Test-retest reliability in performance of persons with hemiparesis tracking by means of compatibly displayed myoelectric feedback derived from upper limb muscles.

The aim of this study was to assess whether persons with hemiparesis will yield statistically reliable test-retest tracking performance on a procedure using limb-generated, compatibly displayed, myoelectric video feedback. A convenience sample of 50 inpatients and outpatients with upper extremity involvement of at least six months were recruited. 30 had hemiparesis and had both upper extremities tested. 20 had hemiplegia and had the nonparetic upper extremity tested. Tracking skill was measured as mean tracking error. Repeated measures analysis of variance yielded statistically significant effects of main factors: Response mode, Cursor Feedback, and Session. Extremity tested was not significant. Performance with involved limb, uninvolved limb, isometric mode, and isotonic mode all yielded positive rest-retest correlations. The reliable range of tracking error obtained from subjects with hemiparesis performing a task requiring modulation of compatibly displayed myoelectric output supports the therapeutic rationale for employing compatibly displayed video feedback in the rehabilitation of motor control.