Chronic low back pain is the leading cause of disability worldwide. Reducing both the personal suffering of patients and the societal economic burden of chronic low back pain is of major concern. However, management of the condition presents significant clinical challenges. Researchers are studying whether medical cannabis has a role for some people whose pain has not improved with standard treatments. Until now, very few studies have followed people for more than 2 years.
A study, published in the European Spine Journal in 2026, followed 1,000 adults with chronic low back pain for 10 years after they started medical cannabis at a specialist pain clinic.
Key Takeaways:
- The study followed 1,000 adults with long-term low back pain linked to problems seen on a scan, such as narrowing of the spinal canal or worn spinal discs. All were taking opioid painkillers every day and had never used cannabis before.
- 638 of the 1,000 people (63.8%) were still in the study at 10 years.
- In these 638 people, average daily opioid use fell by about 90% and average pain scores fell from 8.7 to 1.4 out of 10.
- Day-to-day function improved more slowly. On average, people moved from severe to moderate disability, with most of the change after year 7.
- Side effects such as dry mouth and stomach upset were recorded at 11.4% of follow-up visits. The authors describe most as mild.
- There was no comparison group. The changes were much larger than those seen in placebo-controlled trials. The authors say the results are “hypothesis-generating rather than practice-changing” and need testing in randomised trials.
What is chronic low back pain?
Chronic low back pain is pain in the lower back that lasts for 3 months or longer. In 2020, low back pain affected around 619 million people worldwide. This is expected to rise to more than 800 million by 2050.
For nearly everyone with low back pain, doctors cannot find one specific cause. This is called non-specific low back pain. A smaller group have a structural cause that shows up on a scan. Examples include narrowing of the spinal canal (spinal stenosis) and wear of the discs between the bones of the spine (disc degeneration). Everyone in this study belonged to this second group.
Most new episodes of back pain settle quickly. For some people, the pain keeps coming back or becomes long-lasting. Back pain is shaped by physical, psychological and social factors. Anxiety, low mood and fear of movement can all affect how strong pain feels and how much it limits daily life.
How is chronic low back pain usually treated in the UK?
Treatment usually starts with approaches that do not involve medicines. The National Institute for Health and Care Excellence (NICE) guideline on low back pain (NG59) recommends that doctors consider:
- Exercise programmes, such as group exercise classes
- Manual therapy, such as manipulation or massage, but only alongside exercise
- Talking therapies that use a cognitive behavioural approach, alongside exercise
- Combined physical and psychological programmes for people whose pain persists
For medicines, NICE suggests anti-inflammatory painkillers (NSAIDs) at the lowest dose that works, for the shortest possible time. NICE advises against using opioids, paracetamol on its own, antidepressants or anticonvulsant medicines to manage chronic low back pain.
Your GP, physiotherapist or pain team can talk you through which options may suit you.
Where does medical cannabis fit in?
Chronic pain is the most common reason people use medical cannabis around the world. Cannabis contains compounds called cannabinoids. The two most studied are THC (delta-9-tetrahydrocannabinol) and CBD (cannabidiol).
These compounds act on the endocannabinoid system. This is a network of receptors and signalling molecules in the body that helps to regulate pain, mood, sleep and inflammation. THC binds directly to cannabinoid receptors. CBD binds to them only weakly and acts in other ways.
Research on medical cannabis for chronic low back pain is still limited. A 2020 review found no randomised controlled trials in this condition. One large trial has been published since. Most other studies have followed people for less than 2 years.
In the UK, NICE does not recommend cannabis-based medicines for managing chronic pain in the NHS based on cost-effectiveness (NG144). Most cannabis-based medicines are unlicensed. A doctor on the General Medical Council (GMC) Specialist Register can still prescribe one when they judge that other treatments have not given enough relief or are not suitable for other reasons.
What did the 10-year study look at?
This was a single-arm observational study. Everyone in the study received medical cannabis, and there was no comparison group. The researchers recorded how people got on each year for 10 years.
Who took part?
The study included 1,000 adults who started treatment during 2015 at a specialist orthopaedic pain clinic in Israel. To take part, people had to:
- have chronic low back pain with a structural cause confirmed on a CT or MRI scan
- have taken opioid painkillers continuously for at least 1 year
- have never used cannabis, checked by history and a urine test
- be aged 18 or over
- have not had enough relief from standard treatments, including physiotherapy, NSAIDs and opioids
People were not eligible if they had active cancer, were pregnant, or had a severe mental health condition. People with a substance use disorder (other than prescription opioid use) were also excluded.
At the start, the average age was 49 and 640 of the 1,000 people (64.0%) were men. On average, people had lived with back pain for nearly 10 years. Their main diagnoses were:
- Spinal stenosis, in 486 people (48.6%)
- Disc degeneration, in 269 people (26.9%)
- Vertebrogenic low back pain, which is pain coming from the bones of the spine, in 229 people (22.9%)
Pain was severe at the start, with an average score of 8.6 out of 10. Many people were also taking several other medicines.
How was medical cannabis provided?
