ED crowding is associated with variable perceptions of care compromise

Acad Emerg Med. 2007 Dec;14(12):1176-81. doi: 10.1197/j.aem.2007.06.043.


Objectives: The authors measured the association between emergency department (ED) crowding and patient and provider perceptions about whether patient care was compromised.

Methods: This was a cross-sectional study of patients admitted from the ED and their providers. Surveys of patients, nurses, and resident physicians were linked. The primary outcome was agreement or strong agreement on a five-item scale assessing whether ED crowding compromised care. Logistic regression was used to determine the association between the primary outcome and measures of ED crowding.

Results: Of 741 patients approached, 644 patients consented (87%); 703 resident physician surveys (95%) and 716 nursing surveys (97%) were completed. A total of 106 patients (16%), 86 residents (12%), and 173 nurses (24%) reported that care was compromised by ED crowding. In 252 cases (35%), one or more respondents reported that care was compromised. There was poor agreement over whose care was compromised. For patients, independent predictors of compromised care were waiting room time (odds ratio [OR], 1.05 for each additional 10-minute wait [95% confidence interval {CI} = 1.02 to 1.09]) and being surveyed in a hallway bed (OR, 2.02 [95% CI = 1.12 to 3.68]). Predictors of compromised care for nurses included waiting room time (OR, 1.05 for each additional 10-minute wait [95% CI = 1.01 to 1.08]), number of patients in the waiting room (OR, 1.05 for each additional patient waiting [95% CI = 1.02 to 1.07]), and number of admitted patients waiting for an inpatient bed (OR, 1.08 for each additional patient [95% CI = 1.03 to 1.12]). For residents, predictors of compromised care were patient/nurse ratio (OR, 1.39 for a one-unit increase [95% CI = 1.09 to 1.20]) and number of admitted patients waiting for an inpatient bed (OR, 1.14 for each additional patient [95% CI = 1.10 to 1.75]).

Conclusions: ED crowding is associated with perceptions of compromised emergency care. There is considerable variability among nurses, patients, and resident physicians over which factors are associated with compromised care, whose care was compromised, and how care was compromised.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Cross-Sectional Studies
  • Crowding*
  • Emergency Service, Hospital / standards
  • Emergency Service, Hospital / statistics & numerical data*
  • Health Services Accessibility* / statistics & numerical data
  • Health Services Research
  • Hospitals, University / organization & administration
  • Hospitals, Urban / organization & administration
  • Humans
  • Logistic Models
  • Outcome Assessment, Health Care*
  • Patient Satisfaction
  • Philadelphia
  • Prospective Studies
  • Quality of Health Care*
  • Time Factors
  • Waiting Lists