoriatic arthritisLow certainty
Nerve pain (neuropathy)Very low certainty
Crohn's diseaseVery low certainty
Blood pressure (single dose in healthy men)Very low certainty
Type 2 diabetes (blood sugar and blood fats)Very low certainty
Parkinson's diseaseMixed findings

Lab markers tracked in the studies

Convulsive seizure frequency (Dravet syndrome)Drop seizures (Lennox-Gastaut syndrome)Seizures in tuberous sclerosis complexAnxiety during a simulated public-speaking testPositive psychotic symptoms (PANSS)Days without cannabis per week and urinary THC markerCraving triggered by drug cuesShare of the night in slow-wave plus REM sleepPain intensity (0 to 100 mm scale)At least 50% pain relief (neuropathic pain)Crohn's Disease Activity Index (CDAI)Resting systolic blood pressureMemory delayed-recall score (MoCA)

What is CBD (Cannabidiol) good for?

The uses people ask about and how strong the human evidence is for each. Only the first applies to an approved medicine; the rest are not approved uses.

Seizures in Lennox-Gastaut, Dravet and tuberous sclerosis complex (prescription medicine only)High certaintyFive controlled studies in over 900 patients show a purified, prescription CBD solution added to usual seizure medicines cuts seizures more than placebo. Regulators in the US and EU approved it. This is a medicine with liver tests and interaction checks, not a shop product. AnxietyLow certaintyA 36-person lab speaking test found less anxiety after one CBD dose than placebo, and a 2024 review of 11 trials found mixed results. Not an approved use. Sleep and insomniaVery low certaintyA 2020 review found insufficient evidence. One 2025 trial of a CBD and terpene blend found about 1% more deep and REM sleep; a 30-person pilot found CBD did not harm daytime thinking. Schizophrenia (added to usual medicines)Low certaintyOne 6-week trial of 88 patients found a small drop in psychotic symptoms when CBD was added to antipsychotics. It needs repeating. Cutting down cannabis (cannabis use disorder)Low certaintyA phase 2 trial of 82 adults trying to quit found more days without cannabis at two of three CBD amounts, alongside a therapy session. Cravings in people with heroin use disorderVery low certaintyA small exploratory trial found less drug-cue craving and anxiety after three days of CBD in a lab test. Real-world relapse was not measured. Arthritis painLow certaintyA 12-week trial of 136 people with hand osteoarthritis or psoriatic arthritis found no difference from placebo. Nerve painVery low certaintyA 2026 Cochrane review found no clear effect for CBD-dominant medicines. One tiny skin-oil trial of 29 people reported lower pain scores. Crohn's diseaseVery low certaintyA 20-person trial found low-amount CBD was safe but no better than placebo. Blood pressureVery low certaintyOne dose lowered systolic pressure about 6 mmHg in nine healthy men. This is a physiology finding, not a treatment. Type 2 diabetesVery low certaintyA 13-week pilot of 62 people showed no placebo-controlled benefit of CBD alone. Parkinson's diseaseMixed findingsA 2022 review rated the Parkinsonism evidence moderate, but a 51-person trial of a CBD product with a little THC was null on its main memory test.

Benefits, one by one

Each card is a research question, not a promise. Tap a goal to filter. Open a card to see the studies, doses and limits.

Seizures in Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex (prescription CBD medicine)

High certainty

This is the one use with strong proof, and it applies to a prescription medicine, not to shop products. In five controlled studies in over 900 patients, a purified CBD solution added to usual seizure medicines cut seizures more than placebo. Regulators in the US and the EU approved it for these rare childhood-onset epilepsies.

11 studiesPrescription CBD oral solution (Epidiolex in the US, Epidyolex in the…

Studies, doses and limits

What the trials found

Randomized, double-blind, placebo-controlled trials of the prescription CBD oral solution added to usual medicines: Dravet syndrome (120 children and young adults, then 199 children), Lennox-Gastaut syndrome (171, then 225 patients) and tuberous sclerosis complex (224 patients). A meta-analysis of the four Lennox-Gastaut and Dravet trials found CBD worked with and without clobazam, though the effect looked larger with clobazam. A 2022 review of 152 cannabinoid trials graded the CBD epilepsy evidence as high.

Results

Dravet syndrome: monthly convulsive seizures fell from 12.4 to 5.9 with CBD versus 14.9 to 14.1 with placebo; the share with at least 50% fewer convulsive seizures was 43% versus 27%, which was not statistically significant. Lennox-Gastaut: drop seizures fell by 43.9% versus 21.8% (one trial) and by 41.9% and 37.2% versus 17.2% (the other trial, two amounts). Tuberous sclerosis: 48.6% and 47.5% fewer seizures versus 26.5% with placebo. The European regulator reports reductions of up to 64% versus up to 31% (placebo) for drop seizures with clobazam.

Effect sizes as reported

About 17 to 30 percentage points more seizure reduction than placebo across the main trials. Few patients became seizure-free in the Dravet trial (5% versus 0%).

Why this rating

Several large randomized, double-blind placebo-controlled trials with consistent results, regulator approval in the US and EU, and a 2022 meta-analysis that graded the CBD epilepsy evidence as high. The trials were funded by the manufacturer, and they tested an add-on to other medicines in specific syndromes.

Form and dose studied

Prescription CBD oral solution (Epidiolex in the US, Epidyolex in the EU), taken with other seizure medicines under a doctor's care. It is a medicine with liver testing and interaction checks. It is not the same as CBD oil, gummies or vapes from a shop. Amounts are set by the prescriber by body weight; this page does not give amounts.

Who was studied

Children and adults with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex. The US label covers age 1 and older; the EU page says age 2 and older.

Limits

Add-on to other medicines, 14 to 16 weeks, three specific syndromes. Not shown for other types of epilepsy in the trials we read. More patients stopped on CBD because of side effects, and raised liver enzymes, sleepiness and diarrhea were common.

Why researchers looked

Children with these epilepsies often keep having seizures despite several medicines, and parents had already been trying CBD products. That pushed the company and regulators to test a purified, standard-strength CBD solution in controlled trials.

Traditional use

No traditional use is described in the sources read for these epilepsies. The modern medicine is a purified, plant-derived CBD solution made to a fixed strength.

Proposed mechanism Preliminary mechanism

How it works is not clearly understood. The European regulator says CBD is thought to act on targets that control calcium movement in nerve cells and on adenosine, a brain chemical that helps suppress seizures. A systematic review of laboratory work found more than 65 reported targets and said CBD is very unlikely to work through the body's own cannabinoid system.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. The medicine needs liver blood tests before and during treatment, and interacts strongly with clobazam and valproate.

^{13 28 14 40 41 15 7 50 47 49 21}

Anxiety

Low certainty

Small trials suggest CBD may lower anxiety in some settings, such as a simulated speech test, but results across trials are mixed. A 2024 review of 11 randomized trials found contradictory results overall. This is not an approved use.

5 studiesTrials used oral CBD capsules or solutions made for research. Shop pr…

Studies, doses and limits

What the trials found

One randomized, double-blind trial of 24 treatment-naive adults with social anxiety disorder (12 CBD, 12 placebo) plus 12 healthy controls; a 2024 systematic review of 11 randomized trials across several anxiety disorders; and a retrospective chart review of 72 adults at one psychiatric clinic.

Results

In the speech-test trial, a single oral dose of CBD reduced anxiety, thinking difficulty and discomfort compared with placebo, and the CBD group looked like healthy controls. The 2024 review found results that were often contradictory, with a possible reduction in anxiety versus placebo and few side effects. In the chart review, anxiety scores fell in 57 of 72 patients (79.2%) within a month, with no comparison group.

Effect sizes as reported

No pooled effect size is given in the abstracts. The 2024 review calls the results contradictory.

Why this rating

Small trials of different disorders, products and amounts with mixed results; the best single study is a 36-person lab speaking test. The case series has no control group, so improvement may reflect other treatment or placebo effects.

Form and dose studied

Trials used oral CBD capsules or solutions made for research. Shop products vary a lot in content (see Safety tab). Amounts are not advised here.

Who was studied

Adults with social anxiety disorder (speech test); mixed anxiety disorders (review); adults at a psychiatric clinic (case series).

Limits

Small studies, mixed designs and dosing, single-dose lab tests, no long-term data. Not compared with standard anxiety treatments in the abstracts we read.

Why researchers looked

People use CBD products for worry and stress more than for any other reason, and laboratory and animal work pointed to calming effects.

Traditional use

No traditional use of isolated CBD for anxiety is described in the sources read.

Proposed mechanism Preliminary mechanism

The 2024 review says CBD may act as a partial activator of the 5-HT1A serotonin receptor and as a negative allosteric modulator of CB1 receptors, which could dampen THC's anxiety-causing effects. These are laboratory-based ideas, not proven in people.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. Talk to a doctor before using CBD for anxiety, especially if you take other medicines.

^{6 10 35 26 21}

Sleep and insomnia

Very low certainty

There is not enough good evidence that CBD helps sleep. A 2020 systematic review found insufficient evidence for any sleep disorder. One 2025 trial of a CBD and terpene blend found a very small change in sleep stages; another small 2025 trial found CBD did not harm daytime thinking.

4 studiesSublingual CBD (pilot) and an oral CBD plus eight-terpene formulation…

Studies, doses and limits

What the trials found

A 2020 systematic review (12 clinical studies of THC and CBD); a 2025 randomized crossover trial of 125 adults with insomnia using a CBD and terpene formulation; a 2025 randomized pilot of 30 adults with insomnia taking CBD nightly, testing daytime thinking and mood; and the 72-person case series.

Results

The CBD-terpene blend raised the share of the night in slow-wave plus REM sleep by 1.3 percentage points (95% CI 0.1 to 2.5; P=.03) with no change in total sleep time, measured by a wrist device. In the 30-person pilot, daytime thinking was unaffected, people on CBD reported more calmness, and dry mouth was more common. In the case series, sleep scores improved in 48 of 72 (66.7%) early on but fluctuated.

Effect sizes as reported

About 1% of the night more in deeper sleep stages in the blend trial. No effect sizes for CBD alone.

Why this rating

Reviewers found insufficient evidence; the one sizeable trial tested a blend rather than CBD alone and gave a tiny, borderline result; the other was a small pilot not built to test sleep.

Form and dose studied

Sublingual CBD (pilot) and an oral CBD plus eight-terpene formulation (blend trial). Not the same as other products. No amounts are advised here.

Who was studied

Adults with insomnia or primary insomnia.

Limits

Short trials (2 and 4 weeks), small groups, device-based sleep measures, a blend that cannot be credited to CBD alone, no long-term data.

Why researchers looked

Many people take CBD for sleep, so researchers started testing it in people with insomnia.

Traditional use

No traditional use of isolated CBD for sleep is described in the sources read.

Proposed mechanism Pharmacological theory

The 2020 review says THC and CBD interact with the body's cannabinoid system and with systems that influence anxiety, mood and the sleep and wake cycle. No firm mechanism for sleep effects was established in the sources read.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. CBD can cause drowsiness, and alcohol or other sedatives can add to it.

^{37 43 32 35}

Schizophrenia and psychosis (added to usual medicines)

Low certainty

One 6-week randomized trial of 88 patients found a small improvement in psychotic symptoms when CBD was added to antipsychotic medicine. This needs repeating, and it is not an approved use.

2 studiesOral CBD capsules/solution made for the trial, added to usual antipsy…

Studies, doses and limits

What the trials found

One multicentre randomized, double-blind, placebo-controlled trial: 88 patients with schizophrenia on their usual antipsychotic, CBD (43) or placebo (45) for 6 weeks.

Results

Compared with placebo, the CBD group had lower positive symptom scores (PANSS difference -1.4 points, 95% CI -2.5 to -0.2) and was more likely to be rated improved by clinicians (CGI-I difference -0.5, 95% CI -0.8 to -0.1). Gains in thinking (BACS) and overall functioning (GAF) were not statistically significant. Side effect rates were similar to placebo.

Effect sizes as reported

PANSS positive symptoms about 1.4 points lower than placebo.

Why this rating

One randomized trial of moderate size and short length with a small effect; the review of 152 trials grades most other CBD indications as low or very low.

Form and dose studied

Oral CBD capsules/solution made for the trial, added to usual antipsychotic medicine under medical care. Not advice.

Who was studied

Adults with schizophrenia on antipsychotics.

Limits

One trial; 6 weeks; small effect on one symptom scale; funding not stated in the abstract.

Why researchers looked

CBD does not act like THC, which can trigger psychosis, and laboratory and animal work suggested antipsychotic-like effects.

Traditional use

No traditional use is described in the sources read.

Proposed mechanism Pharmacological theory

The trial authors note CBD's effects do not appear to depend on blocking dopamine receptors, the way current antipsychotics do. The actual mechanism is unknown.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. People with psychosis should only change treatment with their psychiatrist.

^{27 7}

Cannabis use disorder (during a quit attempt)

Low certainty

In one phase 2 trial of 82 adults trying to stop cannabis, two of the three CBD amounts tested were linked to less cannabis use than placebo, alongside a talking-therapy session. It is early research.

1 studiesOral CBD capsules made for the trial; not a shop product.

Studies, doses and limits

What the trials found

One phase 2a randomized, double-blind, placebo-controlled adaptive Bayesian trial in London: 82 adults, 4 weeks of oral CBD at 200, 400 or 800 mg or placebo, with a brief motivational talking-therapy session.

Results

The 200 mg arm was dropped as ineffective at the interim look. With 400 mg, abstinence rose by 0.48 days a week versus placebo (0.15 to 0.82) and the urine THC marker fell by 94.21 ng/mL; with 800 mg the figures were +0.27 days a week (-0.09 to 0.64) and -72.02 ng/mL. No severe adverse events; 94% completed treatment.

Effect sizes as reported

About half a day more abstinence per week at 400 mg.

Why this rating

Small single-site phase 2a trial with a four-week treatment period, promising but unconfirmed; no larger trial read.

Form and dose studied

Oral CBD capsules made for the trial; not a shop product.

Who was studied

Adults with DSM-5 cannabis use disorder trying to stop.

Limits

82 people, one site, short treatment, everyone also received a talking-therapy session.

Why researchers looked

There is no approved medicine for cannabis use disorder, and CBD does not cause a high.

Proposed mechanism Laboratory research

Not established in the sources read. CBD has been suggested to dampen THC signalling at the CB1 receptor in cell studies.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab.

¹⁷

Cravings and anxiety in people with heroin use disorder (lab study)

Very low certainty

A small exploratory trial found that three days of CBD reduced drug-cue craving and anxiety compared with placebo in people who had stopped using heroin. It measured craving in the lab, not actual drug use.

1 studiesOral CBD solution/capsules in a research setting; not advice.

Studies, doses and limits

What the trials found

One exploratory randomized, double-blind, placebo-controlled trial in drug-abstinent adults with heroin use disorder (number of participants not given in the abstract).

Results

Acute CBD reduced craving and anxiety caused by drug cues versus neutral cues, with effects still seen 7 days after the last dose. CBD also lowered heart rate and salivary cortisol after cues. No effects on thinking and no serious adverse effects.

Effect sizes as reported

Not given as numbers in the abstract.

Why this rating

One exploratory lab trial; craving is a surrogate; size not given in the abstract; no study of relapse.

Form and dose studied

Oral CBD solution/capsules in a research setting; not advice.

Who was studied

Adults abstinent from heroin.

Limits

Exploratory, brief, lab measures only.

Why researchers looked

Few non-opioid options exist for opioid use disorder, and craving and anxiety drive relapse.

Proposed mechanism Pharmacological theory

Not established in the sources read.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. Anyone with opioid use disorder should talk to a clinician about proven treatments.

²⁰

Pain from hand osteoarthritis and psoriatic arthritis

Low certainty

A 12-week randomized trial of 136 people found CBD did not ease arthritis pain more than placebo.

1 studiesSynthetic CBD capsules in a trial. Not a shop product. No amounts adv…

Studies, doses and limits

What the trials found

One randomized, double-blind, placebo-controlled trial: 136 adults with hand osteoarthritis or psoriatic arthritis and moderate pain, synthetic CBD capsules or placebo for 12 weeks (129 analysed).

Results

Pain at 12 weeks differed by 0.23 mm on a 100 mm scale (95% CI -9.41 to 9.90; P=0.96). A drop of more than 30 mm happened in 22% on CBD and 21% on placebo. No significant effects on sleep, depression, anxiety or pain catastrophizing.

Effect sizes as reported

No difference from placebo.

Why this rating

One well-run randomized trial with a clear null result for these two conditions; but it is one trial, with one product and a fairly low amount.

Form and dose studied

Synthetic CBD capsules in a trial. Not a shop product. No amounts advised here.

Who was studied

Adults with hand osteoarthritis or psoriatic arthritis.

Limits

Two joint conditions only; 12 weeks; says nothing about other pain types or other products.

Why researchers looked

Many people use CBD for aching joints, but controlled trials of CBD alone for pain were lacking.

Traditional use

No traditional use of isolated CBD for joint pain is described in the sources read.

Proposed mechanism Pharmacological theory

CBD has been suggested to reduce inflammation in animal and cell work; this was not tested in this trial and the sources read do not establish it for people.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab.

⁴²

Nerve pain (neuropathy)

Very low certainty

A 2026 Cochrane review found no clear evidence that CBD-dominant medicines relieve nerve pain. One tiny trial of a skin-applied CBD oil reported lower pain scores in 29 people.

