217 Views

1520 - Evaluating Accelerometry for Physical Activity Intensity Measurement in Stroke Rehabilitation: A Structured Approach Using the V3+ Framework

Thursday, February 27, 2025

5:00 PM - 6:30 PM MST

Presenting Author(s)

OA

Ovuokerie Addoh, MD

Asst Professor
University of Washington
Seattle, Washington, United States

Co-Author(s)

JS

Jessica Sher, MD

Resident Physician
University of Washington
Seattle, Washington, United States
SK

Stanley M. Kamande, MD

Resident Physician, PGY-3
University of Washington, Department of Rehabilitation Medicine
Seattle, Washington, United States
CB

Charles H. Bombardier, PhD

Professor
University of Washington School of Medicine
Seattle, Washington, United States
DR

Dori Rosenberg, PhD, MPH

Senior Investigator
Kaiser Permanente Washington Health Research Institute
Seattle, Washington, United States

Objectives: Physical activity intensity is an underutilized biomarker in stroke rehabilitation. This review examines the potential of accelerometry for assessing physical activity intensity during inpatient stroke rehabilitation, addressing the limitations of current methods like therapy duration and self-reported exertion.

Design: We apply the novel V3 and V3+ frameworks, which recommend accelerometry verification, usability, analytical, and clinical validation to establish its role in stroke rehabilitation. A comprehensive literature search was conducted, and findings were synthesized using a narrative approach.

Results: Verification ensures that accelerometry accurately captures motion data. Although several accelerometers have been verified, the increasing variety of models and device-specific algorithms require more rigorous validation to maintain accuracy for stroke rehabilitation. Usability validation focuses on the practicality of accelerometers during stroke rehabilitation. While recent studies have shown promise in portability and automation, stroke-specific challenges like hemiplegia, spasticity, and cognitive impairments, as well as cost and availability biases, must be addressed to improve real-world applicability. Analytical validation compares the accelerometer’s physical activity intensity measurements against established standards. Current prediction models, such as those by Freedson and Crouter, may be inaccurate for stroke survivors, and device placement (e.g., wrist vs. ankle) significantly impacts accuracy. Clinical validation assesses how accelerometer data correlates with critical stroke rehabilitation outcomes, including improvements in physical and cognitive function. Frameworks like the rehabilitation treatment specification system could help link activity intensity to meaningful clinical outcomes.

Conclusions: While accelerometry holds promise for measuring physical activity intensity in stroke rehabilitation, further research is needed to refine models and algorithms specific to stroke survivors. Future work should focus on developing stroke-specific accelerometry tools to optimize rehabilitation outcomes. Similar principles can be applied to validate other related facets such as step counts.