Portal de Periódicos da CAPES

Heterogeneity of Different Tools for Detecting the Prevalence of Frailty in Nursing Homes: Feasibility and Meaning of Different Approaches

2017; Elsevier BV; Volume: 18; Issue: 10 Linguagem: Inglês

10.1016/j.jamda.2017.06.016

1538-9375

Nicolás Martínez‐Velilla, Pablo Aldaz Herce, Álvaro Casas Herrero, Marta Gutiérrez‐Valencia, Mikel L. Sáez de Asteasu, Alberto Sola Mateos, Ana Conde Zubillaga, Berta Ibáñez, Arkaitz Galbete, Míkel Izquierdo,

Geriatric Care and Nursing Homes

Background and objective The identification of frail individuals has been recognized as a priority for the effective implementation of healthy aging strategies. Only a limited number of studies have examined frailty in nursing homes, and there is a big heterogeneity in the methods used. The primary objective of this study was to determine the prevalence and feasibility of different frailty screening tools in nursing homes as well as its relationship with multimorbidity and disability. Design, setting, and participants Cross-sectional analysis from a concurrent cohort study, which included 110 participants aged over 65 years and with different degrees of disability at 2 nursing homes. Measurements The study used 4 different frailty scales: The Fried frailty criteria, the imputed Fried frailty criteria, the Rockwood clinical frailty scale, and the frailty in nursing home scale, and we analyzed their relationship with disability and multimorbidity. Results The mean age of the study population was 86.3 years (standard deviation 7.3), and 71.8% were female. Most residents had a high percentage of cognitive and functional impairment, multimorbidity, and risk of malnutrition. The following prevalence rates for frailty were determined: 71.8% (62.8, 79.4), 42.7% (33.9, 52.1), and 36.4% (23.8, 51.1) as per according to the Rockwood clinical frailty scale, frailty in nursing home scale, and Fried index (95% confidence interval), respectively. In the case of the Fried index, the prevalence of frailty is based on the percentage of patients meeting the criteria, which is 40% due to the large number of missing values. After the imputation of variables with the multivariate imputation by chained equation software, the prevalence of frailty increased to 66.4% (57.1, 74.5). We observed different statistically significant associations between the frailty scales and the clinical and demographic variables, and also with disability and multimorbidity. Conclusions Most residents of nursing homes are likely to be frail, but there is no single operational definition of frailty. Although all measures of frailty had similar associations with the clinical variables of the study, there are important conceptual differences that must be considered in addressing the relationships between frailty, disability, and multimorbidity. Further research is required, and homogeneous frailty criteria must be used so that studies and interventions can be compared.