2 studiesCBD-dominant oral medicines (Cochrane) and a CBD oil applied to the s…

Studies, doses and limits

What the trials found

A Cochrane review and meta-analysis (searched to January 2025) of 21 randomized trials of cannabis-based medicines for chronic neuropathic pain, of which 5 trials (208 participants) used CBD-dominant medicines; and one 4-week randomized trial of 29 people with lower-limb peripheral neuropathy using a CBD oil on the skin.

Results

Cochrane: for CBD-dominant medicines versus placebo, no clear evidence for at least 50% pain relief (risk difference -0.08, 95% CI -0.20 to 0.05; very low-certainty). The skin oil trial reported significant reductions in intense pain, sharp pain, cold and itchy sensations versus placebo, no adverse events, and no numbers in the abstract.

Effect sizes as reported

No clear effect in the pooled CBD trials.

Why this rating

Cochrane graded the CBD evidence very low or low certainty; the positive skin-oil trial has 29 people and 4 weeks.

Form and dose studied

CBD-dominant oral medicines (Cochrane) and a CBD oil applied to the skin (small trial). Not advice.

Who was studied

Adults with chronic neuropathic pain; adults mean age 68 with lower-limb peripheral neuropathy.

Limits

Few small CBD-only trials, mixed products, short follow-up.

Why researchers looked

Current nerve-pain treatments help only a minority of people, so new options are sought.

Traditional use

Not described in the sources read.

Proposed mechanism Pharmacological theory

Not established in the sources read.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab.

^{2 45}

Crohn's disease

Very low certainty

In a small randomized trial of 20 people, low-amount oral CBD was safe but did not improve Crohn's disease more than placebo.

1 studiesOral CBD capsules in a trial.

Studies, doses and limits

What the trials found

One randomized, placebo-controlled trial: 20 adults with active Crohn's disease who had not responded to standard treatment, 10 mg CBD or placebo twice a day for 8 weeks (19 finished).

Results

Crohn's Disease Activity Index after 8 weeks was 220 (SD 122) with CBD and 216 (SD 121) with placebo (not significant). Blood counts, albumin, kidney and liver tests did not change; no side effects were observed.

Effect sizes as reported

No difference from placebo.

Why this rating

Twenty people, a low amount, eight weeks; the authors themselves say lack of effect, a low amount, small numbers or lack of other cannabinoids could explain the null result.

Form and dose studied

Oral CBD capsules in a trial.

Who was studied

Adults with moderately active Crohn's disease.

Limits

Very small; low amount; a single trial.

Why researchers looked

CBD was anti-inflammatory in a mouse model of inflammatory bowel disease, and the trial authors saw its lack of effects on the brain as an advantage.

Traditional use

Not described in the sources read.

Proposed mechanism Animal research

Anti-inflammatory effects in a mouse model, as cited by the trial authors.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab.

³¹

Blood pressure (single dose in healthy men)

Very low certainty

One small crossover study found a single large oral dose of CBD lowered resting blood pressure by about 6 mmHg in nine healthy men while raising heart rate. This is a physiology finding, not a treatment for high blood pressure.

1 studiesA single oral CBD dose in a research setting.

Studies, doses and limits

What the trials found

One randomized, placebo-controlled, double-blind crossover study in nine healthy men given a single oral dose.

Results

Resting systolic blood pressure fell by 6 mmHg (P<0.05) and stroke volume by 8 ml, with heart rate up by about 10 beats a minute. Blood pressure responses to stress and cold were also lower.

Effect sizes as reported

About 6 mmHg lower systolic pressure after one dose.

Why this rating

Nine healthy volunteers and a single dose; no trial in people with high blood pressure was read.

Form and dose studied

A single oral CBD dose in a research setting.

Who was studied

Healthy adult men.

Limits

Tiny, single dose, healthy men only. The authors say the heart-rate and blood-pressure changes should be considered by people taking CBD.

Why researchers looked

Animal work showed CBD reduced blood pressure responses to stress, so researchers tested people.

Traditional use

Not described in the sources read.

Proposed mechanism Animal research

Not established; the authors note preclinical studies show cardiovascular benefits.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. CBD may lower blood pressure and speed the heartbeat; people on blood pressure medicines should ask their doctor.

²³

Type 2 diabetes (blood sugar and blood fats)

Very low certainty

A 13-week pilot of 62 people with type 2 diabetes did not show a placebo-controlled benefit of CBD alone. The compound THCV, not CBD, improved blood sugar in that trial.

1 studiesOral CBD capsules in a trial.

Studies, doses and limits

What the trials found

One randomized, double-blind, placebo-controlled pilot of 62 adults with non-insulin-treated type 2 diabetes, five arms including a CBD-only arm (100 mg twice daily) for 13 weeks.

Results

The abstract reports no significant difference from placebo for CBD alone. Against its own starting values (not placebo) the CBD arm showed lower resistin and higher GIP. THCV improved fasting glucose versus placebo.

Effect sizes as reported

No placebo-controlled effect for CBD alone reported.

Why this rating

A small five-arm pilot with about a dozen people per arm; null for CBD alone.

Form and dose studied

Oral CBD capsules in a trial.

Who was studied

Adults with non-insulin-treated type 2 diabetes.

Limits

Pilot; small arms; CBD only compared with placebo in one arm.

Why researchers looked

Animal work suggested cannabinoids affect fat and sugar handling.

Traditional use

Not described in the sources read.

Proposed mechanism Animal research

Not established in the sources read.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. CBD can interact with some diabetes and other medicines through liver enzymes.

²²

Parkinson's disease

Mixed findings

A 2022 review graded CBD's effect in Parkinsonism as moderate-quality, but the one recent 12-week trial we read (51 people, with a product that also held a little THC) found no difference on its main memory test.

2 studiesOral or sublingual CBD products in trials; the 2025 product contained…

Studies, doses and limits

What the trials found

A 2022 meta-analysis of cannabinoid trials (152 randomized trials in total) that reports a significant CBD effect in Parkinsonism; and a 2025 randomized trial of 60 people with Parkinson's disease (51 analysed) given a sublingual CBD-enriched product with a little THC for 12 weeks.

Results

Meta-analysis: CBD in Parkinsonism SMD -0.41 (95% CI -0.75 to -0.08), moderate grade of evidence. 2025 trial: no difference on delayed recall; naming improved by 0.37 points (95% CI 0.01 to 0.73); other outcomes did not differ except higher alkaline phosphatase in the CBD group; no serious side effects.

Effect sizes as reported

SMD -0.41 in the meta-analysis; no difference on the main outcome in the 2025 trial.

Why this rating

The meta-analysis rates the evidence moderate but we did not read its individual trials; the trial we did read was null on its main outcome and used a product with some THC.

Form and dose studied

Oral or sublingual CBD products in trials; the 2025 product contained about 5% as much THC as CBD.

Who was studied

People with Parkinson's disease or Parkinsonism.

Limits

Few trials, different outcomes, one trial with a product that was not pure CBD and with CBD found in the blood of only 17 patients.

Why researchers looked

CBD has been tried for movement, mood, sleep and memory problems in Parkinson's disease.

Traditional use

Not described in the sources read.

Proposed mechanism Pharmacological theory

Not established in the sources read.

A proposed mechanism does not show that the effect happens in people.

CBD can raise liver enzymes, cause sleepiness, and change how some other medicines work. See the Safety tab. A liver-related blood test (alkaline phosphatase) was higher with CBD in the 2025 trial.

^{7 29}

Connections to the rest of health

How CBD (Cannabidiol) connects to the rest of our health knowledge graph. Every link says what kind of connection it is. A safety link (for example to the liver) is a warning, not a benefit. A study that merely mentions a topic never creates a benefit link.

Explore every connection: goals, body systems, organs, symptoms, genes

Kind of connection Connected to Search

A. Wellness goals

Connected to	Kind of connection	Evidence status	Context	Studies
Anxiety Support	Potential benefit	Low certainty	Anxiety: Small trials suggest CBD may lower anxiety in some settings, such as a simulated speech test, but results across trials are mixed. A 2024 review of 11 randomized trials found contradictory results overall. This is not an approved use.	^{6 10 35 26 21}
Calm Relaxation	Potential benefit	Low certainty	Anxiety: Small trials suggest CBD may lower anxiety in some settings, such as a simulated speech test, but results across trials are mixed. A 2024 review of 11 randomized trials found contradictory results overall. This is not an approved use.	^{6 10 35 26 21}
Better Sleep		Very low certainty	Sleep and insomnia: There is not enough good evidence that CBD helps sleep. A 2020 systematic review found insufficient evidence for any sleep disorder. One 2025 trial of a CBD and terpene blend found a very small change in sleep stages; another small 2025 trial found CBD did not harm daytime thinking.	^{37 43 32 35}
Sleep Quality		Very low certainty	Sleep and insomnia: There is not enough good evidence that CBD helps sleep. A 2020 systematic review found insufficient evidence for any sleep disorder. One 2025 trial of a CBD and terpene blend found a very small change in sleep stages; another small 2025 trial found CBD did not harm daytime thinking.	^{37 43 32 35}
Joint Support		Low certainty	Pain from hand osteoarthritis and psoriatic arthritis: A 12-week randomized trial of 136 people found CBD did not ease arthritis pain more than placebo. Nerve pain (neuropathy): A 2026 Cochrane review found no clear evidence that CBD-dominant medicines relieve nerve pain. One tiny trial of a skin-applied CBD oil reported lower pain scores in 29 people.	^{42 2 45}
Digestive Support		Very low certainty	Crohn's disease: In a small randomized trial of 20 people, low-amount oral CBD was safe but did not improve Crohn's disease more than placebo.	³¹
Gut Health		Very low certainty	Crohn's disease: In a small randomized trial of 20 people, low-amount oral CBD was safe but did not improve Crohn's disease more than placebo.	³¹
Blood Pressure Support		Very low certainty	Blood pressure (single dose in healthy men): One small crossover study found a single large oral dose of CBD lowered resting blood pressure by about 6 mmHg in nine healthy men while raising heart rate. This is a physiology finding, not a treatment for high blood pressure.	²³
Heart Health		Very low certainty	Blood pressure (single dose in healthy men): One small crossover study found a single large oral dose of CBD lowered resting blood pressure by about 6 mmHg in nine healthy men while raising heart rate. This is a physiology finding, not a treatment for high blood pressure.	²³
Blood Sugar Balance		Very low certainty	Type 2 diabetes (blood sugar and blood fats): A 13-week pilot of 62 people with type 2 diabetes did not show a placebo-controlled benefit of CBD alone. The compound THCV, not CBD, improved blood sugar in that trial.	²²
Metabolic Health		Very low certainty	Type 2 diabetes (blood sugar and blood fats): A 13-week pilot of 62 people with type 2 diabetes did not show a placebo-controlled benefit of CBD alone. The compound THCV, not CBD, improved blood sugar in that trial.	²²
Brain Health		Mixed findings	Parkinson's disease: A 2022 review graded CBD's effect in Parkinsonism as moderate-quality, but the one recent 12-week trial we read (51 people, with a product that also held a little THC) found no difference on its main memory test.	^{7 29}

B. Body systems

Connected to	Kind of connection	Evidence status	Context	Studies
Nervous System	Biological association	High certainty	Studied for: Seizures in Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex (prescription CBD medicine) Studied for: Anxiety	^{13 28 14 40 41 15 7 50 47 49 21 6 10 35 26 37 43 32 27 17 20 2 45 29}
Stress Mood	Biological association	Low certainty	Studied for: Anxiety Studied for: Schizophrenia and psychosis (added to usual medicines)	^{6 10 35 26 21 27 7 17 20}
Sleep Recovery	Biological association	Very low certainty	Studied for: Sleep and insomnia	^{37 43 32 35}
Musculoskeletal System	Biological association	Low certainty	Studied for: Pain from hand osteoarthritis and psoriatic arthritis	⁴²
Digestive System	Biological association	Very low certainty	Studied for: Crohn's disease	³¹
Cardiovascular System	Biological association	Very low certainty	Studied for: Blood pressure (single dose in healthy men)	²³
Endocrine System	Biological association	Very low certainty	Studied for: Type 2 diabetes (blood sugar and blood fats)	²²
Metabolism Weight	Biological association	Very low certainty	Studied for: Type 2 diabetes (blood sugar and blood fats)	²²

C. Body parts and organs

Connected to	Kind of connection	Evidence status	Context	Studies
Liver	Adverse effect or risk	High certainty	ALT over 3 times normal in 12% to 13% versus 1% placebo in controlled epilepsy trials; 5.6% versus 0% in a 2025 healthy-adult trial. Preparation or compound: Prescription CBD solution; higher risk with valproate; also seen in healthy-adult trials	^{50 16 44}
Brain and nervous system	Adverse effect or risk	High certainty	Sleepiness and sedation 32% versus 11% placebo; more with clobazam or alcohol. Preparation or compound: Prescription CBD solution	^{50 9}
Digestive system	Adverse effect or risk	High certainty	Diarrhea and decreased appetite are among the most common effects. Preparation or compound: CBD by mouth	^{50 9}

E. Health conditions

Connected to	Kind of connection	Evidence status	Context	Studies
Pregnancy	Contraindication	Low certainty	Animal data suggest fetal harm; no adequate human data. Preparation or compound: CBD	^{50 19}
Product quality		Moderate certainty	37 of 80 products at least 10% off the label; THC found in 52 of 81 samples. Preparation or compound: Hemp-derived CBD oils	^{24 25}

I. Biomarkers

Connected to	Kind of connection	Evidence status	Context	Studies
Convulsive seizure frequency (Dravet syndrome)	Biological association	High certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	^{13 28}
Drop seizures (Lennox-Gastaut syndrome)	Biological association	High certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	^{40 14}
Seizures in tuberous sclerosis complex	Biological association	High certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	⁴¹
Anxiety during a simulated public-speaking test	Biological association	Low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	⁶
Positive psychotic symptoms (PANSS)	Biological association	Low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	²⁷
Days without cannabis per week and urinary THC marker	Biological association	Low certainty	Measured in trials: raised. A lab change is not the same as a health benefit.	¹⁷
Craving triggered by drug cues	Biological association	Very low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	²⁰
Share of the night in slow-wave plus REM sleep	Biological association	Very low certainty	Measured in trials: raised. A lab change is not the same as a health benefit.	⁴³
Pain intensity (0 to 100 mm scale)	Biological association	Low certainty	Measured in trials: no change. A lab change is not the same as a health benefit.	⁴²
At least 50% pain relief (neuropathic pain)	Biological association	Very low certainty	Measured in trials: no change. A lab change is not the same as a health benefit.	²
Crohn's Disease Activity Index (CDAI)	Biological association	Very low certainty	Measured in trials: no change. A lab change is not the same as a health benefit.	³¹
Resting systolic blood pressure	Biological association	Very low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	²³
Memory delayed-recall score (MoCA)	Biological association	Low certainty	Measured in trials: no change. A lab change is not the same as a health benefit.	²⁹

J. Genes, receptors and enzymes

Connected to	Kind of connection	Evidence status	Context	Studies
CYP2C19		Moderate certainty	CBD is a moderate inhibitor, so levels of CYP2C19 drugs can change. Preparation or compound: Cannabidiol	⁵⁰
CYP2C19	Preliminary mechanism	Human	CBD is a moderate inhibitor, so CYP2C19 drugs (including the active form of clopidogrel) can change; the 2021 healthy-volunteer liver study found no link between liver enzyme rises and CYP2C19 genotype.	^{50 44}
CNR1 (CB1 receptor)	Preliminary mechanism	In vitro	CBD dampens THC and 2-AG signalling at CB1 in cell studies.	²⁶
GPR55	Preliminary mechanism	Lab animal	One of the calcium-related targets judged plausible in a systematic review of laboratory work.	²¹
HTR1A (5-HT1A receptor)	Preliminary mechanism	Lab animal	Proposed target for anxiety effects; not shown in the trials read.	¹⁰

K. Biological pathways

Connected to	Kind of connection	Evidence status	Context	Studies
Drug metabolism (CYP2C8, CYP1A2, CYP2B6, UGT1A9, P-gp)	Preliminary mechanism	Human	Enzymes and transporters that CBD affects, explaining many interactions.	⁵⁰

L. Active compounds

Connected to	Kind of connection	Evidence status	Context	Studies
Cannabidiolic acid (CBDA)	Biological association	Authoritative guidance	The plant's acid precursor of CBD; becomes CBD by decarboxylation with heat, light or age (WHO). Preparation or compound: Not read in the sources	^{46 51}
Cannabigerolic acid (CBGA)	Biological association	Authoritative guidance	The shared precursor: CBDA synthase turns it into CBDA and THCA synthase into THCA. Preparation or compound: Not read in the sources	^{46 38 51}
Delta-9-THC	Biological association	In vitro	The intoxicating cannabinoid. In cells, CBD reduced the strength of THC signalling at CB1 receptors. Under lab acid conditions CBD can be converted to THC; WHO says this likely does not happen after oral CBD in humans. Preparation or compound: Not read in the sources	^{26 46 51}
7-Hydroxycannabidiol (7-OH-CBD)	Biological association	Human	Active breakdown product of CBD (label and early human study).	^{39 30 51}
7-Carboxycannabidiol (7-COOH-CBD)	Biological association	Human	The major human metabolite; animal studies in the label of this metabolite showed developmental effects at higher exposures.	^{39 50 51}

Each connection is stored with the ingredient, the connected record, the kind of connection, the evidence status, the supporting studies, the preparation or compound involved, and a review date (last reviewed October 25, 2026).

