- Email: [email protected]
Trends in CT Utilization for Pediatric Fall Patients in US Emergency Departments
Original Investigations
Trends in CT Utilization for Pediatric Fall Patients in US Emergency Departments Varun Shahi, BS, Waleed Brinjikji, MD, Harry J. Cloft, MD, PhD, Kristen B. Thomas, MD, David F. Kallmes, MD Rationale and Objectives: Falls are a common cause of emergency department (ED) visits in the United States. We evaluated trends in computed tomography (CT) utilization for pediatric fall victims in the United States from 2001 to 2010. Materials and Methods: Using the National Hospital Ambulatory Medical Care Survey from 2001 to 2010, we identified all visits of pediatric (aged <18 years) patients presenting to EDs after falls. This database surveys approximately 500 EDs per year for 4 weeks providing national estimates on ED resource utilization and outcomes. We studied trends in CT utilization and proportion of visits with life-threatening conditions after falls. We also studied the association between CT utilization rates and demographic characteristics and admission status. Results: A total of 9763 unweighted observations for a total of 32,432,686 pediatric fall patients were seen in US EDs from 2001 to 2010. The proportion of pediatric fall patients receiving CT increased from 5.3% in 2001 to a peak of 16.6% in 2009 and decreased to 11.3% in 2010, whereas the proportion of pediatric fall patients with life-threatening conditions fluctuated between 1.2% and 3.3% during this period. In multivariate logistic regression analysis, each increasing year was independently associated with CT utilization (odds ratio [OR], 1.15; 95% confidence interval [CI], 1.14–1.16). Patients aged 0–1 years had higher odds of CT utilization than patients aged 13–17 years (OR, 2.27; 95% CI, 2.26–2.27). Conclusions: There was a twofold increase in CT utilization among pediatric fall visits from 2001 to 2010. When controlling for demographic and clinical variables, increasing year was independently associated with CT utilization. These findings suggest that CT may be overutilized among pediatric fall patients. Key Words: Computed tomography; emergency department; pediatric; falls; utilization. ªAUR, 2015
C
omputed tomography (CT) utilization in the pediatric population has seen a rapid increase over the past decade in pediatric emergency departments (EDs) (1). It is estimated that at least 4 million CT scans are conducted each year on children in the United States (2). Diagnostic capability, wider availability, and prompt results make CT a preferred method of imaging in EDs across the United States (2–6). Falls account for the most ED visits in the pediatric population with relatively high rates of morbidity and mortality (7–10). Recent studies indicate that nearly a quarter of all CTs ordered are inappropriate, thus subjecting patients to unnecessary radiation exposure (11–14). This is particularly concerning in the pediatric population, placing them at a
Acad Radiol 2015; -:1–6 From the Mayo Medical School, College of Medicine, Mayo Clinic, Rochester, Minnesota (V.S.); and Department of Radiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (W.B., H.J.C., K.B.T., D.F.K.). Received November 16, 2014; accepted February 18, 2015. Conflicts of Interest: The authors declare that they have no conflicts and financial disclosures to report. Address correspondence to: W.B. e-mail: [email protected] ªAUR, 2015 http://dx.doi.org/10.1016/j.acra.2015.02.016
potential risk for cancer, a risk that is cumulative over the patient’s lifetime (15–17). Furthermore, unnecessary imaging is a significant contributor to the rising cost of health care in the United States (18). Because of these issues of safety and inappropriate use and cost, we undertook a study to evaluate CT utilization trends in the pediatric population and study the associated demographic and clinical features.
