Research Essay

Vol.:(0123456789)1 3

European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144 https://doi.org/10.1007/s00406-018-0960-9

I N V I T E D R E V I E W

Understanding the evidence for medical cannabis and cannabis-based medicines for the treatment of chronic non-cancer pain

Gabrielle Campbell1  · Emily Stockings1  · Suzanne Nielsen2

Received: 28 August 2018 / Accepted: 21 November 2018 / Published online: 11 January 2019 © Crown 2019

Abstract The use of medical cannabis and cannabis-based medicines has received increasing interest in recent years; with a corre- sponding surge in the number of studies and reviews conducted in the field. Despite this growth in evidence, the findings and conclusions of these studies have been inconsistent. In this paper, we outline the current evidence for medical cannabis and cannabis-based medicines in the treatment and management of chronic non-cancer pain. We discuss limitations of the current evidence, including limitations of randomised control trials in the field, limits on generalisability of previous findings and common issues such as problems with measurements of dose and type of cannabinoids. We discuss future directions for medicinal cannabinoid research, including addressing limitations in trial design; developing frameworks to monitor for use disorder and other unintended outcomes; and considering endpoints other than 30% or 50% reductions in pain severity.

Keywords Cannabis · Chronic pain · Medical cannabis · Cannabis-based medicines

Introduction

Interest in cannabinoids for the treatment of chronic non- cancer pain (CNCP) has increased substantially in recent years. Rapid changes in legislation mean that cannabinoids are increasingly accessible; and treatment of chronic pain is the most commonly cited reason for accessing medicinal cannabinoid products in the United States (US) [1]. Hopes that cannabinoids may help curb the opioid epidemic in the US have piqued the interest of policy makers, research- ers, clinicians and patients alike. Consequently, there has been a recent proliferation of controlled and uncontrolled

trials examining the efficacy of cannabinoids for chronic pain conditions. With that, there has also been a number of recent reviews examining the evidence for cannabis in chronic pain, though conclusions have been conflicting [2], with some reviews reporting moderate to large effects [1, 3], while others have reported minimal [4–7]. Here, we provide a summary of the current body of evidence to use in CNCP, separately for the major pain conditions for which evidence is available, limitations of the current evidence and suggest future directions in our understanding of the effectiveness of cannabinoids for CNCP. Based on the terminology suggested by Hauser et al. [8], we use the following terms throughout, ‘medical cannabinoids’ as an umbrella term and refers to the all plant-derived and synthetic derivatives; ‘medical canna- bis’ refers to the use of cannabis plants and plant material, such as buds, leaves or full plant extracts (such as canna- bis sativa, THC and or CBD extract) for medical reasons and ‘cannabis-based medicines’, are registered medicinal extracts with defined sand standardised THC and THC/CBD content (such as nabiximols and nabilone).

* Gabrielle Campbell [email protected]

Emily Stockings [email protected]

Suzanne Nielsen [email protected]

1 National Drug and Alcohol Research Centre (NDARC), Faculty of Medicine, UNSW Sydney, 22-32 King Street, Randwick, NSW 2031, Australia

2 Monash Addiction Research Centre, Eastern Health Clinical School, Faculty of Medicine Nursing and Health Sciences, Monash University, Level 2, 5 Arnold Street, Box Hill, VIC 3128, Australia

136 European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

Evidence found in the reviews

Any chronic pain

There have been a number of high-profile reviews that have reported that there is ‘moderate’ [3] and ‘substantial’ [1] evidence for the efficacy of medical cannabinoids in the treatment of chronic pain. The 2017 National Academies of Science, Engineering and Medicine (NASEM) report [1] on “The Health Effects of Cannabis and Cannabinoids: The Current State of Evidence and Recommendations for Research” concluded that there is “substantial evidence that cannabis is an effective treatment for chronic pain in adults” (p. 90) [1]. Evidence for this conclusion was based heavily on a 2015 review led by Whiting et al. [3] which reported there is “moderate evidence” for medical cannabinoids for the treatment of chronic pain. The overall conclusions of the NASEM and Whiting reviews indicated that there is evidence for the effectiveness of medical can- nabinoids for all chronic pain conditions, despite both reviews largely only considering evidence from patients with neuropathic pain (see Table 1). The 2017 NASEM review only included two additional studies to the 2015

Whiting review, both of which comprised participants with neuropathic pain. Further, the review by Whiting et  al. [3] has since been criticised as having a “bias towards a positive evaluation of cannabis products” [9] and for inter- preting the results as statistically significant despite esti- mates crossing the line of no effect. Additionally, another recent review by Aviram and Samuelly-Leichtag [6] also states that cannabis-based medicines might be effective for chronic pain management, even though the authors acknowledge this is mainly based on limited evidence for neuropathic pain patients. It is concerning that there have been some broad recommendations for the use of medical cannabinoids in CNCP based on studies of specific condi- tions, primarily neuropathic pain. Of concern, the reviews by Whiting et al. [3] and Aviram and Samuelly-Leichtag [6], and the review of reviews by Allan et  al. [10] also included studies for cancer pain in their analysis. Combin- ing different chronic pain conditions can be problematic. Although both cancer and CNCP can include nocicep- tive and neuropathic pain, the development, course and persistence of CNCP are inherently different form that of cancer pain and treatments for each should be evaluated separately. Given there are important differences in the aetiology of pain conditions, to understand the potential

Table 1 Summary of the evidence for cannabinoids in the treatment of CNCP

References Separate estimates for different can- nabinoids

Separate estimates for different CNCP types

RCT and observational evidence

IMMPACT guidelines?

n studies (partici- pants)

Conclusion

Whiting et al. [3] Yes No RCT only No 28 (22,454) Moderate quality evi- dence that cannabi- noids are beneficial for CNCP

National Academies of Science report [1]

No No RCT only No Review of reviews mainly based on Whiting (2015 with an additional 3 studies of neuro- pathic pain)

There is substan- tial evidence that cannabis is an effective treatment for chronic pain in adults

