The Effectiveness of the Lower Extremity Fugl-Meyer Motor Assessment to Precisely Measure Motor Recovery from the Acute to Chronic Stage Post-Stroke: A Rasch Analysis

Research Objective To investigate how the item-difficulty hierarchy effects measurement precision and validity of the Lower Extremity Fugl-Meyer Motor Assessment (FMMA-LE), a measure of motor impairment in stroke patients. Design A retrospective analysis of pooled datasets from prospective observational studies that tracked stroke motor recovery in a cohort with first-ischemic stroke at 2-5 days and 90±10 days. Setting Academic stroke center. Participants 134 subjects (57.8±13.7 y.o., male=52%) with FMMA-LE at 2-5 and 90±10 days. Interventions Not applicable. Main Outcome Measures Partial Credit Model (PCM), one of the Rasch Measurement Models for multi-category responses, was used to construct Wright Maps to assess the item-difficulty hierarchy and measurement precision of the FMMA-LE. Results Standing Knee Flexion was identified as the most difficult item at both timepoints. The least difficult item evolved from Synergistic Hip Flexion to Synergistic Knee Extension. The results from the PCM analyses demonstrated an increase in average person ability (motor recovery) in subjects from 2-5 to 90±10 days post-stroke. Six misfitting items were identified suggesting that they do not work well with other items of the FMMA-LE (reflexes and coordination). The Wright Maps at 2-5 and 90±10 days revealed item gaps in the measurement of motor impairment encompassing 59% (n=80) and 66% (n=88) of the subjects at the 2-5 and 90±10 days stage, respectively. Therefore, motor impairment may not be precisely measured using all of the items on the FMMA-LE. Conclusions Our findings suggest that the FMMA-LE is a clinically valid but likely not a precise measure of motor impairment at 2-5 and 90±10 days post-stroke. Author Disclosure Dr. Feng received AHA SDG grant (14SDG1829003) for this research work.