Validity and reliability of smartphones in assessing spinal kinematics: a systematic review and meta-analysis

Abstract

Objective Advances in mobile technology have led to the development of smartphones, whose applications present numerous utilities, such as the analysis of human movement based on inertial sensors. The purpose of this review was to investigate validity and reliability of smartphones in assessing the kinematics of the human spine. Methods A systematic search was performed on MEDLINE, Embase, Scopus, and LILACS databases, as well as manual searches. The included studies evaluated psychometric properties of smartphones in assessing kinematic variables of the spine (range of motion [ROM], speed, and acceleration). Two independent reviewers performed the selection, reading, data extraction, and risk of bias assessment of the studies. Results Of the 2651 articles initially found, 9 were included and had their results for ROM analyzed. The meta-analyses for validity showed very high correlation coefficients in the evaluation of cervical flexion, extension, and lateral flexion; high ones in the evaluation of cervical rotation; and also high ones for intrarater and interrater reproducibility of all cervical movements. The meta-analyses for interrater reproducibility showed high correlation coefficients in the evaluation of lumbar flexion and very high ones for intrarater reproducibility. Conclusion The use of smartphones for assessing the ROM of cervical flexion, extension, and lateral flexion and lumbar flexion is feasible. Their use for assessing thoracic rotation is potentially viable, but further validation studies are still needed to ensure a safe use. There is a lack of validation studies that evaluate the applicability of this device in assessing other kinematic characteristics, such as speed and acceleration.