Medical cannabis is increasingly used by people with chronic neuropathic pain, often after standard treatments fall short. However, a major review update found no reliable, high-quality evidence that such products relieve pain beyond placebo.
Published in Cochrane Database of Systematic Reviews, a rigorous, globally trusted evidence library, the review update examined 21 randomized trials involving more than 2,100 adults—including six new studies published since its last update in 2018—and like last time, concluded that benefits of medical cannabis were inconsistent and side effects were common.
However, the research exposes a bigger problem. Instead of offering clear answers on the effectiveness of therapeutic cannabis, it reveals how weak the evidence base still is. The authors of the review and outside experts have said that there are now many broad evidence summaries resting on too few high-quality clinical trials.
“At present, the quality of most of the trials is too poor to draw firm conclusions,” Winfried Häuser, a pain medicine researcher from Technische Universität München and the review’s lead author, said in a press statement.
For now, the authors concluded, cannabis-based products should be viewed at best as cautious, last-line options after established treatments have failed.
The Evidence Gap
Neuropathic pain, which arises from nerve damage rather than tissue injury, affects an estimated one in 10 adults—roughly 16 million people in the United States. It can stem from diabetes, shingles, HIV, multiple sclerosis, or nerve injuries and is notoriously difficult to treat.
Even standard first-line drugs such as gabapentin and certain antidepressants help only a minority of patients, with about one-third achieving a meaningful reduction in pain.
Researchers tested three broad categories of cannabis-based medicines: products dominated by tetrahydrocannabinol (THC); products containing a balance of THC and cannabidiol (CBD), including nabiximols, an oral spray used in several countries; and CBD-dominant products containing little or no THC.
The review found that cannabis-based medicines did not increase the number of people who achieved meaningful pain relief—typically defined as a 50 percent reduction—compared with a placebo. The products studied were regulated, standardized formulations such as prescription sprays and oral cannabinoids, not unregulated products sold online or on the street.
A few patients taking products containing both THC and CBD experienced around a 30 percent reduction in pain, but the effect was limited to a small number of people. The authors judged the improvement to be weak and would unlikely extend to most patients.
CBD-dominant products, often marketed as gentle or broadly therapeutic, showed no clear benefits for pain relief at any threshold.
In contrast, THC has psychoactive effects, and regular use is often associated with addiction.
Why Patient Experience Does Not Reflect Research
As demand for medical cannabis has grown, science has struggled to keep pace. Around 30 percent of adults with chronic pain, in states with medical cannabis laws, use it to manage their pain. Millions of Americans are effectively self‑medicating with cannabis, even though solid evidence that it works has not kept up.
“There’s always this question of whether absence of evidence is evidence of absence,” Kevin Boehnke, a pain researcher at the University of Michigan Medical School, who was not involved in the review, told The Epoch Times. “And that’s an issue that has plagued the therapeutic cannabis literature for as long as I’ve been in this space.”
Many of the trials on cannabis use were small and short, sometimes lasting only a few weeks. Study methods were often imperfect, with incomplete reporting and a higher risk of bias. Results varied widely across studies, making it difficult to draw firm conclusions.
Of the 21 studies the Cochrane review evaluated, the authors noted that only five would meet today’s higher standards for study design and reporting.
Some of the uncertainty reflects a deeper mismatch between clinical trials and real-world use.
“The way people use cannabis in the real world is much different than how it looks in a clinical trial, Boehnke said. In studies, participants are typically given a single product, at a fixed dose, on a rigid schedule. In everyday life, people often mix products and routes—edibles, vaping, topicals—and combine THC with CBD or other cannabinoids, adjusting use over time based on symptoms.
That disconnect, he said, makes it harder for trials to capture how cannabis is actually used, or who might benefit.
Longstanding regulatory barriers have also slowed progress. For years, cannabis’s classification as a Schedule I substance—defined as having a high potential for abuse and no currently accepted medical use—limited funding and constrained study design, slowing the development of large, long-term trials that could provide clearer answers.
Concerns remain about cannabis as a psychoactive, meaning it alters how people think, feel, and perceive the world. An effect that carries risks for some users, especially those with mental health vulnerabilities.
“We need larger, well-designed studies with a treatment duration of at least 12 weeks that include people with comorbid physical illnesses and mental health conditions to fully understand the benefits and harms of cannabis-based medicines,” Häuser said.
Boehnke noted that such studies are possible, pointing to recent large-scale trials of therapeutic cannabis use in other pain conditions that followed hundreds of patients for months and used rigorous methods. Changes in cannabis regulation, he said, could open the door to more research of that caliber.
Where the Evidence Is Clearer: Side Effects
If the benefits of cannabis-based medicines were uncertain, the harms were not.
Across trials, nervous system side effects such as dizziness, drowsiness, and cognitive slowing were consistently more common with cannabis-based medicines. For THC-dominant products, about one additional person would experience side effects for every four treated. The risk was even higher for products containing both THC and CBD.
Psychiatric side effects—including confusion, mood changes, or paranoia—were reported more often with THC/CBD-balanced products, though the evidence was uncertain. These products led to a small number of people stopping treatment because of side effects: about 12 percent on THC-containing products compared with 7 percent on placebo, which is roughly one extra dropout for every 20 people treated. For THC‑dominant products, the effect on dropouts was unclear.
Trial participants were carefully selected. People with serious heart disease, seizure disorders, kidney or liver disease, or major mental illness were usually excluded—limiting how well the findings apply to higher-risk patients.
What the Findings Mean for Patients Right Now
For patients, the uncertainty of the benefits of cannabis-based products leaves medical cannabis between widespread use and weak evidence. Some people report mild to moderate benefits—such as small reductions in pain, anxiety, or reliance on other medication—after standard treatments fail.
Boehnke suggests that the gap between widespread use and limited evidence calls for a harm-reduction approach rather than simple dismissal. “If people are going to use cannabis anyway, the goal should be to minimize harm and maximize benefit,” he said.
That typically means beginning with very small doses, increasing gradually, and paying close attention to symptoms and side effects, he said. It also means understanding how different products work—inhaled forms act quickly but can irritate the lungs, while edibles may take hours to fully kick in, a delay that frequently leads people to take too much.
People with heart disease, serious mental illness, or a history of substance misuse should be especially cautious, since people with these conditions were largely excluded from the trials in the Cochrane review.
“No matter what medicine people are trying, they’re really doing an ‘n of one’ experiment on themselves,” he said. “In some people it works, and in some people it doesn’t.”
Professional guidelines largely echo the review’s caution. An International Association for the Study of Pain task force has concluded that current evidence is insufficient to endorse cannabis-based medicines for chronic pain. A European Pain Federation panel suggested they be considered only as third- or fourth-line options, under specialist supervision.
For some patients, the appeal may lie less in superior pain relief than in avoiding the downsides of other options, such as opioids, Boehnke added, a trade-off that remains highly individual.