VALIDATION OF THE LATE LIFE FUNCTION AND DISABILITY INDEX (LLFDI) AND GAIT EFFICACY SCALE (GES) WITH PERFORMANCE BASED MEASURES OF FUNCTION AND MOBILITY IN COMMUNITY DWELLING OLDER ADULTS.

Abstract

Purpose/Hypothesis: Defining the relation of recently developed self-report measures of function and mobility with established performance-based measures is important to determine construct validity and potential clinical applications. We examined the validity of the LLFDI function, LLFDI disability and GES self report measures by comparison to performance based measures of physical function [7-item Physical Performance Test (PPT)] and mobility (gait speed). We expected participants with better function, less disability and greater confidence in walking would perform better on the PPT and walk faster. We also hypothesized self-reported function and disability would discriminate between participants classified as high vs. low functioning, and selfreported walking confidence would discriminate between those walking at usual vs. slow speeds. Subjects: Twenty community-dwelling older adults (mean age 74.3 [5.3] years; 90% white; 40% male) participated in this cross-sectional validation study. Materials/Methods: Self-report measures included the LLFDI function, LLFDI disability and the GES (all measures score range, 0–100; higher scores=better function, less disability or greater walking confidence). Performance based measures included the PPT (0–28, higher scores=better function) and gait speed. Construct validity was determined using Spearman Rho correlations to define relations between self-reported function, disability, and confidence in walking with the performance based measures. Mann Whitney tests were used to assess differences in self-reported function and disability for participants classified as high (PPT ≥75 percentile) vs. low functioning (PPT <75 percentile), and differences in self-reported walking confidence between those who walked at usual (>1.0m/s) vs. slow (≤1.0m/s) speeds. Results: Self-reported function (LLFDI function) and confidence (GES) were positively related to gait speed (r=0.75, p<0.001; r=0.60, p=0.007 respectively). Self-reported function, differed between participants classified as high (n=15) vs. low (n=5) functioning (LLFDI function: 65.6 vs 52.8, p=0.03). Selfreported disability (LLFDI disability) was not associated with the performance based measures and did not discriminate between functional groups. Reported confidence in walking differed between participants walking at usual (n=14) vs. slow (n=6) speeds (GES: 97.5 vs. 90.5 p=0.025). Conclusions: Construct validity was demonstrated for the LLFDI function and GES scales by the relation with performance-based measures and by expected differences in known functional and mobility groups. Clinical Relevance: These self-report measures may provide complimentary information that adds to, but does not replace, the description of physical function of community-dwelling older adults provided by physical performance-based measures.