Construct validity of a modified bathroom scale that can measure balance in elderly people

To investigate the construct validity of a bathroom scale measuring balance in elderly people. Cross-sectional study. Participants were recruited via nursing homes and an organization that provides exercise classes for community-dwelling elderly people. Nursing home patients were compared with active community-dwelling elderly people. Eligibility criteria for both groups were: aged 65 years or older and being able to step onto a bathroom scale independently. The balance measurement of the bathroom scale was compared with the following three clinical balance measurements: Performance Oriented Mobility Assessment (POMA), Timed Up and Go (TUG), and Four Test Balance Scale (FTBS). An independent samples t-test was performed to determine whether nursing home patients scored lower on these four balance tests compared with community-dwelling elderly people. Correlations were calculated between the bathroom scale balance scores and those of the clinical balance tests for nursing home patients and community-dwelling elderly people separately. Forty-seven nursing home patients with a mean age of 81 years (SD 6.40) and 54 community-dwelling elderly people with a mean age of 76 years (SD 5.06) participated in the study. The results showed that nursing home patients had significantly lower scores on all four balance tests compared with community-dwelling elderly people. Correlations between the bathroom scale scores and the POMA, TUG, and FTBS in nursing home patients were all significant: .49, -.60, and .63, respectively. These correlations were not significant in active community-dwelling elderly people, -.04, -.42, and .33, respectively. Linear regression analyses showed that the correlations for the bathroom scale and POMA, bathroom scale and TUG, and bathroom scale and FTBS did not differ statistically between nursing home patients and community-dwelling elderly people. These results suggest that the modified bathroom scale is useful for measuring balance in elderly people. However, the added value of this assessment method for clinical practice remains to be demonstrated.