| Relationship between Overcrowding and the Result Indices in an Urban Local Emergency Department |
| Chang Bae Park, Tae Yun Kim, Hyun Soo Park, Sang Do Shin |
| Department of Emergency Medicine, Seoul National University College of Medicine, Seoul, Korea. shinsangdo@snuh.org |
|
|
|
|
|
| ABSTRACT |
PURPOSE:
Overcrowding can be caused by specific and complex issues and affects multiple levels of the outcome. We evaluated the relationship between overcrowding and the result indices.
METHODS:
From February 2002 to December 2003, fundamental data were obtained from the Order Communicating System at an urban local emergency department (ED) with 530 inpatient beds and 27 emergency beds. We had also prospectively collected data on the cancellation cases for the last year.
After calculating overcrowding indices based on the daily data, such as the turnover rate, severity indices, the length of stay, and the volume of patients, we evaluated the correlation coefficient and the regression parameter estimate on between those indices and the result indices, including cancellation, death, and number of transfer.
RESULTS:
The total number of patients was 44,534. Of those, 22,793 (58.5%) were male, and a total of 8,572 (19.3%) were hospitalized. 684 (1.5%) died at the ED, and 964 (2.2%) were transferred. The mean counts of cancellations, deaths, and transfers were 13.3, 2.2, and 1.6 per day, respectively. The cancellation count was strongly positively correlated with the turn-over rate (correlation coefficient = 0.421, p=0.000), but was negatively correlated with the length of stay. The parameter estimate of the turnover rate was 4.895 in the simple regression model and 9.952 in the adjusted model.
CONCLUSION:
Overcrowding can be calculated as a hospital-specific indices, which can be correlated with the daily cancellation count. Of the overcrowding indices, the parameter estimate of the turnover rate was the highest. |
| Key words:
Crowding, Outcome, Correlation, Regression, Emergency |
|
PDF Links
Full text via DOI 
Download Citation 
