Year: 2015
Type: article
Study design: Observational Study
Abstract: Falls among the elderly represent a major problem because of the high incident, the involvement of many different risk factors, and the consequential post-fall morbidity and mortality.Many balance scales have been created to help identify the causes and outcomes of elderly falls, and in turn, help health care providers better assess fall risk.The Berg Balance Scale (BBS), 1,2 Fullerton Advanced Balance (FAB) Scale, 3 Timed Upand-Go (TUG), 4,5 and the Activity-specific Balance Confidence (ABC) Scale 6 have been developed to measure balance and functional mobility among older people, as well as quantify confidence and fear of falling in different balance conditions.It is important to identify which tests can discriminate fallers from non-fallers, retrospectively, 3,7-9 and certainly there is more data to be discovered on the prospective abilities of the same tests to predict future falls.Few have considered age as a covariate when evaluating the predictive validity of these tests.Therefore, the purpose of this study was to identify which test(s) best discriminate fallers from non-fallers among 60-79 years and 80-100 years of age.Forty-six community-dwelling heathy older adults (mean age 79.1± 10.2, range 60-100 years, 14 male, 32 female) were recruited from three senior centers and two independent living facilities.Participants filled out an information log which included iii history of falls within the previous six months, prescription medication number, arthritis and joint replacements in either leg, exercise regularity, and assistive device use.Participants completed the ABC scale, and then in random order the TUG, BBS and FAB scale.Logistic regression was used to determine the discriminative power of the ABC, TUG, BBS, and FAB on falls within the six previous months.The models had a binary outcome of self-report history of ≥1 falls vs. no falls.Receiver operating characteristics (ROC) curves and subsequently the area under the curve (AUC) were calculated.Specificity and Sensitivity were calculated for each of the tests for all subjects and by age group (60-79 years and 80-100 years).Significance was set a priori at alpha= .05.Results showed the discriminative abilities for at least one fall versus no falls to be poor.Corresponding area under the curve values were as follows: ABC (0.366), TUG (0.554), BBS (0.426), and FAB (0.437).Although the discriminative abilities were poor, we did see a correlation between all tests and age.As age increased, an appropriate directional change was observed for the ABC (-0.49),TUG (0.41), BBS (-0.58), and FAB (-0.70).Additionally, most models had a greater specificity than sensitivity; they were better at identifying non-fallers than fallers.The present results suggest that the ABC, TUG, BBS, and FAB may be used to assess functional capacity but not necessarily fall risk.They should be used with other fall risk assessment tools such as home safety, polypharmacy risk, vision and hearing, and assistive device evaluations.vii
Author: Jessica Dicke
Institution: The Ohio State University
Language: English
Field: Health Professions
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W2504395828