Medical cannabis and chronic pain: the strongest evidence base in the field, the opioid-substitution question, and what the DEA rescheduling actually changed

Chronic pain is the single most common qualifying condition across the US medical-cannabis map, listed in 42 state programs. This is the largest qualifying-condition cohort of any diagnosis. It is also, by a wide margin, the indication with the strongest published evidence base for cannabis efficacy. The 2017 National Academies consensus report rated cannabis and cannabinoids for chronic pain in adults at the highest evidence tier the report uses: conclusive or substantial.

This article walks the prevalence (chronic pain affects roughly one in five US adults), the evidence base (multiple high-quality systematic reviews and one Cochrane review), the opioid-substitution debate (more nuanced than the popular framing suggests), the 2022 CDC opioid-prescribing guideline (which left cannabis in an uncertain role), and the 2024 DEA Schedule III proposal (which has structural implications for cannabis pain research even where it does not directly change prescribing practice).

The prevalence and the patient population

The CDC's Chronic Pain Among Adults in the United States, 2019 to 2021 MMWR report estimated that 20.9% of US adults (approximately 51.6 million people) experienced chronic pain during 2021, with 6.9% (about 17.1 million) experiencing high-impact chronic pain that substantially limited daily activities or work. The estimates derive from the National Health Interview Survey and are the most current population-level data available.

The prevalence is not uniform across the population. The MMWR analysis identified higher rates in several subgroups: American Indian and Alaska Native adults; adults identifying as bisexual; adults who are divorced or separated; adults with lower household income; adults living in rural areas. The disparities are substantial and persistent across multiple measurement years.

Against this prevalence, the medical-cannabis program patient population is structurally small. Even in the largest state programs (Florida, Pennsylvania, Illinois, Ohio), chronic-pain patients represent a single-digit fraction of the total US chronic-pain population. The medical-cannabis pathway is not a population-scale chronic-pain treatment. It is a structured option that some patients pursue alongside or as an alternative to conventional pain management.

The state-program landscape: 42 programs, varying scope

Chronic pain is a qualifying condition in 42 US medical-cannabis state programs. The list:

• Northeast: Connecticut, Delaware, Maine, Maryland, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont
• South: Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, Oklahoma, Texas, Virginia, West Virginia, District of Columbia
• Midwest: Illinois, Iowa, Michigan, Minnesota, Missouri, North Dakota, Ohio, South Dakota
• Mountain and West: Alaska, Arizona, California, Colorado, Hawaii, Montana, Nevada, New Mexico, Oregon, Utah, Washington

The terminology varies by state ("severe and chronic pain," "intractable pain," "chronic intractable pain," "chronic pain that has not been responsive to other treatments"), and the scope varies with it. A few states require failed-prior-therapy documentation; others authorize chronic pain as a stand-alone qualifying condition. The condition page at /conditions/chronic-pain provides the evidence summary and FAQ that applies regardless of jurisdiction; the per-state pages above cover the specific qualifying-condition language and program rules.

Notably absent from the chronic-pain qualifying list: Texas authorizes "incurable neurodegenerative disease" and certain other narrow conditions under its Compassionate Use Program but does not list "chronic pain" as a stand-alone qualifying condition. The recent HB 46 expansion (covered in Texas HB 46 ten months in) did widen the qualifying list. Idaho, Indiana, Kansas, Nebraska, North Carolina, South Carolina, Tennessee, Wisconsin, and Wyoming do not operate comprehensive medical-cannabis programs at all.

The evidence base: strongest in the field

The 2017 National Academies consensus report on cannabis and cannabinoids is the most rigorous evidence synthesis in the field. The report classified evidence at five tiers: conclusive, substantial, moderate, limited, and insufficient. For chronic pain in adults, the report's bottom-line finding was this:

Conclusive or substantial evidence that cannabis or cannabinoids are effective for the treatment of chronic pain in adults.

This is the highest evidence tier the NASEM committee assigned to any cannabis indication. The category combines "conclusive" (the strongest evidence) and "substantial" (slightly weaker but still strong) into a single top-tier rating.

The NASEM finding rested on a body of randomized controlled trials concentrated in neuropathic pain: diabetic neuropathy, post-herpetic neuralgia, multiple sclerosis-related pain, HIV-associated neuropathy, and chemotherapy-induced peripheral neuropathy. The evidence for nociceptive pain (tissue-injury pain, post-surgical pain) and mixed pain (back pain, headache) was thinner.

In the years since NASEM 2017, several major systematic reviews and meta-analyses have substantially expanded the chronic-pain evidence:

Whiting et al. (JAMA 2015, PMID 26103030) is the most-cited pre-NASEM systematic review and meta-analysis of cannabinoids for medical use. The chronic-pain analysis found moderate-quality evidence supporting use of cannabinoids for chronic pain and spasticity, with effect sizes in the range of a 30% reduction in pain compared with placebo.

