A validation study of two wrist worn wearable devices for remote assessment of exercise capacity

Alexandra Jamieson, Michele Orini, Siana Jones, Nish Chaturvedi, Alun Hughes
UCL


Abstract

The 6-Minute Walk Test (6MWT) is commonly used in clinical settings to assess exercise capacity, a strong independent predictor of cardiovascular and all-cause mortality. Its use in the community would be beneficial but requires remote monitoring technology and a modified protocol allowing individuals to walk freely instead of along 30 m straight paths. Novel wrist-worn technologies provide such an opportunity, however, their magnitude of error in measuring physiological responses to well-established tests is unclear.

The aim of this study was to determine wearable device errors in assessing exercise capacity to develop a framework for a remote, unsupervised, 6MWT.

16 healthy adults (male 7(44%), mean age±SD 27±4 years) performed the 6MWT using two protocols: 1) standard-straight 30 m laps (6MWT-S) and 2) ‘Free range'–circular 240 m laps (6MWT-FR). Each participant was fitted with a Garmin vivoactive4 and Fitbit Sense to measure three parameters: distance, step counts and heart rate (HR) response. Reference measures were a metre-wheel, hand tally counter and ECG, respectively. Mean HR was measured at rest (30 second window), peak exercise (last minute of exercise) and between 1- and 2-minutes recovery. Agreement was measured using Bland-Altman analysis and mean absolute percentage error (MAPE), reported as median [interquartile range].

Distance during the 6MWT-FR was measured with smaller errors than during 6MWT-S for both Garmin (MAPE=9.8%[4.6%,12.6%] vs 18.5%[13.0%,27.4%]; p<0.001) and Fitbit (MAPE=9.4%[4.5%,13.3%] vs 22.7%[18.3%,29.3%]; p<0.001). Garmin measured step count with smaller errors than Fitbit in both 6MWT-S (Fitbit:MAPE=8.1%[5.0%,12.9%]; Garmin:MAPE=2.3%[1.1%,2.9%]; p<0.001) and 6MWT-FR (Fitbit: MAPE=9.5%[3.9%,18.0%]; Garmin: MAPE=1.4%[0.6%,2.1%]; p<0.001). Both devices showed minimal errors in measuring HR at rest (Fitbit: MAPE=2.4%[1.4%,4.3%]; Garmin: MAPE=2.9%[1.5%,6.4%]), 1-minute recovery (Fitbit: MAPE=2.9%[1.1%, 5.1%]; Garmin: MAPE=2.9%[1.0%-5.1%]), and at peak exercise (Fitbit: MAPE=2.0%[1.0%,6.9%]; Garmin MAPE=1.2%[0.4%-9.6%]).

Wearable measurements of the 6MWT are feasible and may provide insights about exercise capacity through remote monitoring.