Effect of early intervention for anxiety on sleep outcomes in adolescents: a randomized trial

Abstract

The potential effect of early intervention for anxiety on sleep outcomes was examined in a sample of adolescents with anxiety (N = 313, mean 14.0 years, SD = 0.84, 84% girls, 95.7% Norwegians). Participants were randomized to one of three conditions: a brief or a standard-length cognitive-behavioral group-intervention (GCBT), or a waitlist control-group (WL). Interventions were delivered at schools, during school hours. Adolescents with elevated anxiety were recruited by school health services. Questionnaires on self-reported anxiety symptoms, depressive symptoms, and sleep characteristics were administered at pre- and post-intervention, post-waitlist, and at 1-year follow-up. Adolescents reported reduced insomnia (odds ratio (OR) = 0.42, p < 0.001) and shorter sleep onset latency (d = 0.27, p < 0.001) from pre- to post-intervention. For insomnia, this effect was maintained at 1-year follow-up (OR = 0.54, p = 0.020). However, no effect of GCBT on sleep outcomes was found when comparing GCBT and WL. Also, no difference was found in sleep outcomes between brief and standard-length interventions. Adolescents defined as responders (i.e., having improved much or very much on anxiety after GCBT), did not differ from non-responders regarding sleep outcomes. Thus, anxiety-focused CBT, delivered in groups, showed no effect on sleep outcomes. Strategies specifically targeting sleep problems in adolescents should be included in GCBT when delivered as early intervention for adolescents with elevated anxiety.