Factors Contributing to Symptom Change in Standardized and Individualized Internet-Based Interventions for Depression: A Randomized-Controlled Trial

摘要

Research suggests 4 categories of outcome predictors in face-to-face therapy (i.e., treatment expectations, extratherapeutic factors, relationship factors, and factors specific to a treatment approach/technique). However, it is unclear whether these factors are relevant in standardized and individualized Internet-based interventions. To investigate this question, a secondary analysis of data from 1,089 mildly to moderately depressed adults undergoing 6 weeks of cognitive-behavioral Internet-based intervention for depression randomized to receive either weekly written feedback individualized by a counselor or automated and fully standardized feedback was performed. The following variables corresponding to the 4 categories were tested regarding associations with depressive symptom change during multiple treatment periods within a multi group structural equation model: (a) outcome expectations, (b) extratherapeutic stressors and stress change during treatment, (c) midtreatment working alliance (task/goal and bond), and (d) uptake of treatment-specific components (login and specific tool use). Results suggest similar regressive associations across treatment conditions: Previous symptom change was the most important predictor for subsequent symptom developments. Stress at baseline and the uptake of specific treatment components only played a minor role, and stronger task/goal ratings were associated with later symptom improvements. Early symptom improvements predicted stronger midtreatment task/goal and bond ratings, whereas only stronger task/goal ratings were associated with later symptom improvements. Outcome expectations were only indirectly related with symptom change mediated through goal/task ratings.