Medicinal Cannabis

Medicinal Cannabis:  the science, evidence, risks and UK prescribing landscape

New Evidence reviewed and updated to 22 August 2026

1. Medicinal cannabis in 2026: the short version

Cannabis-based medicinal products (CBPMs) have a legitimate but limited role in modern medicine. The strongest established UK indications are specific severe epilepsies treated with purified cannabidiol, multiple-sclerosis spasticity treated with nabiximols (THC:CBD spray) in appropriate patients, and refractory chemotherapy-induced nausea and vomiting treated with nabilone. Outside these areas, evidence is much more variable and many prescriptions are for unlicensed products after conventional treatments have been inadequate or unsuitable.

·       CBD is not simply “non-psychoactive cannabis”: it has distinct pharmacology, clinically important drug interactions, and can affect liver enzymes. Highly purified pharmaceutical CBD has robust evidence in selected epilepsies; this does not validate over-the-counter CBD products for other disorders.

·       THC is the principal intoxicating cannabinoid. It may reduce some symptoms in some patients, but higher THC exposure is also associated with impairment, anxiety or dysphoria in some people, dependence, and increased psychosis risk in susceptible individuals.

·       For chronic pain, trials and systematic reviews suggest small average improvements for some cannabinoid products, but adverse effects are common and NICE still advises against routine cannabis-based medicines for chronic pain outside trials.

·       For psychiatric and neurodevelopmental disorders — including ADHD, depression, anxiety disorders, PTSD and autism — evidence remains insufficient for routine first-line treatment. Some observational data are encouraging, but controlled evidence is limited, heterogeneous or low certainty.

·       Prescribed cannabis has been legal in the UK since November 2018, but NHS access remains very restricted. Private prescribing is substantially more common, although robust national patient counts are difficult to establish.

2. What is medicinal cannabis?

Cannabis contains hundreds of biologically active compounds, including phytocannabinoids and terpenes. The two best characterised cannabinoids are delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD). Medical cannabis may refer to licensed medicines, purified cannabinoids, synthetic cannabinoids, or unlicensed cannabis-based medicinal products manufactured to pharmaceutical standards.

3. The endocannabinoid system

Overview

The endocannabinoid system (ECS) is a widespread neuromodulatory network involved in synaptic signalling, stress responses, appetite, pain processing, immune signalling, memory, reward and motor control. Its principal endogenous ligands include anandamide (AEA) and 2-arachidonoylglycerol (2-AG). CB1 receptors are abundant in the central nervous system, while CB2 receptors are prominent in immune and peripheral tissues, although both occur more widely than this simple distinction suggests.

THC can activate cannabinoid receptors directly, especially CB1, which explains both potential therapeutic effects and intoxication-related effects. CBD is pharmacologically different: it does not act as a straightforward CB1 agonist and has multiple proposed molecular targets. Claims that CBD simply “corrects endocannabinoid deficiency” or that particular terpenes reliably determine clinical effects remain unproven in routine clinical practice.

4. UK law, access and prescribing

Cannabis-based products for medicinal use were moved into Schedule 2 of the Misuse of Drugs Regulations in November 2018. This made specialist prescribing lawful; it did not legalise recreational cannabis or make all cannabis/CBD products equivalent to medicines.

·       NHS prescribing: a cannabis-based medicine can be prescribed on the NHS by a specialist hospital doctor, or under specialist supervision. NHS access remains uncommon and is concentrated around defined indications such as severe epilepsy, MS spasticity and refractory chemotherapy-related nausea/vomiting.

·       Private prescribing: doctors on the GMC Specialist Register may prescribe CBPMs privately. NHS England states that specialist doctors in non-NHS settings are legally able to prescribe and should follow equivalent safeguards for unlicensed special medicines.

·       Treatment should be based on a documented clinical assessment, consideration of licensed/evidence-based alternatives, informed discussion of uncertainties and risks, and appropriate monitoring. Local clinic governance and multidisciplinary review processes may go beyond the minimum legal requirements.

·       Possession of cannabis remains unlawful unless it is lawfully prescribed. Patients should keep prescribed medication in its original dispensing packaging and retain supporting prescription/clinical documentation where appropriate.

·       Driving: a prescription is not permission to drive while impaired. In England, Scotland and Wales it is illegal to drive if a medicine impairs driving. A statutory medical defence may apply to specified drug limits when the medicine was prescribed and taken as directed, but it does not protect a driver who is actually impaired.

5. What conditions have evidence?

The quality of evidence differs dramatically by condition and product. The table below separates established uses from areas where evidence is suggestive, inconsistent or insufficient.

