Hypnosis for pain: the strongest evidence in the field, and its limits
Pain is where hypnosis research is deepest and where the effect sizes are largest. It is also where the gap between the laboratory and the clinic is widest, and being straight about that gap matters more than the headline numbers.
What the evidence says, in short
- The largest meta-analysis pooled 85 controlled trials and 3,632 participants, finding analgesic effects across every pain outcome measured, with effect sizes from 0.54 to 0.76. [1]
- That evidence comes almost entirely from pain induced deliberately in healthy volunteers. The authors say plainly that clinical data is still needed. [1]
- In real chronic musculoskeletal and neuropathic pain, nine randomised trials found moderate reductions in pain intensity and interference, and the benefit was clearly dose dependent. [2]
- A 2025 review of twelve trials found a real effect for acute pain and no measurable effect for chronic pain at all. [4]
- Response depends heavily on hypnotic suggestibility, a stable trait that varies between people. [1] [6]
The headline finding, and what it is actually measuring
The single most cited piece of evidence here is a 2019 meta-analysis that pooled 85 controlled experimental trials covering 3,632 participants. It found analgesic effects of hypnosis across all pain outcomes examined, with standardised effect sizes between 0.54 and 0.76. Those are respectable numbers by the standards of psychological research. The analysis also found that participants who scored high on hypnotic suggestibility showed a 42 percent clinically meaningful reduction in pain, and those in the medium range showed 29 percent. Participants who scored low showed minimal benefit. [1]
There is a crucial qualification attached to those figures, and the authors state it themselves. Almost all of these trials used experimentally induced pain in healthy volunteers. That means cold pressor tasks, pressure, heat, or mild electrical stimulation delivered under controlled conditions to people who do not have a pain condition. The review concludes that high quality clinical data is still needed to establish whether the finding generalises to chronic pain populations. [1]
This distinction is not a technicality. Short, predictable, externally caused pain in a laboratory is a genuinely different experience from pain that has been present for years, carries a diagnosis, disrupts sleep, and shapes what a person is willing to do with their day. An intervention can move the first while barely touching the second.
What happens in real, long-standing pain
The picture in clinical chronic pain is more modest and less consistent. A 2022 systematic review and meta-analysis focused specifically on musculoskeletal and neuropathic chronic pain identified nine eligible randomised controlled trials with 530 participants in total. It found a moderate decrease in pain intensity, with a Hedges' g of -0.42 immediately after the intervention and -0.37 at short-term follow-up, and a similar moderate reduction in pain interference at -0.39. [2]
The most practically useful finding in that review is about dose. Trials delivering eight sessions or more produced a moderate to large effect, with a Hedges' g of -0.555. Trials delivering fewer than eight sessions produced a small effect that was not statistically significant. Repetition appears to be doing real work, which is worth knowing before anyone listens once and concludes it did not help. [2]
An earlier meta-analysis comparing hypnosis with standard care and with other psychological interventions found moderate benefit over standard care, and a moderate advantage over other psychological approaches for people whose pain was not headache related. The same authors flagged large heterogeneity between the studies they pooled, which is a polite way of saying the trials disagreed with each other. [3]
And then there is the result that cuts the other way. A 2025 systematic review and meta-analysis of twelve randomised trials published between 2014 and 2024 found that hypnosis reduced acute and perioperative pain by 0.54 standard deviations, a clear and statistically significant effect. For chronic pain, the same review found a Hedges' g of 0.07 with a p value of 0.518. That is no effect. [4]
Is there a plausible mechanism, or is this just expectation?
Expectation is genuinely part of it, and pretending otherwise would be dishonest. But it does not appear to be the whole story. Reviews of the neurophysiological work report that hypnotic analgesia produces measurable changes in brain and spinal cord function, and that those changes differ depending on which specific suggestion is given. A suggestion aimed at the sensory intensity of pain does something different from one aimed at how unpleasant the pain feels. [5]
That specificity is the most interesting part of the mechanistic case. If the entire effect were generic relaxation or belief that something helpful was happening, you would not expect different suggestions to produce different physiological signatures. It suggests hypnotic suggestion engages the descending systems that modulate pain signalling, rather than simply making people feel calmer about a signal that is unchanged.
