Abstract
INTRODUCTION: Frailty likely contributes to disproportionate health care utilization among people living with chronic kidney disease (CKD) and undergoing hemodialysis (HD); but this is poorly captured in nephrology clinical and research practice. We examined Fried frailty phenotype among participants with CKD or on HD and explored associations with health care utilization. We examined frailty transitions in relation to hospitalization. METHODS: We conducted a prospective observational single-center study of patients with advanced CKD or undergoing HD. Frailty was assessed at baseline, 6 and 12 months. Demographic and clinical data, including comorbid burden, disability, and laboratory parameters were recorded. Data linkage with tertiary hospital captured emergency department (ED) presentations, hospital admissions, and days of hospital stay, excluding admissions for maintenance HD. Negative binomial regression was used to model health care utilization patterns. Frailty progression over study follow-up was described using Cox proportional hazards modelling. RESULTS: Among 256 participants, frailty (36.3%) and prefrailty (46.5%) were highly prevalent. Frailty independently predicted ED presentation (incidence rate ratio [IRR]: 1.25, 95% confidence interval [CI]: 1.09-1.43), hospitalization (IRR: 1.22, 95% CI: 1.08-1.37), and total days of hospitalization (IRR: 1.29, 95% CI: 1.06-1.57) independent of demographics, comorbidity, disability, and inflammation. The median occurrence of hospitalization events was 152 days (interquartile range [IQR]: 44-251) after enrolment, suggesting a window of opportunity where frailty recognition might prompt targeted intervention to prevent frailty-related sequelae. Frailty was highly dynamic; frailty progression was not associated with hospitalization or length of stay. CONCLUSION: Frailty is a major contributor to excess health care utilization among people with kidney disease. Recognition of the prognostic implications of frailty might allow timely introduction of interventions to improve patient outcomes.