Estimated use of a pediatric emergency department observation unit

Abstract

Study objective: To estimate the use of a pediatric ED observation unit, including the number of anticipated admissions per 10,000 pediatric ED visits per year and the distribution of those admissions by age group, by month, and by time of day.

Methods: Hospital and ED computer records on all ED patients younger than 18 years who were seen during a 2-year period were abstracted for diagnostic, demographic, and time-flow data. We retrospectively reviewed the charts of patients admitted to the hospital and discharged within 24 hours to determine whether discharge in less than 24 hours could have been anticipated and whether the patient could have been cared for in a pediatric ED observation unit. To refine the estimate, we also reviewed the ICD-9 discharge diagnoses of patients who were not admitted to the hospital but spent more than 6 hours in the pediatric ED.

Results: Of 29,667 pediatric ED visits in a 2-year period, 2,940 (10%) resulted in admission. Of 626 patients discharged in less than 24 hours, only 410 met the anticipation and pediatric ED observation unit level of care criteria. Patients younger than 4 years represented 43% of potential observation unit patients; those aged 16 and 17 years represented 15%. Potential use of an observation unit varied throughout the year. Admission occurred between 3 and 11:59 PM in 60% of the patients. Only 20% of the 176 patients who were not admitted to the hospital but spent more than 6 hours in the pediatric ED were estimated to be candidates for a pediatric ED observation unit.

Conclusion: On the basis of these data, approximately 150 patients per 10,000 each year who visit the University of Virginia pediatric ED would be likely to use an observation unit. Staffing and facility use would be seasonally uneven and would be required during the busiest part of the day. Furthermore, even in a pediatric ED large enough to admit 365 pediatric ED observation unit patients each year, random daily variation in demand means that a single bed would be inadequate 25% of the time and empty 37% of the time. Optimal use of even a single-bed pediatric ED observation unit would not occur until pediatric ED census exceeded 30,000 to 40,000 visits annually.