Wrist Accelerometry for Physical Activity Measurement in Individuals With Spinal Cord Injury—A Need for Individually Calibrated Cut-Points

Abstract

ObjectiveTo create and compare individual and group-based cut-points for wrist accelerometry that correspond to moderate-to-vigorous physical activity (MVPA) in people with spinal cord injury (SCI). DesignParticipants completed a graded treadmill-wheeling test while being assessed for oxygen consumption, wrist-acceleration vector magnitude, and spoke acceleration. Oxygen consumption was converted to SCI metabolic equivalents (METs), and linear regression was applied to determine an individualized vector magnitude cut-point (counts per minute, VM-CPM) corresponding with MVPA (≥3 SCI METs). Multilevel linear regression was applied to determine a group MVPA cut-point. Participants then completed a 6-day monitoring period while wearing the accelerometers. SettingA local SCI research center. ParticipantsManual wheelchair users (N=20; aged 31–64y; injury levels, C5 to L2) with chronic (>1y) SCI. InterventionsNot applicable. Main Outcome MeasuresMean total daily MVPA, wheeled MVPA, and nonwheeled MVPA were calculated using both the individual and group cut-points. Agreement on measures of minutes per day of MVPA between the individual and group mean cut-point method was assessed using Bland-Altman plots. ResultsIndividual cut-points for MVPA ranged from 6040 to 21,540 VM-CPM, with a group cut-point of 11,652 (95% confidence interval, 7395–15,909). For total daily MVPA, Bland-Altman analysis revealed a bias of .22±33.0 minutes, with 95% limits of agreement from −64.5 to 64.9 minutes, suggesting a large discrepancy between total MVPA calculated from individual and group mean cut-points. ConclusionsIndividual calibration of wrist-worn accelerometry is recommended for effective habitual PA monitoring in this population.