Assessment of Accelerometers for Measuring Upper-Extremity Physical Activity
Click name to view affiliation
Search for other papers by Elizabeth Lawinger in
Current site
Google Scholar
PubMed
Search for other papers by Tim L. Uhl in
Current site
Google Scholar
PubMed
Search for other papers by Mark Abel in
Current site
Google Scholar
PubMed
Search for other papers by Srinath Kamineni in
Current site
Google Scholar
PubMed
Objective:
The overarching goal of this study was to examine the use of triaxial accelerometers in measuring upper-extremity motions to monitor upper-extremity-exercise compliance. There were multiple questions investigated, but the primary objective was to investigate the correlation between visually observed arm motions and triaxial accelerometer activity counts to establish fundamental activity counts for the upper extremity.
Study Design:
Cross-sectional, basic research.
Setting:
Clinical laboratory.
Participants:
Thirty healthy individuals age 26 ± 6 y, body mass 24 ± 3 kg, and height 1.68 ± 0.09 m volunteered.
Intervention:
Participants performed 3 series of tasks: activities of daily living (ADLs), rehabilitation exercises, and passive shoulder range of motion at 5 specific velocities on an isokinetic dynamometer while wearing an accelerometer on each wrist. Participants performed exercises with their dominant arm to examine differences between sides. A researcher visually counted all arm motions to correlate counts with physical activity counts provided by the accelerometer.
Main Outcome Measure:
Physical activity counts derived from the accelerometer and visually observed activity counts recorded from a single investigator.
Results:
There was a strong positive correlation (r = .93, P < .01) between accelerometer physical activity counts and visual activity counts for all ADLs. Accelerometers activity counts demonstrated side-to-side difference for all ADLs (P < .001) and 5 of the 7 rehabilitation activities (P < .003). All velocities tested on the isokinetic dynamometer were shown to be significantly different from each other (P < .001).
Conclusion:
There is a linear relationship between arm motions counted visually and the physical activity counts generated by an accelerometer, indicating that arm motions could be potentially accounted for if monitoring arm usage. The accelerometers can detect differences in relatively slow arm-movement velocities, which is critical if attempting to evaluate exercise compliance during early phases of shoulder rehabilitation. These results provide fundamental information that indicates that triaxial accelerometers have the potential to objectively monitor and measure arm activities during rehabilitation and ADLs.
Lawinger and Uhl are with the Div of Athletic Training; Abel, the Dept of Kinesiology and Health Promotion; and Kamineni, the College of Medicine, University of Kentucky, Lexington, KY.
© 2024 Human Kinetics