OBJECTIVE: To examine whether selection of the non-paretic or paretic leg as the weight-bearing leg in items 13 'standing unsupported one foot in front' and item 14 'standing on one leg' of BBS will influence the item scores, and thus the total score.
DESIGN: Cross sectional study. SETTING: University-based rehabilitation laboratory. PARTICIPANTS: Sixty-three community dwelling people with chronic stroke (aged ≥ 50) INTERVENTIONS: None MAIN OUTCOME MEASURE: Berg Balance Scale.
RESULTS: The four BBS total scores ranged from 48.4 to 50.7. The total score was significantly lower when a participant was asked to step forward with the non-paretic leg in item 13, and stand on the paretic leg in item 14. Fewer participants received a maximum score with the BBS1 formulation than the others. In addition, the correlations with walking speed and ABC scores were greatest with the BBS1 score.
CONCLUSIONS: Our findings suggest that BBS1 was the most challenging formulation for our participants; this might serve to minimize the BBS's ceiling effect. These findings provide a rationale for amending the BBS administration guidelines with the BBS1 formulation.