Responsiveness and predictive value of EDSS and MSFC in primary progressive MS

Abstract

We studied the responsiveness and predictive value of two widely used clinical outcome measures that document multiple sclerosis (MS) disease progression-the Expanded Disability Status Scale (EDSS) and the Multiple Sclerosis Functional Composite (MSFC)-in patients with primary progressive (PP) MS. Disease course in PPMS shows less fluctuation than in relapsing remitting (RR) MS. In a group of 161 patients with PPMS, EDSS and MSFC were performed at three timepoints. To assess responsiveness, mean change scores and variances were plotted against baseline scores and effect sizes were calculated. Predictive value was determined by calculating sensitivity, specificity, and likelihood ratios (LRs) of 1-year changes to predict changes over 2 years. Furthermore, multivariate logistic regression models were used to assess the predictive value of short-term worsening on EDSS and MSFC. Responsiveness of both EDSS and MSFC was shown to be limited and mean changes were highly dependent on the baseline scores. Effect sizes for EDSS and MSFC were small and inconclusive (0.239 and 0.161). The predictive value of a short-term worsening (baseline to year 1) to predict worsening in the long term (baseline to year 2) was expressed for EDSS by a sensitivity of 0.55 and a LR+ of 8.64. For MSFC, sensitivity was 0.68 and LR+ was 3.14. However, short-term worsening was a poor predictor of subsequent worsening (year 1 to year 2) for EDSS (LR+ 1.06) and this relationship was actually inverse for MSFC (LR+ 0.61). In this study over a period of 2 years in primary progressive multiple sclerosis, the Multiple Sclerosis Functional Composite (MSFC) was less responsive than the Expanded Disability Status Scale (EDSS). The predictive value of neither EDSS nor MSFC was very powerful.