Clinical validation of smartphone-based activity tracking in peripheral artery disease patients

Abstract Peripheral artery disease (PAD) is a vascular disease that leads to reduced blood flow to the limbs, often causing claudication symptoms that impair patients’ ability to walk. The distance walked during a 6-min walk test (6MWT) correlates well with patient claudication symptoms, so we developed the VascTrac iPhone app as a platform for monitoring PAD using a digital 6MWT. In this study, we evaluate the accuracy of the built-in iPhone distance and step-counting algorithms during 6MWTs. One hundred and fourteen (114) participants with PAD performed a supervised 6MWT using the VascTrac app while simultaneously wearing an ActiGraph GT9X Activity Monitor. Steps and distance-walked during the 6MWT were manually measured and used to assess the bias in the iPhone CMPedometer algorithms. The iPhone CMPedometer step algorithm underestimated steps with a bias of −7.2% ± 13.8% (mean ± SD) and had a mean percent difference with the Actigraph (Actigraph-iPhone) of 5.7% ± 20.5%. The iPhone CMPedometer distance algorithm overestimated distance with a bias of 43% ± 42% due to overestimation in stride length. Our correction factor improved distance estimation to 8% ± 32%. The Ankle-Brachial Index (ABI) correlated poorly with steps (R = 0.365) and distance (R = 0.413). Thus, in PAD patients, the iPhone’s built-in distance algorithm is unable to accurately measure distance, suggesting that custom algorithms are necessary for using iPhones as a platform for monitoring distance walked in PAD patients. Although the iPhone accurately measured steps, more research is necessary to establish step counting as a clinically meaningful metric for PAD.