Objectives: To compare 3 methods of describing the frailty of older adults in nursing homes.
Design: Secondary analysis of a prospective cohort study.
Setting: Canadian long-term care institutions.
Participants: Institutionalized older adults in the second clinical examination cohort of the Canadian Study of Health and Aging (CSHA-2; n = 728).
Measures: Frailty was measured using the Cardiovascular Health Survey definition (Frail-CHS); the CSHA- Clinical Frailty Scale (CSHA-CFS) and a frailty index (FI).
Results: The sample was very elderly (87.7 +/- 6.7 years), disabled (83%), and showed a high level of mobility impairment (83%). Each frailty measure correlated moderately well with each other (0.61-0.71) and with a disability measure (-0.45 to -0.53) but only weakly with age (0.13-0.19). By each measure, frailty was significantly associated (P < .01) with an increased risk of mortality, disability and cognitive decline. In a model that included both the frailty-CHS definition and the Frailty Index only the latter was associated with a higher risk of mortality (P < .01 for FI, P = .18 for Frail-CHS) and decline in the 3MS (P < .01 for FI, P = .20 for the Frail-CHS definition). Both measures were significantly associated with new onset disability (P < .01). Similar results were found when both the CSHA-CFS and Frailty Index were included in the models. Random combinations of 15 variables used to make up alternate 5-item Frail-CHS definitions showed that any stratification based on 5 variables allowed tertiles of risk to be discriminated.
Conclusions: Frailty is a robust concept and however defined, elderly people who are frail have worse outcomes than those who are not frail. The 3 measures showed varying ability to express grades of frailty.