Outcome measures for multiple sclerosis clinical trials: relative measurement precision of the Expanded Disability Status Scale and Multiple Sclerosis Functional Composite

We compared the relative measurement precision (RMP) of the Expanded Disability Status Scale (EDSS) and the Multiple Sclerosis Functional Composite (MSFC) for discriminating between groups of patients known to differ in their extent of multiple sclerosis (MS). A total of 133 patients were rated with the EDSS and MSFC and had magnetic resonance imaging (MRI) scans. Patients were grouped on the basis of MRI appearances (T1- and T2-weighted lesion loads, parenchymal and ventricular fractions - TILL, T2LL, PF, VF, respectively) and RMP was determined using the method of group differences. For each MRI parameter, the total sample was arranged in ascending order of magnitude and divided into two, three, four and five similar sized groups. For each division (two, three, four or five groups), EDSS and MSFC scores for the groups were compared using parametric (paired samples t-tests, one-way ANOVA) and nonparametric (Wilcoxon's rank-sum test, Kruskal-Wallis analysis of variance) statistical methods and RMP was estimated. The EDSS and MSFC were correlated substantially (r= -0.64). Relative to the MSFC, the EDSS had inferior measurement precision regardless of the number of groups into which the total sample was divided, or the statistical method. However, the RMP of the EDSS compared with the MSFC varied from 2% to 86%. Results suggest the MSCF is better than the EDSS for detecting differences between groups of patients, defined by these MRI markers of MS. However, the finding that both scales correlated weakly with MRI markers, indicated that they are limited as predictors of MS pathology as defined by MRI. An explanation for this well-established clinical-MRI paradox is that rating scales and MRI measure fundamentally different manifestations of MS.