Evaluating a treatment selection approach for online single-session interventions for adolescent depression

Area of Science:

• Mental Health
• Adolescent Psychology
• Digital Interventions

Background:

• Personalized treatment selection is crucial for maximizing mental health intervention effectiveness.
• Digitally administered single-session interventions (SSIs) show promise for reducing adolescent depression.
• Identifying which SSIs are most effective for specific adolescents remains an open question.

Purpose of the Study:

• To develop and evaluate a treatment selection algorithm for two digital single-session interventions (SSIs) targeting depression in high-symptom adolescents.
• To determine if a personalized matching approach improves outcomes compared to non-matched assignment.

Main Methods:

• A treatment-matching algorithm was created using the Personalized Advantage Index approach.
• Data from a large-scale randomized controlled trial (N=996) comparing two SSIs (Project Personality and Action Brings Change) were utilized.
• The algorithm's effectiveness was evaluated by comparing outcomes for matched versus non-matched participants.

Main Results:

• No significant difference in 3-month depression outcomes was found between matched and non-matched participants.
• The correlation between predicted and observed response to intervention (RTI) was weak for both interventions.
• Predictors for response to each intervention were highly similar across both SSIs.

Conclusions:

• Current treatment selection algorithms have limited utility for predicting response to digital SSIs in adolescents.
• Predicting individual response to SSIs is challenging, and predictors are similar across different intervention types.
• Increasing access to effective digital SSIs may offer greater public health benefits than further refinement of treatment-matching strategies.