Abstract
ObjectivesThere is a paucity of data for the assessment of frailty in acutely ill hospitalized older adults. We aim to (1) compare the performance of frailty measures [5-item scale of fatigue, resistance, ambulation, illnesses, and loss of weight) (FRAIL), Tilburg Frailty Indicator (TFI), and Clinical Frailty Scale (CFS)] in identifying frailty, using the widely adopted Frailty Index (FI) as “gold standard,” and (2) compare their ability to predict negative outcomes among hospitalized older adults. DesignProspective cohort study. SettingAcute inpatient care. ParticipantsA total of 210 patients (mean age 89.4 ± 4.6 years, 69.5% female) admitted to the Department of Geriatric Medicine. MeasurementsPremorbid frailty status was assessed by FI, FRAIL, TFI, and CFS. We collected data on comorbidities, severity of illness, functional status, and cognitive status. We compared area under receiver operator characteristic curves for FRAIL, TFI, and CFS against the reference FI. Multiple logistic regression was performed to examine the association between frailty and the primary outcome of in-hospital mortality. ResultsFrailty prevalence estimates were 87.1% (FI), 50% (FRAIL), 80% (TFI), and 81% (CFS). Area under receiver operator characteristics against FI ranged from 0.81 [95% confidence interval (CI) 0.72–0.90: FRAIL] to 0.91 (95% CI 0.87–0.95: CFS), with no significant difference on receiver operating characteristic curve contrast. Frailty, as defined by FRAIL score ≥3, was associated with higher in-hospital mortality (6.7% vs 1.0%, P = .031) and length of hospitalization [10 days (6.0–17.5) vs 8 days (5.0–14.0), P = .043]. FI [odds ratio (OR) = 1.15, 95% CI 1.00–1.33, P = .05], FRAIL (OR = 3.31, 95% CI 1.43–7.67, P = .005), and CFS (OR = 2.57, 95% CI 1.14–5.83, P = .023) independently predicted in-hospital mortality adjusted for age, sex, and severity of illness. ConclusionsFRAIL and CFS are simple frailty measures that may identify older adults at highest risk of adverse outcomes of hospitalization. FRAIL performed better in predicting in-hospital mortality.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have