Dysfunctional cognition and self-efficacy as mediators of symptom change in exposure therapy for agoraphobia – Systematic review and meta-analysis

Both dysfunctional and self-efficacy-related cognitions are theorized as etiological and maintaining factors in agoraphobia. Exposure therapy is an effective treatment and central component of CBT for agoraphobia, but the role of changes in these cognitions as a mechanism of action has not been established. The present review aims to evaluate (a) whether exposure without cognitive interventions elicits changes in cognitive variables and (b) whether cognitive changes mediate outcomes in exposure-based treatments. We searched PsycInfo and PubMed for studies on agoraphobia (with or without panic disorder) and exposure as a treatment component. Fifteen articles with 29 relevant study arms (N = 921) were identified for a meta-analysis of cognitive changes after exposure. Seventeen articles (N = 1881) were included in a systematic narrative review of cognitive mediation. A random effects model revealed a large effect of cognitive improvement after pure exposure treatments, d = 1.02 (95% CI 0.81–1.23). The systematic review mostly supported changes in cognition as mediators of symptom change. Improved study designs and statistical methods in future mediation studies are needed to strengthen causal interpretation. Cognitive change is a probable mechanism of action in exposure therapy, especially change in self-efficacy. The present review suggests novel ways in which cognitive interventions can augment exposure therapy. •A meta-analysis examined cognitive changes after pure exposure therapy.•Large effect sizes: d = 0.92 for dysfunctional cognitions (DC) and 1.21 for self-efficacy (SE).•A review mostly supports changes in both DC and SE as mediators of treatment outcome.•Evidence of mediation is stronger for self-efficacy than dysfunctional cognitions.•Facilitating self-efficacy gains related to exposure could improve treatments.