Suggestibility complicates this further. It is a fairly stable individual trait, and it predicts who responds. One review makes the point carefully: higher suggestibility is associated with greater relief, but people in the medium range, roughly a third of the population, also get significant benefit, so high suggestibility is not a requirement. The same review is blunt that the available evidence does not support hypnotic pain interventions for people at the low end of the range. [6]
What this means for an audio session you run yourself
Three things follow, and none of them is a promise. First, the technique has a real evidence base, better than most things marketed as wellness audio, and the mechanism is not purely placebo. Second, the trials that worked used courses of sessions rather than single listens, so a one-off is unlikely to reproduce them. Third, individual variation is large enough that some people will notice a genuine difference and others will notice very little, and that difference is not a matter of effort or willpower.
There is also a boundary that no amount of evidence moves. Pain that is new, that has not been assessed by a doctor, or that comes with warning signs such as unexplained weight loss, fever, new numbness or weakness, or changes in bladder or bowel function needs medical evaluation, not a relaxation recording. Mindmosis screens for those signals during the intake and stops rather than generating a session.
Questions
Frequently asked.
- Is hypnosis for pain clinically proven?
- No, and that phrase would be misleading. Hypnotic analgesia has strong support in laboratory studies of induced pain and moderate, inconsistent support in clinical chronic pain, with one 2025 meta-analysis finding no chronic pain effect at all. It is an evidence-informed approach, not a proven treatment.
- How many sessions does the research suggest are needed?
- In the chronic pain meta-analysis, protocols of eight sessions or more produced a moderate to large effect, while protocols with fewer than eight sessions produced a small effect that was not statistically significant. Repetition appears to matter.
- Does it work for everyone?
- No. Hypnotic suggestibility is a stable trait that varies between people, and it strongly predicts response. Around a third of people fall in the medium range and still benefit meaningfully, but the evidence does not support benefit for people at the low end.
- Can this replace my pain medication or my doctor?
- No. Everything described here was studied alongside medical care, not instead of it. Any change to medication is a conversation with the prescribing clinician.
Sources
References.
Every source below is a real, published paper. Links go to the record on PubMed or the publisher, so you can read the abstract and judge the evidence yourself.
- 1Thompson T, Terhune DB, Oram C, Sharangparni J, Rouf R, Solmi M, Veronese N, Stubbs B (2019). The effectiveness of hypnosis for pain relief: A systematic review and meta-analysis of 85 controlled experimental trials. Neuroscience and Biobehavioral Reviews;99:298-310.Experimentally induced pain in healthy volunteers, not patients with clinical pain. The authors call for clinical data to establish generalisability.PubMed PMID 30790634 →
- 2Langlois P, Perrochon A, David R, Rainville P, Wood C, Vanhaudenhuyse A, et al. (2022). Hypnosis to manage musculoskeletal and neuropathic chronic pain: A systematic review and meta-analysis. Neuroscience and Biobehavioral Reviews;135:104591.PubMed PMID 35192910 →
- 3Adachi T, Fujino H, Nakae A, Mashimo T, Sasaki J (2014). A meta-analysis of hypnosis for chronic pain problems: a comparison between hypnosis, standard care, and other psychological interventions. International Journal of Clinical and Experimental Hypnosis;62(1):1-28.The authors report large heterogeneity between the pooled studies.PubMed PMID 24256477 →
- 4Yerzhan A, Ayazbekova A, Lavage DR, Chelly JE (2025). The Use of Medical Hypnosis to Prevent and Treat Acute and Chronic Pain: A Systematic Review and Meta-Analysis. Journal of Clinical Medicine;14(13):4661.Found a significant effect for acute pain and no significant effect for chronic pain.PubMed PMID 40649035 →
- 5Jensen MP, Patterson DR (2014). Hypnotic approaches for chronic pain management: clinical implications of recent research findings. American Psychologist;69(2):167-77.A narrative expert review, not a meta-analysis. No pooled effect sizes.PubMed PMID 24547802 →
- 6Milling LS (2008). Is high hypnotic suggestibility necessary for successful hypnotic pain intervention?. Current Pain and Headache Reports;12(2):98-102.Narrative review.PubMed PMID 18474188 →