View commentaries on this research

This is a plain English summary of an original research article

Acupuncture was about as effective as long term medication in reducing the number of migraines. There were fewer adverse events amongst people receiving acupuncture (16-17%) compared to drug treatment (34%).

Migraines affect around one in seven people in the UK. Their unpleasant symptoms last between four and 72 hours and can impact on people’s ability to do everyday tasks, such as going to work.

This systematic review looked at acupuncture delivered at least once a week for up to six sessions, similar to the NICE recommendation of up to ten sessions over five to eight - weeks.

Acupuncture may be a preventative treatment option for those willing to have it but NHS provision of acupuncture is variable. This treatment may, in reality, only be available to people living in certain areas or who can afford to pay for treatment themselves.

Why was this study needed?

Migraines are common, affecting around one in five women and one in 15 men. Symptoms include headache, visual disturbances and nausea, which can be severe enough to stop people living their normal daily life. Migraine attacks last anywhere between four and 72 hours and are estimated to cost the UK economy as much as £2 billion a year in lost work.

The cause of migraines remains unclear. Some people can identify triggers such as certain foods, which they can avoid to help prevent migraine attacks. Others take medications to prevent their migraines. Not all migraine sufferers can identify triggers to prevent migraines and they may not want to constantly take medication. Therefore, they may consider other ways to prevent their migraines, such as acupuncture.

This systematic review looked at the effectiveness of acupuncture for preventing migraines.

What did this study do?

This systematic review compared the findings of 22 randomised controlled trials. In the included studies acupuncture was compared to routine care, no treatment, preventive medication or sham acupuncture – which mimics the sensation of acupuncture. The course of acupuncture had to comprise six sessions delivered at least once a week.

This review looked at people who had experienced episodic migraines for over a year. Studies investigating chronic migraine – where people had 15 or more days with a migraine every month – were excluded from the analysis.

The evidence for reducing migraines was graded as moderate quality overall. The review is therefore likely to provide a reasonably close estimate of the true effect.

What did it find?

  • Acupuncture at least halved the frequency of migraines in 41% of people, compared to 17% amongst those receiving no acupuncture (relative risk [RR] 2.40, 95% confidence interval [CI] 2.08 to 2.76) based on four trials including a total of 2519 people.
  • Acupuncture at least halved the frequency of migraines in 50% of those receiving acupuncture compared to 41% receiving sham acupuncture (RR 1.23, 95% CI 1.11 to 1.36) based on 14 trials including a total of 1,825 people.
  • Acupuncture at least halved the frequency of migraines in 57% of people, compared to 46% of those receiving drug treatments after three months (RR 1.24, 95% CI 1.08 to 1.44). There was no difference between acupuncture and drug treatments at six months (RR 1.11, 95% CI 0.97 to 1.26), based on three studies including 743 people at immediate follow-up and 744 at three months.
  • The proportion of people experiencing adverse events associated with their treatment was similar between acupuncture (16%) and sham acupuncture (17%), whereas it was much lower for acupuncture (17%) compared to drug treatment (34%). There was insufficient reporting of adverse events in trials comparing acupuncture with no acupuncture or usual care, so no analysis was possible.

What does current guidance say on this issue?

NICE 2015 guidelines recommend that acupuncture can be used to prevent migraines (with or without aura) when the first choice medications topiramate and propranolol have either been ineffective or are otherwise unsuitable.

NICE recommends a course of up to ten sessions of acupuncture over five to eight weeks. Whether to use preventive treatments and which treatment to use should be discussed with the individual, considering their personal preference and any other medical conditions they may have.

What are the implications?

These findings suggest that acupuncture was more effective at reducing migraines than usual care, sham acupuncture and drug treatments. Fewer adverse events were reported for acupuncture – half the number for migraine medications – suggesting that it is a safe treatment.

NICE recommends acupuncture for migraine prevention. This review reinforces that recommendation and may help to identify potential candidates for acupuncture – for example people who suffer side effects from medication.

The availability of acupuncture on the NHS varies throughout the UK, therefore any GPs considering recommending acupuncture would have to consider local NHS access and the ability of their patients to pay to access this treatment.


Citation and Funding

Linde K, Allais G, Brinkhaus B, et al. Acupuncture for the prevention of episodic migraine. Cochrane Database Syst Rev. 2016;(6):CD001218.

Cochrane UK and the Pain, Palliative and Supportive Care Cochrane Review Group are supported by NIHR infrastructure funding.



NICE. Headaches: Diagnosis and management of headaches in young people and adults. London: National Institute for Health and Care Excellence; 2012.

Migraine Action. Key facts about migraine. Leicester: Migraine Action; 2015.

NHS Choices. Migraine. London: Department of Health; 2016.

NHS Choices. Migraine – treatment. London: Department of Health; 2016.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre


  • Share via:
  • Print article


Expert commentary

This good systematic review echoes what is known about pain: If you do nothing (waiting list), then few people have a good result. If you do something (placebo of any sort, including fake acupuncture), then more people do better.This is also the case for reducing the number of migraine attacks by at least 50% a month, an outcome considered useful by patients, and one known to improve quality of life. Success rates were about 2 in 10 with doing nothing (waiting list), 4 in 10 with fake acupuncture, 4-6 in 10 with real acupuncture, and 5 in 10 with drugs. Acupuncture as practiced by an authentic method is possibly helpful, but to only about 1 in 10 people more than fake acupuncture.Professor Andrew Moore, Pain Research, University of Oxford
Back to top