Objective: The aim of this study was to identify mediators of change in a previously published randomized controlled trial that compared Internet-delivered cognitive behavioral treatment based on exposure exercises (ICBT) with Internet-delivered stress management (ISM) for irritable bowel syndrome (IBS). ICBT and ISM targeted distinct proposed mechanisms of illness maintenance and symptom exacerbation, gastrointestinal symptom-specific anxiety (GSA), and stress reactivity, respectively. The original study found that ICBT was more effective than ISM in improving IBS symptoms.
Method: Weekly measurements of GSA and stress reactivity (putative mediators) and treatment outcome were obtained from 195 participants with IBS, who had been randomized to ICBT or ISM.
Results: Parallel process growth mediational analyses revealed that the larger reduction of IBS symptoms from ICBT compared to ISM was mediated by changes in GSA, αβ = -0.42, 95% CI asymmetric [-0.71, -0.16]. In contrast, changes in stress reactivity did not mediate the difference in outcomes between treatments, αβ = 0.04, 95% CI asymmetric [-0.09, 0.20]. Analyses of the temporal sequence of week-to-week changes in process and outcome measures showed that only GSA displayed a pattern consistent with a causal model in which change in process preceded and contributed to symptom change. Furthermore, engagement in treatment specific activities was related to change in GSA but not to stress reactivity in the ICBT arm, whereas treatment specific activities were not related to change in any of the putative processes in the ISM arm.
Conclusions: We conclude that ICBT works through directly targeting GSA, rather than by means of reducing stress reactivity.