Paper: Medical cannabis in the UK: From principle to practice

The 2017 review by NASEM (2017) provides a wide range of research conclusions on the health effects of cannabis and cannabinoids which are presented in Box 1 with recent additions by Drug Science (2019). The scientific evidence (or lack thereof) is controversial, emphasising the need for further research in most areas.

Focusing on the effectiveness of cannabinoids in the treatment of psychiatric conditions, including depression, ADHD, Tourette’s syndrome, post-traumatic stress disorder (PTSD), psychosis and anxiety, a recent meta-analysis found the evidence to be limited and of a low standard, concluding that a prescription for mental health treatments cannot be justified (Black et al., 2019). However, in that study the main focus was on pharmaceutical cannabinoids, rather than plant-derived medical cannabis for which the therapeutic potential may differ widely, and is broadly used in the USA, Canada and Germany.

Notwithstanding the lack of evidence in many cases, people are using medical cannabis for a broad variety of conditions, including many mental health indications. A survey by the United Patients Alliance (UPA; 2018), a UK patient-led medical cannabis support group, found that indications range from (in order of self-reported use) pain, depression, anxiety, insomnia, arthritis, fibromyalgia, muscle spasms, irritable bowel syndrome and migraines to headaches and more.

These findings were largely replicated with a representative sample, drawing attention to patient-reported outcomes (PROs). PROs emphasise the patient’s wellbeing and have been shown to be more sensitive to the effects of medical cannabis than traditional symptom-based measures. For example, a large recent naturalistic study on pain syndromes using PROs found adding a CBMP to ongoing medication significantly improved outcomes in patients with neuropathic pain (Ueberall et al., 2019). Further real-world benefits from CBMPs using patient reports have been reported for Parkinson’s disease (Balash et al., 2017), autism (Bar-Lev Schleider et al., 2019), pain, depression, and anxiety symptoms (Gulbransen et al., 2020). Many patients experience therapeutic satisfaction when using medical cannabis and report improvements in or relief of a range of symptoms (Gulbransen et al., 2020; Stith et al., 2018; Sexton et al., 2016). Others report using cannabis as an alternative to pharmaceutical prescriptions (Sith et al., 2018) and this can lead to reduced use of opioids for example (Ueberall et al., 2019). Many of these also report that the medical cannabis improves their quality of life (Bar-Leve Schleider et al., 2019; Gulbransen et al., 2020; Sith et al., 2018).

Doctors lack the knowledge of cannabis medicines to have the confidence to prescribe, especially off-license: they have not been trained in prescribing them and may not know the dosage etc. This barrier can be overcome by developing an educational programme, e.g. by Health Education England. A priority should be to provide a range of good quality teaching programmes. The Academy of Medical Cannabis (http://taomc.org) provides a free 12-module programme on the basics of cannabis which has now been used by about 1000 doctors. Drug Science launched a similar online resource (https://mymedic.org.uk/), arranged a series of seminars for health-care professionals (HCPs), and developed a teaching module for medical students (https://drugscience.org.uk/medical-cannabis-education-hub/). Further developments should include a diverse range of other teaching possibilities, especially accredited certificate course programmes.

In addition to the NICE guidelines, doctors are influenced by the guidelines produced by the Royal College of Physicians (2018) (for pain and nausea) and by the British Paediatric Neurology Association (2018) (for childhood epilepsy), which recommend the prescription of medical cannabis only as a last resort when conventional treatment has not been effective. In contrast, the Medical Cannabis Clinicians Society (MCCS) offer more balanced guidelines, proposing that for chronic pain for instance, cannabis medicine could be considered instead of opioids (MCCS, 2019). The British Pain Society (2019) recently released a revised position statement considering the potential role of medical cannabis in pain management, while at the same time continuing to highlight the need for further high quality research, clinical surveillance and patient monitoring. By reference to all these guidelines a physician can now make a more informed decision on prescription in the best interests of their patient.

Concerns about adverse mental health effects, especially psychosis and dependence, have been expressed (Di Forti et al., 2009) but the recent data suggest that these are mainly the result of using street ‘skunk’ with high levels (>10%) of d9THC and negligible levels of CBD (Di Forti et al., 2019). The large-scale database by Health Canada shows very few, if any, cases of psychosis with medicinal use. Surveys of people using medical cannabis revealed some patients with schizophrenia are using it to treat their symptoms and there are several studies providing experimental support for this application (Leweke et al., 2012; McGuire et al., 2018).

