Medical cannabis has become an option for a growing number of conditions, but the evidence suggests its clinical benefits are far more limited than its popularity might imply.

A recent review published in JAMA showed that although medical cannabis may benefit chronic noncancer pain, insomnia, and anxiety, the supporting evidence comes largely from small studies with key methodological limitations.

The review showed that while it may help some patients, current data do not support its use for a number of conditions, said lead author Michael Hsu, MD, clinical professor in the Department of Psychiatry and Biobehavioral Sciences at the University of California, Los Angeles.

However, Hsu cautioned that the study does not mean medical cannabis has no clinical role.

“There are a few conditions that do support the use of medical cannabis, including those approved by the FDA; some others have a moderate degree of evidence, but the jury is still out. It could help and it could not. There just isn’t enough evidence for its support,” Hsu told Medscape Medical News.

Research has also linked cannabis use to adverse health outcomes, including increased risk for stroke, cardiovascular disease, cannabis use disorder, and psychotic symptoms associated with high-potency cannabis.

As questions about the clinical role of medical cannabis persist, the regulatory landscape is also shifting.

In April the US Department of Justice and Drug Enforcement Agency (DEA) issued a directive that immediately moved FDA approved marijuana products distributed under qualifying state medical cannabis programs from Schedule I to Schedule III under the Controlled Substances Act.

Schedule III drugs are recognized as having accepted medical uses and are subject to fewer research restrictions than Schedule I substances. The DEA said the change is intended to facilitate medical research while maintaining federal controls.

The DEA also launched an expedited hearing to consider broader changes to marijuana’s federal scheduling.

The department said the two actions were intended to provide immediate and long-term clarity for researchers, patients, and healthcare providers while maintaining strict federal controls on illicit drug trafficking.

An Urgent Need for Guidance

Regulatory changes may provide greater legal clarity, but they do not resolve the questions clinicians face in clinical practice and given the growing use of cannabis experts say clearer guidance is urgently needed.

Estimates show that nearly 30% of adults in the US and Canada report using cannabis for medical reasons, and about 10% of the US population has used cannabidiol (CBD). The US medical cannabis market is projected to grow by more than 20% annually through the early to mid-2030s.

Currently, 41 states allow the use of medical cannabis, and 24 have also legalized recreational cannabis for adults.

Despite its widespread availability, the FDA has only approved one cannabis-derived medication: Epidiolex (cannabidiol) for the pediatric seizure disorders Lennox-Gastaut syndrome, Dravet syndrome, and Tuberous sclerosis complex (TSC).

The agency has also approved three synthetic cannabinoid drugs: Marinol (dronabinol) and Syndros (dronabinol) for severe nausea and vomiting from chemotherapy and weight loss in patients with HIV/AIDS and Cesamet (nabilone) for chemotherapy-induced nausea.

All of these medications are only available by prescription. Although cannabis has been studied for a wide range of conditions, the FDA has not approved cannabis itself as a treatment for any medical indication. Nor has it approved cannabis-containing products for pain, multiple sclerosis-related spasticity, insomnia, dementia, epilepsy in adults or psychiatric disorders.

Evidence of Efficacy

Even without FDA approval for most indications, research into cannabinoid-based therapies continues to expand.

At the American Academy of Neurology (AAN) 2026 Annual Meeting, investigators presented 6-month interim findings from the 79-patient, phase 3b/4 EpiCom study. Among patients aged 3-42 years with TSC-associated neuropsychiatric disorders (TAND), adding CBD to standard care improved aberrant behaviors, behavioral severity, and irritability after 26 weeks.

“I’m wicked excited because this is the first trial looking at TAND,” study investigator Elizabeth Thiele, MD, PhD, director of the Herscot Center for Tuberous Sclerosis Complex at Mass General Hospital, Boston, told Medscape Medical News.

In addition, a small randomized study presented at the American Society of Clinical Oncology (ASCO) Gastrointestinal Cancers Symposium 2026 showed a link between medical cannabis use and reduced symptom burden in advanced pancreatic cancer, along with minimal side effects.

Cannabis has also shown promise for treating migraine. In findings from the first placebo-controlled trial of vaporized cannabis for acute migraine, participants who inhaled a combination of tetrahydrocannabinol (THC) and CBD reported significant pain reduction and improvement in their most bothersome symptom within 2 hours of treatment.

In another trial, patients with chronic low back pain reported meaningful pain relief after 12 weeks of treatment with a cannabis extract. In a 6-month extension study, nearly three quarters of participants achieved at least a 30% reduction in pain, and more than half reached at least a 50% reduction, with no evidence of dose escalation over time.

A 2024 systematic review and meta-analysis of eight studies involving 316 patients also showed that oral CBD produced large reductions in symptoms of generalized anxiety disorder and social anxiety disorder and related conditions compared with placebo.

However, most of the included studies were small and brief, with treatment durations ranging from 110 minutes to 12 weeks. The review also identified substantial heterogeneity in study design and CBD formulations, making it difficult to draw firm conclusions.

Chronic Pain: An Unmet Need

Other research has reached similar conclusions. An updated Cochrane review published in January showed no high-quality evidence that products high in THC, those primarily containing CBD, or combination formulations were more effective than placebo in reducing neuropathic pain over 2-26 weeks. Although combination products produced small improvements, the benefits were not clinically meaningful.

