Cannabis & Cannabinoids · Cannabinoids · The main intoxicating compound in cannabis, also made as prescription medicines

THC (Delta-9-Tetrahydrocannabinol)

Q: What is THC?

THC (delta-9-tetrahydrocannabinol) is the compound in cannabis that causes the "high". It acts on CB1 and CB2 cannabinoid receptors, mostly CB1. The plant makes it from an acid form called THCA.

Q: Is THC approved as a medicine?

Not the plant. FDA has approved prescription dronabinol (synthetic THC) and nabilone (a THC-like drug) for chemotherapy nausea, and dronabinol for appetite loss in AIDS. In the UK a THC and CBD mouth spray is licensed for MS stiffness. These are prescription medicines and different from shop cannabis.

Q: Does THC help nausea or muscle stiffness?

For chemotherapy nausea, prescription cannabinoids beat placebo in mostly old trials and may help as an add-on. For MS stiffness, people rate it better on the THC and CBD spray, but a clinician-measured trial found no difference.

Q: Can THC cause psychosis or mental health problems?

Heavy use is linked to psychosis (about 3.9 times the odds for the heaviest users) and daily high-potency use about 4.8 times in one study. THC can also cause anxiety and panic. These are links, not proof of cause. Medicine labels advise against use in people with a history of psychosis.

Q: Is THC addictive?

About 22% of people who use cannabis meet criteria for a cannabis use disorder, and 33% of young people who use weekly or daily. Withdrawal can occur. No medicine is approved for cannabis use disorder.

Q: What is cannabis hyperemesis syndrome?

Repeated attacks of severe nausea, vomiting and belly pain after long-term heavy cannabis use. Emergency visits for it rose from 2016 to 2022. Stopping cannabis is the treatment that works best, so see a clinician.

Q: Is it safe to drive after using THC?

No. Cannabis impairs lane control, and cannabis with alcohol is worse than either alone. A blood THC level does not reliably show how impaired a person is.

Q: Is THC safe in pregnancy or when breastfeeding?

Use in pregnancy is linked to low birth weight, preterm birth and NICU admission. THC passes into breast milk. The dronabinol label says to avoid it in pregnancy and the UK spray label says not to use it while breastfeeding.

Q: What about teens and older adults?

In teens, use is linked to depression, suicidal behaviour and a small IQ drop with frequent use. Older adults are more sensitive to dizziness and thinking effects and may have drug interactions.

(6aR,10aR)-6,6,9-trimethyl-3-pentyl-6a,7,8,10a-tetrahydrobenzo[c]chromen-1-ol (delta-9-tetrahydrocannabinol; C21H30O2) · also called THC, delta-9-THC, Δ9-THC; dronabinol (the drug name)

THC is the compound in cannabis that causes the "high". It acts on the CB1 and CB2 cannabinoid receptors. Two forms are prescription medicines: dronabinol (synthetic THC) and nabilone (a THC-like drug). In the UK a THC and CBD mouth spray is licensed for MS stiffness. FDA has not approved the cannabis plant for any disease. Research shows small, uncertain benefits for some uses and clear risks: psychosis and mental health problems, dependence (about 22% of people who use cannabis), driving impairment, harm in pregnancy and poisoning of young children by edibles.

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 THC (Delta-9-Tetrahydrocannabinol)

THC (delta-9-tetrahydrocannabinol) is the main intoxicating compound in cannabis. Prescription forms exist (dronabinol, nabilone, and in some countries a THC and CBD spray), but FDA has not approved the cannabis plant for any use. The research shows small, uncertain benefits for pain, MS stiffness and chemotherapy nausea and clear risks, especially psychosis, dependence, driving impairment, harm in pregnancy and child poisoning.

What it is

THC is a cannabinoid, a compound made by the cannabis plant. The plant makes it first as an acid (THCA), using an enzyme called THCA synthase. PubChem lists it as C21H30O2 with a molecular weight of 314.5, and the drug name dronabinol refers to the same compound.

THC partly activates CB1 and CB2 receptors. Its mind-altering effect comes mainly through CB1; in a randomised trial a drug that blocks CB1 cut the "feeling high" rating by up to 82.8% in healthy volunteers.

THC was isolated in the 1960s and 1970s. It is broken down in the liver, mainly by the enzyme CYP2C9, into an active product called 11-hydroxy-THC. Effects last about 2 to 3 hours when inhaled and about 4 to 12 hours when swallowed.

^{43 30 57 18 28 34 19}

Prescription THC medicines versus cannabis products

FDA says it has not approved a marketing application for cannabis for any disease. It has approved prescription drugs: dronabinol (Marinol, Syndros) and nabilone (Cesamet), plus Epidiolex, which is CBD. Dronabinol is used for loss of appetite with weight loss in AIDS and for chemotherapy nausea and vomiting that has not responded to usual treatment. Nabilone is approved for chemotherapy nausea and vomiting.

A THC and CBD mouth spray called nabiximols (Sativex) is licensed in the UK for adults with moderate to severe MS spasticity who have not responded to other treatment. The label says it must not be used by people with a history of psychotic illness or while breastfeeding.

These medicines come with a set strength, a label and medical supervision. That is not true of smoked, vaped or edible cannabis from shops. Results for one product do not carry over to another.

^{12 31 50 38 8}

What the research shows

Pain: reviews of short trials (1 to 6 months) find a small average pain reduction with some THC products, with more dizziness and sleepiness. The Cochrane review of nerve pain found no clear evidence of benefit for THC-dominant medicines.

Nausea: cannabinoids beat placebo for chemotherapy nausea in mostly old trials but were no better than older anti-nausea drugs. Multiple sclerosis stiffness: patients rate it better with the THC and CBD spray, but a recent clinician-measured trial found no difference.

Other uses: a 2026 Cochrane review found little to no sleep benefit; a 2026 meta-analysis found no benefit for anxiety, PTSD or psychosis; nine dementia trials found no benefit. A major cancer guideline recommends against use as a cancer treatment outside trials.

^{8 2 9 13 6 51 55 47 7}

Teens, pregnancy, older adults and children

Teens and young adults: use before age 18 is linked to depression and suicidal behaviour in young adulthood and to a small drop in IQ with frequent use. Pregnancy: linked to low birth weight, preterm birth and NICU admission; THC passes into breast milk.

Older adults: more sensitive to dizziness, blood pressure drops and thinking effects, with possible interactions. Children: in a study of 138 children and teens with cannabis intoxication in emergency departments, 38% needed intensive care; every child aged 10 or under had eaten edibles.

^{16 35 25 21 50 10}

What the lab numbers show

Our data hold 465 measured THC values for 338 named cannabis strains. They come from state regulator datasets (Maryland and Nevada flower) and from published papers. They are not our own tests.

The biggest set is Nevada flower: the "Total THC" column has medians for 65 strains, from 5.57 to 29.82 percent by weight (the middle of the 65 medians is 20.9). These are medians of 3 to 953 samples each. The "Delta-9 THC" column in the same datasets is a different measurement and holds far smaller numbers: Nevada medians run from 0.20 to 1.59 and Maryland medians from 0.0 to 1.53 (144 strain entries), in percent by weight as reported without a per-column unit label. Do not compare across columns.

Papers add other units and scopes. Greenhouse-grown plants: 3.03 to 21.53 percent of flower dry weight (30 strain means). Single dispensary samples: 7.7 to 27.1 percent (12 samples). A few batches report 0.2 to 25.8 percent with no stated basis. One field-plot paper reports total THC of 0.117 to 0.6 percent of dry matter (50 measurements). One paper's table of single commercial samples lists values from 0.01 to 0.3 "as printed"; it is not clear what these measure, so they are not compared here. Percent by weight, percent of dry weight and percent of dry matter are not the same, so these ranges should not be compared with each other.

The same strain name varies a lot between samples. For example Blue Dream is 19.44 percent total THC in the Nevada data (363 samples) but 10.99 percent of dry weight in a greenhouse study; White Widow is 24.96 percent in Nevada (127 samples), 15.63 in a greenhouse study and 18.7 in one reported batch. Strain names are unreliable labels, and these are real results from specific samples.

Among the 28 strains that have profile pages here, the highest Nevada total THC medians are Chemdawg (28.375 percent, 18 samples), White Widow (24.96, 127 samples) and Ice Cream Cake (24.225, 626 samples); the lowest are Harlequin (5.57, only 3 samples), Girl Scout Cookies (16.55, 19 samples) and Trainwreck (16.96, 30 samples).

These numbers measure how much THC is in the flower. They say nothing about how a product will feel or what it will do to you, and a higher number does not make a product safer or better. NIDA notes that higher THC strength is linked to a greater chance of developing a cannabis use disorder.

^{15 32}

Legal and regulatory status

This is not legal advice and rules change. As the sources say: FDA has approved only the prescription drugs named above (checked 2026-10-25, FDA page current as of February 2023). A US federal rule effective 28 April 2026 placed FDA-approved marijuana-containing drug products, and marijuana subject to state medical licences, in schedule III. A hearing on moving marijuana generally to schedule III was noticed for 29 June 2026; we found no final decision on 2026-10-25.

NCCIH reports that FDA has determined products containing THC or CBD cannot be sold legally as dietary supplements. In the UK, nabiximols is a licensed prescription medicine. Laws in other countries and in US states were not checked here.

^{12 48 49 31 38}

At a glance

Traditional uses history, not proof

THC was isolated and its structure worked out in the 1960s and 1970s (Mechoulam 2014). Traditional use of the whole cannabis plant is not covered by the sources we read.
Dronabinol and nabilone were mostly tested for chemotherapy nausea in the 1980s and 1990s (NCCIH).
FDA approved dronabinol in 1992 for loss of appetite and weight in AIDS (NCCIH).
A THC and CBD mouth spray (Sativex) was first authorised in the UK in June 2010 for MS spasticity (UK label).

Studied for research

Chronic pain, mostly nerve pain: small average benefit for some products, low certainty
Chemotherapy nausea and vomiting: approved medicines; old trials; may help as an add-on, low certainty
Multiple sclerosis stiffness: patients rate it better, clinician measures did not change, mixed
HIV/AIDS appetite and weight loss: some weight gain, low certainty; no benefit in cancer
Sleep and insomnia: little or no clear effect, very low
Anxiety, PTSD, psychosis, depression: no benefit shown
Dementia agitation: no benefit, more sleepiness
Cancer treatment: guideline recommends against outside trials

Main compounds

Delta-9-THC itself (C21H30O2)
THCA, the acid form the plant makes first
11-hydroxy-THC, the active product made in the body
CBD, a different cannabinoid that does not cause a high

Common forms

Dronabinol capsules or liquid (prescription medicine)
Nabilone capsules (prescription medicine)
THC and CBD mouth spray, nabiximols (licensed in some countries, such as the UK)
Cannabis flower, concentrates and edibles (not approved medicines; strength varies and labels are often off)

Research links: FDA: Cannabis research and drug approval process · NIDA: Cannabis (marijuana) · NCCIH: Cannabis and cannabinoids · DailyMed: dronabinol label · UK label: Sativex (emc) · Cochrane: cannabis-based medicines for neuropathic pain (abstract) · Research Library · Genes & Biomarkers Library

What it is and its history

Name: Delta-9-tetrahydrocannabinol (delta-9-THC, Δ9-THC); PubChem IUPAC name (6aR,10aR)-6,6,9-trimethyl-3-pentyl-6a,7,8,10a-tetrahydrobenzo[c]chromen-1-ol
Other names: Tetrahydrocannabinol (PubChem title); Dronabinol (the generic name of the drug form); Delta-9-THC; Δ9-THC; THC
Common names: THC (English); Dronabinol (drug name; brands Marinol and Syndros) (English); Delta-9 (English slang for the compound)
Chemical class: Cannabinoid: a phytocannabinoid, made of an alkylresorcinol part and a monoterpene part (Sirikantaramas 2007). Not a plant itself.
Related or similar compounds: Name: Delta-8-THC; Note: A different form of THC with a slightly different structure (PubChem CID 638026). NIDA says delta-8 THC is intoxicating, has not been approved by FDA for safe use, and that eating large amounts has led to medical emergencies.; Name: THCA (delta-9-THC acid); Note: The acid form made in the plant (CID 98523). The acid form the plant makes first. The sources read here do not describe how it is converted to THC.; Name: CBD (cannabidiol); Note: A different cannabinoid (CID 644019) that a JAMA 2026 review says does not have psychoactive effects. Products labelled CBD can contain THC (NCCIH).

Traditional use and history history, not clinical proof

Tradition or era	Documented use	Source
1960s and 1970s	Isolation and structure of most of the major cannabinoids, including delta-9-THC, "the principal psychoactive molecule in Cannabis sativa", was achieved.	p25315390
1980s to early 1990s	Two cannabinoid receptors were identified, followed shortly by the body's own cannabinoid-like molecules (endocannabinoids).	p25315390
1980s and 1990s	Most of the trials of dronabinol and nabilone for chemotherapy nausea and vomiting were done in this era.	s-nccih-cannabis
1992	FDA approved dronabinol for loss of appetite and weight loss in people with HIV/AIDS, mainly based on one study of 139 people.	s-nccih-cannabis
2010	The first UK authorisation of the THC and CBD mouth spray nabiximols (Sativex), 16 June 2010, for MS spasticity.	s-emc-sativex
28 April 2026	US federal rule placed FDA-approved marijuana-containing drug products, and marijuana covered by state medical licences, in schedule III of the Controlled Substances Act.	s-fr-2026-08176
Traditional and ancient use of cannabis	Not covered in the sources we read. THC itself was only isolated in the 1960s, so traditional use was of the whole plant. Any older history claims have NOT been verified here.	none read

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
Delta-9-THC (THC, dronabinol) PubChem 16078	phytocannabinoid	Cannabis plant; also made synthetically as the drug dronabinol Varies; see the lab-numbers section of the overview	Partial agonist at CB1 and CB2 receptors; the main source of the mind-altering effects. ^{57 19 30}	Review
THCA (delta-9-tetrahydrocannabinolic acid) PubChem 98523	phytocannabinoid acid	Made in the plant by THCA synthase from CBGA Not checked	The acid precursor of THC in the plant. THCA and CBGA induced cell death in lab tests and may play a role in plant defence. ⁴³	Laboratory research
CBGA (cannabigerolic acid) PubChem 6449999	phytocannabinoid acid	Plant; the starting material for THCA, CBDA and CBCA Not checked	The substrate of THCA synthase in the plant. ⁴³	Laboratory research
11-hydroxy-delta-9-THC (11-OH-THC) PubChem 644022	THC metabolite	Made in the body from THC Roughly equal to THC in blood after swallowed dronabinol (label)	The main active metabolite; psychoactive. Formed mainly by CYP2C9. ^{34 50}	Human data
THC-COOH (11-nor-9-carboxy-THC) PubChem 108207	THC metabolite	Made in the body from THC Not checked	An acid metabolite of THC (named in Grotenhermen 2003). ¹⁹	Review
Delta-8-THC PubChem 638026	phytocannabinoid isomer	Minor in the plant; sold as hemp-derived products Not checked	A THC-like intoxicating compound; not FDA approved for safe use (NIDA). ^{32 5}	Authoritative guidance
Nabilone PubChem 5284592	synthetic THC-like drug	Made synthetically; not from the plant n/a	Prescription drug for chemotherapy nausea; in pain trials nabilone reduced pain but dronabinol did not (Annals 2026). ^{12 8}	Authoritative guidance
Cannabidiol (CBD) PubChem 644019	phytocannabinoid	Cannabis plant Not checked	A different cannabinoid that does not cause a high; it may change some effects of THC when taken together (toxicology review). Covered on its own page. ⁵⁷	Review

Mechanisms, receptors and pathways

Target or pathway	Gene or receptor	What was found	Level
CB1 receptor	cannabinoid receptor type 1 (CB1)	THC is a partial agonist at CB1 and CB2. Its mind-altering (psychotomimetic) effect comes mainly through CB1, which is plentiful in the central nervous system except the brain stem. In a randomised trial in healthy volunteers, a drug that blocks CB1 cut the "feeling high" rating by up to 82.8% (95% CI -91.0 to -67.2%). ^{57 18}	Controlled trial
CB2 receptor	cannabinoid receptor type 2 (CB2)	THC is a partial agonist at CB2 too. None of the cannabinoid receptor ligands is entirely specific to CB1 or CB2. ⁵⁷	Review
Other receptors and channels	TRPV1, GPR55, GPR119	Cannabinoids can also interact with these non-CB1/CB2 targets. What this means for THC in people is not shown in the sources read. ⁵⁷	Lab
Endocannabinoid system	CB1, CB2 and the body's own cannabinoid-like molecules	The receptors were identified in the 1980s and early 1990s and the endocannabinoids shortly after. The system is described as a route to treat brain disorders. ²⁸	Review
Dependence and withdrawal	CB1 signalling	Central action at CB1 and the signalling that follows is described as a main component of cannabis dependence and withdrawal; much supporting work is in animals. ²³	Animal research
Drug-metabolising enzymes	CYP2C9, CYP2C19, CYP2D6, CYP3A4, UGT1A9, UGT2B7	CYP2C9 is the main enzyme that clears THC and forms 11-OH-THC in liver tests. THC and other cannabinoids also inhibit several CYP enzymes in test-tube studies. ^{34 36}	Lab

Absorption, metabolism and elimination

Absorption: Inhaled: THC reaches its highest blood level within minutes; effects begin in seconds to a few minutes, peak after 15 to 30 minutes and taper off within 2 to 3 hours. Swallowed: effects start 30 to 90 minutes later, peak after 2 to 3 hours and last about 4 to 12 hours (Grotenhermen 2003). For the dronabinol capsule, 90 to 95% is absorbed but only 10 to 20% reaches the bloodstream because of first-pass liver metabolism; blood levels peak about 0.5 to 4 hours after swallowing, and a high-fat meal delayed the peak by about 4 hours and raised total exposure 2.9-fold (label).
Distribution: Very fat-soluble: apparent volume of distribution about 10 L/kg and about 97% bound to blood proteins for dronabinol and its metabolites (label). THC crosses into breast milk: in 20 breastfeeding mothers who used cannabis, milk THC peaked about 120 minutes after one use and built up with repeated use (Holdsworth 2024).
Metabolism: Extensive first-pass liver metabolism. CYP2C9 is the main enzyme that clears THC and forms the active metabolite 11-OH-THC; CYP2D6, CYP2C19 and CYP3A4 contribute smaller amounts. THC and 11-OH-THC are present in about equal amounts in blood after swallowed dronabinol (label). The label says CYP2C9 genetic variants may lower clearance and raise levels.
Elimination: Two-compartment pattern: an early half-life of about 4 hours and a terminal half-life of 25 to 36 hours for dronabinol (label). About half of a dose is recovered in faeces within 72 hours and 10 to 15% in urine; low levels of metabolites were detected in urine and faeces for more than 5 weeks after a single dose. Blood THC level also did not reliably track driving impairment in a systematic review.
Bioavailability: About 10 to 20% of a swallowed dronabinol dose reaches the circulation (label). Inhaled bioavailability was not given in the sources read.

^{19 50 34 21 4}

Isolated compounds versus whole extracts

Pure THC medicines (dronabinol, nabilone), THC and CBD sprays (nabiximols) and the cannabis plant are different products. Reviews sort products by their THC-to-CBD ratio and source because results differ: in one pain review, nabilone lowered pain but dronabinol did not. Results for one product cannot be assumed for another. Smoked or edible cannabis from a shop has been studied far less than the prescription products, and its THC content varies and often does not match the label.

Nutritional composition

THC is not a nutrient and has no nutritional role. NCCIH reports that FDA has determined that products containing THC or CBD cannot be sold legally as dietary supplements.

14benefits and uses looked at

57studies and sources checked

Moderate certaintybest evidence rating found

9lab markers tracked in trials

1 moderate certainty2 mixed findings4 low certainty1 very low certainty3 not enough evidence3 evidence against

What was looked at

The 14 benefits and uses checked for THC (Delta-9-Tetrahydrocannabinol), each with how strong the evidence is. Tap one to open it.

Chronic pain (mostly nerve pain)Low certainty
Nausea and vomiting from chemotherapyLow certainty
Muscle stiffness (spasticity) in multiple sclerosisMixed findings
Appetite and weight loss (HIV/AIDS and cancer)Low certainty
Felt effects in volunteers: feeling high, relaxed or anxiousModerate certainty
Sleep and insomniaVery low certainty
Anxiety, PTSD, psychosis and depressionNot enough evidence
Treating cannabis withdrawal and cannabis use disorderMixed findings
Tourette syndrome ticsLow certainty
Agitation and thinking problems in dementiaEvidence against
Glaucoma (eye pressure)Evidence against
Treating cancer itselfEvidence against
Reducing the need for opioid pain medicinesNot enough evidence
Crohn's disease, ulcerative colitis and irritable bowel syndromeNot enough evidence

Lab markers tracked in the studies

Pain severity scoreSpasticity rating (patient 0 to 10 scale)Complete response to chemotherapy (no vomiting, no rescue medicine)Body weight in HIV/AIDS"Feeling high" ratingTotal sleep timeTic severity scoreCannabis withdrawal symptoms scoreAgitation score (CMAI)

What is THC (Delta-9-Tetrahydrocannabinol) good for?

