Balance measured by the sway balance smart-device application does not discriminate between older persons with and without a fall history.

Abstract

Clinical functional assessments of balance often lack specificity and sensitivity in discriminating and predicting falls among community-dwelling older adults. We determined the feasibility of using a smart-device application measuring balance to discriminate fall status among older adults. We also evaluated differences between smart-device balance measurements when secured with or without a harness. A cross-sectional study design to determine the ability of the Sway Balance smart-device application (SWAY) to discriminate older adults based on fall history. The Berg Balance Scale (BBS) and Activities-Specific Balance Confidence Scale (ABC) were used as comparative, clinically based assessments. Community-dwelling older adults with (n=25) and without (n=32) a history of fall(s) participated. Multivariate analysis of variance was used to determine differences among assessments based on fall history. Logistic regression models determined the ability of each assessment to discriminate fall history. Older adults with and without a history of falls were not significantly different on SWAY (P=0.92) but were different on BBS (P=0.01), and ABC (P<0.001). Similarly, SWAY did not discriminate fall history (P=0.92), while BBS and ABC both discriminated fall history (P<0.01). Paired t tests between SWAY scores with and without a harness indicated no differences (P≥0.05). Among the older adults studied, the BBS and ABC measures discriminated groups defined by fall history, while the SWAY smart-device balance application did not. Modifications to the application may improve the discriminating ability of the measure in the recognition of fall status in older adults.