3 Years Since the UK Legalised Medical Cannabis, and Only 3 NHS Prescriptions

Medical Cannabis is Legal in the UK?

Medical Cannabis is legal in the United Kingdom, but you’d be forgiven for thinking that it was still prohibited. Access to treatment with medical cannabis has been shamefully slow, and even 3 years on, only 3 prescriptions have ever been issued under the NHS. 

The Path to Medical Cannabis in the UK

Cannabis has been used in British Medicine since the 1800s, but in 1971, parliament decided that it was no longer a useful medicine when it enacted the Misuse of Drugs Act, putting cannabis into Schedule 1 of the regulations. This decision was largely based on the hope that this would stop recreational use, with scant consideration being given to the needs of patients.

The plan failed. In the ensuing 50 years, the number of recreational users increased by over twenty-fold. But patients suffered, as the story of Alfie Dingley – and that of many others just like him – so movingly recounts. Most of the US and Canadian public were granted access to medical cannabis from the first decade of this century. However, the UK medical establishment upheld the ban, exaggerating fears of psychosis and dependence to justify their opposition. Parents of children with severe treatment-resistant epilepsy were forced to live abroad to save their children’s lives.

Eventually, public outcry over the treatment of the children with epilepsy moved the government to legalise medical cannabis in November 2018. However, barriers to prescribing, such as strict NICE guidelines, made NHS access impossible for the vast majority of patients, and in August 2019, the UK government published a report on “Barriers to accessing cannabis-based products for medicinal use on NHS prescription”. Unfortunately, access to medical cannabis has remained slow and almost exclusively limited to private patients. In March 2021, the guidance on the prescription of CBMPs for the treatment of resistant paediatric epilepsy was updated, clarifying that there was no recommendation against their prescription for this population group.

Despite this, the number of NHS prescriptions for full spectrum medical cannabis, three  years after legalisation is three. 

Given the potential cost savings to the NHS from better control of severe childhood epilepsy, which often requires urgent in-patient care, as well as other indications for medical cannabis, this resistance to evidence is perverse and reflects very badly on the medical profession.

It condemns patients and families to incur large costs for necessary healthcare, in direct contradiction to the ‘cradle to the grave’ philosophy that supposedly underpins the NHS.

During the above period, Drug Science has been working on developing the scientific evidence base on CBMPs to help improve patient access. Below, we address the recommendations made in the 2019 report. We focus on what has been achieved since it’s publication in 2019, offering the latest scientific evidence that may be used to overcome current barriers to access and where the remaining shortcomings lie. We also highlight the work that Drug Science has been doing in relation to education about medical cannabis, and the development of scientific evidence, including specific audits and Project Twenty21.

We urge the government to rectify the current NHS impasse and improve access to medical cannabis where there is evidence that it can be effective.

“It’s really important that this issue is addressed, and quickly. The evidence shows that these medicines are safe and effective, yet people simply can’t access them on the NHS as they should. This deadlock has gone on for far too long. It’s time that policy makers and ministers made the necessary changes"

– Lucy Stafford, Advocacy Director for the patient group PLEA

Key Points

Full Drug Science Response

In November 2018, medical cannabis once again became a medicine in the UK. In August 2019, the UK government published a report on “Barriers to accessing cannabis-based products for medicinal use on NHS prescription”. 

Despite of this, access to cannabis-based medicinal products (CBMPs) remains problematic for most patients, with only three NHS prescriptions having been written. In March 2021, the guidance on the prescription of CBMPs for the treatment of resistant paediatric epilepsy was updated, clarifying that there was no recommendation against their prescription for this population group.

During the above period, Drug Science has been working on developing the scientific evidence base on CBMPs to help improve patient access. This response by Drug Science and the Medical Cannabis Working Group focuses on what has been achieved since the publication of the 2019 report, where the remaining shortcomings lie, and how Drug Science and Project Twenty21 can contribute to providing a way forward.

Here will address the recommendations made in the 2019 report, offering the latest scientific evidence that may be used to overcome current barriers to access. We highlight the work that Drug Science has been doing in relation to education about medical cannabis, and the development of scientific evidence, including specific audits and Project Twenty21.

Recommendation 1:   UK-Wide Paediatric Specialist Clinical Network
A UK-wide paediatric specialist clinical network should be established to provide specialist clinical expertise, support discussion of complex cases, provide support to clinicians and to assist in evidence generation.

