Abstract
Purpose: This study was undertaken to validate each item of the Berg Balance Scale (BBS) for elderly people and to simplify the tool.
Methods: The BBS was used for assessment of 88 elderly people (average age 79.5 ± 6.4 years old) who used geriatric healthcare facilities. The BBS items' respective difficulties and their goodness of fit to the Rasch model were analyzed using values measured using Rasch analysis.
Results: Rasch analysis results showed that the order of difficulty of BBS items resembles their order of clinical difficulty. Goodness of fit analysis of the Rasch model showed incompatibility of 8 of 14 BBS items. Consequently, the items were reduced to 5-8 from 14 by selecting BBS items with careful consideration of their difficulty and fit. The eight-item version of the modified BBS showed the best reliability and validity.
Conclusion: The modified eight-item BBS, despite elimination of about half of the items of the longer version, was shown to be appropriate for assessment of motor tasks to evaluate physical functions of independently ambulatory elderly people.
