Evaluating the validity of a smartphone step-counter in adults with asthma: a proof-of-concept study

Introduction: Regular physical activity and structured exercise are often reported to be associated with improved asthma control - however the majority of published evidence is limited by short-term studies employing subjective measures of assessment (i.e. self-report / questionnaires). Modern smartphones typically include built-in activity sensors (i.e. possess the capability to monitor daily step-count) and thus may offer a cost-effective and pragmatic solution to the assessment of physical activity in clinical practice and/or research trials. The primary aim of this proof-of-concept study was therefore to evaluate the validity of the iPhone® (Apple Inc, USA) step-counter in adults with asthma and healthy controls. Methods: The study was conducted as a cross-sectional laboratory based-trial. Ten healthy adults with no prior history of respiratory disease and ten adults with a prior physician diagnosis of asthma were enrolled. All completed baseline clinical assessment followed by a standardised walking treadmill challenge consisting of 3 x 3-minute stages at pre-determined speeds: 2.5kph, 5.0kph and 7.5kph. Steps were recorded using the following devices: (i) Yamax Digiwalker™ SW-200 Pedometer (Yamax, UK), (ii) iPhone® step-counter (upper body arm-band), (iii) iPhone® step-counter (lower body trouser pocket) - and evaluated against a video-verified manual step-count (i.e. gold-standard comparator) conducted by the investigator (CR). Results: No difference was observed in manual total step-count between individuals with asthma (1018 steps) and healthy controls (1038 steps) (P=0.44). The iPhone® step-counter (both upper and lower body) provided close agreement with video-verified manual step-count, and importantly, outperformed the Yamax Digiwalker® SW-200 Pedometer across the majority of test stages. Specifically, the iPhone® (lower body) correlated strongly (r = 0.96; P<0.006) and produced the highest level of agreement with video-verified total step-count (mean bias: -11; limits of agreement: -43 to 21) (Table 1). Conclusion: Our findings indicate that the iPhone® provides an accurate estimate of step-count in adults with asthma and healthy controls completing a standardised laboratory-based treadmill test. Prior to implementation, further research is required to determine the validity and reliability of this approach during daily active / free living conditions.