Related Experiment Video
Updated: Jul 15, 2025
Assessing the Accuracy of Fitness Smartwatch Data for Cardiovascular and Physical Activity Monitoring: A Validation Study in Digital Health
Published on: February 21, 2025
Concurrent validity of walking speed measured by a wearable sensor and a stopwatch during the 10-meter walk test in
Brice T Cleland1, Titus Alex1, Sangeetha Madhavan1
1Brain Plasticity Lab, Department of Physical Therapy, College of Applied Health Sciences, University of Illinois Chicago, Chicago, IL, USA.
Background:
Walking speed is often measured with a stopwatch throughout stroke recovery. Wearable sensors also have been used recently to measure walking speed and provide information about spatiotemporal characteristics of walking.
Research Question:
Do walking speeds measured with stopwatch and APDM wearable sensors have concurrent validity?
Methods:
Individuals with chronic stroke (n = 62) performed the 10-meter walk test at comfortable and maximal speeds. Walking speeds were measured with a stopwatch and APDM Opal wireless wearable sensors (3-unit). Tests of concurrent validity between stopwatch and APDM (Bland-Altman plots, systematic and proportional bias, and intraclass correlations) and test-retest reliability between trials (intraclass correlations, standard error of measurement, and minimal detectable change) were performed.
Results:
Walking speeds measured with APDM were ∼0.07 m/s slower than those measured with stopwatch (systematic bias; t ≥ 13.1, p < 0.001). Intraclass correlations ranged from poor to excellent. There were greater differences in walking speeds between APDM and stopwatch for individuals with faster walking speeds (proportional bias). Test-retest reliability was excellent for both APDM and stopwatch (intraclass correlation≥0.94). Standard error of measurement ranged from 0.04 to 0.07 m/s and minimal detectable change ranged from 0.10 to 0.19 m/s.
Significance:
It may be inappropriate to use walking speed measurements from APDM sensors and stopwatch interchangeably in individuals with chronic stroke. Differences in walking speeds may reflect stopwatch error or the derivation of walking speed from wearable sensors. Test-retest reliability was excellent for both stopwatch and APDM, but minimal detectable change values were large. Large changes in walking speed may be required to be confident that the change is a true and clinically meaningful change and not measurement error. The validity and reliability of measuring walking speed with wearable sensors in individuals with chronic stroke has important implications for determining community ambulation, assessing improvements after rehabilitation, and developing exercise prescriptions.
