Intended for healthcare professionals

Feature Essay

The case for medical cannabis—an essay by M P Barnes

BMJ 2018; 362 doi: (Published 25 July 2018) Cite this as: BMJ 2018;362:k3230
  1. M P Barnes, honorary professor of neurological rehabilitation
  1. Newcastle University, UK
  1. m.p.barnes{at}

Legal restrictions on the grounds that cannabis has no therapeutic use have impeded research, says M P Barnes, but evidence shows cannabinoids may have benefit in several indications, calling for doctors to be allowed to prescribe the drug as they can in many other countries

Key messages

  • Cannabis has many medicinal properties, which are increasingly recognised worldwide

  • It seems a relatively safe product, with a good risk:benefit profile

  • We need much more research to understand the merits of different strains, different THC to CBD ratios, different modes of ingestion and dosages

  • Such research, however, would be facilitated by legalisation for medical use, and the UK should follow over 40 other countries that have done this.

Cannabis has been used as a medicine for thousands of years. The earliest recorded use dates back to 4000 BC in China. It was recognised as a medicine in both ancient Egypt and India and in the Greek and Roman cultures. In the 19th and 20th centuries it was widely used around the world as a treatment for migraine, neuropathic and musculoskeletal pain, and in childbirth.

In the UK, cannabis was made illegal in 1928 but doctors could prescribe it up to the introduction of the Misuse of Drugs Act in 1971. Cannabis is currently a Schedule 1 drug under the UK Misuse of Drugs Regulations 2001, which means that cannabis is deemed to have no legitimate use or medicinal value. It also remains a Schedule IV drug under the United Nations Single Convention on Narcotic Drugs treaty of 1961—along with heroin—for substances “particularly liable to abuse and to produce ill effects” which is “not offset by substantial therapeutic advantages.”

The illegal status of cannabis has impeded modern research. This is changing in many international jurisdictions, enabling both research into, and the use of, cannabis as a medicine. Indeed, cannabis for medical use is legal in 29 US states and in Australia, Austria, Belgium, Canada, Croatia, Czech Republic, Denmark, Germany, Israel, Italy, Malta, Netherlands, Portugal, and Spain, and 26 other countries worldwide. Laws vary from simple decriminalisation to full legal medical use.

The cannabis plant contains over 100 cannabinoids and many terpenes, flavonoids, and other components. The two most studied are tetrahydrocannabinol (THC) and cannabidiol (CBD). THC gives the recreational “high” but CBD does not—to some extent it counteracts the psychoactive effect of THC.1 CBD is unscheduled and legally available as a nutritional supplement in the UK, whereas THC is currently scheduled and illegal, although there are two formulations that can be prescribed in the UK: nabiximols (Sativex) is a natural product with about a 1:1 ratio of THC to CBD, it is available as an oromucosal spray and licensed in the UK for resistant spasticity in multiple sclerosis; and nabilone (Cesamet), a synthetic cannabinoid that mimics THC and is used for chemotherapy induced nausea and vomiting. Epidiolex (a pure form of CBD) has very recently been approved by the Food and Drug Administration in the US for the treatment of seizures associated with Lennox-Gastaut and Dravet syndromes.

Endocannabinoid system

Although the medical benefits of cannabis have been known for centuries, only recently has a scientific rationale been suggested. In 1990, Matsuda and colleagues described a cannabinoid receptor in humans,2 now called the CB1 receptor. Later a CB2 receptor was also identified.3 These receptors are present throughout the central nervous system and in peripheral tissue, including the immune, reproductive, and gastrointestinal systems, as well as the heart, lungs, and bladder. These receptors have natural ligands (anandamide and 2-arachidonoylglycerol). The system, including the natural precursors and the metabolic pathway, is now termed the endocannabinoid system. It is involved in many metabolic, endocrine, neural, and other functions.4 In neurological terms, for example, it is involved in brain protection mechanisms, modulation of pain, regulation of motor activity, and has a role in neurogenesis, neuroplasticity, and memory processing, although more research is needed.

The phytocannabinoids (THC, CBD, and over 100 others) found in the natural cannabis plant can mimic the effects of the endocannabinoid receptor ligands and also interact with other neural transmission systems.4

Evidence of efficacy

There is surprising evidence of efficacy, given the drug has been illegal in most countries for many years. Certainly, more studies are needed, particularly regarding the efficacy of different strains of the plant, ratios of THC to CBD, different methods of ingestion, and whether the whole plant is more efficacious for medicinal purposes than individual cannabinoids, which has been called the “entourage effect.” In brief, the evidence for key conditions is described below.


Several reviews have assessed the efficacy of various cannabinoid preparations for the management of chronic pain. One review found eight studies and concluded there was “moderate quality” evidence of efficacy against placebo to support the use of cannabinoids.5 My non-peer reviewed paper for the All Party Parliamentary Group on Drug Policy Reform in the UK found overall good evidence (defined as at least two class 1 studies backed up by other class II, III, or IV evidence) for pain relief in several conditions, including arthritic, neuropathic, and cancer pain, and with several different products, including the natural plant and as well nabiximols and synthetic cannabinoids.6


There is also evidence, sufficient to satisfy the licensing authorities in the case of Sativex, for the use of cannabinoids in spasticity. Most work has been in the context of nabiximols but other treatments have also been studied.7

Nausea and vomiting in chemotherapy

A recent Cochrane systematic review of 23 randomised controlled trials confirmed the anti-emetic properties of “cannabinoids.” Patients were five times more likely to report complete absence of vomiting against placebo. The authors did not find superiority of “cannabinoids” (undefined) when compared with conventional therapy.8


Recently evidence from published studies has shown that a pure CBD product (Epidiolex) has efficacy in the management of the drug resistant childhood epilepsies, Dravet and Lennox-Gastaut syndromes.910 The recent case of Alfie Dingley and other children,11 whose epilepsy responded to full extract cannabis oils containing CBD and THC, shows that the matter is complex and that some children seem to respond maximally to a combination of low dose THC and higher dose CBD. Clearly more studies will be needed once other cannabinoid preparations are available.

