Predictive validity of the classification schema for functional mobility tests in instrumental activities of daily living decline among older adults

Abstract

Objective: To determine predictive validity for cut points of the Timed Up & Go (TUG) test and life-space assessment (LSA) on decline in instrumental activities of daily living (IADLs) among older adults.

Design: Cross-sectional and 1-year follow-up study.

Setting: Preventive health care services.

Participants: In a cross-sectional study, 2404 older adults (65-100 y) were recruited to determine cut points for the TUG and LSA for IADLs limitation. For longitudinal analysis, 436 older adults (65-100 y) were followed over 1 year to explore the validity of a classification model using the cut points to predict incident IADLs decline.

Interventions: Not applicable.

Main outcome measures: The TUG, LSA, and Tokyo Metropolitan Institute of Gerontology index of IADLs measurement.

Results: The cut points associated with IADLs limitations for the TUG and LSA were 12 seconds and 56 points, respectively. Participants were classified into fast/high (most able; TUG <12 and LSA >56), fast/low, slow/high, and slow/low (vulnerable; TUG > or =12 and LSA < or =56) groups; there were 813 (34%), 385 (16%), 246 (10%), and 960 (40%) participants in each group, respectively. The proportions of participants with IADLs limitation in the most able, fast/low, slow/high, and vulnerable groups were 19%, 64%, 61%, and 89%, respectively. The vulnerable group included significantly more participants with IADLs limitation than any other group (P<.001). Compared with a most able group, the odds ratios of IADLs decline for the fast/low and vulnerable groups were 2.52 (95% confidence interval 1.15-5.53, P<.05) and 2.87 (95% confidence interval 1.38-5.96, P<.01), respectively.

Conclusions: The combination of TUG and LSA identifies persons with future IADLs decline and has the potential to be used by community health care services to target individualized interventions.