The uses people ask about and how strong the human evidence is for each. Most results are for prescription products and short trials, not for cannabis from a shop. None of this is proof that THC treats a disease.

Chronic pain (mostly nerve pain)Low certaintyReviews of short trials find a small average pain reduction with some THC products and more dizziness and sleepiness. The Cochrane review of nerve pain found no clear evidence of benefit. Nausea and vomiting from chemotherapyLow certaintyPrescription dronabinol and nabilone beat placebo in old trials and were no better than older anti-nausea drugs. A guideline says cannabinoids may help refractory nausea as an add-on. Muscle stiffness (spasticity) in multiple sclerosisMixed findingsPeople with MS rate stiffness better on the THC and CBD spray (moderate certainty in Cochrane); a 2024 trial using a clinician measure found no difference. Appetite and weight loss (HIV/AIDS and cancer)Low certaintySome weight gain in HIV/AIDS in older trials; no benefit in cancer; dronabinol is approved for AIDS-related appetite loss. Felt effects in volunteers: feeling high, relaxed or anxiousModerate certaintyIn volunteers who rarely use cannabis, THC causes feeling high, anxiety and tiredness and lowers alertness. Sleep and insomniaVery low certaintyA 2026 Cochrane review found little to no clear effect on sleep quality or time, with more side effects. Anxiety, PTSD, psychosis and depressionNot enough evidenceNo benefit shown for anxiety, PTSD or psychosis; no randomised trials for depression. THC can bring on anxiety. Treating cannabis withdrawal and cannabis use disorderMixed findingsA THC and CBD combination eased withdrawal in small trials, but THC preparations did not make people more likely to stop. Tourette syndrome ticsLow certaintyVery small trials suggest fewer tics; low certainty. Agitation and thinking problems in dementiaEvidence againstNine trials found no improvement and more sleepiness. Glaucoma (eye pressure)Evidence againstOld studies showed a brief drop in eye pressure but not as good as standard treatments. Treating cancer itselfEvidence againstA major guideline recommends against cannabinoids as a cancer treatment outside trials. Reducing the need for opioid pain medicinesNot enough evidenceControlled studies have not shown that cannabis reduces opioid use. Crohn's disease, ulcerative colitis and irritable bowel syndromeNot enough evidenceSmall trials in Crohn's disease and ulcerative colitis showed no remission benefit.

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.

Chronic pain (mostly nerve pain)

Low certainty

Reviews of short trials find that some cannabis medicines that contain THC may lower pain a little, with more dizziness and sleepiness. Some reviews found no clear benefit for nerve pain. No one can say how they work over years.

6 studiesStudied as prescription or research products: synthetic THC (nabilone…

Studies, doses and limits

What the trials found

Several reviews cover trials in people. A 2026 Annals of Internal Medicine review of 25 short trials (2,303 people, about two thirds with nerve pain) found that oral THC-only products may slightly reduce pain and a mouth spray with similar THC and CBD amounts probably slightly reduces pain. A 2026 Cochrane review of 21 trials in nerve pain (2,187 people) found no clear evidence of benefit for THC-dominant medicines on 50% pain relief, with very low certainty. A 2022 review reached similar conclusions. An older 2015 JAMA review found moderate-quality evidence for pain, although only 4 of its 79 trials were at low risk of bias.

Results

Annals 2026: pooled pain differences were -0.78 points (oral THC-only) and -0.54 points (comparable THC and CBD mouth spray) on a 0 to 10 scale; nabilone lowered pain by -1.59 points but dronabinol by only -0.23 points. Cochrane 2026: 50% pain relief with THC-dominant medicines RD 0.14 (95% CI -0.07 to 0.37; 7 studies, 534 people), very low certainty; they may increase nervous system side effects (RD 0.25, 0.14 to 0.37; low certainty). Whiting 2015: average pain score WMD -0.46 on a 0 to 10 scale (95% CI -0.80 to -0.11; 6 trials). A 2023 BMJ umbrella review rated 30% pain reduction as high certainty (odds 0.59, 0.37 to 0.93) across conditions, with more psychological distress.

Effect sizes as reported

Small average pain reduction (about half a point to just over one point on a 10-point scale) for some products. It is not clear that this is large enough to matter to patients (NCCIH).

Why this rating

Trials are short (1 to 6 months), products differ a lot, and the Cochrane review rated the evidence very low to low certainty. Different reviews give different answers. Long-term results are unknown.

Form and dose studied

Studied as prescription or research products: synthetic THC (nabilone, dronabinol), oral THC extracts, and a THC and CBD mouth spray (nabiximols). Smoked or vaped cannabis was studied much less. Shop products were not tested in these trials.

Who was studied

Adults, mostly with nerve pain (diabetic, HIV, multiple sclerosis, injury).

Limits

Short trials; products differ a lot between trials; trial products are not the same as shop products; no long-term safety data.

Why researchers looked

Chronic pain is the most common reason people say they use medical cannabis. Standard pain medicines help only some people, so researchers have tested cannabis medicines that contain THC, alone or with CBD.

Traditional use

Cannabis has been used for pain in many cultures. History is not proof that it works.

Proposed mechanism Pharmacological theory

THC partly activates CB1 and CB2 cannabinoid receptors. CB1 is plentiful in the brain and nervous system. This is a plausible route to changing how pain is felt, but the human trials only measure results, not the route.

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

Dizziness, sleepiness and nausea were the most common effects in trials. See the Safety tab for mental health, driving and dependence.

^{8 2 27 54 44 31}

Nausea and vomiting from chemotherapy

Low certainty

Two THC-type prescription medicines are approved in the United States for chemotherapy nausea when other medicines have not worked. Most of the trials are decades old, and newer anti-nausea regimens are now standard. THC may help as an add-on in hard cases.

8 studiesPrescription capsules or liquid of synthetic THC (dronabinol) and nab…

Studies, doses and limits

What the trials found

A 2025 meta-analysis of 26 trials (23 published before 2000) found cannabinoids beat placebo but were no better than older active nausea drugs. A 2024 ASCO guideline says cannabinoids may improve refractory chemotherapy nausea when added to standard regimens. The FDA label for dronabinol says it is for patients who have not responded adequately to conventional antinausea treatment. NCCIH reports a 2015 review of 23 studies (1,326 people) found dronabinol or nabilone more helpful than placebo and similar to other nausea medicines, with more dizziness and sleepiness.

Results

Chow 2025: cannabinoids vs placebo for overall nausea and vomiting control, RR 2.65 (95% CI 1.70 to 4.12); no difference vs active comparators. Whiting 2015: complete nausea and vomiting response 47% vs 20% (OR 3.82, 95% CI 1.55 to 9.42; 3 trials). Hsu 2026 (JAMA review): small reduction in nausea and vomiting vs placebo or active drugs (SMD -0.29, 95% CI -0.39 to -0.18). A palliative-care review of cancer patients found no significant difference from placebo for nausea (SMD 0.21, -0.10 to 0.52).

Effect sizes as reported

Better than placebo in older trials, about the same as older nausea drugs, unclear benefit on top of modern anti-nausea regimens.

Why this rating

Most trials are 25 to 40 years old, nearly half had concerns about bias, and chemotherapy and anti-nausea care have changed. A newer THC and CBD trial was positive but small in scope.

Form and dose studied

Prescription capsules or liquid of synthetic THC (dronabinol) and nabilone, a THC-like drug, and in one recent trial a THC and CBD mouth spray. These are prescription medicines. They are not the same as smoked or edible cannabis from a shop.

Who was studied

Adults having chemotherapy.

Limits

Old trials; changing treatments; side effects such as dizziness and sleepiness can be hard for sick people.

Why researchers looked

Chemotherapy often causes severe nausea. Research on the prescription drugs dronabinol and nabilone for this was done mostly in the 1980s and 1990s (NCCIH).

Traditional use

Cannabis has a long history of use for nausea. That history is not proof of effect.

Proposed mechanism Pharmacological theory

The sources we read do not explain how THC lowers nausea.

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

Approved medicines are used under medical supervision. Dronabinol labels also warn that THC itself can cause nausea, vomiting and belly pain similar to cannabis hyperemesis syndrome.

^{9 7 54 22 29 50 31 12}

Muscle stiffness (spasticity) in multiple sclerosis

Mixed findings

People with multiple sclerosis who took a THC and CBD mouth spray rated their stiffness better than people who took placebo. When doctors measured muscle tone, a large recent trial saw no difference. In the UK the spray is licensed for people whose stiffness has not responded to other medicines.

5 studiesStudied mostly as the licensed mouth spray nabiximols (a plant-extrac…

Studies, doses and limits

What the trials found

A 2022 Cochrane review of 25 trials (3,763 people) found nabiximols probably increases the number of people who report an important reduction in spasticity (moderate certainty). A 2026 review of 27 trials found a modest benefit in patient-reported spasticity (low certainty) with limited objective benefit. A 2024 randomised crossover trial of 68 people measured muscle tone by a clinician and found no difference from placebo. The UK product label licenses the spray as add-on treatment for adults with moderate to severe MS spasticity who have not responded to other treatments and who show benefit in an initial trial.

Results

Cochrane 2022: nabiximols OR 2.51 (95% CI 1.56 to 4.04; 5 trials, 1,143 people; moderate certainty) for an important reduction in spasticity; about 216 more people per 1,000 reported benefit. Review 2026: THC:CBD extracts reduced patient-reported spasticity by MD -0.82 (95% CI -1.24 to -0.40; 7 studies), certainty low. Trial 2024: change in muscle tone -0.23 with nabiximols vs -0.26 with placebo (difference 0.04, P = 0.7152).

Effect sizes as reported

About one in five more people report a worthwhile improvement in how stiff they feel than with placebo. No measurable change in clinician-rated muscle tone in the 2024 trial.

Why this rating

Patient-rated improvement has moderate certainty in the Cochrane review, but a 2026 review rates certainty low, and the one recent trial that used a clinician measure found no effect.

Form and dose studied

Studied mostly as the licensed mouth spray nabiximols (a plant-extract THC and CBD medicine), and also as synthetic THC, oral extract and smoked cannabis. Licensed medicine; not a consumer product.

Who was studied

Adults with MS who still have spasticity despite other medicines.

Limits

Short trials (3 to 48 weeks); most results are self-rated; drop-outs from side effects.

Why researchers looked

Muscle stiffness and spasms are common and hard to treat in multiple sclerosis (MS). The THC and CBD spray nabiximols has been tested in many trials.

Traditional use

Not a traditional use in the sources we read.

Proposed mechanism Pharmacological theory

Cannabinoid receptors are found in the nervous system. How THC could ease spasticity has not been proven in humans from these sources.

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

The UK label lists dizziness, fainting, falls, psychiatric effects, and a history of schizophrenia or other psychosis as a reason not to use it.

^{13 20 6 38 31}

Appetite and weight loss (HIV/AIDS and cancer)

Low certainty

Dronabinol, synthetic THC, is approved in the US for loss of appetite and weight in AIDS. Reviews find some weight gain in HIV and no clear benefit in cancer. A prescription drug called megestrol worked better.

5 studiesDronabinol capsules (prescription). Smoked cannabis was tested in a f…

Studies, doses and limits

What the trials found

A palliative-care review of nine trials (1,561 people) found weight gain and better appetite in HIV with cannabinoids but not in cancer. A 2022 review of cancer cachexia found no benefit for appetite or weight and a worse quality of life. NCCIH says the 1992 US approval of dronabinol for AIDS appetite loss rested mainly on one study of 139 people and that other studies were short and small.

Results

HIV: weight gain SMD 0.57 (95% CI 0.22 to 0.92) and appetite SMD 0.57 (0.11 to 1.03) vs placebo (Mücke 2018). Cancer: appetite SMD 0.81 (-1.14 to 2.75), not significant (Mücke 2018); appetite SMD -0.02 (-0.51 to 0.46) and quality of life SMD -0.25 (-0.43 to -0.07) in the 2022 cachexia review. In head-to-head trials megestrol beat cannabinoids for appetite and weight.

Effect sizes as reported

Moderate gain in weight in HIV (low certainty), none in cancer.

Why this rating

The authors rate the evidence low or very low. The HIV trials are old and small.

Form and dose studied

Dronabinol capsules (prescription). Smoked cannabis was tested in a few small trials.

Who was studied

Adults with AIDS or cancer who are losing weight.

Limits

Small old trials; the authors found no convincing high-quality evidence.

Why researchers looked

THC makes many people hungrier. It was tested for people who lose weight because of AIDS or cancer.

Traditional use

Increased appetite is a widely known effect of cannabis.

Proposed mechanism Pharmacological theory

The sources we read do not explain how THC changes appetite. THC acts mainly through CB1 receptors in the nervous system (toxicology review).

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

In HIV there were more mental health symptoms with cannabinoids (RD 0.05, 0.00 to 0.11).

^{29 42 22 50 31}

Felt effects in volunteers: feeling high, relaxed or anxious

Moderate certainty

In controlled experiments, THC reliably makes people who rarely use cannabis feel high, tired and more anxious, and less alert. This is the effect people seek, and also the reason for most of the risks.

4 studiesInhaled and swallowed THC in lab studies.

Studies, doses and limits

What the trials found

A 2026 meta-analysis of 41 randomised placebo-controlled experiments in 1,020 healthy adults who rarely use cannabis found THC caused greater feelings of being high, anxiety and tiredness, and lower alertness and contentedness than placebo. Higher inhaled doses raised the "high" and lowered calmness. The effect of swallowed THC was not related to dose in the analysis.

Results

THC vs placebo: more high, anxiety and tiredness (all p < 0.001); lower alertness and contentedness (both p < 0.01); no difference on calmness (p = 0.414). Inhaled THC dose increased feeling high (b = 1.06, p = 0.0024).

Effect sizes as reported

Large, reliable change in how high people feel in volunteers; anxiety also rises on average.

Why this rating

Many randomised, placebo-controlled experiments with consistent direction (the review did not report a formal GRADE rating; this rating is the compiler's reading). Results are for rare users in lab settings.

Form and dose studied

Inhaled and swallowed THC in lab studies.

Who was studied

Healthy adults with infrequent cannabis use.

Limits

Volunteers are not patients; short sessions; frequent users may respond differently.

Why researchers looked

To understand how THC works as a drug and how strong the effects are.

Traditional use

Cannabis has been used for centuries for its mind-altering effects.

Proposed mechanism Controlled trial

THC is a partial agonist at CB1 and CB2 receptors, and its mind-altering effects are produced mainly through CB1 (toxicology review). In a randomised trial, a drug that blocks CB1 cut the "feeling high" rating by up to 82.8% (95% CI -91.0 to -67.2%).

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

Anxiety and panic are the most common bad reactions. See the Safety tab.

^{17 18 57 19}

Sleep and insomnia

Very low certainty

A 2026 Cochrane review found medical cannabis may have little to no clear effect on sleep quality or total sleep time in people with insomnia, with more side effects.

4 studiesPlant-based and synthetic cannabinoid medicines and pharmacy-prepared…

Studies, doses and limits

What the trials found

The Cochrane review pooled 9 randomised trials (537 adults with insomnia). A Lancet Psychiatry 2026 review of mental and substance use disorders found that any cannabinoid increased sleep time measured by an electronic device and by sleep diary in people with insomnia, with low certainty. NCCIH says it is uncertain whether cannabis improves sleep directly or because symptoms of another illness improve.

Results

Cochrane 2026 (medical cannabis vs placebo): sleep quality SMD -0.49 (95% CI -1.06 to 0.08; very low certainty); total sleep time +7.7 minutes (-23.0 to 38.4; very low certainty); daytime functioning SMD -0.34 (-0.83 to 0.16; very low); adverse events OR 11.25 (4.46 to 28.35; low certainty). Lancet Psychiatry 2026: sleep time by device SMD 0.54 (0.14 to 0.95) and by diary 0.55 (0.01 to 1.09).

Effect sizes as reported

No clear, meaningful sleep gain in the Cochrane pooled analysis; small gains in another review.

Why this rating

The Cochrane reviewers rated the key sleep outcomes very low certainty because of risk of bias, inconsistency and imprecision. The JAMA 2026 review says trials do not support use for insomnia.

Form and dose studied

Plant-based and synthetic cannabinoid medicines and pharmacy-prepared cannabis products; mostly short trials.

Who was studied

Adults aged 18 to 76 with insomnia (72.7% women).

Limits

Few small trials; very different products.

Why researchers looked

Many people use cannabis to help them sleep, and THC makes many people tired.

Traditional use

Not covered in the sources read.

Proposed mechanism Pharmacological theory

The sources we read do not test how THC might affect sleep.

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

More adverse events than placebo. THC can cause sleepiness, which is not the same as better sleep.

^{51 55 22 31}

Anxiety, PTSD, psychosis and depression

Not enough evidence

Cannabinoids are not shown to treat anxiety, PTSD or psychosis. No randomised trial evidence was found for depression. THC can bring on anxiety, and heavy use is linked to a higher risk of psychosis.

4 studiesVarious cannabinoids; THC alone, CBD alone and combinations were pool…

Studies, doses and limits

What the trials found

A 2026 Lancet Psychiatry meta-analysis of 54 randomised trials (2,477 people) looked at cannabinoids as the main treatment for mental and substance use disorders. It found no significant effects for anxiety, anorexia nervosa, psychotic disorders, PTSD or opioid use disorder, and no randomised evidence for depression. It rated most evidence low certainty and said routine use is rarely justified. NCCIH notes only one very small study (10 people) in which nabilone cut PTSD nightmares, and no clear answer from observational studies.

Results

Lancet Psychiatry 2026: 44% of trials had high risk of bias; all-cause adverse events OR 1.75 (95% CI 1.25 to 2.46; number needed to harm 7); cocaine craving increased (0.69, 0.22 to 1.15). Nabilone PTSD nightmares: one study of 10 people (NCCIH).

Effect sizes as reported

No benefit shown; some harms signals.

Why this rating

The best review found no significant effect for these conditions and mostly low certainty, with high risk of bias in 44% of trials.

Form and dose studied

Various cannabinoids; THC alone, CBD alone and combinations were pooled by condition.

Who was studied

Adults and some young people with mental or substance use disorders.

Limits

Many small trials; products and doses varied.

Why researchers looked

NIDA notes that older adults mainly use cannabis to self-treat conditions that include anxiety and depression, so trials have tested cannabinoids for mental health conditions.

Traditional use

Not covered in the sources read.

Proposed mechanism Pharmacological theory

THC acts on CB1 receptors, which are plentiful in the central nervous system (toxicology review).

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

See the Safety tab: THC can cause anxiety and panic and is linked to psychosis risk.

^{55 44 31 32}

Treating cannabis withdrawal and cannabis use disorder

Mixed findings

A THC and CBD combination reduced withdrawal symptoms and weekly cannabis use in small trials, but a Cochrane review found THC preparations did not make people more likely to stop. No medicine is approved for cannabis use disorder.

4 studiesPrescription-type THC or THC and CBD products given in a trial settin…

Studies, doses and limits

What the trials found

The 2026 Lancet Psychiatry review found a CBD and delta-9-THC combination reduced withdrawal symptoms and weekly grams of cannabis use in people with cannabis use disorder. The 2025 Cochrane review of 37 randomised trials (3,201 people) found abstinence was no more likely with THC preparations than placebo (moderate certainty), and called all tested medicines experimental.

Results

Lancet Psychiatry 2026: withdrawal symptoms SMD -0.29 (95% CI -0.57 to -0.02); weekly grams of cannabis -1.00 (-1.69 to -0.30). Cochrane 2025: abstinence with THC preparations RR 1.04 (0.71 to 1.52; 4 studies, 290 people; moderate certainty); adverse events RR 1.05 (0.88 to 1.26).

Effect sizes as reported

Small reductions in withdrawal symptoms and use; no more people stopped.

Why this rating

Two reviews point in different directions: a low-certainty benefit on withdrawal and use, but moderate-certainty evidence of no effect on abstinence.

Form and dose studied

Prescription-type THC or THC and CBD products given in a trial setting. Not the same as using cannabis to cut down.

Who was studied

Adults, mostly young men, in treatment trials.

Limits

Few trials per medicine; short follow-up.

Why researchers looked

People who use cannabis daily can have withdrawal when they stop. Researchers have tried giving a controlled amount of a cannabis-based medicine to ease this.

Traditional use

Not covered in the sources read.

