Predictors of critical care admission in emergency department patients triaged as low to moderate urgency

Aim. This paper is a report of a study to identify predictors of critical care admission in emergency department patients triaged as low to moderate urgency that may be apparent early in the emergency department episode of care.

Background. Observations of clinical practice show that a number of emergency department patients triaged as low to moderate urgency require critical care admission, raising questions about the relationship between illness severity and physiological status early in the emergency department episode of care.

Methods. A retrospective case control design was used. All participants were aged over 18 years, triaged to Australasian Triage Scale categories 3, 4 or 5, and attended emergency department between 1 July 2004 and 30 June 2005. Cases were admitted to intensive care unit or coronary care unit and controls were admitted to general medical or surgical units. Cases (n = 193) and controls (n = 193) were matched by age, gender, emergency department discharge diagnosis and triage category.

Results. Critical care admission associated with: (i) a presenting complaint of nausea, vomiting and diarrhoea (OR = 3·40, 95%CI:1·22–9·47, P = 0·019), (ii) heart rate abnormalities at triage (OR = 2·10, 95%CI:1·19–3·71, P = 0·011), (iii) temperature abnormalities at triage (OR = 2·87 95%CI:1·05–7·89, P = 0·041), (iv) respiratory rate at first nursing assessment (OR = 1·66, 95%CI:1·05–2·06, P = 0·31) or (v) heart rate abnormalities at first nursing assessment (OR = 1·57, 95%CI = 1·04−2·39, P = 0·033).

Conclusion. Derangements in temperature, respiratory rate and heart appear to increase risk of critical care admission. Further work using a prospective approach is needed to establish which physiological parameters have the highest predictive validity, the level(s) of physiological abnormality with highest clinical utility, and the optimal timing for collection of physiological data.