Title:

Comparison of Fitness Trackers in Predicting Daily Step Counts in a College Population

Objective: To determine the agreement of popular fitness trackers used by college students, ages 18-24, compared to a validated pedometer in a free-living environment. Methods: Data were collected between 2016-2019 through the College Health and Nutrition Assessment survey, an ongoing, cross-sectional study at a midsize, northeastern university. Students (n=1034; 71% females; 51% first year) wore the Yamax Digi-Walker SW-200 Pedometer (PED) and another step tracking device of their choice for seven days. Agreement in the average steps between devices were examined via Pearson correlations and paired sample t-tests. Steps differences between PED and three common trackers were evaluated via ANCOVA using gender, semester, and year as covariates. Results: Sixty-five percent (1034 of 1683) of all participants provided step data via PED and an additional tracking device. Average daily steps recorded via PED = 9490±108. Of those that provided tracker data, 8.3% (n=86) used the Apple Watch (AW), 16.4% (n=171) used a FitBit (FB), 48.5% (n=501) used the Health iPhone application (HiPA), and 26.8% (n=277) used other tracking devices. All three devices were strongly correlated to the PED (AW, r=0.92, FB, r=.84, HiPA, r=.73; all p<.001). Both AW and FB overestimated steps (4.4 and 8.2 percent, respectively), while the HiPA underestimated steps (6.6 percent) as compared to PED. Average steps did not significantly differ between students who tracked activity via AW compared to the PED (9728±374 vs. 9461±355, p=.07). Conclusion: Our findings show wrist-worn fitness trackers correlate well to a validated pedometer in free-living college students. Given the popularity of fitness trackers and mobile apps, these results may be important for researchers designing health interventions in the college population. Funding Source: New Hampshire Agriculture Experiment Station and USDA National Institute of Food and Agriculture Hatch Project 1010738.

Name

Comment*

Comments are viewable only by submitter

Captcha*