Pediatric ED Utilization in a Destination City: Demographic, Clinical, and Temporal Trends

Abstract

Objectives: Seasonal variation in emergency department (ED) visits has been shown for a variety of pediatric conditions, but previous studies have not considered how geographic location may also influence when and why these patients present to the ED. Our study examined the demographic and clinical characteristics as well as the seasonal variation among 3 patient populations (locals, in-state nonlocals, and out-of-state visitors) presenting to our pediatric ED (PED), which is located in a coastal, destination city.

Methods: This was a retrospective chart review of PED visits from June 2014 to June 2019 at the Medical University of South Carolina Children's Hospital, a tertiary care facility located in Charleston, SC. Pediatric ED encounters were divided into 3 groups, depending on the patient's home address: local patients residing in the 3 surrounding metro counties, in-state but nonlocal patients, and out-of-state patients. Demographic and clinical information was abstracted for each visit and compared among the 3 patient groups. Seasonal variation among PED visits was determined by recording the week of the year during which each visit occurred.

Results: Local patients accounted for more than 90% of PED visits with increases in visits from October to April. In-state nonlocal patients presented at consistent rates throughout the year, whereas out-of-state ED utilization peaked significantly during the summer months, Spring Break, Thanksgiving, and Christmas. Our local patient population was majority African American; our in-state nonlocal patients roughly matched our state's racial demographics, and our out-of-state population was predominantly White. Compared with in-state nonlocal patients, our local patients were more likely to present with an infection-related complaint and be diagnosed with lower-acuity conditions such as viral infection, otitis media, upper respiratory infection, cough, fever, and gastroenteritis. In-state nonlocal patients had the highest average triage acuity, more frequently had laboratory tests and imaging ordered, and were more than 4.5 times as likely to be admitted to the hospital compared with our local patients. In-state nonlocal patients were also more likely to present with a psychiatric chief complaint compared with our local patients. Out-of-state patients had a similar overall acuity to local patients but were more likely to have imaging ordered and be diagnosed with injuries such as fractures.

Conclusion: At our institution, local patients, in-state nonlocal patients, and out-of-state patients exhibited 3 distinct patterns of PED utilization. Knowledge of these trends can be used to optimize resource allocation and follow-up planning, particularly for our out-of-state patient population.