Validity, Reliability, and Measurement Error of the Remote Fugl-Meyer Assessment by Videoconferencing: Tele-FMA

Abstract

Abstract Objective The objective of this study was to investigate the validity, reliability, and measurement error of the Fugl-Meyer Assessment (FMA) when it was remotely administered by videoconferencing (Tele-FMA) and to describe barriers to remote administration of the FMA. Methods Forty-five participants who had strokes and had a smartphone or laptop computer with a camera and internet access were included. An in-person assessment was compared with a remote assessment in 11 participants, and 34 participants completed only the remote assessment. Rater 1 (R1) remotely administered, recorded, and scored the items of the FMA, after which the recording was forwarded to be scored by Rater 2. At least 7 days later, R1 rated the videorecording of the remote assessment a second time for the evaluation of intrarater reliability. In-person assessment was completed by R1 at the participant’s home. Criterion validity was analyzed using the Bland–Altman limits of agreement, and convergent validity was analyzed using Spearman correlation coefficient. The intrarater and interrater reliability was analyzed using the intraclass correlation coefficient, and individual items were analyzed using the weighted kappa. The standard error of measurement and minimal detectable change were calculated to evaluate the measurement error. Results Bland–Altman plots showed adequate agreement of in-person FMA and tele-FMA. A moderate positive correlation was found between Tele-FMA lower extremity (LE) scores and step test results, and a strong positive correlation was found between Tele-FMA-upper extremity (UE) and Stroke Impact Scale hand function domain. Significant and excellent (0.96 ≤ ICC ≤ 0.99) interrater and intrarater reliabilities of the Tele-FMA, Tele-FMA-UE, and tele-FMA-LE were found. Regarding the individual items, most showed excellent reliability (weighted kappa > 0.70). The standard error of measurement for both reliabilities was small (≤3.1 points). The minimal detectable change with 95% CI for both the Tele-FMA and Tele-FMA-UE was 2.5 points, whereas it was 1.3 points for the Tele-FMA-LE. Conclusion Tele-FMA has excellent intrarater and interrater reliability and should be considered as a valid measurement. Impact The FMA is widely used in clinical practice. However, the measurement properties of the remote version applied by videoconferencing were unknown. This study’s results demonstrate the validity and reliability of the Tele-FMA for assessing poststroke motor impairment remotely via videoconferencing. The Tele-FMA may be used to implement telerehabilitation in clinical practice.