The Construct Validity of the Life-Space Assessment in the Canadian Longitudinal Study on Aging (CLSA).

Abstract

Background: Life-space mobility refers to the way individuals maneuver themselves in their home and community and is commonly assessed using the Life-SpaceAssessment (LSA). Before using the LSA, it is important to ensure that it demonstrates validity in the population being used, as culture, built environment, and modes of transportation in different geographical regions can influence mobility. Therefore, the purpose of this study was to assess the convergent, divergent, and known-groups validity of the LSA in community-dwelling Canadian middle-aged and older adults using data from the Canadian Longitudinal Study on Aging (CLSA).

Methods: Spearman's correlation coefficients were calculated for all measures of convergent (hypothesis: r > 0.5) and divergent (hypothesis: r < 0.3) validity. Known-groups validity was assessed by determining if the LSA can differentiate between people with chronic diseases and a history of falls using measures of clinical significance (i.e., LSA score change of 5 or more points).

Results: The sample (n = 24,577) had a median LSA score of 86.0. Correlations between the LSA and comparator measures of convergent validity (r = 0.10-0.29) did not meet the hypothesis; however, measures of divergent validity (r = 0.15-0.17) met the hypothesis. For known-groups validity, the LSA can be used to differentiate between those with (median score = 84; interquartile range [IQR] = 72-100) and without (median score = 90; IQR = 74-100) a history of falls among the whole sample. Further, it can be used to differentiate between those with and without chronic conditions, among males and females aged 75+ years.

Conclusion: The LSA can be used to differentiate between known groups. However, there is limited evidence of its convergent validity as its relationship with other measures of actual mobility, perceived mobility, and locomotor capacity for mobility was weak. Future studies need to assess the convergent validity of the LSA against different comparator measures before its use among community-dwelling Canadians.