Internet-Based Cognitive-Behavioral Program Improves Adolescent Anxiety Symptoms

Improved anxiety symptoms over time in adolescents and reduced use of health care resources may be the result of online self-led cognitive behavioral therapy (CBT), according to study findings published in Journal of Anxiety Disorders.

Researchers conducted a single-blind, randomized controlled trial (ClinicalTrials.gov Identifier: NCT02970734) that included 563 self-referred adolescents (age range, 13-19 years) in Canada who reported anxiety from December 2016 through August 2018. Trial promotion was through social media and brochures placed at primary care and school-based mental health services. A website with anxiety resources was used as the comparator.

The primary endpoint was anxiety change from baseline to postintervention, which was measured using the total Multidimensional Anxiety Scale for Children (MASC2). Secondary endpoints included between-group difference in anxiety change postintervention to 3-month follow-up using the MASC2. Between-group differences were also evaluated in change in quality of life (QOL) using the Youth QOL Instrument-Short Form.

Adolescents with self-reported positive risk screens for suicide, schizophrenia, alcohol use disorders, and residence outside of Canada were excluded from the study. Among the 4587 volunteer participants, 2299 were excluded by criteria, 1314 failed to complete screening questions, 255 failed to consent to participate, and 153 consented but failed to continue with participation, respectively.

Researchers evaluated anxiety symptoms, QOL, and healthcare utilization pre- and postintervention, and again at 3 months postintervention. Participants were randomly assigned to the intervention (6 weeks online CBT; n=280) or the website (n=283), and researchers were blinded to intervention allocation. Participants were girls (71.5%), boys (4.5%), nonbinary (1.3%), agender (0.4%), androgynous (0.2%), genderqueer (0.2%), or gender-fluid (0.4%). Nearly a quarter of participants did not respond (21.6%).

There were no significant group differences in symptom change postintervention (P =.16). Symptom change was significant favoring online CBT at 3 months (P =.04). Age was not a significant factor in this model. Boy participants reported greater symptom change (P =.03). As anxiety symptoms decreased across time, QOL increased (P <.001).

Among participants with available data for before and after intervention health care utilization (n=235), online CBT users experienced the greatest changes in use of clinician visits vs website users (psychiatrist -41.0% vs 18.5%; social worker -42.5% vs -22.1%), self-help or alternative treatments (-60.0% vs 6.6%), and hospital-based care (emergency department visits -80.0% vs 79.4%; hospital admissions -76.1% vs 42.9%).

Limitations of the study include possible selection bias, nonresponse bias, use of self-reported anxiety symptoms as the primary outcome measure, inability to assess degree to which participants practiced learned strategies, and high rate of attrition at postintervention.

Study authors concluded, “[O]ver time, use of online CBT by adolescents can result in improved anxiety symptoms and fewer use of other healthcare resources compared to traditional online information seeking.” They added, “Self-led, online treatment programs can be a critical resource for anxious adolescents looking for credible online resources. […] Augmentation of the online CBT program by introducing more coaching opportunities and content to support self-care and positive functioning may improve immediate treatment responses.”