158 Delirium as a predictor of admission in the emergency department

Abstract

Abstract Background Frailty intervention teams are now commonplace in Emergency Departments (EDs). A geriatric assessment in the ED should be multidimensional and include demographics, details of the presenting complaint, frailty, and delirium. The aim of the present review was to establish which of these dimensions was most predictive of patient disposition (i.e. admit vs. discharge) in the ED of a large university hospital. Methods Data was reviewed from ED attendees aged ≥65 years assessed by the frailty intervention team over a three-month period. Age, sex, and whether the presenting complaint was a fall were recorded. Frailty was measured with the Clinical Frailty Scale (CFS), and delirium with the 4AT. Receiver Operating Characteristic (ROC) curves and Areas Under the Curve (AUC) were compared for the ability of age, CFS and 4AT to predict patient disposition. A multivariate binary logistic regression model was computed to assess how CFS and 4AT independently predicted patient disposition, whilst controlling for age, sex and fall as presenting complaint. Results Data from 217 patients was analysed (mean age 81.7, range 67–99 years; 66.4% women; 29.5% presenting with falls; median CFS 5; mean 4AT 1). The AUCs for age, CFS ad 4AT to predict admission were 0.59 (95% CI: 0.51–0.67, p = 0.033), 0.65 (0.58–0.73, p < 0.001), and 0.70 (0.62–0.78, p < 0.001), respectively. In the regression model, the only significant independent predictor of admission was the 4AT (OR 1.70, 95% CI: 1.37–2.12, p < 0.001). Conclusion In the ED interface, delirium was a stronger prediction of admission than demographics, fall as the presenting complaint, and frailty. While delirium is a marker of illness acuity, frailty refers to the pre-illness baseline. The measurement of delirium in the ED is important as a predictor of patient disposition, while early delirium assessment may help to improve the time-to-disposition interval.