Importance: According to the stepped-care model, there is a medium to large effect size for using cognitive behavioral therapy for insomnia that is delivered digitally, such as a smartphone application. However, it has been reported that studies using fully automated cognitive behavioral therapy for insomnia applications without expert support have high dropout rates.
Objective: To examine the effects of using a fully automated and individually tailored brief behavior therapy for insomnia (BBTI) applications for 2 weeks on insomnia-related symptoms, social disabilities, and work productivity among workers with insomnia in Japan.
Design, setting, and participants: This intent-to-treat prospective parallel-group randomized clinical trial included participants 20 years or older with Insomnia Severity Index (ISI) scores of 8 or higher. Participants were recruited via internet advertisements and workplace flyers and randomized to tailored BBTI, standard BBTI, self-monitoring with sleep diaries, or a waiting list control group. The study was conducted from September 21, 2017, to February 23, 2018. Data were analyzed from February 24, 2018, to February 22, 2019.
Interventions: A personalized BBTI and standard BBTI intervention, both of which included sleep scheduling, relaxation, sleep hygiene, and sleep diaries, administered via smartphone application.
Main outcomes and measures: Primary outcomes were insomnia severity, measured using the Japanese version of the Insomnia Severity Index, and social disabilities, measured using the Japanese version of the Sheehan Disability Scale. Secondary outcomes were dysfunctional beliefs, sleep reactivity, and work productivity. All measures were taken before and after the intervention and at 1-month and 3-month follow-ups.
Results: A total of 92 participants (mean [SD] age, 42.7 [11.5] years; 60 [65%] men) were randomized and included in analysis, with 24 participants assigned to tailored BBTI, 23 participants assigned to standard BBTI, 23 participants assigned to self-monitoring, and 22 participants assigned to the waiting list control group. At baseline, there were no significant differences among groups on any demographic characteristics or outcome measures. The results of the intent-to-treat analysis showed an interaction effect for all outcome measures. Compared with the waiting list control group, the BBTI interventions were more effective for reduction of insomnia severity (tailored BBTI: Hedges g = -1.64 [95% CI, -2.32 to -0.96]; P < .001; standard BBTI: g = -1.28 [95% CI, -1.93 to -0.63]; P < .001), social disabilities relating to social life (tailored BBTI: g = -1.33 [95% CI, -1.97 to -0.68]; P < .001; standard BBTI: g = -0.84 [95% CI, -1.46 to -0.22]; P = .009), and dysfunctional beliefs (tailored BBTI: g = -1.17 [95% CI, -1.80 to -0.54]; P < .001; standard BBTI: g = -0.84 [95% CI, -1.46 to -0.23]; P = .02) at the 3-month follow-up. Tailored BBTI quickly reduced insomnia severity (1-month follow-up: g = -0.85 [95% CI, -1.46 to -0.24]). Tailored BBTI was only more effective for improvement of work performance (g = -1.09 [95% CI, -1.71 to -0.46]; P = .005), social disabilities related to family life (g = -0.89 [95% CI, -1.51 to -0.28]; P = .005), and sleep reactivity (g = -1.09 [95% CI, -1.72 to -0.46]; P = .007) compared with the waiting list control group at the 3-month follow-up. The tailored BBTI led to improved worker productivity compared with standard BBTI (g = 0.94 [95% CI, 0.33 to 1.55]; P = .01) at the 3-month follow-up.
Conclusions and relevance: These findings suggest that an application for individually tailored BBTI is an inexpensive and effective treatment for insomnia. In future research, it would be informative to investigate the reasons for dropout during the follow-up period.
Trial registration: umin.ac.jp/ctr Identifier: UMIN000036572.