Symptom-specific effectiveness of an internet-based intervention in the treatment of mild to moderate depressive symptomatology: The potential of network estimation techniques

Abstract

The internet-based intervention Deprexis® has proven to be effective in improving overall depression severity. The current pragmatic randomized controlled trial included 1013 participants with mild to moderate symptomatology and aimed to identify the symptom-specific effects of the internet-based intervention Deprexis (intervention group) in comparison to care as usual (control group). All participants -in both conditions- were permitted to use any type of treatment. Of the nine considered symptoms (assessed with the Patient Health Questionnaire), seven showed larger improvements in the intervention condition relative to care as usual (effect sizes ranging from 0.15 to 0.31). No significant differences were found for the two other symptoms. In a next step, a network was estimated including treatment condition as well as changes in all nine symptoms. The resulting network suggests that four of the seven identified symptom-specific effects were direct, whereas the three other symptom-specific effects were indirect and could be explained by effects on other symptoms. Lastly, exploratory analyses showed that the intervention was more effective in improving overall depression severity for participants with higher scores on those four symptoms that were directly affected by the intervention; consequently, the network estimation techniques showed potential in precision psychiatry.

Keywords: CBT; Depressive symptoms; Internet-based intervention; Network estimation techniques; Precision psychiatry.