Validation of smartphones and different activity trackers for step counting under free living conditions

Oral Presentation A6.2

Authors

  • Claire Marie Jie Lin Goh National University of Singapore; National University Health System
  • Nan Xin Wang National University of Singapore; National University Health System
  • Andre Matthias Müller National University of Singapore; National University Health System
  • Rowena Yap National University of Singapore; National University Health System
  • Sarah Edney National University of Singapore; National University Health System
  • Falk Müller-Riemenschneider National University of Singapore; National University Health System

DOI:

https://doi.org/10.14288/hfjc.v14i3.397

Keywords:

Validation, Smarphone, Activity Tracker, Step Counts

Abstract

Background: Smartphones and wrist-worn devices have been becoming more popular for step counting purposes and physical activity promotion. However, the validity of these devices under free-living conditions has not been adequately determined. Purpose: To investigate the criterion validity of smartphones and activity trackers used in a nation-wide physical activity program under free living conditions. Methods: Participants (N=34), aged 21-31 years wore a waist-worn pedometer and seven different wrist-worn activity trackers continuously over three days. The number of step counts reflected on the participants smartphone for each of the three days was recorded at the end of the study. To establish criterion validity, smartphones and activity trackers were compared to the pedometer using Pearson’s correlation coefficient, mean absolute percentage error (MAPE) and Intraclass Correlation Coefficient (ICC). Results: The mean reported wear time for the 34 participants across three days was 12.9 hours. All 34 participants had provided tracker data and 23 of them had provided valid smartphone data. On average, five out of the seven activity trackers underestimated step counts while the remaining two and the smartphone overestimated step counts. Criterion validity was consistently higher for the activity trackers (r=0.78-0.92; MAPE 14.5-36.1%; ICC: 0.51-0.91) than the smartphone (r=0.37; MAPE 55.7%; ICC: 0.36). Conclusions: The accuracy of using a smartphone for measuring step counts was substantially lower than that of using wrist-worn activity trackers. Other factors will need to be considered when using a smartphone to track step counts. Funding: Study funded by the Singapore Health Promotion Board.

Published

2021-09-30

How to Cite

Goh, C. M. J. L., Wang, N. X., Müller, A. M., Yap, R., Edney, S., & Müller-Riemenschneider, F. (2021). Validation of smartphones and different activity trackers for step counting under free living conditions: Oral Presentation A6.2. The Health & Fitness Journal of Canada, 14(3). https://doi.org/10.14288/hfjc.v14i3.397