PofoliaShared via Pofolia

Clinical Psychological Science· 2026Q1

Leveraging Machine Learning to Personalize Depression Treatment: A Preregistered Study of 828 Adults Randomly Assigned to a Digital Single-Session Intervention or Waitlist

EJ Salamander Jardas, Jacqueline Howard, Lorenzo Lorenzo‐Luaces

Short summary

A machine-learning algorithm (Personalized Advantage Index) failed to identify adults who would benefit more from a digital single-session depression intervention (COMET) compared to a waitlist control, with no statistically significant interaction found in a study of 828 adults.

AI-generated from the title and abstract; the full text is not read.

Key points

  • A machine-learning algorithm (PAI) was developed to match adults to a digital single-session intervention (COMET) or waitlist control.
  • The study included 828 adults randomly assigned to either COMET or the waitlist.
  • The PAI did not show a statistically significant interaction with treatment assignment in predicting 2-week posttreatment depressive symptoms.
  • The findings indicate difficulty in developing personalized treatment recommendations for digital single-session interventions.

AI-generated from the title and abstract; the full text is not read.

Abstract

Some digital single-session interventions (SSI) for depression appear effective, at least in youths, but not everyone benefits. In the present study, we use machine-learning methods to develop a treatment-matching algorithm for a digital SSI, the Common Elements Toolbox (COMET), versus a waitlist control. Eight hundred twenty-eight adults with a current or past mental-health problem were randomly assigned to COMET or a waitlist control. Elastic-net-regularization models with 10-fold cross-validation were used to develop a Personalized Advantage Index (PAI) indicating the relative benefit of receiving COMET over the waitlist in 2-week posttreatment depressive symptoms. In the 20% held-out test data, PAI did not interact with treatment to predict depression severity after treatment (β = 0.880, SE = 1.21, t = −0.72, p = .47), indicating that our treatment-matching algorithm was not able to provide statistically significant recommendations. Even in a large sample, personalized treatment recommendations for digital SSIs are difficult to develop.

The authors' abstract, as published at the source. Clinical Psychological Science, 2026 · DOI ↗

TakeawaysPremium
Ask the paperFree account

Continue with a free account

Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.

Continue free on the web

Sign in with Google or Apple; no card needed. You come back to this paper.

On your phone:

Field: Applied Psychology

Applied PsychologyPsychology