Journal of Rehabilitation Medicine 51-9

J Rehabil Med 2019; 51: 652–659 ORIGINAL REPORT INTRA- AND INTER-RATER RELIABILITY OF FUGL-MEYER ASSESSMENT OF UPPER EXTREMITY IN STROKE Edgar D. HERNÁNDEZ, PT, MSc 1 , Claudia P. GALEANO, PT 2 , Nubia E. BARBOSA, PT 2 , Sandra M. FORERO, PT 2 , Åsa NORDIN, PT, MSc 3 , Katharina S. SUNNERHAGEN , MD, PhD 3 and Margit ALT MURPHY , PT, PhD 3 From the 1 Universidad Nacional de Colombia, Departamento del Movimiento Corporal Humano, 2 Central Military Hospital of Colombia, Bogota, Colombia, 3 Institute of Neuroscience and Physiology, Clinical Neuroscience, Rehabilitation Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden Objective: The Fugl-Meyer Assessment of Upper Ex- tremity (FMA-UE) is recommended for evaluation of sensorimotor impairment post stroke, but the item- level reliability of the scale is unknown. This study aims to determine intra- and inter-rater reliability of the FMA-UE at item-, subscale- and total score level in patients with early subacute stroke. Design: Intra/inter-rater reliability. Subjects: Sixty consecutively included patients with stroke (mean age 65.9 years) admitted to Central Military Hospital of Colombia, Bogota. Methods: Two physiotherapists scored FMA-UE inde- pendently on 2 consecutive days within 10 days post stroke. A rank-based statistical method for paired ordinal data was used to assess the level of agree ment, systematic and random disagreements. Results: Systematic disagreements either in posi- tion or concentration were detected in 4 items of the shoulder section. The item level intra- and inter-rater agreement was high (79–100%). The 70% agree ment was also reached for the subscales and the to- tal score when 1–3-point difference was accepted. Conclusion: The FMA-UE is reliable both within and between raters in patients with stroke in the early subacute phase. A wider international use of FMA- UE will allow comparison of stroke recovery between regions and countries and thereby potentially impro- ve the quality of care and rehabilitation in persons with stroke worldwide. Key words: reliability; psychometrics; non-parametric statis- tics; stroke rehabilitation; upper extremity; motor activity; physical therapy specialty; reproducibility of results. Accepted Aug 9, 2019; Epub ahead of print Aug 26, 2019 J Rehabil Med 2019; 51: 652–659 Correspondence address: Margit Alt Murphy, Per Dubbsgatan 14, plan 3, 413 45 Gothenburg, Sweden. E-mail: margit.alt-murphy@neuro. gu.se H emiparesis is one of the most frequent sequelae of stroke, causing significant disability (1, 2). Motor deficits of the upper limb are common and af- fect approximately 50–70% of patients admitted to the hospital in the acute, subacute phase (3–5) and 40% in the chronic phase (6, 7). Impaired function of the upper limb makes it difficult to carry out basic movements and daily tasks in an efficient way (8). Upper limb LAY ABSTRACT The Fugl-Meyer Assessment of Upper Extremity (FMA- UE) is one of the most used and recommended assess- ment scales of sensorimotor function in stroke. This study investigated the reliability of the scale when dif- ferent therapists assessed the patient’s performance at the same test session and when the assessment was performed by the same therapist but on 2 different oc- casions. Sixty individuals with stroke at the Central Mi- litary Hospital of Colombia were included. The results showed that the agreement in each individual move- ment (FMA-UE includes 33 movements/items) was 79% or above. Disagreements in scorings between raters were noted for 4 single items. These disagreements were probably caused by the spontaneous recovery that occurred in the early subacute phase after stroke. The item, subscale and total score level reliabilities were high and the scale can be recommended for use in ge- neral, including in Spanish-speaking countries. It is im- portant, however, that standardized testing procedures are followed. sensorimotor impairment after stroke is commonly assessed by using the Fugl-Meyer Assessment for Upper Extremity (FMA-UE). It is considered as gold standard and is the only impairment level measure recommended for stroke trials (9). The FMA-UE is well-established internationally, clinically feasible and shows excellent reliability, validity and responsiveness (10, 11). FMA-UE is widely used to determine the severity of stroke and to quantify recovery (12). Both the intra- and inter-rater reliability of the FMA-UE, by means of the intraclass correlation coefficient (ICC), have demonstrated to be excellent, with reported values above 0.90, both for the total and subscale level in the chronic and subacute phase (13–17). The ICC and other correlation methods are valid for measuring the strength of association, but are limited for evaluation of agreement between assessments. Clinical scales, such as the FMA-UE, produce ordinal data, in which the ordered categories represent only rank order and not a numerical value (18). Previous reliability studies of FMA-UE have predominantly used statistical analyses appropriate to continuous data rather than non-parametric statistical methods and evaluated the reliability of the FMA-UE This is an open access article under the CC BY-NC license. www.medicaljournals.se/jrm doi: 10.2340/16501977-2590 Journal Compilation © 2019 Foundation of Rehabilitation Information. ISSN 1650-1977

Journal of Rehabilitation Medicine 51-9 | Page 28