To the Editor.
We read with interest the article by James et al1 concerning the association between race/ethnicity and emergency department (ED) wait times. As pediatricians, we obviously are concerned about disparities in care, and as pediatric emergency medicine physicians, the potential existence of this disparity within our own ED is particularly concerning. We found the study and the analysis to be very well done but did have some questions regarding potential analyses that may further elucidate the disparity that is under investigation.
Our first question concerns time of day, a potentially important confounder. ED census, and wait time, varies over the course of the day. It is plausible to hypothesize that, because of work or child care issues, minority children may be more likely to present for care in the ED during the evening hours. These are the hours at which the ED is busiest and wait times for any given level of triage category would be longer than at other times during the day. If this information is available, we would be interested in knowing its effect on the disparity presented.
Second, we wondered if any analysis was done by stratifying the triage category. Although the authors controlled for triage urgency as a possible confounding variable, it is also possible that the association between wait time and race/ethnicity differs according to urgency (ie, there may be effect modification). It seems that additional information might be gained from evaluating if there are differences in the wait time within each of the triage categories or if the regression results are driven by longer waits for only the most nonurgent of triage categories. Although this analysis may be difficult using the percent-change-in-wait-time outcome that was used by the authors, the National Hospital Ambulatory Medical Care Survey database certainly has a large enough sample that stratified analysis (using minutes as the outcome) may be possible. It might be interesting to determine if there was a similar delay between minorities (most notably Hispanic white children) and non-Hispanic white children in each category. Using the percent change in wait time, a similar difference in actual wait time for the different triage categories would be a decreasing percentage difference between the groups as the triage status became more nonurgent.
Third, the authors state that they desired to analyze differences between the EDs in children's hospitals and nonchildren's hospitals. Although we agree that it is not possible explicitly to determine which children are treated in general or pediatric EDs, it is possible to infer this relationship. The hospitals within the National Hospital Ambulatory Medical Care Survey are dummy coded, and the percentage of that hospital's patients that are <18 years old can be calculated. An ED that sees 95% to 100% children could most likely be determined to be a pediatric ED.
In summary, we appreciate the outstanding work done by the authors and hope that future studies will not only provide more information about disparities in wait times but also seek to eliminate these disparities when the root causes are better understood.
REFERENCE
This article has been cited by other articles:
![]() |
D. W. Lewis, P. Winner, A. D. Hershey, W. W. Wasiewski, and on behalf of the Adolescent Migraine Steering Comm Efficacy of Zolmitriptan Nasal Spray in Adolescent Migraine Pediatrics, August 1, 2007; 120(2): 390 - 396. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Ahonen, M. L. Hamalainen, M. Eerola, and K. Hoppu A randomized trial of rizatriptan in migraine attacks in children Neurology, October 10, 2006; 67(7): 1135 - 1140. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Andrasik and M. S. Schwartz Behavioral Assessment and Treatment of Pediatric Headache Behav Modif, January 1, 2006; 30(1): 93 - 113. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||