Computed Tomography Use in Children With Minor Head Trauma Presenting to 21 Community Emergency Departments Within an Integrated Health-Care System


Judy Shan1; E Margaret Warton, MPH1; Mary E Reed, DrPH1; David R Vinson, MD1,2; Nathan Kuppermann, MD, MPH3; Peter S Dayan, MD4; Stuart R Dalziel, MD5; Adina S Rauchwerger, MPH1; Dustin W Ballard, MD, MBE1,6

Perm J 2021;25:21.096  E-pub: 04/05/2022

Volume 26, Issue 1

Introduction: Decreasing unnecessary cranial computed tomography (CT) use in pediatric head trauma patients remains important for emergency departments (EDs) across the US. Our study evaluated CT use in children with minor blunt head trauma in 21 community EDs within an integrated health-care system.

Methods: We studied all children younger than 18 years old presenting to 21 community EDs between 2016 through 2018 with acute minor blunt head trauma, defined by an algorithm of ED chief complaints and diagnoses. We excluded patients with traumatic brain injuries diagnosed in the prior year, a CT within 24 hours prior to the ED visit, or an ED Glasgow Coma Scale score of less than 14.

Results: Among 39,792 pediatric minor head trauma ED visits, the aggregate CT use proportion across all EDs was 12.9% [95% confidence interval (CI), 12.6–13.3%; facility-level range, 5.4–21.6%]. The 7 facilities that had previously received a clinical decision support system intervention implementing the Pediatric Emergency Care Applied Research Network rules during 2013 through 2014 had an aggregate mean CT ordering rate of 11.2% (95% CI, 10.7–11.7%; facility-level range, 5.4–14.3%) compared to 14.1% (95% CI, 13.6–14.5%; facility-level range, 7.3–21.6%) for the nonintervention facilities.

Conclusion: CT use for children with minor blunt head trauma in the community EDs of an integrated health-care system was low and stable across facilities from 2016 through 2018. This may be indicative of the safe stewardship of resources in the system, including the absence of financial or medicolegal incentives to scan very low-risk patients as well the availability of resources for close patient follow-up.


More than 800,000 emergency department (ED) visits related to pediatric blunt head trauma were reported in the US in 2014, with roughly 50% of these children undergoing cranial computed tomography (CT) imaging.1–3 To reduce the risk of malignancy associated with ionizing radiation exposure, considerable efforts have been made to decrease unnecessary imaging in children. In 2009, the Pediatric Emergency Care Applied Research Network (PECARN) developed and published traumatic brain injury (TBI) prediction rules, which identify children who are at very low risk of clinically important TBI and can safely forgo CT imaging.2 Since then, several studies have demonstrated that the implementation of clinical decision support (CDS) based on the PECARN rules can result in modest and safe decreases in CT use for children presenting to EDs with minor blunt head trauma.4–8 However, cranial CT use in North American community ED settings remains highly variable, with reports of use rates ranging anywhere from 15% to 70%.3,9–11

Our observational study aimed to evaluate CT use within a large integrated health-care system that emphasizes responsible stewardship of medical resources and has distinctive features that make it well poised to achieve a safe floor of CT use in the pediatric head injury population. These features include a largely capitated payment model, an integrated system that allows for close patient follow-up, stable ED physician staffing models, region-wide emphasis on iterative feedback to physicians on imaging practices, and a comprehensive electronic health record (EHR) that supports CDS tools (7 of the 21 EDs had previously received a CDS system intervention implementing the PECARN rules during 2013 and 2014).11 This CDS tool remained available but was not promoted actively at the 7 sites for the duration of this study.


We performed a retrospective observational study of pediatric (< 18 years old) minor blunt head trauma encounters from January 2016 to December 2018 across 21 community EDs within Kaiser Permanente Northern California (KPNC). KPNC is a private, nonprofit integrated health-care system that covers 4.4 million members, or approximately one-third of the region’s population. KPNC members are comparable to the surrounding population with respect to age, gender, race, and ethnicity.12 All care facilities (emergency, outpatient, inpatient) within KPNC use the same comprehensive integrated EHR (Epic, Verona, WI).13

To define our study patient population, we used a novel hierarchical algorithm based on previously validated “groupers” of ED chief complaints and diagnoses (Supplemental Figure S1).2,14 The criteria were informed and refined through iterative medical record review. Prior retrospective work in our system had derived a grouper of 13 ED chief complaints with 86% sensitivity and 90% specificity for an ED head trauma diagnosis (unpublished data using previously described methods).14 In our current study, we refined this grouper to define our head trauma inclusion criteria with the goal of capturing children with minor blunt head trauma who would likely be eligible for application of the PECARN rule. If a patient had a head trauma diagnosis or a chief complaint of head trauma or head laceration, they were included directly in the study cohort. For other encounters, we screened for alternative validated combinations of ED chief complaints and established two main groupers for alternative chief complaints: a mechanism grouper and a symptom grouper with an associated inclusion algorithm (Supplemental Figure S1). Effectively, this algorithm excluded patients without head trauma-related diagnoses from the ED encounter if their only chief complaint was headache (a symptom grouper), but included those with headache who also had a mechanism grouper or additional symptom grouper.

Because our patient population of interest did not include those with severe, prior, or subacute head trauma, we excluded those with a TBI diagnosis in the prior year, a cranial CT for any reason within 24 hours prior to the ED visit, or any documented Glasgow Coma Scale (GCS) scores < 14 (from nursing flowsheet data). We also excluded patients transferred in from other facilities. We compared CT use rates between facilities that had previously received a CDS system intervention and those that had not, and tested for statistical significance with a 2-tailed t-test.


Our study included 39,792 pediatric head trauma-related ED visits. We excluded 153 patients with TBI diagnoses in the prior year, 128 patients with a CT scan or claim in the prior 24 hours, and 108 patients with ED GCS scores < 14 (Figure 1). Our cohort consisted of 156 (0.4%) patient encounters with a GCS score of 14 and 8204 (20.6%) patients who were younger than 2 years of age. By year, aggregate CT ordering rates and facility-level ranges were as follows: 13.2% (range, 6.8–22.7%) in 2016, 12.3% (range, 3.5–21.9%) in 2017, and 13.2% (range, 6.0–23.3%) in 2018. Aggregate CT ordering rate across the 21 EDs throughout the study period was 12.9% [95% confidence interval (CI), 12.6–13.3%]. Facility-level ED CT ordering rates across the entire study period ranged from 5.4% to 21.6%. The 7 facilities that had received the CDS system intervention previously during 2013 and 2014 had an aggregate mean CT ordering rate of 11.2% (95% CI, 10.7–11.7%; facility-level range, 5.4–14.3%) compared to 14.1% (95% CI, 13.6–14.5%; facility-level range, 7.3–21.6%) for the nonintervention facilities (difference, 2.9%; 95% CI, 2.2–3.5%; p < 0.00001). The overall study period CT ordering rates for the 2 trauma sites (sites L and site T) were 14.3% (95% CI, 12.7–15.9%; yearly range, 13.6–14.7%) and 12.4% (95% CI, 11.3–13.5%; yearly range, 11.4–13.0%), respectively (Table 1, Figure 2).


Figure 1. Children with head trauma presenting to 21 community emergency departments, 2016 through 2018 cohort assembly. ADT = arrival, departure, and transfer; CC = chief complaint; CT = computed tomography; dx = diagnosis; ED = emergency department; GCS = Glasgow Coma Scale; LOS = length of stay; TBI = traumatic brain injury.


Figure 2. Cranial computed tomography (CT) use in children with head trauma in 21 community emergency departments (EDs), 2016 through 2018. * = yearly maximum, minimum, and median facility CT use rates among the 21 EDs.

Table 1. Cranial computed tomographic scan rate for children presenting with mild traumatic head injury to 21 emergency departments, 2016 through 2018

Site Year Overall study period
2016 2017 2018
No scans, n CT scans, n Scan rate, % No scans, n CT scans, n Scan rate, % No scans, n CT scans, n Scan rate, % No scans, n CT scans, n Scan rate, %
A 584 120 17.0 562 128 18.6 598 122 16.9 1744 370 17.5
B 361 56 13.4 348 40 10.3 358 52 12.7 1067 148 12.2
C 249 50 16.7 293 39 11.7 252 56 18.2 794 145 15.4
D 595 100 14.4 602 75 11.1 569 66 10.4 1766 241 12.0
E 332 75 18.4 392 90 18.7 371 84 18.5 1095 249 18.5
F 395 94 19.2 381 107 21.9 452 137 23.3 1228 338 21.6
G 346 52 13.1 376 61 14.0 332 70 17.4 1054 183 14.8
H 924 89 8.8 917 82 8.2 930 108 10.4 2771 279 9.2
I 240 24 9.1 244 18 6.9 262 17 6.1 746 59 7.3
J 568 74 11.5 535 80 13.0 578 83 12.6 1681 237 12.4
K 661 79 10.7 678 83 10.9 583 77 11.7 1922 239 11.1
L 502 86 14.6 515 81 13.6 503 87 14.7 1520 254 14.3
M 485 59 10.8 476 69 12.7 446 62 12.2 1407 190 11.9
N 570 167 22.7 583 133 18.6 626 136 17.8 1779 436 19.7
Oa 501 74 12.9 490 56 10.3 493 44 8.2 1484 174 10.5
Pa 509 53 9.4 527 39 6.9 461 50 9.8 1497 142 8.7
Qa 1176 221 15.8 1331 205 13.3 1377 224 14.0 3884 650 14.3
Ra 812 77 8.7 756 100 11.7 765 95 11.0 2333 272 10.4
Sa 231 18 7.2 277 19 6.4 250 20 7.4 758 57 7.0
Ta 1060 157 12.9 974 125 11.4 892 133 13.0 2926 415 12.4
Ua 398 29 6.8 418 15 3.5 374 24 6.0 1190 68 5.4
Grand total 11,499 1754 13.2 11,675 1645 12.4 11,472 1747 13.2 34,646 5146 12.9

aThese sites received a clinical decision support intervention during 2013 and 2014.

CT = computed tomography.


Our observational study revealed CT ordering rates to be low across 21 community EDs in an integrated health-care system over 2016 through 2018. Although intrafacility variation in yearly rates of CT use in our study was small (variation range, 0.3–6.5%), study-long variation between facilities was comparatively large, with CT use rates ranging from a low of 3.5% to a high of 23.3%. This was somewhat surprising given that all sites are part of an integrated delivery system, and use the same EHR, standardized documentation templates, order sets, and physician staffing group. Furthermore, the rate of severe pediatric head trauma is extremely low across the entire system (we found and excluded only 108 patients with GCS scores < 14). Although CT use was lower at PECARN CDS sites in this investigation (11.2% vs 14.1%), prior analyses did not reveal a significant pre-/post-implementation change in CT use at intervention sites compared to control sites, so it is unclear whether the observed difference is a result of the prior implementation or other unmeasured facility-level differences.11

Across the US, there is even greater variation in CT use for children presenting to EDs with minor blunt head trauma, with average reported rates hovering around 50%.2,3 The cross-sectional study of Marin et al.3 of 324,435 pediatric head trauma visits to 848 general EDs found a risk-adjusted median CT use rate of 56% (interquartile range, 46.4–64.7%), with nontrauma centers 10% less likely to perform a CT than trauma centers. The 5-year retrospective study of Mannix et al.10 of 161,319 pediatric minor head injury encounters across 40 pediatric EDs revealed a median imaging rate of 36% (interquartile range, 29–42%; range, 19–58%) and found no significant association between institution-specific rates of serious head injury patients and CT use among minor head injury patients (p = 0.44). Lower CT use rates are found more frequently at sites using computerized CDS and/or with ongoing quality improvement initiatives.11,15

Our study used a unique case ascertainment method. Although previous studies have largely used International Classification of Diseases codes, singular chief complaints, and ED disposition diagnoses to derive their cohorts, we used GCS scores as well as combinations of ED diagnoses and chief complaints to define a minor blunt head trauma cohort.5,7,16 We used previously validated inclusion criteria of chief complaints and diagnostic groupings in our cohort derivation and, rather than excluding patients with missing GCS scores, as has been done frequently with previous studies, we opted simply to exclude any patient with an explicitly documented GCS score that was less than 14.14 We decided to include patients without documented GCS scores because iterations of chart review revealed that many of these patients were otherwise eligible for our study. Hence, by including these patients, we were able to assemble a more robust cohort that better captured the population of interest.

Although many studies discuss and explore the impact of CDS or quality improvement interventions on rates of CT use, there are few studies on trends of CT use in facilities not actively undergoing these initiatives. Our study captures both populations—demonstrating low CT use in settings with and without CDS interventions with rates that are near the floor (10–15%) of those reported in prior studies.11,15,16 Our findings may be explained by the distinctives of our care delivery model, including comparably lower amounts of financial incentives and medicolegal risk to image very low-risk patients, as well the presence of an integrated system that allows for close patient follow-up. Such dynamics may reassure physicians and parents who choose observation or expectant management rather than immediate CT.17 This use pattern is similar to those seen in New Zealand and Australia, where the retrospective review of Wilson et al.18 of 31 EDs (a mix of tertiary, urban/suburban, and rural EDs) revealed head CT use to be less than 10% on average. Our findings may also be explained by the natural diffusion of the PECARN prediction rules into clinical practice, as the evidence mounts for their effectiveness in reducing CT imaging in children with minor blunt head trauma without compromising safety.4–8 The implementation of these rules into a EHR-based decision support tool in a subset of our study cohort may have spurred more rapid diffusion at other sites in the system.


Our study is limited by its observational nature, and thus we were unable to control for specific factors associated with interfacility variation. In addition, we did not assess potentially missed TBI, although a review of medicolegal cases for our health system did not suggest this to be an issue. We were also unable to exclude all populations (such as known or suspected abusive head trauma) that might be excluded in a prospective study. However, the volume of such patients in our system is small and we have no reason to believe there was variation in such patients over time or medical facility. Last, our study sites are part of an integrated health-care system, and thus our observed CT practice patterns may not be generalizable to other community EDs.


Cranial CT use rates for children with minor blunt head trauma were low and stable at 21 community EDs in this integrated health system between 2016 and 2018. This may be indicative of the safe stewardship of resources in the system, including the absence of financial or medicolegal incentives to image very low-risk patients, as well the availability of resources for close patient follow-up. Because a large proportion of pediatric emergency care is provided by community hospitals, it remains important to generate high-quality imaging use data that can inform future quality improvement studies and interventions.

Supplemental Material

Supplemental material is available at

Disclosure Statement

The author(s) have no conflicts of interest to disclose.

Financial Support

This study was supported by the Kaiser Permanente Northern California Community Benefit Program.

Author Affiliations

1 Kaiser Permanente Division of Research, Oakland, CA

2 Kaiser Permanente Roseville, Roseville, CA

3 University of California Davis School of Medicine, Sacramento, CA

4 Columbia University College of Physicians and Surgeons, New York City, NY

5 University of Auckland, Auckland, New Zealand

6 Kaiser Permanente San Rafael, San Rafael, CA

Corresponding Author

Judy Shan (

Author Contributions

Dustin W Ballard, MD, MBE, and David R Vinson, MD, obtained funding. David R Vinson, MD, Nathan Kuppermann, MD, MPH, Peter S Dayan, MD, E Margaret Warton, MPH, and Dustin W Ballard, MD, MBE, conceptualized and designed the study. E Margaret Warton, MPH, acquired and analyzed the data. E Margaret Warton, MPH, and Mary E Reed, DrPH, provided statistical expertise. Dustin W Ballard, MD, MBE, E Margaret Warton, MPH, and Judy Shan interpreted the data. Dustin W Ballard, MD, MBE, and Judy Shan drafted the manuscript. All authors reviewed and revised the manuscript for important intellectual content.


