Irritable bowel syndrome (IBS) is a common gastrointestinal condition characterized by cramping, bloating, gas, and diarrhea or constipation. It likely has multiple causes, and many patients continue to suffer despite taking medication. Cognitive behavioral therapy (CBT) has sometimes been recommended as a treatment based on the notion that it can help patients reduce stress, change their eating patterns, and modify how they think about the illness. But now the largest study to date of CBT for IBS has shown that it has little impact in reducing symptom severity.
The UK study, conducted by investigators at King's College London and the University of Southampton, included 558 people with unresolved IBS. Participants were randomly assigned to receive telephone-delivered CBT, web-based CBT, or no intervention. All three groups continued to receive their usual medical treatment. The study did not test in-person CBT. The 12-month and 24-month results were published in 2019 in, respectively, the journals Gut and Lancet Gastroenterology & Hepatology.
At 12 months, those in both CBT groups reported a statistically significant reduction on the standard scale for assessing IBS symptoms. However, the average benefit for telephone-delivered CBT over treatment-as-usual was only slightly above the threshold considered clinically significant. At 24 months, the average benefit in this group was no longer clinically significant.
For the web-based group, the average benefit at 12 months was already below the critical threshold for clinical significance. At 24 months, the average benefit was neither clinically significant nor statistically significant.
Both groups reported modest benefits in other domains, such as social adjustment and depression—not surprising after a course of CBT. Given the weak results for reductions in symptom severity, however, these reported improvements likely have little or nothing to do with treating the illness itself.
Moreover, this was an unblinded trial relying solely on self-reported rather than objective outcomes. Since this kind of study design is likely to generate an unknown amount of bias, any positive findings should be taken with caution.
First published October 2019. Updated February 2020. A version of this article first appeared in the UC Berkeley Wellness Letter.
Also see A Better Treatment for Insomnia.