Doctors prescribed and supervised treatment in line with national rules. Products came from licensed producers. They included dried flower for vaporising (heating without burning) and cannabis oils taken by mouth. Most products were THC-dominant.
Treatment started at a low dose and was increased gradually over 2 to 4 weeks. Patients were taught how to use the products, which side effects to look out for, and the rules on driving. Other medicines, including opioids, could be continued or reduced at the doctor’s discretion. There was no set plan for reducing them.
What did the researchers measure?
Research staff who were not involved in treatment collected data once a year. They recorded:
- Opioid use, the main outcome. All opioid doses were converted to the equivalent dose of morphine, in milligrams per day.
- Pain intensity, on a scale from 0 (no pain) to 10 (worst pain imaginable)
- Disability, using the Oswestry Disability Index (ODI). This questionnaire asks how back pain affects everyday activities. Scores run from 0% to 100%, and higher scores mean more disability.
- How much cannabis people used
- Side effects
- Use of other medicines
Before the study began, the researchers set the smallest change that patients would notice as meaningful. This is called the minimal clinically important difference (MCID). The thresholds were:
- Opioid use falling by half or more
- Pain falling by 30% or more
- Disability score improving by 10 points or more
What did the study find?
How many people stayed in the study?
Of the 1,000 people who started, 638 (63.8%) completed 10 years of follow-up. The 362 people who left gave these reasons:
- personal choice, including moving to an area without access to cannabis (42.0%)
- losing contact with the study team (33.3%)
- moving away (18.2%)
- death from causes not related to back pain, such as heart disease (6.5%)
The authors report that no one left the study because of side effects, lack of benefit, or cannabis use disorder. People who completed the study were similar at the start to those who left. The results below are for the 638 people who completed 10 years.
Did opioid use change?
Yes. Average daily opioid use fell by about 90%, from the equivalent of 62.8 mg of morphine a day to 6.4 mg a day at 10 years. Most of the fall happened in the first year, when use dropped to 3.1 mg a day. Use rose slightly after that but stayed well below the starting level.
- 582 of 638 people (91.2%) reduced their opioid use by half or more.
- 42 of 638 people (6.6%) stopped opioids completely.
Did pain change?
Yes. Average pain scores fell by 84%, from 8.7 to 1.4 out of 10. Most of the fall happened in the first year.
- 616 of 638 people (96.6%) had a fall in pain of 30% or more.
- 569 of 638 people (89.2%) had mild pain or less (a score of 3 or below) at 10 years.
Did day-to-day function change?
Yes, but more slowly and to a smaller degree. The average disability score fell from 52.9% to 36.8%, an improvement of 16.1 points. In everyday terms, this is a move from severe to moderate disability.
There was little change in the first 5 years. Most of the improvement came between years 7 and 10. 396 of 638 people (62.1%) improved by 10 points or more.
The authors suggest that regaining function may take years of steady pain relief and gradual reconditioning. This idea has not been tested directly.
Did people need more cannabis over time?
The amount of cannabis people used rose in the first 2 years while doses were adjusted. After that, it stayed broadly steady. The authors report no significant increase in dose over the 10 years. They suggest this is evidence against people developing tolerance.
By year 10, 62% of people vaporised dried flower, 31% took oils by mouth, and 7% used both. Products contained around 4 times as much THC as CBD, on average.
Did use of other medicines change?
Yes. Use of several other medicines fell, mostly in the first 2 years. These changes were agreed between each patient and their doctor based on symptoms. They were not part of a set study plan.
| Medicine type (people who completed 10 years, n = 638) | At the start | At 10 years |
| Tramadol or tapentadol (opioid painkillers) | 89.7% (572 people) | 5.6% |
| Benzodiazepines (sedatives used for anxiety or sleep) | 78.8% (503 people) | 5.3% |
| SSRIs (a type of antidepressant) | 77.7% (495 people) | 5.8% |
| Gabapentinoids (used for nerve pain) | 31.3% (200 people) | 0.6% |
What side effects were reported?
Over the 10 years, there were 8,089 follow-up visits. Some kind of problem was recorded at 2,142 visits (26.5%). Many of these were reports that pain relief was not good enough, rather than side effects.
True side effects were recorded at 925 visits (11.4%). The most common were:
- dry mouth (5.2% of visits)
- stomach symptoms, such as nausea or changes in appetite (3.6%)
- red or irritated eyes (1.6%)
- low blood pressure (0.4%)
- problems with thinking or concentration (0.3%)
- palpitations (0.2%)
- dizziness (0.1%)
Serious mental health side effects (psychosis or a major mood disorder) were recorded at 2 visits (0.02%). There were no hospital admissions, deaths or treatment withdrawals because of side effects.
No cases of cannabis use disorder were identified. However, the researchers did not use a validated questionnaire to check for it. People with severe mental health conditions or substance use disorders were excluded, so these findings may not apply to them.
Medical cannabis can cause side effects and can interact with other medicines. Any decision about treatment should be made with a specialist doctor.
How do these results compare with other research?
The improvements in this study were much larger than those seen in randomised controlled trials, where some people receive a placebo.
