Estimating the minimal clinically important difference for balance and gait outcome measures in individuals with vestibular disorders

BACKGROUND: Vestibular Rehabilitation Therapists (VRT) utilize outcome measures to quantify gait and balance abilities in individuals with vestibular disorders (IVD). The minimal clinically important difference (MCID) in gait and balance outcome measures for IVD is unknown. OBJECTIVE: The purpose of this study is to estimate the MCID of the Activities-specific Balance Confidence Scale (ABC), Functional Gait Assessment (FGA), and Gait Speed (GS) using distribution and anchor-based methods relative to the Dizziness Handicap Inventory (DHI) in IVD. METHODS: Data were collected using a retrospective chart review from two outpatient Vestibular Rehabilitation (VR) clinics. Data included demographic characteristics, diagnosis, VR course, and pre and post outcome measures including DHI, ABC, FGA, and GS. The DHI was used to classify subjects as "responders" or "non-responders" in order to calculate MCID values. RESULTS: The total number of subjects analyzed for each outcome measure was 222 for the ABC, 220 for FGA, and 237 for GS. Subjects made statistically significant improvements in ABC, DHI, FGA, and GS (p < 0.001) from pre to post VR. The MCID calculated for ABC, FGA, and GS using the anchor-based approach was 18.1%, 4 points, and 0.09 m/s respectively. The MCIDs calculated using distribution-based approach for the ABC ranged between 7.5-23.5%, FGA ranged between 1.31-4.15 points, and GS ranged between 0.07 m/s-0.22 m/s. CONCLUSIONS: The anchor-based calculations of the MCID of 18.1%, 4 points, and 0.09 m/s for ABC, FGA, and GS respectively for IVD should be used over distribution-based calculations. This is due to strength of DHI as the anchor and statistical analysis. VRT and researches can use these values to indicate meaningful changes in gait and balance function in IVD.