Aviriam et al. [6] No Yes RCT only No 43 (2437) Might be effective for neuropathic pain

Nugent et al. [13] No Yes Yes No 27 (3281) Limited evidence suggests that can- nabis may alleviate neuropathic pain in some patients, but insufficient evidence exists for other types of chronic pain

Mücke et al. [31] Yes Yes—only neuro- pathic

RCT only Yes 16 (1750) Any benefit of can- nabinoids out- weighed by harms

Stockings et al. [15] Yes Yes Yes Yes 104 (9958) Unlikely that can- nabinoids are highly effective medicines for CNCP

137European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

benefit of cannabinoids in CNCP, it is crucial to examine the evidence for specific pain conditions, where evidence is available, as concluded by Deshpande et al. [11] “Gen- eralizing the use of medical marijuana to all CNCP condi- tions does not appear to be supported by existing evidence. Clinicians should exercise caution when prescribing medi- cal marijuana for patients, especially in those with non- neuropathic CNCP” (p. e372) [11]. Below we focus on the evidence of for specific CNCP pain conditions, where evidence is available.

Neuropathic pain

Neuropathic pain, that is pain caused by a lesion or disease of the somatosensory nervous system [12], has been the most rigorously studied pain condition in trials of medici- nal cannabinoids [3, 13, 14]. A number of reviews of the evidence for medicinal cannabinoids for neuropathic pain have been published, some of which report moderate to large reductions in pain [1, 3], while others have reported minimal benefit and risk of potential harms [4, 6, 7]. A recent review [5] found that in randomised controlled trials (RCTs), medicinal cannabinoids were superior to placebo in producing a 30% reduction in pain and a reduction in pain scores as measured on continuous numeric or visual analogue scale (VAS) for people with neuropathic pain conditions, but these effects were modest: number needed to treat for benefit (NNTB) for one person to achieve a 30% reduction in pain was 27 (38.5% response in treatment ver- sus 33.0% response in placebo), and change in pain scores was equivalent to a reduction of 4  mm on a 0–100  mm VAS. Cannabinoids were not significantly different to pla- cebo in producing a 50% reduction in pain. Observational evidence was largely consistent, but study quality has been poor overall. Importantly, patients using cannabinoids were two to three times more likely to experience adverse events (primarily dizziness, drowsiness, confusion and low mood) and were more likely to withdraw from treatment than people receiving placebo. It is also important to note that the majority of studies have tested nabiximols (trade name Sativex), which delivers a combined dose of tetrahy- drocannabinol (THC) and cannabidiol (CBD), with little evidence for other types of cannabis-based medicines or medical cannabis in neuropathic pain.

In a recent Cochrane review examining the efficacy of cannabis-based medicines for chronic neuropathic pain in adults, Mücke and colleagues [7] report a NNTB of 20 for 50% or greater reduction in pain, suggesting a similar efficacy and profile of cannabinoids for pain as reported in the review by Stockings et al. [5].

Most of the evidence to date on the effectiveness of cannabis-based medicine for CNCP has been from studies of neuropathic pain. In a recent review of reviews, Allan

et al. [10] concluded that there is uncertainty whether can- nabinoids improve pain, though if they do, it is most likely to be neuropathic pain and the benefit is likely minimal. Based on this evidence, recent guidelines in Canada [15] and Europe [8] recommend cannabis-based medicines be used only in patients whose conditions are refractory to standard medical therapies and as a third-line therapy. Medical cannabis, in both guidelines, is only to be used when trials of cannabis-based medicines are ineffective.

Multiple sclerosis-related pain

Pain arising from multiple sclerosis (MS) may either be caused by damage to the nervous system (i.e. MS-related neuropathic pain) or may be due to musculoskeletal pain associated with spasms. Medicinal cannabinoids for MS- related pain have received substantial attention in recent years: a recent overview of systematic reviews [16] identi- fied 11 systematic reviews of the evidence for medicinal cannabinoids in MS-related pain, drawing on 32 studies, 10 of which were moderate to high quality RCTs. Of the seven reviews examining medicinal cannabinoids for pain in MS, two concluded that THC or nabiximols (THC:CBD) are likely effective in reducing pain or painful spasms in MS, one review reported no significant effect, and four reviews cited insufficient evidence or mixed findings [16]. In the systematic review by Stockings et al. [5], cannabi- noids produced significant but modest reductions in pain scores relative to placebo for MS-related neuropathic pain (equivalent to a reduction of 4.3 mm on a 0–100 VAS) but not for MS-related musculoskeletal pain. No differences between cannabinoids and placebo were found for 30% or 50% reduction in pain among people with MS; however, only two studies examined these outcomes. As with neuro- pathic pain, patients using cannabinoids were two to three times more likely to experience adverse events or to with- draw from treatment than those using placebo. It is possible that medicinal cannabinoids may play a role in the treatment of pain among people with MS; however, the number of high-quality studies to date has been small. It is also diffi- cult to isolate the potential benefit on pain symptoms alone, as cannabinoids may also alleviate other symptoms of MS, including muscular spasticity (which can be painful), sleep, and quality of life. Further, there have been concerns that cannabinoids may impair cognitive function among people with MS who have pre-existing cognitive dysfunction [16]. These factors need to be considered when determining the overall suitability of cannabinoids in managing MS-related pain.

138 European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

Visceral pain

Visceral pain is pain that occurs in the trunk of the body, including the heart, lungs, abdominal and pelvic organs, and includes conditions such as pancreatitis, gallstones, pelvic pain and gastrointestinal conditions such as inflammatory bowel disease, ulcerative colitis and Crohn’s disease. While there is a growing literature exploring the potential thera- peutic mechanisms of medicinal cannabinoids in visceral pain [17] and demonstrating that people with visceral pain are more likely to use medical cannabis to manage their pain [18], there are currently too few clinical trials to provide a meaningful conclusion regarding their efficacy in managing visceral pain. Two RCTs comprising just 98 patients with chronic abdominal pain after surgery or chronic pancrea- titis [19, 20] were identified in a recent review of canna- bis for CNCP conditions [5], with meta-analysis indicating that medicinal cannabinoids had no significant impact on pain scores. Interest in this field in increasing and a num- ber of clinical trials examining cannabis for visceral pain are underway which will soon be synthesised in a planned Cochrane review [21].

