Factors related to delayed intensive care unit admission from emergency department—A retrospective cohort study

• Published in: Acta anaesthesiologica Scandinavica Vol. 63; no. 7; pp. 939 - 946
• Main Authors: Aitavaara‐Anttila, Mia, Liisanantti, Janne H., Raatiniemi, Lasse, Ohtonen, Pasi, Ala‐Kokko, Tero
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.08.2019
• Subjects: anesthesiology; Cohort analysis; Coma; Delayed admission; delayed ED LOS; Diagnosis; Diagnostic systems; emergency department length of stay; Emergency medical services; emergency medicine; Health risk assessment; Intensive care; intensive care medicine; Organizational aspects; Overcrowding; Patients; Platelets; Population studies; Regression analysis; Sepsis; Thrombocytes; Trauma; delayed ED LOS; Delayed admission; anesthesiology; intensive care medicine; emergency department length of stay; emergency medicine
• ISSN: 0001-5172; 1399-6576
• DOI: 10.1111/aas.13355

Background The delays in transferring patients from emergency department (ED) to intensive care unit (ICU) are known to be linked with several adverse events, including prolonged ICU stay and increased hospital mortality. The factors associated with delayed ICU admission include shortage of ICU beds, organizational factors, ED overcrowding, and patient‐related factors, including sepsis as admission diagnosis. The aim of this study was to examine ED‐related factors associated with prolonged ED stay. Methods The study population consisted of adult patients admitted (n = 479) from ED to ICU between 31 May 2016 and 19 March 2017 in Oulu University Hospital. A patient's ED length of stay (LOS) exceeding 180 minutes was considered delayed. Results Most of the patients (380, 79.3%) were admitted to the ICU within 3 hours of hospital admission. In a logistic regression analysis, odds ratios (ORs) for ED LOS > 180 minutes were as follows: for Glasgow Coma Scale score > 9, 2.73 (1.39‐5.32); for thrombocytes < 100 × 109/mmol, 6.69 (2.32‐19.26); for absence of pre‐arrival notification, 5.27 (3.04‐9.14); and for radiological examination, 3.95 (1.72‐9.10). Trauma and intoxicated patients had shorter ED LOS while patients with medical conditions had more often prolonged admissions. Conclusion The delays in ICU admissions were linked to therapeutic and diagnostic procedures and absence of pre‐arrival notification. Patients were admitted to the ICU on the basis of diagnosis instead of clinical risk. However, the delays were not associated with worsening outcome, which indicates that sufficient care can be provided at the ED while the ICU admission is pending.