Drug Science is keen to understand whether this was done and what has been achieved so far. We think most people assume that the Refractory Epilepsy Specialist Clinical Advisory Service (RESCAS) committee was set up to approve and monitor the use of CBMPs but now understand from the founding chair – Prof Helen Cross- that this is not what they are supposed to be doing. Yet this committee continues to act as if they have this role and have denied NHS prescription access to children whose epilepsy has responded well to privately paid-for prescriptions for CBMPs.

Medcan Support has discovered that five families have been through the RESCAS referral process and all been denied CBMPs with the committee citing lack of evidence of safety and efficacy. Furthermore, families have citied clinicians not referring their children due to the lack of any positive recommendation from RESCAS meaning they would have no guarantee to a prescription due to no funding being made available through NHSE.

Complicating matters is the fact that RESCAS had apparently approved CBMPs for an individual child which lead to an NHS funded prescription, but have not divulged the criteria for this decision, which is very concerning to five parents whose children have been denied access to the same medication.

Recommendation 2:   Guidance for Prescribers
The National Medical Director and Chief Pharmaceutical Officer for England will write to doctors and pharmacists reminding them of General Medical Council (GMC) guidance on the prescribing and use of unlicensed medicines – and to clarify the procedure for prescribing and supplying cannabis-based products for medicinal use (CBPMs). Clinicians will also be made aware of how they can access the Health Education England (HEE) cannabis education package, commissioned by NHS England, and published alongside this report. 

Drug Science welcomes the March 2021 NICE clarifications of the guidelines on prescribing CBMPs for refractory paediatric epilepsy. It is important for clinicians to understand that there is NO recommendation against the use of cannabis-based medicinal products. This means that patients in this population can be prescribed cannabis-based medicinal products if their healthcare professional considers that that would be appropriate on a balance of benefit and risk, and in consultation with the patient, and their families/carers or guardian.

Together with further education, as outlined above, we hope these clarifications will make it easier for clinicians to prescribe in cases where CBMPs have shown efficacy. But this must be subject to proper monitoring and evaluation from both professional and patient perspectives.

Drug Science has developed a range of educational resources on medical cannabis, which are freely available. These include educational slide decks, videos, and podcasts. Drug Science have joined with the Medical Cannabis Clinicians Society to run a monthly training session for any medical practitioner who wants to prescribe or just learn about medical cannabis.

Reommendation 3:   Information for Patients
NHS England and NHS Improvement and the Department of Health and Social Care (DHSC) should work together to develop clear information for patients and patient groups on the prescribing of cannabis-based products for medicinal use.

The current information available on the NHS webpage requires updating and further clarification for patients. It is not beneficial to refer patients to recreational cannabis links, for the reasons outlined below. Rather, it is now vital to also include the recently updated information on effectiveness of CBMPs for treatment resistant epilepsies. (Zafar et al., 2020) and prescribing.

Also note that Drug Science has developed a range of freely available educational materials on medical cannabis (including webinars, podcasts and slide decks) and is working together with various patient groups, such as Patient-led engagement for access (PLEA) and we would welcome the opportunity to contribute further details, as requested by patients and prescribers. Drug Science would be very happy if these educational materials are made available via links from NHS websites.

Recommendation 4:   NIHR Support for Research in Priority Research Areas
The National Institute for Health Research (NIHR) should support research into the five priority research areas that have been identified by the draft National Institute for Health and Care Excellence (NICE) clinical guideline on the use of cannabis-based products for medicinal use.

These five research areas were – fibromyalgia and persistent treatment-resistant neuropathic pain in adults; chronic pain in children and young people; cannabidiol for severe treatment-resistant epilepsy; THC/cannabidiol combinations for severe treatment-resistant epilepsy; and spasticity.

So far there seems to have been little published evidence of progress on these targets from NHS research. To assist in this process Drug Science launched Project Twenty21 (T21) in August 2020, Europe’s largest observational medical cannabis registry addressing several of these five research areas.

To date, data of over 650 patients has been systematically collected and analysed. The majority (64%) of these patients were male and their average age was 38.7 years (range = 18-80). The most commonly reported primary conditions were chronic pain (55.6% of the sample) and anxiety disorders (32.0%).  Patients receiving prescribed cannabinoids reported a very low quality of life, assessed using the EQ-5D-5L and they also reported high levels of multi-morbidity, including high rates of insomnia and depression. A peer-reviewed paper on the outcomes of the first 200 patients is due to be published later this month in the journal Psychopharmacology.

