Predicting Outcomes from Cognitive Behavioural Therapy for Social Anxiety Disorder: A Bayesian Network Analysis
Author(s):
Emily Rogerson1, Mel Simmonds Buckley2, Stephen Kellett2* and Jaime Delgadillo1
Better understanding of predictors of treatment response to cognitive behavioural therapy (CBT) for social anxiety disorder (SAD) holds potential for improving outcomes. This study sought to identify which specific social anxiety symptoms measured pretreatment were associated with post-treatment outcomes. A pre-registered retrospective cohort study was used in a sample of N=1315 patients treated with CBT for SAD in routine clinical practice. The outcome was a reliable and clinically significant improvement (RCSI) on the Generalized Anxiety Disorder-7 outcome measure. A Bayesian network symptom model of the Social Phobia Inventory (SPIN) was trained using Bayesian network analysis with 10- fold cross-validation (n=658). Predictive accuracy was evaluated in an external test sample (n=657). The performance of the network model generalised to an external test sample (AUC = 0.59; PPV = 0.53; NPV = 0.59) with moderate prediction shrinkage relative to performance in the test sample (AUC = 0.67). A network of four interrelated SAD symptoms were found to be reliable outcome predictors: fear of embarrassment, avoiding talking, avoiding criticism and fear of being observed. Identifying important SAD symptoms at assessment could enable these to be targeted during CBT to help maximize treatment efficiency and effectiveness.