MATERIAL AND METHODS Patient Population
This was a cross-sectional study using the ED component of the National Hospital Ambulatory Medical Care Survey (NHAMCS) from 2001 to 2010. The NHAMCS is a nationally representative survey of hospital outpatient and ED encounters in the United States administered by the National Center for Health Statistics, Centers for Disease Control and Prevention. Data are collected by trained interviewers who survey EDs using patient record forms. Patient record forms are obtained for a systematic random sample of patient visits during a randomly assigned 4-week reporting period. Data are collected on demographic characteristics of patients, payer 1
Academic Radiology, Vol -, No -, - 2015
SHAHI ET AL
status, patient complaints and diagnoses, diagnostic and screening services, procedures, medication therapy, cause of injury, and other variables. Further details regarding the data collection process and estimation procedures are provided at http://www.cdc.gov/nchs/ahcd/ahcd_questionnaires.htm# methodology. NHAMCS uses a complex design to sample ED visits in nonfederal hospitals. A weight is applied to each ED visit to create a nationwide estimate. Because this is based on probabilistic sampling of select visits, the values in our study should be interpreted as estimates rather than true values. In our data set, the mean weight per patient visit was 3345 (standard deviation, 2808) and the median was 2480 (interquartile range [IQR], 1462–4243). Thus, each visit in the database was estimated to reflect, on average, 3345 nationwide ED visits. Patients aged <18 years who presented to the ED with one or more reasons for visit or mechanism of injury were queried from the database and included in our study. A text search for the word ‘‘fall’’ in the reason for visit and mechanism of injury columns was performed. Any patients aged <18 years with a reason for visit or mechanism of injury of ‘‘fall’’ were included. We abstracted the following information on each patient: gender, age, ethnicity (white, black, Hispanic, and other), insurance status (Medicare, private, Medicaid, and self-pay), life-threatening injury (yes and no), arrival method (ambulance and no ambulance), and whether the patient was admitted. Our patients were broken down into the following groups depending on age: 0–1, 2–5, 6–12, and 13–17 years. Outcomes
We were particularly interested in the utilization of CT in the pediatric population from 2001 to 2010 and the correlation between the diagnosis of life-threatening injuries and the ratio of ED visits in which CTs were used. Through the utilization of International Classification of Diseases, Ninth Revision codes, we classified the following conditions under life-threatening injuries: cardiac tamponade (423.3), tension pneumothorax (512.0), liver lacerations (864.xx), spine fractures (805.x–806.x), skull fracture (800.x–802.xx, 803.x– 804.xx), spleen lacerations (865.xx), and intracranial hemorrhage (852.xx–843.xx). We studied the variables associated with CT utilization between 2001 and 2010 and the relationship between CT utilization, stratified by each age group, and the ratio of patients presenting with life-threatening injuries during this time. The NHAMCS has changed its classification and the coding of CT scans during the study period between 2001 and 2010. Particularly, magnetic resonance imaging (MRIs) and CTs were coded together from 2001 to 2004, and we resolved this conflict by making the assumption that CTs and MRIs for all pediatric patients were CTs. This assumption was based on the fact that from 2006 to 2010, 96.1% of cross-sectional imaging performed among patients in our database were CTexaminations and 3.9% were MRIs. During the period between 2007 and 2010, CTs were further classified as ‘‘CT head’’ and ‘‘CT other than head’’. 2
Statistical Analysis
We derived annual estimates of total ED visits related to falls in our pediatric population which were calculated using sample design and weighting variables from the NHAMCS. Demographic variables were independently studied to look for a relationship between CT utilization using a multivariate logistic regression analysis and included gender, age, year (modeled as a continuous variable), ethnicity, hospital admission, insurance status, arrival method to emergency room (ambulance vs. no ambulance), and the presence of a lifethreatening condition. Furthermore, Student t test was used for comparison of continuous variables, whereas chi-squared testing was used for comparison of categorical variables. JMP 10.0 (SAS Institute Inc., NC, USA) was used for statistical analysis (www.jmp.com). RESULTS Patient Population