Mücke et al. (Cochrane 2018, PMID 29513392) is a Cochrane systematic review of cannabis-based medicines for chronic neuropathic pain in adults. The review pooled 16 randomized controlled trials with 1,750 participants. The conclusion was cautious: cannabis-based medicines may produce small benefits in pain reduction, sleep, and global impression of change, but the quality of the evidence was downgraded for risk of bias and small effect sizes. Numbers needed to treat were in the range of 20 for moderate pain reduction.

Wang et al. (BMJ 2021, PMID 34497047) is a systematic review and meta-analysis of medical cannabis or cannabinoids for chronic pain (both non-cancer and cancer-related). The analysis included 32 trials with over 5,000 patients. The headline finding was small to very small improvements in pain relief, physical functioning, and sleep quality, with high-certainty evidence for the pain-relief outcome but small effect sizes. The authors noted that any improvement should be weighed against modest harms (cognitive impairment, dizziness, sedation).

Solmi et al. (BMJ 2023, PMID 37648266) is a 2023 BMJ umbrella review of meta-analyses across cannabis and cannabinoid indications. The chronic-pain conclusion was consistent with the prior reviews: there is evidence of benefit, the effect sizes are small to moderate, and the balance of benefits and harms is patient-specific.

The combined evidence shows a real effect of modest size, best documented for neuropathic pain, with a consistent secondary signal for sleep and global functioning improvements. This is not a wonder drug. It is a clinically meaningful adjunct that benefits some patients more than others.

The opioid-substitution debate

The most consequential clinical question for chronic-pain cannabis is whether it functions as an opioid substitute, allowing patients to reduce or discontinue opioids that carry substantially higher dependence, overdose, and mortality risk.

The evidence here is more contested than the headline framing suggests.

Population-level observational studies through the 2014-2019 period suggested that state medical-cannabis legalization was associated with reductions in opioid prescribing rates and opioid-related mortality. These findings drew substantial attention and were cited extensively in state-program advocacy. They have aged less well than initially hoped. Subsequent re-analyses with longer follow-up windows have produced more mixed results, and the early signals were not consistently replicated as state programs scaled.

Individual-patient observational studies consistently report that a substantial fraction of medical-cannabis program patients (often 50-80% in survey samples) report reducing opioid use after starting cannabis. These findings are constrained by self-report, selection effects (patients who tried cannabis and felt it worked report the outcome; patients who tried it and did not benefit may have discontinued program participation), and the absence of randomized comparison.

Randomized clinical trial evidence is much thinner. The 2021 Noori et al. BMJ Open systematic review examined the opioid-sparing question directly. The review included randomized and observational studies. The conclusion was cautious: there is observational evidence of opioid-sparing effects, but the randomized-trial evidence is limited and of low certainty. The authors specifically noted that the observational findings should not be over-interpreted as strong evidence of a substitution effect.

The current consensus view (reflected in the 2022 CDC opioid-prescribing guideline, the VA's clinical-care guidance, and most pain-medicine specialty society positions) is that cannabis may produce opioid-sparing effects in some patients but the evidence does not support cannabis as a routine clinical substitute for opioids in chronic pain. Patients on long-term opioid therapy who introduce cannabis should do so in coordination with their prescribing clinician, both to monitor pain control and to manage potential interactions.

The 2022 CDC opioid-prescribing guideline and what it said about cannabis

The 2022 CDC Clinical Practice Guideline for Prescribing Opioids for Pain (also summarized at CDC's at-a-glance overview) replaced the 2016 CDC guideline that had been controversial for its perceived role in restricting opioid access for legitimate chronic-pain patients. The 2022 update softened several of the 2016 dose-threshold recommendations, emphasized individualized clinical judgment, and explicitly addressed harms from over-rapid opioid tapering.

The guideline's treatment of cannabis was deliberately limited. It recommended non-opioid therapies (including non-pharmacologic interventions and non-opioid medications) as the preferred approach for subacute and chronic pain, but it did not specifically endorse or recommend cannabis as a preferred non-opioid therapy. The guideline-development working group cited insufficient evidence to make a specific recommendation about medical cannabis for chronic pain.

The structural implication is that the CDC guideline left cannabis in an ambiguous role. It is not contraindicated. It is not endorsed. It is a treatment option that clinicians and patients can consider, with the same evidence base and the same caveats that apply outside the opioid-tapering context.

The 2024 DEA Schedule III proposal and its 2026 implementation

The DEA's May 21, 2024 Notice of Proposed Rulemaking initiated the rescheduling process that culminated in the April 22, 2026 DOJ order moving cannabis from Schedule I to Schedule III. For chronic-pain medicine specifically, the rescheduling has three substantive implications:

Research access. Schedule I research access for cannabis has been restrictive for decades, requiring specialized DEA registration and limiting research to NIDA-supplied product. Schedule III status materially eases the research-registration burden. The practical implication: expect more federally permitted clinical trials of cannabis for chronic pain over the next several years, including more head-to-head comparisons with conventional analgesics and more trials of specific cannabinoid formulations.

