The ability of a scale to detect clinical relevant changes over time, i.e., its "responsiveness," may help clinicians to choose among valid and reliable measures. Therefore, we investigated the responsiveness' rank ordering (best to worse) of six selected valid and reliable scales, namely the Medical Research Council (MRC)-sumscore, sensory-sumscore, grip-strength (Vigorimeter), nine-hole peg, ten-meters walking, and a disability-sumscore, in immune-mediated polyneuropathies. Patients with newly diagnosed Guillain-Barré syndrome (n = 7) or chronic inflammatory demyelinating polyneuropathy (n = 13) were examined over 52 weeks. Responsiveness of each scale was measured using different methods (effect-size, standardized response mean score, Wilcoxon matched-pairs signed-rank, and a newly devised Schmitz's distribution-free responsiveness score), and the obtained scores in each method were plotted against the follow-up period, thus allowing area-under-the-curve calculations (higher area-under-the-curve indicating better responsiveness). Also, longitudinal correlations were performed between the scales' values and patients' own clinical judgments (deteriorated, unchanged, improved) (higher correlation = better responsiveness). A consistent rank ordering was observed in each technique with the disability-sumscore, MRC-sumscore, and Vigorimeter being among the best responsive scales. Hence, the primary use of these measures is suggested in studies of immune-mediated polyneuropathies.
Copyright 2003 Wiley Periodicals, Inc.