As the digital health landscape continues to evolve, researchers are increasingly looking to maximize the power of their data by pooling datasets across multi-year longitudinal studies, or by combining independent datasets across different cohorts within a specific disease or therapeutic area. However, a critical question keeps coming up: how comparable are physical behavior measurements across different device generations/models?
Whether you are harmonizing data over a decade-long study or attempting to aggregate data from different clinical trials evaluating a specific disease area, minor variations in device performance or characteristics can subtly alter how we interpret physical activity, sedentary time, and sleep at both the individual and group levels. To address this, our team set out to establish a standardized, reproducible verification protocol to test device comparability directly.
Dr. LaMunion presented this research and key study findings at the annual ADDS 2026 scientific conference in February 2026. Watch the on-demand recording below or dive even deeper with the complete technical white paper.
Watch the On-Demand Presentation:
Device Comparability and Data Compatibility Across ActiGraph Generations, Samuel R. LaMunion, PhD
This original research, now published in Medicine & Science in Sports & Exercise (MSSE), represents a multi-year collaborative effort between the National Institutes of Health (NIH) and Ametris, (formerly ActiGraph).
About the Author
Disclosures: a) The views presented in this article and the accompanying presentation are entirely my own and do not constitute an endorsement by the National Institutes of Health or the U.S. government. Samuel LaMunion has engaged in paid consulting work for Ametris, however, the entirety of this work pre-dates that affiliation.
Related Resources: