A Comparison of Hip-Worn and Wrist-Worn ActiGraph Data in Low-Active Older Adults with Obesity.

Jason Fanning, Michael E Miller, Shyh-Huei Chen, W. Jack Rejeski

Key findings:

Light home chores produced higher and more variable vector magnitude (VM) values than fast walking via wrist ActiGraph.

Vector magnitude values increased with intensity of ambulatory tasks, this was as expected due to the fact that wrist and waist movement both increase with ambulatory effort.

Click Here to Access the Paper

Why does this topic matter?

The issue with incorrectly categorizing accelerometer data has important implications. An accelerometer at the wrist can move fast with low metabolic output, and thus misclassifies activities to be vigorous when they are actually not. When these activities are categorized this way, individuals are thought to be engaging in moderate to vigorous physical activity, and this can change their profiles of daily activity. This trend could potentially mislead an interventionist into thinking their physical activity intervention program is achieving the appropriate movement patterns when in reality it is not.

This study shows that we cannot be confident in the information wrist accelerometers give us in regards to activity profiles of older adults (a population which engages in vastly more activities of daily living as their physical activity than exercise). Promising methodology that uses machine learning might prove useful in this situation as we grapple with accuracy from wrist accelerometry when concerned with studying movement patterns of older adult populations.

About the author

Justin Robison is a first year graduate student at Wake Forest University. His profile is here