Proposed mechanism Pharmacological theory

A review says action on the CB1 receptor and the signalling that follows it is a main part of cannabis dependence and withdrawal; much of this work is in animals.

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

Using a THC product to treat THC dependence is only a research approach; talk to a clinician.

^{55 45 32 23}

Tourette syndrome tics

Low certainty

Two small older trials and a pooled 2026 analysis suggest THC or THC with CBD may reduce tic severity in Tourette syndrome. The evidence is small and low certainty.

2 studiesSynthetic THC capsules or THC with CBD combinations.

Studies, doses and limits

What the trials found

The Lancet Psychiatry 2026 meta-analysis found a CBD and delta-9-THC combination reduced tic severity versus placebo. NCCIH describes a 2015 review of 2 small placebo-controlled studies (36 people) that suggested synthetic THC capsules may improve tic severity.

Results

Lancet Psychiatry 2026: tic severity SMD -0.68 (95% CI -1.03 to -0.34). NCCIH: 2 studies, 36 people.

Effect sizes as reported

A moderate-sized drop in tic scores in very small groups.

Why this rating

Only a few small trials; the review rated most evidence low certainty.

Form and dose studied

Synthetic THC capsules or THC with CBD combinations.

Who was studied

People with Tourette syndrome or tic disorders.

Limits

Very few participants.

Why researchers looked

Tics are movements or sounds a person cannot easily control. Some small trials tested THC capsules.

Traditional use

Not covered in the sources read.

Proposed mechanism Pharmacological theory

The sources we read do not explain the mechanism.

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

Dizziness, sleepiness and psychiatric effects of THC apply.

^{55 31}

Agitation and thinking problems in dementia

Evidence against

A 2026 meta-analysis of 9 trials found cannabinoids did not improve agitation, overall symptoms or thinking test scores in dementia and caused more sleepiness.

2 studiesVarious cannabinoid medicines in small trials.

Studies, doses and limits

What the trials found

The review included 9 randomised trials (334 people) of cannabinoid-based treatments in Alzheimer disease or dementia. It found no improvement in agitation, neuropsychiatric symptoms or thinking, with moderate certainty for behaviour outcomes.

Results

Agitation SMD -0.58 (95% CI -1.71 to 0.55); neuropsychiatric symptoms SMD -0.02 (-1.00 to 0.96); thinking (MMSE) SMD 0.86 (-16.33 to 18.06). Sleepiness RR 2.03 (1.29 to 3.20).

Effect sizes as reported

No average improvement; wide confidence ranges.

Why this rating

The reviewers report moderate-certainty evidence that behaviour did not improve and more sleepiness occurred.

Form and dose studied

Various cannabinoid medicines in small trials.

Who was studied

People with Alzheimer disease or dementia, mostly in care settings.

Limits

Small total sample; trials differed a lot.

Why researchers looked

Agitation is a hard-to-treat problem in dementia and families look for options.

Traditional use

Not covered in the sources read.

Proposed mechanism Pharmacological theory

The sources we read do not test a mechanism.

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

Older adults can be more sensitive to the nervous system and blood pressure effects of THC (dronabinol label).

^{47 50}

Glaucoma (eye pressure)

Evidence against

THC and cannabis lowered eye pressure briefly in old studies but not as well as current treatments. NCCIH says cannabis is not helpful for glaucoma.

1 studiesNot applicable.

Studies, doses and limits

What the trials found

NCCIH says studies in the 1970s and 1980s showed cannabis or substances derived from it could lower eye pressure, but not as effectively as existing treatments, and only for a short time. NCCIH states that cannabis is not helpful for glaucoma.

Results

No numbers were available in the source read.

Effect sizes as reported

No practical benefit compared with standard treatment.

Why this rating

The only source read is NCCIH (last updated November 2019). Older studies, no new trials read.

Who was studied

People with glaucoma.

Limits

We did not read the original old studies.

Why researchers looked

Raised eye pressure can damage the optic nerve; cannabis was tested in the 1970s and 1980s.

Traditional use

Not covered in the sources read.

Proposed mechanism Pharmacological theory

The sources we read do not explain the mechanism.

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

Untreated glaucoma can cause lasting vision loss; do not delay standard care.

³¹

Treating cancer itself

Evidence against

A major cancer guideline recommends against using cannabis or cannabinoids to treat cancer outside clinical trials.

1 studiesNot applicable.

Studies, doses and limits

What the trials found

The 2024 ASCO guideline reviewed 13 systematic reviews and 5 primary studies. It says clinicians should recommend against using cannabis or cannabinoids as a cancer-directed treatment unless in a clinical trial. It also says use "has outpaced the science".

Results

No effect sizes in the abstract; evidence rated low or very low for most outcomes.

Effect sizes as reported

No evidence that THC treats cancer in people.

Why this rating

A guideline panel recommends against use as a cancer treatment outside trials.

Who was studied

Adults with cancer.

Limits

Guideline abstract only.

Why researchers looked

People with cancer sometimes ask whether cannabis can treat the cancer itself.

Traditional use

Not covered in the sources read.

Proposed mechanism Preclinical only

Lab findings were not reviewed here.

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

Do not replace or delay cancer treatment with cannabis.

⁷

Reducing the need for opioid pain medicines

Not enough evidence

Studies have not shown that cannabis or THC lets people use fewer opioids. Findings are inconsistent.

2 studiesCannabinoids given with opioids in trials.

Studies, doses and limits

What the trials found

NCCIH reports a 2017 review of 9 studies (750 people) in which none of the 3 high-quality randomised studies showed cannabinoids lowered opioid use. Data from states with medical marijuana laws were inconsistent. A 2018 clinic study of 2,736 people over 65 reported 18.1% stopped or reduced opioids, but it had no control group.

Results

NCCIH: 9 studies, 750 people; 3 high-quality studies showed no reduction. Clinic cohort: 18.1% reduced or stopped opioids (uncontrolled).

Effect sizes as reported

Inconsistent.

Why this rating

Controlled studies show no benefit; uncontrolled clinic data show a possible one.

Form and dose studied

Cannabinoids given with opioids in trials.

Who was studied

Adults with pain.

Limits

NCCIH page last updated November 2019.

Why researchers looked

Animal studies suggested THC might let pain be controlled with smaller opioid doses.

Traditional use

Not covered in the sources read.

Proposed mechanism Animal research

Animal studies suggest THC with opioids may allow smaller opioid doses (NCCIH).

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

Combining THC with opioids adds sedation and impairment risk (see Safety).

^{31 1}

Crohn's disease, ulcerative colitis and irritable bowel syndrome

Not enough evidence

Small trials of cannabis or cannabis oil did not show remission in Crohn's disease. Little research exists for ulcerative colitis or irritable bowel syndrome.

2 studiesSmoked cannabis, cannabis oil, CBD capsules.

Studies, doses and limits

What the trials found

NCCIH summarises 2018 Cochrane reviews: 3 studies (93 people) of smoked cannabis or cannabis oil in active Crohn's disease showed no difference in clinical remission versus placebo; 2 studies (92 people) in ulcerative colitis showed no remission difference with CBD capsules. For irritable bowel syndrome NCCIH says it is unknown whether cannabinoids help. A 2023 BMJ umbrella review rated improved quality of life in inflammatory bowel disease as high certainty (odds 0.34, 0.22 to 0.53).

Results

NCCIH: Crohn's 3 studies, 93 people; ulcerative colitis 2 studies, 92 people. BMJ umbrella: quality of life OR 0.34 (0.22 to 0.53).

Effect sizes as reported

No change in remission; possible quality-of-life gain in one review.

Why this rating

Few small studies, mostly of smoked cannabis or CBD, not THC alone.

Form and dose studied

Smoked cannabis, cannabis oil, CBD capsules.

Who was studied

Adults with Crohn's disease or ulcerative colitis.

Limits

Tiny studies; mixed products; sources partly from 2018 and 2019.

Why researchers looked

People with digestive illness ask whether cannabis helps their symptoms.

Traditional use

Not covered in the sources read.

Proposed mechanism Pharmacological theory

The sources we read do not test a mechanism.

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

Cannabis hyperemesis syndrome can cause vomiting and belly pain in heavy users (see Safety).

^{31 44}

Connections to the rest of health

How THC (Delta-9-Tetrahydrocannabinol) 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

B. Body systems

Connected to	Kind of connection	Evidence status	Context	Studies
Digestive system	Adverse effect or risk	Low certainty	Nausea, vomiting and belly pain can be a side effect, even of the synthetic medicine. Preparation or compound: THC (including dronabinol)	⁵⁰
Nervous System	Biological association	Low certainty	Studied for: Chronic pain (mostly nerve pain) Studied for: Muscle stiffness (spasticity) in multiple sclerosis	^{8 2 27 54 44 31 13 20 6 38 17 18 57 19 51 55 22 32 45 23 47 50 1}
Digestive System	Biological association	Low certainty	Studied for: Nausea and vomiting from chemotherapy Studied for: Crohn's disease, ulcerative colitis and irritable bowel syndrome	^{9 7 54 22 29 50 31 12 44}
Musculoskeletal System	Biological association	Mixed findings	Studied for: Muscle stiffness (spasticity) in multiple sclerosis	^{13 20 6 38 31}
Metabolism Weight	Biological association	Low certainty	Studied for: Appetite and weight loss (HIV/AIDS and cancer)	^{29 42 22 50 31}
Stress Mood	Biological association	Moderate certainty	Studied for: Felt effects in volunteers: feeling high, relaxed or anxious Studied for: Anxiety, PTSD, psychosis and depression	^{17 18 57 19 55 44 31 32 45 23}
Sleep Recovery	Biological association	Very low certainty	Studied for: Sleep and insomnia	^{51 55 22 31}
Immune System	Biological association	Evidence against	Studied for: Treating cancer itself	⁷

C. Body parts and organs

Connected to	Kind of connection	Evidence status	Context	Studies
Brain	Adverse effect or risk	Moderate certainty	Feeling high, anxiety, slower thinking; linked to psychosis, dependence and lower IQ with heavy youth use. Preparation or compound: THC, mainly inhaled or edible cannabis	^{17 26 35}
Lungs	Adverse effect or risk	Low certainty	Cough, phlegm and wheeze. Preparation or compound: Smoked cannabis	¹⁴
Heart	Adverse effect or risk	Low certainty	Fast heart rate and blood pressure changes; ever-use linked to combined cardiovascular events (RR 1.48). Preparation or compound: THC	^{39 50}
Liver	Biological association	Low certainty	LiverTox: not linked to clinically apparent liver injury; older reports of mild test abnormalities and scarring in hepatitis C. Preparation or compound: Marijuana	³³

E. Health conditions

Connected to	Kind of connection	Evidence status	Context	Studies
Psychosis	Adverse effect or risk	Moderate certainty	Odds about 3.9 times higher with the heaviest use; 4.8 times for daily high-potency use. Preparation or compound: Heavy or daily, high-potency cannabis	^{26 11}
Cannabis use disorder	Adverse effect or risk	Moderate certainty	About 22% of people who use cannabis; 33% of young weekly or daily users. Preparation or compound: Cannabis with THC	²⁴
Cannabis hyperemesis syndrome	Adverse effect or risk	Low certainty	Repeated vomiting; emergency visits rose 2016 to 2022. Preparation or compound: Long-term heavy cannabis use	^{40 46 52}
Pregnancy	Contraindication	Moderate certainty	Lower birth weight, preterm birth; avoid per dronabinol label. Preparation or compound: Cannabis and THC medicines	^{25 50}

I. Biomarkers

Connected to	Kind of connection	Evidence status	Context	Studies
Pain severity score	Biological association	Low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	^{8 54 31}
Spasticity rating (patient 0 to 10 scale)	Biological association	Mixed findings	Measured in trials: lowered. A lab change is not the same as a health benefit.	^{20 6}
Complete response to chemotherapy (no vomiting, no rescue medicine)	Biological association	Low certainty	Measured in trials: raised. A lab change is not the same as a health benefit.	^{54 9}
Body weight in HIV/AIDS	Biological association	Low certainty	Measured in trials: raised. A lab change is not the same as a health benefit.	²⁹
"Feeling high" rating	Biological association	Moderate certainty	Measured in trials: raised. A lab change is not the same as a health benefit.	^{17 18}
Total sleep time	Biological association	Very low certainty	Measured in trials: mixed. A lab change is not the same as a health benefit.	^{51 55}
Tic severity score	Biological association	Low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	⁵⁵
Cannabis withdrawal symptoms score	Biological association	Mixed findings	Measured in trials: lowered. A lab change is not the same as a health benefit.	^{55 45}
Agitation score (CMAI)	Biological association	Evidence against	Measured in trials: not changed. A lab change is not the same as a health benefit.	⁴⁷

J. Genes, receptors and enzymes

Connected to	Kind of connection	Evidence status	Context	Studies
CB1 receptor (gene CNR1)	Preliminary mechanism	Rct	Main target of THC's mind-altering effects. The gene symbol is standard naming and was not stated in the sources read.	^{57 18}
CB2 receptor (gene CNR2)	Preliminary mechanism	Review	THC is a partial agonist here too. The gene symbol was not stated in the sources read.	⁵⁷
CYP2C9	Preliminary mechanism	Lab	Main enzyme clearing THC; genetic variants may lower clearance (dronabinol label).	^{34 50}
CYP2C19, CYP2D6, CYP3A4, CYP1A2	Preliminary mechanism	Lab	THC and other cannabinoids inhibit some of these in test-tube work; smoking cannabis may induce CYP1A2.	^{36 34}

L. Active compounds

Connected to	Kind of connection	Evidence status	Context	Studies
Delta-9-THC (THC, dronabinol)	Biological association	Review	Partial agonist at CB1 and CB2 receptors; the main source of the mind-altering effects. Preparation or compound: Cannabis plant; also made synthetically as the drug dronabinol	^{57 19 30}
THCA (delta-9-tetrahydrocannabinolic acid)	Biological association	In vitro	The acid precursor of THC in the plant. THCA and CBGA induced cell death in lab tests and may play a role in plant defence. Preparation or compound: Made in the plant by THCA synthase from CBGA	⁴³
CBGA (cannabigerolic acid)	Biological association	In vitro	The substrate of THCA synthase in the plant. Preparation or compound: Plant; the starting material for THCA, CBDA and CBCA	⁴³
11-hydroxy-delta-9-THC (11-OH-THC)	Biological association	Human	The main active metabolite; psychoactive. Formed mainly by CYP2C9. Preparation or compound: Made in the body from THC	^{34 50}
THC-COOH (11-nor-9-carboxy-THC)	Biological association	Review	An acid metabolite of THC (named in Grotenhermen 2003). Preparation or compound: Made in the body from THC	¹⁹
Delta-8-THC	Biological association	Authoritative guidance	A THC-like intoxicating compound; not FDA approved for safe use (NIDA). Preparation or compound: Minor in the plant; sold as hemp-derived products	^{32 5}
Nabilone	Biological association	Authoritative guidance	Prescription drug for chemotherapy nausea; in pain trials nabilone reduced pain but dronabinol did not (Annals 2026). Preparation or compound: Made synthetically; not from the plant	^{12 8}
Cannabidiol (CBD)	Biological association	Review	A different cannabinoid that does not cause a high; it may change some effects of THC when taken together (toxicology review). Covered on its own page. Preparation or compound: Cannabis plant	⁵⁷

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 8 lab markers

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

Other431

The evidence

The research centre for THC (Delta-9-Tetrahydrocannabinol). Every record here was checked against PubMed or the publisher's page; we never list a study we have not verified.

Evidence by outcome

Chronic pain (mostly nerve pain) 6 studies Low certainty Nausea and vomiting from chemotherapy 8 studies Low certainty Muscle stiffness (spasticity) in multiple sclerosis 5 studies · 68 people in trials Mixed findings Appetite and weight loss (HIV/AIDS and cancer) 5 studies Low certainty Felt effects in volunteers: feeling high, relaxed or anxious 4 studies · 85 people in trials Moderate certainty Sleep and insomnia 4 studies Very low certainty Anxiety, PTSD, psychosis and depression 4 studies Not enough evidence Treating cannabis withdrawal and cannabis use disorder 4 studies Mixed findings Tourette syndrome tics 2 studies Low certainty Agitation and thinking problems in dementia 2 studies Evidence against Glaucoma (eye pressure) 1 study Evidence against Treating cancer itself 1 study Evidence against Reducing the need for opioid pain medicines 2 studies · 2,736 people in trials Not enough evidence Crohn's disease, ulcerative colitis and irritable bowel syndrome 2 studies Not enough evidence

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 abstracts of the 49 papers cited, read the DailyMed dronabinol label, NCCIH, NIDA, FDA, the UK Sativex label and two Federal Register notices on 2026-10-25, and confirmed the chemical name and compound IDs in PubChem. Lab numbers come from state regulator datasets and published papers, not from our own testing. Certainty ratings follow the reviewers' own GRADE where given; others are the compilers' reading. Only abstracts were read for papers.

