Open AccessSymptoms at presentation to the emergency department: Predicting outcomes and changing clinical practice?
Author(s) -
Weigel Karin,
Nickel Christian H.,
Malinovska Alexandra,
Bingisser Roland
Publication year - 2018
Publication title -
international journal of clinical practice
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.756
H-Index - 98
eISSN - 1742-1241
pISSN - 1368-5031
DOI - 10.1111/ijcp.13033
Subject(s) - medicine , triage , emergency department , observational study , comorbidity , prospective cohort study , logistic regression , emergency medicine , severity of illness , disease , pediatrics , psychiatry
Background The type and number of symptoms in emergency patients differ widely. It remains unclear, if outcomes can be predicted by the number of symptoms. Furthermore, it is unknown, whether clinical practice could be influenced by presenting symptoms. Methods Prospective observational study, performed in the emergency department of the University Hospital Basel, a tertiary hospital. A consecutive sample of patients was interviewed at presentation for a predefined set of 35 symptoms. The number of symptoms was correlated with outcomes using linear and logistic regression models. Clinical practice was observed using prospective data on disease severity ratings, triage category, use of resources, length‐of‐stay and follow‐up presentations. Results Data of 3472 patients were analysed. The number of symptoms ranged between 1 and 25, the mean being 2.74. Women reported more symptoms than men. Age and comorbidity indices were not associated with the number of symptoms. After adjusting for age and gender, there was no correlation between the number of symptoms and adverse outcomes, such as ICU ‐admission or in‐hospital mortality (OR: 1.03, CI: 0.88‐1.18, P = .68). The number of symptoms at presentation was associated with hospitalisation, disease severity rating by patients and emergency physicians, triage categories, use of resources, length‐of‐stay and follow‐up presentations. Conclusion The number of symptoms did not correlate with the main adverse outcomes ( ICU ‐admission and in‐hospital mortality). However, clinical practice was influenced by the nature and number of symptoms. This was shown by associations with hospitalisation, length‐of‐stay, use of resources and follow‐up presentations. Furthermore, the number of symptoms correlated with the caregivers’ disease severity ratings and the attributed triage categories. This may indicate that caregivers respond to the number of symptoms by a higher investment in their immediate work‐ups and later follow‐ups.