Acupuncture for Migraines: What the Evidence Says

What the Cochrane review actually found

The most authoritative synthesis is a Cochrane systematic review of acupuncture for the prevention of episodic migraine, covering 22 trials and almost 5,000 participants. Compared with no prophylactic treatment or routine care alone, true acupuncture produced at least a 50% reduction in headache frequency in around 41% of patients, versus about 17% with routine care — a clear, clinically meaningful difference. Compared with sham acupuncture (needling at non-traditional points, used to control for the ritual of treatment), true acupuncture still came out ahead, though by a smaller margin — roughly 50% of patients achieved a 50%+ reduction versus 41% with sham, an effect that persisted at follow-up. Compared with drug prophylaxis (the standard preventive medications), acupuncture performed at least as well after three months — around 57% of the acupuncture group achieved a 50%+ reduction in frequency versus 46% on drugs — with fewer reported adverse effects and fewer people dropping out of treatment. The review authors rated the overall quality of this evidence as moderate, which is a notably stronger rating than acupuncture receives for many other conditions (Linde et al., Cochrane Database of Systematic Reviews, 2016).

It is worth being precise about what this does and doesn't show. The advantage over sham acupuncture, while real and statistically significant, is modest — sham needling itself produces a meaningful improvement in many trials, which is a reminder that context, expectation and the treatment ritual are doing real work alongside whatever is specific to needle placement. The advantage over doing nothing, and the rough parity with drug prophylaxis on effectiveness with fewer side effects, are the more clinically striking findings.

What NICE guidance says

UK guidance reflects this evidence base. NICE's guideline on headaches in over-12s (CG150) recommends considering a course of up to 10 sessions of acupuncture over 5 to 8 weeks for migraine prophylaxis specifically where standard preventive medications — propranolol, topiramate and amitriptyline — have not worked, are not tolerated, or are unsuitable because of safety concerns, taking into account the person's preference, other health conditions and risk of side effects. The same guideline separately recommends a course of up to 10 sessions of acupuncture over 5 to 8 weeks for chronic tension-type headache prophylaxis (NICE, Clinical Guideline CG150). In practice this places acupuncture as a considered option once first-line preventive medication has been tried and hasn't worked out, rather than as a default first step — worth knowing before you go in, so you have realistic expectations about where it sits in the treatment pathway.

What a treatment course involves

• A course of around 10 sessions over 5 to 8 weeks is the pattern reflected in NICE guidance and used in many of the trials in the Cochrane review.
• Sessions are aimed at reducing the frequency of migraine attacks over the course of treatment and beyond, not at stopping an attack that has already started.
• Practitioners typically ask you to keep a headache diary before and during treatment, so any change in frequency, intensity or duration can be judged against your own baseline.
• As with any use of acupuncture, choose someone registered with the British Acupuncture Council, and make sure your migraine diagnosis itself has been confirmed by a GP, since headache red flags always warrant medical assessment first.

Who it may help most, and realistic limitations

The clearest role for acupuncture, based on both the Cochrane review and NICE guidance, is as a prophylactic (preventive) option for people with frequent episodic migraine who have already tried standard preventive medication without success or cannot tolerate it. It is not designed to treat an acute attack once it has started — for that, standard acute treatments (such as triptans, taken early in an attack, as advised by a GP) remain the appropriate approach. It is also worth being realistic that the advantage over sham treatment, while genuine, is moderate rather than dramatic, and that — as with any preventive migraine treatment — a meaningful minority of people will not respond even when the average trial result is positive. Track your own frequency and severity over the full course before judging whether it has made a real difference for you, and keep your GP or headache specialist informed either way, particularly if attacks are changing in character or becoming more frequent, which always warrants review.

FAQ

Does acupuncture actually work for migraines?

The evidence here is comparatively strong for a complementary therapy: a Cochrane review found moderate-quality evidence that acupuncture reduces migraine frequency more than no treatment and more than sham acupuncture, and performs about as well as standard preventive medication with fewer side effects.

Can acupuncture stop a migraine attack once it has started?

No. The evidence and NICE guidance both frame acupuncture as prophylaxis — a course of treatment aimed at reducing how often attacks happen over weeks and months, not a treatment for an attack already underway.

How many acupuncture sessions do I need for migraines?

NICE guidance and most trials use a course of around 10 sessions over 5 to 8 weeks, after which you and your practitioner review whether your migraine frequency has genuinely changed.

Is acupuncture better than migraine medication?

Trial evidence suggests broadly comparable effectiveness to standard preventive drugs, with fewer reported side effects and lower dropout rates — but NICE guidance positions it as an option to consider after standard preventives have been tried, not necessarily as a first choice.