Why isn’t there more research into hypnosis?

Hypnosis has been studied for decades. So why don’t we have clearer answers about how well it works?

There is research. I write about it regularly on this site, and I keep adding to it. Clinical hypnosis has been used for centuries, and modern scientific study of it goes back more than a hundred years. There is now a substantial body of work looking at its effects on pain, anxiety, depression, IBS, sleep, stress, medical procedures and other health problems, including dozens of systematic reviews and meta-analyses.

And yet, if you search for evidence on hypnotherapy, you will often find the same frustrating line on the bottom of the paper: more research is needed.

That isn’t because nobody has bothered to study hypnosis, it’s because hypnosis is unusually difficult to study well. Let me tell you why…

Hypnosis isn’t one thing

The word “hypnosis” covers a surprisingly wide range of interventions. One study might use a formal hypnotic induction followed by direct suggestions. Another might use guided imagery. Another might teach self-hypnosis. Another might combine hypnosis with cognitive behavioural therapy. Another might use hypnosis during a medical procedure.

Researchers do not always describe these interventions in enough detail for anyone else to know exactly what happened. That matters, because a treatment cannot be properly tested if it cannot be reproduced.

Also, imagine a study of “exercise” in which one group went for a gentle walk, another lifted weights and another ran a marathon. You could calculate an average result, but it would tell you very little about which approach actually worked. Hypnosis research has sometimes had a similar problem.

Future studies would be stronger if they published detailed treatment manuals, scripts or recordings, described exactly how hypnosis was induced, and explained precisely what the researchers meant by “hypnosis”.

You can’t easily give someone a placebo hypnosis session

If you are testing a drug, you can give one group the active tablet and another an identical-looking placebo. The participants usually don’t know which they received.

With hypnosis, that is much harder. If I tell you that you are going to have a hypnosis session, you know you are having a hypnosis session. I know it too. Those expectations are part of the process and positive expectancy can help.

So how do you give someone a hypnosis session that is not hypnosis and still create positive expectancy? They will usually know the difference. So will the practitioner. That makes clean experimental design considerably harder than it is with medication.

The therapist is part of the treatment

Drugs do not have a personality, a voice, or a relationship with the person taking them, but a hypnotherapist does.

Tone, manner, confidence, language, warmth and rapport can all influence the result of the treatment. That is not unique to hypnotherapy; it is a problem shared by many psychological and behavioural treatments. But it is especially important with hypnosis because suggestion, trust and responsiveness are vitally important.

A good hypnotherapy session is not simply someone reading a script while you lie down with your eyes closed. The relationship, the “chemistry”, and the client’s responsiveness matter too.

People vary in hypnotisability

That does not mean there are “hypnotisable people” and “non-hypnotisable people” in the cartoon sense often seen on stage. Hypnotic responsiveness exists along a spectrum and can be measured with standardised tests.

In some circumstances, hypnotisability is associated with treatment response. That creates a research problem. Imagine a treatment that works extremely well for one group of people and poorly for another. If you put everybody together and report one average result, you may conclude that the effect is only small.

The opposite problem can occur too. If you recruit only highly hypnotisable people, you may get an impressive result that does not apply to everyone who walks into a hypnotherapy practice.

Good hypnosis research needs to measure and report hypnotisability rather than treating everyone as identical. Even then, the scales are imperfect. They are often incomplete or self-reported, and people respond differently to different kinds of suggestion. Someone may have a profound inner experience even if they do not show an obvious outward response, such as an ideomotor movement.

We still don’t fully agree on how hypnosis works

You might think that after all these years scientists would have settled this but they haven’t!

There are competing theories of hypnosis and hypnotic suggestion. Some emphasise altered states of consciousness measurable in brain activity. Others focus on attention, expectation, imagination, social cognition, executive control, or the way people interpret and respond to suggestions. This is the classic “state” versus “non-state” debate. In my view, hypnosis involves both — but it is not helpful that there is still no scientifically definitive account.

It would be wrong to say that hypnosis has barely been studied. There is research across mental and physical health: pain, IBS, anxiety, stress, depression, sleep, cancer-related symptoms, medical procedures, smoking cessation, palliative care, performance, PTSD, immune system, weight management and many more.

But quantity and quality are not the same thing. You can have hundreds of studies and still have an evidence base that is hard to interpret if the studies are small, use different interventions, measure different outcomes and have methodological weaknesses. That is where hypnosis currently finds itself. We need better studies!

Perception is part of the problem

Hypnosis has been investigated scientifically for a very long time, but many people’s mental image of it still comes from entertainment rather than medicine.

Stage hypnosis — swinging watches, clucking like a chicken — has done clinical hypnosis no favours. While in both cases I think it is true to say hypnosis is powerful, there are now genuine misconceptions about what hypnosis is and is not.

If clinicians and researchers are sceptical about an intervention, it is harder for that intervention to attract the same research interest and funding as more established treatments. Limited high-quality research then feeds more scepticism, which leads to less investment, and so the cycle continues. Conducting research is expensive and you need buy in and funding.

So, does hypnosis work?

That is actually the wrong question, because it is too broad. Just like “Does medicine work?” is not a useful question. Neither is “Does psychotherapy work?” What we need to know is whether a particular intervention works for a particular problem, in particular people, compared with the alternatives, and whether the benefits outweigh the risks and costs.

Hypnosis deserves the same treatment.

There are areas where the evidence is now very promising. There are conditions where hypnosis appears to be a useful adjunct to conventional care. And acceptance in mainstream medicine is growing. I work with doctors, nurses, nutritionists, functional medicine practitioners and others who are increasingly open to hypnotherapy because they have seen the results in our shared patients.

If you are curious about hypnotherapy and would like to understand whether it might be appropriate for you, please get in touch.