A 2025 systematic review examined 12 randomized controlled trials on cannabis and cannabinoids for orthopaedic and trauma pain. While a modest analgesic signal emerged compared to placebo, the evidence quality was predominantly low, and head-to-head comparisons against standard painkillers were virtually absent. The findings highlight an urgent need for larger, longer, and more rigorous trials before clinical recommendations can be made.

Orthopaedic pain affects hundreds of millions of people worldwide, and the fallout from the opioid crisis has created an urgent clinical vacuum for safer analgesic alternatives. Cannabis and cannabinoid products have entered that space with substantial patient interest and expanding legal access, yet clinicians are left to navigate prescribing decisions in real time with remarkably little high-quality evidence. This review directly confronts the gap between enthusiasm and data, and understanding what it does and does not establish is essential for responsible clinical practice.

Orthopaedic pain, encompassing arthritis, chronic back pain, and trauma-related neuropathic syndromes, represents one of the largest and most therapeutically challenging domains in clinical medicine. The endocannabinoid system is expressed in bone, cartilage, and synovial tissue, providing a biological rationale for cannabinoid analgesia in these conditions. This systematic review by Shah and colleagues searched PubMed and Cochrane for double-blind RCTs published between 2003 and 2023 that evaluated medical cannabis or cannabinoid products for pain relief in orthopaedic and trauma populations. From 422 records, only 12 trials survived the screening process, a number that itself reveals the scarcity of rigorous research.

Across these 12 trials, cannabinoids showed some analgesic benefit relative to placebo, but the evidence quality was predominantly low or very low by GRADE standards. The vast majority of comparisons were against placebo or no treatment rather than against active analgesics such as NSAIDs or opioids, which sharply limits the clinical translation of findings. Adverse effects were generally described as mild to moderate within trial durations, though no long-term safety data were captured. No meta-analysis was performed, so pooled effect sizes are unavailable. The authors acknowledge that the data remain insufficient for practice guidelines and call for larger, longer, more rigorously designed RCTs with standardised dosing, active comparators, and comprehensive adverse event reporting.

Better Than Placebo, But How Much Better? Cannabis and Orthopaedic Pain Relief

The opioid crisis has made the search for safer pain relievers feel urgent, almost desperate. Cannabis has arrived into that vacuum with cultural momentum, growing legal access, and genuine biological plausibility. But urgency is not the same as evidence, and a new systematic review of the best available clinical trials asks the uncomfortable question: do we actually know enough to recommend cannabinoids for the millions of people living with orthopaedic pain? This review, published in Cureus, does something genuinely useful by restricting its analysis to double-blind RCTs, applying GRADE quality ratings and Cochrane risk-of-bias assessments, and focusing exclusively on orthopaedic and trauma populations rather than lumping all chronic pain conditions together. That methodological discipline deserves recognition. It also deserves to be understood alongside the review’s real limitations. Only two databases were searched. No protocol was registered in advance. No meta-analysis was performed. Without quantitative pooling, we are left trying to estimate the average height of a population by reading 12 individual case reports rather than taking standardised measurements. You get a directional sense, but not a reliable number.

The central problem is the comparator. Almost all of these trials measured cannabinoids against placebo or no treatment. That is a necessary first step, but it answers a preliminary question, not the one patients actually need answered. It is the difference between asking whether a new painkiller is better than doing nothing versus asking whether it is better than ibuprofen. The lower bar tells you that a compound has some pharmacological activity. It does not tell you whether it should replace or augment the treatments a patient is already using. Compounding this gap is the absence of minimum clinically important difference thresholds: we do not know, from this review, whether the pain reductions that reached statistical significance would actually feel meaningful to a person with chronic knee arthritis or a healing fracture. And testing a moisturiser for one week tells you very little about whether to use it for a decade-long skin condition. These were short-duration trials being used to draw inferences about chronic, often lifelong, pain syndromes.

In my practice, I would use this review to have an honest conversation, not to write a prescription or refuse one. To a patient, I would say that the evidence shows cannabinoids may take the edge off orthopaedic pain somewhat better than a sugar pill, but that we do not yet know how they compare to existing pain medications, what the right dose or formulation is, or what happens with long-term use. To a colleague, I would say this review confirms what we suspected: the RCT evidence base is thin and does not yet justify displacing NSAIDs, physiotherapy, or guided injections. And to a policymaker, I would say that the field needs investment in large, long-term, head-to-head trials before cannabis can be responsibly woven into orthopaedic prescribing guidelines. The most important lesson here may not be clinical but methodological: a signal of efficacy versus placebo in short-duration trials is necessary but far from sufficient to justify clinical adoption of a new analgesic. Active comparator data, long-term safety, and clinically meaningful outcome thresholds are the non-negotiable evidentiary floor for practice-changing claims.

This review sits squarely at the research-mapping stage of the evidence arc for cannabinoids in orthopaedic pain. It confirms the existence of a modest analgesic signal but cannot position cannabinoids within the existing treatment hierarchy because head-to-head data against NSAIDs, acetaminophen, physiotherapy, or even opioids are essentially absent from the included trials. Clinicians should interpret the review as cataloguing what trials exist and grading their quality, not as synthesising a quantitative answer about efficacy. The GRADE ratings of low and very low for most included evidence mean that future, better-designed trials could substantially shift the conclusions in either direction.