Other pain

Other pain conditions that have been considered potential targets for medicinal cannabis treatment include fibromy- algia, rheumatoid arthritis and musculoskeletal pain condi- tions of the back and neck; however, the number of studies is currently too few, and samples sizes are insufficient to draw meaningful conclusions. In a recent Cochrane review, two small RCTs (comprising n = 32 and 40 participants) examined nabilone for fibromyalgia [22], however, neither study measured 30 nor 50% reduction in pain. While one of the studies noted improvements in pain scores, anxiety and health-related quality of life [23], no statistically significant effect was found, and review authors concluded that there is currently no convincing evidence that nabilone is of value in treating pain among people with fibromyalgia [14]. One small RCT (n = 58) examining nabiximols relative to pla- cebo for pain relief among people with rheumatoid arthri- tis reported a significant reduction in pain scores among patients receiving nabiximols, however, study quality was low, thus caution is warranted in interpreting trial findings [24]. Finally, one small RCT (n = 30) examining nabilone versus placebo for people with musculoskeletal pain condi- tions [25] [predominantly spinal pain (cervical syndrome), lower back pain (lumbalgia), and thoracic syndrome] reported that nabilone was superior to placebo in achieving reductions in overall pain intensity, however, most findings were borderline non-significant, and replication with larger samples is needed. We identified no trials specifically exam- ining the efficacy of cannabinoids for migraine-related pain.

Harms A recent review of medicinal cannabinoids for neuropathic pain by  Mücke et  al. found that the Num- ber Needed To Harm (NNTH) was three for nervous sys- tem disorders (95% CI 2–6), based on nine studies with 1304 patients and 10 for psychiatric disorders (95% CI 7–16) among nine studies with 1314 participants. Stock- ings et al. [5] reported that the estimated pooled rate of all- cause adeverse events was 81.2% among people receiving cannabinoids, compared with 66.2% of those receiving placebo; and the NNTH was 6 (95% CI 5–8). Importantly, patients using cannabinoids were two to three times more likely to experience adverse events (primarily dizziness, drowsiness, confusion and low mood) and were more likely to withdraw from treatment than people receiving placebo [5]. Allan et al. [10] also found that the NNTH for psychiatric adverse events was 9, sedation 5, dysphoria 8 and disorientation and confusion 15. In the review by Nugent et al. [13], they found consistent evidence between cannabis use (specifically related to THC content) and the development of psychotic symptoms. Allan et  al. [10] concluded that the most consistent effects for medicinal cannabis are adverse events which are also likely to be underestimated as many studies enrol patients with a prior cannabis use history, rather than cannabis naïve individu- als. In the comprehensive systematic review of all RCT and observational studies by Stockings et  al. [5], only 3% of studies (n = 104) reported that patients were can- nabis naïve at study entrance. Experienced users are less likely to experience adverse events as they are preselected as resistant, have developed tolerance and may like some adverse events such as ‘feeling high’ [10]. The review by Mucke et  al. concluded that the potential benefits of cannabis-based medicine in chronic neuropathic pain in adults might be outweighed by their potential harms.

Cannabinoid type In a recent review of 104 clinical tri- als and observational studies on the use of medical can- nabis and cannabis-based medicines for CNCP [5], 42 studies were based on medical cannabis (cannabis sativa, THC and or CBD extract), and 59 were based on canna- bis-based medicines (e.g. 24 studies for nabiximols, 18 for dronabinol and 17 studies for nabilone). As cannabis contains over 100 distinct cannabinoid constituents and THC and CBD concentrations can vary considerably, it is difficult in studies of medical cannabis to determine the most beneficial dose and type. Both the Canadian [15] and European [8] guidelines for use of medicinal cannabinoids recommend the use cannabis-based medicines first (i.e. nabilone or nabiximols) before the use of medical can- nabis. Although some reviews have aimed to examine the evidence for different cannabinoids separately [3, 5, 7], others have grouped the cannabinoids together [6, 13, 26]

139European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

which makes guidelines and recommendations difficult to determine.

Limitations of current evidence

This section provides an overview of the limitations of previous evidence of the effectiveness of cannabinoids in CNCP.

Limitations of evidence from RCT’s

With the exception of the reviews by Stockings et al. [5] and Hauser et al. [2], most reviews of the evidence for cannabi- noids in chronic pain have focused solely on findings from RCTs. Although RCT’s are the Gold Standard in determin- ing efficacy, there are a number of issues that need to be con- sidered when evaluating the generalisability of the evidence.

In an attempt to control for variables that may influence treatment outcomes, RCTs typically exclude people with complex physical and mental health comorbidities and people with a history of substance use disorders, which comprise a substantial number of people living with CNCP [27–30]. Previous research has found that people living with CNCP often experience multiple pain and other health conditions, have poorer overall physical health and have also experience high rates of childhood abuse and neglect. People living with CNCP also experience elevated rates of depression and anxiety, suicidal behaviours and a notable proportion have a history of substance use disorder. A recent Cochrane review examining the evidence for cannabis-based medicines for neuropathic pain found the majority (10 out of 16 included studies) explicitly stated that they excluded peo- ple with a history of substance use [31]. To understand the overall effectiveness of medicinal cannabinoids in CNCP, we need to understand the impact they may have on people with complex comorbidities.

Second, most RCT’s are of limited duration (i.e. median of 8 weeks in the review by Stockings et al.) [5]. Given that CNCP is a chronic long-term condition, the examination of the appropriateness of long-term use of cannabinoids in CNCP is lacking, in terms of both treatment efficacy and safety. Importantly, the review by Stockings et al. [5] found that reductions in pain intensity were largest for 1-day stud- ies, and smaller or non-significant in studies of 13 weeks duration or longer, providing some initial suggestion that the effectiveness of cannabinoids for CNCP may diminish over time. This finding is consistent with the findings of a recent systematic review of systematic reviews [10] that large or longer duration studies were less likely to find an effect of cannabis-based medicines on pain.