Study library: every study we checked 57

Design Research type Year from Search

Study	Design	People	Product, dose, length	Result
Abuhasira 2018 Epidemiological characteristics, safety and efficacy of medical cannabis in the elderly. European journal of internal medicine · PMID 29398248 · DOI	Cohort study	2736 2,736 patients aged over 65 starting medical cannabis in one Israeli clinic, 2015 to 2017	Medical cannabis (herbal and other products as supplied) Not summarised here 6 months · vs see abstract	Mean age 74.5. Most common reasons were pain (66.6%) and cancer (60.8%). After 6 months, 93.7% of respondents reported improvement and median pain fell from 8 to 4 on a 0 to 10 scale. Common adverse events were dizziness (9.7%) and dry mouth (7.1%). 18.1% stopped or reduced opioids. The authors call it safe and efficacious but ask for randomised trials. Uncontrolled clinic data: people reported feeling better and few side effects; not proof Methods, limits and funding Outcomes: Self-reported pain, quality of life, adverse events at 6 months Limitations: No control group; self-report; people who stopped early are not counted; one clinic. Funding and conflicts: Not stated in the abstract. Read: abstract
Ateş 2026 Cannabis-based medicines for chronic neuropathic pain in adults. The Cochrane database of systematic reviews · PMID 41548880 · DOI	Systematic review	2187 Adults with chronic neuropathic pain in 21 randomised double-blind trials (2 to 26 weeks)	THC-dominant, THC/CBD-balanced and CBD-dominant medicines (herbal, plant-based, synthetic) Not summarised here 2 to 26 weeks · vs Placebo (20 studies); dihydrocodeine (1)	THC-dominant medicines versus placebo: no clear evidence for 50% pain relief (RD 0.14, 95% CI -0.07 to 0.37; 7 studies, 534 participants; very low certainty); they may increase nervous system adverse events (RD 0.25, 0.14 to 0.37; low certainty). THC/CBD-balanced medicines: no clear evidence for 50% pain relief (RD 0.04, 0.00 to 0.08; very low certainty); they may increase "much or very much improved" ratings (RD 0.07, 0.02 to 0.11) and withdrawals from adverse events, low certainty, and the authors say these effects were not clinically relevant. CBD-dominant medicines: no clear evidence of benefit (very low certainty). No clear evidence of meaningful benefit for neuropathic pain; very low to low certainty Methods, limits and funding Outcomes: Pain relief of 50% or more, Patient Global Impression of Change, serious adverse events, withdrawals from adverse events Limitations: Risk of bias was low in 6 studies, unclear in 10 and high in 5. Short trials. Certainty rated very low to low. Funding and conflicts: No funding (Cochrane review). Read: abstract
Bassalov 2024 Prenatal cannabis exposure and the risk for neuropsychiatric anomalies in the offspring: a systematic review and meta-analysis. American journal of obstetrics and gynecology · PMID 38908654 · DOI	Meta-analysis	534445 534,445 children in 17 observational studies of prenatal cannabis exposure	Cannabis use during pregnancy Not summarised here see abstract · vs Unexposed pregnancies	After adjustment, pooled odds were 1.13 (95% CI 1.01 to 1.26) for ADHD and 1.20 (1.01 to 1.42) for the child's own later cannabis use. No significant link was found with autism (1.04), psychotic symptoms (1.29, 0.97 to 1.72), anxiety (1.34) or depression (0.72). The authors still advise caution during pregnancy. Small possible rise in ADHD and later cannabis use; no clear link with autism, anxiety or depression Methods, limits and funding Outcomes: Child ADHD, autism, psychotic symptoms, anxiety, depression, own cannabis use Limitations: Observational studies; wide uncertainty for some outcomes. Funding and conflicts: Not stated in the abstract. Read: abstract
Behzad 2025 Association of driving with blood delta-9-tetrahydrocannabinol: a systematic review. The international journal of neuropsychopharmacology · PMID 40172477 · DOI	Systematic review	12 studies (from 4,845 records) comparing blood THC with simulated driving	Blood THC level Not summarised here see abstract · vs see abstract	Ten of 12 studies found no significant straight-line link between blood THC and driving measures (for example 8 of 9 for weaving or lane control). The studies that did find a link used complex driving situations. The authors say current blood-level detection methods may suit only some situations. A blood THC number does not reliably show how impaired someone is Methods, limits and funding Outcomes: Whether higher blood THC lines up with worse driving Limitations: Only 12 studies; mostly lab simulators. Funding and conflicts: Not stated in the abstract. Read: abstract
Beneke 2025 Dazed and confused: variability in reported and measured tetrahydrocannabinol content in cannabis edibles. Clinical toxicology (Philadelphia, Pa.) · PMID 41025251 · DOI	Product analysis	12 12 cannabis edible packages bought from dispensaries and convenience stores in Jackson, Mississippi	Edibles (some containing delta-8-THC and delta-9-THC) Not summarised here see abstract · vs see abstract	Measured delta-8-THC and delta-9-THC differed from the labels by from 288 mg less to 5,491 mg more than advertised; no CBD was detected. Three products of the same brand differed from their labels by 1,542, 719 and 5,491 mg. Edible labels can be far from the real THC amount Methods, limits and funding Outcomes: Labelled versus lab-measured THC Limitations: Only 12 packages from one city. Funding and conflicts: Not stated in the abstract. Read: abstract
Bethoux 2024 A randomized, double-blind, placebo-controlled trial to evaluate the effect of nabiximols oromucosal spray on clinical measures of spasticity in patients with multiple sclerosis. Multiple sclerosis and related disorders · PMID 39106541 · DOI	Controlled trial	68 68 adults with multiple sclerosis spasticity despite baclofen, tizanidine or dantrolene (crossover, two 21-day periods)	Nabiximols oromucosal spray (THC:CBD) versus placebo spray Not summarised here Two 21-day periods · vs Placebo spray	The change in muscle tone was -0.23 with nabiximols and -0.26 with placebo; the treatment difference was 0.04 (P = 0.7152). The main goal was not met. Other muscle tone scores also did not differ. The authors note the product has shown benefit on patient-reported spasticity scores in earlier trials. No difference from placebo on clinician-measured muscle tone Methods, limits and funding Outcomes: Spastic muscle tone on the clinician-rated Modified Ashworth Scale (lower limb, 6 muscle groups) Limitations: Small crossover trial, three weeks per period; the endpoint was a clinician scale, not patient rating. Funding and conflicts: Sponsor not stated in the abstract. Trials of this branded product are usually run or paid for by its manufacturer (not checked here). Registration: NCT04657666 Read: abstract
Braun 2024 Cannabis and Cannabinoids in Adults With Cancer: ASCO Guideline. Journal of clinical oncology : official journal of the American Society of Clinical Oncology · PMID 38478773 · DOI	Guideline	Adults with cancer; evidence base of 13 systematic reviews and 5 primary studies (4 randomised trials, 1 cohort)	Cannabis and cannabinoids (synthetic, purified, combination and whole-plant) Not summarised here see abstract · vs Various	Certainty of evidence for most outcomes was low or very low. The panel recommends against cannabis or cannabinoids as a cancer treatment outside clinical trials. Cannabis or cannabinoids may improve refractory chemotherapy-induced nausea and vomiting when added to guideline-concordant antinausea regimens. Whether they help other supportive-care outcomes remains uncertain. Guideline: not a cancer treatment; may help refractory chemotherapy nausea as an add-on Methods, limits and funding Outcomes: Anti-cancer effect, treatment toxicity, symptoms, quality of life Limitations: A guideline built on low or very low certainty evidence. Funding and conflicts: PubMed lists research support from non-US government and NIH sources; conflicts are not stated in the abstract. Read: abstract
Chou 2026 Cannabis-Based Products for Chronic Pain : An Updated Systematic Review. Annals of internal medicine · PMID 41429020 · DOI	Systematic review	2303 Adults with chronic pain (64% neuropathic) in 25 short-term randomised placebo-controlled trials	Cannabinoids sorted by THC-to-CBD ratio, source (synthetic, purified, extracted) and route Not summarised here 1 to 6 months · vs Placebo	25 trials of 1 to 6 months (n = 2,303). Oral synthetic or purified THC-only products may slightly reduce pain and oral mucosal spray extracts with a comparable THC-to-CBD ratio probably slightly reduce pain (pooled differences -0.78 and -0.54 points on a 0 to 10 scale), with moderate or large increases in dizziness, sedation and nausea. Among THC-only products nabilone moderately reduced pain (-1.59 points) but dronabinol did not (-0.23 points). Low-THC mixtures may not improve outcomes. Small average pain reduction with the product types that contain THC; more dizziness, sedation and nausea Methods, limits and funding Outcomes: Pain severity on a 0 to 10 scale; dizziness, sedation, nausea Limitations: Products vary a lot; product details were often missing; US availability of some products is unclear; English-language studies only; no long-term data. Funding and conflicts: Agency for Healthcare Research and Quality (US government). PROSPERO CRD42021229579. Registration: CRD42021229579 Read: abstract
Chow 2025 Efficacy of cannabinoids for the prophylaxis of chemotherapy-induced nausea and vomiting-a systematic review and meta-analysis. Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer · PMID 39953210 · DOI	Meta-analysis	Adults with cancer in 26 randomised trials of cannabinoids to prevent chemotherapy-induced nausea and vomiting (23 published before the 2000s)	Cannabinoids (older single-agent trials and one recent THC:CBD trial) Not summarised here see abstract · vs Placebo and active antinausea drugs	Cannabinoids gave better overall control than placebo (RR 2.65, 95% CI 1.70 to 4.12). They were not better than the active medicines they were compared with, which were mostly older single-agent regimens. A recent phase II/III trial found THC:CBD helped as an add-on to modern antinausea drugs for secondary prevention, mostly without olanzapine. The authors call the evidence for cannabinoids in the era of triple and quadruple antinausea regimens "scant". Better than placebo; no better than older active drugs; unclear value on top of modern regimens Methods, limits and funding Outcomes: Complete response, no nausea, no vomiting, no rescue medication Limitations: Nearly half of trials had some concern for bias; most trials are decades old. Funding and conflicts: Not stated in the abstract. Read: abstract
Cohen 2023 Severe outcomes following pediatric cannabis intoxication: a prospective cohort study of an international toxicology surveillance registry. Clinical toxicology (Philadelphia, Pa.) · PMID 37603042 · DOI	Cohort study	138 138 children and teens (under 18) seen in emergency departments with cannabis intoxication, 2017 to 2020	Cannabis exposure (edibles, other forms, mixed with other drugs) Not summarised here see abstract · vs see abstract	Fifty-two of 138 (38%) were admitted to intensive care and one died. Mixing with other substances (adjusted OR 16.3, 95% CI 4.6 to 58.3) and eating cannabis edibles (adjusted OR 5.5, 1.9 to 15.9) predicted severe outcomes. All children aged 10 or younger had eaten edibles. Edibles and mixed substances were the strongest predictors of severe outcomes in children Methods, limits and funding Outcomes: Severe outcome (intensive care or death) Limitations: Toxicology centres only; severe cases are over-represented. Funding and conflicts: Not stated in the abstract. Read: abstract
Di Forti 2019 The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI): a multicentre case-control study. The lancet. Psychiatry · PMID 30902669 · DOI	Case control	2138 901 adults aged 18 to 64 with first-episode psychosis and 1,237 population controls at 11 sites in Europe and Brazil	Cannabis use patterns; potency split into low (THC under 10%) and high (THC 10% or more) using regional data on expected THC Not summarised here see abstract · vs Never users	Daily use was linked to higher odds of psychotic disorder than never use (adjusted OR 3.2, 95% CI 2.2 to 4.1), rising to nearly five times for daily use of high-potency types (OR 4.8, 2.5 to 6.3). If causal, removing high-potency cannabis would prevent 12.2% of first-episode psychosis across sites (95% CI 3.0 to 16.1), 30.3% in London and 50.3% in Amsterdam. Site rates of psychosis were correlated with local use of high-potency cannabis (r = 0.7) and daily use (r = 0.8). Daily and high-potency use linked to much higher odds of psychotic disorder Methods, limits and funding Outcomes: Odds of psychotic disorder by frequency and potency of use; share of cases linked to high-potency cannabis if causal Limitations: Case-control design; potency was estimated from regional data, not measured in each person's own cannabis; cause cannot be proved. Funding and conflicts: Public funders (Medical Research Council, EU, São Paulo Research Foundation, NIHR, Wellcome Trust). Read: abstract
FDA 2023 FDA: FDA and Cannabis: Research and Drug Approval Process (content current as of 24 Feb 2023) FDA · Source	Authoritative guidance n/a	General public	n/a	FDA has not approved a marketing application for cannabis for the treatment of any disease. It has approved one cannabis-derived drug (Epidiolex, cannabidiol) and three synthetic cannabis-related drugs: Marinol (dronabinol), Syndros (dronabinol) and Cesamet (nabilone). These are available only by prescription. FDA has not approved any other cannabis, cannabis-derived or CBD products on the market. Methods, limits and funding Outcomes: FDA position on approved cannabis-related drugs Limitations: Content current as of 24 February 2023. Read: full-text
Filippini 2022 Cannabis and cannabinoids for symptomatic treatment for people with multiple sclerosis. The Cochrane database of systematic reviews · PMID 35510826 · DOI	Systematic review	3763 Adults with multiple sclerosis in 25 randomised trials (3 to 48 weeks); 2,290 received a cannabinoid	Nabiximols (13 trials), synthetic THC-like drugs (7), oral cannabis extract (2), inhaled herbal cannabis (1) Not summarised here 3 to 48 weeks · vs Placebo	Nabiximols probably increases the number of people reporting an important reduction in spasticity versus placebo (OR 2.51, 95% CI 1.56 to 4.04; 5 trials, 1,143 people; moderate certainty; about 216 more per 1,000). Cannabinoids probably increase "much or very much improved" ratings (OR 1.80, 1.37 to 2.36; moderate certainty). Treatment stopped because of side effects slightly more often (OR 2.41, 1.51 to 3.84; low certainty). They may increase nervous system disorders (OR 2.61, 1.53 to 4.44) and psychiatric disorders (OR 1.94, 1.31 to 2.88), both low certainty. Effect on chronic neurological pain and quality of life is uncertain. Moderate certainty that nabiximols reduces spasticity in the short term; more side effects Methods, limits and funding Outcomes: Spasticity (30% reduction on a rating scale), pain, global impression, quality of life, adverse events Limitations: Short trials; certainty limited by short duration. Funding and conflicts: Research support from a non-US government source; Cochrane review. Read: abstract
Ghasemiesfe 2018 Marijuana Use, Respiratory Symptoms, and Pulmonary Function: A Systematic Review and Meta-analysis. Annals of internal medicine · PMID 29971337 · DOI	Meta-analysis	22 studies of adolescents and adults who smoked marijuana	Smoked marijuana Not summarised here see abstract · vs see abstract	Pooled prospective studies linked marijuana use to cough (RR 2.04, 95% CI 1.02 to 4.06) and sputum (RR 3.84, 1.62 to 9.07). Cross-sectional studies linked it to cough (RR 4.37), sputum (3.40), wheezing (2.83) and breathlessness (1.56). Data on lung function and obstructive lung disease were insufficient. Smoking cannabis goes with cough, phlegm and wheeze (low-strength evidence) Methods, limits and funding Outcomes: Cough, sputum, wheeze, breathlessness, lung function, obstructive lung disease Limitations: Few low-risk-of-bias studies; exposure was limited and young cohorts. Funding and conflicts: The abstract says no primary funding source; PubMed also lists NIH extramural support. PROSPERO CRD42017059224. Registration: CRD42017059224 Read: abstract
Giordano 2025 Accuracy of labeled THC potency across flower and concentrate cannabis products. Scientific reports · PMID 40592871 · DOI	Product analysis	277 277 cannabis products (178 flower, 99 concentrate) bought from 52 Colorado dispensaries	Cannabis flower and concentrates (labelled THC versus lab-measured THC) Not summarised here see abstract · vs see abstract	96.0% of concentrates were within 15% of the labelled THC, compared with 56.7% of flower. Measured THC was lower than labelled in both flower and concentrate. The authors call for independent, blinded verification of labels. Flower labels were often off by more than 15%, usually reading higher than the lab result Methods, limits and funding Outcomes: Whether measured THC was within 15% of the label Limitations: Colorado only; one set of lab methods. Funding and conflicts: Not stated in the abstract. Read: abstract
Gobbi 2019 Association of Cannabis Use in Adolescence and Risk of Depression, Anxiety, and Suicidality in Young Adulthood: A Systematic Review and Meta-analysis. JAMA psychiatry · PMID 30758486 · DOI	Meta-analysis	23317 23,317 people in 11 longitudinal studies of cannabis use before age 18 and mood in young adulthood (18 to 32)	Cannabis use in adolescence Not summarised here see abstract · vs Non-users	Compared with non-users, adolescents who used cannabis had higher odds of depression in young adulthood (OR 1.37, 95% CI 1.16 to 1.62), suicidal ideation (OR 1.50, 1.11 to 2.03) and suicide attempt (OR 3.46, 1.53 to 7.84). Anxiety was not significantly raised (OR 1.18, 0.84 to 1.67). The authors say individual-level risk is moderate to low. Association between teen cannabis use and later depression and suicidal behaviour Methods, limits and funding Outcomes: Later depression, anxiety, suicidal ideas and attempts (adjusted for earlier depression or anxiety) Limitations: Observational; other factors may explain part of the link. Funding and conflicts: Research support from a non-US government source (PubMed record). Read: abstract
Goodwin 2026 The subjective effects of △9-tetrahydrocannabinol: A systematic review and dose-response meta-regression. Neuroscience and biobehavioral reviews · PMID 42242480 · DOI	Meta-analysis	1020 1,020 healthy adults with infrequent cannabis use (under 2 times a week) in 41 randomised placebo-controlled experiments	THC (inhaled or oral) Not summarised here see abstract · vs Placebo	THC caused greater feelings of being high, anxiety and tiredness (all p < 0.001), and lower alertness and contentedness (p < 0.01) compared with placebo. Higher inhaled doses raised the high and contentedness and lowered calmness. Oral dose was not linked to subjective effects in this analysis. In people who rarely use cannabis, THC reliably causes feeling high, anxiety and tiredness Methods, limits and funding Outcomes: Peak subjective effects: high, anxiety, tiredness, alertness, contentedness, calmness Limitations: Healthy volunteers who rarely use cannabis; lab setting. Funding and conflicts: Not stated in the abstract. Read: abstract
Gorbenko 2025 CB1 Receptor Antagonist Selonabant (ANEB-001) Blocks Acute THC Effects in Healthy Volunteers: A Phase II Randomized Controlled Trial. Clinical pharmacology and therapeutics · PMID 39898464 · DOI	Controlled trial	85 Healthy adults given THC with a CB1 blocker (selonabant) or placebo, in two parts (60 and 25 people)	THC with selonabant or placebo Not summarised here see abstract · vs Placebo	The CB1 blocker reduced the "feeling high" rating by up to 82.8% versus placebo (95% CI -91.0 to -67.2%), increased alertness and reduced body sway; heart rate effects were not significant. The authors say THC intoxication is mainly mediated by CB1 receptor activation. Blocking CB1 blocked most THC effects in people, supporting CB1 as the main target Methods, limits and funding Outcomes: Feeling high, alertness, body sway, heart rate Limitations: Small phase II trial of an experimental drug; healthy volunteers. Funding and conflicts: PubMed lists non-US government research support; the abstract does not name the sponsor, and the blocker is an experimental drug. Read: abstract
Grotenhermen 2003 Pharmacokinetics and pharmacodynamics of cannabinoids. Clinical pharmacokinetics · PMID 12648025 · DOI	Review	Review of THC and other cannabinoids in humans	THC by inhaled, oral and other routes Not summarised here see abstract · vs see abstract	THC is the main source of the effects of cannabis. Inhaled THC reaches a peak in blood within minutes; effects start in seconds to minutes, peak at 15 to 30 minutes and taper off in 2 to 3 hours. After swallowing, effects start 30 to 90 minutes later, peak at 2 to 3 hours and last about 4 to 12 hours. Overdosing mainly causes anxiety and panic and a faster heart rate. Regular use can lead to dependency and a mild withdrawal syndrome. Cannabinoids act on G-protein-coupled cannabinoid receptors and also by other routes. Methods, limits and funding Outcomes: Pharmacokinetics and pharmacodynamics Limitations: Published 2003, so it predates modern high-potency products. Funding and conflicts: Not stated in the abstract. Read: abstract
Hamm-Buiar 2026 Efficacy and safety of cannabinoids in the treatment of spasticity in multiple sclerosis: A systematic review of randomized clinical trials. British journal of clinical pharmacology · PMID 42543771 · DOI	Systematic review	3000 People with multiple sclerosis spasticity in 27 randomised trials (more than 3,000 participants)	THC:CBD extracts, other cannabis extracts, isolated cannabinoids and smoked cannabis Not summarised here see abstract · vs Placebo	THC:CBD extracts reduced patient-reported spasticity versus placebo (MD -0.82; 95% CI -1.24 to -0.40; 7 studies). Network meta-analysis agreed (MD -0.80; 95% CrI -1.61 to -0.32). Adverse events were frequent but mostly mild to moderate; serious events were rare. Certainty was low and the trials differed a lot. The authors conclude cannabinoids give modest improvements in patient-reported spasticity with limited objective benefit. Small improvement in how people rate spasticity; certainty low Methods, limits and funding Outcomes: Patient-reported spasticity; adverse events Limitations: Low certainty; large differences between trials; objective benefit limited. Funding and conflicts: Not stated in the abstract. Read: abstract
Holdsworth 2024 Human Milk Cannabinoid Concentrations and Associations with Maternal Factors: The Lactation and Cannabis (LAC) Study. Breastfeeding medicine : the official journal of the Academy of Breastfeeding Medicine · PMID 38695182 · DOI	Cohort study	20 20 breastfeeding mothers who used cannabis at least weekly, fewer than 6 months after birth	Cannabis use (various products, self-chosen) Not summarised here 8 to 12 hours · vs see abstract	THC was measurable in milk after cannabis use and peaked about 120 minutes after one use. More uses in the study period meant more THC in milk (rho = 0.65), which the authors read as THC building up. Milk THC at baseline went with how often mothers used cannabis. Variation between mothers was large. THC passes into breast milk and can build up with repeated use Methods, limits and funding Outcomes: THC concentrations in breast milk over 8 to 12 hours Limitations: Only 20 women; not a study of infant outcomes. Funding and conflicts: Research support from a non-US government source (PubMed record). Read: abstract
Hsu 2026 Therapeutic Use of Cannabis and Cannabinoids: A Review. JAMA · PMID 41296368 · DOI	Review	Narrative review of therapeutic cannabis and cannabinoids, with adults in the US and Canada as the main context	Cannabis and cannabinoids (approved drugs and consumer products) Not summarised here see abstract · vs see abstract	FDA-approved cannabinoid uses are HIV/AIDS-related anorexia, chemotherapy-induced nausea and vomiting, and certain pediatric seizure disorders. Prescribed cannabinoids gave a small reduction in nausea and vomiting versus placebo or active comparators (SMD -0.29, 95% CI -0.39 to -0.18). In HIV/AIDS, cannabinoids moderately increased body weight versus placebo (SMD 0.57, 0.22 to 0.92). Guidelines do not recommend inhaled or high-potency cannabis (10% or 10 mg THC or more) for medical purposes. High-potency use was linked to more psychotic symptoms (12.4% vs 7.1%) and generalised anxiety disorder (19.1% vs 11.6%). In a meta-analysis of observational studies, 29% of people using cannabis for medical purposes met criteria for cannabis use disorder. Evidence is insufficient for most medical indications. Evidence insufficient for most uses; approved uses are limited Methods, limits and funding Outcomes: Approved uses, evidence by condition, harms Limitations: A narrative review, not a new systematic search. Funding and conflicts: Not stated in the abstract. Read: abstract
Kesner 2021 Cannabis use, abuse, and withdrawal: Cannabinergic mechanisms, clinical, and preclinical findings. Journal of neurochemistry · PMID 33891706 · DOI	Review mixed (human and animal)	Review of human and animal research on cannabis use disorder and withdrawal	THC Not summarised here see abstract · vs see abstract	THC is the compound that drives cannabis abuse potential. A significant subgroup of frequent users develop cannabis use disorder, with craving, tolerance and withdrawal. Animal work points to action on the CB1 receptor and downstream signalling as a main component. There are no efficacious treatments for cannabis use disorder or withdrawal. Methods, limits and funding Outcomes: Brain mechanisms of use, dependence and withdrawal Limitations: A narrative review; much of the mechanism work is in animals. Funding and conflicts: PubMed lists NIH extramural research support. Read: abstract
Leung 2020 What is the prevalence and risk of cannabis use disorders among people who use cannabis? a systematic review and meta-analysis. Addictive behaviors · PMID 32485547 · DOI	Meta-analysis	People in the general population who had used cannabis, in 21 studies published 2009 to 2019	Cannabis use Not summarised here see abstract · vs see abstract	Among people who used cannabis, 22% (95% CI 18 to 26%) had a cannabis use disorder, 13% (8 to 18%) abuse and 13% (10 to 15%) dependence. In cohort studies the risk of dependence rose to 33% (22 to 44%) among young people who used weekly or daily. 22% of people who use cannabis meet criteria for a use disorder Methods, limits and funding Outcomes: Share meeting criteria for cannabis use disorder (CUD), abuse or dependence Limitations: Few cohort studies; data on regular users were lacking for CUD and abuse. Funding and conflicts: Not stated in the abstract. Read: abstract
Marchand 2022 Birth Outcomes of Neonates Exposed to Marijuana in Utero: A Systematic Review and Meta-analysis. JAMA network open · PMID 35084479 · DOI	Meta-analysis	59138 59,138 pregnant women in 16 studies comparing marijuana-exposed and unexposed pregnancies	Marijuana use during pregnancy Not summarised here see abstract · vs Unexposed pregnancies	Exposure was linked to higher risk of birth weight under 2,500 g (RR 2.06, 95% CI 1.25 to 3.42), small for gestational age (RR 1.61, 1.44 to 1.79), preterm delivery (RR 1.28, 1.16 to 1.42) and NICU admission (RR 1.38, 1.18 to 1.62), and with lower mean birth weight (-112 g), lower Apgar score at 1 minute and smaller head circumference (-0.34 cm). Several adverse newborn outcomes are more common after cannabis use in pregnancy (association) Methods, limits and funding Outcomes: Birth weight, small for gestational age, preterm delivery, NICU admission, Apgar scores, head size Limitations: Observational; tobacco and other factors may mix with the effect. Funding and conflicts: Not stated in the abstract. Read: abstract
Marconi 2016 Meta-analysis of the Association Between the Level of Cannabis Use and Risk of Psychosis. Schizophrenia bulletin · PMID 26884547 · DOI	Meta-analysis	66816 66,816 people in 10 studies (18 in the systematic review) with data on cannabis use before the onset of psychosis	Cannabis (self-reported level of use; THC content not measured in the pooled model) Not summarised here see abstract · vs Non-users	Higher levels of use were linked to more psychosis in all included studies. The heaviest users had about four times the odds of schizophrenia or other psychosis-related outcomes versus non-users (OR 3.90, 95% CI 2.84 to 5.34). The authors say a causal link cannot be proved but there is enough evidence to justify harm-reduction programmes. Dose-response: more cannabis use, more psychosis risk (association) Methods, limits and funding Outcomes: Schizophrenia and other psychosis-related outcomes by level of cannabis use Limitations: Observational data; cause cannot be proved; people at higher risk may use more cannabis. Funding and conflicts: Not stated in the abstract. Read: abstract
McDonagh 2022 Cannabis-Based Products for Chronic Pain : A Systematic Review. Annals of internal medicine · PMID 35667066 · DOI	Systematic review	14835 Adults with chronic pain: 18 randomised placebo-controlled trials (n = 1,740) and 7 cohort studies (n = 13,095)	Cannabinoids sorted by THC-to-CBD ratio and source (synthetic, extract/purified, whole plant) Not summarised here Mostly 1 to 6 months · vs Placebo (trials); none (cohorts)	Synthetic products with high THC-to-CBD ratios (more than 98% THC) may be linked to moderate improvement in pain severity and response and increase sedation; they are probably linked to a large increase in dizziness. Extracted high-ratio products may cause large increases in study withdrawal because of adverse events and in dizziness. A sublingual spray with a comparable THC-to-CBD ratio (1.1:1) probably gives a small improvement in pain and function, with large increases in dizziness and sedation and moderate increase in nausea. Evidence for other products and for longer-term harms was insufficient. Short-term improvements for some products; evidence for others and for long-term harms is insufficient Methods, limits and funding Outcomes: Pain severity, response (30% or more improvement), function, adverse events Limitations: Studies were mostly 1 to 6 months; products differ; some are not available in the US; some outcomes had inadequate evidence. Funding and conflicts: Agency for Healthcare Research and Quality (US government). PROSPERO CRD42021229579. Registration: CRD42021229579 Read: abstract
Mechoulam 2014 Early phytocannabinoid chemistry to endocannabinoids and beyond. Nature reviews. Neuroscience · PMID 25315390 · DOI	Review	Historical review of cannabinoid chemistry and the endocannabinoid system	Cannabinoids Not summarised here see abstract · vs see abstract	Isolation and structure elucidation of most major cannabinoids, including delta-9-THC (the main psychoactive molecule in Cannabis sativa), was achieved in the 1960s and 1970s. Two cannabinoid receptors were identified in the 1980s and early 1990s, followed by the endocannabinoids. Methods, limits and funding Outcomes: History of discovery Limitations: A short historical abstract. Funding and conflicts: Research support from non-US government sources (PubMed record). Read: abstract
Mücke 2018 Systematic review and meta-analysis of cannabinoids in palliative medicine. Journal of cachexia, sarcopenia and muscle · PMID 29400010 · DOI	Meta-analysis	1561 People with cancer or HIV in 9 randomised trials in palliative medicine	Cannabinoids (e.g. dronabinol and others) Not summarised here see abstract · vs Placebo and megestrol	In cancer, cannabinoids were not better than placebo for calorie intake, appetite, nausea and vomiting, pain relief of 30% or more, or sleep. In HIV, cannabinoids beat placebo for weight gain (SMD 0.57, 95% CI 0.22 to 0.92) and appetite (SMD 0.57, 0.11 to 1.03) but not for nausea. In HIV there were more mental health symptoms (RD 0.05, 0.00 to 0.11). Megestrol beat cannabinoids for appetite and weight in head-to-head trials. The authors found no convincing high-quality evidence of value for anorexia or cachexia in cancer or HIV. Some weight and appetite benefit in HIV; none in cancer; evidence low or very low Methods, limits and funding Outcomes: Appetite, weight, calorie intake, nausea, pain, sleep, side effects Limitations: Quality of evidence low or very low; searches ended March 2017. Funding and conflicts: Research support from a non-US government source (PubMed record). Read: abstract
