Predicting outcome in computerized cognitive behavioral therapy for depression in primary care: A randomized trial

Area of Science:

• Psychiatry
• Digital Health
• Clinical Psychology

Background:

• Depression treatment effectiveness varies.
• Online computerized cognitive behavioral therapy (CCBT) offers a scalable intervention.
• Identifying predictors of treatment success is crucial for personalized care.

Purpose of the Study:

• To identify pretreatment and short-term improvement variables that predict 12-month outcomes.
• To explore these variables as moderators for three depression treatment arms: unsupported CCBT, usual care (TAU), and CCBT combined with TAU (CCBT&TAU).

Main Methods:

• 303 depressed patients were randomized into three groups: unsupported CCBT, TAU, or CCBT&TAU.
• Demographic, clinical, cognitive, and short-term improvement variables were assessed as potential predictors and moderators.
• Outcomes included the Beck Depression Inventory-II score and reliable change at 12-month follow-up.

Main Results:

• Patients with extreme positive responding fared better with CCBT versus TAU.
• Parental psychiatric history or major depressive disorder diagnosis predicted better outcomes with CCBT&TAU versus TAU.
• Independent of treatment type, current employment, lower pretreatment illness severity, and short-term clinical improvement predicted better outcomes.

Conclusions:

• Optimistic patients with approach-oriented coping may benefit most from CCBT.
• CCBT&TAU appears most suitable for patients with greater vulnerability.
• Patients with less severe depression showed the most improvement across all treatment types.