• Noortje Zelis, Jacqueline Buijs, Peter W de Leeuw, van KuijkSander M JSMJDepartment of Clinical Epidemiology and Medical Technology Assessment, Maastricht University Medical Centre, Maastricht University, Maastricht, the Netherlands., and Patricia M Stassen.
• Department of Internal Medicine and Gastroenterology, Zuyderland Medical Centre, Heerlen, the Netherlands; Department of Internal Medicine, Division of General Internal Medicine, Section Acute Medicine, Maastricht University Medical Centre, Maastricht University, Maastricht, the Netherlands; CARIM School for Cardiovascular Diseases, Maastricht University Medical Centre, Maastricht University, Maastricht, the Netherlands. Electronic address: noortje.zelis@mumc.nl.
• Eur. J. Intern. Med. 2020 Jul 1; 77: 36-43.

Background/ObjectivesCurrently, accurate clinical models that predict short-term mortality in older (≥ 65 years) emergency department (ED) patients are lacking. We aimed to develop and validate a prediction model for 30-day mortality in older ED patients that is easy to apply using variables that are readily available and reliably retrievable during the short phase of an ED stay.MethodsProspective multi-centre cohort study in older medical ED patients. The model was derived through logistic regression analyses, externally validated and compared with other well-known prediction models (Identification of Seniors at Risk (ISAR), ISAR-Hospitalised Patients, Acute Physiology and Chronic Health Evaluation II (APACHE II) and Modified Early Warning Score (MEWS)).ResultsWithin 30 days after presentation, 66 (10.9%) of 603 patients in the derivation cohort and 105 (13.3%) of 792 patients in the validation cohort died. The newly developed model included 6 predictors: age, ≥2 abnormal vital signs, serum albumin, blood urea nitrogen, lactate dehydrogenase, and bilirubin. The discriminatory value of the model for mortality was very good with an AUC of 0.84 in the derivation and 0.83 in the validation cohort. The final model was excellently calibrated (Hosmer-Lemeshow p-value 0.89). The discriminatory value of the model was significantly higher than that of the four risk stratification scores (highest AUC of 0.69 for ISAR score, p-value 0.007).ConclusionWe developed and externally validated an accurate and simplified prediction model for 30-day mortality in older ED patients. This model may be useful to identify patients at risk of short-term mortality and to apply personalised medical care.Copyright © 2020 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.

15 keywords...

hide…