NCBI PubChem 2026 PubChem: Tetrahydrocannabinol (delta-9-THC), CID 16078 PubChem REST API · Source	Database n/a	Chemical record	n/a	PubChem CID 16078, title "Tetrahydrocannabinol", formula C21H30O2, molecular weight 314.5, IUPAC name (6aR,10aR)-6,6,9-trimethyl-3-pentyl-6a,7,8,10a-tetrahydrobenzo[c]chromen-1-ol. The name dronabinol resolves to the same record. Methods, limits and funding Outcomes: Name, formula and structure Limitations: A database record, retrieved 2026-10-25. Read: API
NCCIH 2019 NCCIH (NIH): Cannabis (Marijuana) and Cannabinoids: What You Need To Know (last updated November 2019) NCCIH · Source	Authoritative guidance n/a	General public	n/a	FDA has not approved the cannabis plant for any medical use. FDA has approved Marinol and Syndros (dronabinol, synthetic THC) and Cesamet (nabilone, a synthetic substance similar to THC) for chemotherapy nausea and vomiting, and dronabinol for loss of appetite and weight loss in HIV/AIDS. Some evidence suggests modest benefits for chronic pain and multiple sclerosis symptoms; cannabis is not helpful for glaucoma. Safety concerns listed include motor vehicle crashes, lower birth weight with smoking in pregnancy, cannabis use disorder (adolescents four to seven times more likely than adults to develop it), injury in older adults, a higher risk of schizophrenia or other psychoses in people predisposed, orthostatic hypotension, accidental consumption by children, recurrent severe vomiting in some long-term high-dose users, contamination and label amounts that differ substantially from contents. Nabiximols (Sativex) is approved in more than 25 countries outside the United States. Methods, limits and funding Outcomes: What NIH says about approved cannabinoid drugs, evidence by condition and safety Limitations: Last updated November 2019; some statements may be out of date. Read: full-text
NIDA 2026 NIDA (NIH): Cannabis (Marijuana) research topic page NIDA · Source	Authoritative guidance n/a	General public	n/a	THC is the intoxicating compound and delta-9 THC is the most abundant form in the plant. Higher THC concentrations are linked to a greater likelihood of progressing to cannabis use disorder. Studies estimate 22% to 30% of people who use cannabis have it. Cannabis use is linked to earlier onset of psychosis in people with genetic risk, and an episode of intoxication-induced psychosis can occur at high doses. Cannabis hyperemesis syndrome means nausea, vomiting and abdominal pain after long-term heavy use and only resolves when use stops completely. Cannabis is the drug most often found in the blood of drivers in crashes, but a National Highway Traffic Safety Administration study found no significant increased crash risk attributable to cannabis. Use in pregnancy is linked to lower birth weight and preterm birth; ACOG recommends counselling against use when trying to get pregnant, during pregnancy and while breastfeeding. Heavy adolescent use is linked to poorer working memory and academic functioning. Older adults may face interactions with warfarin, opioids and benzodiazepines. Children eating edibles is a growing health concern. Delta-8 THC products are not FDA-approved. Methods, limits and funding Outcomes: What NIDA says about THC effects, cannabis use disorder, mental health, driving, pregnancy, adolescents and older adults Limitations: Page date was not shown in the text we read (read on 2026-10-25). ACOG statements are reported by NIDA, not read directly. Read: full-text
NIDDK 2023 LiverTox: Marijuana (NIDDK, record updated 2023) LiverTox (NCBI Bookshelf NBK590051) · PMID 37023245	Authoritative guidance	Published case reports and small series on cannabis and liver tests	Marijuana (cannabis plant material)	States that marijuana use has not been linked to serum enzyme elevations during therapy or to instances of clinically apparent liver injury with jaundice. Its reference list notes older reports of mild liver test abnormalities in frequent users and of faster liver scarring in chronic hepatitis C among daily cannabis smokers. Methods, limits and funding Outcomes: Whether marijuana use is linked to raised liver enzymes or clinically apparent liver injury Limitations: Short entry; the overview is dated 2012 and was updated in 2023; it does not cover modern high-potency products. Read: abstract
Patilea-Vrana 2019 Hepatic Enzymes Relevant to the Disposition of (-)-∆9-Tetrahydrocannabinol (THC) and Its Psychoactive Metabolite, 11-OH-THC. Drug metabolism and disposition: the biological fate of chemicals · PMID 30567877 · DOI	Laboratory research in vitro	Human liver microsomes and recombinant enzymes	THC and its active metabolite 11-OH-THC Not summarised here see abstract · vs see abstract	CYP2C9 was the main enzyme depleting THC and forming 11-OH-THC (fraction metabolised 0.82 and 0.99). CYP2D6 and CYP2C19 were minor. 11-OH-THC was cleared by UGT and CYP enzymes (UGT2B7, UGT1A9, CYP2C9, CYP3A4 contribute). Methods, limits and funding Outcomes: Which enzymes break down THC Limitations: Test-tube study; amounts are for modelling. Funding and conflicts: Research support from a non-US government source (PubMed record). Read: abstract
Power 2021 Intelligence quotient decline following frequent or dependent cannabis use in youth: a systematic review and meta-analysis of longitudinal studies. Psychological medicine · PMID 33501901 · DOI	Meta-analysis	6116 7 cohort studies: 808 young people with frequent or dependent cannabis use and 5,308 controls, with IQ measured before and after	Frequent or dependent cannabis use in youth Not summarised here see abstract · vs see abstract	Frequent or dependent use in youth was linked to a small IQ decline (Cohen's d -0.132, 95% CI -0.198 to -0.066), roughly 2 IQ points on average. Study quality was moderate to high. About 2 IQ points lower after frequent or dependent youth use (association) Methods, limits and funding Outcomes: Change in IQ Limitations: Follow-up was short; effect size is small; other factors may play a role. Funding and conflicts: Research support from a non-US government source (PubMed record). PROSPERO CRD42019125624. Registration: CRD42019125624 Read: abstract
Qian 2019 The Potential for Pharmacokinetic Interactions Between Cannabis Products and Conventional Medications. Journal of clinical psychopharmacology · PMID 31433338 · DOI	Systematic review	Laboratory and clinical studies on cannabis-drug interactions (search to March 2019)	THC, CBD and CBN (cannabinoids) Not summarised here see abstract · vs see abstract	In test-tube work, THC, CBD and CBN are likely to inhibit CYP2C9, CYP1A1/2 and CYP1B1; THC and CBD inhibit CYP2D6, CYP2C19, CYP2B6 and CYP2J2. THC also activates CYP2C9 and induces CYP1A1. Clinical studies suggest inhibition of CYP2C9 by various cannabis products and induction of CYP1A2 by cannabis smoking. Medicines cleared mainly by CYP2C19, CYP2C9 and CYP1A2 may be at risk of altered levels. No clinical studies were found showing an effect on drug transporters. Possible changes in levels of drugs cleared by CYP2C9, CYP2C19 and CYP1A2 Methods, limits and funding Outcomes: Effects on drug-metabolising enzymes and transporters Limitations: Mostly test-tube data; few clinical interaction studies. Funding and conflicts: Not stated in the abstract. Read: abstract
Rogeberg 2019 A meta-analysis of the crash risk of cannabis-positive drivers in culpability studies-Avoiding interpretational bias. Accident; analysis and prevention · PMID 30468948 · DOI	Meta-analysis	13 published culpability (crash responsibility) studies	THC-positive drivers Not summarised here see abstract · vs see abstract	Average increase in crash risk for THC-positive drivers was estimated at 1.28 (95% credibility interval 1.16 to 1.40); culpable crash risk 1.42 (1.11 to 1.75). The attributable risk fraction of cannabis-impaired driving was below 2% for all but two studies. The author concludes the raised crash risk linked to THC-positive drivers in these studies is low. Crash risk raised modestly in THC-positive drivers (about 28%) Methods, limits and funding Outcomes: Total and culpable crash risk using a Bayesian model Limitations: Culpability studies; THC-positive does not always mean impaired. Funding and conflicts: Not stated in the abstract. Read: abstract
SVX Therapeutics / emc 2026 Electronic Medicines Compendium (UK): Sativex oromucosal spray, Summary of Product Characteristics (text revised 2 April 2026) emc (UK) · Source	Authoritative guidance n/a	Adults with moderate to severe MS spasticity	Nabiximols (Sativex) oromucosal spray	Sativex (a plant-extract spray with THC and CBD in a roughly equal ratio) is licensed in the UK for symptom improvement in adults with moderate to severe spasticity due to multiple sclerosis who have not responded adequately to other anti-spasticity medicines and who show clinically significant improvement during an initial trial. Treatment must be started and supervised by a specialist physician. It is contraindicated with a known or suspected personal or family history of schizophrenia or other psychotic illness, a history of severe personality disorder or other significant psychiatric disorder, and during breastfeeding. Warnings cover dizziness, fainting and heart rate and blood pressure changes (not recommended in serious cardiovascular disease), psychiatric symptoms including paranoid ideas and transient psychotic reactions, falls, and caution with epilepsy. It may reduce the effectiveness of hormonal contraceptives. Patients should not drive if they have dizziness or sleepiness; the label says driving may be affected and that driving while impaired is an offence in the UK. Dependence is described as unlikely. Methods, limits and funding Outcomes: UK licensed use, contraindications and warnings for nabiximols Limitations: UK document; other countries have their own labels. emc page last updated 14 July 2026. Read: full-text
Sebastian 2025 Cannabis use and atherosclerotic cardiovascular disease outcomes: A meta-analysis of multinational cohort data. Disease-a-month : DM · PMID 39800612 · DOI	Meta-analysis	1902481 1,902,481 adults aged 18 to 74 in 17 cohort studies, mean follow-up 8.5 years	Ever use of cannabis Not summarised here see abstract · vs see abstract	Ever use was not significantly linked to heart attack (RR 1.25, 95% CI 0.91 to 1.71) or stroke (RR 1.38, 0.88 to 2.16) but was linked to the combined outcome of any adverse cardiovascular event (RR 1.48, 1.16 to 1.90). Association with any cardiovascular event; individual outcomes not significant Methods, limits and funding Outcomes: Heart attack, stroke, any adverse cardiovascular event Limitations: Observational; cause not shown; "ever use" mixes light and heavy use. Funding and conflicts: Not stated in the abstract. PROSPERO CRD42024530366. Registration: CRD42024530366 Read: abstract
Senderovich 2022 A Systematic Review on Cannabis Hyperemesis Syndrome and Its Management Options. Medical principles and practice : international journal of the Kuwait University, Health Science Centre · PMID 34724666 · DOI	Systematic review	Case reports and studies on cannabis hyperemesis syndrome (17 of 225 records reviewed)	Long-term cannabis use (6 months to 11 years before symptoms) Not summarised here see abstract · vs see abstract	Cannabis hyperemesis syndrome means repeated severe nausea and vomiting in long-term cannabis users. Rome IV criteria need cannabis use and nausea and vomiting symptoms for at least 6 months. Stopping cannabis was the most successful management. Hot-water bathing, topical capsaicin cream, haloperidol, droperidol, benzodiazepines, propranolol and aprepitant relieved symptoms in some reports. Stopping cannabis is the treatment that works best; other treatments help symptoms Methods, limits and funding Outcomes: Description and treatment of repeated severe nausea and vomiting Limitations: Small case-based evidence; no controlled trials reviewed. Funding and conflicts: Not stated in the abstract. Read: abstract
Simmons 2022 The effects of cannabis and alcohol on driving performance and driver behaviour: a systematic review and meta-analysis. Addiction (Abingdon, England) · PMID 35083810 · DOI	Meta-analysis	1725 1,725 participants in 57 experimental driving studies (simulator, closed course, on road)	Cannabis and/or alcohol given alone or together Not summarised here see abstract · vs Placebo and alcohol	Cannabis alone impaired lateral control (lateral position variability g = 0.331, 95% CI 0.212 to 0.451; lane excursions g = 0.198, 0.001 to 0.395) and lowered driving speed (g = -0.176). Cannabis plus alcohol caused larger driving decrements than either alone. Effects of cannabis were similar to those of low blood alcohol concentrations. Cannabis impairs driving; cannabis with alcohol is worse than either alone Methods, limits and funding Outcomes: Lane position variability, lane excursions, speed, reaction measures Limitations: Few data on some measures; studies differed. Funding and conflicts: PubMed lists research support from a non-US government source; conflicts are not stated in the abstract. Read: abstract
Simon 2022 Cannabinoid interventions for improving cachexia outcomes in cancer: a systematic review and meta-analysis. Journal of cachexia, sarcopenia and muscle · PMID 34881518 · DOI	Meta-analysis	People with cancer cachexia (loss of appetite and weight) in 4 randomised trials and 6 non-randomised studies	Cannabinoids Not summarised here see abstract · vs Control	Very low-quality evidence showed no benefit of cannabinoids for appetite (SMD -0.02, 95% CI -0.51 to 0.46). Moderate-quality evidence showed cannabinoids did worse than control on quality of life (SMD -0.25, -0.43 to -0.07). Low-quality evidence showed no benefit for weight gain, appetite or quality of life. No clear benefit for appetite or weight in cancer cachexia; quality of life was worse Methods, limits and funding Outcomes: Appetite, weight, quality of life, adverse effects Limitations: Only 10 studies, mostly small; very low to low quality. Funding and conflicts: Research support from a non-US government source (PubMed record). Read: abstract
Sirikantaramas 2007 Recent advances in Cannabis sativa research: biosynthetic studies and its potential in biotechnology. Current pharmaceutical biotechnology · PMID 17691992 · DOI	Review plant	Review of cannabis plant biosynthesis research	Cannabis sativa enzymes Not summarised here see abstract · vs see abstract	THC, CBD and CBC are made by enzymes in the plant, including THCA synthase, which turns CBGA (cannabigerolic acid) into THCA. THCA and CBGA were shown to induce cell death in tests and may play a role in plant defence. Tobacco hairy roots given the THCA synthase gene made THCA when fed CBGA. Methods, limits and funding Outcomes: How the plant makes cannabinoids Limitations: Review from 2007; plant and lab work. Funding and conflicts: Research support from a non-US government source (PubMed record). Read: abstract
Solmi 2023 Balancing risks and benefits of cannabis use: umbrella review of meta-analyses of randomised controlled trials and observational studies. BMJ (Clinical research ed.) · PMID 37648266 · DOI	Systematic review	101 meta-analyses of randomised trials (51) and observational studies (50) on cannabis, cannabinoids and cannabis-based medicines	Cannabis, cannabinoids and cannabis-based medicines Not summarised here see abstract · vs see abstract	From trials with moderate to high certainty, cannabis-based medicines raised central nervous system adverse events (equivalent OR 2.84, 95% CI 2.16 to 3.73), psychological effects (3.07, 1.79 to 5.26) and vision effects (3.00, 1.79 to 5.03). They improved nausea and vomiting, pain and spasticity but increased psychiatric and gastrointestinal adverse events and sleepiness (moderate). For chronic pain, 30% pain reduction odds were 0.59 (0.37 to 0.93; high certainty) with more psychological distress. In the general population cannabis worsened positive psychotic symptoms (5.21, 3.36 to 8.01). Convincing observational evidence: small for gestational age (1.61), low birth weight (1.43), car crash (1.27) and psychosis (1.71). The authors conclude convincing or converging evidence supports avoiding cannabis in adolescence, in people prone to or with mental health disorders, in pregnancy, and before and while driving. Medicines help some symptoms with added side effects; clear harms signals for pregnancy, driving, psychosis Methods, limits and funding Outcomes: Efficacy and safety outcomes across conditions Limitations: An umbrella review of other reviews; different products pooled. Funding and conflicts: No funding. PROSPERO CRD42018093045. Registration: CRD42018093045 Read: abstract
Spiga 2025 Pharmacotherapies for cannabis use disorder. The Cochrane database of systematic reviews · PMID 41025421 · DOI	Systematic review	3201 People with cannabis use disorder or likely dependence in 37 randomised trials	Medicines tested for cannabis withdrawal or cutting down, including THC preparations, cannabidiol, N-acetylcysteine and others Not summarised here see abstract · vs Placebo or no medication	Abstinence at the end of treatment was no more likely with THC preparations (RR 1.04, 95% CI 0.71 to 1.52; 4 studies, 290 people; moderate certainty) or N-acetylcysteine than with placebo. There was probably little difference in adverse events with THC preparations (RR 1.05, 0.88 to 1.26). The authors say no medicine is approved for cannabis use disorder and that all those tested remain experimental. No medicine shows clear benefit for cannabis use disorder Methods, limits and funding Outcomes: Abstinence, withdrawal and craving, adverse events Limitations: Medicines and outcomes differed, which limited pooling; certainty often low. Funding and conflicts: National Institute for Health and Care Research (UK) and the National Health and Medical Research Council (Australia) supported the authors. Eleven studies received study medicines from a manufacturing company. Read: abstract
Swartz 2025 Cannabinoid Hyperemesis Syndrome, 2016 to 2022. JAMA network open · PMID 41284293 · DOI	Cross sectional	188610906 188,610,906 US emergency department visits (people 12 and older), 2016 to 2022	Cannabis-related diagnoses with cyclic vomiting (a proxy for cannabis hyperemesis syndrome) Not summarised here see abstract · vs see abstract	Cases coded as the CHS proxy rose from 4.4 per 100,000 visits, peaked at 33.1 in the second quarter of 2020 and stayed elevated at 22.3 in 2022. Risk was highest at ages 18 to 25 (RRR 3.59, 95% CI 3.38 to 3.81) and 26 to 35 (RRR 2.26, 2.14 to 2.40). Females had lower risk (RRR 0.92). Emergency visits for CHS rose sharply and stayed high Methods, limits and funding Outcomes: Rate of CHS-proxy visits per 100,000 visits over time Limitations: Uses diagnosis codes as a proxy; coding changes can affect counts. Funding and conflicts: Research support from a non-US government source (PubMed record). Read: abstract
Tudella 2026 Efficacy and safety of cannabinoid-based interventions for behavioral and cognitive symptoms in dementia: systematic review and meta-analysis. Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology · PMID 42704497 · DOI	Meta-analysis	334 People with Alzheimer disease or dementia in 9 randomised trials	Cannabinoid-based therapies Not summarised here see abstract · vs Placebo	Cannabinoids did not improve agitation (SMD -0.58, 95% CI -1.71 to 0.55), overall neuropsychiatric symptoms or thinking test scores. Overall adverse events were similar to placebo but sleepiness was more frequent (RR 2.03, 1.29 to 3.20). Certainty was moderate for behaviour outcomes and low for thinking. No benefit for agitation or thinking in dementia; more sleepiness Methods, limits and funding Outcomes: Agitation (CMAI), neuropsychiatric symptoms (NPI-NH), thinking (MMSE), adverse events Limitations: Small total number of people; trials differed a lot. Funding and conflicts: Not stated in the abstract. Read: abstract
U.S. Department of Justice / 2026 Federal Register: Schedules of Controlled Substances: Rescheduling of Food and Drug Administration Approved Products Containing Marijuana From Schedule I to Schedule III (final rule, effective 28 April 2026) Federal Register · Source	Authoritative guidance n/a	US law	n/a	The abstract says the final rule (a final order) places drug products containing marijuana that have been approved by the FDA in schedule III of the Controlled Substances Act. It applies to marijuana, marijuana extracts and delta-9-THC derived from the marijuana plant to the extent they are in an FDA-approved drug product or covered by a state-issued medical marijuana licence, and creates an expedited registration process for state medical licence holders. Effective 28 April 2026. Methods, limits and funding Outcomes: US controlled-substance status for FDA-approved marijuana-containing products and state medical licence holders Limitations: Read the Federal Register abstract and part of the text; it does not by itself reschedule all marijuana. Read: abstract
U.S. Department of Justice / 2026 Federal Register: Schedules of Controlled Substances: Rescheduling of Marijuana (notice of hearing, 28 April 2026) Federal Register · Source	Authoritative guidance n/a	US law	n/a	DEA announced it would hold a hearing on the proposed rescheduling of marijuana into schedule III beginning 29 June 2026. The proposal was first published in May 2024. The abstract cites Executive Order 14370 and says DEA is completing the process as expeditiously as federal law allows. A companion notice (2026-08178) withdrew the earlier hearing notice. Methods, limits and funding Outcomes: Status of the broader proposal to move marijuana to schedule III Limitations: Notice of a hearing, not a final decision. We found no final rule on this broader proposal in a Federal Register search on 2026-10-25, but the search was not exhaustive. Read: abstract
U.S. National Library of Medicine / FDA label 2026 DailyMed: dronabinol capsules, prescribing information (revised April 2026) DailyMed · Source	Authoritative guidance n/a	Prescribing information for adults	Dronabinol (synthetic delta-9-THC) capsules	Dronabinol capsules are indicated in adults for anorexia with weight loss in AIDS and for chemotherapy nausea and vomiting in patients who have not responded adequately to conventional antinausea treatment. The label warns of psychiatric effects (it can worsen mania, depression or schizophrenia), cognitive impairment, hazardous activities such as driving, additive effects with sedatives and alcohol, low or high blood pressure, fainting and fast heart rate, seizures, risk of abuse, and paradoxical nausea, vomiting or abdominal pain with symptoms like cannabis hyperemesis syndrome. It may cause fetal harm and should be avoided in pregnancy. It is a Schedule III controlled substance. CYP2C9 and CYP3A4 are the main enzymes; the label notes possible displacement of highly protein-bound drugs such as warfarin. Methods, limits and funding Outcomes: Approved uses, warnings, interactions, pharmacokinetics and pregnancy/older-adult statements for dronabinol (synthetic delta-9-THC) Limitations: A label for a synthetic THC medicine, not for cannabis plant products. Our copy is the Ascend/Major version of the generic label; other makers' labels may differ in wording. Read: full-text
Varela 2026 Medical cannabis for the treatment of insomnia. The Cochrane database of systematic reviews · PMID 42863850 · DOI	Systematic review	537 Adults aged 18 to 76 with insomnia in 9 randomised trials	Medical cannabis (licensed plant-based and synthetic cannabinoids and pharmacy-prepared products) Not summarised here see abstract · vs Placebo (8 trials); amitriptyline (1)	Versus placebo, medical cannabis may have little to no clinically meaningful effect on sleep quality (SMD -0.49, 95% CI -1.06 to 0.08; very low certainty), total sleep time (7.7 minutes, -23.0 to 38.4; very low certainty) or daytime functioning (very low certainty). Adverse events were slightly more common (OR 11.25, 4.46 to 28.35; low certainty; 3 studies, 98 people). Little to no clear sleep benefit; more adverse events Methods, limits and funding Outcomes: Sleep quality, total sleep time, daytime functioning, adverse events Limitations: Few small trials; high inconsistency; very low certainty. Funding and conflicts: No dedicated funding (Cochrane review). Read: abstract
Vivolo-Kantor 2026 Trends in Emergency Department Visits Involving Cannabis Hyperemesis Syndrome Identified Using a New Diagnosis Code - United States, January 2023-May 2026. MMWR. Morbidity and mortality weekly report · PMID 42560855 · DOI	Cross sectional	199565 Emergency department visits in the CDC syndromic surveillance system, January 2023 to May 2026	Visits involving cannabis hyperemesis syndrome Not summarised here see abstract · vs see abstract	199,565 visits involved CHS in January 2023 to May 2026. The share rose from 3.35 per 10,000 ED visits in September 2025 to 11.26 in October 2025, when the new code began, and averaged 3.7 times the earlier level over the next eight months. Higher shares were seen at ages 15 to 24 and in females. The authors say the jump may partly reflect better recognition and coding rather than a true rise. CHS is common in emergency departments and likely under-counted Methods, limits and funding Outcomes: Share of ED visits involving CHS before and after a new diagnosis code (1 October 2025) Limitations: Surveillance data; coding change makes trends hard to read. Funding and conflicts: US CDC authors; no conflict statement in the abstract. Read: abstract
Weigel 2026 Pediatric Hemp-Derived Cannabinoid and Δ9-Tetrahydrocannabinol Exposures: A National Poison Data Study. The Journal of pediatrics · PMID 42025788 · DOI	Cross sectional	11836 Children aged 5 years or younger in the US National Poison Data System, 2023 to 2024: 9,667 delta-9-THC edible cases and 2,169 hemp-derived cannabinoid edible cases	Edibles containing delta-9-THC or hemp-derived cannabinoids (delta-8-THC and others) Not summarised here see abstract · vs see abstract	Hemp-derived cannabinoid cases had higher rates of respiratory depression (5.9% vs 4.6%) and low blood pressure (3.87% vs 2.97%), more critical care admission (11.3% vs 8.58%) and more overall admission (28.35% vs 26.13%) than delta-9-THC cases. No deaths were reported. Small children who eat THC edibles are often admitted; hemp-derived products were as bad or worse Methods, limits and funding Outcomes: Clinical effects, interventions, hospital admission Limitations: Reports are voluntary and amounts were not recorded. Funding and conflicts: Not stated in the abstract. Read: abstract
Whiting 2015 Cannabinoids for Medical Use: A Systematic Review and Meta-analysis. JAMA · PMID 26103030 · DOI	Meta-analysis	6462 Adults in 79 randomised trials of cannabinoids for nausea and vomiting from chemotherapy, HIV/AIDS appetite, chronic pain, spasticity (MS or paraplegia), depression, anxiety, sleep disorders, psychosis, glaucoma and Tourette syndrome	Cannabinoids (several products and routes pooled) Not summarised here Mostly short term · vs Placebo or other treatment	Of 79 trials (6,462 participants), 4 were at low risk of bias. Complete nausea and vomiting response was more common with cannabinoids than placebo (47% vs 20%; OR 3.82, 95% CI 1.55 to 9.42; 3 trials). Pain reduction: 37% vs 31% (OR 1.41, 0.99 to 2.00; 8 trials); average pain score on a 0 to 10 scale WMD -0.46 (-0.80 to -0.11; 6 trials); Ashworth spasticity scale WMD -0.36 (-0.69 to -0.05; 7 trials). More short-term adverse events, including serious ones (dizziness, dry mouth, nausea, fatigue, sleepiness, euphoria, disorientation, confusion, hallucination). Small average benefit for pain and spasticity; larger response for chemotherapy nausea in 3 old trials (See results) Methods, limits and funding Outcomes: Complete nausea and vomiting response; pain reduction; pain and spasticity scale scores; adverse events Limitations: Only 4 of 79 trials were at low risk of bias. Search ended April 2015, so it does not include newer trials. Products and conditions were pooled together. Funding and conflicts: Research support from a non-US government source (PubMed record); no conflict statement in the abstract. Read: abstract
Wilson 2026 The efficacy and safety of cannabinoids for the treatment of mental disorders and substance use disorders: a systematic review and meta-analysis. The lancet. Psychiatry · PMID 41856154 · DOI	Meta-analysis	2477 People in 54 randomised trials of cannabinoids as the main treatment for mental or substance use disorders	Cannabinoids (THC, CBD and combinations, various products) Not summarised here see abstract · vs Placebo	44% of trials were at high risk of bias; certainty was mostly low. A CBD plus delta-9-THC combination reduced cannabis withdrawal symptoms (SMD -0.29, 95% CI -0.57 to -0.02) and weekly cannabis use in people with cannabis use disorder, and reduced tic severity in Tourette syndrome (SMD -0.68, -1.03 to -0.34). Any cannabinoid increased sleep time in insomnia, and autistic traits fell. Cannabinoids increased cocaine craving. No significant effects were found for anxiety, anorexia nervosa, psychosis, PTSD or opioid use disorder; no randomised evidence was found for depression. All-cause adverse events were more common (OR 1.75, 1.25 to 2.46; number needed to harm 7); serious events and withdrawals were not. Some low-certainty signals (cannabis use disorder, tics, insomnia); no benefit seen for anxiety, PTSD, psychosis Methods, limits and funding Outcomes: Remission or symptom reduction; all-cause and serious adverse events Limitations: Many trials at high risk of bias; certainty mostly low. The authors say routine use for mental disorders is rarely justified. Funding and conflicts: National Health and Medical Research Council (Australia). PROSPERO CRD42023392718. Registration: CRD42023392718 Read: abstract
Zamarripa 2026 Impact of Cannabis Edibles Combined With Alcohol on Driving, Field Sobriety Performance, and Subjective Effects: A Within-Participant Crossover Trial. JAMA network open · PMID 42065887 · DOI	Controlled trial	25 25 healthy adults (mean age 25.6) who co-used cannabis and alcohol, in a seven-session double-blind crossover trial	THC brownies at two strengths with or without alcohol drinks (breath alcohol 0.05% or 0.08%) Not summarised here 7 sessions, 1 week apart · vs Placebo brownie and placebo drink	Compared with placebo, all active drug conditions except the lower THC dose alone negatively affected driving. Driving impairment and feeling intoxicated were often greater when cannabis and alcohol were combined than with either alone. Field sobriety tests did not detect several conditions in which driving was clearly worse. Alcohol did not change THC levels in blood. The authors say the 0.08% alcohol limit may be too liberal for someone who has co-used cannabis. Combining edible THC with alcohol worsened simulated driving Methods, limits and funding Outcomes: Simulated driving composite score, lane position variability, field sobriety tests, subjective effects Limitations: Small trial of healthy young adults who already co-used; simulator, not real roads. Funding and conflicts: Not stated in the abstract. ClinicalTrials.gov NCT04931095. Registration: NCT04931095 Read: abstract
Černe 2020 Toxicological properties of Δ9-tetrahydrocannabinol and cannabidiol. Arhiv za higijenu rada i toksikologiju · PMID 32597140 · DOI	Review mixed (human and animal)	Review of toxicology of THC and CBD	THC and CBD Not summarised here see abstract · vs see abstract	Cannabis contains more than 100 phytocannabinoids that can interact with CB1 and CB2 receptors; none is entirely CB1- or CB2-specific. THC is a partial agonist at both receptors, and its psychotomimetic effect comes mainly through CB1, which is strongly expressed in the central nervous system except the brain stem. THC and CBD also touch targets such as TRPV1, GPR55 and GPR119. The risk of lasting neuropsychological effects of THC needs more research. Methods, limits and funding Outcomes: Receptor actions and toxic effects Limitations: A narrative review. Funding and conflicts: Not stated in the abstract. Read: abstract