Third, current evidence is limited due to small sample sizes in trials conducted to date. A recent Cochrane review for cannabis-based medicine in neuropathic pain found that 9 of the 16 studies were at high risk of bias due to small sample size. The review by Stocking et al. [5] found that for some estimates, effect sizes were notably larger in studies with < 30 participants per treatment arm compared to stud- ies of 100 + per arm, however, these estimates fell within overlapping bounds of uncertainty. There is a growing body of evidence indicating that effect estimates tend to be larger in studies with small sample sizes [32], and as such, cau- tion should be taken when interpreting outcomes based on studies with small sample sizes. Well-conducted, large RCTs comprising at least 100 participants per treatment arm should be considered a priority in this space.

There are also concerns regarding the quality of evi- dence to date. There is a paucity of high-quality studies on which evidence-based decisions can reasonably be made. Of the 104 studies included in the review by Stockings et al. [5], (which included observational studies) only 15 studies were graded as high according to an adapted version of the standard Grades of Recommendation, Assessment, Develop- ment and Evaluation (GRADE) tool on study methodology. Forty-three were graded as moderate, 24 graded as low and 22 studies were graded as very low. Further, the authors also reported that most parallel and crossover RCT’s were rated as having an unclear risk of bias across all domains as all required information were not reported or could not be obtained from the authors of the studies. Several studies in the review were rated as a high risk of bias because of selective reporting or other biases, such as omission of data and CIs, changes in selection of the primary endpoint, or a failure to take account of within-subject effects in crossover studies [5].

Finally, it is possible that publication bias exists, whereby authors are more likely to publish studies with positive find- ings. Mücke et al. [7] reported they found three-industry sponsored studies with negative results that had not been fully published and a further three studies were the results unknown despite the study authors attempting to gain the results. Other reviews [6] have not included unpublished studies and may have overestimated the evidence for canna- bis-based medicines in CNCP [33].

Lack of studies on main pain conditions

The Global Burden of Disease 2016 study found back pain was the leading cause of non-fatal health burden globally (as measured by years of life lived with disability, YLDs), followed by migraines [34], with neck pain and other mus- culoskeletal problems among the top ten causes of non- fatal health burden. Not surprisingly then, “severe pain” is the most commonly cited reason for accessing medicinal

140 European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

cannabis in the US [1]. Despite the substantial variation in pain conditions experienced by people seeking medicinal cannabinoids, most of the clinical evidence to date has been based specifically on studies of participants with neuropathic pain. There is scant, low quality evidence on cannabinoids used for fibromyalgia or visceral pain, and very few studies of cannabinoids’ use in the most common and burdensome CNCP conditions, namely, back/neck problems, migraines and arthritides. While some reviews have extended the find- ings of studies conducted among people with neuropathic pain to apply to CNCP overall, it is important to acknowl- edge that there is a lack of studies for the most common pain conditions for which people commonly report using cannabis to manage. Future studies focusing on these most common pain conditions should become a priority.

Cannabinoid dose and type

Lynch et al. [35] argue that it is ill-advised to treat all can- nabinoids the same and that the term ‘medical marijuana’ often groups together all compounds and formulations. In fact, there are very important pharmacokinetic differences between modes of delivery (e.g. oral versus inhaled), differ- ences in cannabinoid profiles (e.g. the presence of CBD in different amounts), and source of cannabinoid (plant-based complex botanicals versus synthetic single molecules). Dis- tinctions must be made when citing these studies to ensure that the conclusions are not drawn more widely than is jus- tified. Among the evidence that is currently available, can- nabis dose is often poorly recorded. Often only a maximum recommended dose is reported and data on participants’ actual cannabinoid consumption are seldom provided, thus it is difficult to make strong recommendations on doses that are maximally effective and safe. Medical cannabis prod- ucts vary in strength and formulation and the optimal dose range in pain is yet to be determined, and may be affected by a range of factors, including tolerance with long-term use. In recent Canadian [15] and European [8] guidelines, both recommended the use of pharmaceutically developed products (such as nabiximols and nabilone) and recommend against the use of medical marijuana (specifically smoked cannabis), as the evidence for smoked cannabis is likely to have a very high risk of bias and long-term consequences are unknown. To improve recommendations around type and dosing future trials, specific trial designs, including adaptive trial designs [36] utilising products with known dosages, to explore optimal dosing in different populations to provide appropriate guidelines. Research in this area is difficult due to the current legal status of cannabis and cannabinoids in a number of countries. Although U.S states may decide to legalise cannabis, it has not been legalised at the federal

level making research into the use of cannabis for medicinal purposes difficult.

As mentioned above, another important issue is that a large proportion of participants that have participated in the research to date are unlikely to be cannabis naïve and therefore are more likely to have a better chance of hav- ing better outcomes and less chance of adverse events. The severity and number of adverse events is most likely to be greater than reported and more likely to occur in cannabis naïve patients. It is important that future studies document whether participants are cannabis naïve and should focus on the safety profile of cannabis-based medicines among people who are cannabis naïve.

Most studies included cannabinoids as an adjunctive therapy

Most studies used a placebo comparator and added cannabi- noids to stable doses of analgesics, NSAIDS and anti-spas- ticity drugs, thus the evidence for cannabinoid use in CNCP is largely around cannabinoids as adjuvant medicines, as is common in other RCT’s, with other centrally acting pain drugs. Often multiple analgesics were used, which varied between groups, and the ways they were used was not con- sistently reported. Most studies held doses of other analge- sic medications constant, though some studies documented changes in breakthrough medication or adjunctive analgesia [5]. To really understand the effect of cannabinoids in people with CNCP, placebo-controlled trials involving patients with no other medication are warranted.

