Predictive validity of two frailty tools for mortality in Chinese nursing home residents: frailty index based on common laboratory tests (FI-Lab) versus FRAIL-NH.

Abstract

Little is known about frailty in Chinese nursing home residents. (1) To evaluate the prevalence of frailty in nursing home residents according to the FI-Lab or FRAIL-NH; and (2) to compare the predictive validity of these two tools for mortality. We conducted a prospective study in four nursing homes in China. Frailty was assessed using the fatigue, resistance, ambulation, illness, loss of weight, nutrition, and help with dressing questionnaire (FRAIL-NH) and frailty index based on common laboratory tests (FI-Lab), respectively. The survival status was collected via medical records or telephone interviews. Receiver-operating characteristic (ROC) curves were calculated to estimate the area under the ROC curves (AUCs) for FI-Lab and FRAIL-NH in relation to mortality. Cox proportional hazard models were applied to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs) for mortality by FRAIL-NH and FI-Lab, separately. We included 329 participants. The FI-Lab score was significantly and strongly associated with the FRAIL-NH score (r = 0.799, p < 0.001). Frailty was defined as the FI-Lab score ≥ 0.3 or the FRAIL-NH score ≥ 6, and the prevalence of frailty was 56.2% and 58.7%, respectively. Seventy-three participants (22.7%) died during the 1-year follow-up. The FI-Lab (AUC 0.700, 95% CI 0.647-0.750) was slightly better than the FRAIL-NH (AUC 0.676, 95% CI 0.622-0.727) for predicting mortality (p = 0.025). After adjusted for age and gender, the increment of the FI-Lab score was associated with mortality (adjusted HR per 0.01 increment in score 1.07, 95% CI 1.05-1.09), the increment of the FRAIL-NH score was also associated with mortality (adjusted HR per 1 increment in score 1.28, 95% CI 1.19-1.46). The FI-Lab and FRAIL-NH are valuable for predicting mortality in Chinese nursing home residents.