A total of 9763 unweighted observations for a total of 32,432,686 pediatric fall patients were seen in US EDs from 2001 to 2010. The number of fall patients fluctuated between a minimum of 2,876,799 patients in 2002 and a maximum of 3,665,240 patients in 2010. Median patient age was 7 years (IQR, 3–12); 4,716,880 patients (14.4%) were 0–1 years; 9,238,272 patients (28.3%) were 2–5 years; 11,379,168 patients (34.8%) were 6–12 years; 7,324,462 patients (22.4%) were 13–17 years; 13,192,248 patients (40.4%) were female; 22,407,019 patients (68.6%) were white; 16,719,694 patients (52.8%) had private insurance; 1,310,740 patients (6.8%) arrived to the ER in an ambulance; and 549,714 patients (1.7%) were admitted to the hospital. These data are summarized in Table 1. Trends in CT Utilization and Hospital Admission
Figure 1 summarizes trends in hospital admission rates and CT utilization of pediatric fall patients and by age group. The proportion of pediatric fall patients receiving CT increased from 5.3% in 2001 to a peak of 16.6% in 2009 and decreased to 11.3% in 2010, whereas the proportion of pediatric fall patients with life-threatening conditions fluctuated between 1.2% and 3.3% during this period. The proportion of patients aged 0–1 years receiving CT after a fall increased from 9.4% in 2001 to 27.6% in 2008 and trended back down to 12.2% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 1.0% and 4.5% during this period. The proportion of fall patients aged 2–5 years receiving CT for fall increased from 3.2% in 2001 to 12.2% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 0.3% and 2.6% during this period. The proportion of fall patients aged 6–12 receiving CT increased from 21.1% in 2001 to 48.1% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 0.8% and 4.1% during this period. The
Academic Radiology, Vol -, No -, - 2015
CT UTILIZATION TRENDS FOR PEDIATRIC FALLS
TABLE 1. Univariate Analysis of Variables Associated with Computed Tomography Imaging
Attribute Age (years) 0–1 2–5 6–12 13–17 Gender Female Male Insurance status Private Medicaid Self-pay Ethnicity White Black Hispanic Other Admission No Yes Life-threatening injury No Yes Arrival method No ambulance Ambulance
Patients Presenting with Fall, N (%) 4,716,880 (14.4) 9,238,272 (28.3) 11,379,168 (34.8) 7,324,462 (22.4)
Presenting with Fall Who Are Also Receiving CT Imaging, N (%) 684,887 (14.5) 804,689 (8.7) 873,404 (7.7) 683,347 (9.3)
Odds Ratio (95% Confidence Interval) 1.56 (1.55–1.56) 0.93 (0.92–0.93) 0.81 (0.81–0.81) Ref
13,192,248 (40.4) 19,466,534 (59.6)
1,260,621 (9.6) 1,785,706 (9.2)
0.96 (0.95–0.96)
16,719,694 (52.8) 10,281,238 (32.5) 2,711,403 (8.6)
1,618,127 (9.7) 953,315 (9.3) 204,450 (7.5)
Ref 0.95 (0.95–0.96) 0.76 (0.76–0.76)
22,407,019 (68.6) 5,631,363 (17.2) 3,397,591 (10.4) 1,222,809 (3.7)
2,094,860 (9.4) 479,026 (8.5) 348,655 (10.3) 123,786 (10.1)
Ref 0.90 (0.90–0.90) 1.11 (1.10–1.11) 1.09 (1.08–1.10)
32,109,068 (98.3) 549,714 (1.7)
2,884,419 (9.0) 161,908 (29.5)
4.23 (4.21–4.26)
31843166 (97.5) 815616 (2.5)
2763996 (8.7) 282331 (34.6)
5.57 (5.54–5.60)
17983261 (93.2) 1310740 (6.8)
1437118 (8.0) 263712 (20.1)
2.90 (2.89–2.91)
Figure 1. All pediatric patients. CT, computed tomography.
proportion of fall patients aged 13–17 receiving CT increased from 6.5% in 2001 to 14.5% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 2.1% and 5.7% during this period. These data are summarized in Figures 2–5. Variables Associated with CT Utilization
Univariate analysis of variables associated with CT utilization is summarized in Table 1. In multivariate logistic
Figure 2. Patients aged <2 years. CT, computed tomography.
regression analysis, each increasing year was independently associated with CT utilization (odds ratio [OR], 1.15; 95% confidence interval [CI], 1.14–1.16). On the adjusted analysis, the odds of CT utilization in 2010 compared to 2001 was 2.21 (95% CI, 2.20–2.22). Patients aged 0–1 years had higher odds of CT utilization than patients aged 13–17 years (OR, 2.27; 95% CI, 2.26–2.27), and the same was true for patients aged 2–5 years (OR, 1.01; 95% CI, 1.01–1.02). Patients aged 6–12 years had lower odds of 3
Academic Radiology, Vol -, No -, - 2015
SHAHI ET AL
TABLE 2. Multivariate Analysis
Attribute
Figure 3. Patients aged 2–5 years. CT, computed tomography.