Lack of evidence of the effectiveness of cannabinoids in other important CNCP outcomes

Pain is considered by leading clinicians and researchers to be only one of a range of core outcomes that must be con- sidered evaluating interventions for CNCP [37]. Other out- comes include physical functioning, emotional functioning and participants (or their carers’) ratings of improvement. To date, however, there have been few studies or reviews that have examined these other important outcome domains in CNCP. Research examining the efficacy of cannabinoids in managing pain has often focussed on pure changes in pain scores (e.g. scores on a numerical or visual rating scale), whereas increasingly the pain field is moving towards more global measures of functioning, and quality of life [38]. Two recent reviews [5, 31] examined the impact of cannabinoids on these other outcomes found although there may be evi- dence for a reduction in sleep problems, psychological dis- tress and patient rating of improvement, this was based on low quality evidence and more research is needed in this area before conclusive recommendation can be made. As the field increasingly focuses on functional improvements, in addition

141European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

to considering potential effects of cannabinoids on quality of life and sleep, different conclusions about the effectiveness of cannabinoids may arise. Some of these outcomes may also be more clinically meaningful than reductions in pain scores for patients with chronic disabling pain who experi- ence substantial impairment in their daily functioning.

There is also the possibility that cannabis does not reduce pain severity per se but may help a person living with CNCP tolerate the pain more effectively [39]. It is important that future research begins to focus not only on pain, but that we have high-quality placebo-controlled trials that examine a range of outcomes including effect on sleep, patient’s rat- ings of improvement and distress associated with pain. In a recent study [40], patient reports indicated cannabis pro- vided 7/10 effectiveness in reducing their pain (on a scale of 0–10 where 10 was ‘completely effective’), suggesting patients perceived that the cannabis was effective on their pain. This finding came even though patients who reported cannabis use had greater pain scores and that there were no long-term associations between cannabis and pain severity or interference. Patient reports of perceived benefit should be considered in future research alongside ‘pure’ measures of pain severity.

Future directions

There are a growing number of controlled trials examin- ing the efficacy of cannabinoids in managing CNCP, many with important limitations as discussed above. To advance the field, future trials may consider the following recom- mendations to improve the quality and applicability of the evidence.

Issues with blinding

Few studies with cannabinoids in pain have addressed limi- tations with the lack of blinding resulting from the psycho- active effects of THC [41]. Wilsey et al. suggest to address this limitation future studies could consider strategies such as the use of active placebos with similar side effect profiles to cannabinoids. One such example was the successful use of amitriptyline in a study of fibromyalgia, with blinding demonstrated to be preserved as participants estimation of the condition they received being less than 50% (i.e. no better than random chance) [42]. Other proposed strategies include measuring patients’ beliefs about the efficacy of the treatment allocations and controlling for this in analyses, and reporting on the effectiveness of the blinding procedures [41]. Such approaches may address the powerful expectancy effects that can contribute to bias in clinical trial results [43]. There is arguably no ‘gold standard’ analgesic, with limita- tions surrounding the use of all analgesics. However, the

use of active (and psychoactive placebo) controls will help to reduce bias and help to determine if cannabinoids offer additional benefits over existing medications.

The emergence of cannabis use disorders and developing precaution frameworks

Given the already rapid rates at which cannabinoids are becoming increasingly available in many countries, future research should determine the most appropriate ways to assess for the emergence of cannabis use disorders. Cannabi- noid use disorders are one of the most prevalent substance use disorders in the general population, estimated at 5.4% in the Australian general population and 6.3% in the US [44, 45]. Among those who use cannabis, rates of dependence are around one in ten [46]. Substance use disorders are identified among people with chronic pain at higher rates than the gen- eral population, with a systematic review reporting the life- time prevalence of a substance use disorder ranged from 16 to 74% [47]. Given this, and the known dependence liability of cannabinoids, it would be prudent to establish frameworks to address this risk proactively and put appropriate manage- ment strategies in place for patients. In addition, designing studies with a sufficiently long follow-up period, and includ- ing in both RCTs and patient registries standardised meas- ures for use disorders related to cannabis-based medicines will inform often cannabis use disorders emerges and the likely consequences for patients with pain. This would allow a better understanding of these phenomena, and any likely adverse effects for patients using cannabinoids for pain. Rec- ommended frameworks for monitoring these outcomes with opioids provide a good starting point, with suggestions to monitor for substance use disorder outcomes including the use of validated screening tools to assess risks, and collect- ing routine urine drug screens to detect other substance use [48] being equally applicable to cannabinoids. The experi- ence with prescribed opioids has demonstrated that failure to adequately appreciate and address risk with dependence or use disorders can have devastating consequences. It is inevi- table that with the increasing use of cannabinoids for pain we will see some patients develop dependence. Although the likelihood of fatal overdose with cannabinoids is extremely low, the lessons learned from the rapid expansion of opioids, where dependence liability was not only underestimated, but was also not routinely monitored for, can be put into play here. We now have frameworks to monitor for emerg- ing dependence with prescribed opioids [49]. Implementing similar universal precaution frameworks with cannabinoids is one strategy to consider.

With changing legislation leading to growing acceptance of cannabinoids and reduced risk perceptions around its use [50, 51], it is likely that—whether the evidence demonstrates effectiveness—more patients will be interested in trialling

142 European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

them. Pharmacovigilance in this area will be important to identify any emerging signals of harm. Monitoring for unin- tended consequences of therapeutic use of cannabinoids will be an area of ongoing interest. Globally, there are a wide range of legal frameworks for therapeutic cannabinoid prod- ucts. Greater opportunities for monitoring for such conse- quences exist in countries with legal frameworks to prescribe medicinal cannabinoids, and particularly where registries of prescriptions and patient are utilised [52]. Given that medi- cal cannabis is often supplied outside therapeutic channels even when used for pain, capturing adverse drug reactions associated with therapeutic use may be more challenging than with traditional pharmaceutical drug supply, but this will be important in developing our understanding of which patients benefit and which patients might be most suscepti- ble to side effects.

