Abstract
Introduction
Suicide is a leading cause of death among adolescents in the U.S., with especially high rates among adolescents recently discharged from acute psychiatric care. Insomnia has been linked to increased suicidal ideation (SI) and serves as a promising treatment target to reduce SI in adolescents. Cognitive behavioral therapy for insomnia (CBT-I) has proven efficacious at reducing insomnia, but few people have access to this intervention. Digital versions have the potential to be scaled to reach vulnerable populations, such as those during high risk periods. This open pilot tested the feasibility and accessibility of a smartphone-based version of dCBT-I (SleepioTM) among high-risk adolescents for reducing insomnia and SI.
Methods
This open pilot included twelve adolescents, ages 14-18 (M=15.8, SD=1.2), recently (past 90 days) hospitalized (inpatient or emergency department) for suicide risk and had co-occurring insomnia. In this project, all adolescents were offered Sleepio, which delivers core CBT-I components (i.e., sleep hygiene, sleep restriction, stimulus control, cognitive therapy, relaxation) through 6 sessions in a smartphone app. This study assessed the feasibility (sessions attended) and acceptability of the intervention from both adolescents and caregivers (post-treatment assessments).
Results
All 12 adolescents completed at least one Sleepio session, and three adolescents (25%) completed all six sessions (i.e., the full treatment package). Partial intervention completion rates were as follows: one session (n=3), two sessions (n=1), three sessions (n=3), and four sessions (n=2).
Adolescents’ post-treatment survey ratings (n=10; Scale 0=not at all to 4=very) indicated that they somewhat liked the program (M=2.10, SD=1.10), it was mostly easy to use (M=2.80, SD=1.23), convenient (M=2.90, SD=1.10), and easy to understand (M=3.20, SD=0.79). In addition, adolescents rated that it was moderately useful for improving their sleep (M=2.30, SD=1.42). The lowest ratings were for the program’s ability to keep their interest and attention (M=1.60, SD=1.07). Finally, in a post-treatment interview, all teens identified at least one dCBT-I skill they used to improve their sleep (e.g., relaxation).
Caregivers’ post-treatment survey ratings indicated that their adolescents, overall, liked the intervention (Scale: 1=low to 5=high; M=3.45, SD=0.69). Most (64%) observed a notable improvement in their adolescent’s sleep patterns, and 82% reported an improvement in their adolescent’s mood.
Conclusion
This pilot study shows preliminary feasibility and acceptability of dCBT-I among high-risk adolescents, as most completed at least one session and were able to identify new skills learned. However, adolescents on average reported that the intervention did not hold their attention, and only 25% completed the full treatment package. Limitations of this study include small sample size and lack of a control group. A small randomized control trial is currently in progress to address these limitations.