What we do not know yet

No long-term randomised trials of THC for any condition; most trials last 1 to 6 months.
Most benefit trials test prescription products (dronabinol, nabilone, THC and CBD spray), not smoked or edible shop cannabis.
Safety links (psychosis, dependence, pregnancy outcomes) come from observational studies and cannot prove cause.
Rates of psychosis, dependence and hyperemesis by product strength are not well known.
EMA and Health Canada were not read, ACOG advice was read only as reported by NIDA, the nabilone label was not found in DailyMed, and legal status in countries and US states was not checked.

How new research is added

New studies are found by a saved PubMed search and are placed in a review queue. A person checks each one before the page changes. This page was last updated 2026-10-25.

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 9

▼

Pain severity score

Lowered in trials

A rating of how bad pain is, usually on a 0 to 10 scale where 10 is the worst imaginable.

Low certainty

Results and what it means

Why it was measured

It is the main way pain trials judge whether a treatment helps.

Statistical vs. clinical meaning

Small average changes. NCCIH says a similarly small pain benefit in an earlier review may be too small to be meaningful to patients.

Safety implication

More dizziness, sedation and nausea.

Study	Result	People	Product
⁸	Oral THC-only products: -0.78 points; THC and CBD mouth spray: -0.54 points; nabilone -1.59 and dronabinol -0.23 points, on a 0 to 10 scale.	25 trials, 2,303 adults with chronic pain	Cannabinoid products vs placebo, 1 to 6 months
⁵⁴	Average pain score WMD -0.46 (95% CI -0.80 to -0.11) on a 0 to 10 scale; 6 trials.	Adults with chronic pain	Cannabinoids vs placebo

^{8 54 31}

▼

Spasticity rating (patient 0 to 10 scale)

Lowered in trials

A score people give for how stiff or spastic their muscles feel, on a 0 to 10 scale.

Mixed findings

Results and what it means

Why it was measured

Spasticity is mainly a symptom people feel, so trials ask patients to rate it.

Statistical vs. clinical meaning

Patient scores improve; clinician measures in the 2024 trial did not.

Safety implication

Dizziness, sleepiness and falls risk.

Study	Result	People	Product
²⁰	THC:CBD extracts reduced patient-reported spasticity versus placebo: MD -0.82 (95% CI -1.24 to -0.40; 7 studies).	Adults with MS spasticity (27 trials in the review)	THC:CBD extracts vs placebo
⁶	Clinician-rated muscle tone did not differ: -0.23 with nabiximols vs -0.26 with placebo (P = 0.7152).	68 adults with MS spasticity	Nabiximols vs placebo, crossover

^{20 6}

▲

Complete response to chemotherapy (no vomiting, no rescue medicine)

Raised in trials

The share of people who have no vomiting and need no extra anti-nausea medicine during a set period after chemotherapy.

Low certainty

Results and what it means

Why it was measured

It is the standard way chemotherapy anti-nausea treatments are compared.

Statistical vs. clinical meaning

Better than placebo; not shown better than older active medicines.

Safety implication

Dizziness and sleepiness are more common.

Study	Result	People	Product
⁵⁴	Complete nausea and vomiting response 47% with cannabinoids vs 20% with placebo (OR 3.82, 95% CI 1.55 to 9.42; 3 trials).	Adults on chemotherapy	Cannabinoids vs placebo
⁹	Overall control better than placebo (RR 2.65, 95% CI 1.70 to 4.12) but no better than active comparators.	26 trials, adults on chemotherapy	Cannabinoids vs placebo or active drugs

^{54 9}

▲

Body weight in HIV/AIDS

Raised in trials

Body weight or weight gain over the trial period in people losing weight because of HIV/AIDS.

Low certainty

Results and what it means

Why it was measured

Weight loss in AIDS is dangerous, so weight gain is the key goal.

Statistical vs. clinical meaning

Statistically significant; certainty low.

Safety implication

More mental health symptoms in HIV (RD 0.05).

Study	Result	People	Product
²⁹	Weight gain SMD 0.57 (95% CI 0.22 to 0.92) vs placebo; megestrol gave more weight gain than dronabinol in one comparison.	People with HIV	Cannabinoids (e.g. dronabinol) vs placebo

²⁹

▲

"Feeling high" rating

Raised in trials

A rating (0 to 100 scale) volunteers give for how high they feel after a dose.

Moderate certainty

Results and what it means

Why it was measured

It shows how strongly the brain is affected and is used to test drugs that block THC.

Statistical vs. clinical meaning

A large, consistent effect, and strong support that the CB1 receptor mediates it.

Safety implication

The "high" comes with anxiety, impairment and dizziness.

Study	Result	People	Product
¹⁷	THC increased feelings of being high compared with placebo (p < 0.001) in 41 experiments.	1,020 healthy adults with infrequent cannabis use	THC vs placebo
¹⁸	A CB1 blocker reduced the "feeling high" rating by up to 82.8% (95% CI -91.0 to -67.2%) vs placebo.	Healthy adults given THC	THC with the blocker selonabant vs placebo

^{17 18}

↕

Total sleep time

Mixed results in trials

How long a person sleeps in a night, measured by a device or diary.

