Does clinical hypnosis work? The clinical evidence base
The short answer
The short answer is: Yes, hypnotherapy is helpful for some conditions. There is good research evidence showing that hypnosis is an effective treatment for conditions such as pain, or irritable bowel syndrome. For many other conditions there is much less research evidence, or lower-quality studies. The strongest evidence for the use of hypnosis tends to be for conditions where it is easiest to research a 'pure hypnosis' treatment. For conditions where a mixed 'hypnosis + other therapy' approach might be more appropriate it has been harder to conduct research.
Some of the research has looked at adding hypnosis to other therapies, such as cognitive-behavioural therapy. For some conditions, for example, CBT has been compared to CBT + hypnosis. It is important to remember that hypnotherapy isn't a treatment in its own right, so you really need a therapist who is trained in other evidence-based therapies as well as hypnosis.
The long answer
Evidence for the effectiveness of psychological therapies such as hypnosis comes in a variety of forms, some more reliable than others. The use of hypnosis for a variety of conditions is reviewed on these pages, and the quality of the evidence is assessed. Very simply put, the more evidence we have supporting a particular therapy, and the less conflicting evidence we have, then the surer we can be that such a therapy is effective. Additionally, the higher the quality, or rigour, of the research studies (e.g. randomised controlled trials vs. single case studies) the stronger the evidence.
We can rank the quality of different types of evidence, in this case clinical research involving hypnosis, according to how reliable we think it is. One such ranking in psychology might include:
Description | Features | Reliability |
---|---|---|
Systematic review |
Finds all of the relevant studies according to clear inclusion criteria. Tries to appraise, select and synthesise all of the high quality research relevant to that question. |
High |
Meta-analysis |
Finds all of the relevant studies according to clear inclusion criteria. Groups the key results (in the form of 'effect sizes') from lots of studies |
High |
Randomised-controlled trial |
Randomises groups of participants to receive either treatment A or treatment B |
Medium |
Controlled trial |
Participants receive treatment A or treatment B but are not randomised to each group. |
Medium |
Uncontrolled trial |
No participants receive a placebo or alternate therapy |
Lower |
Single case study |
More or less detailed report on how a patient or group of patients responded to a particular therapy. Can be improved with good reporting of baselines |
Lower |
Meta-analyses of hypnosis
Rather than looking at individual studies to try to gauge the effectiveness of a treatment, a meta-analysis combines the results of many such studies to obtain a mathematically more reliable estimate of the effectiveness of a treatment, given as an 'effect size'. Effect sizes refer to group differences in standard deviation units on the normal distribution. Cohen (1988) defined the magnitude of effect sizes, an effect size of 0.2 is considered small, one of 0.5 is medium, and 0.8 is large. Inclusion criteria are set in advance (e.g. I might decide to look at studies investigating the effect of hypnosis on pain which report certain useful statistics, or only decide to look at studies which meet certain standards of quality) and all studies which meet the criteria are included. Studies conducted on larger numbers of people can be judged to be more 'reliable' and can be given more weight in the final analysis. A meta-analysis takes advantage of the large sample size endowed by multiple studies all looking at a similar topic and is considered to be a useful way of evaluating the size of effects. Meta analysis is not without problems though, including the difficulty of aggregating across different study types, the difficulty of appropriately classifying studies, and the problem of publication bias. Some of these criticisms are considered by Roth & Fonagy (2005).
A number of meta-analyses have been conducted regarding the effectiveness of hypnosis:
- Kirsch, I., Montgomery, G., & Sapirstein, G. (1995). Hypnosis as an adjunct to cognitive-behavioral psychotherapy: A meta-analysis. Journal of Consulting & Clinical Psychology, 63, 214-220.
- In this meta-analysis studies were considered for inclusion if hypnosis was used as an adjunct to cognitive behavioural therapy. This approach is consistent with the view that hypnosis is not a therapy in its own right, but instead can make valid forms of therapy more effective. Accordingly, studies conducted on a wide variety of patients with a range of disorders were included in the meta-analysis. Problems treated included phobia, obesity, and pain, and it is important to note that the result is therefore not specific to any one disorder. The results showed that the addition of hypnosis substantially enhanced treatment outcome, with the average client who received cognitive-behavioural hypnotherapy showing greater improvement than at least 70% of clients receiving non-hypnotic treatment. The results were particularly pronounced for studies on obesity, and effect sizes for other treatments was generally slightly lower.
- Lynn, S. J., Kirsch, I., Barabasz, A., CardeƱa, E., & Patterson, D. (2000). Hypnosis as an empirically supported clinical intervention: The state of the evidence and a look to the future. International Journal of Clinical and Experimental Hypnosis, 48, 235-255.
- Flammer, E., Bongartz, W. (2003). On the efficacy of hypnosis: A meta-analytic study. Contemporary Hypnosis, 20, 179-197.
- This meta-analysis included the results of 57 randomised clinical trials which compared pateints treated exclusively by hypnosis to either an untreated control group or to a group of patients treated by conventional medical procedures. A medium effect size of d=0.56 was observed indicating superiority of treatment in hypnosis groups, but this must be understood in the context of comparisons of hypnosis with no-treatment or standard-treatment. Of six studies that reported correlations between hypnotic susceptibility and outcome the average was r=0.44, indicating that high hypnotisables achieved greater benefit than low hypnotisables.
What is hypnosis?
Definitions of hypnosis
Types of suggestion
FAQ
Scientific theories of hypnosis
History of hypnosis
Animal hypnosis
Key people in hypnosis
Demand characteristics
Scientific research
States of consciousness
Neuroscience
Modification of suggestibility
Attention and hypnosis
Pain research
Hypnosis as a research tool
Genes and hypnotizability
What is hypnotherapy?
Is it effective?
Finding a therapist
Depression
Irritable bowel syndrome
Pain
PTSD
Smoking
Surgery
Weight loss
Hypnosis research papers
Suggestibility scales
Scripts
Videos
Forum
Organisations
Journals
Book reviews
© 2007-2019 Dr Matthew Whalley