Frailty in community-dwelling older adults: a comparative study of screening instruments

Abstract

Objective to analyze the agreement between the Edmonton Frail Scale (EFS) and the Clinical Functional Vulnerability Index (CFVI-20). Methods cross-sectional study, during which the Edmonton Frail Scale and the Clinical Functional Vulnerability Index were applied, at home, to older adults, registered in units of the Family Health Strategy of Montes Claros (MG) and randomly selected by lot. To evaluate the correlation and agreement between the instruments, Pearson's correlation coefficient and the weighted Kappa were calculated, considering three levels of frailty classification, as follows: "robust", "risk of frail" and "frail" for the IVCF-20 and "not frail", "vulnerable" and “frail’” for the EFS. Results We evaluated 673 older adults, predominantly brown, between 60 and 74 years old and female. According to the IVCF-20, 153 (22.7%) of the older adults were classified as "frail", 195 (29%) as "risk of frail" and 325 (48.3%) as "robust". According to the EFS, 159 older adults (23.6%) were classified as "frail"; 112 (16.6%) older adults "apparently vulnerable" and 402 (59.7%) "not frail". Pearson's correlation coefficient was 0.865 (p<0.001) and showed a positive correlation between the instruments and Kappa statistics showed a value of 0.532 (p=0.027), revealing moderate agreement. Conclusion The instruments evaluated showed moderate agreement and strong positive correlation, despite the differences between some of their components. Both showed to be compatible for the assessment of frailty in older adults in the context of Primary Health Care.