The association between time to disposition plan in the emergency department and in-hospital mortality of general medical patients

Intern Med J. 2012 Apr;42(4):444-50. doi: 10.1111/j.1445-5994.2011.02502.x.

Abstract

Background: A perceived risk of time-limited emergency department (ED) assessment of patients is inadequate workup leading to inappropriate disposition. The aim of this study was to examine the association of time to disposition plan (TDP) on ED length of stay (LOS) and correlate this to mortality.

Methods: A retrospective review of data collected from ED information systems at three hospitals was conducted between June 2008 and October 2009. Included patients were admitted to a general medical unit. Patients were excluded if admitted to intensive care, coronary care, a cardiac monitored bed or required surgery in first 24 h or had an expected LOS of <48 h. Multivariate regression analysis was used to identify independent associations with mortality.

Results: A total of 10,107 patient episodes was analysed, of which 6768 patients (67.0%) had an ED LOS of ≥8 h. There was significant effect modification by ED LOS in the association of TDP and mortality. In the setting of longer ED LOS, a TDP of <4 h was associated with significantly higher mortality (OR 1.57, 95% CI: 1.28-1.92, P < 0.001), corrected for age, gender and triage category. This association was not significant when ED LOS was <8 h (OR 0.88, 95% CI: 0.60-1.27, P = 0.49).

Conclusions: In the setting of prolonged ED LOS, completing ED assessment and management within 4 h of presentation was associated with significantly higher mortality. Further prospective studies are required to understand the relationship between rapid decision making in the ED and patient safety.

MeSH terms

  • Aged
  • Aged, 80 and over
  • Australasia
  • Emergency Service, Hospital / statistics & numerical data*
  • Female
  • Hospital Mortality*
  • Humans
  • Length of Stay / statistics & numerical data*
  • Male
  • Middle Aged
  • Patient Admission / statistics & numerical data*
  • Retrospective Studies
  • Time Factors
  • Triage