Evidence at a glance: what changed in 9 lab markers

Tap a number to show only those markers. Counts are markers, not people.

Other72

The evidence

The research centre for CBD (Cannabidiol). Every record here was checked against PubMed or the publisher's page; we never list a study we have not verified.

Evidence by outcome

Seizures in Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex (prescription CBD medicine) 11 studies · 939 people in trials High certainty Anxiety 5 studies · 108 people in trials Low certainty Sleep and insomnia 4 studies · 227 people in trials Very low certainty Schizophrenia and psychosis (added to usual medicines) 2 studies · 88 people in trials Low certainty Cannabis use disorder (during a quit attempt) 1 study · 82 people in trials Low certainty Cravings and anxiety in people with heroin use disorder (lab study) 1 study Very low certainty Pain from hand osteoarthritis and psoriatic arthritis 1 study · 136 people in trials Low certainty Nerve pain (neuropathy) 2 studies · 29 people in trials Very low certainty Crohn's disease 1 study · 20 people in trials Very low certainty Blood pressure (single dose in healthy men) 1 study · 9 people in trials Very low certainty Type 2 diabetes (blood sugar and blood fats) 1 study · 62 people in trials Very low certainty Parkinson's disease 2 studies · 60 people in trials Mixed findings

Bars show how many studies we found for each outcome, not how well it works. People are added up across trials only (reviews are not added, to avoid double counting).

How we rate evidence

High certainty

Several well-run trials or reviews agree; further research is unlikely to change the picture much.

Moderate certainty

Good trials point the same way, but some doubt remains.

Low certainty

Small, short or inconsistent trials. The finding could change with better research.

Very low certainty

Very little or very weak human data. Treat as a possibility only.

Mixed findings

Trials or reviews disagree with each other.

Not enough evidence

Not enough human research to say either way.

Evidence against

Good research found no benefit, or found harm.

We take the reviewers' own risk-of-bias and GRADE statements where they exist, and otherwise judge from study size, design, consistency, directness and whether a product matches what is sold. Many publications do not make evidence high. Method note: We searched PubMed and read the abstracts of the studies cited, and read the FDA consumer update and questions-and-answers page, the US prescribing information for the CBD medicine, the EMA overview and the WHO 2018 critical review in full. Compound identity was checked in PubChem. Lab numbers come from a table of published and state testing data and are not our own testing.

