Interrater variability with the Expanded Disability Status Scale (EDSS) and Functional Systems (FS) in a multiple sclerosis clinical trial

We describe the interrater variability in the Expanded Disability Status Scale (EDSS) and Functional Systems (FS) in a multiple sclerosis clinical trial. Two physicians blinded to their previous assessments and to each other's scores consecutively examined 168 patients (545 paired examinations). Perfect agreement on the assignment of the disability scores ranged from 48% (cerebellar functional group) to 69% (EDSS and pyramidal functional group). Only 31% to 62% of this agreement occurred independently of that expected by chance (kappa). With the exception of the cerebellar and sensory functional groups, agreement within 1 step occurred in at least 92% of cases. These findings suggest that differences of a single step on these scales may not reflect an important functional change. We recommend that at least a 2-step change (1.0 point on the EDSS and 2 points on the FS) is needed to be confident of an important change in the degree of disability or response to treatment in this disease.