Visit medbud.wiki ↗

6. ADHD, autism and mental health: a more careful interpretation

ADHD

Cannabis is commonly discussed as “self-medication” in ADHD, but patient preference or subjective calming does not establish treatment efficacy. The principal controlled trial remains the small 2017 EMA-C study of nabiximols in 30 adults. It found no statistically significant benefit on the primary cognitive/activity outcome; some secondary symptom signals were reported, but they did not remain significant after correction for multiple comparisons. A 2024 scoping review still described the relationship between the ECS, cannabis use and ADHD symptoms as poorly understood.

This caution matters clinically because ADHD is associated with a higher prevalence of cannabis use disorder. Cannabis can also affect attention, short-term memory, reaction time, motivation and driving performance, which may overlap with the very functions an ADHD treatment is intended to improve. Cannabis should therefore not be presented as a substitute for established ADHD medication, psychological interventions or environmental support.

Autism

The autism evidence base has grown but remains uncertain. A 2025 systematic review of randomised trials found only four autism RCTs with available results and rated the certainty of evidence from very low to low. A 2026 systematic review in children and adolescents found 12 studies (four RCTs), with a signal for global clinical improvement and social communication in one trial but no consistent benefit across sleep, overall symptom severity or repetitive behaviour. This is a research-active area, not an established treatment for core autism.

Psychosis, bipolar disorder and vulnerability to adverse psychiatric effects

THC-containing cannabis requires particular caution in people with a personal or strong family history of psychotic illness, previous cannabis-related paranoia or psychosis, unstable bipolar disorder, severe dissociation, or other states in which intoxication may destabilise mental state. Risk is influenced by THC dose/potency, frequency of exposure, age at exposure and individual vulnerability. CBD should not be assumed to neutralise all THC-related psychiatric risk.

7. Safety, adverse effects and interactions

·       Common short-term effects: dizziness, somnolence/fatigue, dry mouth, nausea, appetite change, impaired concentration, slowed reaction time, altered perception and intoxication with THC-containing products.

·       Psychiatric effects: anxiety, panic, dysphoria, paranoia, hallucinations and, in vulnerable individuals, psychosis. Mood changes and suicidal thoughts require prompt clinical review.

·       Dependence: cannabis use disorder can develop. Risk is higher with frequent/high-THC exposure and in people with previous substance-use disorders or other vulnerability factors.

·       Cardiovascular effects: THC can acutely increase heart rate and alter blood pressure. Clinical assessment should consider cardiovascular disease, syncope, arrhythmia risk and interacting medicines.

·       Cognition and performance: THC can impair attention, memory, coordination and reaction time. This is especially important for driving, operating machinery and safety-critical work.

·       Liver and interactions: CBD can inhibit drug-metabolising enzymes and may raise concentrations of some medicines. Pharmaceutical CBD can elevate liver transaminases, especially with particular antiseizure regimens, so monitoring may be required.

·       Pregnancy and breastfeeding: cannabis exposure is not considered a benign option. Avoidance is generally advised unless a specialist determines that exceptional benefits outweigh risks; evidence does not establish safety for fetal or infant development.

·       Smoking: combustion exposes the lungs to irritants and toxic products and is not an appropriate pharmaceutical delivery strategy. Prescribed products are supplied in specified formulations and should be used exactly as directed.

8. What good specialist prescribing should include

9. T21 / Project Twenty21: what it was and what changed

Visit ukmccs.org ↗

The original Neurohaven page describes Project Twenty21 as an active initiative aiming to recruit 20,000 patients. That is now out of date. Drug Science states that T21 (formerly Project Twenty21) ran from 2020 to the end of 2024 and has concluded data collection. More than 4,000 patients were helped to access medical cannabis through the project.

T21 remains important because it generated longitudinal UK real-world evidence across conditions including chronic pain, PTSD, sleep problems and older populations. These studies can identify associations, tolerability patterns and patient-reported changes in routine practice. However, registry evidence does not provide the same protection against placebo effects, regression to the mean, selection bias and confounding as randomised controlled trials. It should therefore complement — not replace — controlled evidence.

10. Context, stigma and “Reefer Madness”

Visit en.wikipedia.org ↗

The 1936 film Reefer Madness is a useful historical example of sensationalised drug messaging. Modern evidence supports rejecting both extremes: cannabis is neither the uniformly catastrophic substance portrayed by old propaganda nor a harmless universal medicine. Good public information should distinguish recreational use from controlled medical prescribing, differentiate THC from CBD and other cannabinoids, and discuss both therapeutic evidence and harms with the same standard of scientific scrutiny.

11. External resources and links — cleaned and consolidated

The original page distributes links across multiple headings and embeds. The following table consolidates the most useful external resources, removes duplicate destinations, and separates authoritative guidance from education, research and commercial directories. Commercial links are included for navigation only and should not be treated as endorsements.