Accuracy and Precision of Smartphone Applications and Commercially Available Motion Sensors for Measuring Steps in Multiple Sclerosis

Background: There is increasing interest in the adoption of smartphone applications and wearable motion sensors for measuring community-based, ambulatory physical activity in persons with multiple sclerosis (MS). Researchers recently examined the feasibility of using a Fitbit One for monitoring physical activity among people with MS, providing a strong case for the usability of wearable sensors. The next step requires research examining the accuracy and precision of the applications and sensors for measuring steps taken while ambulating using a standard protocol. 

Objectives: This study examined the accuracy and precision of three smartphone applications and five wearable motion sensors for measuring actual steps taken, using direct observation as a gold standard, while walking 500 steps on a treadmill in a sample of persons with MS. 

Methods: The wearable motion sensors included in this study were the Digi-Walker SW-200 pedometer (Yamax), UP2 and  Up Move (Jawbone), and the Flex and One (Fitbit). The three smartphone applications were My Health (Apple), Health Mate (Withings), and Moves (ProtoGeo Oy). Forty-five ambulatory participants with MS underwent neurological examinations for generation of EDSS scores, followed by two 500 step walks at comfortable walking speed while wearing all of the devices and the smartphone. The accuracy and precision of the motion sensors and smart phone applications was expressed in absolute and relative metrics using SPSS Statistics 22.0.

Results: The FitBit One had the best absolute (M=490.6 steps, 95%CI=485.6-495.5 steps) and relative accuracy (1.9% error), and absolute (SD=16.4) and relative precision (CV=0.0) of all 8 applications and sensors for the first trial of walking 500 steps. The FitBit One again had the best absolute (M=492.2 steps, 95%CI=487.6-496.8 steps) and relative accuracy (1.9% error) and absolute (SD=15.4) and relative precision (CoV=0.0) of all applications and sensors for the second trial of walking 500 steps

Conclusions: The waist worn FitBit One had the best accuracy and precision of all applications and sensors tested. Such evidence is important and indicates the FitBit One has the necessary accuracy and precision for measurement of steps taken in clinical research and practice involving persons with MS.