Preliminary Efficacy of a Digital Intervention for Adolescent Depression: A Randomized Clinical Trial (Preprint)

Abstract

Adolescent depression is a significant public health concern, yet access to effective mental health care is limited. Digital therapeutics (DTx) can improve access to evidence-based interventions, but their efficacy in adolescents is sparsely documented.

To examine the efficacy of a mobile app DTx vs. an active control as an adjunct treatment for adolescent depression symptoms.

A virtual, open-label, randomized clinical trial was conducted nationwide with a partial cross-over design and recruited 168 adolescents aged 13-21 years with symptoms of depression between November 2020 and September 2021. Participants were randomized (1:1) to the cognitive behavioral therapy (CBT)-based treatment app (Spark), or to a psychoeducational control app (Control) to use for 5 weeks. Primary outcome was a between-group (Spark vs. Control) difference in the change in depression symptoms from baseline to post-intervention as measured by the Patient Health Questionnaire (PHQ-8) using a linear mixed-effects analysis. The PHQ-8 ranges from 0-24, with scores of 5-9 indicating mild depression symptoms, scores of 10-14 indicating moderate symptoms, 15 to 19 indicating moderately severe symptoms, and 20-24 indicating severe symptoms. A minimally clinically important difference (MCID; 5 point reduction between baseline and post-intervention) in the Spark group and group differences in remission and treatment response rates based on the PHQ-8 at post-intervention were also investigated.

160 participants were randomized, 80 to the Spark group (x̄age=16.89 [2.5]) and 80 to the Control (x̄age=16.79 [2.59]). Data from 121 participants (Spark N=63; Control N=58) with moderate to severe (PHQ-8≥10) symptoms at baseline were included in primary analyses following a modified intention to treat principle. A linear mixed-effect analysis revealed a non-significant difference between groups in depression symptom change over the intervention period. The Spark group met an MCID threshold (x̄=-5.08 [95% CI -6.72 to -3.42]). Remission rate in the Spark group was significantly higher than the Control group (17% vs. 3%), Χ21=6.183, P=.01, FDR-adj. P=.03). Treatment response rates were not significant between groups. Post hoc analyses including participants with mild to severe (PHQ≥5) symptoms at baseline revealed promising evidence that Spark is effective in those with mild to severe symptoms.

There is initial evidence that a self-guided, CBT-based DTx intervention may effectively treat mild to severe depression symptoms in adolescents. DTx may improve access to mental health care for adolescents or serve as an important adjunct to standard of care.

ClinicalTrials.gov Identifier: NCT04524598