Objective: To investigate the inter-rater reliability and validity of the Swedish version of the Modified Elderly Mobility Scale (Swe M-EMS).
Design: A correlation study.
Setting: A stroke unit which provides acute care and rehabilitation for all stroke admissions in Northern Alvsborg County Hospital, Sweden.
Subjects: Thirty consecutive patients with an acute first-time stroke. Seventeen were women and 13 were men, with a median age of 79 years (41-92) and with a median hospital stay of 21 days (8-35).
Intervention: The patients were assessed twice with the Swe M-EMS on admission and at discharge, before and after midday, by two independent physiotherapists. They were assessed by an independent physiotherapist using the three evaluation instruments on admission, once a week during their hospital stay and at discharge.
Main outcome measures: The primary evaluation instrument was the Swe M-EMS. The two other evaluation instruments were the Modified Motor Assessment Scale of Uppsala Akademiska Sjukhus-95 (M-MAS UAS-95) and the Berg Balance Scale (BBS).
Results: The Swe M-EMS was a reliable instrument (ICC 0.98-0.99). It correlated well with both the M-MAS UAS-95 (rs=0.69-0.88) and the BBS (rs=0.86-0.94). The Swe M-EMS was sensitive to change over time, but the patients obtained the maximum score quickly.
Conclusion: The instrument shows a high inter-rater reliability. The Swe M-EMS correlates well with both the M-MAS UAS-95 and the BBS and is sensitive to change. However, it is not sensitive enough to use as a single instrument in evaluating the improvement of a patient with acute stroke.