Potential “opioid‑sparing” effects

When considering potential population level benefits of cannabinoids, the potential for cannabinoids to reduce reli- ance on high dose opioids, and consequently reduce opioid- related overdose and mortality, is an area of both clinical and public health interest. Pre-clinical evidence certainly supports this possibility [53], and ecological [54–57] and epidemiological [58–61] studies also suggest that this is pos- sible [62], though not all studies find cannabis use is associ- ated with reduced opioid use [40]. The evidence base sug- gesting that cannabinoids may reduce opioid consumption and related mortality is not without its limitations [62]. Few studies have controlled for other interventions that reduce opioid-related mortality (e.g. opioid dependence treatment availability). Epidemiological studies have been limited by their design and recruitment strategies [62]. To date, despite promising pre-clinical and ecological evidence, no well-con- ducted clinical trials that address these limitations and can demonstrate clear causality have confirmed the potential for cannabinoids to reduce opioid use or harms [53]. Whether cannabinoids represent a potential strategy to help patients taper off prescribed opioids or support the use of lower total doses of opioids would be an area of immense clinical and public health interest.

Fragmented models of care

Unique challenges may arise from the fragmented medical care that may emerge as cannabinoids are prescribed by a separate ‘cannabis doctor’ and ‘cannabis dispensary’ in some countries, for example in the United States [63], with proposals for similar models in Australia. Some legal frame- works, such as in operation in many countries in Europe [52], prohibit these models and may facilitate greater com- munications between healthcare professionals.

These models of care mean that the same health profes- sionals (or retailers in the case of cannabis dispensaries) do not manage the rest of a patient’s care or medications. Being able to incorporate therapeutic use of cannabinoids into a patient’s medical records may help to identify adverse drug reactions or interactions and also estimate potential benefits from cannabinoids where administrative datasets are used to examine long-term outcomes. These actions are not possi- ble where cannabinoids are supplied outside the mainstream health system, and the lack of information may be of detri- ment to individual patient care and slow the development of the broader knowledge base relating to benefits and harms.

Finally, it is important to note that many guidelines sug- gest that for the management of CNCP a multidisciplinary approach with an emphasis on non-drug techniques is best [64–66]. Non-pharmacotherapy options include patient education, cognitive behavioural therapy, physical therapy, surgery and other non-invasive procedures. If medicines are to be used, management of CNCP should not rely on phar- macological therapy alone and a combination of pharmaco- logical and non-pharmacological should be continued. As research and interest of the role of medicinal cannabinoids increases, it is crucial that patients are aware of other treat- ments that may assist in the management of their pain.

Conclusion

Cannabinoids may offer important advances in a range of areas of medicine, yet currently the evidence in many areas is in its infancy. Despite this, CNCP is now the most com- monly cited reason for accessing medicinal cannabis. There are a number of limitations that need to be addressed to advance the field. Public perception around the efficacy of cannabinoids for pain so far are not consistent with evidence [67]. It will be important to ensure that patient expectations are appropriately managed, particularly where there are large out of pocket expenses for these medications. Further, poten- tial side effects may limit utility, particularly among older adults where the prevalence of chronic pain is highest.

Funding No direct funding was received to write this article. SN, ES and GC are supported by NHMRC research fellowships (#1132433, #1104600 and #1119992). The National Drug and Alcohol Research Centre at the University of New South Wales is supported by funding from the Australian Government under the Substance Misuse Preven- tion and Service Improvements Grant Fund.

Compliance with ethical standards

Conflict of interest SN has been an investigator on untied investigator- driven educational grants funded by Indivior and Reckitt-Benckiser and has had travel costs covered and honoraria paid to her institution to provide training on identification and management of codeine de-

143European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

pendence by Indivior. GC has been an investigator on untied investiga- tor driven educational grants from Reckitt-Benckiser. ES declares no conflicts of interest.

References

1. National Academies of Sciences Engineering and Medicine (2017) The health effects of cannabis and cannabinoids: the current state of evidence and recommendations for research. The National Academies Press, Washington, DC

2. Hauser W, Fitzcharles MA, Radbruch L, Petzke F (2017) Can- nabinoids in pain management and palliative medicine. Deutsch Arztebl Int 114(38):627–634

3. Whiting PF, Wolff RF, Deshpande S et al (2015) Cannabinoids for medical use: a systematic review and meta-analysis. JAMA 313(24):2456–2473

4. Petzke F, Enax-Krumova EK, Hauser W (2016) Efficacy, toler- ability and safety of cannabinoids for chronic neuropathic pain: a systematic review of randomized controlled studies. Schmerz (Berlin Germany) 30(1):62–88

5. Stockings E, Campbell G, Hall WD et al (2018) Cannabis and can- nabinoids for the treatment of people with chronic non-cancer pain conditions: a systematic review and meta-analysis of controlled and observational studies. Pain

6. Aviram J, Samuelly-Leichtag G (2017) Efficacy of cannabis- based medicines for pain management: a systematic review and meta-analysis of randomized controlled trials. Pain Phys 20(6):E755–E796

7. Mücke M, Phillips T, Radbruch L, Petzke F, Hauser W (2018) Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database Syst Rev 3:CD012182

8. Hauser W, Finn DP, Kalso E et al (2018) European Pain Federa- tion (EFIC) position paper on appropriate use of cannabis-based medicines and medical cannabis for chronic pain management. Eur J Pain (London England) 22(9):1547–1564

9. Hauser W, Petzke F, Fitzcharles MA (2018) Efficacy, tolerability and safety of cannabis-based medicines for chronic pain manage- ment—an overview of systematic reviews. Eur J Pain (London England) 22(3):455–470

10. Allan GM, Finley CR, Ton J et al (2018) Systematic review of systematic reviews for medical cannabinoids: pain, nausea and vomiting, spasticity, and harms. Can Fam Phys (Med Fam Can) 64(2):e78–e94

