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

Isaac L. Ahuvia*, Michael C. Mullarkey, Jenna Y. Sung, Kathryn R. Fox, Jessica L. Schleider

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

Background: The question ‘what works for whom’ is essential to mental health research, as matching individuals to the treatment best suited to their needs has the potential to maximize the effectiveness of existing approaches. Digitally administered single-session interventions (SSIs) are effective means of reducing depressive symptoms in adolescence, with potential for rapid, large-scale implementation. However, little is known about which SSIs work best for different adolescents. Objective: We created and tested a treatment selection algorithm for use with two SSIs targeting depression in high-symptom adolescents from across the United States. Methods: Using data from a large-scale RCT comparing two evidence-based SSIs (N = 996; ClinicalTrials.gov: NCT04634903), we utilized a Personalized Advantage Index approach to create and evaluate a treatment-matching algorithm for these interventions. The two interventions were Project Personality (PP; N = 482), an intervention teaching that traits and symptoms are malleable (a ‘growth mindset’), and the Action Brings Change Project (ABC; N = 514), a behavioral activation intervention. Results: Results indicated no significant difference in 3-month depression outcomes between participants assigned to their matched intervention and those assigned to their nonmatched intervention. The relationship between predicted response to intervention (RTI) and observed RTI was weak for both interventions (r =.39 for PP, r =.24 for ABC). Moreover, the correlation between a participants' predicted RTI for PP and their predicted RTI for ABC was very high (r =.79). Conclusions: The utility of treatment selection approaches for SSIs targeting adolescent depression appears limited. Results suggest that both (a) predicting RTI for SSIs is relatively challenging, and (b) the factors that predict RTI for SSIs are similar regardless of the content of the intervention. Given their overall effectiveness and their low-intensity, low-cost nature, increasing youths' access to both digital SSIs may carry more public health utility than additional treatment-matching efforts.

Original languageEnglish (US)
Pages (from-to)1679-1688
Number of pages10
JournalJournal of Child Psychology and Psychiatry and Allied Disciplines
Volume64
Issue number12
DOIs
StatePublished - Dec 2023

Funding

This study was supported by the Office of the Director, National Institutes of Health under an ‘Emergency COVID‐19 Competitive Revision Award’ linked with grant no. DP5OD028123 (principal investigator: J.S.). The authors have declared that they have no competing or potential conflicts of interest. Key points

Keywords

  • Single-session interventions
  • adolescents
  • depression
  • machine learning
  • response to intervention
  • treatment matching
  • treatment selection

ASJC Scopus subject areas

  • Pediatrics, Perinatology, and Child Health
  • Developmental and Educational Psychology
  • Psychiatry and Mental health

Fingerprint

Dive into the research topics of 'Evaluating a treatment selection approach for online single-session interventions for adolescent depression'. Together they form a unique fingerprint.

Cite this