Study library: every study we checked 52

Design Research type Year from Search

Study	Design	People	Product, dose, length	Result
Arkell 2020 Effect of Cannabidiol and Δ9-Tetrahydrocannabinol on Driving Performance: A Randomized Clinical Trial. JAMA · PMID 33258890 · DOI	Controlled trial	26 26 healthy occasional cannabis users (mean age 23)	Vaporized cannabis: THC-dominant, CBD-dominant, THC/CBD-equivalent or placebo 13.75 mg THC and/or 13.75 mg CBD per session Four sessions with driving tests at 40 and 240 minutes · vs Placebo cannabis	At 40 to 100 minutes, lane weaving rose with THC-dominant (+2.33 cm) and THC/CBD-equivalent (+2.83 cm) cannabis but not with CBD-dominant cannabis (-0.05 cm; 95% CI -1.49 to 1.39). Nothing differed at 240 to 300 minutes. The authors caution that the CBD result may not exclude clinically important impairment and the amounts tested may not represent common use. No measurable weaving increase with CBD-dominant cannabis (-1.49 to 1.39 cm) Methods, limits and funding Outcomes: Lane weaving (SDLP) over 100 km on-road driving Limitations: Vaporized cannabis, not oral CBD; 26 young occasional users; 8.5% of drives stopped for safety. Funding and conflicts: Not stated in the abstract. Read: abstract
Ateş 2026 Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database Syst Rev · PMID 41548880 · DOI	Systematic review	2187 21 randomized trials (2,187 participants) of cannabis-based medicines for chronic neuropathic pain; 5 trials of CBD-dominant medicines (208 participants)	Cannabis-based medicines (THC-dominant, balanced THC/CBD, CBD-dominant) Varied 2 to 26 weeks · vs Placebo (one trial dihydrocodeine)	For CBD-dominant medicines versus placebo there was no clear evidence for an effect on at least 50% pain relief (risk difference -0.08, 95% CI -0.20 to 0.05; 5 studies, 208 participants; very low-certainty). Effects on global impression, serious adverse events and withdrawals were uncertain (low-certainty). THC-dominant and balanced products also showed no clear effect on 50% pain relief (very low-certainty). No clear effect for CBD-dominant medicines (-0.20 to 0.05) Methods, limits and funding Outcomes: At least 50% pain relief; patient global impression; adverse events Limitations: Only 5 small CBD-dominant trials; mixed neuropathic pain conditions. Funding and conflicts: No funding (as stated in the abstract). Read: abstract
Azubuike 2026 Nationwide Prevalence of Cannabidiol Use in Pregnancy and in Women of Reproductive Age. Obstet Gynecol · PMID 41343836 · DOI	Cross sectional	US National Survey on Drug Use and Health, 2022 and 2023 (pregnant women versus women of reproductive age)	Self-reported CBD use	In 2023, about 323 per 1,000 pregnant women and 361 per 1,000 reproductive-age women reported ever using CBD; past-30-day use was about 39 versus 113 per 1,000. The authors say clinical guidelines recommend avoiding cannabinoids including CBD in pregnancy and lactation, effects in pregnancy are unknown and clinicians should screen for use. Use is common, including in pregnancy (95% CI 259.2 to 387.6 (pregnant, ever use, 2023)) Methods, limits and funding Outcomes: How many report using CBD Limitations: Survey of use; not a safety study. Funding and conflicts: Not stated in the abstract. Read: abstract
Babalonis 2017 Oral cannabidiol does not produce a signal for abuse liability in frequent marijuana smokers. Drug Alcohol Depend · PMID 28088032 · DOI	Controlled trial	31 31 healthy frequent marijuana users	Oral CBD alone and with smoked marijuana 200, 400 or 800 mg Eight weekly sessions · vs Placebo and smoked marijuana (5.3 to 5.8% THC)	Active marijuana reliably produced abuse-related effects, but CBD was placebo-like on all measures collected (p>0.05). The authors conclude CBD did not display signals of abuse liability at the doses tested. Placebo-like on abuse-related measures (n/a) Methods, limits and funding Outcomes: Abuse-related ratings such as feeling "high" Limitations: Short single-session exposures in people who already use cannabis. Funding and conflicts: Not stated in the abstract. Read: abstract
Balachandran 2021 Cannabidiol Interactions with Medications, Illicit Substances, and Alcohol: a Comprehensive Review. J Gen Intern Med · PMID 33515191 · DOI	Review human (review)	Published reports of CBD interactions with medicines, illicit substances and alcohol	CBD in various forms	The review reports CBD interacting with anti-epileptic drugs, antidepressants, opioid analgesics and THC, and also with other common medicines such as acetaminophen and with alcohol. It notes few controlled studies and says the effect of CBD on other medicines should be tested in randomized trials. Many possible interactions; few tested (n/a) Methods, limits and funding Outcomes: Possible interacting drugs and mechanisms Limitations: Narrative review; the abstract gives no rates or sizes. Funding and conflicts: Not stated in the abstract. Read: abstract
Bergamaschi 2011 Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naïve social phobia patients. Neuropsychopharmacology · PMID 21307846 · DOI	Controlled trial	36 24 treatment-naive adults with social anxiety disorder (12 CBD, 12 placebo) and 12 healthy controls	Oral CBD, single dose, before a simulated public-speaking test 600 mg once Single session, 1.5 hours before the test · vs Placebo; plus a healthy-control group with no drug	Pretreatment with CBD significantly reduced anxiety, cognitive impairment and discomfort during the speech compared with placebo. The CBD group looked similar to healthy controls. The placebo group had higher anxiety than controls. Lower anxiety ratings than placebo in a simulated speech test (n/a) Methods, limits and funding Outcomes: Anxiety, cognitive impairment, discomfort and alert ratings; blood pressure, heart rate, skin conductance Limitations: Preliminary study; 12 people per group; one dose; a lab speaking test, not everyday anxiety. Funding and conflicts: Not stated in the abstract. Read: abstract
Bilbao 2022 Medical cannabinoids: a pharmacology-based systematic review and meta-analysis for all relevant medical indications. BMC Med · PMID 35982439 · DOI	Meta-analysis	12123 152 randomized trials (12,123 participants) of dronabinol, nabilone, CBD and nabiximols across many conditions	Medical cannabinoids, analysed by type (CBD results used here) Various Various · vs Placebo and active comparators	For CBD: significant effect for epilepsy (SMD -0.5, 95% CI -0.62 to -0.38; high grade of evidence) and for Parkinsonism (SMD -0.41, CI -0.75 to -0.08; moderate grade). The authors state all other significant therapeutic effects in the review had low, very low or no grade of evidence. SMD -0.5 for epilepsy (CBD) (-0.62 to -0.38) Methods, limits and funding Outcomes: Disease-specific outcomes, retention and adverse events Limitations: Pools many different products and conditions. The Parkinsonism result was not checked against the individual trials. Funding and conflicts: Not stated in the abstract. Read: abstract
Bonn-Miller 2017 Labeling Accuracy of Cannabidiol Extracts Sold Online. JAMA · PMID 29114823 · DOI	Product analysis product analysis	CBD extracts sold online	Commercial CBD products	The abstract states only that the study compares the ingredients on the labels of CBD products sold online with what lab analysis found. It gives no figures. Not stated in the abstract (n/a) Methods, limits and funding Outcomes: Label claims compared with laboratory analysis Limitations: We read only a one-sentence abstract. We do not quote numbers from this paper. Funding and conflicts: Not stated in the abstract. Read: abstract
Chesney 2020 Adverse effects of cannabidiol: a systematic review and meta-analysis of randomized clinical trials. Neuropsychopharmacology · PMID 32268347 · DOI	Meta-analysis	803 12 randomized trials (803 participants) of CBD for various conditions, lasting 7 days or longer	Oral CBD (mostly the prescription solution in childhood epilepsy trials) Varied At least 7 days · vs Placebo	Compared with placebo, CBD raised the odds of: stopping for any reason (OR 2.61, 95% CI 1.38 to 4.96), stopping because of adverse events (OR 2.65, 1.04 to 6.80), any serious adverse event (OR 2.30, 1.18 to 4.48), serious adverse events with abnormal liver tests (OR 11.19, 2.09 to 60.02) or pneumonia (OR 5.37, 1.17 to 24.65), decreased appetite (OR 3.56), diarrhoea (OR 2.61), somnolence (OR 2.23) and sedation (OR 4.21). Liver, sleepiness, sedation and pneumonia findings were limited to childhood epilepsy studies where CBD may have interacted with clobazam or valproate. Outside childhood epilepsy, only diarrhoea was linked to CBD (OR 5.03, 1.44 to 17.61). More adverse events than placebo, mainly in childhood epilepsy trials (see results) Methods, limits and funding Outcomes: Adverse events Limitations: Few trials outside epilepsy; no data on over-the-counter products. Funding and conflicts: Not stated in the abstract. Read: abstract
Coelho 2024 The Impact of Cannabidiol Treatment on Anxiety Disorders: A Systematic Review of Randomized Controlled Clinical Trials. Life (Basel) · PMID 39598172 · DOI	Systematic review human (review)	11 randomized trials (from 284 articles, searched 2013 to 2023) in people with various anxiety disorders	CBD in various forms and doses versus placebo or usual treatments Varied widely Varied · vs Placebo and conventional treatments	Results were often contradictory. The authors say the data suggest CBD may reduce anxiety with minimal adverse effects compared with placebo, and call for better trials with a wider range of doses and continuous dosing in specific anxiety disorders. Mixed results; possible small benefit (n/a) Methods, limits and funding Outcomes: Anxiety outcomes Limitations: Very different disorders, products and designs pooled. No numbers in the abstract. Not a meta-analysis. Funding and conflicts: Not stated in the abstract. Read: abstract
Cortopassi 2020 Warfarin dose adjustment required after cannabidiol initiation and titration. Am J Health Syst Pharm · PMID 33016308 · DOI	Case report	1 46-year-old man taking warfarin for a deep vein clot who started the prescription CBD solution for epilepsy	Prescription CBD oral solution Titrated Monitored over start and dose increase	He needed a nearly 20% warfarin dose reduction to keep INR in range. The authors cite one other report of an INR of 6.86 needing a 30% weekly warfarin reduction. They advise frequent INR checks when CBD is started. About 20% lower warfarin dose needed (n/a) Methods, limits and funding Outcomes: INR (blood clotting test) Limitations: Single case; cause not proven. Funding and conflicts: Not stated in the abstract. Read: abstract
Crockett 2020 A phase I, randomized, pharmacokinetic trial of the effect of different meal compositions, whole milk, and alcohol on cannabidiol exposure and safety in healthy subjects. Epilepsia · PMID 32012251 · DOI	Controlled trial	Healthy adults given one dose of the prescription CBD solution with different meals, milk or alcohol, or fasted	Pharmaceutical CBD oral solution 750 mg once Single doses, 96-hour sampling · vs Fasted state	Compared with fasting, a high-fat meal raised CBD AUC 3.8-fold and Cmax 5.2-fold; a low-fat meal 2.7- and 3.8-fold; whole milk 2.4- and 3.1-fold; alcohol 1.6- and 1.9-fold. Variability between people was moderate to high. CBD was tolerated with no severe or serious adverse events. Meals, milk and alcohol all raised CBD levels (n/a) Methods, limits and funding Outcomes: Blood levels of CBD and metabolites Limitations: Single doses in healthy volunteers. Funding and conflicts: Not stated in the abstract. Read: abstract
Devinsky 2017 Trial of Cannabidiol for Drug-Resistant Seizures in the Dravet Syndrome. N Engl J Med · PMID 28538134 · DOI	Controlled trial	120 120 children and young adults with Dravet syndrome and drug-resistant seizures	Pharmaceutical CBD oral solution added to usual seizure medicines 20 mg per kg of body weight per day 14 weeks of treatment after a 4-week baseline · vs Placebo added to usual seizure medicines	Median monthly convulsive seizures fell from 12.4 to 5.9 with CBD and from 14.9 to 14.1 with placebo (adjusted median difference -22.8 percentage points, P=0.01). At least 50% fewer convulsive seizures: 43% with CBD versus 27% with placebo (OR 2.00, 95% CI 0.93 to 4.30, P=0.08, not statistically significant). Seizure-free: 5% versus 0% (P=0.08). Caregivers rated overall improvement in 62% versus 34% (P=0.02). Diarrhea, vomiting, fatigue, fever, sleepiness and abnormal liver tests were more common with CBD, and more people left the trial on CBD. -22.8 percentage points in convulsive seizure frequency versus placebo (-41.1 to -5.4) Methods, limits and funding Outcomes: Change in convulsive-seizure frequency Limitations: Add-on to other seizure medicines, one syndrome, 14 weeks. More withdrawals and more adverse events with CBD. Manufacturer-funded. Funding and conflicts: Funded by GW Pharmaceuticals (the maker of the medicine). Read: abstract
Devinsky 2018 Effect of Cannabidiol on Drop Seizures in the Lennox-Gastaut Syndrome. N Engl J Med · PMID 29768152 · DOI	Controlled trial	225 225 patients aged 2 to 55 with Lennox-Gastaut syndrome and drop seizures	Pharmaceutical CBD oral solution added to usual seizure medicines 20 mg/kg/day or 10 mg/kg/day, in two divided doses 14 weeks · vs Matching placebo	Median drop-seizure reduction from baseline: 41.9% (20 mg/kg group), 37.2% (10 mg/kg group) and 17.2% (placebo); P=0.005 and P=0.002 versus placebo. Most common adverse events: sleepiness, decreased appetite and diarrhea, more often at the higher dose. Six patients (20 mg/kg group) and one (10 mg/kg group) stopped because of adverse events. Fourteen CBD-treated patients (9%) had raised liver aminotransferases. Drop seizures reduced more than with placebo at both amounts (n/a) Methods, limits and funding Outcomes: Percent change in drop-seizure frequency Limitations: Add-on trial in one syndrome, 14 weeks. Manufacturer-funded. Funding and conflicts: Funded by GW Pharmaceuticals. Read: abstract
Devinsky 2020 Cannabidiol efficacy independent of clobazam: Meta-analysis of four randomized controlled trials. Acta Neurol Scand · PMID 32592183 · DOI	Meta-analysis	Four randomized trials: two in Lennox-Gastaut syndrome, two in Dravet syndrome	Pharmaceutical CBD oral solution, analysed with and without clobazam 10 and 20 mg/kg/day · vs Placebo	Seizure frequency ratio of CBD over placebo: 0.59 (95% CI 0.52 to 0.68) with clobazam and 0.85 (0.73 to 0.98) without. 50% responder odds ratio: 2.51 (1.69 to 3.71) with clobazam and 2.40 (1.38 to 4.16) without. Sleepiness, rash, pneumonia and aggression were more common with clobazam. CBD helped with and without clobazam; the effect looked larger with clobazam (see results) Methods, limits and funding Outcomes: Seizure frequency and 50% responder rate by clobazam use Limitations: Subgroup analysis; a synergy with clobazam cannot be excluded. Same trials as above. Funding and conflicts: Not stated in the abstract. Read: abstract
Florian 2025 Cannabidiol and Liver Enzyme Level Elevations in Healthy Adults: A Randomized Clinical Trial. JAMA Intern Med · PMID 40622698 · DOI	Controlled trial	201 201 healthy adults (median age 36)	CBD at an amount the authors say falls within the range consumers take with unregulated products 5 mg/kg/day (2.5 mg/kg twice daily) 28 days · vs Placebo	8 participants on CBD (5.6%; 95% CI 1.8% to 9.3%) and 0 on placebo had ALT or AST above 3 times the upper limit. Seven met withdrawal criteria for possible drug-induced liver injury (2 at day 21, 5 at day 28). Increased eosinophils were also noted. No differences in testosterone, inhibin B or thyroid hormones. The authors call for study of long-term effects and of lower amounts. 5.6% versus 0% with liver enzymes above 3 times the upper limit (1.8 to 9.3) Methods, limits and funding Outcomes: ALT or AST above 3 times the upper limit of normal Limitations: Healthy volunteers, 4 weeks; product and amount were chosen by the researchers. Funding and conflicts: Support from a US government agency (U.S. Gov't, P.H.S., as listed in PubMed). Read: abstract
Freeman 2020 Cannabidiol for the treatment of cannabis use disorder: a phase 2a, double-blind, placebo-controlled, randomised, adaptive Bayesian trial. Lancet Psychiatry · PMID 32735782 · DOI	Controlled trial	82 82 adults meeting criteria for cannabis use disorder, during a quit attempt	Oral CBD at three amounts, plus a brief talking-therapy intervention 200 mg, 400 mg or 800 mg 4 weeks · vs Placebo	The 200 mg amount was dropped as ineffective at the interim look. At 400 mg, the urine THC-COOH ratio fell by 94.21 ng/mL versus placebo (95% interval -161.83 to -35.56) and abstinence rose by 0.48 days a week (0.15 to 0.82). At 800 mg: -72.02 ng/mL (-135.47 to -19.52) and +0.27 days a week (-0.09 to 0.64). No severe adverse events; 94% completed treatment. 400 mg: +0.48 abstinent days per week versus placebo (0.15 to 0.82 days) Methods, limits and funding Outcomes: Urinary THC-COOH to creatinine ratio; days a week without cannabis Limitations: Phase 2a, small, 4 weeks, adaptive Bayesian design in one clinic. Everyone also received a talking-therapy session. Funding and conflicts: Funded by the Medical Research Council (UK). Read: abstract
Gaston 2023 Final analysis of potential drug-drug interactions between highly purified cannabidiol and anti-seizure medications in an open-label expanded access program. Epilepsia Open · PMID 37593907 · DOI	Observational	169 169 children and adults (80 adults) in an expanded-access program for the prescription CBD solution	Pharmaceutical CBD oral solution with other seizure medicines Weight-based, rising Repeated blood-level checks over time	With higher CBD amounts, blood levels rose for clobazam, N-desmethylclobazam, free valproate, felbamate and topiramate (both groups), levetiracetam (children) and perampanel (adults). The authors say the clinical significance is likely minor. Several seizure-medicine levels rose with higher CBD (n/a) Methods, limits and funding Outcomes: Blood levels of seizure medicines as CBD amount rose Limitations: Observational; no placebo group. Funding and conflicts: Not stated in the abstract. Read: abstract
Gomez Wulschner 2026 Oral Consumption of Cannabidiol During Pregnancy Alters Behavior in Mouse Offspring. Cannabis Cannabinoid Res · PMID 41840986 · DOI	Animal research mouse	Pregnant mice and their offspring	CBD by mouth during pregnancy 10 mg/kg/day by gavage, from embryonic day 5 to birth Pregnancy · vs Control	Male offspring had higher thermal pain sensitivity; female offspring had impaired problem solving and less excitable prefrontal cortex cells. The authors say CBD crosses the placenta and enters the fetal brain. Behaviour changed in mouse offspring (n/a) Methods, limits and funding Outcomes: Offspring behaviour and brain-cell activity Limitations: Mice; results may not apply to people. Funding and conflicts: Not stated in the abstract. Read: abstract
Hurd 2019 Cannabidiol for the Reduction of Cue-Induced Craving and Anxiety in Drug-Abstinent Individuals With Heroin Use Disorder: A Double-Blind Randomized Placebo-Controlled Trial. Am J Psychiatry · PMID 31109198 · DOI	Controlled trial	Drug-abstinent adults with heroin use disorder	Oral CBD for three days, then follow-up 400 mg or 800 mg once daily for 3 days 3 days plus 7 days follow-up · vs Placebo	CBD, compared with placebo, significantly reduced craving and anxiety triggered by drug cues, and the effect was still seen 7 days after the last dose. CBD also lowered heart rate and salivary cortisol after the cues. No effects on cognition and no serious adverse effects. Less cue-induced craving and anxiety than placebo (n/a) Methods, limits and funding Outcomes: Cue-induced craving and anxiety; heart rate; salivary cortisol Limitations: Exploratory; short; measures craving in a lab, not actual drug use or relapse. Size of the trial not stated in the abstract. Funding and conflicts: Not stated in the abstract. Read: abstract
Ibeas Bih 2015 Molecular Targets of Cannabidiol in Neurological Disorders. Neurotherapeutics · PMID 26264914 · DOI	Narrative review mixed (mostly lab and animal)	Published laboratory and animal studies of CBD's molecular targets	CBD	The authors found more than 65 reported molecular targets. They conclude CBD is very unlikely to act in neurological disease through the endocannabinoid system, many reported effects occur only at concentrations too high to reach in the body, and the plausible targets (for example VDAC1, GPR55, CaV3.x) relate to intracellular calcium. No causal proof exists yet. Mechanism unresolved (n/a) Methods, limits and funding Outcomes: Molecular targets of CBD relevant to neurological disease Limitations: Review of cell and animal work. Funding and conflicts: Not stated in the abstract. Read: abstract
Jadoon 2016 Efficacy and Safety of Cannabidiol and Tetrahydrocannabivarin on Glycemic and Lipid Parameters in Patients With Type 2 Diabetes: A Randomized, Double-Blind, Placebo-Controlled, Parallel Group Pilot Study. Diabetes Care · PMID 27573936 · DOI	Controlled trial	62 62 adults with non-insulin-treated type 2 diabetes	Oral CBD (and THCV, and combinations) CBD 100 mg twice daily (one arm) 13 weeks · vs Placebo	The abstract reports no significant difference from placebo for CBD alone. Against its own starting values (not against placebo) the CBD arm showed lower resistin and higher GIP, a gut hormone. THCV, not CBD, lowered fasting glucose versus placebo, and the authors suggest THCV as a possible new agent for glycemic control. CBD and THCV were well tolerated. No significant difference from placebo for CBD alone (n/a) Methods, limits and funding Outcomes: HDL cholesterol (main); glucose, insulin and other measures Limitations: Pilot with five arms and about 12 people each. Funding and conflicts: Not stated in the abstract. Read: abstract
Jadoon 2017 A single dose of cannabidiol reduces blood pressure in healthy volunteers in a randomized crossover study. JCI Insight · PMID 28614793 · DOI	Controlled trial	9 9 healthy men	Oral CBD, single dose 600 mg once Single-dose crossover · vs Placebo	CBD lowered resting systolic blood pressure by 6 mmHg (P<0.05) and stroke volume by 8 ml, with heart rate up about 10 beats a minute. Blood pressure responses to stress were also lower. The authors suggest these effects should be considered by people taking CBD. Systolic BP -6 mmHg after a single dose (n/a) Methods, limits and funding Outcomes: Blood pressure, heart rate, stroke volume, forearm blood flow Limitations: Nine healthy men, one dose; not a trial in people with high blood pressure. Funding and conflicts: Not stated in the abstract. Read: abstract
Johnson 2022 Label accuracy of unregulated cannabidiol (CBD) products: measured concentration vs. label claim. J Cannabis Res · PMID 35658956 · DOI	Product analysis product analysis	80 80 hemp-derived CBD products bought online and from local shops in central Kentucky, with Epidiolex as a control	Commercial hemp-derived CBD oils Label claims 7.5 to 60 mg/mL · vs Label claim and Epidiolex	Measured CBD ranged from 2.9 to 61.3 mg/mL. 37 of 80 products were at least 10% different from the label: 12 held under 90% and 25 held over 110% of the claim. Agreement within 10% was 54%. The authors say nearly half were not properly labelled and call for good manufacturing practice and testing standards. 37 of 80 products off the label by 10% or more (n/a) Methods, limits and funding Outcomes: Measured CBD versus label claim (LC-MS/MS) Limitations: One region and time point; oils only. Funding and conflicts: Not stated in the abstract. Read: abstract
Johnson 2022 Cannabidiol (CBD) product contamination: Quantitative analysis of Δ9-tetrahydrocannabinol (Δ9-THC) concentrations found in commercially available CBD products. Drug Alcohol Depend · PMID 35690015 · DOI	Product analysis product analysis	80 The same 80 hemp-derived CBD products plus Epidiolex	Commercial hemp-derived CBD oils · vs Epidiolex as a regulated control	THC was detected above the limit of quantification in 52 of the 81 samples tested (80 products and Epidiolex), from 0.008 to 2.071 mg/mL. Twenty-one products were labelled "THC-free", and 5 of those held detectable THC (0.015 to 0.656 mg/mL). The authors warn about accidental THC intake, impaired driving and drug-test consequences. THC found in most samples tested (n/a) Methods, limits and funding Outcomes: Delta-9 THC content (LC-MS/MS) Limitations: One region; THC level in mg/mL per product, not per serving. Funding and conflicts: Support from US National Institutes of Health (as listed in PubMed). Read: abstract
Laprairie 2015 Cannabidiol is a negative allosteric modulator of the cannabinoid CB1 receptor. Br J Pharmacol · PMID 26218440 · DOI	Laboratory research in-vitro	Cultured human kidney cells and a mouse striatal neuron cell model carrying CB1 receptors	CBD with 2-AG or THC Laboratory concentrations · vs Cannabinoid alone	CBD reduced the strength and potency of 2-AG and THC signalling at CB1 receptors and stopped receptor internalization. The authors describe CBD as a non-competitive negative allosteric modulator of CB1. CBD dampens THC and 2-AG signalling at CB1 in cells (n/a) Methods, limits and funding Outcomes: CB1 receptor signalling Limitations: Cell work only; not shown in people. Funding and conflicts: Not stated in the abstract. Read: abstract