From a pharmacological and safety standpoint, the short-term adverse-effect profile described as mild to moderate is reassuring but insufficient. Cannabinoids interact with multiple receptor systems, carry psychoactive potential that complicates blinding integrity in clinical trials, and may interact with common orthopaedic co-medications including anticoagulants and muscle relaxants. Long-term risks, including cognitive effects and dependence potential in chronic use populations, remain uncharacterised in the included trials. The single most actionable recommendation for clinicians today is to engage patients in explicit shared decision-making when cannabinoid use is being considered for orthopaedic pain, clearly communicating that the evidence base remains preliminary and that established analgesic options with stronger supporting data should not be abandoned prematurely.

This is a systematic review restricted to double-blind randomized controlled trials, placing it near the top of the traditional evidence hierarchy. However, the absence of meta-analytic pooling means it functions as a structured narrative synthesis rather than a quantitative evidence integration. The single most important inference constraint is that without pooled effect sizes and confidence intervals, readers cannot estimate the magnitude or precision of any analgesic benefit, and the predominantly low GRADE ratings mean conclusions remain provisional and subject to revision by future, higher-quality research.

This review’s findings are broadly consistent with the landmark Whiting et al. (2015) JAMA meta-analysis, which identified moderate-quality evidence for cannabinoids reducing chronic pain generally but noted significant heterogeneity and limited orthopaedic-specific data. The Madden et al. systematic review of cannabinoids in musculoskeletal and rheumatic pain similarly concluded that evidence was sparse and quality was low. What the current review adds is a tighter clinical focus on orthopaedic and trauma populations specifically, confirming that even within this narrowed scope, the same fundamental evidence gaps persist. The opioid-sparing potential suggested by Nielsen et al. (2017) remains theoretically attractive but is not directly tested by any of the 12 included trials in this review.

The most consequential analytic choice was the decision not to perform a meta-analysis. Had the authors pooled the 12 included trials quantitatively, even acknowledging heterogeneity through random-effects modelling, readers would have obtained an aggregate effect estimate with a confidence interval, providing a far more interpretable answer about the magnitude and precision of any analgesic benefit. Additionally, expanding the database search to include EMBASE, trial registries such as ClinicalTrials.gov, and grey literature might have identified additional qualifying trials that could have shifted the overall direction or strength of conclusions. Applying minimum clinically important difference thresholds to pain outcomes would have clarified whether statistically significant reductions were genuinely meaningful to patients, potentially tempering or strengthening the directional signal that emerged.

The most likely overinterpretation is reading the review’s language about cannabinoids as a “safer alternative or adjunct to opioid pain management” as an evidence-based clinical recommendation. The included trials do not directly compare cannabinoids to opioids, do not measure long-term safety, and do not establish opioid-sparing effects. This phrasing reflects a plausible hypothesis, not a demonstrated outcome. Similarly, the generally mild adverse-effect profile reported across short-duration trials should not be mistaken for evidence of long-term safety. Extrapolating a few weeks of tolerability data to chronic use in populations with degenerative joint disease or recurrent trauma significantly exceeds what the evidence permits.

This systematic review contributes a focused, methodologically structured catalogue of the 12 existing RCTs evaluating cannabinoids for orthopaedic and trauma pain. It identifies a consistent but modest analgesic signal versus placebo and a manageable short-term adverse-effect profile. It does not establish that cannabinoids are clinically superior to existing analgesics, does not define optimal dosing or formulations, and cannot address long-term safety. For now, this review is best understood as a research roadmap, not a practice guide.

Does this review prove that cannabis works for joint or bone pain?

Not definitively. The review found that cannabinoid products reduced pain scores more than placebo in some short-term trials, but the overall evidence quality was rated low. This means the signal is suggestive but not strong enough to form the basis of a clinical recommendation. Better-designed studies are needed to confirm whether the benefit is real and meaningful.

Is cannabis safer than opioids for orthopaedic pain based on this research?

This review did not directly compare cannabis to opioids, so it cannot answer that question. The adverse effects observed in short-term trials were generally mild, but the trials were too brief to assess long-term safety risks. Until head-to-head comparisons are conducted over longer periods, claims about relative safety remain theoretical rather than evidence-based.

Should I ask my doctor about trying cannabis for my arthritis or back pain?

It is always reasonable to discuss treatment options with your physician. If cannabis is legal in your jurisdiction and you are interested in exploring it, your doctor can help weigh the potential benefits and risks in the context of your specific condition, current medications, and overall health. This review suggests the conversation is worth having, but it does not yet provide the evidence needed to make cannabis a standard recommendation for orthopaedic pain.

Which type of cannabis product is best for bone and joint pain?

This review could not identify an optimal cannabinoid type, dose, or administration route. The 12 included trials used a variety of products including THC, CBD, nabiximols, and nabilone in different formulations. Determining which product works best for which condition will require future trials designed to answer that specific question.

Have thoughts on this? Share it: