Comparison of Reliability, Validity, and Responsiveness of the Mini-BESTest and Berg Balance Scale in Patients With Balance Disorders

TLDR: The Mini-BESTest appears to have a lower ceiling effect, slightly higher reliability levels, and greater accuracy in classifying individual patients who show significant improvement in balance function.

Abstract: Background Recently, a new tool for assessing dynamic balance impairments has been presented: the 14-item Mini-BESTest. Objective Aim of this study was to compare the psychometric performance of Mini-BESTest vs. Berg Balance Scale (BBS). Methods Ninety-three subjects (mean age 66.2; 53 females) with balance deficits were recruited. Interrater (3 raters) and test-retest (1-3 days) reliability were calculated using intraclass correlation coefficients (ICCs). Responsiveness and minimal important change (MIC) were assessed (after 10 sessions of physical therapy) using both distribution-based and anchor-based methods (external criterion: a 15-point Global Rating of Change [GRC] scale). Results At baseline, neither floor nor ceiling effects were found in either scale. After treatment, the maximum score was found in 12 patients (12.9%) with BBS and in 2 (2.1%) with Mini-BESTest. Test-retest reliability for total scores was significantly higher in Mini-BESTest (ICC=0.96) than in BBS (ICC=0.92), whereas interrater reliability was similar (ICC=0.98 vs. 0.97). The standard error of measurement (SEM) was 1.26 and minimum detectable change at the 95% confidence level (MDC95) 3.5 points for Mini-BESTest, whereas SEM was 2.18 and MDC95 6.2 points for BBS. In receiver operating characteristic (ROC) curves, the area under the curve was 0.92 for Mini-BESTest and 0.91 for BBS. The best Minimal Important Change (MIC) was 4 points for Mini-BESTest and 7 for BBS. After treatment, 38 patients at Mini-BESTest vs. only 23 at BBS (out of the 40 subjects who had a GRC ≥3.5) showed a score change equal or greater than MICs. Conclusions The two scales behave similarly, but the Mini-BESTest scale seems to have a lesser ceiling effect, slightly higher reliability levels, and greater accuracy in classifying individual patients who show significant improvement in balance function.