Does hypnosis work?
The evidence is strongest for pain and anxiety, mixed for sleep, and thin for focus. Effects in trials ranged from moderate to large depending on the goal and the study, and were larger for people who respond more to hypnosis. Below is what GodHelm App's session is built on, how strong each finding is, and where it comes from.
What studies found, by goal
| Goal | What studies found | Strength | Source |
|---|---|---|---|
| Anxiety | Hypnosis beat control conditions, effect size d = 0.79 at the end of treatment (17 trials), and 0.99 at the longest follow-up. | Good | Valentine et al., 2019 |
| Pain | Effect sizes of 0.54 to 0.76 across 85 trials. Pain fell 42% in highly responsive people and 29% in moderately responsive people, with very little change for low responders. | Strong | Thompson et al., 2019 |
| Hypnosis added to CBT | CBT with hypnosis did better than CBT alone, by at least d = 0.53. | Good | Kirsch et al., 1995 |
| Stress, per session | In 84,395 users of another hypnosis app, stress fell about 1.5 points out of 10 in each session, by as much in the tenth session as in the first. Observational, with no control group, run by that app's own team. | Moderate | Tran et al., 2025 |
| Sleep | 58% of 24 studies found a benefit. The better-designed studies were less positive. | Mixed | Chamine et al., 2018 |
| Focus | Only small studies, mostly about test anxiety rather than focus itself. | Thin | No strong source found |
| Calling it hypnosis | The same procedure was rated 2 to 3 points deeper (out of 10) when it was called hypnosis. Expectation is part of how it works. | Good | Kekecs et al., 2025 |
Does it matter how responsive you are?
Yes. People differ in how strongly they respond to hypnosis, and the stronger responders gain more, most clearly for pain. Lower responders still benefit, but less. That is why GodHelm App's first session measures responsiveness with a spoken, remote adaptation of Spiegel's Hypnotic Induction Profile (scored 0 to 8, not the full clinical test), and guides low, middle and high responders differently.
Is self-hypnosis safe?
For most people, yes. Brief after-effects such as a headache, drowsiness or dizziness are common and pass within minutes to hours. Serious reactions are rare and have occurred almost only in clinical or stage settings, mostly in vulnerable people (Gruzelier, 2000; Trance and Treatment, chapter 1). GodHelm App is designed around those findings:
- Screened first. Anyone who has had seizures, psychosis or a dissociative condition, or is in crisis, is offered relaxation only, with no hypnosis.
- A way out at any moment. Say stop or I'm scared, press Stop, or press Escape on a keyboard. The guide grounds you and brings you out gently.
- No digging. No age regression and no searching through memories.
- A checked wake-up. The session ends only when you say your head is clear, and asks about headache or dizziness.
What is not shown yet
Whether GodHelm App itself helps. The studies above are about hypnosis in general, not about this app, and GodHelm App has not been clinically validated. The app asks you to rate your goal from 0 to 10 before and after each session, so you can see how you felt. Those ratings are self-reported, with no control group, so they cannot show that the app works, or that it works better than other apps or recordings.
The method
Herbert Spiegel and David Spiegel, Trance and Treatment: Clinical Uses of Hypnosis, 2nd edition. GodHelm App is independent and not affiliated with David Spiegel, Stanford University or Reveri.