Validity of the Fitbit Charge 2 for measuring steps taken at moderate to fast walking speeds in healthy adults

Abstract

Purpose: To explore the validity of the Fitbit Charge 2 during treadmill walking in a healthy adult population. Design: A validation study with a within-subjects design. Participants: 26 healthy adults (18 females and 8 males, mean age 25.1±8.9 years) able to walk at a brisk walking speed for 2 minutes. Methods: The participants wore the Fitbit Charge 2 on their non-dominant wrist during treadmill walking at two different walking speeds (1.0 and 1.4 m/s) and hand positions (hands by side and hands in pockets) for 2 minutes each. Steps measured by the Fitbit Charge 2 were compared to video observation. Results: Inter-rater reliability of video observations was excellent (all ICC > 0.99). There were no significant differences in steps measured by the Fitbit and visual observation for hands in pockets 1.0 and 1.4 m/s (p>0.05) with mean differences (Fitbit – mean of raters) of 0.5 ± 3.1 and 1.5 ± 5.9 respectively. There were significant differences in steps measured by Fitbit and visual observation for hands by side 1.0 and 1.4 m/s (p<0.05) with mean differences of -5.9 ± 9.6, -9.3 ± 17.0 respectively. Conclusion: The Fitbit Charge 2 is considered valid for measuring steps taken during treadmill walking when walking with hands in pockets. Walking with hands by side at 1.0 and 1.4 m/s underestimated steps. Future research is needed to establish the validity of the Fitbit Charge 2 for this population.