11. Deshpande A, Mailis-Gagnon A, Zoheiry N, Lakha SF (2015) Efficacy and adverse effects of medical marijuana for chronic non- cancer pain: systematic review of randomized controlled trials. Can Fam Phys 61(8):e372–e381

12. International Association for the Study of Pain (IASP) (2018) IASP terminology. http://www.iasp-pain.org/termi nolog y?navIt emNum ber=576#Neuro pathi cpain . Accessed 30 July 2018

13. Nugent SM, Morasco BJ, O’Neil ME et al (2017) The effects of cannabis among adults with chronic pain and an overview of gen- eral harms: a systematic review. Ann Intern Med 167(5):319–331

14. Walitt B, Klose P, Fitzcharles MA, Phillips T, Hauser W (2016) Cannabinoids for fibromyalgia. Cochrane Database Syst Rev 7:CD011694

15. Allan GM, Ramji J, Perry D et al (2018) Simplified guideline for prescribing medical cannabinoids in primary care. Can Fam Phys (Med Fam Can) 64(2):111–120

16. Nielsen S, Germanos R, Weier M et al (2018) The use of cannabis and cannabinoids in treating symptoms of multiple sclerosis: a systematic review of reviews. Curr Neurol Neurosci Rep 18(2):8

17. Malik Z, Baik D, Schey R (2015) The role of cannabinoids in regulation of nausea and vomiting, and visceral pain. Curr Gas- troenterol Rep 17(2):9

18. Kerlin AM, Long M, Kappelman M, Martin C, Sandler RS (2018) Profiles of patients who use marijuana for inflammatory bowel disease. Digest Dis Sci 63(6):1600–1604

19. de Vries M, van Rijckevorsel DCM, Vissers KCP, Wilder-Smith OHG, van Goor H (2017) Tetrahydrocannabinol does not reduce pain in patients with chronic abdominal pain in a phase 2 placebo- controlled study. Clin Gastroenterol Hepatol 15(7):1079–86.e4

20. de Vries M, Van Rijckevorsel DC, Vissers KC, Wilder-Smith OH, Van Goor H (2016) Single dose delta-9-tetrahydrocannabinol in chronic pancreatitis patients: analgesic efficacy, pharmacokinetics and tolerability. Br J Clin Pharmacol 81(3):525–537

21. Kafil TS, Nguyen TM, MacDonald JK, Chande N (2018) Cannabis for the treatment of ulcerative colitis. Cochrane Database Syst Rev 11(2):CD012954

22. Walitt B, Klose P, Fitzcharles MA, Phillips T, Häuser W (2016) Cannabinoids for fibromyalgia. Cochrane Database Syst Rev 7:CD011694

23. Skrabek RQ, Galimova L, Ethans K, Perry D (2008) Nabilone for the treatment of pain in fibromyalgia. J Pain 9(2):164–173

24. Blake DR, Robson P, Ho M, Jubb RW, McCabe CS (2006) Pre- liminary assessment of the efficacy, tolerability and safety of a cannabis-based medicine (Sativex) in the treatment of pain caused by rheumatoid arthritis. Rheumatology 45(1):50–52

25. Pinsger M, Schimetta W, Volc D, Hiermann E, Riederer F, Pölz W (2006) Nutzen einer add-on-therapie mit dem synthetischen cannabinomimetikum nabilone bei patienten mit chronischen Schmerzzuständen—eine randomisierte kontrollierte studie. Wien Klin Wochenschr 118(11):327–335

26. National Academies of Sciences, Engineering, and Medicine (2017) The health effects of cannabis and cannabinoids: the current state of evidence and recommendations for research. National Academies Press, Washington

27. Campbell G, Nielsen S, Bruno R et al (2015) The Pain and Opi- oids IN Treatment study: characteristics of a cohort using opi- oids to manage chronic non-cancer pain. Pain 156(2):231–242

28. Rogers KD, Kemp A, McLachlan AJ, Blyth F (2013) Adverse selection? A multi-dimensional profile of people dispensed opioid analgesics for persistent non-cancer pain. PLoS ONE 8(12):e80095

29. Tunks ER, Crook J, Weir R (2008) Epidemiology of chronic pain with psychological comorbidity: prevalence, risk, course, and prognosis. Can J Psychiatry 53(4):224–234

30. Walker ER, Druss BG (2017) Cumulative burden of comor- bid mental disorders, substance use disorders, chronic medical conditions, and poverty on health among adults in the U.S.A. Psychol Health Med 22(6):727–735

31. Mücke M, Phillips T, Radbruch L, Petzke F, Häuser W (2018) Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database Syst Rev 3:CD012182

32. Dechartres A, Trinquart L, Boutron I, Ravaud P (2013) Influ- ence of trial sample size on treatment effect estimates: meta- epidemiological study. BMJ Br Med J 346:f2304

33. Hauser W, Fitzcharles MA (2018) The perils of overestimat- ing the efficacy of cannabis-based medicines for chronic pain management. Pain Phys 21(1):E79–E80

34. Vos T, Barber RM, Bell B et al (2016) Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Dis- ease Study 2013. Lancet 386(9995):743–800

35. Lynch ME, Ware MA (2015) Cannabinoids for the treatment of chronic non-cancer pain: an updated systematic review

144 European Archives of Psychiatry and Clinical Neuroscience (2019) 269:135–144

1 3

of randomized controlled trials. J Neuroimmune Pharmacol 10(2):293–301

36. Bhatt DL, Mehta C (2016) Adaptive designs for clinical trials. N Engl J Med 375(1):65–74

37. Turk DC, Dworkin RH, Allen RR et al (2003) Core outcome domains for chronic pain clinical trials: IMMPACT recommen- dations. Pain 106(3):337–345

38. Levy N, Sturgess J, Mills P (2018) “Pain as the fifth vital sign” and dependence on the “numerical pain scale” is being aban- doned in the US: why? Br J Anaesth 120(3):435–438

39. Piper BJ, Beals ML, Abess AT et  al (2017) Chronic pain patients’ perspectives of medical cannabis. Pain 158(7):1373–1379