McGuire 2018 Cannabidiol (CBD) as an Adjunctive Therapy in Schizophrenia: A Multicenter Randomized Controlled Trial. Am J Psychiatry · PMID 29241357 · DOI	Controlled trial	88 88 patients with schizophrenia on their usual antipsychotic medicine	Oral CBD added to antipsychotics 1000 mg/day 6 weeks · vs Placebo	Compared with placebo, the CBD group had lower positive psychotic symptoms (PANSS difference -1.4, 95% CI -2.5 to -0.2) and was more likely to be rated improved by the clinician (CGI-I difference -0.5, -0.8 to -0.1). Improvements in cognition (BACS 1.31, -0.10 to 2.72) and overall functioning (GAF 3.0, -0.4 to 6.4) fell short of statistical significance. CBD was well tolerated; adverse event rates were similar to placebo. PANSS positive symptoms -1.4 points versus placebo (-2.5 to -0.2) Methods, limits and funding Outcomes: PANSS (psychotic symptoms), BACS (cognition), GAF (functioning), CGI Limitations: Exploratory trial, 6 weeks, small difference in symptom score; needs replication. Funding and conflicts: Not stated in the abstract. Read: abstract
Miller 2020 Dose-Ranging Effect of Adjunctive Oral Cannabidiol vs Placebo on Convulsive Seizure Frequency in Dravet Syndrome: A Randomized Clinical Trial. JAMA Neurol · PMID 32119035 · DOI	Controlled trial	199 199 children aged 2 to 18 with Dravet syndrome	Pharmaceutical CBD oral solution added to usual seizure medicines 10 or 20 mg/kg/day 14 weeks · vs Matched placebo	Reduction from baseline: 48.7% (10 mg/kg), 45.7% (20 mg/kg) and 26.9% (placebo). Most common adverse events: decreased appetite, diarrhea, sleepiness, fever and fatigue. Raised liver transaminases: 13 patients in the 20 mg/kg group and 3 in the 10 mg/kg group, all of them also taking valproate. 29.8% reduction beyond placebo at 10 mg/kg (P=.01); 25.7% at 20 mg/kg (P=.03) (8.4 to 46.2 (10 mg/kg); 2.9 to 43.2 (20 mg/kg)) Methods, limits and funding Outcomes: Change in convulsive seizure frequency Limitations: Children only; add-on; 14 weeks. Funding and conflicts: Not stated in the abstract. Read: abstract
Mitarnun 2025 Cannabidiol and cognitive functions/inflammatory markers in Parkinson's disease: A double-blind randomized controlled trial at Buriram Hospital (CBD-PD-BRH trial). Parkinsonism Relat Disord · PMID 40267585 · DOI	Controlled trial	60 60 people with Parkinson's disease randomized (51 analysed)	Sublingual CBD-enriched product that also contained a little THC Mean CBD 26 mg/day; THC about 1.2 mg/day 12 weeks · vs Placebo	No group difference on delayed recall. CBD group improved on naming (mean difference 0.37, 95% CI 0.01 to 0.73); other outcomes did not differ, except higher alkaline phosphatase (a liver-related blood test) in the CBD group. No serious side effects. CBD was detected in only 17 patients and no THC was found in blood. No difference on the main memory test (0.01 to 0.73 for naming) Methods, limits and funding Outcomes: MoCA delayed recall score; other thinking, movement, mood and inflammatory measures Limitations: Product was not pure CBD; a low amount; 9 people lost to follow-up. Funding and conflicts: Not stated in the abstract. Read: abstract
Morrison 2019 A Phase I, Open-Label, Pharmacokinetic Trial to Investigate Possible Drug-Drug Interactions Between Clobazam, Stiripentol, or Valproate and Cannabidiol in Healthy Subjects. Clin Pharmacol Drug Dev · PMID 30791225 · DOI	Non randomised trial	Healthy volunteers in three drug-interaction sub-studies	Pharmaceutical CBD oral solution with clobazam, stiripentol or valproate Fixed-sequence, open label	Adding CBD left clobazam levels about the same (1.2-fold) but raised its active metabolite N-desmethylclobazam 3.4-fold; stiripentol rose slightly (AUC 1.6-fold); no clinically relevant effect on valproate levels. Clobazam raised the active CBD metabolite 7-OH-CBD (AUC 1.5-fold). Valproate did not change CBD levels. No deaths or serious adverse events. N-desmethylclobazam 3.4-fold higher with CBD (n/a) Methods, limits and funding Outcomes: Blood levels of CBD, its metabolites and each partner drug Limitations: Healthy volunteers; short; open label. Funding and conflicts: Not stated in the abstract. Read: abstract
Naftali 2017 Low-Dose Cannabidiol Is Safe but Not Effective in the Treatment for Crohn's Disease, a Randomized Controlled Trial. Dig Dis Sci · PMID 28349233 · DOI	Controlled trial	20 20 adults with active Crohn's disease not responding to standard treatment	Oral CBD 10 mg twice a day 8 weeks treatment plus 2 weeks follow-up · vs Placebo	CDAI after 8 weeks was 220 (SD 122) with CBD and 216 (SD 121) with placebo (not significant). Blood counts, albumin, kidney and liver tests unchanged; no side effects. The authors say CBD was safe but had no beneficial effects, and note the low amount and small group. No difference from placebo (n/a) Methods, limits and funding Outcomes: Crohn's Disease Activity Index (CDAI) Limitations: 20 people, low amount, 8 weeks. Funding and conflicts: Not stated in the abstract. Read: abstract
Narayan 2025 The effect of nightly use of 150 mg cannabidiol on daytime neurocognitive performance in primary insomnia: a randomized controlled pilot trial. Psychopharmacology (Berl) · PMID 39153080 · DOI	Controlled trial	30 30 adults with primary insomnia	CBD taken under the tongue nightly (cognition and mood were the main outcomes) 150 mg nightly 2 weeks double-blind after a 1-week placebo run-in · vs Placebo	Daytime cognitive performance was unaffected by nightly CBD (all p>0.05). People on CBD reported more calmness, clear-headedness and coordination and were more likely to report dry mouth. The authors describe an overall favourable safety profile. No change in daytime thinking; some mood ratings better (n/a) Methods, limits and funding Outcomes: Daytime thinking tests, daily mood, side effects Limitations: Small pilot, 2 weeks. It was not designed to test whether CBD improves sleep itself. Funding and conflicts: Not stated in the abstract. Read: abstract
Patsalos 2020 Clinical implications of trials investigating drug-drug interactions between cannabidiol and enzyme inducers or inhibitors or common antiseizure drugs. Epilepsia · PMID 32918835 · DOI	Review human (review)	Six trials of the prescription CBD solution in healthy volunteers or people with epilepsy	Pharmaceutical CBD oral solution	CBD did not affect CYP3A4 activity. One potentially important interaction was identified: with clobazam, which raised exposure to the major metabolites of both drugs; reducing clobazam may be considered if its side effects appear. A small rise in stiripentol; no clinically important change with valproate. The authors judged the potential for interactions low except for clobazam. Low interaction potential except clobazam (n/a) Methods, limits and funding Outcomes: Drug-drug interactions with enzyme inducers, inhibitors and seizure medicines Limitations: Company-linked trials of one product; seizure-medicine focus. Funding and conflicts: Not stated in the abstract. Read: abstract
Perucca 2020 Critical Aspects Affecting Cannabidiol Oral Bioavailability and Metabolic Elimination, and Related Clinical Implications. CNS Drugs · PMID 32504461 · DOI	Review human (review)	Published human data on oral CBD absorption and elimination	Oral CBD	The authors estimate absolute oral bioavailability of CBD is about 6% when fasting and rises about fourfold with a high-fat meal. They estimate roughly 70 to 75% of an absorbed oral dose is removed by the liver before reaching the blood, and say this variability affects response, side effects and interactions. Low and variable oral absorption (n/a) Methods, limits and funding Outcomes: Bioavailability and elimination Limitations: Estimates built from other studies; no new data. Funding and conflicts: Not stated in the abstract. Read: abstract
Shannon 2019 Cannabidiol in Anxiety and Sleep: A Large Case Series. Perm J · PMID 30624194 · DOI	Case series	72 72 adults at one psychiatric clinic with anxiety (47) or poor sleep (25), from 103 chart reviews	CBD added to usual care (a clinic-chosen product) Monthly scores over the review period	Anxiety scores fell within the first month in 57 patients (79.2%) and stayed lower. Sleep scores improved within the first month in 48 patients (66.7%) but fluctuated over time. CBD was well tolerated in all but 3 patients. Scores improved in most who stayed in the review (n/a) Methods, limits and funding Outcomes: Anxiety and sleep scores Limitations: Retrospective chart review, no comparison group, CBD given alongside usual treatment; improvement may reflect other care or placebo effects. Funding and conflicts: Not stated in the abstract. Read: abstract
Smythe 2023 Anticoagulant drug-drug interactions with cannabinoids: A systematic review. Pharmacotherapy · PMID 37740600 · DOI	Systematic review human (review)	7 case reports of cannabis products with warfarin among 959 citations screened	Cannabis products including CBD and THC Mostly very high or unknown Varied	Apart from warfarin, evidence for any anticoagulant interaction is non-existent. In the 7 warfarin reports, INR rose in 6, bleeding occurred in 1 (14.2%), 4 needed a 22% to 31% warfarin reduction. Doses involved were very high or unknown (for example above 600 mg/day CBD). The authors call the data very low quality and say a drug-interaction study is needed. INR rose in 6 of 7 case reports (n/a) Methods, limits and funding Outcomes: Bleeding, clots and INR change Limitations: Case reports only; high or unknown amounts; THC and CBD mixed. Funding and conflicts: Not stated in the abstract. Read: abstract
Suraev 2020 Cannabinoid therapies in the management of sleep disorders: A systematic review of preclinical and clinical studies. Sleep Med Rev · PMID 32603954 · DOI	Systematic review human and animal (review)	14 preclinical and 12 clinical studies of THC and CBD in sleep disorders	THC and CBD therapies Varied Varied · vs Varied	The authors found insufficient evidence to support routine clinical use of cannabinoid therapies for any sleep disorder, because of little research and a moderate-to-high risk of bias in most studies. They say preliminary evidence justifies larger controlled trials in sleep apnea, insomnia, PTSD-related nightmares, restless legs syndrome, REM sleep behaviour disorder and narcolepsy. Insufficient evidence (n/a) Methods, limits and funding Outcomes: Efficacy and safety for sleep disorders Limitations: Review of mixed cannabinoids and designs. Funding and conflicts: Not stated in the abstract. Read: abstract
Taura 2007 Cannabidiolic-acid synthase, the chemotype-determining enzyme in the fiber-type Cannabis sativa. FEBS Lett · PMID 17544411 · DOI	Laboratory research plant enzyme	Enzyme cloned from fiber-type cannabis plants and tested in the laboratory	Recombinant enzyme	CBDA synthase converts cannabigerolic acid (CBGA) into CBDA, the main cannabinoid of fiber-type cannabis. It is a covalently flavinylated oxidase with structure and function similar to THCA synthase. Identifies the plant enzyme that makes CBDA (n/a) Methods, limits and funding Outcomes: Enzyme identity and function Limitations: Laboratory enzyme work. Funding and conflicts: Not stated in the abstract. Read: abstract
Taylor 2018 A Phase I, Randomized, Double-Blind, Placebo-Controlled, Single Ascending Dose, Multiple Dose, and Food Effect Trial of the Safety, Tolerability and Pharmacokinetics of Highly Purified Cannabidiol in Healthy Subjects. CNS Drugs · PMID 30374683 · DOI	Controlled trial	Healthy adult volunteers in single-dose, multiple-dose and food-effect arms	Pharmaceutical CBD oral solution Single 1,500 to 6,000 mg; multiple 750 or 1,500 mg twice daily; food effect 1,500 mg Single doses, 7 days of twice-daily dosing · vs Placebo	CBD was generally well tolerated; diarrhea, nausea, headache and sleepiness were the most common events, all mild or moderate. Peak time 4 to 5 hours; terminal half-life about 60 hours after twice-daily dosing (effective half-life 10 to 17 hours). 7-carboxy-CBD was the main circulating metabolite. A high-fat meal raised peak level 4.85-fold and total exposure 4.2-fold. High-fat meal raised exposure about 4 to 5 times (n/a) Methods, limits and funding Outcomes: Safety, tolerability and blood levels Limitations: Healthy volunteers; high amounts of one pharmaceutical solution. Funding and conflicts: Not stated in the abstract. Read: abstract
Thiele 2018 Cannabidiol in patients with seizures associated with Lennox-Gastaut syndrome (GWPCARE4): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet · PMID 29395273 · DOI	Controlled trial	171 171 patients aged 2 to 55 with treatment-resistant Lennox-Gastaut syndrome (USA, Netherlands, Poland)	Pharmaceutical CBD oral solution added to usual seizure medicines 20 mg/kg/day 14 weeks · vs Matched placebo	Median reduction 43.9% with CBD versus 21.8% with placebo (estimated median difference -17.21, 95% CI -30.32 to -4.09, p=0.0135). Adverse events in 86% with CBD versus 69% with placebo, mostly mild or moderate (diarrhoea, sleepiness, fever, decreased appetite, vomiting). 14% stopped because of adverse events on CBD versus 1% on placebo. One death in the CBD group, judged unrelated to treatment. Estimated median difference -17.21 percentage points in drop-seizure frequency (-30.32 to -4.09) Methods, limits and funding Outcomes: Percent change in monthly drop-seizure frequency Limitations: Add-on trial in one syndrome, 14 weeks. Manufacturer-funded. Funding and conflicts: Funding: GW Pharmaceuticals. Read: abstract
Thiele 2021 Add-on Cannabidiol Treatment for Drug-Resistant Seizures in Tuberous Sclerosis Complex: A Placebo-Controlled Randomized Clinical Trial. JAMA Neurol · PMID 33346789 · DOI	Controlled trial	224 224 patients aged 1 to 65 with tuberous sclerosis complex and medication-resistant epilepsy	Pharmaceutical CBD oral solution added to usual seizure medicines 25 mg/kg/day or 50 mg/kg/day 16 weeks · vs Matched placebo	Seizure reduction from baseline: 48.6% (25 mg/kg), 47.5% (50 mg/kg) and 26.5% (placebo); reduction beyond placebo 30.1% (95% CI 13.9 to 43.3, P<.001) for 25 mg/kg. Diarrhea (31% and 56% versus 25%) and sleepiness (13% and 26% versus 9%) were more frequent on CBD. 28 CBD patients (18.9%) had raised liver transaminases versus none on placebo. The 25 mg/kg amount had the better safety profile. 30.1% greater reduction than placebo at 25 mg/kg/day (13.9 to 43.3) Methods, limits and funding Outcomes: Change in TSC-associated seizures from baseline Limitations: Add-on trial in one condition, 16 weeks. Funding and conflicts: Not stated in the abstract. Read: abstract
Vela 2022 Cannabidiol treatment in hand osteoarthritis and psoriatic arthritis: a randomized, double-blind, placebo-controlled trial. Pain · PMID 34510141 · DOI	Controlled trial	136 136 adults with hand osteoarthritis or psoriatic arthritis and moderate pain despite treatment (129 analysed)	Synthetic CBD capsules as add-on pain treatment 20 to 30 mg daily 12 weeks · vs Placebo	Between-group difference in pain at 12 weeks was 0.23 mm (95% CI -9.41 to 9.90; P=0.96). A pain drop of more than 30 mm occurred in 22% on CBD and 21% on placebo. No significant effects on sleep quality, depression, anxiety or pain catastrophizing. The authors found neither clinically nor statistically significant effects of CBD for pain intensity. No difference from placebo (-9.41 to 9.90 mm) Methods, limits and funding Outcomes: Pain intensity over the past 24 hours (0 to 100 mm) Limitations: Two joint conditions, 12 weeks; a fairly low amount, which the authors consider in context. Funding and conflicts: Not stated in the abstract. Read: abstract
Wang 2025 Effects of a cannabidiol/terpene formulation on sleep in individuals with insomnia: a double-blind, placebo-controlled, randomized, crossover study. J Clin Sleep Med · PMID 39167421 · DOI	Controlled trial	125 125 adults with insomnia	Oral formulation of CBD plus eight terpenes (not CBD alone) CBD 300 mg plus terpenes 1 mg each (linalool, myrcene, phytol, limonene, alpha-terpinene, alpha-terpineol, alpha-pinene, beta-caryophyllene), 4 or more days a week 4 weeks, crossover · vs Placebo	The CBD-terpene regimen marginally increased slow-wave plus REM sleep: +1.3% (SE 0.60%; 95% CI 0.1 to 2.5; P=.03). No effect on total sleep time, resting heart rate or heart rate variability. No adverse events reported. Larger increases were seen in people with low baseline slow-wave plus REM sleep. +1.3 percentage points of the night in slow-wave plus REM sleep (0.1 to 2.5) Methods, limits and funding Outcomes: Share of the night in slow-wave plus REM sleep (wrist tracker) Limitations: A blend, so the effect cannot be credited to CBD alone. Sleep measured by a wrist device, not a sleep lab. Small effect, borderline p-value. Funding and conflicts: Not stated in the abstract. Read: abstract
Watkins 2021 Cannabidiol and Abnormal Liver Chemistries in Healthy Adults: Results of a Phase I Clinical Trial. Clin Pharmacol Ther · PMID 33022751 · DOI	Non randomised trial	16 16 healthy adults	Pharmaceutical CBD oral solution, repeated daily doses, in a drug-interaction study 1,500 mg a day About 3.5 weeks (open label, single sequence)	Seven participants (44%) had peak ALT above the upper limit of normal; five (31%) had more than 5 times the upper limit, meeting international criteria for drug-induced liver injury. All rises began within 2 to 4 weeks. No link to age, sex, CYP2C19 genotype or CBD blood levels. Six people with rises were taken off the protocol, some with symptoms like hepatitis or hypersensitivity. 31% had ALT over 5 times the upper limit (n/a) Methods, limits and funding Outcomes: Liver enzymes (ALT) Limitations: Small open-label study at a high amount used in prescription epilepsy care. Funding and conflicts: Not stated in the abstract. Read: abstract
Xu 2020 The Effectiveness of Topical Cannabidiol Oil in Symptomatic Relief of Peripheral Neuropathy of the Lower Extremities. Curr Pharm Biotechnol · PMID 31793418 · DOI	Controlled trial	29 29 patients (mean age 68) with symptomatic peripheral neuropathy of the lower limbs	CBD oil applied to the skin (topical) Product containing 250 mg CBD per 3 fluid ounces; amount applied not stated in the abstract 4 weeks (placebo group could then cross over) · vs Placebo	The CBD group had a statistically significant reduction in intense pain, sharp pain, cold and itchy sensations compared with placebo. No adverse events were reported. The authors conclude the product was well tolerated. Lower scores on some pain-scale items than placebo (n/a) Methods, limits and funding Outcomes: Neuropathic Pain Scale Limitations: Only 29 people and 4 weeks; one product; no numbers in the abstract; not independent confirmation. Funding and conflicts: Not stated in the abstract. Read: abstract
n/a 2018 WHO Expert Committee on Drug Dependence, 40th meeting: Cannabidiol (CBD) Critical Review Report (June 2018) WHO · Source	Authoritative guidance n/a	n/a	n/a	Describes CBD as a 21-carbon terpenophenolic compound formed from its acid precursor CBDA by decarboxylation. Says CBD shows no effects in humans indicating abuse or dependence potential, is generally well tolerated, and that reported adverse effects may come from interactions with a patient's other medicines. Says CBD can be converted to THC in the laboratory under acid conditions but that this likely does not happen when oral CBD is taken by humans. Notes unsanctioned medical use of CBD products sold online. Methods, limits and funding Outcomes: WHO critical review of CBD: identity, plant origin, convertibility to THC, pharmacology, abuse and dependence potential, medical use. Limitations: Written in 2018, before the first CBD medicine was approved; the report calls Epidiolex still in development. The committee's later scheduling decisions were not read. Read: full-text
n/a 2019 EMA: Epidyolex (cannabidiol): EPAR overview (marketing authorisation 19 Sep 2019) EMA · Source	Authoritative guidance n/a	n/a	n/a	Authorised in the EU on 19 September 2019 as add-on treatment with clobazam for Lennox-Gastaut or Dravet syndrome from age 2, and with other epilepsy treatments for tuberous sclerosis complex from age 2. Prescription only. Five studies in over 900 patients; up to 64% fewer drop seizures with Epidyolex plus clobazam versus up to 31% with placebo plus clobazam in Lennox-Gastaut syndrome. Common side effects: sleepiness, decreased appetite, diarrhoea, fever, tiredness, vomiting. Most common reason for stopping was raised liver enzymes. Must not be used if liver enzymes are over three times the normal limit together with bilirubin over twice the normal limit. Methods, limits and funding Outcomes: European regulator overview of the CBD medicine Epidyolex: use, benefits in studies, risks and reason for authorisation. Limitations: A regulator overview page. Age limits quoted are the ones on this page; the US label uses 1 year and older. Read: full-text
n/a 2020 FDA: What You Need to Know (And What We're Working to Find Out) About Products Containing Cannabis or Cannabis-derived Compounds, Including CBD (content current as of 5 Mar 2020) FDA · Source	Authoritative guidance n/a	n/a	n/a	Says the FDA has approved only one CBD product (a prescription drug). Says CBD can cause liver injury and can change how other drugs work. Says using CBD with alcohol or other drugs that slow the brain raises the risk of sedation. Lists male reproductive toxicity in animals, and unknowns about long-term daily use, the developing brain, pregnancy and breastfeeding. Says FDA tests found many products did not contain the CBD amount claimed, and it is investigating contaminants including pesticides, heavy metals and THC. Methods, limits and funding Outcomes: FDA consumer update on CBD: one approved product, limited safety data, liver injury, drug interactions, sedation, male reproductive toxicity in animals, unknowns, product quality. Limitations: Page content is dated 5 March 2020 and has not been updated since by its own date. Read in full on 2026-10-25. Read: full-text
n/a 2024 FDA: FDA Regulation of Cannabis and Cannabis-Derived Products, Including Cannabidiol (CBD): Questions and Answers (content current as of 16 Jul 2024) FDA · Source	Authoritative guidance n/a	n/a	n/a	FDA has approved Epidiolex, a purified CBD medicine, for seizures with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex in patients 1 year and older. FDA has concluded CBD products are excluded from the dietary supplement definition and that adding CBD to food in interstate commerce is prohibited; no regulation has been issued to change that. The 2018 Farm Bill removed hemp (0.3% delta-9 THC or less on a dry-weight basis) from the Controlled Substances Act while keeping FDA authority. On pregnancy and breastfeeding the page discusses cannabis products containing THC and cites ACOG advice to avoid cannabis. Methods, limits and funding Outcomes: FDA questions and answers on approved cannabis medicines, the legal status of CBD in food and supplements, the 2018 Farm Bill, children, pregnancy and lactation. Limitations: Page is dated 16 July 2024. Later changes to US hemp law or FDA policy were not checked. Read: full-text
n/a 2026 Epidiolex (cannabidiol) oral solution: US prescribing information (DailyMed label, Jazz Pharmaceuticals; label version published 15 Jun 2026) DailyMed · Source	Authoritative guidance n/a	n/a	n/a	Approved for seizures with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex in patients 1 year and older. Warns of liver injury (ALT above 3 times the upper limit of normal in 13% and 12% of treated patients in controlled studies versus 1% on placebo), sleepiness and sedation (32% versus 11% in Lennox-Gastaut and Dravet studies), suicidal thoughts (class warning for seizure medicines) and hypersensitivity. Lists interactions with clobazam, valproate, stiripentol, everolimus and other CYP and transporter substrates. Says animal data suggest it may cause fetal harm and that there are no data on human milk. States it is not a controlled substance. Methods, limits and funding Outcomes: US prescribing information for the only FDA-approved CBD medicine: indications, liver, sedation, interactions, pregnancy, abuse and dependence. Limitations: A product label for one prescription medicine under medical supervision. It does not describe shop products. Read: full-text
n/a 2026 PubChem: related compounds, retrieved through the REST API on 2026-10-25 (CBDA CID 160570, CBGA CID 6449999, delta-9-THC CID 16078, 7-hydroxycannabidiol CID 11301963, 7-carboxycannabidiol CID 146592489) PubChem · Source	Reference database n/a	n/a	n/a	CBDA (Cannabidiolic acid) C22H30O4, CID 160570. CBGA (Cannabigerolic Acid) C22H32O4, CID 6449999. Tetrahydrocannabinol (delta-9-THC) C21H30O2, CID 16078. 7-Hydroxycannabidiol C21H30O3, CID 11301963. 7-Carboxycannabidiol C21H28O4, CID 146592489. Methods, limits and funding Outcomes: Identity of compounds related to CBD. Limitations: Database record; names and formulas as PubChem returned them. Read: api-record
n/a 2026 PubChem: Cannabidiol, compound CID 644019 (REST API record) PubChem · Source	Reference database n/a	n/a	n/a	PubChem CID 644019, title Cannabidiol, formula C21H30O2, molecular weight 314.5, IUPAC name 2-[(1R,6R)-3-methyl-6-prop-1-en-2-ylcyclohex-2-en-1-yl]-5-pentylbenzene-1,3-diol, InChIKey QHMBSVQNZZTUGM-ZWKOTPCHSA-N. CAS 13956-29-1 appears in PubChem synonyms and in the WHO report. Methods, limits and funding Outcomes: Identity of cannabidiol: CID, formula, weight, IUPAC name. Limitations: A database record. Retrieved through the PubChem REST API on 2026-10-25. Read: api-record