Similarly, whilst the risk of dependence is around 9% for recreational users of street cannabis, it is more common with high potency THC strains with a low CBD content, large ‘doses’, high frequency use (heavy, daily) and starting use in adolescence (Curran et al., 2016). Risk of dependence can therefore be mitigated with these factors in mind by giving harm-reduction advice. The Canadian database reveals that only about half of patients initiated on medical cannabis continue beyond six months, suggesting dependence liability is low. Moreover, the very easy access to street cannabis further argues that medical cannabis is unlikely to be sought for recreational purposes.

Still, regulations should mitigate against adolescent uptake and against the availability of high potency THC products lacking CBD. Additionally, in light of the recent outbreak in the USA of respiratory illness, including linked fatalities, associated with the vaping of black market (THC) cannabis oils (Centres for Disease Control and Prevention (CDC), 2019), it is vital to further communicate with the public about related risks and to more effectively regulate products and routes of administration in order to limit illicit products.

It is a serious shortcoming of current research that adverse effects of cannabis have largely been studied in relation to recreational (i.e. non-medical) use, rather than medical use. This is complicated by recent findings indicating that a large proportion of medical cannabis users also report recreational use (Han et al., 2018). Drug Science is currently reviewing existing research to discern if, how, and to what extent, adverse effects apply to prescribed medical use.

The cost of medical cannabis in the UK is currently high. Some families are forced to pay for private prescriptions, costing up to £40,000 a year, after their National Health Service (NHS) clinicians do not prescribe it (Wickware, 2019). Yet medical cannabis is potentially cheap – saving money on conventional treatments as well as on opioid prescription costs (Boehnke et al., 2016), anxiety prescriptions (Baron et al., 2018) and hospital admissions, for example, epilepsy (Bellnier et al., 2018). Medical cannabis could work out well economically, and a full health economics analysis is vital for conditions for which clinical efficacy has been shown.

The high cost of private prescriptions is related to the import challenges of medical cannabis. It is difficult to get access to the right products in a timely manner. Licensing and imports generally are for one patient for one month at a time as bulk import is still limited (Barnes, 2019). However, in March 2020, these restrictions were loosened and the UK government is now allowing bulk, non-patient specific, importation which should improve delivery time to the patient and begin to bring down costs. Current UK standards of regulation as well as of practice need to be fully developed and regularly revised, as is done, for example, by the Royal College of Physicians and Surgeons in Alberta/Canada in their ‘advice to the profession’ (http://www.cpsa.ca/wp-content/uploads/2018/05/AP_Cannabis-for-Medical-Purposes.pdf).

Many patients who request cannabis have not responded to standard treatments and are desperate to find something that helps ease their symptoms. In such cases, the fact that other treatments might be statistically more effective may not be relevant as a contra-indication to use of cannabinoids (Stockings et al., 2018). Now that cannabis substances have been legalised for medical uses, it is the duty of clinicians to assess the balance of legitimate patient need against potential harms as in any other area of medicine, particularly taking into account informed choice on behalf of the patient.

The ethical importance of autonomy interests in medical decision-making means that patients’ rights to information that enables them to properly weigh up potential goods and potential harms is paramount. When there is insufficient evidence and/or insufficient clinical understanding to adequately inform patients, patients are at risk of making bad decisions that can lead to harm. In the current context, most patients who use cannabis and their caregivers have decided to access cannabis without the benefit of clinical guidance or support. In so doing, patients are exercising autonomy and it would be excessively paternalistic to argue that such patients are intrinsically wrong in their actions. The desire for access to potentially beneficial treatments is certainly not wrong. Moreover, when a treatment is legally sanctioned by a national healthcare system for the application desired by the patient, then the patient has a right to access, within parameters. A healthcare system that legalises a treatment but then leaves patients to independently access and use those treatments because of a lack of evidence and clinical confidence, is arguably in danger of shirking its duty of care to vulnerable patients. At the same time, the challenges for prescribers have to be addressed as key gaps exist not only in the scientific evidence but also in the detailed information needed (but not yet available) about dosages, types, duration and formulations.