The largest trial so far was published in 2025. It included 820 adults with chronic low back pain. Over 12 weeks, a cannabis extract reduced pain by 0.6 points more than placebo on a 0 to 10 scale. A 30% fall in pain was seen in 54.1% of people taking the extract and 39.5% of people taking placebo. In the 10-year study, average pain fell by 7.3 points.
For opioid use, a 2021 review of observational studies found an average reduction of 22.5 mg of morphine equivalent a day when cannabis was added. The certainty of this evidence was very low. In the 10-year study, the average reduction was 56.4 mg a day.
The authors suggest some reasons for the difference. Participants had never used cannabis. The long follow-up allowed slow, gradual reductions in opioids. Starting opioid doses were high, which left more room for change. Treatment was supervised by doctors.
However, the authors also state that the size of the changes “likely reflects multiple factors beyond pharmacological effect”. These are explained in the next section.
What are the limitations of this study?
The authors are clear that the results should be read with caution.
- No comparison group. The study cannot show that cannabis caused the improvements. Other explanations include the placebo effect, people’s expectations, other treatments, and the natural course of back pain.
- Regression to the mean. People often seek new treatment when their pain is at its worst. Scores then tend to drift back towards their usual level over time, whatever the treatment.
- Many people left the study. 362 of 1,000 people (36.2%) did not complete 10 years. Those who stayed may have been the people who did best. Extra analyses including all 1,000 people gave similar results, but they cannot fully rule this out.
- One specialist clinic. All participants were referred to a specialist orthopaedic clinic in Israel and had a structural cause for their pain. The results may not apply to people with non-specific back pain, people seen in primary care, or people in the UK, where products and regulations differ.
- No set plan for medicine changes. Opioids and other medicines were reduced at each doctor’s discretion, which makes the changes harder to interpret.
- Some outcomes are not measured. The study did not measure quality of life, ability to work, or psychological symptoms such as anxiety and depression.
- Cannabis use disorder not formally assessed. No validated questionnaire was used.
What does this mean for people with chronic low back pain?
This is the longest study so far of people with chronic low back pain who were prescribed medical cannabis. In this group, long-term treatment under specialist supervision was associated with lower opioid use, lower pain scores and, more slowly, better day-to-day function. The study design means it cannot show whether cannabis caused these changes.
The authors call for randomised trials that:
- compare medical cannabis with other active treatments
- include a wider range of people, including those seen in primary care
- use standardised cannabis products
- measure function and quality of life over the long term
- look at psychological factors
- check for cannabis use disorder using validated tools
Speak to a specialist about your options
Living with chronic back pain can affect work, sleep, relationships and mood. Your GP, physiotherapist or pain team are the best people to talk to first about managing your pain. The charity Arthritis UK also provides free information and support on back pain.
If you take opioids or other medicines for pain, do not change or stop them without speaking to the doctor who prescribes them.
Curaleaf Clinic is a specialist medical cannabis clinic. To find out whether a consultation with one of our specialist doctors may be appropriate for your circumstances, you can complete our online eligibility check. Any treatment decisions are made by a specialist on an individual basis after a full assessment.
Frequently asked questions
Can medical cannabis cure chronic low back pain?
No. Medical cannabis is not a cure for back pain. Research is looking at whether it may help some people manage long-term pain alongside other treatments. The evidence is still developing.
What did the 10-year study find?
The study followed 1,000 people with chronic low back pain who started medical cannabis at a specialist clinic in Israel. Among the 638 people who completed 10 years, average opioid use fell by about 90% and average pain scores fell from 8.7 to 1.4 out of 10. Disability improved more slowly. There was no comparison group, so the study cannot show that cannabis caused these changes.
Did people in the study stop taking opioids?
Most people reduced their opioid use. 582 of 638 people (91.2%) halved their dose or more, and 42 of 638 (6.6%) stopped completely. These changes were made with their doctors over time. Do not reduce or stop opioids without speaking to your prescriber.
What side effects were reported?
True side effects were recorded at 11.4% of follow-up visits. The most common were dry mouth, stomach symptoms and red eyes. Serious mental health side effects were recorded at 2 of 8,089 visits. People with severe mental health conditions were not included in the study.
Why are the results so different from clinical trials?
In a placebo-controlled trial of 820 people with chronic low back pain, a cannabis extract reduced pain by 0.6 points more than placebo on a 0 to 10 scale. The 10-year study had no comparison group. The authors say its much larger changes probably reflect several factors, including the placebo effect, the natural course of back pain, and the fact that people who did well were more likely to stay in the study.
What does NICE say about cannabis and chronic low back pain?
NICE does not recommend cannabis-based medicines for managing chronic pain in the NHS. For low back pain, NICE recommends exercise first, sometimes with manual therapy or talking therapy, and advises against opioids for chronic low back pain. You can read more on our chronic pain page.
Is medical cannabis legal in the UK?
Yes. Since November 2018, specialist doctors in the UK have been able to prescribe cannabis-based medicines when other treatments have not worked well enough. Most of these medicines are unlicensed, and they are only available on prescription following a specialist assessment.