What we do not know yet

What the evidence supports: only the prescription solution for three epilepsies is high certainty. Every other use here is low, very low or mixed.
Well-run trials of CBD alone for anxiety, sleep and pain in everyday life are missing.
Long-term safety of daily CBD in healthy adults, older adults and teenagers is unknown, as is its effect on a fetus or breastfed infant.
How often liver injury occurs with typical shop-bought CBD is unknown.
Controlled studies of CBD with warfarin are missing; only case reports exist.

How new research is added

New studies are found by a saved PubMed search and are placed in a review queue. A person reads each one before it changes any evidence rating or safety wording.

Lab markers

What changed in blood or saliva tests during trials. A changed number is not the same as feeling or being healthier, and we give no "normal ranges".

Other 13

▼

Convulsive seizure frequency (Dravet syndrome)

Lowered in trials

How many convulsive seizures happen each month.

High certainty

Results and what it means

Why it was measured

It is the main way epilepsy trials judge whether a medicine is helping.

Statistical vs. clinical meaning

A clinically meaningful drop in seizures for many patients, though few became seizure-free.

Safety implication

More diarrhea, sleepiness, raised liver enzymes and withdrawals on CBD.

Study	Result	People	Product
¹³	Median monthly convulsive seizures fell from 12.4 to 5.9 with prescription CBD and from 14.9 to 14.1 with placebo (adjusted difference -22.8 percentage points, 95% CI -41.1 to -5.4, P=0.01).	120 children and young adults with Dravet syndrome	Prescription CBD oral solution added to usual medicines versus placebo, 14 weeks
²⁸	Reduction from baseline of 48.7% (lower amount) and 45.7% (higher amount) versus 26.9% with placebo.	199 children with Dravet syndrome	Prescription CBD oral solution at two amounts versus placebo, 14 weeks

^{13 28}

▼

Drop seizures (Lennox-Gastaut syndrome)

Lowered in trials

Sudden seizures in which the person falls or drops; counted per 28 days.

High certainty

Results and what it means

Why it was measured

Drop seizures cause injuries and are the standard outcome in Lennox-Gastaut trials.

Statistical vs. clinical meaning

Consistent in two large trials.

Safety implication

Sleepiness, decreased appetite, diarrhea, raised liver enzymes.

Study	Result	People	Product
⁴⁰	Median reduction 43.9% with prescription CBD versus 21.8% with placebo (estimated median difference -17.21, 95% CI -30.32 to -4.09, p=0.0135).	171 patients aged 2 to 55 with Lennox-Gastaut syndrome	Prescription CBD oral solution added to usual medicines versus placebo, 14 weeks
¹⁴	Median reduction 41.9% (higher amount) and 37.2% (lower amount) versus 17.2% with placebo (P=0.005 and P=0.002).	225 patients aged 2 to 55 with Lennox-Gastaut syndrome	Prescription CBD oral solution at two amounts versus placebo, 14 weeks

^{40 14}

▼

Seizures in tuberous sclerosis complex

Lowered in trials

Seizures linked to tuberous sclerosis complex, counted over the treatment period.

High certainty

Results and what it means

Why it was measured

Standard efficacy outcome for this condition.

Statistical vs. clinical meaning

Significant in a large trial; the lower amount had the better safety profile.

Safety implication

Raised liver enzymes in 18.9% of CBD patients versus none on placebo; diarrhea and sleepiness more common.

Study	Result	People	Product
⁴¹	Reduction from baseline of 48.6% (25 mg/kg/day), 47.5% (50 mg/kg/day) versus 26.5% with placebo; 30.1% beyond placebo at 25 mg/kg (95% CI 13.9 to 43.3, P<.001).	224 patients aged 1 to 65 with tuberous sclerosis complex	Prescription CBD oral solution versus placebo, 16 weeks

⁴¹

▼

Anxiety during a simulated public-speaking test

Lowered in trials

Self-rated anxiety on a mood scale while giving a speech in a lab test.

Low certainty

Results and what it means

Why it was measured

A standard way to test anxiety medicines in a lab before trials in patients.

Statistical vs. clinical meaning

A single-dose lab result in a small group.

Safety implication

Single dose; no long-term safety data from this study.

Study	Result	People	Product
⁶	CBD significantly reduced anxiety, thinking difficulty and discomfort versus placebo; CBD and healthy controls looked similar (24 patients, 600 mg single oral dose).	24 treatment-naive adults with social anxiety disorder plus 12 healthy controls	Single oral CBD dose versus placebo before the test

⁶

▼

Positive psychotic symptoms (PANSS)

Lowered in trials

A clinician-rated score of symptoms such as hallucinations and delusions.

Low certainty

Results and what it means

Why it was measured

Main way schizophrenia trials rate symptom change.

Statistical vs. clinical meaning

Statistically significant but a small difference on the scale.

Safety implication

Adverse event rates were similar to placebo.

Study	Result	People	Product
²⁷	PANSS positive score difference vs placebo -1.4 (95% CI -2.5 to -0.2) after 6 weeks.	88 patients with schizophrenia on antipsychotics	CBD added to usual antipsychotic versus placebo

²⁷

▲

Days without cannabis per week and urinary THC marker

Raised in trials

Self-reported abstinent days per week, and a urine THC-COOH to creatinine ratio.

Low certainty

Results and what it means

Why it was measured

Direct measures of whether a medicine helps people cut down or stop.

Statistical vs. clinical meaning

Promising in phase 2, not confirmed.

Safety implication

No severe adverse events; 94% completed.

Study	Result	People	Product
¹⁷	400 mg: +0.48 abstinent days a week (0.15 to 0.82) and THC-COOH ratio -94.21 ng/mL; 800 mg: +0.27 days a week (-0.09 to 0.64) and -72.02 ng/mL.	82 adults with cannabis use disorder	4 weeks of oral CBD or placebo, plus brief therapy

¹⁷

▼

Craving triggered by drug cues

Lowered in trials

How strongly a person craves the drug after being shown drug-related pictures or items.

Very low certainty

Results and what it means

Why it was measured

Craving contributes to relapse.

Statistical vs. clinical meaning

A lab surrogate; real-world relapse was not measured.

Safety implication

No serious adverse effects.

Study	Result	People	Product
²⁰	CBD reduced cue-induced craving and anxiety versus placebo, acutely and 7 days after the last of three daily doses.	Drug-abstinent adults with heroin use disorder	400 or 800 mg CBD once daily for 3 days versus placebo

²⁰

▲

Share of the night in slow-wave plus REM sleep

Raised in trials

The percentage of sleep time spent in deep (slow-wave) and dream (REM) sleep, estimated by a wrist-worn tracker.

Very low certainty

Results and what it means

Why it was measured

Tests whether a product changes sleep stages, not just how long someone sleeps.

Statistical vs. clinical meaning

A very small, borderline result from a blend, measured by a wearable.

Safety implication

No adverse events were reported.

Study	Result	People	Product
⁴³	+1.3 percentage points (95% CI 0.1 to 2.5, P=.03) with a CBD plus terpene blend; no change in total sleep time.	125 adults with insomnia	Oral CBD (300 mg) plus eight terpenes versus placebo, 4 weeks, crossover

⁴³

●

Pain intensity (0 to 100 mm scale)

No change in trials

A person marks how bad their pain has been over the past 24 hours on a 100 mm line.

Low certainty

Results and what it means

Why it was measured

The standard way to judge pain relief in trials.

Statistical vs. clinical meaning

No difference from placebo.

Safety implication

Adverse events were assessed as a safety outcome; the abstract gives no rates.

Study	Result	People	Product
⁴²	Between-group difference at 12 weeks 0.23 mm (95% CI -9.41 to 9.90; P=0.96); a drop of more than 30 mm in 22% on CBD and 21% on placebo.	136 adults with hand osteoarthritis or psoriatic arthritis	Synthetic CBD capsules versus placebo for 12 weeks

⁴²

●

At least 50% pain relief (neuropathic pain)

No change in trials

The share of people whose pain falls by half or more.

Very low certainty

Results and what it means

Why it was measured

A common benchmark for a worthwhile pain treatment.

Statistical vs. clinical meaning

No clear effect.

Safety implication

Serious adverse events and withdrawals: effect uncertain (low-certainty).

Study	Result	People	Product
²	CBD-dominant medicines versus placebo: risk difference -0.08 (95% CI -0.20 to 0.05) in 5 studies and 208 participants; very low-certainty.	Adults with chronic neuropathic pain	CBD-dominant medicines versus placebo

●

Crohn's Disease Activity Index (CDAI)

No change in trials

A score that combines symptoms and exam findings to rate how active Crohn's disease is. Lower is better.

Very low certainty

Results and what it means

Why it was measured

The standard activity measure in Crohn's trials.

Statistical vs. clinical meaning

No difference from placebo.

Safety implication

Blood, kidney and liver tests unchanged; no side effects reported.

Study	Result	People	Product
³¹	After 8 weeks: 220 (SD 122) with CBD versus 216 (SD 121) with placebo; not significant.	20 adults with active Crohn's disease	Oral CBD 10 mg twice daily versus placebo for 8 weeks

³¹

▼

Resting systolic blood pressure

Lowered in trials

The upper blood pressure number at rest.

Very low certainty

Results and what it means

Why it was measured

To see whether CBD changes blood pressure, which matters for people taking blood pressure medicines.

Statistical vs. clinical meaning

Statistically significant in a tiny single-dose study of healthy men.

Safety implication

Slower stress response of blood pressure and faster heart rate; the authors say this should be considered by people taking CBD.

Study	Result	People	Product
²³	Fell by 6 mmHg (P<0.05) after a single oral dose of CBD versus placebo; heart rate rose about 10 beats a minute.	9 healthy men	Single oral CBD dose versus placebo, crossover

²³

●

Memory delayed-recall score (MoCA)

No change in trials

Part of a short memory and thinking test: how many words a person recalls after a delay.

Low certainty

Results and what it means

Why it was measured

Main outcome of the 2025 Parkinson's trial.

Statistical vs. clinical meaning

No difference on the planned main outcome.

Safety implication

Higher alkaline phosphatase with CBD; no serious side effects.

Study	Result	People	Product
²⁹	No group difference in delayed recall; naming score improved by 0.37 (95% CI 0.01 to 0.73).	51 people with Parkinson's disease analysed	Sublingual CBD-enriched product with some THC versus placebo for 12 weeks

²⁹

Safety

What studies and official sources say about side effects, medicines and who should check first. Each point shows what kind of evidence it comes from.

Main points to know 12

One approved CBD medicine (prescription only); shop products are not the same thing. The FDA has approved only one CBD product: a prescription solution for seizures with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex in people 1 year and older. The EU authorised it for age 2 and older. It is made to a fixed strength and comes with liver tests and interaction checks. Everything else sold as CBD (oils, gummies, vapes, creams) is not an approved medicine, and its safety has not been reviewed the same way. Authoritative guidance ^{49 48 50 47}
Liver injury is the main serious risk. In controlled studies of the prescription CBD solution, liver enzyme (ALT) rises above 3 times normal happened in 12% to 13% of treated patients versus 1% on placebo, mostly in the first two months but sometimes much later. Rises were far more common with valproate. In healthy adults, a 2021 study of a high amount found 31% had ALT over 5 times the limit, and a 2025 randomized trial at 5 mg/kg a day found 5.6% with a rise above 3 times the limit versus 0% on placebo. The FDA says CBD can cause liver injury. Controlled trial ^{50 44 16 48}
Interactions with other medicines are common and some matter a lot. With clobazam, CBD raised its active breakdown product N-desmethylclobazam 3.4-fold in healthy volunteers, and sleepiness and sedation rose. Valproate with CBD raises the chance of liver enzyme rises. CBD is a moderate inhibitor of CYP2C19 and also affects CYP2C8, CYP1A2, UGT1A9 and the P-gp transporter, so blood levels of other drugs can change. Case reports describe higher INR in people on warfarin after starting CBD. Ask a pharmacist before combining CBD with any prescription medicine. Pharmacokinetic study ^{50 30 33 11 36 5}
Sleepiness and sedation. In Lennox-Gastaut and Dravet trials, 32% of treated patients had sleepiness or sedation versus 11% on placebo, and 46% of those also taking clobazam. Alcohol and other sedating drugs can add to it. The FDA says CBD with alcohol or other drugs that slow the brain raises the risk of sedation and injuries. Authoritative guidance ^{50 48 9}
Tummy and appetite effects. Diarrhea and decreased appetite are among the most common effects. In a 2020 pooled analysis of 12 randomized trials, diarrhoea was the only adverse outcome still linked to CBD after childhood epilepsy trials (where interactions with other medicines may play a part) were left out. Meta-analysis ^{9 50}
Shop CBD products often do not match their labels, and may contain THC. In 80 hemp-derived CBD oils bought in Kentucky, 37 held at least 10% more or less CBD than the label said (12 under 90%, 25 over 110%). THC was detected in 52 of 81 samples tested (80 products plus a control), including 5 of 21 products labelled THC-free. The FDA also says its own tests found many products did not contain the CBD claimed and it is investigating pesticides, heavy metals and THC. Product analysis ^{24 25 48 8}
Pregnancy and breastfeeding: avoid unless a doctor says otherwise. The prescription label says animal data suggest CBD may cause fetal harm and that there are no data on human milk. The FDA says the effects of CBD on a developing fetus or breastfed newborn are unknown. A 2026 US survey found about 39 per 1,000 pregnant women reported recent CBD use, and its authors say guidelines recommend avoiding CBD in pregnancy and lactation. In mice, CBD during pregnancy changed offspring behaviour. Authoritative guidance ^{50 48 3 19}
Mental health and mood. The prescription label carries the warning that all seizure medicines have: monitor for new or worsening depression, suicidal thoughts or unusual mood or behaviour changes. In trials, irritability, agitation and aggression occurred more often with CBD than placebo (for example 9% and 5% at two amounts versus 2%). In one trial in people with schizophrenia, CBD added to usual antipsychotics had side effect rates similar to placebo. Authoritative guidance ^{50 48 27}
Driving and impairment. In an on-road test of 26 occasional cannabis users, vaporized CBD-dominant cannabis did not increase lane weaving versus placebo, while THC-containing cannabis did. The authors say that result may not rule out meaningful impairment and the amounts may not match everyday use. The prescription label tells patients not to drive or operate machinery until they know how it affects them. Controlled trial ^{1 50}
Abuse and dependence: low signal. WHO's 2018 review found no effects in humans suggesting abuse or dependence potential. A trial in frequent cannabis users found oral CBD placebo-like on abuse-related measures. The label reports no withdrawal signs in a 28-day study of healthy adults, and that the medicine is not a controlled substance. Stopping seizure medicine suddenly is dangerous, so it should be tapered under medical care. Authoritative guidance ^{46 4 50}
Does CBD turn into THC in the body?. WHO says CBD can be converted to THC in the laboratory under acid conditions, but this likely does not happen when oral CBD is taken by humans. The authors of a 2022 product study describe the THC found in shop products as contamination. Authoritative guidance ^{46 25}
Food, and amounts in the blood. In healthy volunteers, a high-fat meal raised the peak level of the prescription solution about 5-fold and total exposure about 4-fold, and alcohol raised total exposure 1.6-fold. Oral absorption is low and varies a lot between people, so the same product can act differently from person to person. Pharmacokinetic study ^{39 12 34}