Age (years) 0–1 2–5 6–12 13–17 Male versus female Insurance status Private Medicaid Self-pay Ethnicity White Black Hispanic Other Arrival in ambulance Admitted versus not admitted Year (unit change) Year (over range) Life-threatening condition
Odds Ratio (95% Confidence Interval)* 2.27 (2.26–2.27) 1.01 (1.01–1.02) 0.92 (0.92–0.93) Ref 0.86 (0.86–0.86) Ref 0.90 (0.89–0.90) 0.55 (0.55–0.55) Ref 0.98 (0.98–0.99) 1.14 (1.14–1.15) 1.14 (1.13–1.15) 2.30 (2.29–2.31) 3.25 (3.22–3.28) 1.15 (1.14–1.16) 2.21 (2.20–2.22) 3.78 (3.76–3.80)
Analysis adjusted for age, gender, insurance status, ethnicity, arrival in ambulance, hospital admission, year, and presence of lifethreatening condition. *All P values <.0001 unless otherwise indicated. Figure 4. Patients aged 6–12 years. CT, computed tomography.
DISCUSSION
Figure 5. Patients aged 13–18 Years. CT, computed tomography.
CT utilization than patients aged 13–17 years (OR, 0.92; 95% CI, 0.92–0.93). Male patients had lower odds of CT utilization than females (OR, 0.86; 95% CI, 0.86–0.86). Medicaid patients (OR, 0.90; 95% CI, 0.89–0.90) and self-pay patients (OR, 0.55; 95% CI = 0.55–0.55) had lower odds of CT utilization than private patients. Patients arriving in an ambulance had higher odds of CT utilization than patients arriving through self-transport (OR, 2.30; 95% CI, 2.29–2.31). Patients with life-threatening conditions had higher odds of CT utilization than those without (OR, 3.78; 95% CI, 3.76–3.80). These data are summarized in Table 2. 4
Our study demonstrated that CTutilization among pediatric fall patients doubled between 2001 and 2010, although the prevalence of life-threatening conditions remained relatively stable. On our adjusted analysis, each incremental year saw an increase in odds of getting a CT (OR, 1.15). Additionally, we found that the 0–1 year age group had the highest likelihood of getting a CT. This may be due to the fact that physical examination findings in this age group may be less reliable thus leading physicians to depend on CT findings to rule out life-threatening conditions. The 13–17 years age group had the second highest likelihood of getting CT in 2010, perhaps because this group is at high risk of conducting ‘‘risky’’ behaviors (19). Overall, increased utilization of CT imaging in the pediatric population is concerning, especially in the context of wider campaigns to reduce pediatric imaging and radiation exposure. A number of previously published studies have demonstrated increased utilization of CT imaging in the pediatric population (1,20–22). Roudsari et al. (21) used a trauma registry containing 4589 pediatric patients from 1996 to 2006 and found that utilization of repetitive head CT studies for adolescents admitted to an adult level 1 trauma center increased during this period from 2% to 19%. In that study, the greatest increase in imaging utilization occurred in the adolescent age group, similar to our study, with an incident risk ratio of 1.16. Another study used data from the
Academic Radiology, Vol -, No -, - 2015
NHAMCS database from 1995 to 2003 to show that the use of CT in pediatric patients with acute closed head trauma increased from 12.8% to 22.4% during this period (22). Similarly, a study by Adelgais et al. (23) used patient data from a pediatric trauma registry from 2002 to 2011 to show that cervical spine injury in pediatric trauma patients remained stable; however, rate of cervical spine CT, especially in patients originally assessed at general EDs, increased significantly during this period from 6.8% to 42.0%. Increased CT utilization among pediatric patients has been reported for a number of nontraumatic conditions as well. Tsze et al. (24) demonstrated that CT utilization for right lower quadrant pain in pediatric patients increased from 0% to 59.8% from 1996 to 2006 although the frequency of appendicitis diagnosis in children did not see an appreciable increase during this period. One interesting finding from our study was that CTutilization in patients aged <13 years was lower in 2010 than in 2008–2009. Other studies of pediatric CT utilization have demonstrated a similar decline in pediatric CT utilization in the latter half of the decade. For example, Hoshiko et al. (25) showed that CT scans in certain California EDs in children and pregnant women increased initially in 2005 then began to decline in 2008. An article published by Brenner et al. (2) in 2007, and those similar to it, highlighted the harmful effects of radiation exposure from CT use. Thus, this likely suggests that the decline seen in both California EDs noted previously and in our study after 2009 is probably not a random fluctuation but rather a real change in practice. The decline seen may be partly attributed to ‘‘Image Gently’’ campaign (2008) and the growing emphasis on ALARA radiation concept: As Low As