40. Campbell G, Hall WD, Peacock A et al (2018) Effect of can- nabis use in people with chronic non-cancer pain prescribed opioids: findings from a 4-year prospective cohort study. Lancet Public Health 3(7):e341–e350

41. Wilsey B, Deutsch R, Marcotte TD (2016) Maintenance of blinding in clinical trials and the implications for studying analgesia using cannabinoids. Cannabis Cannabinoid Res 1(1):139–148

42. Ware MA, Fitzcharles M-A, Joseph L, Shir Y (2010) The effects of nabilone on sleep in fibromyalgia: results of a randomized con- trolled trial. Anesth Analg 110(2):604–610

43. Colagiuri B (2010) Participant expectancies in double-blind ran- domized placebo-controlled trials: potential limitations to trial validity. Clin Trials 7(3):246–255

44. Goldstein RB, Chou SP, Smith SM et al (2015) Nosologic com- parisons of DSM-IV and DSM-5 alcohol and drug use disorders: results from the national epidemiologic survey on alcohol and related conditions-III. J Stud Alcohol Drugs 76(3):378–388

45. Mewton L, Slade T, Teesson M (2013) An evaluation of the pro- posed DSM-5 cannabis use disorder criteria using Australian National Survey Data. J Stud Alcohol Drugs 74(4):614–621

46. Hall W (2015) What has research over the past two decades revealed about the adverse health effects of recreational cannabis use? Addiction 110(1):19–35

47. Morasco BJ, Gritzner S, Lewis L, Oldham R, Turk DC, Dobs- cha SK (2011) Systematic review of prevalence, correlates, and treatment outcomes for chronic non-cancer pain in patients with comorbid substance use disorder. Pain 152(3):488–497

48. Turk DC, O’Connor AB, Dworkin RH et  al (2012) Research design considerations for clinical studies of abuse-deter- rent opioid analgesics: IMMPACT recommendations. Pain 153(10):1997–2008

49. Gourlay DL, Heit HA, Almahrezi A (2005) Universal precautions in pain medicine: a rational approach to the treatment of chronic pain. Pain Med 6(2):107–112

50. Schuermeyer J, Salomonsen-Sautel S, Price RK et al (2014) Tem- poral trends in marijuana attitudes, availability and use in Colo- rado compared to non-medical marijuana states: 2003–11. Drug Alcohol Depend 140:145–155

51. Troutt WD, DiDonato MD (2015) Medical cannabis in Arizona: patient characteristics, perceptions, and impressions of medical cannabis legalization. J Psychoact Drugs 47(4):259–266

52. Krcevski-Skvarc N, Wells C, Hauser W (2018) Availability and approval of cannabis-based medicines for chronic pain

management and palliative/supportive care in Europe: a survey of the status in the chapters of the European Pain Federation. Eur J Pain 22(3):440–454

53. Nielsen S, Sabioni P, Trigo J et al (2017) Opioid-sparing effects of cannabinoids: a systematic review and meta-analyses. Neuro- spychopharmacology 49(9):1752–1765

54. Bachhuber MA, Saloner B, Cunningham CO, Barry CL (2014) Medical cannabis laws and opioid analgesic overdose mor- tality in the United States, 1999–2010. JAMA Intern Med 174(10):1668–1673

55. Bradford AC, Bradford W, Abraham A, Bagwell Adams G (2018) Association between us state medical cannabis laws and opioid prescribing in the medicare part D population. JAMA Intern Med

56. Livingston MD, Barnett TE, Delcher C, Wagenaar AC (2017) Recreational cannabis legalization and opioid-related deaths in Colorado, 2000–2015. Am J Public Health 107(11):1827–1829

57. Pardo B (2017) Do more robust prescription drug monitor- ing programs reduce prescription opioid overdose? Addiction 112(10):1773–1783

58. Abuhasira R, Schleider LB, Mechoulam R, Novack V (2018) Epidemiological characteristics, safety and efficacy of medical cannabis in the elderly. Eur J Intern Med 49:44–50

59. Boehnke KF, Litinas E, Clauw DJ (2016) Medical cannabis use is associated with decreased opiate medication use in a retrospec- tive cross-sectional survey of patients with chronic pain. J Pain 17(6):739–744

60. Corroon JM Jr, Mischley LK, Sexton M (2017) Cannabis as a substitute for prescription drugs—a cross-sectional study. J Pain Res 10:989

61. Haroutounian S, Ratz Y, Ginosar Y et  al (2016) The effect of medicinal cannabis on pain and quality-of-life outcomes in chronic pain: a prospective open-label study. Clin J Pain 32(12):1036–1043

62. Campbell G, Hall W, Nielsen S (2018) What does the ecologi- cal and epidemiological evidence indicate about the potential for cannabinoids to reduce opioid use and harms? A comprehensive review. Int Rev Psychiatry (in press, accepted 6 Aug 2018)

63. Klieger SB, Gutman A, Allen L, Pacula RL, Ibrahim JK, Burris S (2017) Mapping medical marijuana: state laws regulating patients, product safety, supply chains and dispensaries, 2017. Addiction 112(12):2206–2216

64. Dowell D, Haegerich TM, Chou R (2016) CDC guideline for pre- scribing opioids for chronic pain—United States, 2016. JAMA 315(15):1624–1645

65. NSW Health (2018) Chronic pain management: information for medical practitioners. https ://www.healt h.nsw.gov.au/pharm aceut ical/docto rs/Pages /chron ic-pain-medic al-pract ition ers.aspx. Accessed 29 Oct 2018

66. Oliveira CB, Maher CG, Pinto RZ et al (2018) Clinical practice guidelines for the management of non-specific low back pain in primary care: an updated overview. Eur Spine J 27(11):2791–2803

67. Häuser W, Fitzcharles M-A, Radbruch L, Petzke F (2017) Can- nabinoids in pain management and palliative medicine. Dtsch Arztebl Int 114(38):627–634

European Archives of Psychiatry & Clinical Neuroscience is a copyright of Springer, 2019. All Rights Reserved.