1. Centers for Disease Control and Prevention. Traumatic brain injury & concussion. 2021. Accessed September 23, 2021.

2. Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: A prospective cohort study. Lancet 2009 Oct;374(9696):1160–70. DOI:

3. Marin JR, Weaver MD, Barnato AE, Yabes JG, Yealy DM, Roberts MS. Variation in emergency department head computed tomography use for pediatric head trauma. Acad Emerg Med 2014 Sep;21(9):987–95. DOI:

4. Ballard DW, Kuppermann N, Vinson DR, et al. Implementation of a clinical decision support system for children with minor blunt head trauma who are at nonnegligible risk for traumatic brain injuries. Ann Emerg Med 2019 May;73(5):440–51. DOI:

5. Nigrovic LE, Stack AM, Mannix RC, et al. Quality improvement effort to reduce cranial CTs for children with minor blunt head trauma. Pediatrics 2015 Jul;136(1):e227–33. DOI:

6. Jennings RM, Burtner JJ, Pellicer JF, et al. Reducing head CT use for children with head injuries in a community emergency department. Pediatrics 2017 Apr;139(4):e20161349–11. DOI:

7. Puffenbarger MS, Ahmad FA, Argent M, et al. Reduction of computed tomography use for pediatric closed head injury evaluation at a non-pediatric community emergency department. Acad Emerg Med 2019 Jul;26(7):784–95. DOI:

8. Miescier MJ, Dudley NC, Kadish HA, Mundorff MB, Corneli HM. Variation in computed tomography use for evaluation of head injury in a pediatric emergency department. Pediatr Emerg Care 2017 Mar;33(3):156–60. DOI:

9. Marin JR, Rodean J, Mannix RC, et al. Association of clinical guidelines and decision support with CT use in pediatric mild traumatic brain injury. J Pediatr 2021 Apr;21. DOI:

10. Mannix R, Meehan WP, Monuteaux MC, Bachur RG. Computed tomography for minor head injury: Variation and trends in major United States pediatric emergency departments. J Pediatr 2012 Jan;160(1):136–9.e1. DOI:

11. Dayan PS, Ballard DW, Tham E, et al. Use of traumatic brain injury prediction rules with clinical decision support. Pediatrics 2017 Apr;139(4):e20162709. DOI:

12. Gordon N, Lin T. The Kaiser Permanente Northern California adult member health survey. Perm J 2016;20(4):15–225. DOI:

13. Bornstein S. An integrated EHR at Northern California Kaiser Permanente: Pitfalls, challenges, and benefits experienced in transitioning. Appl Clin Inform 2012 Aug;3(3):318–25. DOI:

14. Deakyne SJ, Bajaj L, Hoffman J, et al. Development, evaluation and implementation of chief complaint groupings to activate data collection: A multi-center study of clinical decision support for children with head trauma. Appl Clin Inform 2015 Aug;6(3):521–35. DOI:

15. Jennings RM, Burtner JJ, Pellicer JF, et al. Reducing head CT use for children with head injuries in a community emergency department. Pediatrics 2017 Apr;139(4):e20161349. DOI:

16. Stiell IG, Wells GA, Vandemheen K, et al. Variation in ED use of computed tomography for patients with minor head injury. Ann Emerg Med 1997 Jul;30(1):14–22. DOI:

17. Bellolio MF, Bellew SD, Sangaralingham LR, et al. Access to primary care and computed tomography use in the emergency department. BMC Health Serv Res 2018 Mar;18(1):154. DOI:

18. Wilson CL, Tavender EJ, Phillips NT, et al. Variation in CT use for paediatric head injuries across different types of emergency departments in Australia and New Zealand. Emerg Med J 2020 Nov;37(11):686–9. DOI:

Keywords: clinical decision support, computed tomography, emergency department, head trauma, pediatric, traumatic brain injury


Click here to join the eTOC list or text ETOC to 22828. You will receive an email notice with the Table of Contents of The Permanente Journal.


2 million page views of TPJ articles in PubMed from a broad international readership.


Indexed in MEDLINE, PubMed Central, EMBASE, EBSCO Academic Search Complete, and CrossRef.




ISSN 1552-5775 Copyright © 2021

All Rights Reserved