Gut-directed hypnotherapy for IBS: the most replicated result in the field

The largest trial, and what a control group changes

The IMAGINE study is the biggest randomised trial in this area. It allocated 354 patients across three arms: 150 to individual hypnotherapy, 150 to group hypnotherapy, and 54 to educational supportive therapy as a control. In the intention-to-treat analysis at three months, adequate response was 40.8 percent for individual hypnotherapy, 33.2 percent for group hypnotherapy, and 16.7 percent for the control group. At twelve months the figures were 40.8 percent, 49.5 percent and 22.6 percent. Hypnotherapy beat control at both time points. [2]

Two details deserve attention. The control arm was small, only 54 patients, which widens the uncertainty around the comparison. And group hypnotherapy performed about as well as individual hypnotherapy, which suggests the personal relationship with a therapist may matter less than the content of the sessions themselves. That has obvious implications for whether a recording can do useful work.

Broader syntheses reach compatible conclusions. A 2019 update pooling 53 randomised trials found the relative risk of symptoms not improving with psychological therapies was 0.69, with hypnotherapy among the approaches beneficial when two or more trials were pooled. [4] A network meta-analysis of 41 trials and 4,072 participants placed gut-directed hypnotherapy at a relative risk of 0.67, alongside self-administered cognitive behavioural therapy at 0.61 and face-to-face CBT at 0.62. [3]

The problem every review names

You cannot blind someone to whether they received hypnotherapy. One meta-analysis puts it directly: performance bias was high in all trials because it was impossible to blind participants and therapists in this type of intervention. [6] That is not a flaw in any particular study. It is a structural limit of the research question.

The consequences show up in the syntheses. The 2019 review noted significant heterogeneity between studies, significant funnel plot asymmetry suggesting publication bias, and issues with trial design including lack of blinding, concluding that treatment effects may be overestimated as a result. [4] The network meta-analysis was equally direct: risk of bias of trials was high, with evidence of funnel plot asymmetry, so the efficacy of psychological therapies is likely to have been overestimated. That same review found that several psychological therapies work for IBS and none was superior to another. [3]

The most recent and largest synthesis, covering 67 trials and 7,441 participants, is the most cautious of all. It found gut-directed hypnotherapy efficacious at a relative risk of 0.79 across twelve trials, ranking it below minimal-contact CBT and telephone self-management. It reported that no trial was at low risk of bias across all domains, and rated certainty in the evidence for every direct and indirect comparison in the network as low or very low. [5]

One further honesty note. A 2021 meta-analysis with subgroup analysis found that the headline effect of hypnotherapy on global gastrointestinal symptoms was not statistically significant, at a standardised mean difference of 0.24. Benefit appeared in the subgroups receiving at least weekly sessions or at least eight sessions with six or more hours of contact. [7] Dose, again, appears to matter more than the existence of the technique.

Does it survive being delivered through an app?

This is the question that matters most for a product like ours, and the answer is genuinely mixed. A 2025 randomised trial of a digitally delivered gut-directed hypnotherapy program against an active digital control randomised 240 participants and found 81 percent met the primary endpoint with the hypnotherapy program against 63 percent with the control, and 71 percent reported a 30 percent reduction in pain against 35 percent. [8]

A different digital trial did not replicate that. It randomised 378 patients to digital gut-directed hypnotherapy or muscle relaxation and found a similar proportion met the primary endpoint in each group, 30.4 percent against 27.1 percent, with no significant difference. Some secondary endpoints did separate later in treatment. This trial missed its primary endpoint, and anyone citing it as a success is misreading it. [9]

A retrospective evaluation of app-delivered gut-directed hypnotherapy is sobering on a different axis. Of 2,843 people who started the free sessions, 1,428 purchased the app and 253, or 9 percent, completed all 42 sessions. Among the small group who provided outcome data, 64 percent responded. The authors state plainly that adherence was low. [10] Finishing the course turns out to be the hard part.

For historical completeness, the 2007 Cochrane review on this topic exists and reached no conclusion. It found four studies with 147 patients, declined to pool them, and concluded that trial quality was inadequate to allow any conclusion about efficacy. [11] It has been superseded by the larger work above, but it is a useful reminder of how recently this evidence base was too thin to say anything at all.