Differential Effects of Cognitive vs. Motor Dual-Task Training in Stroke Rehabilitation: A Precision-Focused Meta-Analysis

This systematic review and meta-analysis primarily aimed to investigate the differential effectiveness of motor dual-task training (MDT) and cognitive dual-task training (CDT) on gait performance, balance control, and motor function in stroke survivors, and explored other important moderating factors such as stroke chronicity and individual functional profiles to inform a precision-based, personalized approach. Twenty-one RCTs involving 786 stroke survivors were included. Dual-task training demonstrated a medium overall beneficial effect on both temporal and spatial gait performance (SMD=0.50, p=0.03; SMD=0.5, p=0.04) and balance control (SMD=0.71, p=0.02), whereas no statistically significant improvement was observed in lower-extremity motor function. Subgroup analysis revealed that dual-task training modality was a critical determinant of treatment response. MDT was significantly superior for gait performance on both gait speed and stride length (SMD=1.15, p=0.01; SMD=0.89, p<0.01), while CDT demonstrated a significant benefit for balance control (SMD=0.59, p<0.01). Those modality-specific effects were further supported by meta-regression analysis. Stroke survivors at high risk of falls showed greater balance improvements following dual-task training. Furthermore, improvements in balance control and motor function were observed in non-chronic stroke survivors ([&le;]6 months post-stroke) but not in chronic stroke survivors. These results offer crucial prescriptive insights, guiding clinicians to match the dual-task modality and timing of intervention to the individual patients functional profile. However, the high heterogeneity among studies and the lack of direct comparative trials between CDT and MDT limit the conclusive strength of these recommendations.