A second peer-reviewed article, focusing specifically on pain patients, is currently in preparation, with further condition-specific articles (including anxiety disorders, Post-Traumatic Stress Disorder (PTSD), Multiple Sclerosis (MS), Tourette’s Syndrome and Adult Epilepsy) expected to be published throughout 2021 and 2022.

Recommendation 5a:   Randomised Controlled Trials for Severe Treatment-Resistant Paediatric Epilepsy
For severe treatment-resistant paediatric epilepsy, evidence generation should support one or more randomised control trials (RCTs) using a standardised approach with several comparative treatment arms.

We are disappointed that an NHS RCT trial on CBD isolate with placebo Vs CBD/THC isolate with placebo was planned at the time, but we were unable to find further details?  We assume therefore that it has not been started which is a great disappointment to the field. Could the reasons for this delay be made public?

Further, we would like to emphasise that the use of RCTs is not the most appropriate methodology for this specific population group. The reasons for this, and possible alternative approaches are discussed in our recent article in which we highlight the importance of patient-centred approaches including patient-reported outcomes, pharmacoepidemiology and n=1 trials, which can contribute to the development of the evidence-base for medical cannabis. 

We recently published an article on how the pathway to approval for the drug Zolgensma, used to treat infants with spinal muscular atrophy could provide a roadmap to easing access to medical cannabis for children suffering with epilepsy in the UK.

Particularly, we draw attention to how, like in the case of childhood epilepsy, a double blind RCT cannot be performed and therefore legislators must use other data to make approvals. With Zolgensma, a single-arm open label study design was adopted. Similarly, we argue that such study designs in rare paediatric epilepsies could also be used to effectively gather evidence on the safety and efficacy on novel pharmacological interventions such as whole-plant cannabis extracts which we have shown in our recent observational study to be highly effective in such patient groups.

"Conducting medical cannabis RCTs on children with epilepsy faces enormous obstacles. Not least that they would require half of participants to stop their current treatment, which would be wildly unethical, dangerous and could even prove fatal. Therefore medical cannabis for severe childhood epilepsy is effectively blocked under current NHS rules”

– Professor David Nutt, Founder & Chair of Drug Science

Recommendation 5b:   Gathering Evidence Beyond Randomised Controlled Trials
For severe treatment-resistant paediatric epilepsy, NHS England and NHS Improvement and NIHR in conjunction with the specialist network will work together to determine an appropriate alternative study design that will enable evidence generation for those patients who cannot be enrolled into a standard RCT. This scope of study should include those children and young adults who are currently in receipt of a CBPM.”

In February 2020, the Secretary of State for Health confirmed that an observational trial for this population group would be conducted but to the best of our knowledge this does not appear to have taken place yet. We are keen to understand more about the current alternative study being conducted, and can offer expertise from patients, physicians, and academic researchers, to move this evidence base forward.

Moreover, we believe that NHSE and NIHR should be looking at alternative ways to gather evidence.  This should include data from children who are already prescribed CBPMs. To this end, Drug Science, together with the charities End our Pain and MedCan’s support, has been gathering evidence, which is widely available, with further audits currently being conducted. This Drug Science research will be essential to be included in future benefit-safety assessments. Figure 1 below shows the results of our audit, showing the effectiveness of CBPMs for treatment resistant paediatric epilepsy in 10 young patients.

As a result of the reductions in seizures and use of mentally impairing medicines many parents reported great improvements in behavioural, psychological and cognitive faculties associated with medical cannabis use.

Figure 1.

Zafar et al. (2020)  


Figure 2. Follow up analysis of patients 11-20 of monthly seizure frequency pre and post CBMP

 Zafar et al. (Manuscript in preparation)

*P < 0.01

Additionally, Drug Science has collected qualitative evidence on the above patients from interviews with the families systematically highlighting the benefits and limitations of using CBMPs for treatment resistant paediatric epilepsy. In addition to improved seizure control, we found that patients (and their families) also benefit from much improved quality of life, including better sleep, improved digestion, and enhanced general well-being, mirroring the findings and key themes of the NHS report.

