ABSTRACT
OBJECTIVE: The purpose of our study was to assess the differences between the Balance Error Scoring System (BESS), modified BESS (mBESS), and both measures with the double-leg (DL) stances removed [BESS-revised (BESS-R) and mBESS-revised (mBESS-R)] among healthy and concussed collegiate student-athletes. DESIGN: Retrospective, repeated-measures cohort study. SETTING: Clinical. PATIENTS OR OTHER PARTICIPANTS: Healthy and concussed collegiate athletes (baseline n = 622, postinjury n = 41) from 12 National Collegiate Athletic Association Division I sports cheer and dance at a single university. INTERVENTION OR INDEPENDENT VARIABLES: Balance Error Scoring System, mBESS, BESS-R, and mBESS-R test versions from baseline and postinjury testing. MAIN OUTCOME MEASURES: The mBESS and BESS and their revised versions with DL stances removed (mBESS-R and BESS-R) scores were compared at baseline. Baseline and postinjury scores for all 4 BESS variations and the 6 BESS conditions were compared for those who sustained a concussion. RESULTS: The BESS and BESS-R were statistically different at baseline for the entire sample (99.6% confidence interval 0.32, 0.38, P > 0.0001). None of the other comparisons were significantly different ( P > 0.004). CONCLUSION: Although our results do suggest statistically significant differences between the BESS and BESS-R test versions, they do not represent clinically meaningful differences. The greatest mean difference between all test versions was <1 error; therefore, these BESS versions may not be specific enough to identify balance deficits at baseline or postinjury. Elimination of the time intensive DL measures in the revised BESS variations may be a more clinically practical alternative.