Outpatient and self-referred participants: Adherence to treatment components and outcome in an internet intervention targeting anxiety disorders.

Objective: While adherence is an important factor influencing the effectiveness of internet interventions, many studies operationalize adherence only by the number of sessions and do not report adherence to specific treatment components. The goal of this study was to investigate adherence to treatment components as well as outcome in outpatients and self-referred participants who participated in an internet intervention targeting anxiety.

Method: Outpatients ( = 50) were compared to self-referred ( = 37) participants and a matched outpatient waitlist sample (based on nearest neighbor matching): Using -test and χ2 tests adherence to treatment components based on the number of completed exercises was compared between participant groups. A 2 × 2 repeated measures ANOVA was used to compare pre-to post symptom change between participant groups. Primary measures included the Generalized Anxiety Disorder Scale-7 (GAD-7) and the Mini Social Phobia Inventory (Mini-SPIN). Using nonparametric bootstrap analyses number of sessions and adherence to treatment components were investigated as potential mediators of the relationship between participant group and outcome. Finally, predictors of adherence to treatment components in outpatient participants were investigated using LASSO and logistic regression.

Results: Self-referred participants were more adherent than outpatient participants, however the groups did not differ significantly in outcome. Outpatient participants who adhered to relaxation showed greater improvement during the waiting period than the matched outpatient waitlist sample. The effect of participant group on outcome was mediated via adherence to exposure and number of sessions.

Conclusions: In internet interventions adherence to treatment components differs between participant groups and has a mediating effect on treatment outcome. Therefore, it should be fostered, especially when participants are not self-referred. In line with these findings more studies should investigate relevant participant characteristics in more depth.