Medical marijuana: Should it be sold in pharmacies?


This year will bring a Canada Day for the history books. Only July 1, 2018, recreational marijuana (also called cannabis) will be legalized and regulated in Canada. The federal Cannabis Act creates a legal framework for producing, possessing and selling marijuana across Canada, meaning that each Canadian province will set its own rules to oversee its distribution, subject to federal government conditions. Provincial and federal governments will share in the responsibility for the oversight of this new system, and will also share in the tax revenue. Different provinces are taking different approaches, similar to how alcohol purchases vary between jurisdictions. This trend follows what we’re seeing at the state level in the United States, with different states moving to decriminalize recreational use.

Marijuana has been legal to some extent in Canada (and in many US states) for some time, in the form of “medical” marijuana. The Canadian government authorized the sale of marijuana for that purpose, while it simultaneously emphasizes that cannabis is not an approved therapeutic product. The medical market, for many, appears to simply be a means to access products for recreational, or non-medical use, and has generated wildly unsubstantiated claims about the medical merits of marijuana for conditions like autism and the treatment of cancer. Dispensaries have appeared across Canada and the US, usually with very easy referrals for prescriptions. Some dispensaries ignore any prescription requirement entirely and will sell marijuana directly to the public without any medical assessment or advice. With the introduction of government-overseen (and in some Canadian provinces, government-delivered) retail sales in Canada later this year, it’s reasonable to assume that unregulated dispensaries will eventually disappear.

With recreational sales imminent in Canada (and already here, in states like California), there are questions about the future market for “medical” marijuana. Should use for medical purposes be treated like recreational use, where consumers make their own selections, and purchases are taxed like other consumer products? Or should some forms or uses of marijuana be treated like prescription drugs, where a health professional remains involved, and products may be even be covered by insurance plans? Given the major changes we are seeing in how we can access marijuana, it’s worth summarizing the current state of evidence for marijuana when used for specific medical purposes. With marijuana becoming much more accessible, physicians, other health professionals, and their patients need high-quality information about its value for different medical conditions. David Gorski reviewed much of the evidence in a series of posts over the past three years. Now, three new documents prepared for Canadian physicians and health professionals concisely summarize the current evidence base for medical marijuana.

The pharmacy profession seemingly sees a bright future in medical marijuana, with big chains striking deals with producers and even hiring “brand managers” in anticipation of the shuttering of unregulated dispensaries and a continued demand for “medical” uses. The argument being made by the pharmacy profession seems to be that (1) marijuana is a legitimate drug for medical purposes, and should be treated as such, which includes (2) a pharmacy and pharmacist being involved in the provision. The latter we can set aside for now, and focus first on whether or not marijuana is indeed a drug that should be treated like other prescription drugs.

Before I continue, I should state my personal position on marijuana. I am fully supportive of the legalization of marijuana for recreational use and support regulation and taxation, treating it along the lines of alcohol or tobacco. I should also state that I have no “skin in the game” when it comes to marijuana in pharmacies, or medical marijuana more generally – I don’t work in retail pharmacy, and while pharmacy professional associations seem enamored with the idea of medical marijuana in pharmacies, I have no personal opinion on it, other that wanting pharmacies to be places that offer and promote science-based and medically useful products, not pseudoscience or harmful/ineffective products (see my post on the commercial and professional ethical obligations of pharmacists for more).

It’s worth mentioning as an aside that there’s a somewhat similar set of circumstances in US history, when alcohol was available only by prescription during Prohibition. This prescription (via the Smithsonian Institute) could be used by physicians to prescribe alcohol for an array of ailments:

Naturally, the prescription market for alcohol disappeared once Prohibition ended. But marijuana is not alcohol. It contains an array of potentially medically useful chemical substances, several of which have been clinically investigated for the treatment of different medical conditions.