Likewise a separate analysis published in December 2025 linked products with a high or comparable THC-to-CBD ratio to only modest reductions in neuropathic pain severity, while increasing the risk for adverse events including dizziness, nausea, and sedation.

Julia H. Arnsten, MD, chief in the Division of General Internal Medicine at Montefiore Health System and professor in the Department of Psychiatry and Behavioral Sciences at Montefiore Einstein College of Medicine, New York City, co-authored a December study showing that adults with chronic pain who participated in a state medical cannabis program were less likely to require prescription opioids.

“While we need to look at all the indications for which cannabinoids have been tested, the burning need in this country is for patients with chronic pain,” she told Medscape Medical News.

Still Arnsten said the study findings do not necessarily conflict with the conclusions of the JAMA review. Although her study suggested a potential benefit in chronic pain, she agreed that substantial heterogeneity in cannabinoid formulations, dosing strategies, and study designs underscore the need for larger, high-quality trials before these products can be broadly recommended.

“We do not recommend medical cannabis as a first-line treatment for chronic pain,” Arnsten said. However, she added that it may be considered for patients who have not responded to other therapies and have no contraindications.

Also presented at the AAN 2026 annual meeting, a review of six randomized controlled trials in patients with Parkinson’s disease showed no significant improvement in disease severity, motor function, nonmotor symptoms, quality of life, or depression after treatment with CBD.

Co-investigator Khaled Mohamed, MD, of North Dakota State University in Fargo, North Dakota, said the findings highlight the limitations of the current evidence. “Even if you see a little difference, we can’t rely on that difference” because the quality of the research is not strong, he told Medscape Medical News. “We need more data to be confident in our decisions.”

Broader reviews have reached similar conclusions. A large systematic review and meta-analysis published in Lancet Psychiatry also showed no significant association between cannabinoids and improved outcomes in anorexia nervosa, posttraumatic stress disorder, or opioid use disorder.

The review did identify potential benefits for cannabis withdrawal, Tourette syndrome, insomnia, and autism spectrum disorder traits. However, the certainty of the evidence for most outcomes was low.

What Are the Risks, Benefits?

The American Psychiatric Association and the American Society of Addiction Medicine have also cautioned that cannabis may exacerbate mental illness, particularly in younger users.

Supporting these concerns, a recent review of 99 studies linked high levels of THC with increased risk for cannabis use disorder and psychosis or schizophrenia.

Cardiovascular safety is another concern. A meta-analysis of 24 studies published last June showed a twofold increased risk for major adverse cardiovascular events (MACEs) and a doubling in the likelihood of death from MACEs for cannabis use.

The uncertain balance of benefits and risks is reflected in ASCO’s current clinical guidelines, which conclude that despite some evidence of benefit for cancer pain, the evidence remains insufficient to recommend cannabis-derived products for its management.

In 2025, the American College of Physicians (ACP) suggested that nabiximols, a cannabis-derived oral spray containing THC and CBD, may be considered for neuropathic pain that has not responded to first-line therapies, such as tricyclic antidepressants. However, this recommendation was based on moderate-certainty evidence.

However, the ACP does not extend this recommendation to inhaled cannabis, advising against its use for chronic pain because of concerns about adverse effects, including chronic bronchitis.

Even as clinical guidance remains cautious, researchers are hopeful that the recent federal policy changes will make it easier to generate much-needed higher-quality evidence.

Hsu noted the recent federal rescheduling of certain marijuana products could ease some of the longstanding barriers to cannabis research.

“Some of the burdensome administrative challenges that have made cannabis difficult to study in the US will definitely be mitigated with the rescheduling,” he said. Hsu added that he hopes the change will also reduce funding challenges and some of the institutional caution that has surrounded cannabis research.

Even so, he cautioned that rescheduling alone will not answer many of the remaining clinical questions. It could make large randomized controlled trials more feasible and attract more investigators to the field, he said, but funding remains a major barrier.

Hsu also questioned whether research will ever keep pace with the rapidly expanding cannabis marketplace. Although pragmatic, real-world studies could provide greater clarity, he said the gap between the standardized products evaluated in clinical trials and the diverse products marketed for medical use is likely to remain.

Talking With Patients

Although research to date has questioned the benefits of medical cannabis for many conditions, some patients continue to report that it helps them, Arnsten said.

Arnsten said clinicians should engage patients in open, informed discussions about medical cannabis rather than dismissing its use outright. “Let’s talk about that. What else have you tried? How are you using it? And are you using it safely?” she said.

She noted that medical cannabis can benefit some patients for certain indications. The challenge, she said, is identifying who is most likely to benefit and helping those patients use it safely.

Thiele, who has been researching CBD for more than a decade, said she is optimistic about the field’s future.

“I think we’re at the tip of the iceberg in understanding what CBD and other cannabinoids can do to benefit human health,” she said.

Even so, she acknowledged that skepticism remains, including concerns about placebo effects. High-quality, rigorous studies are essential, she said, because many clinicians will remain skeptical until stronger evidence catches up with anecdotal reports.

“And I think that’s fair. We’re big believers in Class I data and big randomized controlled trials,” she said.

Thiele encouraged clinicians to educate themselves about cannabinoid-based therapies so they can have informed conversations with patients and better understand the products they may already be using.

The experts cited in this article had no relevant disclosures.