Skriner, Laura Christine. Trajectories and predictors of treatment response in CBT for youth anxiety. Retrieved from https://doi.org/doi:10.7282/T3FF3VC1
DescriptionObjective: To enhance knowledge of symptom change and treatment response in cognitive behavioral therapy (CBT) for youth anxiety by adopting an Integrative Data Analysis (IDA) approach. The study used an IDA framework to combine existing data from nine clinical trials of CBT for anxious youth (N = 832) and identify distinct trajectories of anxiety symptoms during and following treatment, as well as predictors of these trajectories. Method: Youth and parent-reported anxiety symptom data from the nine contributing clinical trials were combined using item response theory (IRT) models. Growth mixture modeling was used to identify distinct trajectories of treatment response using IRT-scored anxiety symptom data across four time points including: pre-, mid-, and post-treatment, and 1-year follow-up. Once identified, several pre-treatment client demographic and clinical profile traits were tested as predictors of trajectory classes. Results: Growth mixture modeling identified three trajectory classes based on parentreported symptoms: steady responders (71.0%), rapid responders (7.2%), and delayed improvement (21.4%). Four classes were identified based on youth-reported symptoms: steady responders (55.1%), rapid responders (12.6%), delayed improvement (7.1%), and iii low symptom responders (25.3%). Number of diagnoses, youth age, treatment type, and youth gender predicted trajectory class in both child- and parent-reported anxiety models. Delayed improvement classes were predicted by number of pre-treatment diagnoses (based on parent and youth report);receiving family versus individual CBT predicted membership in the delayed improvement compared to all other response classes and also in the steady responder compared to rapid responder class (based on youth report); rapid responders were predicted by older age (parent report) and number of pre-treatment diagnoses (parent report); and low symptom responders were more likely to be male (youth report) compared with those in the steady responder class. Conclusions: The use of an IDA framework allowed for the identification of distinct patterns of symptom change during and following CBT for youth anxiety that have not been previously identified in individual trials. Diagnostic complexity, age, gender, and treatment modality differentiated response classes.