Factors Affecting Users' Behavioral Intention, Behavior and Therapy Outcome of a Mobile-Phone-Delivered Cognitive Behavioral Therapy for Insomnia (Preprint)

Abstract

A mobile app could be a powerful medium for providing individual support for cognitive behavioral therapy (CBT), as well as to facilitate the therapy adherence. Many studies have reported about the efficacy of such apps for insomnia treatment. However, little is known about factors that may explain the acceptance and uptake of such applications.

This study, therefore examines factors that may explain variation between people's behavioral intention to use a CBT for insomnia (CBT-I) app and their use behavior, and the influence of the behavior on therapy outcomes.

From literature, related factors were identified. Data were gathered from a field trial involving people with relatively mild insomnia using a CBT-I app. Applying the Partial Least Squares-Structural Equation Modeling method, the study examines a three-tier model. The first tier explored seven aspects of behavioral intention: performance expectancy, effort expectancy, social influence, self-efficacy, trust, affect, and anxiety. The second tier investigated the influence of behavioral intention and facilitating condition on user behavior, which was specifically defined as engaging with the app (i.e. sleep diaries and conversations with the app), doing relaxation exercises, and following sleep restriction exercises. Here, the relationship between app engagement and the other two exercise behaviors was also examined. Finally, the third tier tested the influence of the behavior on therapy outcomes: insight into own sleep pattern, general sleep knowledge, and insomnia severity. The latter was measured using Insomnia Severity Index.

Performance expectancy, effort expectancy, social influence, self-efficacy, and trust all explained part of the variation in behavioral intention, but not beyond the explanation provided by affect, which accounted for R2 = .59 (n = 89). Behavioral intention could explain two behavior factors, i.e. app engagement (R2 = .30, n = 89) and relaxation exercise (R2 = .42, n = 89). Engagement with the app was a determinant for the other behaviors, i.e. relaxation exercise (R2 = .42, n = 89) and sleep restriction (R2 = .54, n = 47). Furthermore, app engagement was the only determinant for one of the therapy outcomes, i.e. understanding own sleep pattern (R2 = .09, n = 72). We did not find an association between user behavior and insomnia severity.

We anticipate that the findings will help researchers and developers to focus on: (1) users' positive feelings about the app, as this was an indicator of their acceptance of the mobile app; and on (2) therapeutic activities delivered via the app, as this correlated with their sleep pattern awareness and their involvement in the exercises recommended by the CBT-I app.

Netherlands Trial Register: NTR5560; http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=5560 (archived by WebCite at http://www.webcitation.org/6noLaUdJ4)