Recommendation 6:   Gather Evidence As Soon As Possible
These should commence as soon as possible, pending ethical review, and if necessary, the alternative study should commence in advance of the RCT.
The Drug Science audits – now of 21 patients [see data above] provide this data and prediction of response on this patients group. A straightforward Bayesian analysis of the propensity for CBMPs to reduce seizures for this cohort of patients, assuming an uninformative prior, shows that the posterior probability of the next patient showing improvement is 0.954. The magnitude of improvement is so large, that no inferential statistics are required to establish that CBMP works very well indeed. These simple AB trials with no control have established causality by virtue of the before and after treatment measures.
Recommendation 7:    National UK Cannabis Patient Registry
NHS England and NHS Improvement, DHSC and Devolved Administrations should work with industry and academia to scope the development of a national UK patient registry to collect a uniform data set, across all indications, for patients prescribed a cannabis-based product for medicinal use in the United Kingdom.

Drug Science welcomes the launch of the NHS medical cannabis registry in April 2021. This registry is a positive step, showing that the UK government is actively engaging in the conversation around medical cannabis. 

However, if data collection is limited to Epidyolex and Sativex, it will severely limit patient numbers as well as the conditions these medicines will be prescribed for. Our research to date [see above] clearly highlights the value to patients of access to full-plant extract products. To achieve the best efficacy, medical cannabis products often need to be tailored to individual patients’ needs by optimising CBD:THC ratios, as well as products based on minor cannabinoids/strains where possible.

If no other (unlicensed) products are being prescribed in the NHS the evidence base for other CBMPs will not be developed, and indeed, patient access to these may be hampered rather than enhanced.

If prescriptions for whole-plant medicines are excluded from the study, this registry will not provide much further information to that provided from the trials of Epidyolex and Sativex.

With Project T21, Drug Science has already developed a database as suggested, incorporating both industry and academia to collect a uniform dataset (utilising the well-established CB2 data collection software), for the following indications: Anxiety Disorder, Chronic Pain, Multiple Sclerosis (MS), Post-Traumatic Stress Disorder (PTSD), Substance Use Disorder, Tourette’s Syndrome, and Adult Epilepsy .

We would welcome a dialogue and involvement with NHS England, and the possibility to further widen our current indications.

Recommendation 8:   Update National Genomic Test Directory
NHS England and NHS Improvement will also explore how to update the National Genomic Test Directory so that whole genome sequencing would be offered to all children with severe treatment-resistant epilepsy under consideration for treatment with medical cannabis.

This makes sense we are not aware of any child having been offered any genome test, we guess this is because they are not under any consideration for CBPMs.

Recommendation 9:   Regulation of Cannabis Products
DHSC and Medicines and Healthcare products Regulatory Agency (MHRA) should work to provide access to information on good quality products, manufactured to Good Manufacturing Practice standard and with consistent ratios of cannabidiol (CBD) to delta-9-Tetrahydrocannabinol (THC) between batches.

What is the outcome of this proposal? Also, more clarity on carriers used, and any added ingredients, other phytocannabinoids and terpenes would be helpful.

Recommendation 10:   Ensuring Supply of Cannabis Products 
NHS England and NHS Improvement should work with suppliers to ensure that sufficient stock of good quality CBPMs are available and that the products available offer the best value for the NHS, including scoping options for UK manufacture.

We agree with the recommendation and it is relatively easy to ensure supply. The issue before this is to ensure doctors are confident to prescribe and what steps are being put in place taken so that CBMPs as Specials can be easily prescribed on the NHS.

In terms of this recommendation, it will be helpful to enter into a discussion with importers/producers who already have product available. Discussions should be opened now on the assumption that the NHS will start to prescribe in the coming months.

Furthermore, if the Home Office created clear guidance on high THC licenses and issued these, that would help start UK cultivation which could bring costs right down and ensure the supply chain?

In addition to the points outlined above, we would like to briefly address the issues raised in relation to medicinal cannabis use and mental health in recreational users. Our recently published review highlights the differences in terms of effects and adverse effects in medical versus recreational use and users. We find notable differences between medical and recreational cannabis users in terms of routes of administration, doses, forms of cannabis as well as set and setting factors that significantly reduce risks such as dependence and psychosis. This makes it unjustifiable to directly extrapolate findings and concerns from recreational use to medical use.

Also because of these differences, the tools traditionally used to assess recreational cannabis dependence are less relevant to measure medical cannabis dependence. This is why Prof Val Curran and Prof Michael Lynskey (Drug Science) developed the Cannabis-Based Medicines Questionnaire (CBM-Q) to specifically address problematic medical cannabis use. The CBM-Q is included as data collection within Project T21, ensuring a high number of respondents over a longitudinal timeframe.