Background:

Increasing the use of computed tomography (CT) imaging over the past 10 years has been accompanied by rising concerns about the effects of radiation exposure, especially in children. One approach to limiting radiation exposure is to minimize the number of referrals for CT imaging. Understanding the patterns of use for CT imaging may help to determine opportunities for decreasing utilization. The objective of this study was to characterize the frequency of CT imaging for pediatric patients seen in and admitted from the ED and to determine the role of hospital characteristics in CT imaging practices.

Methods:

Data are from the 2005–2007 National Hospital Ambulatory Medical Care Survey, a nationally representative sample of ED visits. We included all pediatric patient visits (< 18 years of age). We determined the proportion of patients in whom CT imaging was utilized and the most common chief complaints and diagnoses in those patients who received CT imaging Using SUDAAN, we generated national estimates of CT utilization in the ED. We determined the association between weighted frequencies of CT imaging and patient and hospital characteristics using the chi‐square test.

Results:

The sample included 27,052 ED visits by pediatric patients. CT scans were obtained in 5.3% of visits, representing a national estimate of 1.54 million scans annually. Seventeen percent of patients admitted or transferred from the ED had CT scans obtained, with an annual estimate of 295,000 patients nationally. CT imaging was performed in 26% of patients presenting with head injury. 22% of patients with headache, and 15% of patients with abdominal pain. Patient characteristics such as older age (2.5% of infants vs. 9.2% of adolescents, P < 0.001) and being admitted or transferred (17.6% vs. 4.5%, P < 0.001) were associated with obtaining CT scans. Frequency of obtaining CT scans differed by type of hospital (children's hospital, 3.9%, versus general hospital, 5,2%, P < 0.001), There wene no statistically significant differences in weighted frequencies between metropolitan versus nonmetropolitan location (5.4% vs. 4.3%, P = 0.21), not‐for‐profit hospital ownership versus government‐owned versus for‐profit (5.3% vs. 4.5% vs. 5.9%, P = 0.39), or teaching versus nonleaching hospitals (5.9% vs. 5.1%, P = 0.22).

Conclusions:

Being seen in a children's hospital ED versus a general hospital ED is significantly associated with a decreased likelihood of obtaining CT imaging. Efforts to reduce radiation exposure by reducing overuse of CT imaging should include a focus on emergency departments in general hospitals. As many patients who receive imaging in the ED are admitted or transferred, hospitalists may have an important role to play in these efforts.

Author Disclosure:

L. Johnson, none.