At T1, 52 participants were recruited through flyers from 2 community centers and 2 independent living apartments. For the comparison group, 17 individuals from one of the community centers and from an independent living apartment were recruited by flyers and 41 individuals were recruited from a registry for older adults who were interested in participating in studies. Ten participants were excluded because they were too young (n = 2), never started the program (n = 5), or had missing data at Time 1 (n = 3). Consequently, 45 participants in the MOB group and 55 in the comparison group were included in the study. At T2, 10 participants in the MOB group and 15 in the comparison group were lost to follow-up. In the MOB group, 11 participants were from community center 1; 7 were from community center 2; 9 were from the independent living apartment 1; and 8 were from the independent living apartment 2. In the comparison group, 27 were from the participant registry, 1 from community center 2, and 12 from an independent living apartment. The attrition was examined by multivariate analysis of variance. Thirty-five individuals in the MOB group and 40 in the comparison group were included in the final data analysis.

Figure 1. Process for including participants in the A Matter of Balance (MOB) group and the comparison group in the analysis, Tampa, Florida, 2013. Abbreviations: T1, Time 1; T2, Time 2; MANCOVA, multivariate analysis of covariance.

Return to Article

Top

Figure 2. Comparisons of group × time interactions for the number of falls, the Performance-Oriented Mobility Assessment (POMA), the Timed Up and Go (TUG) test, and the Functional Reach (FR) test for participants in the A Matter of Balance (MOB) program and a comparison group, Tampa, Florida, 2013^. All outcome variables were adjusted for covariates (age, race, chronic conditions, functional limitations, Montreal Cognitive Assessment scores, geriatric fear of falling measure scores, and Modified Falls Efficacy Scale scores) at Time 1. For the number of falls and the TUG test, lower numbers are better. For the POMA and the FR test, higher numbers are better. For POMA, TUG test, and FR test, differences between the MOB and comparison group were significant, P < .001.

Return to Article

Top