Development of a wrist-worn accelerometer feedback intervention to enhance arm recovery after stroke

Abstract

Background Arm recovery after stroke is enhanced by frequent practice of functional activities. Remaining motivated to practice and remembering to integrate the impaired limb into daily activities can be difficult. Methods A mixed methods approach developed a novel intervention to promote use of the impaired arm after stroke: 1. Systematic review examined reports of self-directed interventions for arm rehabilitation after stroke. 2. Development of a novel intervention using a wrist-worn accelerometer with vibrating alert to prompt arm activity. 3. An un-blinded observational proof of concept study within 4 weeks of stroke refined the intervention. 4. A multi-centre, observer-blind pilot randomised controlled trial (RCT) evaluated feasibility of the intervention for 8 weeks within 3 months of stroke and provided descriptive clinical and biomarker data. Results The systematic review showed that high doses of independent practice are possible, with benefits from functional task practice. A novel intervention was developed consisting of feedback from a wristband accelerometer to prompt increases in functional therapy practice within daily routines. The proof of concept study showed that feedback was acceptable with refinements to the technology and therapy programme. Thirty-three participants were recruited to the Pilot RCT. Research assessments were completed for 28/29 and 25/28 patients at four and eight weeks. Wristbands were worn for 79% of the recommended time with a median of 8[IQR: 6-10] prompts delivered per participant/day. ii Clinical outcomes were better for intervention participants and continued to improve post-intervention although the small cohort size precluded statistical analysis. Statistical analysis of the accelerometer data showed impaired arm activity increased for the intervention group and continued to increase further over the follow-up period. In contrast, arm activity in the control group changed marginally. Conclusion Feedback from a wristband accelerometer to prompt greater use and independent practice of the impaired arm after stroke is feasible and should be considered for further research evaluation.