Cross-cultural adaptation and psychometric properties of the Groningen Frailty Indicator (GFI) among Chinese community-dwelling older adults

Abstract

The objective was to examine the feasibility, reliability and validity of the Groningen Frailty Indicator (GFI) among Chinese community-dwelling older adults. Of the 1230 participants, 1202 (97.7%) completed all items on the GFI. The internal consistency was acceptable (Cronbach's α = 0.64), and the test–retest reliability within a 7–15-day interval was good (ICC = 0.87). The GFI showed good diagnostic accuracy in the identification of frailty with reference to the frailty index (AUC = 0.84), and the optimal frailty cut-point was 3. Convergent validity was supported by significant correlations between each domain of the GFI and the corresponding alternative measurement(s). Higher proportions of frailty (GFI ≥ 3) were found in those who were older, female, less-educated, lived alone, and had 2 or more chronic diseases than in their counterparts, supporting its known-group discriminant validity. The Chinese GFI has good feasibility, acceptable reliability and satisfactory validity among community-dwelling older adults.