The Reliability And Validity Of Groningen Frailty Indicator And Its Application

ObjectiveIntroduce and translate the Groningen Frailty Indicator(GFI)into Chinese,and assess reliability and validity of the GFI to explore its applicability in Chinese nursing homes.Evaluating among nursing home residents,the extent to which the GFI and FP predict mortality,hospital admission,falls and development of disabilities at 1 year,so as to provide a theoretical basis for the development of targeted interventions in the future.MethodAuthorized by the GFI author team,the scale was translated by Brislin's pre-translation-back translation method to form the first draft.Then,according to the opinions of experts and residents,the Chinese version of the GFI was formed.Taking the convenient sampling method,four nursing homes in Changchun City were selected.The Fried Phenotype(FP),the Activities of Daily Living(ADL),the Mini Nutritional Assessment(MNA),the Mini-Mental State Examination(MMSE),the Social Support Rating Scale(SSRS),the 20-item Epidemiologic Studies Depression Scale(CES-D)and the Short Form 36(SF-36)mental component summary were used to assess the reliability and validity of the GFI.The internal consistency reliability was calculated by Cronbach's ?,and the test-retest reliability was analyzed by the Pearson correlation coefficient.The content validity was analyzed by calculating the item content validity(I-CVI)and the scale average content validity(S-CVI);the convergent validity was measured by using Spearman rank correlations between the GFI domains and alternative measurements;the known-group divergent validity was assessed by chi-square test;the criterion validity used the FP as the "gold standard",and the ROC is analyzed.One year later,the elderly in the nursing homes were asked about mortality,hospital admission,falls and development of disabilities during the year.AUC for the adverse outcomes was calculated to test the predictive validity of the GFI and FP;the binary logistic regression analysis was performed,and odds ratios were calculated to compare frail versus non-frail groups for their risk for the adverse outcomes.Results1.Using the Chinese version of GFI to investigate the prevalence of frailty in the Chinese nursing homes was 61.6%,and the pre-frailty was 10.5%.2.The Chinese version of GFI has a Cronbach's ? coefficient of 0.712 and a test-retest reliability coefficient of 0.939,which indicates good reliability.3.The Chinese version of GFI has good content validity with a S-CVI of 0.966;the prevalence of frailty in older group and younger group is statistically different,indicating good known-group divergent validity;correlations of each dimension with related external scale are medium to high,indicating that the convergent validity of GFI is excellent;the area under the ROC curve with FP as a reference is 0.823,which indicates that the criterion validity of the Chinese version of GFI is good,and 4 is chosen as the best cut-off to distinguish frailty and robust..4.The mortality of the residents in the nursing homes was 11.9% within one year.The age,monthly income and the number of chronic diseases were the influencing factors of mortality.The ROC curve showed that the predictive validity of the GFI on mortality was higher than that of FP.The binary logistic regression analysis of mortality found the adjusted OR of GFI was 1.467 and that of FP was 1.739.5.The falls rate of the residents in the nursing homes was 16.6% within one year.The age and the number of chronic diseases were the influencing factors of the falls;the ROC curve showed that the predictive validity of GFI was lower than that of FP;the binary logistic regression analysis of the falls found the adjusted OR of GFI was 1.200.6.The hospital admission rate of residents in the nursing homes was 23.3% within one year.The age and the number of chronic diseases were the influencing factors of hospital admission;the ROC curve showed that the predictive validity of GFI and FP was similar in hospital admission;the binary logistic regression analysis of hospital admission on found the adjusted OR of GFI was 1.272 and that of FP was 1.552.7.The rate of development of disabilities in the nursing home residents is 23.9% within one year.The age and the number of chronic diseases are the influencing factors of development of disabilities;the ROC curve shows that the predictive validity of GFI for development of disabilities is higher than that of FP;Binary logistic regression analysis found that the adjusted OR value of GFI was 3.278 and that of FP was 2.384.Conclusion1.Chinese version of GFI has good reliability and validity,and can be applied to the screening of frailty in Chinese nursing homes.2.Frailty has a good predictive validity on the morality,hospitalization and development of disabilities of the nursing home residents within one year.3.The predictive validity of GFI on the adverse outcomes in nursing home residents is higher than that of FP.