Patient-Centric Real-world Functional Outcomes for Amyotrophic Lateral Sclerosis Using Wearable Digital Health Technologies



Poster Number: 176


Rakesh Pilkar, PhD, ActiGraph LLC, Cory Holdom, The University of Queensland, Dawid Gerstel, PhD, ActiGraph LLC, Matthew Patterson, PhD, ActiGraph LLC, Ali Neishabouri, PhD, ActiGraph LLC, Frederik Steyn, PhD, The University of Queensland, Christine Guo, PhD, ActiGraph LLC

Background – Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that progressively damages parts of the nervous system, causes muscle degradation, and significantly affects movement and physical activity (PA). Quantification of PA and mobility, an essential component for understanding ALS progression as well as intervention outcomes, is usually performed by subjective clinical evaluations or burdensome laboratory-based experiments. Remote monitoring of patients with ALS (pwALS) using wearable sensors can reduce the patient burden of lab visits and provide objective and sensitive continuous data to quantify real-world behavior (PA, mobility, sleep). Current work presents the feasibility of unobtrusive wrist-based accelerometry to extract a set of PA outcomes in pwALS.

Methods – ActiGraph GT9X (ActiGraph, Pensacola, FL) were placed on non-dominant wrists of 24 pwALS and 57 age-matched controls. The accelerometer data were continuously recorded for 7 to 8 days at 30 Hz. The data processing pipeline equipped with state-of-the-art PA algorithms combined with wear and sleep algorithms as filters provided daily total minutes of physical activity (light, moderate, vigorous, moderate-to-vigorous (MVPA), sedentary, and locomotion. All activity metrics were presented as a percent of total time and were aggregated over days with at least 10 hours of valid data.

Results – Man-Whitney U tests showed significantly lower light PA (control: 39.35±11.75 %, ALS: 28.73±16.02 %; U=1027, p<0.05), and MVPA (control: 7.11±3.94 %, ALS: 2.2±2.97 %; U=1189, p<0.05). The ALS group showed significantly higher daily sedentary time (69.07±16.59 %) compared to the control group (53.54±12.83 %; U=286, p<0.05). Real-life mobility, as measured by total time spent in locomotion, was significantly lower for the ALS (4.65±3.18 %) compared to the controls (5.56±2.45 %; U=876, p=0.047). Conclusion – Wrist-accelerometry can provide rich data to extract objective and sensitive measures of PA in free-living. In the future, gait and sleep metrics will be examined.