The pharmacology of marijuana

As David Gorski has pointed out in previous posts, there are a number of biologically active chemicals in marijuana. The main psychoactive ingredients are called cannabinoids, and the primary cannabinoid produced is delta-9-tetrahydrocannabinol (Δ9-THC, or simply THC.). Cannabinoids are produced in the stalk, leaves, flowers, seeds and resin of marijuana plants. Marijuana can be smoked, vaporized, or eaten, among other forms of ingestion. THC is rapidly absorbed, and when inhaled, reaches the brain within minutes. (Oral absorption is lower owing to a significant reduction in available drug after passing though the liver.) These chemicals bond to cannabinoid receptors on cells throughout the body, triggering or modulating different effects. Marijuana immediately affects and impairs attention, concentration, memory, learning and motor coordination, proportional to the dose. You might wonder why our cells have receptors for THC and other cannabinoids. That’s because we (and other mammals) have an endocannabinoid system, and we naturally produce endocannabinoids. It is absolutely plausible that drugs that target endocannbinoid receptors, like THC (or derivatives), have the potential to produce beneficial medicinal effects, given the presence of receptors on nearly every organ system. With the growing understanding of the endocannabinoid system, and the identification of different types of receptors, there’s the potential for targeting specific effects on specific organs. That could mean products that produce beneficial effects and minimize any adverse effects (e.g., fewer psychoactive effects).

Cannabinoids are highly fat soluble and so are difficult for the body to eliminate – the complete elimination of a single dose may take up to one month. With repeated doses, levels can rapidly accumulate. While the liver eliminated cannabinoids, even the metabolites of THC can persist in the body, and there is little relationship between the levels of THC found in the blood and the degree of THC-induced effects. Owing to metabolism in the liver, THC has the potential to interact with other drugs. The overall impact has not been well studied. As a drug, there is lot we do not know about marijuana. However, we can be confident in observing that there is little acute toxicity of marijuana, unlike many other drugs and substances. While not addictive, there are also cases of cannabis use disorder, which while infrequent, can occur. It should be acknowledged that cannabis use disorder is a minor public health issue compared to the widespread harms and mortality caused by substances like alcohol and opioids.

The evidence check

Let’s now look at a trio of documents prepared by the Alberta College of Family Physicians. They routinely produce “Tools for Practice” which are concise, actionable answers to clinical questions. All the documents are available online, and are fully referenced, but I will summarize each document here:

Are medical cannabinoids (MC) effective for the treatment of pain?

Bottom Line: Evidence for inhaled marijuana for pain is too sparse and poor to provide good evidence-based guidance. Synthetic MC-derived products may modestly improve neuropathic pain for one in 11-14 users but perhaps not for other pain types. Additionally, longer and larger studies (better evidence) show no effect. Adverse events are plentiful.

The full document is available here.

  • Chronic pain: 39% experience pain reduction of >30% versus 30% with placebo, resulting in a number needed to treat (NNT)=11. Larger and longer RCT show no effects. The mean pain improvement is 0.5 on a 0-10 scale, which isn’t clinically meaningful.
  • Neuropathic pain: With inhaled MC, the NNT=6. With any MC, the NNT=14.
  • Cancer pain: In six randomized controlled trials, the pain reduction was not statistically significant.
  • HIV neuropathy: with smoked MC: NNT=4.
  • Pain from multiple sclerosis: Mean pain improvement over placebo was 0.8 on a 1-10 scale which was borderline insignificant.
  • Acute pain: One positive trial, one negative trial, and five trials showing MC is equivalent to placebo.
  • When compared to medication, MC was no better than amitriptyline (with more adverse events), or worse than dihydrocodeine with similar adverse events.
  • No overall differences shown in quality of life.
  • Little evidence for back pain, fibromyalgia, or osteoarthritis.

What are the harms associated with medical cannabinoid therapy?

Bottom Line: Compared to placebo, medical cannabinoids cause multiple different adverse events in patients, from visual disturbance or hypotension (1 in 3-10) to hallucination or paranoia (1 in 20). Stopping due to adverse effects occurs in 1 in every 8-20 patients. Regardless of the type of medical cannabinoid used, adverse events are common and likely underestimated. Given the extensive harms, potential benefits must be impressive to warrant a trial of therapy.

The full document is available here.

