FRAILTY INDICES AND THE CLINICAL FRAILTY SCALE EACH PREDICT 30-DAY MORTALITY IN ED PATIENTS. (11th November 2018)

• Record Type: Journal Article
• Authors: Rockwood, K; van der Valk, A; Eendebak, R; Lines, K; Howlett, S; Theou, O
• Is Part Of: Innovation in aging. Volume 2(2018)Supplement 1
• Page Start: 491
• Page End: 491
• Publication Date: 2018-11-11
• Subjects: Aging -- Periodicals; Gerontology -- Periodicals; 612.67
• DOI: 10.1093/geroni/igy023.1831
• Languages: English
• ISSNs: 2399-5300

Abstract

Abstract: We examined the ability of frailty indices and the Clinical Frailty Scale to predict mortality in acutely ill, older (aged 65+ years) Emergency Department (ED) patients referred to internal medicine. Here we report on the first 415 patients (Mage = 80.6 ± 8.3, 58.1% women) with complete data. Acuity was assessed using the Canadian Triage and Acuity Scale (CTAS). A geriatrician (KR) conducted a Comprehensive Geriatric Assessment from which an FI-score was calculated based on the patient's current state (FI-CGA). The Clinical Frailty Scale (CFS) was also completed. An FI-Lab was constructed using 30 common laboratory tests collected within 2 days of CGA administration. Of the 415 patients, 321 (77.3%) were admitted, staying an average of 26.9 ± 45.8 days. Mortality within 30 days was 2.6% (CFS), 8% (FI-CGA) and 7.9% (FI-Lab) for the least frail groups and 41.8% (CFS), 28.2% (FI-CGA), and 29.7% (FI-Lab) for the frailest groups. After adjusting for age, sex, and CTAS, all three frailty tools independently predicted 6-month mortality. The hazard ratio was 1.67 (95% CI: 1.48–1.89) per 1-grade increase in CFS, 1.04 (1.03–1.06) per 0.01-point increase in the FI-CGA and 1.05 (1.02–1.07) per 0.01-point increase in the FI-Lab. ED patients requiring CTAS-defined urgent assistance were at higher risk for mortality than less urgent cases; 2.65 (1.71–4.12) CFS model, 2.84 (1.80–4.48) FI-CGA model, 1.81 (1.08–3.02) FI-Lab model. An FI from common laboratory tests is a promising tool for grading frailty and predicting mortality for ED patients. It can be automated readily and adds prognostic information to the CFS.