Refining the Analysis of Mechanism–Outcome Relationships for Anxiety Treatment: A Preliminary Investigation Using Mixed Models

Abstract

Although efficacious treatments exist for anxiety disorders, issues remain regarding how best to conceptualize and measure purported change processes in clinical research. In the current study, we examined the relationship between treatment-specific (exposure therapy, attention bias modification [ABM]) as well as more general change processes with symptoms within a transdiagnostic sample using mixed models. Results indicated that slope of self-efficacy across treatment and between-session habituation across identical exposures was associated with slope of symptom change. Although slope of anxiety ratings within session was not associated with slope of symptom change, it did interact with other candidate exposure processes to predict symptoms. Purported ABM change processes were not associated with outcome. Our use of mixed models exemplifies an emerging trend in this research aimed at minimizing loss of data through aggregation, and our results highlight the utility of integrating treatment-specific as well as more general change processes in mechanistic research.