Measurement Properties of Smartphone Approaches to Assess Physical Activity in Healthy Young People: Systematic Review
Belinda Parmenter,Claire Burley,Courtney Stewart,Jesse Whife,Katrina Champion,Bridie Osman,Nicola Newton,Olivia Green,Annie B Wescott,Lauren A Gardner,Rachel Visontay,Louise Birrell,Zachary Bryant,Cath Chapman,David R Lubans,Matthew Sunderland,Tim Slade,Louise Thornton
DOI: https://doi.org/10.2196/39085
2022-10-21
Abstract:Background: Physical inactivity is a preventable risk factor for several chronic diseases and one of the driving forces behind the growing global burden of disease. Recent evidence has shown that interventions using mobile smartphone apps can promote a significant increase in physical activity (PA) levels. However, the accuracy and reliability of using apps is unknown. Objective: The aim of our review was to determine the accuracy and reliability of using mobile apps to measure PA levels in young people. We conducted a systematic review guided by PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). Methods: Studies published from 2007 to 2020 were sourced from 8 databases-Ovid MEDLINE, Embase (Elsevier), Cochrane Library (Wiley), PsychINFO (EBSCOhost), CINAHL (EBSCOhost), Web of Science (Clarivate), SPORTDiscus (EBSCOhost), and IEEE Xplore Digital Library database. Studies were conducted in young people aged 10-24 years and without chronic illnesses, who evaluated a mobile app's ability to measure PA. Primary outcomes included validity, reliability, and responsiveness of the measurement approach. Duplicate screening was conducted for eligibility, data extraction, and assessing the risk of bias. Results were reported as a systematic review. The main physical activity measures evaluated for each study were the following: total PA time (min/day or min/week), total moderate to vigorous PA per week, daily step count, intensity measure (heart rate), and frequency measure (days per week). Results: Of the 149 identified studies, 5 met the inclusion criteria (322 participants, 176 female; mean age 14, SD 3 years). A total of 3 studies measured criterion validity and compared PA measured via apps against PA measured via an Actigraph accelerometer. The 2 studies that reported on construct validity identified a significant difference between self-reported PA and the objective measure. Only 1 of the 5 apps examined was available to the public, and although this app was highly accepted by young people, the app recorded PA to be significantly different to participants' self-reported PA. Conclusions: Overall, few studies assess the reliability, validity, and responsiveness of mobile apps to measure PA in healthy young people, with studies typically only reporting on one measurement property. Of the 3 studies that measured validity, all concluded that mobile phones were acceptable and valid tools. More research is needed into the validity and reliability of smartphone apps to measure PA levels in this population as well as in populations with other characteristics, including other age groups and those with chronic diseases. Trial registration: PROSPERO CRD42019122242; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=122242.