Hypnosis for Migraine and Tension Headaches: What the Evidence Shows

Migraine affects over one billion people. The drugs are real but imperfect — side effects, rebound headaches, attacks that break through anyway.

So does hypnosis actually help?

Does hypnosis help migraines?

Probably — but the evidence is older and thinner than the confident headlines suggest. Randomized trials going back to 1975 consistently show hypnosis reducing migraine frequency and severity, and a 2026 systematic review of nine studies found it effective across the board. But a rigorous 2025 meta-analysis of behavioral migraine treatments rated the hypnosis evidence insufficient to permit conclusions — not negative, just too small and too weak to be sure. Both readings are true at once, and understanding why is more useful than either headline.

The classic trial: hypnosis vs. an actual drug

The study everyone cites is worth knowing in detail. In 1975, Anderson, Basker, and Dalton (International Journal of Clinical and Experimental Hypnosis) randomized 47 migraine patients to either hypnosis (including self-hypnosis practice) or prochlorperazine — a real prophylactic medication — and followed them for a full year with monthly assessments and independent evaluation.

The hypnosis group had significantly fewer attacks per month and significantly fewer Grade 4 “blinding” attacks than the medication group. The sharpest number: in the final three months of the trial, 10 of 23 hypnosis patients achieved complete remission — versus 3 of 24 on prochlorperazine. Prochlorperazine, the authors noted, performed about as well as the patients’ previous treatments. Hypnosis beat both.

One trial, 47 people, half a century old. But it’s a randomized comparison against an active drug — a higher bar than most wellness interventions ever clear — and it has never been convincingly overturned.

What the reviews say

Two systematic reviews have since pulled the migraine literature together, and both point the same direction.

A 2018 systematic review (Flynn, International Journal of Clinical and Experimental Hypnosis) identified 8 studies of hypnotic techniques for migraine — alone or combined with imagery and relaxation — and concluded hypnosis is effective in reducing both short- and long-term headache activity in migraine sufferers.

A 2026 systematic review (Cardinal et al., Complementary Therapies in Medicine) analyzed 9 randomized or quasi-experimental studies with 406 participants, using hypnosis as a standalone intervention. Results consistently showed hypnosis reducing migraine symptoms, often outperforming other non-pharmacological interventions — with fewer resources required. The same review is blunt about the catch: the studies were too heterogeneous to meta-analyze, sample descriptions were inadequate, and most lacked statistical power calculations.

For tension-type headache, the signal is similar. A 1991 controlled study (Melis et al., Headache) tested hypnotherapy against a waiting-list control in chronic tension-type headache and found significant reductions in headache days, headache hours, and headache intensity (all p < 0.05), plus a significant drop in anxiety scores. And a 2007 review (Hammond, International Journal of Clinical and Experimental Hypnosis) concluded hypnosis for headaches and migraines meets clinical psychology’s research criteria for a “well-established and efficacious” treatment — while being virtually free of the side effects and ongoing costs of medication.

The honest counterweight

Here’s what the enthusiastic summaries skip. A 2025 systematic review and meta-analysis of behavioral interventions for migraine prevention (Treadwell et al., Headache) — 50 trials, 6,024 adults — is the most rigorous synthesis in this space. It found low-strength evidence that CBT, relaxation training, and mindfulness-based therapies may reduce migraine frequency. For hypnotherapy specifically, it judged the evidence insufficient to permit any conclusion.

How does a treatment go from “well-established” in a hypnosis journal to “insufficient” in a headache journal? Standards. The hypnosis trials are old, small, and methodologically loose by modern criteria — unblinded, underpowered, inconsistent outcomes. Reviews from within the hypnosis field weight the consistent positive direction; strict evidence-grading weights the absence of large, well-controlled modern trials. Nobody has run the definitive RCT. That’s the actual state of play: a consistently positive but genuinely under-powered evidence base — promising, cheap, safe, and unproven at the standard we hold drugs to.

Newer trials are appearing but haven’t fixed the size problem. A 2023 RCT (Khazraee et al., Life) randomized 38 women with chronic migraine to mindful hypnotherapy or treatment as usual: headache disability scores fell from 73.6 to 23.3 and diary-rated headache intensity from 7.3 to 2.8 post-treatment (both p < 0.001). Striking numbers — from a small, single-site trial against usual care. Treat it as consistent with the pattern, not proof on its own.

How it would work, mechanically

The plausibility case is solid even where the trial base is thin. Hypnosis has robust, replicated effects on pain processing — the chronic pain literature shows moderate effect sizes across dozens of controlled trials, with suggestibility as the key moderator. Migraine adds two levers: attacks are commonly stress-triggered, and hypnosis reliably downshifts sympathetic arousal — the same reason relaxation training earns a positive rating in the 2025 meta-analysis. Neuroimaging shows hypnotic suggestion modulating the brain regions that assign pain its unpleasantness, which is precisely the pathway a migraine-directed protocol targets.

The honest takeaway

If you’re weighing hypnosis for migraine or tension headaches, the evidence supports it as a low-risk adjunct, not a replacement for medical care. Every controlled comparison points the right way; none is definitive. The practical case is the risk-reward asymmetry Hammond’s review flagged: essentially no side effects, no rebound-headache risk, low cost — against a real chance of fewer and milder attacks. Keep your neurologist, keep a headache diary so you can see your own numbers, and give any protocol several weeks of consistent practice before judging it. For the broader evidence base on what hypnosis can and can’t do, see the science of clinical hypnosis series.