Acupuncture for migraine prevention: the comparison that changes the answer

If you’ve been offered acupuncture for recurring migraine, the headline numbers can sound persuasive: 41% versus 17%. But that comparison isn’t with a placebo-like procedure. In a separate set of trials comparing real acupuncture with sham acupuncture, the numbers were 50% versus 41%. The difference matters if you’re trying to decide what the needles themselves add - and whether a course of appointments fits alongside your existing prevention plan. Both figures come from a 2016 Cochrane review of adults with episodic migraine, not a guarantee of what will happen to you.

Here, prevention means trying to have fewer headaches over time, not treating a migraine already in progress. The percentages refer to people whose headache frequency fell by at least half after treatment. They aren’t cure rates. And the two control groups aren’t interchangeable: people in the usual-care group could still use treatment for attacks, whereas the sham group went through a simulated acupuncture procedure.

Against usual care: the bigger gap

#

In four trials covering 2,519 people for this outcome, about 41 in 100 assigned acupuncture had their headache frequency at least halved after treatment, compared with 17 in 100 receiving no acupuncture or routine care that could include medication for attacks (Cochrane). That’s a meaningful difference between those groups. It’s not, though, a clean measurement of where a needle goes or how deeply it’s inserted.

People knew whether they were getting appointments. Expectations, the time spent with a practitioner and other parts of care can influence reported outcomes; the NCCIH notes that acupuncture’s mechanisms aren’t fully understood. Traditional explanations of acupuncture are part of its history, not clinical proof of a particular migraine mechanism. The usual-care result tells us something useful about the whole intervention, not the size of a specific needling effect.

Against sham: a smaller difference

#

Sham acupuncture in a trial is a procedure designed to resemble acupuncture without the same needling protocol. It’s not necessarily the same as doing nothing. In 14 trials involving 1,825 people, half of those assigned real acupuncture met the at-least-50%-reduction threshold after treatment, versus 41% assigned sham (Cochrane). A small pooled difference in headache frequency also favored real acupuncture. At the studied follow-up, the responder comparison was 53% versus 42% across 11 trials and 1,683 people - but that doesn’t establish a benefit lasting for years.

So yes, the trials found an edge over sham. Small is the operative word (and sham procedures themselves may have effects). It would be a mistake to recast that finding as “acupuncture works twice as well as placebo” by borrowing the 41%-versus-17% figures from the different, unblinded comparison.

A 2025 systematic review of 23 trials and 2,295 migraine patients also reported fewer migraine days over four weeks versus sham, with a pooled mean difference of 1.38 days. Its included population and outcome measures don’t match the older adult episodic-migraine review exactly. And some underlying trials may overlap, so these two reviews aren’t two wholly independent batches of patients. That average isn’t a promise of one fewer migraine day for each person who tries acupuncture.

Against preventive drugs: no simple winner

#

The older Cochrane drug-comparison trials looked more favorable to acupuncture shortly after treatment. At roughly three months, 57% assigned acupuncture versus 46% assigned preventive drugs met the responder threshold. By around six months it was 59% versus 54%, a difference that wasn’t statistically clear; only three trials contributed useful drug-comparator information (Cochrane). Those comparisons weren’t blinded either. You can’t use them to say acupuncture beats today’s preventive medicines across the board.

A separate 2025 synthesis comparing acupuncture with standard medical care found no statistically significant between-group difference in migraine days or pain intensity, although it reported favorable quality-of-life and analgesic-use outcomes. The studies and protocols varied, and the review wasn’t a head-to-head test against every newer prevention option. An absence of a statistically significant difference isn’t proof the treatments are identical, either.

What would I take into a clinical appointment? Not “which one wins?” but “what’s realistic for my migraine pattern, previous treatments and willingness to attend repeated visits?” That question leaves room for acupuncture without quietly turning it into a replacement for prescribed care.

Where acupuncture fits in a prevention plan

#

NICE’s UK headache guidance, amended in 2025, says to consider a course of up to 10 acupuncture sessions over five to eight weeks for migraine prevention if propranolol, topiramate and amitriptyline have not worked, aren’t tolerated or aren’t suitable because of safety concerns. That’s a conditional guideline option, not a session prescription from this article or a rule that everyone must try three drugs first. Treatment choices also depend on preferences, other conditions and adverse-effect risks. NICE separately flags restrictions on topiramate, including pregnancy-related risks - don’t make medication changes on your own.

It helps to bring a headache diary: when the headaches happened, how long they lasted, how severe they were and what medication you used. If you try acupuncture, the same record gives you and your clinician something more useful than a vague memory of a good week. Some weeks are good anyway. That’s precisely why a before-and-after impression needs a little caution.

The needle-safety conversation

#

Acupuncture is generally considered relatively safe when properly performed, but “relatively” matters. Minor soreness, bleeding or bruising can happen; badly performed treatment or nonsterile needles can cause infection or serious injury (Mayo Clinic; NCCIH). Ask about the practitioner’s credentials and experience with migraine, and whether they use sterile, single-use needles. Don’t try self-needling.

And before booking, tell both your migraine clinician and the acupuncturist about blood thinners or bleeding problems, an immune condition or active skin infection, chronic illness, and any implanted electronic device if electrical stimulation is proposed. Mention pregnancy, breastfeeding or recent childbirth too. These are reasons to discuss individual suitability - not a blanket list of conditions that always rule acupuncture out. Needling doesn’t come with a demonstrated drug-interaction list in these sources; bleeding and procedure safety, along with keeping your migraine care coordinated, are the practical medication concerns (Mayo Clinic; NCCIH). Don’t stop a blood thinner to make an appointment possible.

The adult trial averages don’t establish safety or effectiveness in pregnancy, breastfeeding, children or people with chronic migraine. If you’re pregnant or have recently given birth and develop migraine symptoms, seek medical advice rather than extrapolating from these studies; the NHS specifically calls for urgent advice in that situation. And if a headache is suddenly extremely painful, or comes with new speech or vision trouble, one-sided weakness, seizure or confusion, seek emergency assessment - not an acupuncture booking (NHS). A new or changing pattern deserves a clinician’s assessment too.

Bottom line

#

Acupuncture has randomized-trial evidence for fewer headaches in some adults with episodic migraine. Its advantage over usual care looks larger than its advantage over sham, and comparisons with preventive drugs don’t establish a lasting winner. If the visits appeal to you, discuss them as one possible part of a clinician-guided prevention plan - with your diary, medication list and a few safety questions in hand. Keep your prescribed acute and preventive treatment in place unless the clinician managing your migraine changes it with you.