The development of the evidence base for medical cannabis needs to go hand in hand with the pursuit of clinicians’ education if the risk to patients of using cannabis from illicit sources is to be minimised. As evidence is being gathered, professional organisations must do what they can to educate clinicians. Given the problematics of clinical responsibility in an area where there is significant medical uncertainty and confusion (Singh et al., 2017), the education process needs to engage explicitly with the ideal qualities of a practitioner. One such important quality is professional integrity, which may invite clinicians to acknowledge the limits of knowledge and confidence in this context, while still seeking to provide the best possible care and support for their patients. Physicians need to comply with the standards of practice, and develop a healthy physician-patient relationship, rather than simply being ‘prescribers employed by a cannabis clinic’. It is essential to develop the UK regulatory regime for medical cannabis so that it can allow for patient access while at the same time avoiding a ‘free for all’ scenario as in some US states (Schlag, 2020).

It is essential to monitor patient outcomes and adverse effects, as is already being done by Health Canada. In the UK, Drug Science launched Project Twenty21 in November 2019 to create Europe’s largest national medical cannabis database registry (https://drugscience.org.uk/project-twenty21/). Continued monitoring and regulation can play a major role to mitigate risks and to collect and collate experimental and trial data. Efforts to collect valid ‘real world data’ in responsible and ethical ways need to be further improved. A ‘real world data’ approach, specifically promoting a digital solution to the multiple complexities of data, evidence, uses and formulations can offer a key resource. In the longer term, anonymous electronic patient records, such as Clinical Records Interactive Search (CRIS) can establish ‘real world’ data in large quantities.

When new compounds with misuse potential are licensed and deployed, the risks associated with misuse need to be mitigated so that patients are protected, but also so that the compound is not overused so widely that its use attracts stigma. A variety of elements to assure the safe use of such compounds can be instituted. In the UK, the packaging of opiates is to acquire warnings similar to those on cigarettes (Gregory and Wheeler, 2019). In the USA, the use of Spravato, Janssen’s intranasal esketamine for depression, is conditional on patients confirming they understand the risks, and on doctors and pharmacies undergoing training (Janssen Care Paths, 2019).

Registries, in which individual prescriptions or treatments are tracked, may offer a partial solution. At one end, the model of clozapine shows how a clear risk is mitigated by a pharmacist-operated algorithm based on a blood test result: no blood result, no dispensing. In some US and Australian jurisdictions, prescribers must check whether a patient is being prescribed an opiate elsewhere before writing the first prescription. In relation to medical cannabis, prescribers and regulators still need to develop and decide on the exact details of registry implementation.

In the UK, medical cannabis at present is offered as a ‘last resort’, when other licensed medicines have been shown to be unsuccessful. It would be useful to develop a hierarchy of evidence to see where cannabis medicines sit, which likely would be indication-specific. However, this is a challenging task as cannabis is not one medicine but a whole family of medicines. They would come out poorly in a double-blind trial as, for example, in chronic pain some people might respond to a high CBD product, some to a high THC product and some to a product that combines both. As such, whilst the effect of a single cannabis product might lack statistical significance in a clinical trial, an analysis of different combinations of ‘cannabis’ could be statistically significant (Namdar et al., 2019).

Best-practice guidelines for prescribing medical cannabis could be informed by comparing its benefit-safety balance with those of drugs already in use. Multi-criteria decision analysis provides a well-demonstrated basis for making such comparisons based on available evidence and clinical practice (Moore et al., 2017). For example, a multi-criteria decision analysis comparison of drugs for relapsing–remitting multiple sclerosis (RRMS) revealed very different benefit-safety profiles between drugs, which would enable prescribers to accurately select the most suitable drug for their patient (Vermersch et al., 2019). Such an analysis could include options of alternative products, doses and timing in order to aid practitioners’ decision-making.

There is a need for collaboration between different stakeholders (including patients, prescribers, clinics and scientists), and to develop an overarching mechanism to convene different parties together. A network of clinical studies could report on clinical practice and monitor outcomes (both risks and benefits), to enable innovative use of medical cannabis with care, precaution and foresight. This network group could address particular diseases or more general controversies related to medical cannabis. All these concerns are part of society’s move towards increasingly personalised medicine.

The 2018 re-scheduling of medical cannabis has been rightly criticised by the public and media alike based on the misperception that these medicines would become freely available on the NHS. Public communication about medical cannabis needs to be much better. Whilst information is available on the NHS website for the public, communication efforts need to be increased. Accusing the public of not understanding is unhelpful and may lead to a lack of trust which is vital for the doctor-patient relationship. If a communication vacuum occurs, this will be filled by other interest groups, such as industry lobbying groups. Hereby, medical cannabis should not be presented as a panacea for all ailments – rather the public needs to understand that not all patients may benefit from the use of cannabis.