Validity of Consumer Activity Monitors and an Algorithm Using Smartphone Data for Measuring Steps during Different Activity Types

Abstract

Background: Consumer activity monitors and smartphones have gained relevance for the assessment and promotion of physical activity. The aim of this study was to determine the concurrent validity of various consumer activity monitor models and smartphone models for measuring steps. Methods: Participants completed three activity protocols: (1) overground walking with three different speeds (comfortable, slow, fast), (2) activities of daily living (ADLs) focusing on arm movements, and (3) intermittent walking. Participants wore 11 activity monitors (wrist: 8; hip: 2; ankle: 1) and four smartphones (hip: 3; calf: 1). Observed steps served as the criterion measure. The mean average percentage error (MAPE) was calculated for each device and protocol. Results: Eighteen healthy adults participated in the study (age: 28.8 ± 4.9 years). MAPEs ranged from 0.3–38.2% during overground walking, 48.2–861.2% during ADLs, and 11.2–47.3% during intermittent walking. Wrist-worn activity monitors tended to misclassify arm movements as steps. Smartphone data collected at the hip, analyzed with a separate algorithm, performed either equally or even superiorly to the research-grade ActiGraph. Conclusion: This study highlights the potential of smartphones for physical activity measurement. Measurement inaccuracies during intermittent walking and arm movements should be considered when interpreting study results and choosing activity monitors for evaluation purposes.

Evaluation of a Low-Cost Commercial Actigraph and Its Potential Use in Detecting Cultural Variations in Physical Activity and Sleep

“…In a recent study, Höchsmann et al [ 52 ] found that the wrist GT3X+ step count was overestimating steps especially during low, but active (e.g., public transport or car driving) or even inactive (e.g., sitting or laying) situations, when compared with ankle-based measurements (see also Figure A1 ). Thus, in principle, the differences between the MB and GT3X step counts in the current study could also be explained by an overestimation of the GT3X wrist step count in free-living conditions, which is also in line with the observation of Hartung et al [ 29 ]. Based on the current results, we therefore concluded that the MB adequately captured activity fluctuations across days, but raw absolute values were hard to compare and differed widely between actigraphy devices.…”

“…Part of this disagreement could be expailned by including only 44 days of recorded data under free-living conditions in a sample of 20 participants in Degroote’s study, compared to 162 days and 21 participants analyzed in the current study (mean = 8; range: 5–13). According to Hartung et al [ 29 ], MB counted 63 out of 250 manually counted steps too few, whereas the GT3X counted even 30 steps more in a natural walking condition, confirming the general underestimation of the MB device for step counts. In a recent study, Höchsmann et al [ 52 ] found that the wrist GT3X+ step count was overestimating steps especially during low, but active (e.g., public transport or car driving) or even inactive (e.g., sitting or laying) situations, when compared with ankle-based measurements (see also Figure A1 ).…”

“…In Study 1, we evaluated the performance of MB to objectively measure physical activity and sleep measures. The low-cost of MB in combination with its promising performance in extracting physical activity and sleep-related measures [ 23 , 24 , 29 , 30 ] endorsed the utilization of this device for a cross-cultural actigraphy study. We revealed that step counts were not influenced by the exact actigraph wrist position, neither for MB, nor for the GT3X.…”

“…Xie et al [ 23 ] also evaluated the accuracy of MB by comparing it with manually counted steps and sleep-time and found a high MB accuracy for both measures and even higher accuracy than a high-cost commercial smartwatch (Apple Watch). In another study that examined the performance of commercial actigraphs in continuous and intermittent walking, Hartung et al [ 29 ] found that the MB underestimated on average 10/250 steps compared to the actual observed step count in the continuous walking condition, while 63/250 steps in the intermittent walking condition. In both conditions, the MB performed better than all high-cost smartwatches, but worse than the GT3X.…”

Step Count in Patients With Lumbar Spinal Stenosis

“…It limits the generalizability to everyday life, where further pitfalls not addressed in this study can impact measures of step count. 47,[57][58][59] The implemented method is made available to the research community as open source. This enables further development by any party and ensures transparency and reproducibility by making decisions during data processing that impact results clear.…”