Hypnosis for Anxiety: What the Evidence Shows

Anxiety would seem like a natural target for hypnosis: the intervention centers on altering subjective experience, and anxiety is a subjective experience that many people want altered. But a plausible rationale is not evidence, and the gap between what you might expect and what trials actually show is revealing.

What the meta-analysis found

A 2019 meta-analysis identified 17 studies testing hypnosis for anxiety outcomes. The pooled effect was large (g = 0.79), suggesting meaningful benefit [1].

That number deserves scrutiny:

• Small studies. Most of the 17 trials had small samples. As meta-epidemiological research has shown, small trials reliably produce larger effects.
• Weak comparisons. Many trials compared hypnosis with a wait-list or no treatment. Against active comparisons the effects were smaller.
• Heterogeneity. The studies covered different types of anxiety (test anxiety, dental anxiety, medical procedure anxiety, generalized anxiety) with different interventions and outcome measures.
• Quality. A 2024 overview rated the methodological quality of hypnosis meta-analyses as variable [6].

The CBT comparison problem

CBT for anxiety disorders has one of the largest evidence bases in all of psychology. A comprehensive meta-analysis found large effects for CBT across anxiety disorders, supported by hundreds of trials [2]. NICE recommends CBT as the psychological treatment for generalized anxiety disorder and panic disorder; hypnosis does not appear in the guideline [5].

The critical question for hypnosis is not whether it beats a wait-list, which is a low bar, but whether it adds to or approaches the effectiveness of CBT. The data here are scarce:

• Kirsch’s 1995 meta-analysis found that adding hypnosis to CBT improved outcomes across conditions, but anxiety was not analyzed separately and the studies were few [3].
• No large trial has directly compared hypnosis with CBT for a specific anxiety disorder.

Without this comparison, hypnosis occupies an odd position: possibly helpful, but tested against the wrong benchmark.

Specific anxieties

Some types of anxiety have been studied more than others:

• Test anxiety has a handful of small trials, generally positive but not replicated at scale.
• Dental anxiety has been tested in several trials, with some evidence of benefit, though sessions are usually brief and the anxiety is situational.
• Medical procedure anxiety has the most data, overlapping with the surgical evidence base discussed in Hypnosis in Surgery.
• Generalized anxiety disorder and social anxiety have very few hypnosis-specific trials.

Relaxation-induced anxiety

An important caution: relaxation-based interventions, including hypnosis, can occasionally increase anxiety rather than reduce it. A study found that relaxation-induced anxiety was more common in people with generalized anxiety disorder than in healthy controls, and was linked to discomfort with reduced control and the unfamiliarity of relaxation itself [4]. Practitioners should screen for this possibility.

What would stronger evidence look like?

Feature	Current state	What is needed
Sample sizes	Mostly under 50	Trials of 100+ per arm
Comparison groups	Mostly wait-list	CBT or another active comparison
Diagnoses	Mixed anxiety presentations	Trials in specific disorders
Follow-up	Typically end-of-treatment	Six to twelve months
Mechanisms	Not studied	Whether hypnotizability predicts outcome

What this means in practice

Hypnosis for anxiety is plausible and sometimes helpful, but it has not been shown to match or exceed treatments with much stronger evidence. A person seeking help for an anxiety disorder should be aware that CBT has guideline-level support and that hypnosis, if tried, should supplement rather than replace evidence-based treatment.

For how to judge a specific trial’s comparison group, see Why Comparison Groups Matter.

When to see a clinician first. Persistent anxiety that disrupts daily life, work or relationships should be assessed by a GP or mental health professional. In England, you can self-refer to NHS Talking Therapies for CBT. If you are in distress, contact emergency services or a crisis line: in the UK, Samaritans on 116 123; in the US, call or text 988.