The Clinical Frailty Scale and Health Services Use for Older Adults in the Emergency Department

Abstract

The 2013 consensus geriatric emergency department (GED) guidelines, endorsed by the American Geriatrics Society and other professional societies,1 recommend “routine screening for all older adult patients at higher risk for adverse outcomes.” However, current screening tools are not predictive of health services use.2 Frailty, as measured by the Clinical Frailty Scale (CFS), is associated with poor outcomes for older adults, including mortality and functional decline.3-6 We evaluated whether CFS may be a useful ED screening tool by examining its association with health services use for older adults. This study was a retrospective cohort analysis of ED patients 65 years and older enrolled at a U.S., urban, academic Level 1 GED from September 1, 2019 to January 31, 2020 and who met inclusion criteria. A detailed description of the study methods are provided in the Supplementary Materials (Supplementary File S1). CFS was recorded using a standardized instrument by trained ED nurses at the bedside. Data on patient characteristics and study outcomes were extracted from the electronic health record. The primary outcome was hospital admission at the time of ED visit within the study period. Secondary outcomes included ED return visit within 9 days and subsequent admission within 30 days of ED visit at the same institution. Odds ratios (ORs) were calculated for each outcome using univariable logistic regression with CFS score as the independent variable. We chose a univariable rather than multivariable model a priori in order to determine whether CFS alone was associated with health outcomes and could function as a single-variable screening tool for older adults in the ED. A stepwise receiver operator characteristic curve was constructed for each outcome. Area under the receiver operating curve (AUC) was calculated. CFS cut-points were determined for each outcome using the Youden index. During the study period, 8,258 patients 65 years and older visited the ED. CFS screening was performed for 7,304 (88.4%) patients who met inclusion criteria. Patients with and without documented CFS scores were similar in regard to age, Emergency Severity Index, and sex. Patients with CFS scores were more likely to have increased ED length of stay (542 minutes vs 395), a higher admission rate (54.9% vs 42.2%), and lower ED return visit rate (7.09% vs 10.0%) compared to those without. The most common CFS score was 3 (managing well), 2,214/7,304 (26.8%) (Supplementary Tables S1 and S2). Higher CFS was associated with increased odds of admission (OR = 1.42 (95% confidence interval (CI) = 1.38–1.46)) and subsequent admission within 30 days (OR = 1.29 (95% CI = 1.23–1.35)) but not ED return visits within 9 days (OR = 1.01 (95% CI = 0.966–1.07)) (Supplementary Figure S1). The AUC for admission was 0.66 (95% CI = 0.65–0.67) (Figure 1). The optimal cut-point was CFS ≥4 (vulnerable) for admission and subsequent admission. For prediction of admission, CFS score 4–9 had a sensitivity of 54.0% (95% CI = 52.4–70.1) and specificity of 70.1% (95% CI = 68.5–71.6). The optimal cut-point for ED return visit was ≥3. CFS score 3–9 had a sensitivity of 78.8% (CI 75.0–82.1) and specificity of 27.0% (CI 26.0–28.1) for ED return visits (Supplementary Table S3). The results of our study demonstrate an association between CFS score and admission at ED visit and within 30 days. Previous studies have demonstrated an association between CFS score and mortality and institutionalization for older adults, both in community and ED populations.3-6 CFS has also been demonstrated to be feasible and reliable for ED use.7, 8 These features make CFS an appealing candidate to help identify older patients at higher risk for adverse outcomes who might benefit form GED interventions, as recommended in the GED guidelines.1 Our results suggest that CFS alone does not adequately identify older adults at risk for admissions or return ED visits within the specified time frame. However, its reliability and association with ED outcomes points to its potential value as a component of a composite score including acuity (e.g., estimated severity index) and complexity (e.g., CFS) to improve upon previously published risk scores.9, 10 The study limitations included a single study site, data limited to return ED visits and future admissions only at the same institution, and the finding that patients with CFS recorded had higher admission rates and longer ED stays than those without; however, the direction of any bias on results is unclear given the limited research in this area. Although CFS may not be optimal as a standalone screening tool, our results suggest frailty is an important risk factor for health services use among older ED patients. This study was funded by Agency for Healthcare Research and Quality (1R01HS026489-01) and Northwestern University Clinical and Translational Sciences Institute Pilot and Voucher Program (110-5430000, CF 1292). The authors have no conflicts. PS and SMD developed the concept, designed the study, and obtained the research funding. DC acquired the data. PS, MK, and EG performed analysis and provided statistical advice. PS, AXL, SMD interpreted the data. PS drafted the manuscript and all authors contributed substantially to its revision. PS takes responsibility for the paper as a whole. The sponsors did not play a role in the design, methods, data collection, analysis or preparation of this paper. Supplementary Table S1. Sample Size, Age, Sex, Race, Emergency Department Length of Stay (ED LOS), and Admission Stratified by CFS Supplementary Table S2. Sample Size, Age, Sex, Race, Emergency Department Length of Stay (ED LOS), and Admission Stratified by Recorded Versus Missing CFS Supplementary Figure S1. Probability of admission, subsequent admission and ED return visit as a function of CFS score. Gray bars are estimated rates. Red line represents predicted values based on logistic regression models, with 95% CI Supplementary Table S3. Sensitivity and Specificity for Admission (Inpatient, Observation, Combined) During First ED Visit, ED Return Visits Within 9 Days, and Subsequent Admission Within 30 Days of First ED Visit Supplementary File S1. Detailed description of study methods Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts