Inter‐ and intrarater scoring agreement using grades 1.0 to 3.5 of the Kurtzke Expanded Disability Status Scale (EDSS)

Abstract
We determined inter- and intrarater Kurtzke Expanded Disability Status Scale (EDSS) scoring agreement for four trained examining physicians who evaluated 10 clinically stable multiple sclerosis patients. These patients had previously been determined to have EDSS scores of 1.0 to 3.5 and were scheduled to participate in a funded clinical trial of intramuscular recombinant interferon-beta. Intrarater reliability was greater than interrater reliability for scoring the EDSS and all of its component functional systems scores (FSS). Specifically, individual examiners were able to reproduce three serial examination scores on the same patient on the same day (intrarater agreement) within 1.0 EDSS or 2.0 individual FSS points. Reproducible scoring across examiners (interrater agreement), however, could only be accomplished within 1.5 EDSS or 3.0 individual FSS points. Additionally, the interrater scoring variability in our patients is greater than that previously reported for patients with higher EDSS scores. We conclude that clinical trials that employ the EDSS as an outcome measure of treatment efficacy should include inter- and intrarater agreement data for all examining physicians. Most importantly, studies using a single examining physician to evaluate individual patients throughout the course of a clinical trial will require less change in the EDSS to reliably measure disease activity than will studies using more than one examining physician to evaluate individual patients throughout the trial.