Article Title



K. Kracher, N. Holliday, A. Vahk, N. H. Lawton

Eastern Washington University, Cheney, WA

A recent ACSM Sports Medicine Bulletin reported the number one fitness trend for 2016 would be wearable fitness tracking technology. Smart phone exercise applications were ranked 17th in the survey. The final note mentioned that like all fitness tools, technology must be put to use to be effective, begging the question of how a multipurpose device like a cell phone will affect fitness tracking data. PURPOSE: The purpose of this study was to compare the effect of cell phone use during walking on activity tracker accuracy between a hip worn pedometer (PD), wrist worn activity tracker (AT), and cell phone (CP). METHODS: 37 healthy adults (aged 24.8 ± 6y) who own a smart phone and utilize a wrist worn AT participated in this study. Height and weight data were collected and entered into a pedometer placed on the left hip. Participants were instructed to walk on a treadmill at a self-selected casual walking pace during a 5 min warm-up period to establish an appropriate speed and to test all activity tracking devices. Following the warm-up period, subjects completed two 5 min data collection periods, one with the CP in the front right pocket (CPP) and the second with the CP in their hands (CPH). The wrist worn AT remained on the same wrist during all data collection sessions. During the CPH period subjects were instructed to use the CP as if they were on a casual walk. Researchers did not specify if one or two hands should be used. At the end of each data collection period subjects stepped off of the treadmill belt and step counts were recorded from all three devices. Data was analyzed using a 2x3 repeated measures ANOVA and Pearson Correlation Coefficients were calculated. RESULTS: No significant differences (p < .05) were found between devices within or between CPP and CPH conditions (PD: 495.3 ± 95.9 vs. 498.7 ± 95.3; AT: 496.2 ± 67.4 vs. 484.5 ± 87.8; CP: 487.7 ± 109.7 vs. 469 ± 127.9). A strong positive correlation was found for PD and AT step counts (PD: r = .979, p = .00; AT: r = .724, p = .00), but a weak positive correlation was found for CP (r = .048, p = .78) between conditions. CONCLUSION: High step data variability was likely a result of allowing subjects to use personal devices and self-selected walking speeds. Future studies should control these variables. High step data variability may have affected statistical significance. Notably, trend data suggested CP use while walking had virtually no effect on PD and little effect on AT, but did have a large effect on the ability of the CP to measure steps. These findings suggest that the CP activity tracking applications are more affected by CP use during a walking task than either wrist worn ATs or hip mounted PDs.

This document is currently not available here.