Studies have also shown that cannabis may have therapeutic use for anxiety,12 sleep disorders, appetite stimulation in the context of chemotherapy, fibromyalgia, post-traumatic stress disorder, and some aspects of the motor symptoms of Parkinson’s disease, as well as for the management of agitation in dementia, bladder dysfunction, glaucoma, and Tourette’s syndrome.6 Other indications are often cited anecdotally but so far lack a firm evidence base.

Side effects

In the appraisal of any “new” drug, a risk-benefit analysis is needed because any benefit may be outweighed by unacceptable side effects. For cannabis this seems not to be the case. Side effects generally depend on the amount of THC in the product. Varieties available for recreational use with very high THC (commonly called “skunk” in the UK) can cause serious mental health problems.13 In medicinal cannabis, however, lower THC levels, often combined with CBD which counteracts the effects of THC, are usually recommended. Certainly, in the short term, products with predominant THC can have effects such as dizziness, euphoria, drowsiness, dry mouth, confusion, disorientation, somnolence, balance problems, and fatigue, whereas these effects are generally not the case in pure CBD products or full extract oils high in CBD, where the effects are usually mild and well tolerated.

There is legitimate concern about the link between cannabis and schizophrenia or psychosis. The evidence suggests a likely causal link between cannabis use (particularly with high THC) and psychosis among people who already have psychotic symptoms or a family history of schizophrenia or other psychosis.14 Thus, prescription in such people would be a relative contraindication. Clearly, we need better understanding of the relation between cannabis use and psychosis if the drug is to be more widely available.

Dependency on cannabis is around 9% of users, usually those using a high THC product, which compares with about 15% for alcohol and 32% for tobacco.15

There is a theoretical risk of lung cancer from smoked cannabis but there is no definite association, and smoking cannabis is not the recommended form of medical administration.

Cannabis varieties with predominant THC can impair psychomotor performance and cognition (and driving) in the short term but there is conflicting evidence on whether there are neurocognitive deficits in the long term and these are probably only for heavy use of THC products in people who start using at a young age.16

Alternative drugs for the main indications should also be considered. As an example, severe pain is often treated with opioids, and the problem of opioid addiction and death either deliberately or accidentally, is a substantial social issue. No death has ever been reported from an overdose of cannabis. Prescription of cannabis can lead to less opioid use and occasionally cessation of opioids.17 Allowing doctors to prescribe cannabis might prevent tens of thousands of opioid deaths worldwide every year. Reduction of mortality from opioid death after introduction of medical cannabis has now been demonstrated in the US.18

Given the limited evidence available, an analysis of the benefits and risks of cannabis leads to the conclusion that in various formulations it is both an efficacious product in some indications and is reasonably safe.


In the past few weeks the UK government stance has changed from one of insistence that cannabis should remain a Schedule 1 drug to an acceptance that it has medicinal value. The report of the chief medical officer19 concluded that there was “conclusive evidence of the therapeutic benefit of cannabis based medicinal products for certain conditions and reasonable evidence of therapeutic benefit in several other conditions.” The recommendation was that “the whole class of cannabis based medicinal products be moved out of Schedule 1.” As a result, the home secretary has now asked the Advisory Council on the Misuse of Drugs to consider the matter and report back within three weeks (from early July 2018). The result of their deliberations is awaited at the time of writing. In the interim, a panel has been established for individual applications to be made for consideration of a special licence.20

While this is real progress there are still many questions yet to be answered. Not least, who will be able to prescribe what, and for what conditions. Will responsibility be moved from the Home Office to the Department of Health and Social Care, thus allowing devolved governments to determine their own criteria for supply and prescription, with consequent variation in availability?

A significant hurdle is likely to be the licensing of a natural plant product. Approval of drugs in the UK generally focuses on a single compound, rather than the whole variety of cannabinoids as well as terpenes and flavonoids contained in the cannabis plant. In addition, there are many different strains of cannabis plant with varying proportions of THC and CBD. The product can be ingested in many different ways, with wide variations in bioavailability. The UK’s drug approval system is simply not geared to recognising such complexity. It would take decades for each type and variety of cannabis to go through the current clinical trial system, and in the meantime thousands of people with disabilities in the UK are deprived of a potentially useful and relatively safe medicine.

Other countries have resolved this problem simply by developing alternative licensing systems for the plant product. Many jurisdictions, both in the US and other countries, control the quality of cannabis by approving specific suppliers and monitoring the quality of the product. The final cannabis product is often available only through licensed pharmacies with the appropriate medical recommendation.

Of course, we need more research into the efficacy and side effects, the most beneficial type of cannabis, and the best mode of ingestion for particular conditions and best dosage. Much work needs to be done, and this work will be facilitated by legalisation. It is certainly time that we moved beyond “reefer madness” to more enlightened use of a plant which has so much potential benefit for many tens of thousands of people in the UK.


  • Conflict of Interest: MBB is an ambassador for End our Pain, a cannabis lobbying organisation ( He is a trustee of the United Patients Alliance and CLEAR, which are both organisations promoting the use of medical cannabis. He is chief medical officer at Scythian Biosciences, a Canadian company promoting cannabis research.


View Abstract