Hypnosis for Procedural Anxiety: What Clinical Trials Measure

The dinner-party sentence for this one is “of course anxious patients feel better when someone sits with them and talks softly; that’s not hypnosis, that’s manners.” I said a version of it for years. It’s half-right, which is my favorite kind of wrong, and the trials that finally shut me up are the ones that tested the manners separately. Out of how many? The biggest of the procedure trials I’m going to walk through had a few hundred patients, and most had far fewer.

I’ve covered the overall surgical evidence in Hypnosis in Surgery and Medical Procedures. This post narrows in on one outcome, anxiety around a procedure, and asks the boring question that turned out to matter most: what exactly did these trials measure, and against what?

What “procedural anxiety” means in these trials

Procedural anxiety is the fear around a specific medical event: the biopsy, the operation, the catheter, the mask going on. It’s short, predictable and tied to a date on a calendar, which makes it unusually well suited to research. You know when the anxiety will peak, you can measure it before and during, and the intervention can be as short as 15 minutes.

What gets measured varies more than the headlines suggest:

• Self-rated anxiety on a simple scale. In the largest breast biopsy trial, women rated anxiety from 0 (none) to 10 (worst possible) at intervals throughout the procedure [1]. Others use a 0-100 visual analog scale for “emotional upset” or similar [4] [5].
• Mood questionnaires. One presurgery trial used a short version of the Profile of Mood States alongside visual analog scales [5].
• Observer ratings in children. Young children can’t fill in a scale, so a pediatric trial used the modified Yale Preoperative Anxiety Scale, rated by staff at arrival, at entry to the operating room and when the mask was fitted [6].
• Indirect outcomes. Sedative and painkiller use, adverse events, procedure time and cost [2] [4]. These are less vulnerable to a patient wanting to please the researcher, which is why I always read them second, right after the control group.

That last point matters because almost none of this can be blinded. The patient knows whether they heard a hypnosis script. So a self-rated anxiety score is exactly the kind of outcome that expectation can nudge, and a medication count is a little harder to nudge.

Who got what: the comparison groups

This is where my old joke died. A trial that compares hypnosis with standard care tells you hypnosis beats the usual routine. A trial that adds a third arm, where someone gives the same warm, structured attention without the hypnotic suggestions, tells you whether the hypnosis part adds anything. Several procedure trials did exactly that.

Procedure, comparison, outcomeWhat each trial compared hypnosis against, and what it found

Procedure and comparison groupAnxiety outcome

Breast biopsy, 236 women: standard care and empathic attention

→

Anxiety rose with standard care, stayed flat with empathy, fell with hypnosis

Tumor treatment, 201 patients: standard care and empathic attention

→

Less anxiety with hypnosis; adverse events 48% with empathy, 12% with hypnosis

Before surgery, 76 adults: standard care and attention control

→

Anxiety down 56% with hypnosis, up 10% with attention, up 47% with standard care

Breast surgery, 200 patients: empathic listening

→

Emotional upset 8.67 vs 33.46 on a 0-100 scale at discharge

Children aged 2 to 11, 50 patients: oral midazolam

→

Anxious at mask induction: 39% vs 68%

Breast biopsy, 236 women. Women were randomized to standard care, structured empathic attention or self-hypnotic relaxation that included the same empathic behaviors plus a script. Anxiety rose during the procedure with standard care, stayed flat with empathy, and fell with hypnosis [1]. Pain rose in all three groups, though less steeply with hypnosis and with empathy than with standard care. That’s a three-way split I didn’t expect: kindness helped, and hypnosis helped more.

Interventional radiology, 201 patients. Same three-arm design, during percutaneous tumor treatment. Hypnosis reduced pain and anxiety compared with both other groups and used fewer medication units. The surprise was in the safety data: 48% of the empathy group had adverse events, against 12% with hypnosis and 26% with standard care [2]. The authors’ subtitle was “when being nice does not suffice.” I’ve reread that result several times, and I still don’t think anyone fully knows why attention alone went worse, but it is a good reason never to assume an attention control is harmless.

Before surgery, 76 adults. Patients were randomized to hypnosis (26), an attention control with supportive listening (26) or standard care (24). By the time they entered the operating room, anxiety had fallen 56% in the hypnosis group, risen 10% in the attention group and risen 47% with standard care [3]. Small trial, single center, but the attention arm is there and it didn’t do the same job.

Breast surgery, 200 patients. A 15-minute presurgery session with a psychologist was compared with nondirective empathic listening, and patients weren’t blinded. On a 0-100 scale at discharge, emotional upset averaged 8.67 with hypnosis against 33.46 with attention control, and the hypnosis group needed less propofol and lidocaine and cost the hospital $772.71 less per patient, mainly through shorter surgery [4].

