The FitBit Charge 2 is a valid measure of steps at slow walking speeds: an experimental study

Abstract

Question: Does walking at a slower speed affect the concurrent validity of the FitBit Charge 2 when counting steps in a healthy population? Design: A quantitative validity study which investigated the validity of the FitBit Charge 2 compared to visual observation. Participants wore the FitBit and walked for two minutes on a treadmill at 0.6m/s and 1.4m/s. All walks were video recorded, and 2 researchers counted the number of steps taken for comparison. Participants: Twenty-six healthy adults recruited from a university. Outcome measures: The mean number of steps counted by the two raters from the video recordings were compared to the steps counted by the FitBit at both walking speeds. Results: Inter-rater reliability of the two raters was excellent (0.996-0.999). The mean difference in steps counted (FitBit – visual observation) was 8.846 and -9.236 steps for 0.6m/s and 1.4m/s respectively. There was no significant difference in the number of steps counted by the FitBit and visual observation at 0.6m/s (p=0.065), showing that it is a valid measure at this slow walking speed. There was a significant difference in the number of steps counted between methods at 1.4m/s (p=0.015). However, Bland Altman graphs showed a larger spread of data at 0.6m/s which corresponded with a larger percentage error at this speed. Conclusion: The FitBit Charge 2 is a valid measure for counting steps at a slow walking speed of 0.6m/s in healthy participants. However, it should be used with caution due to the variation in results between participants.