  • Eleven systematic review with meta-analyses of harm were identified.
  • 79-92% of patients using MC experienced any adverse event versus 56-78% with placebo, giving a number needed to harm (NNH)=5-8.
  • Serious adverse events measured in 3 meta-analyses, with two studies reporting no significant differences versus placebo. The third reported an odds ratio of 1.41.
  • Seven meta-analyses reporting stopping MC due to adverse events, with a range of 7-14% for MC versus 1-5% for placebo, giving an estimated NNH of 8-22. One of the seven analyses showed no significant difference versus placebo.
  • Side effects, overall, are very frequent: Sedation (NNH=5), feeling high (NNH=2-4) and euphoria (NNH=9). Other adverse events include visual blurring/hallucinations (NNH=3), dizziness (NNH=5), speech disorders (NNH=5), ataxia/muscle twitching (NNH=6), disconnected thought (NNH=7), dysphoria (NNH=8), hypotension (NNH=8), impaired memory (NNH=12), and disorientation (NNH=15).
  • MC increased adverse effects compared to drugs like prochlorperazine, causing more sedation and dizziness.
  • Adverse event rates varied little between different MC products, including synthetic cannabinoids and inhaled marijuana (NNH=4-7).

Besides pain, are medical cannabinoids effective for other conditions?

Bottom Line: For most conditions (example anxiety), cannabinoid evidence is sparse (at best), low quality and non-convincing. Dronabinol/nabilone improve control of nausea/vomiting post chemotherapy for 1 in 3 users over placebo. Nabiximols likely improve multiple sclerosis spasticity ≥30% for ~1 in 10 users over placebo. Patients’ preference for cannabinoids exceeds cannabinoids effectiveness.

The full document is available here.

  • Two comprehensive systematic review suggest reasonable evidence for nausea and vomiting due to chemotherapy (NNT=3) and for spasticity (NNT=7).
  • Evidence is sparse but suggestive for potential for children with Dravet epilepsy.
  • Evaluations show that patient preference exceeds effectiveness measures.
  • In other conditions, high level evidence is sparse, low quality, or negative, for conditions such as glaucoma, anxiety, or other causes of nausea and vomiting.
  • Blinding in trials was noted to be difficult, with 85-95% of patients and clinicians identifying who receiving MC.

Developing an evidence base for marijuana

Studying marijuana under rigorous circumstances has been difficult until fairly recently. The plant itself isn’t patented, so even ignoring the legal access issues, there may be a lack of industry enthusiasm in conducting clinical trials. The other issue is the challenge of a proper placebo control, particularly for non-oral forms of use. Given the psychoactive effects and the widely heralded effects on conditions that can only be assessed subjectively, like nausea, fatigue or appetite, a proper placebo is essential to separate out actual from placebo effects. While some commercial products have been developed and marketed with standardized ingredients and quality control (e.g., nabilone), these products are exceptions. However, these purified and standardized products have allowed for proper placebo controls and more rigorous assessments of effectiveness. Regrettably, these products haven’t been shown to be that effective which may suggest that the perceived beneficial effects may be largely placebo effects. Hopefully, clinical trials will become more common and more marijuana-based drugs can be more rigorously evaluated.

Evidence is lacking

The use of psychoactive drugs like marijuana is a health issue, particularly when used for medical purposes. Regrettably, there is a lack of high-quality data that shows marijuana for most medical purposes is both safe and effective. What little evidence exists is of poor quality and may not even be representative of the purposes for which medical marijuana is sought. There are significant gaps in information necessary to treat marijuana like other forms of medicine: Dosage standardization and overall quality control may not be in place. Overall effectiveness, contraindications, drug interactions, adverse events and long-terms risks when marijuana is used as medicine are not well understood. The best evidence suggests that marijuana may be a reasonable treatment option only when safer, more effective, and better tolerated treatment options have been tried first. If marijuana is to be treated as medicine, then it needs to meet the same standards of quality, effectiveness, and safety we would expect of any other prescription drug. That standard has not yet been met.

Images from flickr users Chuck Coker used under a CC licence.