Very low certainty

Results and what it means

Why it was measured

It is a basic measure of whether a treatment helps insomnia.

Statistical vs. clinical meaning

The two reviews disagree; the Cochrane result is not clearly different from zero.

Safety implication

More adverse events in the Cochrane analysis.

Study	Result	People	Product
⁵¹	Total sleep time +7.7 minutes (95% CI -23.0 to 38.4) with medical cannabis vs placebo; very low certainty.	6 trials, 194 adults with insomnia	Medical cannabis vs placebo
⁵⁵	Sleep time by device SMD 0.54 (0.14 to 0.95) and by diary SMD 0.55 (0.01 to 1.09) with any cannabinoid.	People with insomnia in the 54-trial review	Cannabinoids vs placebo

^{51 55}

▼

Tic severity score

Lowered in trials

A rating of how often and how strong tics are.

Low certainty

Results and what it means

Why it was measured

It is the main goal in Tourette trials.

Statistical vs. clinical meaning

Statistically significant but based on very few people.

Safety implication

THC side effects apply.

Study	Result	People	Product
⁵⁵	Tic severity SMD -0.68 (95% CI -1.03 to -0.34) with CBD plus delta-9-THC vs placebo.	People with tic disorders or Tourette syndrome	CBD and delta-9-THC combination vs placebo

⁵⁵

▼

Cannabis withdrawal symptoms score

Lowered in trials

A score of symptoms such as irritability, poor sleep and low appetite after stopping heavy cannabis use.

Mixed findings

Results and what it means

Why it was measured

It shows how hard stopping is and whether a medicine eases it.

Statistical vs. clinical meaning

Small reduction in symptoms; no gain in people stopping.

Safety implication

THC dependence concerns apply.

Study	Result	People	Product
⁵⁵	Withdrawal symptoms SMD -0.29 (95% CI -0.57 to -0.02) and weekly grams of cannabis -1.00 (-1.69 to -0.30) with CBD plus THC vs placebo.	People with cannabis use disorder	CBD and delta-9-THC combination vs placebo
⁴⁵	Abstinence at end of treatment RR 1.04 (0.71 to 1.52) with THC preparations vs placebo; moderate certainty.	290 people, 4 studies	THC preparations vs placebo

^{55 45}

●

Agitation score (CMAI)

Not changed in trials

A caregiver-rated score of agitated behaviours such as pacing, shouting or aggression in dementia.

Evidence against

Results and what it means

Why it was measured

It is the main measure in dementia agitation trials.

Statistical vs. clinical meaning

No improvement.

Safety implication

More sleepiness.

Study	Result	People	Product
⁴⁷	SMD -0.58 (95% CI -1.71 to 0.55): no clear change. Sleepiness RR 2.03 (1.29 to 3.20).	9 trials, 334 people with dementia	Cannabinoids vs placebo

⁴⁷

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

Prescription THC medicines are separate from cannabis from a shop or dispensary. FDA has not approved the cannabis plant for any disease. It has approved prescription medicines: dronabinol (synthetic THC, as Marinol and Syndros) and nabilone (a THC-like drug, as Cesamet). A THC and CBD mouth spray (nabiximols, Sativex) is licensed in the UK for multiple sclerosis spasticity. These medicines come with a label, a set amount and medical supervision. Smoked, vaped or edible cannabis does not. Authoritative guidance ^{12 31 50 38}
Psychosis and mental health. Heavy cannabis use is linked to psychosis. A meta-analysis of 66,816 people found the heaviest users had about 3.9 times the odds of psychosis outcomes versus non-users (OR 3.90, 95% CI 2.84 to 5.34). In a European study, daily use carried 3.2 times the odds of a first psychotic disorder and daily use of high-potency cannabis (THC 10% or more) nearly 4.8 times. These are links, not proof of cause, but the pattern grows with use. THC itself can cause anxiety and panic, and in lab experiments it raised anxiety in people who rarely use cannabis. Both the UK Sativex label and the dronabinol label say to avoid or not use THC medicines in people with psychotic illness. Meta-analysis ^{26 11 17 44 38 50 32}
Cannabis use disorder (dependence). About 22% of people who use cannabis (95% CI 18 to 26%) meet criteria for a cannabis use disorder, and the risk of dependence was 33% among young people who use weekly or daily. NIDA gives a range of 22% to 30% and links higher THC strength to a greater chance of progressing to the disorder. Withdrawal can happen when stopping. No medicine is approved for cannabis use disorder and the medicines tested in trials are still experimental. Meta-analysis ^{24 32 45 23}
Cannabis hyperemesis syndrome (CHS). Long-term heavy cannabis use can cause repeated attacks of severe nausea, vomiting and belly pain. People may relieve symptoms with hot showers, but the NIDA page says it only resolves when use stops completely. Emergency visits for CHS rose sharply in the US from 2016 to 2022 and stayed high, and a 2026 CDC report found a likely under-count. The dronabinol label warns that even synthetic THC can cause similar nausea, vomiting or belly pain in some patients. Systematic review ^{40 46 52 32 50}
Driving and impairment. Cannabis impairs driving. A meta-analysis of 57 experiments found cannabis alone impaired lane control, and cannabis with alcohol was worse than either alone. Studies of real crashes find a modest rise in risk (about 1.28 times in one model). A blood THC level does not reliably show how impaired someone is, so a person can be impaired even if the number looks low. In a trial, THC edibles combined with alcohol worsened simulated driving, and standard roadside tests missed several of the impaired conditions. The UK and US medicine labels tell patients not to drive until they know how the medicine affects them. Meta-analysis ^{41 37 4 56 38 50}
Pregnancy and breastfeeding. Use in pregnancy is linked to lower birth weight (RR 2.06 for under 2,500 g), small-for-gestational-age babies (RR 1.61), preterm birth (RR 1.28) and NICU admission (RR 1.38). A larger look at children's development found a small possible rise in ADHD (OR 1.13) and later cannabis use (OR 1.20), and no clear link with autism, anxiety or depression. THC passes into breast milk and builds up with repeated use. The dronabinol label says to avoid it in pregnancy and the UK Sativex label says not to use it while breastfeeding. NIDA reports that ACOG recommends counselling against use when trying to get pregnant, during pregnancy and while breastfeeding. Meta-analysis ^{25 3 21 50 38 32}
Teens and young adults. Cannabis use before age 18 is linked to depression in young adulthood (OR 1.37), suicidal ideas (OR 1.50) and suicide attempts (OR 3.46). Frequent or dependent use in youth is linked to about a 2-point IQ decline. NIDA reports regular heavy use in adolescence is linked to poorer working memory and school performance, and NCCIH says adolescents are four to seven times more likely than adults to develop cannabis use disorder. A 2023 umbrella review says convincing evidence supports avoiding cannabis in adolescence and early adulthood. Meta-analysis ^{16 35 44 32 31}
Older adults. Older adults may be more sensitive to the thinking and blood pressure effects of THC, and dizziness on standing raises fall risk. NCCIH links cannabis use to injury in older adults. Possible interactions with warfarin, opioids and benzodiazepines are noted by NIDA. One uncontrolled clinic study of 2,736 people over 65 reported mostly good results and few side effects (dizziness 9.7%, dry mouth 7.1%), but it had no control group and is weak evidence. Authoritative guidance ^{50 31 32 1}
Accidental swallowing by children (edibles). Edibles that look like sweets can seriously poison young children. In a study of 138 children and teens seen in emergency departments, 38% needed intensive care; eating edibles and mixing substances were the strongest predictors of severe outcomes, and every child aged 10 or younger had eaten edibles. In US poison data for children 5 and under, hemp-derived (delta-8 and similar) edibles led to similar or more severe outcomes than delta-9-THC edibles. Keep all THC products locked up and out of sight. Cohort study ^{10 53 32}
Lungs and heart. Smoking cannabis is linked to cough, phlegm and wheeze (low-strength evidence). Ever-use of cannabis was linked to a higher combined risk of cardiovascular events (RR 1.48, 95% CI 1.16 to 1.90) though heart attack and stroke alone did not reach significance. Dronabinol and nabiximols labels warn of fast heart rate, blood pressure changes and fainting. Meta-analysis ^{14 39 50 38}
Alcohol and sedating medicines add to THC effects. THC adds to the effects of alcohol and other medicines that cause sleepiness or dizziness. The dronabinol label lists barbiturates, benzodiazepines, ethanol, lithium and opioids among medicines that can add to the effect. Combining THC with alcohol worsened driving more than either alone in a randomised trial. Authoritative guidance ^{50 56 41}
Product labels are often wrong. In a Colorado study of 277 products, only 56.7% of flower products were within 15% of their labelled THC (96.0% of concentrates were), and lab-measured THC was lower than labelled on average. In 12 edible packages from Mississippi, measured THC differed from the label by from 288 mg less to 5,491 mg more. This makes it hard to know how much THC you are getting. Product analysis ^{15 5}

Side effects 14

Effect	How often	Severity and recovery	Evidence
Feeling high, euphoria In volunteers who rarely use cannabis, THC raised "feeling high" ratings in 41 randomised experiments.	Reported in 24% of people taking dronabinol for chemotherapy nausea and 8% at the lower appetite-trial dose (dronabinol label)	mild to moderate Wears off	Authoritative guidance ^{50 17}
Dizziness Among the most common effects in trials (3% to 10% in dronabinol labelling). In an uncontrolled study of people over 65, 9.7% reported dizziness.	Not established	mild to moderate Usually reversible	Systematic review ^{27 8 1 50}
Sleepiness, tiredness Moderate or large increases in sedation in pain trials of THC-containing products.	Not established	mild to moderate Usually reversible	Systematic review ^{8 27 17}
Anxiety, panic, paranoia Anxiety rose in volunteers who rarely use cannabis. Paranoid reaction is one of the most common adverse reactions (3% to 10%) in the dronabinol label.	Not established	mild to severe Usually reversible	Meta-analysis ^{17 50 38}
Thinking and memory problems, confusion Daily or frequent youth use was linked to a small IQ decline.	Dronabinol label: "thinking abnormal" in 3% to 10%	mild to moderate Reversible in most; long-term effects need more research	Authoritative guidance ^{50 35}
Nausea, vomiting, belly pain Nausea and vomiting can be both a use and a side effect of THC.	Among the most common events (3% to 10% in dronabinol labelling)	mild to severe Usually reversible; repeated severe vomiting in long-term heavy users may be CHS	Authoritative guidance ^{50 40}
Dry mouth 7.1% in an uncontrolled study of people over 65.	Not established	mild Reversible	Systematic review ^{54 1}
Fast heart rate, blood pressure changes, fainting NCCIH notes dizziness on standing raises the risk of fainting and falls.	Not given in the sources	mild to serious Usually reversible	Authoritative guidance ^{50 38 31}
Hallucinations, transient psychotic reactions UK label: stop treatment immediately and monitor until symptoms resolve.	Not established	moderate to serious Usually reverse when the medicine is stopped (labels)	Authoritative guidance ^{38 50 44}
Psychosis, depression, suicidal thoughts (long-term risk) Associations from observational studies; not proof of cause.	Not established	serious Not always reversible	Meta-analysis ^{26 11 16 44}
Dependence, cannabis use disorder See the main safety points.	About 22% of people who use cannabis	moderate to severe Treatable; behavioural support is the main approach	Meta-analysis ^{24 32}
Cannabis hyperemesis syndrome May need hospital care for dehydration.	Rare but rising emergency visits (4.4 to 33.1 per 100,000 ED visits at the 2020 peak, 22.3 in 2022)	serious Resolves when use stops completely (NIDA)	Cross sectional ^{46 52 40}
Cough, phlegm, wheeze (when smoked) Low-strength evidence.	RR 2 to 4 in different study types	mild to moderate Not well studied	Meta-analysis ¹⁴
Vision effects, central nervous system and psychological effects as a group High-certainty trial evidence for cannabis-based medicines: central nervous system events (equivalent OR 2.84), psychological effects (3.07), vision (3.00).	Not established	mild to moderate Reversible	Systematic review ⁴⁴

Liver

LiverTox (NIH) says marijuana use has not been linked to serum enzyme elevations during therapy or to clinically apparent liver injury with jaundice. Its reference list does describe older reports of mild, non-specific liver test abnormalities in frequent users, and studies in chronic hepatitis C in which daily cannabis use went with more liver scarring (fibrosis) and fat in the liver. The dronabinol label lists raised liver enzymes as an uncommon adverse reaction (1% or less in trials). This is different from CBD, which has been linked to raised liver enzymes in trials of the CBD prescription medicine. ^{33 50 31}

How often. No rate of liver injury from THC could be found. Cases are rare or absent in LiverTox.

Monitoring. No agency advice on liver monitoring for THC was read. People with hepatitis C or other liver disease should ask a clinician before daily use.

p37023245. LiverTox overview: no link to clinically apparent liver injury; older reports of mild liver test abnormalities in frequent users.
s-dailymed-dronabinol. Dronabinol label lists "hepatic enzyme elevation" among less common adverse reactions (incidence 1% or less).

This is a safety connection. It is not evidence that THC (Delta-9-Tetrahydrocannabinol) supports liver health.

Medicines and supplements 7

With	Why it matters	Importance	Evidence	What to do
Alcohol recreational	Both slow thinking and movement and add to each other.	In a randomised crossover trial of 25 people, THC edibles combined with alcohol worsened simulated driving and subjective intoxication more than either alone, and field sobriety tests missed several impaired conditions. A meta-analysis found cannabis plus alcohol worse than either alone.	Controlled trial Moderate certainty ^{56 41 50}	Do not combine, and do not drive.
Sedating medicines: benzodiazepines, barbiturates, opioids, lithium and other central nervous system depressants prescription	Additive sleepiness, dizziness and confusion.	The dronabinol label lists these as medicines that may add to the effect; NIDA notes possible interactions with benzodiazepines and opioids in older adults.	Authoritative guidance Low certainty ^{50 32}	Tell your prescriber and pharmacist about cannabis use.
Warfarin and other highly protein-bound medicines (for example cyclosporine, amphotericin B) prescription	THC is about 97% bound to blood proteins and could displace other drugs.	The dronabinol label says to monitor for adverse reactions to these drugs when starting or increasing THC. NIDA names warfarin among medicines that cannabis may interact with.	Authoritative guidance Low certainty ^{50 32}	Ask a pharmacist; blood tests may be needed.
Medicines that block or boost CYP2C9 and CYP3A4 liver enzymes prescription	THC is broken down mainly by CYP2C9 (and CYP3A4) in the liver.	CYP2C9 was the main enzyme depleting THC and forming 11-OH-THC in liver tests. The dronabinol label says inhibitors and inducers of CYP2C9 and CYP3A4 may change THC exposure.	Laboratory research Low certainty ^{34 50}	Ask a pharmacist if you take a medicine known to interact with these enzymes.
Medicines cleared by CYP2C9, CYP2C19 or CYP1A2 enzymes prescription	Test-tube and some clinical work suggest THC and other cannabinoids may inhibit CYP2C9, CYP2D6 and CYP2C19, and smoking cannabis may induce CYP1A2.	A review concludes drugs that depend mainly on CYP2C19, CYP2C9 and CYP1A2 may be at risk of altered levels. Mostly test-tube data.	Systematic review Low certainty ³⁶	Ask a pharmacist.
Medicines that raise or lower heart rate or blood pressure prescription	THC causes fast heart rate, and sometimes low blood pressure on standing.	The dronabinol label lists additive cardiac effects such as low or high blood pressure, fainting and fast heart rate with similar drugs.	Authoritative guidance Low certainty ^{50 38}	Tell your doctor if you have heart disease.
Hormonal contraceptives prescription	Not explained in the label text we read.	The UK Sativex label says it may reduce the effectiveness of hormonal contraceptives; it did not specify for THC alone.	Authoritative guidance Not enough evidence ³⁸	Ask a clinician about additional contraception if on this medicine.

Who should check first 11

Situation	Advice	Evidence
Personal or family history of schizophrenia or other psychosis, or severe mood or personality disorder	The UK Sativex label contraindicates the medicine in these patients. The dronabinol label says to avoid it in people with a psychiatric history. Heavy and high-potency use is linked to higher psychosis risk.	Authoritative guidance ^{38 50 11}
Pregnancy or trying to conceive	Avoid. The dronabinol label says it may cause fetal harm and to avoid it in pregnancy. Use in pregnancy is linked to lower birth weight and preterm birth.	Meta-analysis ^{50 25 32}
Breastfeeding	Avoid. The UK Sativex label contraindicates use while breastfeeding, and THC passes into breast milk and builds up with repeated use.	Authoritative guidance ^{38 21 32}
Teenagers and young adults	Avoid recreational use. Evidence links use in adolescence to depression, suicidal behaviour, lower IQ and cannabis use disorder. The dronabinol label says its safety and effectiveness in children have not been established.	Meta-analysis ^{16 35 32 50}
Adults 65 and over	Use extra caution: more sensitive to thinking and blood pressure effects, higher risk of falls, possible drug interactions.	Authoritative guidance ^{50 31 32}
Heart disease or cardiac disorders	The UK label does not recommend Sativex in serious cardiovascular disease. The dronabinol label warns of low or high blood pressure, fainting and fast heart rate in patients with cardiac disorders.	Authoritative guidance ^{38 50 39}
History of seizures or epilepsy	The dronabinol label says to weigh risk versus benefit and to stop if seizures occur. The UK label says to take caution.	Authoritative guidance ^{50 38}
History of substance misuse or cannabis use disorder	The dronabinol label says to assess the risk of abuse or misuse in people with a history of substance abuse and to monitor them.	Authoritative guidance ^{50 24}
Allergy to sesame oil (dronabinol capsules)	Contraindicated: the capsules contain sesame oil.	Authoritative guidance ⁵⁰
Driving or using machinery	Do not drive while impaired. Do not drive after using a THC medicine until you know how it affects you.	Authoritative guidance ^{38 50 41}
Homes with young children	Store every THC product locked away. Edibles can look like sweets and have caused serious poisoning in young children.	Cohort study ^{10 53}

Overdose, product quality and regulators

Summary. THC is intoxicating. Most serious harms in the research are linked to heavy or frequent use, high-potency products, edibles eaten by children, driving while impaired, and use in people with mental illness, in pregnancy or in youth.

Psychosis and mental health. Heavy use: OR 3.90 for psychosis outcomes (66,816 people). Daily high-potency use: OR 4.8 in a European case-control study. Adolescent use: depression OR 1.37, suicidal ideas OR 1.50, attempts OR 3.46. These are associations.

Dependence. About 22% of people who use cannabis meet criteria for a use disorder (95% CI 18 to 26%); 33% (22 to 44%) among young people who use weekly or daily.

Cannabis hyperemesis syndrome. Repeated attacks of severe nausea, vomiting and belly pain after long-term heavy use; stopping use is the treatment that works best.

Driving and impairment. Impairs lane control and speed; worse with alcohol; blood THC does not reliably track impairment (10 of 12 studies found no linear relationship).

Adolescents. Linked to depression, suicidal behaviour and a small IQ decline; NCCIH says adolescents are four to seven times more likely than adults to develop cannabis use disorder.

Pregnancy and breastfeeding. Linked to low birth weight, small size for age, preterm birth and NICU admission; THC passes into breast milk.

Older adults. More sensitive to thinking and blood pressure effects; fall risk; interaction risk.

Children accidental ingestion. Of 138 children and teens seen in emergency departments with cannabis intoxication, 38% were admitted to intensive care and one died. Edibles and mixed substances predicted severe outcomes. Among children 5 and under in poison data (2023 to 2024), 28.35% of hemp-derived cannabinoid cases and 26.13% of delta-9-THC cases were admitted.

Overdose response. Large amounts cause anxiety, panic, fast heart rate, confusion and psychosis-like reactions; a UK label notes a transient toxic psychosis in 3 of 41 people given a very large multiple of the recommended amount of the THC and CBD spray, which resolved after stopping. The UK label says treatment of overdose should be symptomatic and supportive. A research drug that blocks the CB1 receptor reduced THC effects in volunteers, but it is not an approved treatment.

Genotoxicity. Not reviewed.

THC products sold outside pharmacies are not held to the standards of approved medicines. Labels can be off, and some products contain unexpected cannabinoids such as delta-8-THC. FDA has not approved any cannabis or CBD product other than the prescription medicines named above.