Side effects 14

Effect	How often	Severity and recovery	Evidence
Sleepiness, sedation, lethargy More with clobazam (46% versus 16% without).	32% versus 11% placebo in Lennox-Gastaut and Dravet trials (prescription CBD); 19% versus 17% in the tuberous sclerosis trial	mild to moderate Often early in treatment and may fade; reversible	Controlled trial ^{50 9 40}
Diarrhea The one adverse outcome still linked to CBD after excluding childhood epilepsy trials in a 2020 meta-analysis.	9% to 20% versus 9% placebo (Lennox-Gastaut/Dravet); 31% versus 25% (tuberous sclerosis)	mild to moderate Reversible	Controlled trial ^{50 9}
Decreased appetite, weight loss Dose-related in the label.	16% to 22% versus 5% placebo; weight down 5% or more in 16% versus 8%	mild to moderate Reversible on stopping or lowering	Controlled trial ^{50 9}
Raised liver enzymes (ALT/AST) See Liver.	ALT over 3 times normal in 12% to 13% versus 1% placebo (prescription CBD); 5.6% versus 0% in healthy adults at 5 mg/kg/day for 28 days	mild to serious Resolved after stopping or reducing in about two-thirds of label cases; about one-third resolved on continued treatment	Controlled trial ^{50 16 44}
Nausea, vomiting	Vomiting 17% versus 9% placebo and nausea 9% versus 3% (tuberous sclerosis)	mild to moderate Reversible	Controlled trial ^{50 13}
Tiredness, weakness	Fatigue, malaise, asthenia 11% to 12% versus 4% placebo	mild to moderate Reversible	Controlled trial ⁵⁰
Fever	19% versus 8% placebo (tuberous sclerosis trial)	mild to moderate Reversible	Controlled trial ^{50 47}
Rash and hypersensitivity	Rash 7% to 13% versus 3% placebo; the label's most frequent cause of stopping in the tuberous sclerosis trial was rash (5%)	mild to serious The label says to stop and not restart after hypersensitivity	Authoritative guidance ⁵⁰
Irritability, agitation, aggression, sleep problems	Irritability/agitation 5% to 9% versus 2%; aggression 3% to 5% versus under 1%; insomnia/poor sleep 5% to 11% versus 4%	mild to moderate Improves when stopped or reduced, per FDA	Authoritative guidance ^{50 48}
Pneumonia and infections (mainly with clobazam) Seen in childhood epilepsy trials.	Pneumonia 17% with 10 mg/kg and 10% with 20 mg/kg when taken with clobazam versus 1% placebo; OR 5.37 in the pooled analysis	serious Not stated in the sources read	Meta-analysis ^{50 9}
Blood changes (low red cells or platelets, high eosinophils) Eosinophilia was also noted in the 2025 healthy-adult trial.	Anemia 7% versus 1%, low platelets 5% versus 1%, eosinophils up 5% versus 0% (tuberous sclerosis trial)	mild to moderate Not stated in the sources read	Controlled trial ^{50 16}
Dry mouth	More likely than placebo in a 2-week insomnia pilot	mild Reversible	Controlled trial ³²
Lower blood pressure, faster heartbeat Not studied in people on blood pressure medicines.	About 6 mmHg lower systolic pressure and about 10 beats a minute faster in 9 healthy men after one large dose	mild in healthy men Single-dose effect	Controlled trial ²³
Suicidal thoughts or behaviour (class warning) Applies to antiseizure medicines as a class.	Pooled seizure-medicine data in the label; no CBD-specific rate read	serious Not stated	Authoritative guidance ⁵⁰

Liver

CBD can injure the liver. In controlled trials of the prescription solution, ALT above 3 times the upper limit of normal occurred in 13% (two amounts for Lennox-Gastaut and Dravet) and 12% (tuberous sclerosis) of treated patients versus 1% on placebo, and fewer than 1% had ALT or AST above 20 times the limit. Cases were mostly in the first two months but some arose up to 18 months in, especially with valproate. Some cases led to hospital stays. After approval, cholestatic or mixed liver injury was reported. The ALT rates were 30% with valproate plus clobazam, 21% with valproate alone, 4% with clobazam alone and 3% with neither (Lennox-Gastaut and Dravet). In healthy adults, a 2021 early safety study (16 people, 1,500 mg a day for about 3.5 weeks) found 44% above the upper limit and 31% above 5 times the limit, starting within 2 to 4 weeks. A 2025 randomized trial in 201 healthy adults (5 mg/kg a day for 28 days) found 5.6% with ALT or AST above 3 times the limit versus 0% on placebo, with 7 meeting withdrawal criteria for possible drug-induced liver injury. The FDA says CBD can cause liver injury and that the risk is harder to manage without medical supervision. ^{50 47 44 16 48 14 41 28}

How often. Rates come from prescription-strength trials and two healthy-volunteer studies at high or moderate amounts. How often liver injury occurs with typical shop products is not known.

Monitoring. The US label says to obtain ALT, AST and total bilirubin before starting the prescription medicine, advises dose adjustment in moderate or severe liver impairment, and says to consider stopping or lowering the dose of CBD or valproate if liver enzymes rise. The EMA page says it must not be used if ALT/AST is over 3 times normal together with bilirubin over twice normal. Anyone with liver disease or who drinks heavily should talk to a doctor first.

p33022751.
p40622698.

This is a safety connection. It is not evidence that CBD (Cannabidiol) supports liver health.

Medicines and supplements 10

With	Why it matters	Importance	Evidence	What to do
Clobazam prescription	CBD raises levels of clobazam's active breakdown product, N-desmethylclobazam (3.4-fold in healthy volunteers); CBD's own active metabolite also rises.	More sleepiness and sedation (46% versus 16% without clobazam), more pneumonia, and more liver enzyme rises. The label says to consider lowering clobazam if its side effects appear.	Pharmacokinetic study High certainty ^{50 30 33 18 15}	Only combine under a prescriber's care. Do not adjust either medicine yourself.
Valproate (valproic acid) prescription	Not a simple blood-level effect: valproate levels were not changed in healthy volunteers, but liver enzyme rises are much more common together. Free valproate rose with higher CBD in an expanded-access program.	ALT over 3 times normal in 21% with valproate alone versus 3% with neither (Lennox-Gastaut and Dravet). In Dravet, all the liver enzyme rises seen at the lower amount were in people taking valproate.	Controlled trial High certainty ^{50 28 30 18}	Needs liver tests and medical supervision.
Warfarin and possibly other blood thinners anticoagulant	Thought to be CYP-enzyme related; not proven. The prescription label sections read do not name warfarin.	Case reports: a 46-year-old needed about a 20% warfarin dose cut; another case had an INR of 6.86 and needed a 30% cut. A 2023 systematic review found INR rose in 6 of 7 cannabis-warfarin case reports, with one bleed, and called the evidence very low quality.	Case report Very low certainty ^{11 36}	Anyone on warfarin should tell their prescriber before using any CBD and have INR checked more often if they start.
Medicines cleared by CYP2C19, CYP2C8, CYP1A2, CYP2B6 or UGT1A9 prescription	Per the label, CBD is a moderate inhibitor of CYP2C19, an inhibitor of UGT1A9, a weak inhibitor of CYP1A2, and an inducer and inhibitor of CYP2B6; it may also cause clinically significant interactions with CYP2C8 substrates.	Levels of medicines such as clobazam's metabolite, theophylline and tizanidine (CYP1A2) may rise; for clopidogrel, which needs CYP2C19 to become active, effect may fall. The label advises considering dose changes.	Authoritative guidance Moderate certainty ^{50 33}	Ask a pharmacist to check every medicine you take.
P-glycoprotein substrates: everolimus, sirolimus, tacrolimus, digoxin prescription	CBD raised exposure to everolimus about 2.5-fold; other P-gp substrates may also rise.	Transplant and heart medicines have narrow safety margins. The label advises blood-level monitoring and dose changes.	Authoritative guidance Moderate certainty ⁵⁰	Transplant and heart patients should not use CBD without their specialist.
Stiripentol prescription	CBD raised stiripentol exposure slightly (AUC 1.6-fold in healthy volunteers).	The label says to monitor for stiripentol side effects.	Pharmacokinetic study Moderate certainty ^{30 50 33}	Only under a prescriber's care.
Strong CYP3A4/CYP2C19 inducers such as rifampin prescription	Rifampin lowered CBD by about 32% and its active metabolite 7-OH-CBD by about 63%.	CBD may work less well; the label says to consider a higher amount of the prescription medicine.	Authoritative guidance Moderate certainty ⁵⁰	A prescriber should decide.
Alcohol and other sedating drugs (sleep aids, anti-anxiety drugs) alcohol	Added sedation. In healthy volunteers alcohol also raised CBD exposure 1.6-fold.	The FDA and the label warn of more sleepiness and injuries.	Authoritative guidance Moderate certainty ^{48 50 12}	Avoid combining unless a clinician says it is fine; do not drive.
Antidepressants, opioid pain medicines, acetaminophen, THC prescription	Reported in a 2021 review of published interaction reports; mechanisms differ.	The review's abstract names these groups but gives no rates or sizes; evidence is mostly reports and lab work.	Narrative review Low certainty ⁵	Mention CBD use to your prescriber or pharmacist.
Food (especially a high-fat meal) and milk food	Absorption rises sharply with food; high-fat meal 3.8-fold AUC, 5.2-fold peak; milk 2.4- and 3.1-fold.	The same product can give very different blood levels depending on what you ate, which affects both effect and side effects.	Pharmacokinetic study High certainty ^{12 39 34}	Be consistent and talk to a prescriber if on the prescription medicine.

Who should check first 10

Situation	Advice	Evidence
Allergy to CBD or to sesame oil (in the prescription solution)	Do not use the prescription solution. The label lists hypersensitivity to CBD or its ingredients, which include sesame oil, as a contraindication; stop and seek care for rash, swelling or trouble breathing.	Authoritative guidance ⁵⁰
Liver disease or high liver enzymes	Talk to a doctor first. The US label requires liver tests and lowers amounts in moderate or severe liver impairment; the EU page says not to use it with enzymes over 3 times normal plus bilirubin over twice normal.	Authoritative guidance ^{50 47}
Pregnancy	Avoid unless a doctor decides the benefit outweighs risk. Animal data suggest possible fetal harm (label); human effects are unknown (FDA). Women taking the prescription medicine in pregnancy are encouraged to join a pregnancy registry.	Authoritative guidance ^{50 48 3 19}
Breastfeeding	Avoid unless advised by a doctor. The label has no data on CBD in human milk or effects on the infant; the FDA lists effects on breastfed newborns as unknown. A medical reference such as LactMed was not read for this page.	Authoritative guidance ^{50 48}
Children and teenagers	Only as a prescription medicine for approved seizure disorders. The FDA lists effects on the developing brain as unknown; juvenile rats given CBD had delayed sexual maturation and behaviour changes in the label's animal data. Keep all CBD products away from young children.	Authoritative guidance ^{50 48 49}
Older adults	Few people over 55 were in the prescription trials, so the label cannot say if they respond differently and advises cautious dosing because liver, kidney and heart function and other medicines vary. Shop-product safety in older adults has not been established in the sources read.	Authoritative guidance ⁵⁰
Taking warfarin, clobazam, valproate, transplant medicines or many other prescriptions	Check with a pharmacist or prescriber before use; see Interactions.	Authoritative guidance ^{50 11 30}
Depression or suicidal thoughts	Monitor and tell a clinician about mood changes; the prescription label carries a class warning for antiseizure medicines.	Authoritative guidance ⁵⁰
Needing to drive, operate machinery or work with heavy equipment	Do not until you know how CBD affects you; the label says so for the prescription medicine.	Authoritative guidance ^{50 1}
Drug testing at work, sport or in custody cases	CBD products can contain THC even when labelled THC-free, and THC drug-test findings could affect employment, military or sports eligibility, say the authors of a 2022 product study.	Product analysis ²⁵

Overdose, product quality and regulators

Summary. At the amounts used in prescription trials CBD is not described as highly toxic, but it commonly causes sleepiness, diarrhea and liver enzyme rises. Animal studies in the label show developmental toxicity in rats and rabbits at high exposures, and the FDA lists male reproductive toxicity in animals as an unresolved question in humans.

Animal. Prescription label animal data: embryo-fetal deaths in rats at the highest dose, lower fetal weights in rabbits, and in rat offspring decreased growth, delayed sexual maturation, behaviour changes and small testes with lower fertility at mid and high doses; juvenile rats had delayed sexual maturation, behaviour changes, higher bone density and liver cell changes. The FDA also describes lower testicular size, inhibited sperm growth and lower testosterone in animals.

Human doses studied. Studies used very different amounts. Prescription trials gave 10 to 50 mg per kg of body weight a day. Healthy-volunteer studies gave 5 mg/kg a day for 28 days (2025 trial), 1,500 mg a day for about 3.5 weeks (2021 study), and single doses up to 6,000 mg (early safety study). Many consumer products are far lower. These are what studies used, not advice.

High dose or long term signals. Liver enzyme rises and sleepiness are dose-related in the label. No long-term (many years) data in healthy adults were found in the sources read. The FDA says daily long-term use and which amounts trigger known risks are unknown.

Overdose response. In the single-ascending-dose study (up to 6,000 mg), adverse events were mild or moderate, with no serious events. The sources read do not describe a poisoning treatment. If someone takes a lot, or a child swallows any CBD product, call a poison center or emergency number.

Genotoxicity. Not read.

CBD products sold outside pharmacies are not checked against the same quality rules as medicines. Studies and the FDA report products with the wrong amount of CBD and with THC.

CBD amount versus label. In 80 hemp-derived CBD products bought in central Kentucky (online and in shops), labels claimed 7.5 to 60 mg/mL and tests found 2.9 to 61.3 mg/mL. 37 of 80 were at least 10% off the label: 12 held under 90% and 25 held over 110%. Only 54% were within 10%. ²⁴

THC contamination. Of 80 products plus Epidiolex as a regulated control, THC was found above the quantification limit in 52 samples (0.008 to 2.071 mg/mL). Five of 21 products labelled THC-free held detectable THC (0.015 to 0.656 mg/mL). ²⁵

Earlier online-product study. A 2017 JAMA study compared the labels of CBD extracts sold online with lab results. The one-sentence abstract gives no figures, so none are quoted here. ⁸

FDA findings. The FDA says it tested the chemical content of some products and many did not contain the CBD level claimed, and it is investigating reports of unsafe levels of contaminants such as pesticides, heavy metals and THC (page dated March 2020). ⁴⁸

Body	Action	Date
US Food and Drug Administration (via the DailyMed label) (United States)	Approved one CBD medicine, Epidiolex (cannabidiol oral solution), for seizures with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex in patients 1 year and older; prescription only. The label says it is not a controlled substance. source	2026-10-25
US Food and Drug Administration (United States)	Says THC and CBD products are excluded from the dietary supplement definition and that adding CBD to food in interstate commerce is prohibited; no regulation has changed this; the agency says it has not approved other CBD products. Hemp with 0.3% delta-9 THC or less (dry weight) is not controlled under federal law, but FDA authority over products is kept. source	2026-10-25
US Food and Drug Administration (United States)	Consumer update: limited safety data point to real risks including liver injury, drug interactions, sedation, and (in animals) male reproductive toxicity; effects on the developing brain, fetus and breastfed newborn are unknown; some products do not contain the CBD claimed. source	2026-10-25
European Medicines Agency (European Union)	Epidyolex (cannabidiol) authorised on 19 September 2019 as add-on to clobazam for Lennox-Gastaut or Dravet syndrome from age 2, and with other treatments for tuberous sclerosis complex from age 2; prescription only; liver monitoring required. source	2026-10-25
World Health Organization, Expert Committee on Drug Dependence (International)	2018 critical review: no effects indicating abuse or dependence potential in humans; generally well tolerated; adverse effects may come from interactions with existing medicines; no evidence of recreational use or public-health problems associated with pure CBD. source	2026-10-25

When to get medical help

Yellow skin or eyes, dark urine, pain in the upper right belly, severe tiredness, or nausea and vomiting that will not go away: these can be signs of liver injury. Stop and call a doctor now.
Rash, swelling of the face, lips or throat, or trouble breathing: possible allergic reaction. Get emergency care.
Thoughts of suicide or self-harm, or new severe mood or behaviour changes: get urgent help.
Very hard to wake, slow or shallow breathing, or confusion, especially after mixing with alcohol or sedating medicines: call emergency services.
Easy bruising, nosebleeds or other bleeding if you take warfarin: call your prescriber today.
A child has swallowed any CBD or cannabis product, or an adult took far more than intended: call a poison center or emergency number.
Do not stop seizure medicine suddenly; talk to the prescriber first.

Stop using the product and contact a clinician or emergency service. In the US, Poison Control is 1-800-222-1222.

General information only, not medical advice. Ask a clinician or pharmacist before use, especially with medicines, in pregnancy or with a health condition. See the Natural Remedy Safety guide.

Frequently asked questions

What is CBD?

CBD (cannabidiol) is a compound made by the cannabis plant, formed from its acid form CBDA when heated or aged. It does not cause a high. It is sold in oils, gummies, vapes and creams, and one purified form is an approved prescription medicine for certain epilepsies.

^{46 52 48}

Is there an approved CBD medicine?

Yes, one. The FDA approved a prescription CBD solution (Epidiolex) for seizures with Lennox-Gastaut syndrome, Dravet syndrome or tuberous sclerosis complex in people 1 year and older. The EU authorised Epidyolex from age 2. The FDA says it has approved no other CBD products.

^{49 50 47}

Does CBD work for epilepsy?

The purified prescription solution does, for those three syndromes, when added to usual medicines. In trials, seizures fell about 17 to 30 percentage points more than with placebo, though few patients became seizure-free. Shop products have not been tested this way.

^{13 40 41 28 7}

Does CBD help with anxiety?

Maybe a little in some settings, but the evidence is mixed. A 24-patient lab speaking test found less anxiety after a single CBD dose than placebo, and a 2024 review of 11 trials found contradictory results. It is not an approved treatment for anxiety.

^{6 10}

Does CBD help me sleep?

Not proven. A 2020 systematic review found insufficient evidence for any sleep disorder. A 2025 trial of a CBD and terpene blend found about 1% more deep and REM sleep, which is small and cannot be credited to CBD alone.

^{37 43}

Does CBD relieve pain?

