Defining accelerometer cut-points for different intensity levels in motor-complete spinal cord injury

Study design Descriptive. Objective The present aim was to define accelerometer cut-point values for wrist-worn accelerometers to identify absolute- and relative-intensity physical activity (PA) levels in people with motor-complete paraplegics (PP) and tetraplegics (TP). Settings Rehabilitation facility in Sweden. Methods The participants were 26 (19 men, 7 women) with C5–C8, AIS A and B (TP) and 37 (27 men, 10 women) with T7–T12 (PP), AIS A and B. Wrist-worn accelerometer recordings (Actigraph GT3X+) were taken during seven standardized activities. Oxygen consumption was measured, as well as at-rest and peak effort, with indirect calorimetry. Accelerometer cut-points for absolute and relative intensities were defined using ROC-curve analyses. Results The ROC-curve analyses for accelerometer cut-points revealed good-to-excellent accuracy (AUC >0.8), defining cut-points for absolute intensity (2, 3, 4, 5, 6, 7 METs for PP and 2 to 6 METs for TP) and relative intensity (30, 40, 50, 60, 70, and 80% for PP and 40–80% for TP). The cut-points for moderate-to-vigorous physical activity was defined as ≥9515 vector magnitude counts per minute (VMC) for PP and ≥4887 VMC/min for TP. Conclusion This study presents cut-points for wrist-worn accelerometers in both PP and TP, which could be used in clinical practice to describe physical activity patterns and time spent at different intensity levels.