Which frailty scale for patients admitted via Emergency Department? A cohort study. (January 2019)

• Record Type: Journal Article
• Authors: Lewis, Ebony T.; Dent, Elsa; Alkhouri, Hatem; Kellett, John; Williamson, Margaret; Asha, Stephen; Holdgate, Anna; Mackenzie, John; Winoto, Luis; Fajardo-Pulido, Diana; Ticehurst, Maree; Hillman, Ken; McCarthy, Sally; Elcombe, Emma; Rogers, Kris; Cardona, Magnolia
• Is Part Of: Archives of gerontology and geriatrics. Volume 80(2019)
• Page Start: 104
• Page End: 114
• Publication Date: 2019-01
• Subjects: Geriatric assessment -- Emergency Department -- Frail elderly -- Self-report -- Prospective studies; Aging -- Periodicals; Geriatrics -- Periodicals; Gerontology -- Periodicals; Electronic journals; 305.26
• DOI: 10.1016/j.archger.2018.11.002
• Languages: English
• ISSNs: 0167-4943

Abstract

Highlights: It is feasible to establish frailty levels routinely among older patients on ED admission. Substantial variation exists in determination of frailty status using different scales. Self-reported frailty at hospital admission can accurately predict poor follow-up outcomes in the short term. CFS is the most practical and non-disruptive tool for rapid frailty assessment. Abstract: Objectives: To determine the prevalence of frailty in Emergency Departments (EDs); examine the ability of frailty to predict poor outcomes post-discharge; and identify the most appropriate instrument for routine ED use. Methods: In this prospective study we simultaneously assessed adults 65+yrs admitted and/or spent one night in the ED using Fried, the Clinical Frailty Scale (CFS), and SUHB (Stable, Unstable, Help to walk, Bedbound) scales in four Australian EDs for rapid recognition of frailty between June 2015 and March 2016. Results: 899 adults with complete follow-up data (mean (SD) age 80.0 (8.3) years; female 51.4%) were screened for frailty. Although different scales yielded vastly different frailty prevalence (SUHB 9.7%, Fried 30.4%, CFS 43.7%), predictive discrimination of poor discharge outcomes (death, poor self-reported health/quality of life, need for community services post-discharge, or reattendance to ED after the index hospitalization) for all identical final models was equivalent across all scales (AUROC 0.735 for Fried, 0.730 for CFS and 0.720 for SUHB). Conclusion: This study confirms that screening for frailty in older ED patients can inform prognosis and target discharge planning including community services required. The CFS was as accurate as the Fried and SUHB in predicting poor outcomes, but more practical for use in busy clinical environments with lower level of disruption. Given the limitations of objectively measuring frailty parameters, self-report and clinical judgment can reliably substitute the assessment in EDs. We propose that in a busy ED environment, frailty scores could be used as a red flag for poor follow-up outcome.