The Original and Short Versions of the Activities-Specific Balance Confidence (ABC) Scale in Community-Dwelling Older Adults: Clinical Implications Based on Scale Agreement, Internal Consistencies, and Associations With Self-Rated Health

Abstract

Background and purpose: Balance confidence assessment in older adults has implications for falls prevention and quality of life. It remains unclear whether the original Activities-specific Balance Confidence Scale (ABC-16) or the shortened 6-item scale (ABC-6) is recommended clinically. This study examined ABC-16 and ABC-6 association and agreement, internal consistencies, and relationships with self-rated health (SRH) in community-dwelling older adults.

Methods: Secondary analysis of an existing dataset (N = 77) was performed. Scale association and agreement were assessed with a Spearman ρ correlation (rs), intraclass correlation coefficient, 95% limits of agreement (LoA), and Bland-Altman plot. Cronbach α values were calculated to determine internal consistencies. Separate multiple linear regression models with SRH as the outcome and ABC-6 and ABC-16 scores as primary predictors were estimated and subsequently used to conduct Hotelling t test.

Results and discussion: Participants were primarily female (80.5%) with a median age of 68 years living in the metro Detroit area. The ABC-6 and ABC-16 were closely associated [rs = 0.97, P < .001; intraclass correlation coefficient (2,1) = 0.80] but demonstrated discrepancy (95% LoA range of -3.9 to +18.2; mean difference = 7.2 points in the direction of the ABC-16). Cronbach α values were 0.95 (ABC-16) and 0.89 (ABC-6). Regression model 1 (ABC-6 = primary predictor) explained more of the variance (R2 = 0.36) in SRH compared with model 2 (ABC-16 = primary predictor; R2 = 0.29). Hotelling t test [t(74) = 2.4, P = .008] found that the predicted values from the ABC-6 model were significantly more highly correlated with SRH than those from the ABC-16 model.

Conclusions: Despite a high correlation between the ABC-16 and ABC-6, the 2 scales showed limited agreement and should not be considered interchangeable. Given that the ABC-16 takes longer to administer, does not relate to SRH as strongly, and could have redundant items, the ABC-6 may be preferable to the ABC-16 for balance confidence assessment in older adults living in cold weather, urban, or well-resourced areas.