Clinical Application of Wearable Devices in Stroke Rehabilitation: A Scoping Review of Safety, Feasibility, and Adherence

Abstract

Highlights What are the main findings? Wearable devices are generally safe for stroke rehabilitation, with most adverse events being mild and manageable. Patient adherence is high in clinical settings but more variable in home-based rehabilitation due to device burden and self-management challenges. What are the implications of the main findings? Wearable devices can enable personalized and high-intensity rehabilitation, highlighting the importance of user-friendly and comfortable designs. Successful transition from clinical to home-based rehabilitation requires integrating remote monitoring and structured guidance to optimize adherence and safety. Abstract Background/Objectives: Stroke is a major cause of long-term disability worldwide. Wearable devices have shown potential in supporting rehabilitation, yet evidence on their safety, feasibility, and adherence remains fragmented. This scoping review aimed to evaluate the safety, feasibility, and adherence of wearable devices in stroke rehabilitation. Methods: A systematic search of PubMed, EMBASE, Web of Science, and CINAHL Complete was conducted in January 2026. Studies involving participants with a confirmed diagnosis of stroke were included regardless of study design. Studies were eligible if wearable devices were used as part of rehabilitation interventions and reported outcomes related to safety, feasibility, or adherence. Adverse events were categorized according to severity, while feasibility was assessed using indicators such as recruitment rate, dropout rate, and adherence rate. Results: A total of 38 studies involving 1121 participants were included in this review. Available evidence generally suggested a favorable safety profile for wearable device-assisted rehabilitation, with most reported adverse events being mild and primarily consisting of skin irritation, pressure-related discomfort, muscle fatigue, localized pain, and transient discomfort. Recruitment rates ranged from 5.8% to 100%, indicating substantial variability in implementation feasibility across studies. Only four studies reported adherence-related outcomes, and considerable variation existed in the definitions and assessment methods used. Most included studies reported intervention completion, attendance, or dropout outcomes rather than directly evaluating adherence. Conclusions: Current evidence suggests that wearable devices are generally safe and potentially feasible for stroke rehabilitation. However, substantial heterogeneity across studies, inconsistent adverse event reporting, and limited adherence data restrict the strength of current conclusions. Adherence remains an important evidence gap, as it was infrequently reported and inconsistently defined. Future research should establish standardized reporting criteria for safety, feasibility, and adherence outcomes and evaluate wearable device interventions in larger and more diverse populations with longer follow-up periods.