Breast biopsy, 90 patients. A 15-minute session before surgery, again against attention control. Before the operation, emotional upset on a 0-100 scale was 16.5 with hypnosis against 38.2 with attention (d = .85) [5]. This trial and the 200-patient one came from the same research group, which is worth knowing when you count them as separate confirmations.

Children before anesthesia, 50 patients. This one used an active drug as the comparison: hypnosis against oral midazolam, the usual sedative premedication, in children aged 2 to 11. Fewer children in the hypnosis group were anxious when the mask went on (39% against 68%) [6]. Fifty children is small, and the anxiety ratings came from observers rather than the children, but beating a real sedative on its home turf is not beating the lobby.

Trial setting	Comparison	Main anxiety finding
Breast biopsy, 236 women	Standard care and empathic attention	Anxiety rose, stayed flat and fell, in that order
Tumor treatment, 201 patients	Standard care and empathic attention	Less anxiety with hypnosis; more adverse events with empathy alone
Before surgery, 76 adults	Standard care and attention control	Anxiety down 56% with hypnosis by entry to the operating room
Breast surgery, 200 patients	Empathic listening	Lower emotional upset at discharge, less propofol and lidocaine
Children, 50 patients	Oral midazolam	Fewer anxious children at mask induction

What the meta-analyses add

Individual trials are the receipts. Meta-analyses are the totals, and the totals here are smaller than the best individual trials, which is normal.

• A 2025 meta-analysis of hypnosis during invasive procedures pooled 20 randomized trials with 1,250 patients and found less anxiety than with standard care, a standardized mean difference of -0.43 [7]. That’s a small-to-moderate effect, and it’s against standard care, not attention.
• A 2025 review for anesthetists gathered 142 studies of any design with 9,238 patients and ran meta-analyses of the randomized ones by timing. Hypnosis before the intervention reduced post-intervention anxiety (SMD -0.76), and hypnosis during the intervention reduced it less (SMD -0.44). The same authors concluded that for most other outcomes the impact remains unknown [8].
• A 2014 systematic review of hypnosis before procedures found 18 trials with 968 patients. Fourteen of them had 60 or fewer patients, and 12 of the 18 had major quality limitations such as unclear allocation concealment, uncertain provider experience and problems with intention-to-treat design [9].
• And the broader surgical meta-analysis of 50 trials found that the 95% prediction interval for emotional distress included no effect: a new hospital could plausibly see no benefit at all [10].

That last one is the result I’d tape to every clinic leaflet. The average is encouraging. The spread is wide.

Limitations worth naming

Limitation	Consequence
Patients can’t be blinded	Self-rated anxiety is open to expectation and politeness
Many small trials	Estimates are noisy, and small positive trials are easier to publish
Developer-run trials	The strongest trials often come from the teams that wrote the scripts
Anxiety measured many ways	0-10 ratings, 0-100 scales, mood questionnaires and observer scales are hard to pool cleanly
Mostly standard care as the comparison	The pooled averages don’t isolate hypnosis from attention
Results vary by setting	Prediction intervals suggest some programs won’t replicate the average

What remains uncertain

I’d like to know three things nobody has nailed down. First, how much of the effect survives when the people rating anxiety, in children’s trials especially, don’t know who got hypnosis. Second, whether a recorded script does as well as a trained person in the room, which matters a great deal for whether hospitals can afford this. Third, why attention alone sometimes helps and once went badly. The old me would have said “it’s all just attention.” The trials say attention is part of it, and not the whole of it.

What this means in practice

For anxiety in the hours around a procedure, hypnosis has better evidence than most hypnosis claims I’ve looked at: randomized trials, some with credible attention controls, one against a real sedative, and pooled effects in the small-to-moderate range. It is not a guaranteed calm, the trials are mostly small and unblinded, and results vary a lot between settings. If you’re anxious about an upcoming procedure, asking your anesthesia or procedure team whether hypnosis or a similar structured approach is available is a reasonable request, as an addition to standard care rather than a substitute for sedation.

For how to judge the control group in any of these trials, see Expectation, Placebo and Hypnosis: Why Comparison Groups Matter. For children specifically, see Pediatric Hypnosis: What the Evidence Shows.

When to see a clinician first. Talk to your procedure or anesthesia team before changing anything about sedation, pain relief or preparation; hypnosis in these trials was added to standard care, not used instead of it. If fear of procedures is stopping you from getting tests or treatment you need, or brings on panic attacks, ask your doctor about a referral for anxiety treatment, since established therapies for specific fears exist. Tell your team about any heart, lung or mental health conditions before any procedure.