Abstract 14796: Validity of Smartphone Pedometer Apps as a Tool to Monitor Physical Activity Remotely
Introduction: Over 65% of cellphones in the US are smartphones and the market share is rapidly growing. These devices allow many applications through mobile apps. The objective of this study was to evaluate the readiness of free smartphone pedometer apps as a tool to monitor activity.
Hypothesis: Free pedometer Apps are a valid tool for activity monitoring.
Methods: We reviewed all the pedometer apps from the app stores, Apple (iOS A) and Android (AA)(combined 93% of market share), and selected them for further review if they were free of charge and capable of storing and sharing data (step count and time). Two independent raters downloaded and evaluated all the selected apps to confirm compliance with requisites, background running and user friendliness. We recruited 12 volunteers to test the Apps without adjusted sensitivity settings. We randomized testing to phone location (waist vs shirt pocket) and walking speed on a treadmill (1, 2, and 3 mph for 6 minutes at each speed) and during a 6-minute walk test. All tests were performed at least twice by each volunteer, and the criterion measure of steps was obtained by manual counting using a hand tally counter. For statistical analysis we used Wilcoxon’s signed rank test, in addition to Pearson correlation coefficients and Bland-Altman plots.
Results: 219 iOS A and 181 AA were reviewed. Of those, 13 iOS A and 9 AA were selected, downloaded and rated by two independent reviewers. Three apps from each system fulfilled the criteria and were analyzed. Accupedo and Pedometer SP (iOS A) and Accupedo (AA) step counts were not statistically different (p>0.05) than the counted steps, Pearson correlation coefficients were significant (p< 0.05) and were 0.76, 0.93 and 0.91, respectively. The Bland-Altman analysis revealed that these apps underestimate the step counts at the treadmill speed of 1 mph.
Conclusion: Despite the number of free pedometer apps, only a small subset is suitable for remote physical activity monitoring. Apps perform better at walking speeds higher than 1 mph without sensitivity adjustment. The technology and ubiquitous nature of smartphones appears that they are ready to use for activity monitoring in a clinical setting, which may aid in monitoring home walking programs.
Author Disclosures: G.G. Fontana: None. A.J. Tafur: None. A.W. Gardner: None. A.I. Casanegra: None.
- © 2014 by American Heart Association, Inc.