Abstract
BACKGROUND: The use of short geriatric tools in the emergency department (ED) is increasing, but the literature is still conflicting. The aim of this study is to compare the precision and the accuracy of two short geriatric assessment tools to predict mortality in a cohort of older patients attending the ED. METHODS: A retrospective study was conducted including patients ≥ 65 years, attending the ED and transferred to a medical assessment unit from February to July 2022. Clinical Frailty Scale (CFS) and Brief Multidimensional Prognostic Index (Brief MPI) were administered. The association between Brief MPI and CFS and mortality was analysed via area under the curve (AUC) with its 95% confidence intervals (CIs), the C-statistics and a multivariate Cox's regression analysis, in the latter case reporting the data as hazard ratios (HRs) with their 95% CI. RESULTS: Among the 579 patients enrolled (mean age: 77 years), both Brief MPI and CFS showed a good accuracy in predicting mortality (AUC: 0.72; 95% CI: 0.61-0.83 for Brief MPI; 0.754; 95% CI: 0.65-0.83 for CFS). The discrimination of Brief MPI and CFS in predicting mortality was excellent, since the C-index of the Brief MPI was 0.85 and of CFS = 0.84. In the multivariate analysis, the risk for mortality was significantly increased for frailer subjects (HR 4.65; 95% CI: 1.45-15.00 for Brief MPI > 0.66; HR = 9.24; 95% CI: 1.16-76.90 for CFS > 6). CONCLUSIONS: Brief MPI and CFS showed a good accuracy/precision to predict mortality in older patients attending the ED. Considering that they are quick to perform, their introduction in ED clinical practice could be extremely helpful.