Clinical-trial infrastructure. Schedule III status places cannabis in the same regulatory category as several existing prescription pain medications (including ketamine and certain anabolic steroids). The infrastructure for running Schedule III pain-medication trials is well established. Investigators and IRBs that have not previously worked with Schedule I cannabis will find Schedule III protocols substantially easier.

No direct change to prescribing. The rescheduling does not authorize physicians to prescribe whole-plant cannabis. FDA-approved cannabis-derived products (Epidiolex, Marinol, Cesamet, Syndros) remain the only cannabis-related substances that can be prescribed under conventional pharmacy-distribution pathways. State medical-cannabis programs continue to operate as state-licensing structures, not as federal prescription pathways.

For chronic-pain patients tracking this: the most important near-term change is the expanded research pipeline, which will accumulate evidence at a faster rate than has been possible under Schedule I. The state-program clinical pathway remains structurally unchanged.

Which pain types respond best

Across the systematic reviews, the strongest signal is in neuropathic pain, meaning pain originating from injury to nerves rather than tissue. The clinical contexts with the most cannabis evidence:

• Diabetic peripheral neuropathy: multiple small RCTs supporting modest pain reduction
• Chemotherapy-induced peripheral neuropathy: evidence from cancer-population trials
• HIV-associated neuropathy: early evidence from the pre-NASEM era
• Multiple sclerosis-related neuropathic pain and spasticity: relatively strong evidence; Sativex (nabiximols) is approved in Europe and Canada for MS spasticity
• Post-herpetic neuralgia: limited but supportive evidence
• Complex regional pain syndrome: case-series and small-trial evidence

For nociceptive pain (tissue-injury pain), the evidence is more mixed:

• Post-surgical pain: limited evidence; cannabis is not a primary post-surgical analgesic
• Inflammatory pain (rheumatoid arthritis, osteoarthritis): mixed evidence
• Cancer pain (mixed mechanism): moderate evidence, particularly for refractory pain

For chronic pain syndromes with mixed or centralized mechanisms:

• Fibromyalgia: limited evidence; small trials suggest possible benefit for sleep and overall functioning, less for direct pain reduction
• Chronic low back pain: mixed evidence; this is the most common chronic pain indication and the evidence quality is generally lower
• Migraine and headache disorders: limited controlled-trial evidence; observational data more positive

The pattern matters clinically. A patient with diabetic neuropathy can reasonably expect modest benefit from a well-designed medical-cannabis trial. A patient with chronic low back pain has less evidentiary support for the same expectation.

What patients should weigh

Five practical considerations:

1. Cannabis is rarely a stand-alone chronic-pain treatment. The best-documented benefits are modest and additive. Patients who use cannabis as part of a multi-modal pain plan (which may include physical therapy, non-opioid medications, behavioral approaches, and selective use of opioids when clinically indicated) typically do better than patients who substitute cannabis for those other components.
2. Product variability matters. State-program cannabis is not FDA-regulated. THC content varies; CBD content varies; product purity varies; terpene profiles vary. Patients should identify operators with rigorous third-party testing and consistent batch documentation, and should expect to spend time finding the chemotype and dosage that works for their specific pain pattern.
3. Start low, go slow. The most consistent clinical advice across cannabis-prescribing literature is to start at low doses (2.5-5 mg THC equivalent or lower for cannabis-naive patients) and titrate upward gradually. Most adverse effects (cognitive impairment, anxiety, dizziness) are dose-related.
4. Smoke is rarely the best route. Inhalation provides rapid onset but variable dosing and respiratory irritation. Edibles, tinctures, and capsules provide more consistent dosing for sustained-pain management. Topicals are useful for localized pain but provide limited systemic effect.
5. The opioid question is patient-specific. Patients on long-term opioid therapy who add cannabis should coordinate with their prescribing clinician. The opioid-sparing question is not settled at the population level, but at the individual level, some patients do reduce opioid use after introducing cannabis. Do not discontinue prescribed opioids abruptly without clinical supervision.

Related reading

• Chronic pain condition overview: evidence summary, FAQ, and safety considerations
• Federal employment and medical cannabis: workforce implications for federally regulated chronic-pain patients
• How to get a medical marijuana card: the qualification process every state has in common
• Federal rescheduling 2026 status: current status of the Schedule III implementation and its clinical implications

[Last reviewed 2026-05-24. This is informational only, not medical advice. Chronic-pain treatment decisions should be made in consultation with a qualified pain-medicine clinician.]