Acupuncture and Headache Relief: Can It Reduce Migraine Frequency?

If you live with migraines, you already know the toll they take. Beyond the throbbing pain, nausea, and sensitivity to light and sound, migraine is one of the leading causes of disability worldwide, and it disproportionately affects women in their most active years. For many people, standard preventive medications either don't work well enough or come with side effects that make daily life harder, not easier. That's led a growing number of patients — and researchers — to take a closer look at acupuncture as a way to reduce how often migraines strike.

So what does the evidence actually say? Over the past decade, several large systematic reviews and randomized controlled trials have tackled this question head-on. Here's what they found.

The gold-standard review: acupuncture beats "no treatment," and holds its own against drugs

The most authoritative summary of the evidence comes from a Cochrane review, the international gold standard for evaluating medical treatments. Researchers pooled data from 22 trials involving nearly 5,000 people with migraine and compared acupuncture against three things: no preventive treatment, sham (fake) acupuncture, and standard preventive medications.

The results were encouraging. When acupuncture was added to usual care, 41 out of 100 people saw their headache frequency cut at least in half, compared to just 17 out of 100 people who received usual care alone. To put that in concrete terms: if you were having six migraine days a month, the review's data suggest you might expect that to drop to around three and a half days a month with a course of acupuncture, versus roughly five days a month with usual care alone.

Just as notably, acupuncture performed at least as well as prophylactic drugs like beta-blockers, and did so with meaningfully fewer side effects. After three months, 57% of people on acupuncture had their migraine frequency cut in half, compared to 46% on medication — and people receiving acupuncture were less likely to drop out of treatment due to adverse effects.

What about the placebo question?

This is where things get more nuanced. When acupuncture was compared against "sham" acupuncture — needling at incorrect points, or using needles that don't actually penetrate the skin — the difference between real and fake acupuncture was smaller, though still statistically real. About 50% of people getting genuine acupuncture responded, compared to 41% getting sham acupuncture.

Why is the gap between real and fake acupuncture narrower than the gap between acupuncture and no treatment at all? Researchers offer a few explanations, and they're not mutually exclusive. Sham acupuncture may not be a true "do-nothing" placebo — even a needle that doesn't fully penetrate the skin can stimulate nerve endings and trigger some of the same physiological responses as real acupuncture. It's also possible that acupuncture is simply a particularly powerful placebo, given the ritual, the one-on-one attention, and the expectation of relief that come with the treatment itself. Either way, several more recent and larger trials have found the real-versus-sham difference to be statistically significant, even if modest.

More recent research strengthens the picture

Since the original Cochrane analysis, several newer studies have added weight to the case for acupuncture:

A 2020 systematic review focused specifically on acupuncture versus pharmacological prevention (drugs like flunarizine, metoprolol, valproate, and topiramate) across nine trials and found that acupuncture was mildly more effective at reducing migraine days and pain intensity, and was associated with a much lower dropout rate — both from treatment in general and specifically due to side effects.

A 2022 review of 15 randomized trials reached a similar conclusion: seven out of ten trials comparing real acupuncture to sham found a meaningfully bigger drop in attack frequency and pain intensity with real acupuncture, and in trials comparing acupuncture to medication, acupuncture matched or beat the drugs while causing fewer adverse effects. Side effects from acupuncture itself were generally mild — things like minor bruising or brief soreness at the needle site — and no serious adverse events were reported across any of the included trials.

One large single trial worth mentioning, published in the Canadian Medical Association Journal, tested whether it mattered which acupuncture points were used — comparing "migraine-specific" points against other points and against sham treatment in 480 patients. It found that the specific points chosen made little difference, but that acupuncture overall produced a modest, statistically significant benefit over sham treatment by around three to four months. This suggests that some of acupuncture's effect may come from the general experience of treatment rather than the precise location of each needle — an important, if slightly humbling, finding for the field.

How much acupuncture is actually needed?

If you're considering trying acupuncture, a natural question is: how many sessions, how often, and for how long? A 2024 analysis pooled data from 32 trials and more than 1,500 patients specifically to answer this "dosing" question, and it found a clear pattern.

The sweet spot appears to be around 16 acupuncture sessions, delivered three times per week, over a total treatment period of about 1.5 to 2 months. Benefits started showing up after as few as seven sessions, climbed steadily, and then leveled off — meaning that piling on many more sessions beyond that point didn't add much extra benefit. In practical terms, this points to a fairly intensive but time-limited course of treatment, rather than an indefinite, ongoing commitment.

Is it safe?

Across the studies reviewed here, acupuncture had a strong safety record. The most common side effects were minor and local — slight bleeding, bruising, or tenderness where the needles were inserted — and these typically resolved on their own within a day or two. Serious adverse events were rare to nonexistent. By comparison, several trials noted that people on prophylactic medications experienced more frequent and more bothersome side effects, including nausea, drowsiness, weight gain, and in some cases needing to stop treatment altogether.

The bottom line

The research on acupuncture for migraine prevention isn't perfect — many trials are hard to fully blind (patients often know whether they're getting a needle inserted the "normal" way or not), and study quality varies. But the overall pattern across multiple independent reviews is fairly consistent: acupuncture appears to meaningfully reduce migraine frequency, performs comparably to standard preventive medications, and does so with a notably better side-effect profile. The effect over sham acupuncture is real but modest, suggesting that both the physical stimulation and the broader treatment experience likely play a role.

For someone who has struggled with medication side effects, or who is simply looking for a non-drug option to try alongside their current care, a course of roughly 16 sessions over one and a half to two months represents a reasonable, evidence-informed starting point to discuss with a healthcare provider.