Reliability parameters of the Timed 25-Foot-Walk (T25FW) in patients with Multiple Sclerosis: Lower walking speed is associated with greater Smallest Detectable Change

Abstract

Background: Reliability parameters of clinical measures should be sufficient in order to adequately monitor disease course and evaluate treatment in patients with Multiple Sclerosis (MS). The aim of this study was to assess the reliability of the Timed 25-Foot-Walk (T25FW) in patients with MS. Methods: MS patients performed the T25FW twice with approximately one year in between. After the second measurement, they answered an anchor question using a three-point Likert scale. Taking the non-normal distribution of the T25FW into account, intraclass correlation coefficient (ICCagreement), standard error of the mean (SEMagreement) and smallest detectable change (SDC) were computed. Results: 118 MS patients (76.3 % females, mean age 48.2 years) were included. Of these patients, 73 reported no change on the anchor question. They completed the T25FW at baseline in 4.7s (IQR 4.3-5.4s, n = 72) and at follow-up in 4.9s (IQR 4.3-5.9s, n = 73). The ICCagreement was 0.895, the SEMagreement was 0.037. The SDC was higher for higher mean T25FW and can be computed for each mean T25FW by 0.23*mean T25FW. Conclusion: Overall, in patients with MS, the T25FW has a sufficient reliability as measured with the ICC, however the SDC of the T25FW increased when patients perform worse on the T25FW (i.e. have a lower walking speed). Because this test is often used in MS patients with limited walking ability, these findings are important to keep in mind when interpreting the re-test scores of the T25FW.