Icon Legend

This session is not in your schedule.

This session is in your schedule. Click again to remove it.

Presentation Icons

Recorded Session

Ticketed Session

24 Views

Open Paper

Child and adolescent mental health

Supporting Youth at Risk: CBT, DBT, and Technology-Based Interventions for Mental Health, Sleep, and Problematic Internet Use

4 - (OP19) Linguistic Features Predicting Treatment Response to a CBT Smartphone App for Adolescent Depression

Friday, June 26, 2026

2:56 PM - 3:13 PM PDT

Location: Yerba Buena Salon 12, B3 Level

Keywords: Depression, CBT, Longitudinal
Recommended Readings: https://www.nature.com/articles/s44184-025-00140-y, , , ,

Co-Author(s)

BO

Bridianne O'Dea, Ph.D.

Professor
Flinders University
Adelaide, South Australia, Australia

Author(s)

TB

Taylor Braund, Ph.D.

Senior Research Associate
Black Dog Institute
Sydney, New South Wales, Australia

Co-Author(s)

BY

Bethany Yu, Other

Graduate
Black Dog Institute
Sydney, New South Wales, Australia
SL

Sophie Li, Ph.D.

Clinical Psychologist
Black Dog Institute, UNSW Sydney
Sydney, New South Wales, Australia
MS

Mirjana Subotic-Kerry, Ph.D.

Senior Research Program Manager
Black Dog Institute, UNSW Sydney
Sydney, New South Wales, Australia
MA

Melinda Achilles, Ph.D.

Research Program Manager
Black Dog Institute
Sydney, New South Wales, Australia
JN

Jill M. Newby, Ph.D. (she/her/hers)

Professor
UNSW Sydney at the Black Dog Institute
Randwick, New South Wales, Australia
HC

Helen Christensen, Ph.D.

Professor
University of New South Wales
Sydney, New South Wales, Australia
AM

Andrew Mackinnon, Ph.D.

Professor
Black Dog Institute
Sydney, New South Wales, Australia
Aliza Werner-Seidler, Ph.D.

Professor
Black Dog Institute
Randwick, New South Wales, Australia
PB

Philip Batterham, Ph.D.

Professor
Australian National University
Canberra, Australian Capital Territory, Australia

Background: Self-directed digital Cognitive Behavioural Therapy (CBT) programs can be efficacious for reducing depressive symptoms in adolescents. However, many users disengage from these interventions. Early linguistic features expressed in individuals’ responses to self-directed activities may indicate their response to treatment. This study explored the linguistic features associated with depressive symptom change, treatment response, and depression remission in symptomatic adolescents using a self-directed CBT-based smartphone application (ClearlyMe®).

Methods: This study utilised data from adolescent participants allocated to the intervention conditions (n=128) of a three-arm Randomised Controlled Trial (RCT). Depressive symptoms were measured at baseline, 6-weeks (endpoint), and 4-months (follow-up) using the self-report Patient Health Questionnaire-A (PHQ-A). Linguistic features were extracted from a self-directed therapy activity within the app using the Linguistic Inquiry and Word Count-22 (LIWC-22). Regression models examined whether linguistic features predicted changes in symptoms, treatment response, and depression remission measured at 6-weeks and 4-months. Models were controlled for baseline depression, age, sex, and education.

Results: Greater use negate words (e.g., not, never) and reduced use of health words (e.g., patients, physician) predicted decreases in depression severity at 4-month follow-up (p = .030 and p = .038, respectively). Similarly, greater use of negate words at baseline predicted treatment response for depression at follow-up (OR = 1.06, p = .045). No other significant predictors were found.

Conclusion: Linguistic features were modestly associated with symptom change and treatment response in adolescents using a self-directed digital CBT app. While findings suggest potential for language-based markers in tailoring interventions, replication in larger samples is needed to establish reliability and clinical utility.

Learning Objectives:

Describe how early linguistic features predict depressive symptom change, treatment response, and remission in adolescents using self-directed digital CBT interventions.