Determinants of Dual-Task Training Effect Size in Parkinson Disease: Who Will Benefit Most?

Background and Purpose: Dual-task interventions show positive effects in people with Parkinson disease (PD), but it remains unclear which factors determine the size of these benefits. As a secondary analysis of the DUALITY trial, the aim of this study was to assess the determinants of the effect size after 2 types of dual-task practice. Methods: We randomly allocated 121 participants with PD to receive either integrated or consecutive dual-task training. Dual-task walking performance was assessed during (i) a backward digit span task (digit), (ii) an auditory Stroop task (Stroop), and (iii) a functional mobile phone task. Baseline descriptive, motor, and cognitive variables were correlated with the change in dual-task gait velocity after the intervention. Factors correlated with the change in dual-task gait velocity postintervention (P < 0.20) were entered into a stepwise forward multiple linear regression model. Results: Lower dual-task gait velocity and higher cognitive capacity (Scales for Outcomes in Parkinson's Disease-Cognition [ScopaCog]) at baseline were related to larger improvements in dual-task gait velocity after both integrated and consecutive dual-task training for all 3 tasks (beta[gait] = -0.45, beta[ScopaCog] = 0.34, R-2 = 0.23, P < 0.001, for digit; beta[gait] = -0.52, beta[ScopaCog] = 0.29, R-2 = 0.26, P < 0.001, for Stroop; and beta[gait] = -0.40, beta[ScopaCog] = 0.30, R-2 = 0.18, P < 0.001, for mobile phone task). Discussion and Conclusions: Participants with PD who showed a slowdual-task gait velocity and good cognitive functioning at baseline benefited most from the dual-task training, irrespective of the type of training and type of dual-task outcome.