Label accuracy of flower and concentrates. A study of 277 products from 52 Colorado dispensaries found 96.0% of concentrates were within 15% of the labelled THC but only 56.7% of flower; measured THC was lower than labelled in both. ¹⁵

Label accuracy of edibles. Twelve edible packages from Mississippi differed from the label by from 288 mg less to 5,491 mg more THC (delta-8 and delta-9); three products from one brand differed by 1,542, 719 and 5,491 mg. ⁵

Delta-8 and other hemp-derived THC-like products. NIDA says delta-8 THC products are not FDA-approved for safe use and that eating large amounts has caused medical emergencies, including in children. ^{32 53}

Body	Action	Date
US Food and Drug Administration (United States)	Has not approved a marketing application for cannabis for any disease. Has approved Epidiolex (cannabidiol) and three synthetic cannabis-related prescription drugs: Marinol and Syndros (dronabinol) and Cesamet (nabilone). Has not approved other cannabis or CBD products on the market. source	2023-02-24
US Food and Drug Administration (drug label) (United States)	Dronabinol capsules are indicated in adults for anorexia with weight loss in AIDS and for chemotherapy nausea and vomiting after conventional treatment has failed. The label says dronabinol is a Schedule III controlled substance. source	2026-04
US Department of Justice / Drug Enforcement Administration (United States)	Final rule: drug products containing marijuana approved by the FDA, and marijuana subject to a state medical marijuana licence, are placed in schedule III of the Controlled Substances Act, with an expedited registration process for state licence holders. source	2026-04-28
US Department of Justice / Drug Enforcement Administration (United States)	Notice of a hearing beginning 29 June 2026 on the proposal to move marijuana to schedule III. The earlier hearing notice from 2024 was withdrawn. We found no final decision on this broader proposal on 2026-10-25. source	2026-04-28
NIH National Center for Complementary and Integrative Health (reporting FDA) (United States)	Says FDA has determined that products containing THC or CBD cannot be sold legally as dietary supplements, and foods with added THC or CBD cannot be sold legally in interstate commerce; state law governs sales within a state. source	2019-11
UK medicines licence for Sativex (nabiximols), via the Electronic Medicines Compendium (United Kingdom)	Licensed for symptom improvement in adults with moderate to severe MS spasticity who have not responded to other treatment and who show benefit in an initial trial; specialist initiation; contraindicated with psychotic illness history and in breastfeeding; driving warnings. source	2026-04-02

When to get medical help

A child or adult who swallowed a THC product and is very sleepy, floppy, hard to wake, breathing slowly or shallowly, having a seizure, or vomiting repeatedly: call emergency services or poison control (in the US, Poison Help 1-800-222-1222) now. Children who eat edibles can need intensive care.
Repeated severe vomiting and belly pain in someone who uses cannabis often: get medical care. Dehydration can be serious and stopping cannabis is the treatment that works.
Hallucinations, strong paranoia, confusion that does not settle, or thoughts of harming yourself: get urgent help. In the US you can call or text 988.
Chest pain, fainting, or a racing heart that will not slow down.
Pregnancy: tell your care team about any cannabis use so they can support you; do not stop care.

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 THC?

THC (delta-9-tetrahydrocannabinol) is the compound in cannabis that causes the "high". It acts on CB1 and CB2 cannabinoid receptors, mostly CB1. The plant makes it from an acid form called THCA.

^{32 57 43}

Is THC approved as a medicine?

Not the plant. FDA has approved prescription dronabinol (synthetic THC) and nabilone (a THC-like drug) for chemotherapy nausea, and dronabinol for appetite loss in AIDS. In the UK a THC and CBD mouth spray is licensed for MS stiffness. These are prescription medicines and different from shop cannabis.

^{12 50 38}

Does THC help with pain?

Reviews of short trials find a small average pain reduction with some THC products, with more dizziness and sleepiness. The 2026 Cochrane review of nerve pain found no clear evidence of benefit for THC-dominant medicines. Talk to a doctor about pain.

^{8 2}

Does THC help nausea or muscle stiffness?

For chemotherapy nausea, prescription cannabinoids beat placebo in mostly old trials and may help as an add-on. For MS stiffness, people rate it better on the THC and CBD spray, but a clinician-measured trial found no difference.

^{9 7 13 6}

Can THC cause psychosis or mental health problems?

Heavy use is linked to psychosis (about 3.9 times the odds for the heaviest users) and daily high-potency use about 4.8 times in one study. THC can also cause anxiety and panic. These are links, not proof of cause. Medicine labels advise against use in people with a history of psychosis.

^{26 11 17 38}

Is THC addictive?

About 22% of people who use cannabis meet criteria for a cannabis use disorder, and 33% of young people who use weekly or daily. Withdrawal can occur. No medicine is approved for cannabis use disorder.

^{24 45}

What is cannabis hyperemesis syndrome?

Repeated attacks of severe nausea, vomiting and belly pain after long-term heavy cannabis use. Emergency visits for it rose from 2016 to 2022. Stopping cannabis is the treatment that works best, so see a clinician.

^{40 46 32}

Is it safe to drive after using THC?

No. Cannabis impairs lane control, and cannabis with alcohol is worse than either alone. A blood THC level does not reliably show how impaired a person is.

^{41 4 56}

Is THC safe in pregnancy or when breastfeeding?

Use in pregnancy is linked to low birth weight, preterm birth and NICU admission. THC passes into breast milk. The dronabinol label says to avoid it in pregnancy and the UK spray label says not to use it while breastfeeding.

^{25 21 50 38}

What about teens and older adults?

In teens, use is linked to depression, suicidal behaviour and a small IQ drop with frequent use. Older adults are more sensitive to dizziness and thinking effects and may have drug interactions.

^{16 35 50 32}

What if a child eats a THC edible?

Treat it as an emergency. Edibles have caused serious poisoning and intensive care admissions in young children. Call poison control (in the US, 1-800-222-1222) or emergency services, and keep THC products locked away.

^{10 53}

Are the THC amounts on labels accurate?

Often not. In a Colorado study, only 56.7% of flower products were within 15% of their labelled THC. A small study of edibles found the THC amount far from the label.

^{15 5}

Explore by health goal

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 abstracts of the 49 papers cited, read the DailyMed dronabinol label, NCCIH, NIDA, FDA, the UK Sativex label and two Federal Register notices on 2026-10-25, and confirmed the chemical name and compound IDs in PubChem. Lab numbers come from state regulator datasets and published papers, not from our own testing. Certainty ratings follow the reviewers' own GRADE where given; others are the compilers' reading. Only abstracts were read for papers.
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.

Sources 57

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

Abuhasira R, Schleider LB, Mechoulam R, et al. Epidemiological characteristics, safety and efficacy of medical cannabis in the elderly.. European journal of internal medicine. 2018. PMID 29398248. DOI 10.1016/j.ejim.2018.01.019. (abstract)
Ateş G, Welsch P, Klose P, et al. Cannabis-based medicines for chronic neuropathic pain in adults.. The Cochrane database of systematic reviews. 2026. PMID 41548880. DOI 10.1002/14651858.CD012182.pub3. (abstract)
Bassalov H, Yakirevich-Amir N, Reuveni I, et al. Prenatal cannabis exposure and the risk for neuropsychiatric anomalies in the offspring: a systematic review and meta-analysis.. American journal of obstetrics and gynecology. 2024. PMID 38908654. DOI 10.1016/j.ajog.2024.06.014. (abstract)
Behzad D, Zhao S, Besa R, et al. Association of driving with blood delta-9-tetrahydrocannabinol: a systematic review.. The international journal of neuropsychopharmacology. 2025. PMID 40172477. DOI 10.1093/ijnp/pyaf021. (abstract)
Beneke LL, Kyle PB, Benton JB, et al. Dazed and confused: variability in reported and measured tetrahydrocannabinol content in cannabis edibles.. Clinical toxicology (Philadelphia, Pa.). 2025. PMID 41025251. DOI 10.1080/15563650.2025.2559183. (abstract)
Bethoux FA, Farrell R, Checketts D, et al. A randomized, double-blind, placebo-controlled trial to evaluate the effect of nabiximols oromucosal spray on clinical measures of spasticity in patients with multiple sclerosis.. Multiple sclerosis and related disorders. 2024. PMID 39106541. DOI 10.1016/j.msard.2024.105740. (abstract)
Braun IM, Bohlke K, Abrams DI, et al. Cannabis and Cannabinoids in Adults With Cancer: ASCO Guideline.. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2024. PMID 38478773. DOI 10.1200/JCO.23.02596. (abstract)
Chou R, Fu R, Ahmed AY, et al. Cannabis-Based Products for Chronic Pain : An Updated Systematic Review.. Annals of internal medicine. 2026. PMID 41429020. DOI 10.7326/ANNALS-25-03152. (abstract)
Chow R, Basu A, Kaur J, et al. Efficacy of cannabinoids for the prophylaxis of chemotherapy-induced nausea and vomiting-a systematic review and meta-analysis.. Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer. 2025. PMID 39953210. DOI 10.1007/s00520-025-09251-w. (abstract)
Cohen N, Mathew M, Brent J, et al. Severe outcomes following pediatric cannabis intoxication: a prospective cohort study of an international toxicology surveillance registry.. Clinical toxicology (Philadelphia, Pa.). 2023. PMID 37603042. DOI 10.1080/15563650.2023.2238121. (abstract)
Di Forti M, Quattrone D, Freeman TP, et al. The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI): a multicentre case-control study.. The lancet. Psychiatry. 2019. PMID 30902669. DOI 10.1016/S2215-0366(19)30048-3. (abstract)
FDA. FDA: FDA and Cannabis: Research and Drug Approval Process (content current as of 24 Feb 2023). FDA. 2023. (full-text)
Filippini G, Minozzi S, Borrelli F, et al. Cannabis and cannabinoids for symptomatic treatment for people with multiple sclerosis.. The Cochrane database of systematic reviews. 2022. PMID 35510826. DOI 10.1002/14651858.CD013444.pub2. (abstract)
Ghasemiesfe M, Ravi D, Vali M, et al. Marijuana Use, Respiratory Symptoms, and Pulmonary Function: A Systematic Review and Meta-analysis.. Annals of internal medicine. 2018. PMID 29971337. DOI 10.7326/M18-0522. (abstract)
Giordano G, Brook CP, Ortiz Torres M, et al. Accuracy of labeled THC potency across flower and concentrate cannabis products.. Scientific reports. 2025. PMID 40592871. DOI 10.1038/s41598-025-03854-3. (abstract)
Gobbi G, Atkin T, Zytynski T, et al. Association of Cannabis Use in Adolescence and Risk of Depression, Anxiety, and Suicidality in Young Adulthood: A Systematic Review and Meta-analysis.. JAMA psychiatry. 2019. PMID 30758486. DOI 10.1001/jamapsychiatry.2018.4500. (abstract)
Goodwin I, Oliver D, Chesney E, et al. The subjective effects of △9-tetrahydrocannabinol: A systematic review and dose-response meta-regression.. Neuroscience and biobehavioral reviews. 2026. PMID 42242480. DOI 10.1016/j.neubiorev.2026.106795. (abstract)
Gorbenko AA, Heuberger JAAC, Juachon M, et al. CB1 Receptor Antagonist Selonabant (ANEB-001) Blocks Acute THC Effects in Healthy Volunteers: A Phase II Randomized Controlled Trial.. Clinical pharmacology and therapeutics. 2025. PMID 39898464. DOI 10.1002/cpt.3581. (abstract)
Grotenhermen F. Pharmacokinetics and pharmacodynamics of cannabinoids.. Clinical pharmacokinetics. 2003. PMID 12648025. DOI 10.2165/00003088-200342040-00003. (abstract)
Hamm-Buiar L, Gorski D, Kagueiama MH, et al. Efficacy and safety of cannabinoids in the treatment of spasticity in multiple sclerosis: A systematic review of randomized clinical trials.. British journal of clinical pharmacology. 2026. PMID 42543771. DOI 10.1002/bcp.70719. (abstract)
Holdsworth EA, Berim A, Gang DR, et al. Human Milk Cannabinoid Concentrations and Associations with Maternal Factors: The Lactation and Cannabis (LAC) Study.. Breastfeeding medicine : the official journal of the Academy of Breastfeeding Medicine. 2024. PMID 38695182. DOI 10.1089/bfm.2024.0021. (abstract)
Hsu M, Shah A, Jordan A, et al. Therapeutic Use of Cannabis and Cannabinoids: A Review.. JAMA. 2026. PMID 41296368. DOI 10.1001/jama.2025.19433. (abstract)
Kesner AJ, Lovinger DM. Cannabis use, abuse, and withdrawal: Cannabinergic mechanisms, clinical, and preclinical findings.. Journal of neurochemistry. 2021. PMID 33891706. DOI 10.1111/jnc.15369. (abstract)
Leung J, Chan GCK, Hides L, et al. What is the prevalence and risk of cannabis use disorders among people who use cannabis? a systematic review and meta-analysis.. Addictive behaviors. 2020. PMID 32485547. DOI 10.1016/j.addbeh.2020.106479. (abstract)
Marchand G, Masoud AT, Govindan M, et al. Birth Outcomes of Neonates Exposed to Marijuana in Utero: A Systematic Review and Meta-analysis.. JAMA network open. 2022. PMID 35084479. DOI 10.1001/jamanetworkopen.2021.45653. (abstract)
Marconi A, Di Forti M, Lewis CM, et al. Meta-analysis of the Association Between the Level of Cannabis Use and Risk of Psychosis.. Schizophrenia bulletin. 2016. PMID 26884547. DOI 10.1093/schbul/sbw003. (abstract)
McDonagh MS, Morasco BJ, Wagner J, et al. Cannabis-Based Products for Chronic Pain : A Systematic Review.. Annals of internal medicine. 2022. PMID 35667066. DOI 10.7326/M21-4520. (abstract)
Mechoulam R, Hanuš LO, Pertwee R, et al. Early phytocannabinoid chemistry to endocannabinoids and beyond.. Nature reviews. Neuroscience. 2014. PMID 25315390. DOI 10.1038/nrn3811. (abstract)
Mücke M, Weier M, Carter C, et al. Systematic review and meta-analysis of cannabinoids in palliative medicine.. Journal of cachexia, sarcopenia and muscle. 2018. PMID 29400010. DOI 10.1002/jcsm.12273. (abstract)
NCBI PubChem. PubChem: Tetrahydrocannabinol (delta-9-THC), CID 16078. PubChem REST API. 2026. (API)
NCCIH. NCCIH (NIH): Cannabis (Marijuana) and Cannabinoids: What You Need To Know (last updated November 2019). NCCIH. 2019. (full-text)
NIDA. NIDA (NIH): Cannabis (Marijuana) research topic page. NIDA. 2026. (full-text)
NIDDK. LiverTox: Marijuana (NIDDK, record updated 2023). LiverTox (NCBI Bookshelf NBK590051). 2023. PMID 37023245. (abstract)
Patilea-Vrana GI, Anoshchenko O, Unadkat JD. Hepatic Enzymes Relevant to the Disposition of (-)-∆9-Tetrahydrocannabinol (THC) and Its Psychoactive Metabolite, 11-OH-THC.. Drug metabolism and disposition: the biological fate of chemicals. 2019. PMID 30567877. DOI 10.1124/dmd.118.085548. (abstract)
Power E, Sabherwal S, Healy C, et al. Intelligence quotient decline following frequent or dependent cannabis use in youth: a systematic review and meta-analysis of longitudinal studies.. Psychological medicine. 2021. PMID 33501901. DOI 10.1017/S0033291720005036. (abstract)
Qian Y, Gurley BJ, Markowitz JS. The Potential for Pharmacokinetic Interactions Between Cannabis Products and Conventional Medications.. Journal of clinical psychopharmacology. 2019. PMID 31433338. DOI 10.1097/JCP.0000000000001089. (abstract)
Rogeberg O. A meta-analysis of the crash risk of cannabis-positive drivers in culpability studies-Avoiding interpretational bias.. Accident; analysis and prevention. 2019. PMID 30468948. DOI 10.1016/j.aap.2018.11.011. (abstract)
SVX Therapeutics / emc. Electronic Medicines Compendium (UK): Sativex oromucosal spray, Summary of Product Characteristics (text revised 2 April 2026). emc (UK). 2026. (full-text)
Sebastian SA, Shah Y, Arsene C, et al. Cannabis use and atherosclerotic cardiovascular disease outcomes: A meta-analysis of multinational cohort data.. Disease-a-month : DM. 2025. PMID 39800612. DOI 10.1016/j.disamonth.2024.101849. (abstract)
Senderovich H, Patel P, Jimenez Lopez B, et al. A Systematic Review on Cannabis Hyperemesis Syndrome and Its Management Options.. Medical principles and practice : international journal of the Kuwait University, Health Science Centre. 2022. PMID 34724666. DOI 10.1159/000520417. (abstract)
Simmons SM, Caird JK, Sterzer F, et al. The effects of cannabis and alcohol on driving performance and driver behaviour: a systematic review and meta-analysis.. Addiction (Abingdon, England). 2022. PMID 35083810. DOI 10.1111/add.15770. (abstract)
Simon L, Baldwin C, Kalea AZ, et al. Cannabinoid interventions for improving cachexia outcomes in cancer: a systematic review and meta-analysis.. Journal of cachexia, sarcopenia and muscle. 2022. PMID 34881518. DOI 10.1002/jcsm.12861. (abstract)
Sirikantaramas S, Taura F, Morimoto S, et al. Recent advances in Cannabis sativa research: biosynthetic studies and its potential in biotechnology.. Current pharmaceutical biotechnology. 2007. PMID 17691992. DOI 10.2174/138920107781387456. (abstract)
Solmi M, De Toffol M, Kim JY, et al. Balancing risks and benefits of cannabis use: umbrella review of meta-analyses of randomised controlled trials and observational studies.. BMJ (Clinical research ed.). 2023. PMID 37648266. DOI 10.1136/bmj-2022-072348. (abstract)
Spiga F, Parkhouse T, Tang VM, et al. Pharmacotherapies for cannabis use disorder.. The Cochrane database of systematic reviews. 2025. PMID 41025421. DOI 10.1002/14651858.CD008940.pub4. (abstract)
Swartz JA, Franceschini D. Cannabinoid Hyperemesis Syndrome, 2016 to 2022.. JAMA network open. 2025. PMID 41284293. DOI 10.1001/jamanetworkopen.2025.45310. (abstract)
Tudella GCN, Klostermann AU, Menegucci G, et al. Efficacy and safety of cannabinoid-based interventions for behavioral and cognitive symptoms in dementia: systematic review and meta-analysis.. Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology. 2026. PMID 42704497. DOI 10.1007/s10072-026-09347-z. (abstract)
U.S. Department of Justice / DEA. Federal Register: Schedules of Controlled Substances: Rescheduling of Food and Drug Administration Approved Products Containing Marijuana From Schedule I to Schedule III (final rule, effective 28 April 2026). Federal Register. 2026. (abstract)
U.S. Department of Justice / DEA. Federal Register: Schedules of Controlled Substances: Rescheduling of Marijuana (notice of hearing, 28 April 2026). Federal Register. 2026. (abstract)
U.S. National Library of Medicine / FDA label. DailyMed: dronabinol capsules, prescribing information (revised April 2026). DailyMed. 2026. (full-text)
Varela LB, Miño Zambrano JE, Maura Goncalves Costa B, et al. Medical cannabis for the treatment of insomnia.. The Cochrane database of systematic reviews. 2026. PMID 42863850. DOI 10.1002/14651858.CD016216.pub2. (abstract)
Vivolo-Kantor AM, Liu S, Tanz LJ, et al. Trends in Emergency Department Visits Involving Cannabis Hyperemesis Syndrome Identified Using a New Diagnosis Code - United States, January 2023-May 2026.. MMWR. Morbidity and mortality weekly report. 2026. PMID 42560855. DOI 10.15585/mmwr.mm7530a2. (abstract)
Weigel B, Wahl M. Pediatric Hemp-Derived Cannabinoid and Δ9-Tetrahydrocannabinol Exposures: A National Poison Data Study.. The Journal of pediatrics. 2026. PMID 42025788. DOI 10.1016/j.jpeds.2026.115113. (abstract)
Whiting PF, Wolff RF, Deshpande S, et al. Cannabinoids for Medical Use: A Systematic Review and Meta-analysis.. JAMA. 2015. PMID 26103030. DOI 10.1001/jama.2015.6358. (abstract)
Wilson J, Dobson O, Langcake A, et al. The efficacy and safety of cannabinoids for the treatment of mental disorders and substance use disorders: a systematic review and meta-analysis.. The lancet. Psychiatry. 2026. PMID 41856154. DOI 10.1016/S2215-0366(26)00015-5. (abstract)
Zamarripa CA, Lin S, Klausner M, et al. Impact of Cannabis Edibles Combined With Alcohol on Driving, Field Sobriety Performance, and Subjective Effects: A Within-Participant Crossover Trial.. JAMA network open. 2026. PMID 42065887. DOI 10.1001/jamanetworkopen.2026.9842. (abstract)
Černe K. Toxicological properties of Δ9-tetrahydrocannabinol and cannabidiol.. Arhiv za higijenu rada i toksikologiju. 2020. PMID 32597140. DOI 10.2478/aiht-2020-71-3301. (abstract)

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.