Cannabinoids for Chronic Pain: Translating Systematic Review Findings Into Clinical Action

In their article, McDonagh and colleagues reported a systematic review of placebo-controlled trials and cohort studies to investigate the benefits and harms of cannabinoids for treatment of chronic pain. The editorialists discuss the limitations and strength of available evidence on cannabis as treatment of chronic pain and share their own advice for engaging with patients interested in using cannabis to treat chronic pain.

In their article, McDonagh and colleagues (1) report a systematic review of placebo-controlled trials and cohort studies to investigate the benefits and harms of cannabinoids for treatment of chronic pain. They categorized cannabis products by delta-9-tetrahydrocannabinol (THC) to cannabidiol (CBD) ratio, route of administration (oral, topical, or sublingual), and source (plant based, synthetic, or extract). Overall, use of comparable THC-to-CBD ratio products (1.1:1) and synthetic, high THC-to-CBD ratio products were generally associated with small improvements in pain severity and increased risk for sedation and dizziness. These effects were primarily seen among patients with chronic neuropathic pain, as studies of patients with other pain conditions were lacking. As with other such systematic reviews, included studies were of short duration (1 to 6 months) and did not include formulations available in the medical cannabis market (for example, CBD-dominant products and smoked or vaporized products). These limitations are well documented in the cannabinoid and chronic pain literature and are due in part to “War on Drugs” policies that have overwhelmingly favored studying cannabis-related harms over therapeutic effects (2). Unfortunately, this means that this well-conducted review found limited generalizable evidence to inform long-term use of available cannabis products for chronic pain, which is the most common reason for medical cannabis licensure in the United States (3).

Faced with this continued deficiency of clinical evidence, some physicians cite lack of data as rationale for not engaging with patients who wish to use or currently use cannabis. Such practices may reflect consideration of cannabis solely as a drug of misuse (even in the 37 states where medical cannabis is legal) and requirements to refer patients who disclose or test positive for cannabis use to addiction services or decline to refill opioid prescriptions. This stance harms the trust necessary for shared, patient-centric decision making that is already hampered by the stigma associated with cannabis use (4). Further, it punishes persons using legal medical cannabis products or CBD products, the latter of which are available over the counter and not regulated under the federal Controlled Substances Act. Indeed, a recent study showed that 50% (7 of 14) of study participants taking a standardized descheduled hemp product (1.04% CBD, 0.02% THC) for 1 month tested positive for THC metabolites via urinalysis (5). By analogy, it’d be incredibly problematic if 50% of persons having a drink of alcohol on the weekend had a positive drug test result that resulted in medical, employment, or legal consequences.

As such, we counsel clinicians to move in a different direction—one that centers on pragmatism, patient experience, known cannabinoid effects, and harm reduction. Doing so reframes disclosure of cannabis use or a positive drug screen result as an opening for a thoughtful clinical conversation of patients’ rationale for, patterns of, and outcomes of cannabis use. Such open information exchange enables clinicians to guide patients wishing to trial cannabis products, knowing that, as with other analgesics, some persons will benefit and others will not. Indeed, as noted by Nutt and colleagues (6), “every time a medicine is prescribed an n = 1 experiment is being conducted.” A useful n-of-1 trial requires targeted education, mutual creation of a plan defining treatment success and failure, and tracking symptom changes and adverse effects.

Clinicians should provide patients with guidance on routes of administration, the effects of CBD versus THC, and dosing—topics informed by McDonagh and colleagues’ review and previous reviews of the scientific literature (1, 7, 8). Clinicians could suggest using tinctures (effect onset, 15 to 45 minutes) for breakthrough pain and edibles or capsules (which last about 6 to 8 hours) for extended relief. If patients insist on inhaled administration, vaporizing cannabis flower is preferable to smoking (7). There are very little data on whether topicals are effective, but it seems plausible that those that contain an ingredient that allows absorption of lipid-soluble cannabinoids could be useful for musculoskeletal pain. Although human research on CBD’s analgesic effects is very limited, its remarkable safety profile (nonintoxicating, no misuse potential, and minimal adverse effects [7]) makes CBD alone an ideal starting point for cannabis-naive patients. THC may also be useful for some patients, especially those with sleep issues or those who have pain that is unresponsive to CBD (7, 8). The scientific literature suggests that CBD doses could start at 5 to 10 mg twice daily and increase to 40 to 50 mg daily, whereas THC doses could start at 0.5 to 3 mg (initially at night) and increase to 30 to 40 mg/d (7–9). These dose ceilings are flexible if patients are partial responders and not experiencing meaningful adverse effects that outweigh clinical benefit.

Other patients may benefit from advice to avoid common harms, such as sedation and dizziness (1), advice to avoid driving or other potentially dangerous activities while using THC products, information on cannabis use disorder, and discussion of drug–drug interactions. The cannabis literature suggests that a subset of persons with chronic pain successfully substitute cannabis for pain medications (10). If a patient is able to substitute cannabis products for opioid medications, the risk–benefit calculation is relatively straightforward given the disproportionate all-cause mortality associated with opioid use. However, patients should be counseled against substituting cannabis for disease-modifying drugs.

Given the slow pace of clinical trials, we believe it likely that McDonagh and colleagues’ findings will be the best available evidence for some time. While we await better evidence, we believe that clinicians should meet patients with chronic pain “where they are.” Conventional analgesic medications are effective only in a subset of persons, so it is no wonder that many patients are drawn to widely available cannabis products. Clinicians can compassionately witness, record, and offer guidance to help patients with chronic pain use cannabis wisely.