RESUMEN
BACKGROUND: Trunk control (TC) constitutes one of the main objectives in the rehabilitation of people with a spinal cord injury (SCI), but there are few clinically validated tests to assess it. Accelerometers have been proposed as sensitive and suitable procedures to assess TC. OBJECTIVE: To evaluate test-retest reliability, construct and criterion validity of accelerometer parameters to assess TC in people with SCI. METHODS: A cross-sectional study of simultaneous application of Clinical Trunk Control Test (CTCT) and accelerometer evaluation was conducted. Accelerometers were placed on the trunks of 27 people with SCI and 15 people without SCI. Reliability was assessed by three repeated measures in random order of selected static and dynamic TC tasks. Acceleration on three axes was analyzed using five metrics. Criterion validity was assessed by analyzing correlation of acceleration to CTCT scores. Construct validity was assessed by analyzing capacity of inertial measurement units (IMU) to differentiate individual's characteristics, ASIA Impairment scale, gait capacity, level of TC, and neurological level of injury. RESULTS: Reliable IMU data were obtained in people with SCI and without SCI, of all accelerometer axes, metrics, and tested items of the CTCT. Reliability of acceleration decreases with the increasing demand for TC tasks. Ten acceleration parameters showed construct and criterion validity. CONCLUSION: Accelerometer parameters are reliable, valid, and sensitive to evaluate TC in people with SCI. SIGNIFICANCE: A set of IMU parameters were validated as reliable and valid measures to evaluate TC, which could be useful for the assessment of progression of people with SCI and clinical interventions.