Responsiveness of postural performance measures following balance rehabilitation in multiple sclerosis patients

Abstract

Introduction: Evaluating responsiveness has an important role in design and interpretation of the interventional studies. The aim was to estimate the responsiveness and minimally important difference (MID) of postural performance measures following balance rehabilitation in patients with Multiple Sclerosis (MS, n = 38). Methods: Postural measures were evaluated at baseline and after 4 weeks intervention. Laboratory-based measures were center of pressure parameters. Clinically-based measures were Activities-specific Balance Confidence (ABC), Berg Balance Scale (BBS), Functional Gait Assessment (FGA); and walking measures 2 Minute Walk (2 MW), 10 Meter Timed Walk (10 MTW) and Timed Up and Go (TUG) performed under single and dual-task conditions. To evaluate responsiveness, we calculated the Receiver Operating Characteristics (ROC) and the Area Under the ROC Curve (AUC). The optimal values for the MID were the cutoffs corresponding to the upper left corner of the ROC curve. Results: The AUCs for mean and standard deviation of sway velocity were above the cutoff of 0.50 in most conditions. For the clinically-based measures, the highest AUCs were found for the ABC, and cognitive-2MW, followed by the BBS and 10 MTW. Conclusions: In this preliminary study, the most appropriate postural performance measures and the MID values for detecting meaningful changes in MS undergoing balance rehabilitation have been provided.