The evidence is weak. The best controlled trial, in 136 people with hand osteoarthritis or psoriatic arthritis, found no difference from placebo. A 2026 Cochrane review found no clear evidence for CBD-dominant medicines in nerve pain.

^{42 2}

Can CBD harm the liver?

Yes. In trials of the prescription solution, liver enzymes rose above 3 times normal in 12% to 13% of treated patients versus 1% on placebo, and more with valproate. A 2025 trial in healthy adults found 5.6% on CBD versus 0% on placebo after 28 days. Ask a doctor about liver tests if you use CBD regularly.

^{50 16 44}

Can CBD interact with my medicines?

Yes. With clobazam, CBD raised its active breakdown product 3.4-fold and sleepiness was more common. With valproate, liver enzyme rises are more common. Case reports describe higher INR with warfarin. CBD also affects CYP2C19 and other liver enzymes, so blood levels of many drugs can change. Ask a pharmacist.

^{50 30 11 36}

Will CBD make me sleepy or affect driving?

It can cause sleepiness (32% versus 11% on placebo in the main trials, more with clobazam), and alcohol adds to it. In one on-road test, vaporized CBD-dominant cannabis did not increase lane weaving, but the authors say that does not rule out impairment. The label advises not driving until you know how it affects you.

^{50 1}

Do CBD products contain what the label says?

Often not. In a test of 80 hemp CBD oils, 37 were at least 10% off their label, and THC was found in 52 of 81 samples, including 5 of 21 labelled THC-free. The FDA says many products it tested did not contain the CBD claimed.

^{24 25 48}

Is CBD safe in pregnancy or while breastfeeding?

Not shown to be safe. The prescription label says animal data suggest it may cause fetal harm and that there are no data on human milk. The FDA says effects on a developing fetus or breastfed newborn are unknown. Talk to a doctor.

^{50 48 3}

Is CBD addictive, and does it turn into THC?

WHO's 2018 review found no signs of abuse or dependence potential in humans, and in a trial CBD was placebo-like in frequent cannabis users. WHO also says CBD can be changed to THC in the lab under acid conditions, but this likely does not happen after oral CBD in people. Some shop products do contain THC, though.

^{46 4 25}

Explore by health goal

Joint Support Anxiety Support Calm Relaxation Better Sleep Sleep Quality Digestive Support Gut Health Blood Pressure Support Heart Health Blood Sugar Balance Metabolic Health Brain Health

Editorial transparency

Written by: AlternativeMed editorial team. Research compiled with AI assistance, then checked against PubMed records.
Medical review: None. This page has not been reviewed by a physician or pharmacist.
First published: 2026-10-25
Last substantive update: 2026-10-25
How we research: We searched PubMed and read the abstracts of the studies cited, and read the FDA consumer update and questions-and-answers page, the US prescribing information for the CBD medicine, the EMA overview and the WHO 2018 critical review in full. Compound identity was checked in PubChem. Lab numbers come from a table of published and state testing data and are not our own testing.
How we rate evidence: See "How we rate evidence" in the Evidence tab.
Independence: No product is promoted on this page. No affiliate links are shown, and none will be added for CBD. This page is information, not a recommendation to use any product.

Sources 52

Each source was opened and checked. Where we read only the abstract, it says so. The small numbers in the text point here.

Arkell TR, Vinckenbosch F, Kevin RC, et al. Effect of Cannabidiol and Δ9-Tetrahydrocannabinol on Driving Performance: A Randomized Clinical Trial.. JAMA. 2020. PMID 33258890. DOI 10.1001/jama.2015.6613. (abstract)
Ateş G, Welsch P, Klose P, et al. Cannabis-based medicines for chronic neuropathic pain in adults.. Cochrane Database Syst Rev. 2026. PMID 41548880. DOI 10.1002/14651858.CD012182. (abstract)
Azubuike C, Deng A, Goodin A. Nationwide Prevalence of Cannabidiol Use in Pregnancy and in Women of Reproductive Age.. Obstet Gynecol. 2026. PMID 41343836. DOI 10.1097/AOG.0000000000005603. (abstract)
Babalonis S, Haney M, Malcolm RJ, et al. Oral cannabidiol does not produce a signal for abuse liability in frequent marijuana smokers.. Drug Alcohol Depend. 2017. PMID 28088032. DOI 10.1016/j.drugalcdep.2016.11.030. (abstract)
Balachandran P, Elsohly M, Hill KP. Cannabidiol Interactions with Medications, Illicit Substances, and Alcohol: a Comprehensive Review.. J Gen Intern Med. 2021. PMID 33515191. DOI 10.1007/s11606-020-06504-8. (abstract)
Bergamaschi MM, Queiroz RH, Chagas MH, et al. Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naïve social phobia patients.. Neuropsychopharmacology. 2011. PMID 21307846. DOI 10.1038/npp.2011.6. (abstract)
Bilbao A, Spanagel R. Medical cannabinoids: a pharmacology-based systematic review and meta-analysis for all relevant medical indications.. BMC Med. 2022. PMID 35982439. DOI 10.1177/1352458505070618. (abstract)
Bonn-Miller MO, Loflin MJE, Thomas BF, et al. Labeling Accuracy of Cannabidiol Extracts Sold Online.. JAMA. 2017. PMID 29114823. DOI 10.1001/jama.2017.11909. (abstract)
Chesney E, Oliver D, Green A, et al. Adverse effects of cannabidiol: a systematic review and meta-analysis of randomized clinical trials.. Neuropsychopharmacology. 2020. PMID 32268347. DOI 10.1016/j.forsciint.2018.10.019. (abstract)
Coelho CF, Vieira RP, Araújo-Junior OS, et al. The Impact of Cannabidiol Treatment on Anxiety Disorders: A Systematic Review of Randomized Controlled Clinical Trials.. Life (Basel). 2024. PMID 39598172. DOI 10.1176/appi.ajp.2013.13020178. (abstract)
Cortopassi J. Warfarin dose adjustment required after cannabidiol initiation and titration.. Am J Health Syst Pharm. 2020. PMID 33016308. DOI 10.1093/ajhp/zxaa268. (abstract)
Crockett J, Critchley D, Tayo B, et al. A phase I, randomized, pharmacokinetic trial of the effect of different meal compositions, whole milk, and alcohol on cannabidiol exposure and safety in healthy subjects.. Epilepsia. 2020. PMID 32012251. DOI 10.1111/epi.16419. (abstract)
Devinsky O, Cross JH, Laux L, et al. Trial of Cannabidiol for Drug-Resistant Seizures in the Dravet Syndrome.. N Engl J Med. 2017. PMID 28538134. DOI 10.1056/NEJMoa1611618. (abstract)
Devinsky O, Patel AD, Cross JH, et al. Effect of Cannabidiol on Drop Seizures in the Lennox-Gastaut Syndrome.. N Engl J Med. 2018. PMID 29768152. DOI 10.1056/NEJMoa1714631. (abstract)
Devinsky O, Thiele EA, Wright S, et al. Cannabidiol efficacy independent of clobazam: Meta-analysis of four randomized controlled trials.. Acta Neurol Scand. 2020. PMID 32592183. DOI 10.1111/epi.16355. (abstract)
Florian J, Salcedo P, Burkhart K, et al. Cannabidiol and Liver Enzyme Level Elevations in Healthy Adults: A Randomized Clinical Trial.. JAMA Intern Med. 2025. PMID 40622698. DOI 10.3928/00989134-20210610-02. (abstract)
Freeman TP, Hindocha C, Baio G, et al. Cannabidiol for the treatment of cannabis use disorder: a phase 2a, double-blind, placebo-controlled, randomised, adaptive Bayesian trial.. Lancet Psychiatry. 2020. PMID 32735782. DOI 10.1017/S003329171800226X. (abstract)
Gaston TE, Bebin EM, Cutter GR, et al. Final analysis of potential drug-drug interactions between highly purified cannabidiol and anti-seizure medications in an open-label expanded access program.. Epilepsia Open. 2023. PMID 37593907. DOI 10.1002/epi4.12815. (abstract)
Gomez Wulschner LE, Chang VN, Oh WC, et al. Oral Consumption of Cannabidiol During Pregnancy Alters Behavior in Mouse Offspring.. Cannabis Cannabinoid Res. 2026. PMID 41840986. DOI 10.1177/25785125251413416. (abstract)
Hurd YL, Spriggs S, Alishayev J, et al. Cannabidiol for the Reduction of Cue-Induced Craving and Anxiety in Drug-Abstinent Individuals With Heroin Use Disorder: A Double-Blind Randomized Placebo-Controlled Trial.. Am J Psychiatry. 2019. PMID 31109198. DOI 10.1176/appi.ajp.2019.18101191. (abstract)
Ibeas Bih C, Chen T, Nunn AV, et al. Molecular Targets of Cannabidiol in Neurological Disorders.. Neurotherapeutics. 2015. PMID 26264914. DOI 10.1017/S1461145709990848. (abstract)
Jadoon KA, Ratcliffe SH, Barrett DA, et al. Efficacy and Safety of Cannabidiol and Tetrahydrocannabivarin on Glycemic and Lipid Parameters in Patients With Type 2 Diabetes: A Randomized, Double-Blind, Placebo-Controlled, Parallel Group Pilot Study.. Diabetes Care. 2016. PMID 27573936. DOI 10.2337/dc16-0650. (abstract)
Jadoon KA, Tan GD, O'Sullivan SE. A single dose of cannabidiol reduces blood pressure in healthy volunteers in a randomized crossover study.. JCI Insight. 2017. PMID 28614793. DOI 10.1038/sj.jhh.1001639. (abstract)
Johnson E, Kilgore M, Babalonis S. Label accuracy of unregulated cannabidiol (CBD) products: measured concentration vs. label claim.. J Cannabis Res. 2022. PMID 35658956. DOI 10.1089/can.2020.0166. (abstract)
Johnson E, Kilgore M, Babalonis S. Cannabidiol (CBD) product contamination: Quantitative analysis of Δ9-tetrahydrocannabinol (Δ9-THC) concentrations found in commercially available CBD products.. Drug Alcohol Depend. 2022. PMID 35690015. DOI 10.1089/can.2020.0166. (abstract)
Laprairie RB, Bagher AM, Kelly ME, et al. Cannabidiol is a negative allosteric modulator of the cannabinoid CB1 receptor.. Br J Pharmacol. 2015. PMID 26218440. DOI 10.1111/bph.13250. (abstract)
McGuire P, Robson P, Cubala WJ, et al. Cannabidiol (CBD) as an Adjunctive Therapy in Schizophrenia: A Multicenter Randomized Controlled Trial.. Am J Psychiatry. 2018. PMID 29241357. DOI 10.1176/appi.ajp.2017.17030325. (abstract)
Miller I, Scheffer IE, Gunning B, et al. Dose-Ranging Effect of Adjunctive Oral Cannabidiol vs Placebo on Convulsive Seizure Frequency in Dravet Syndrome: A Randomized Clinical Trial.. JAMA Neurol. 2020. PMID 32119035. DOI 10.1111/j.1528-1167.2010.02915.x. (abstract)
Mitarnun W, Kanjanarangsichai A, Junlaor P, et al. Cannabidiol and cognitive functions/inflammatory markers in Parkinson's disease: A double-blind randomized controlled trial at Buriram Hospital (CBD-PD-BRH trial).. Parkinsonism Relat Disord. 2025. PMID 40267585. DOI 10.1016/j.parkreldis.2025.107841. (abstract)
Morrison G, Crockett J, Blakey G, et al. A Phase I, Open-Label, Pharmacokinetic Trial to Investigate Possible Drug-Drug Interactions Between Clobazam, Stiripentol, or Valproate and Cannabidiol in Healthy Subjects.. Clin Pharmacol Drug Dev. 2019. PMID 30791225. DOI 10.1007/s40263-018-0578-5. (abstract)
Naftali T, Mechulam R, Marii A, et al. Low-Dose Cannabidiol Is Safe but Not Effective in the Treatment for Crohn's Disease, a Randomized Controlled Trial.. Dig Dis Sci. 2017. PMID 28349233. DOI 10.1007/s10620-017-4540-z. (abstract)
Narayan AJ, Hayley AC, Rose S, et al. The effect of nightly use of 150 mg cannabidiol on daytime neurocognitive performance in primary insomnia: a randomized controlled pilot trial.. Psychopharmacology (Berl). 2025. PMID 39153080. DOI 10.1155/2019/8961748. (abstract)
Patsalos PN, Szaflarski JP, Gidal B, et al. Clinical implications of trials investigating drug-drug interactions between cannabidiol and enzyme inducers or inhibitors or common antiseizure drugs.. Epilepsia. 2020. PMID 32918835. DOI 10.1002/jcph.1634. (abstract)
Perucca E, Bialer M. Critical Aspects Affecting Cannabidiol Oral Bioavailability and Metabolic Elimination, and Related Clinical Implications.. CNS Drugs. 2020. PMID 32504461. DOI 10.1007/s40263-020-00741-5. (abstract)
Shannon S, Lewis N, Lee H, et al. Cannabidiol in Anxiety and Sleep: A Large Case Series.. Perm J. 2019. PMID 30624194. DOI 10.1080/02791072.2017.1354409. (abstract)
Smythe MA, Wu W, Garwood CL. Anticoagulant drug-drug interactions with cannabinoids: A systematic review.. Pharmacotherapy. 2023. PMID 37740600. DOI 10.1002/phar.2881. (abstract)
Suraev AS, Marshall NS, Vandrey R, et al. Cannabinoid therapies in the management of sleep disorders: A systematic review of preclinical and clinical studies.. Sleep Med Rev. 2020. PMID 32603954. DOI 10.1016/j.smrv.2020.101339. (abstract)
Taura F, Sirikantaramas S, Shoyama Y, et al. Cannabidiolic-acid synthase, the chemotype-determining enzyme in the fiber-type Cannabis sativa.. FEBS Lett. 2007. PMID 17544411. DOI 10.1016/j.febslet.2007.05.043. (abstract)
Taylor L, Gidal B, Blakey G, et al. A Phase I, Randomized, Double-Blind, Placebo-Controlled, Single Ascending Dose, Multiple Dose, and Food Effect Trial of the Safety, Tolerability and Pharmacokinetics of Highly Purified Cannabidiol in Healthy Subjects.. CNS Drugs. 2018. PMID 30374683. DOI 10.3390/molecules200814451. (abstract)
Thiele EA, Marsh ED, French JA, et al. Cannabidiol in patients with seizures associated with Lennox-Gastaut syndrome (GWPCARE4): a randomised, double-blind, placebo-controlled phase 3 trial.. Lancet. 2018. PMID 29395273. DOI 10.1016/S0140-6736(18)30136-3. (abstract)
Thiele EA, Bebin EM, Bhathal H, et al. Add-on Cannabidiol Treatment for Drug-Resistant Seizures in Tuberous Sclerosis Complex: A Placebo-Controlled Randomized Clinical Trial.. JAMA Neurol. 2021. PMID 33346789. DOI 10.3390/molecules24091694. (abstract)
Vela J, Dreyer L, Petersen KK, et al. Cannabidiol treatment in hand osteoarthritis and psoriatic arthritis: a randomized, double-blind, placebo-controlled trial.. Pain. 2022. PMID 34510141. DOI 10.7812/TPP/18-041. (abstract)
Wang M, Faust M, Abbott S, et al. Effects of a cannabidiol/terpene formulation on sleep in individuals with insomnia: a double-blind, placebo-controlled, randomized, crossover study.. J Clin Sleep Med. 2025. PMID 39167421. DOI 10.5664/jcsm.11324. (abstract)
Watkins PB, Church RJ, Li J, et al. Cannabidiol and Abnormal Liver Chemistries in Healthy Adults: Results of a Phase I Clinical Trial.. Clin Pharmacol Ther. 2021. PMID 33022751. DOI 10.1002/cpt.2071. (abstract)
Xu DH, Cullen BD, Tang M, et al. The Effectiveness of Topical Cannabidiol Oil in Symptomatic Relief of Peripheral Neuropathy of the Lower Extremities.. Curr Pharm Biotechnol. 2020. PMID 31793418. DOI 10.2174/1389201020666191202111534. (abstract)
n/a. WHO Expert Committee on Drug Dependence, 40th meeting: Cannabidiol (CBD) Critical Review Report (June 2018). WHO. 2018. (full-text)
n/a. EMA: Epidyolex (cannabidiol): EPAR overview (marketing authorisation 19 Sep 2019). EMA. 2019. (full-text)
n/a. FDA: What You Need to Know (And What We're Working to Find Out) About Products Containing Cannabis or Cannabis-derived Compounds, Including CBD (content current as of 5 Mar 2020). FDA. 2020. (full-text)
n/a. FDA: FDA Regulation of Cannabis and Cannabis-Derived Products, Including Cannabidiol (CBD): Questions and Answers (content current as of 16 Jul 2024). FDA. 2024. (full-text)
n/a. Epidiolex (cannabidiol) oral solution: US prescribing information (DailyMed label, Jazz Pharmaceuticals; label version published 15 Jun 2026). DailyMed. 2026. (full-text)
n/a. PubChem: related compounds, retrieved through the REST API on 2026-10-25 (CBDA CID 160570, CBGA CID 6449999, delta-9-THC CID 16078, 7-hydroxycannabidiol CID 11301963, 7-carboxycannabidiol CID 146592489). PubChem. 2026. (api-record)
n/a. PubChem: Cannabidiol, compound CID 644019 (REST API record). PubChem. 2026. (api-record)

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.