Comparing Diagnostic Properties of the FRAIL-NH Scale and 4 Frailty Screening Instruments among Chinese Institutionalized Older Adults.

Abstract

To examine the diagnostic test accuracy (DTA) of the FRAIL-NH and four frailty screening instruments among institutionalized older adults. Cross-sectional study. Institutionalized setting, Jinan, China. A total of 305 older adults (mean age 79.3 ± 8.4 years, 57.0% female) were enrolled from nursing homes. Frailty was assessed by the FRAIL-NH, Physical Frailty Phenotype (PFP), FRAIL, Tilburg Frailty Indicator (TFI), and Groningen Frailty Indicator (GFI), respectively. The Comprehensive Geriatric Assessment (CGA) was used as a reference standard of frailty. The receiver operating characteristic (ROC) curve was plotted to examine the DTA of five frailty screening instruments against the CGA. The optimal cut-point was determined by the maximum value of the Youden index (YI, calculated as sensitivity + specificity - 1). The prevalence of frailty ranged from 25.9% (FRAIL) to 56.4% (GFI). Areas under the curve (AUCs) against the CGA ranged from 0.80 [95% confidence interval (CI) 0.74 - 0.85: FRAIL] to 0.83 (95% CI 0.78 - 0.88: PFP). At their original cut-points, all five frailty screening instruments presented low sensitivity (32.9% - 69.3%) and high specificity (80.0% - 93.8%), as well as high positive predictive values (90.7% - 94.9%) and low negative predictive values (33.2% - 48.1%). At their optimal cut-points, the sensitivity and specificity of the FRAIL-NH, PFP, and FRAIL tended to be balanced, and their correctly classified rates (76.1% - 81.3%) and kappa values (0.465 - 0.523) increased a lot. ROC contrasts showed that all five frailty screening instruments had similarly good diagnostic accuracy (χ2: 0.0003 - 1.38, P > .05). In the institutionalized setting, the specific FRAIL-NH, self-report FRAIL, TFI, and GFI as well as hybrid PFP, show similarly good diagnostic properties in identifying frailty against the CGA.