Letter to the Editor: Cannabis as a Solution to the Opioid Crisis: Is the Cart Before the Horse Again?

Dear Editor,

The opioid crisis has fueled multifaceted harm reduction efforts to curtail the use of opioids for chronic pain relief. One proposed approach is the use of cannabis to spare or replace opioids. For example, New York, Pennsylvania, and Illinois have recently modified existing medical cannabis laws to allow patients to substitute their prescription opioid medications with cannabis. Furthermore, New York and Pennsylvania have added opioid use disorder (OUD) to the list of qualifying conditions for medical cannabis, and other states are following suit.

The gold standard evidence to get new medications approved is primarily based on data from randomized double-blind placebo-controlled (RCTs) studies. To receive United States Food and Drug Administration (FDA) approval, a pharmaceutical product must go through a rigorous review process that requires at least two adequately powered (typically n>200) RCTs. The approval of medical cannabis for opioid substitution and OUD has, thus far, relied on low-quality scientific evidence, as there are no still no adequately powered controlled RCTs testing the efficacy and safety of cannabinoids for OUD.

The popular press has increasingly reported that cannabis—and some of its constituent cannabinoids, such as delta-9-tetrahydrocannabinol—may either safely substitute for opioids or have opioid-sparing effects—decreasing the effective opioid dose necessary for pain control. The hope is that reducing the opioid dose will, in turn, reduce the opioid-related harms. That cannabis and cannabinoids are viewed as a solution for the opioid epidemic aligns with the growing public perception that cannabinoids are benign and have medical benefits. However, despite the growing public enthusiasm for cannabinoids to spare or substitute opioids, only one out of nine human studies in a meta-analysis provided evidence of opioid-sparing effects for pain control.¹ More recently, one small human laboratory study found evidence of acute opioid-sparing effects of cannabis on experimental pain outcomes, among healthy persons who use cannabis regularly, although with some increase in abuse liability.² Although these preliminary findings are promising, the opioid-sparing effects of cannabinoids need to be rigorously tested in the patient populations of interest.

In the absence of good-quality evidence, there are many logical missteps in assuming the opioid-sparing hypothesis of cannabis/cannabinoids is true—and for various loosely related outcomes. For instance, it is problematic to conflate putative opioid-sparing effects of cannabis for pain relief, with suppression of opioid withdrawal, and with reducing non-medical opioid use—especially in the context of OUD.

Studies testing the efficacy of cannabis/cannabinoids for pain relief have typically excluded participants with a history of substance use disorders (SUDs)—including OUD. It follows that even though pain relief is a commonly identified reason for cannabis use among individuals with OUD, even the analgesic efficacy of cannabis/cannabinoids among persons with OUD—who often experience opioid-induced hyperalgesia, or a heightened pain response—has not yet been established. Moreover, persons with OUD may be more likely to develop cannabis use disorder—which affects 1 in 10 regular persons who use cannabis in the general adult population. Furthermore, the alternate hypothesis that cannabis/cannabinoids use could lead to more extensive opioid use among persons with OUD remains plausible and has yet to be studied systematically. Unfortunately, the previous studies do not answer these critical questions, and should be interpreted in light of significant limitations.

First, the ecological fallacy—which asserts that associations observed at the ecological level cannot provide conclusions about processes at the individual level. Second, the recognition that state-level policy-impact analyses cannot address the confounding effects of relevant clinical factors, such as whether cannabis/cannabinoids may have specific effects among persons with OUD. Third, the appreciation that some methodologically sound studies have relied on databases with limited external validity, as exemplified by Medicare D and Medicaid claims data. Fourth, the possibility that other confounding factors, such as the expansion of evidence-based pharmacotherapies for OUD, such as buprenorphine, may have reduced the use of other opioids at the state level.

Drawing individual-level inferences about the opioid-sparing effects of cannabis/cannabinoids from aggregated data in policy studies can be deceptive. This is illustrated by the fact that epidemiological data at the individual respondent level have been less supportive of the opioid substitution and opioid-sparing hypotheses of cannabis/cannabinoids. For example, longitudinal analyses of waves I and II of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) showed that cannabis use was associated with an increased incidence of both OUD and nonmedical use of prescription opioids.³ Furthermore, studies of large samples of individuals with chronic pain have found that persons with chronic pain who use cannabis may have higher rates of SUDs than the general population, and may not spontaneously lower their daily use of prescription opioids by using cannabis.⁴

Since the specific impact of cannabis/cannabinoid use in OUD has not been experimentally studied, there are reasons for caution. Joint use of opioids and cannabis/cannabinoids may increase their abuse liability, and converging data suggest that a bidirectional relationship between OUD and a shared genetic liability for distinct psychiatric and SUDs. Collectively, it is conceivable that these factors may exacerbate the unwanted consequences of concurrent opioid and cannabis/cannabinoid use.

It has been argued that the lack of high-quality evidence reflects the regulatory challenges of conducting cannabinoid research. If this is the case, research should be urgently encouraged to generate high-quality evidence before changing public policy. Studies should focus on the specific dose-dependent impact of cannabis and its constituent cannabinoids, administered through various routes, among persons with OUD, using experimental human laboratory studies and clinical trials—which can generate more conclusive and interpretable information than observational studies. For instance, emerging data indicate anti-craving effects of oral cannabidiol among recently abstinent and untreated persons with OUD.⁵ As the use of cannabis and its constituent cannabinoids becomes more common among treatment-seeking individuals with OUD, another critical question is whether they may have distinct or interactive effects when used with the currently available pharmacotherapies for OUD—buprenorphine, methadone, and intramuscular long-acting naltrexone.

Some epidemiological approaches examining the relationship between cannabinoids and opioids could be conducted quickly. For instance, future prospective studies should separately examine the impact of cannabis/cannabinoid use on distinct populations with varying levels of long-term opioid exposure: (1) individuals with OUD; (2) individuals with aberrant use of opioid analgesics who do not yet meet the criteria for OUD; and (3) individuals with chronic pain receiving opioid analgesics, but without any evidence of nonmedical opioid use. The reason for studying these different populations is that, conceivably, various levels of opioid use can cause distinct neuroadaptations on nociceptive, rewarding, and mood regulation systems—thereby influencing the analgesic, subjective, and behavioral response to cannabis/cannabinoids. Finally, the influence of co-occurring psychiatric disorders should also be considered when identifying risk factors for adverse outcomes of concurrent cannabis and opioid use.

The opioids crisis and the shifting legal and public attitudes toward cannabis underscore the urgent need for higher-quality evidence on how cannabinoids affect opioid use. As health care professionals and cannabinoid scientists, we believe it is time to put the horse in front of the cart, once again.

Abbreviations Used

OUD

opioid use disorder

RCT

randomized double-blind placebo-controlled

SUDs

substance use disorders

Author Disclosure Statement

No competing financial interests exist.

Funding Information

J.P.D. is supported by the NIH grant 1K23DA05682-01 and the Robert E. Leet and Clara Guthrie Patterson Trust. In the past 2 years, DCD has received grant funding—administered through Yale University—from NIH, Takeda, Biogen, CH-TAC, Ceruvia, the Wallace Foundation and the Heffter Institute. He has served as a consultant to Jazz Pharma and Abide Therapeutics. A.B. is supported by the NIH grants R25DA037756 and R25DA044211.

Cite this article as: De Aquino JP, Bahji A, D'Souza DC (2022) Letter to the Editor: Cannabis as a solution to the opioid crisis: Is the cart before the horse again?, Cannabis and Cannabinoid Research 7:6, 898–900, DOI: 10.1089/can.2021.0155.