Factors associated with treatment response in the COMPARE randomized controlled trial of intervention for chronic poststroke aphasia.

Abstract

Introduction: Limited evidence exists of the of different interventions for post-stroke aphasia or participant factors associated with treatment response. COMPARE was a Phase III multicentre, single blinded, randomised controlled trial (n=216). COMPARE investigated whether Constraint-Induced Aphasia Therapy Plus (CIATPlus) and/or Multimodality Aphasia Therapy (M-MAT) resulted in improved language, functional communication, multimodal communication, and quality of life outcomes compared to usual care (UC) in people with chronic post-stroke aphasia. Compared to UC, CIAT Plus and M-MAT were more effective for word retrieval, functional communication, and quality of life. Aims/Hypothesis: We aimed to investigate participant baseline demographic, linguistic, and cognitive variables associated with treatment response in COMPARE. Method(s): Factors associated with improvements in outcomes of treated participants were assessed at post-intervention and 12-week follow-up using multivariable logistic regression analysis. Outcomes assessed were clinically meaningful improvements on the Western Aphasia Battery Revised Aphasia Quotient (WAB-R-AQ), Naming Battery, Communicative Effectiveness Index (CETI), Stroke and Aphasia Quality of Life Scale-39, and Scenario Test. The predictive performance of models was assessed using the area under the receiver operating characteristic curve (AUC). Result(s): We analysed 140 participants (70 CIAT-Plus; 70 M-MAT). There was considerable variability in treatment response. Predictive performance of models was mostly satisfactory (AUC>0.6). There were no factors consistently associated with achieving a clinically meaningful improvement across all outcome measures, but age, WAB-R-AQ score at baseline, and time since stroke were the most common factors that were most strongly associated with outcomes. Conclusions/Discussion: Age, aphasia severity, and time since stroke predicted treatment response. Further work is required to determine the critical baseline thresholds across a range of variables that underpin response to M-MAT and CIAT-Plus in different outcomes of interest. This evidence is required to support individualised treatment prescription.