Reasonably Achievable both highlighting using as little radiation as possible because of the adverse effects of radiation exposure. Similarly, another study used the Gachon University Gil Hospital Emergency Center Database to show that the utilization of CTs in pediatric ED increased from 2001–2006, a trend that was more pronounced in adolescents, and then declined in 2006 (26). Pediatric radiation exposure from CT is especially harmful because they are potentially at a higher risk of cancer which is cumulative over patient’s lifetime (15). This is complicated by the fact that many ER physicians fail to realize the potential risk of cancer associated with radiation from CTuse. Previous research indicated that before the 2007 article by Brenner et al., only 9% of ER physicians recognized an increased risk of cancer associated with CT scans—a number that is perhaps, quite different now (2,27). A number of studies have examined the role of decision tools in an attempt to curb imaging utilization among pediatric trauma patients. Recently, randomized control trials by Hess et al. and Babl et al. have been initiated to study the effectiveness of decision aids and rules to assist in deciding whether a CT on a child is necessary (28,29). Furthermore, a study conducted by Bressan et al. (30) demonstrates that the validated Pediatric Emergency Care Applied Research Network rule is effective in identifying children
CT UTILIZATION TRENDS FOR PEDIATRIC FALLS
who are at a low risk for traumatic brain injury and thus are not in need for an emergent CT. Additionally, Hershkovitz et al. (31) also advocate a decrease in CT utilization in cases of pediatric intra-abdominal injury because they describe that serial physical examinations are more reliable in diagnosing intra-abdominal injury and only two patients from their cohort of 42 patients required a change in management based off CT results alone. Schachar et al. (32) found that use of clinical decision aids resulted in a 10.9% reduction in CT scans in pediatric patients with head trauma. Limitations
Our study has limitations. The NHAMCS data set used is based on probabilistic sampling of select visits in select EDs. It is possible that the national estimates generated may not accurately represent true values. A number of investigators have raised concerns regarding the accuracy of NHAMCS data and the methodology of data collection (33,34). NHAMCS chart abstraction is performed by a hospitaldesignated employee or a Census Bureau field representative who is not required to have medical background or training. There is no independent review of data collection and no assessment of interobserver agreement on data collection. It is not clear which parts of the medical record are available for data abstractors as well. Thus, it is clear that the data collection process for NHAMCS is far from ideal (33). However, the NHAMCS does have a number of strengths. The NHAMCS includes broad spectrum of hospitals including small community hospitals and large academic medical centers thus giving a complete representation of health care utilization. Validity of the NHAMCS has been confirmed by corroborating findings from this database to other, large data samples (34). Given the limitations and advantages of the NHAMCS, findings from this study should be corroborated with those of single and multi-institutional ED registries. One major limitation of the data collection process in this setting is that pediatric injuries are thought to have substantial seasonal variability (more injuries in warmer months than cooler months). However, data collection at centers is performed only over 4 weeks and at different times at different centers. Thus, we cannot exclude the possibility that data collection was more likely to occur during warmer months than cooler months from one year to another. An assumption was made that life-threatening injuries mostly require CT imaging to show that doubling of CT utilization in pediatric patients between 2001 and 2010 is not likely attributed to an increase in life-threatening injuries during this time. The presence of a short list of ‘‘life-threatening injuries’’ was used as a surrogate for trends in severe injuries over the period. This is not a comprehensive list and there are certainly many indications for imaging beyond the perceived presence of a life-threatening injury. We did not adjust for fall type or fall severity (ie, fall from height vs. fall from chair, etc.). Finally, our study did not include patient outcomes after admission or arrival to the ED. 5
SHAHI ET AL
CONCLUSIONS Our data indicated a significantly greater increase in CT utilization than an increase in life-threatening injuries from 2001 to 2010. Even when controlling for injury severity, the odds of CTutilization from 2001 to 2010 was 2.2 in the pediatric population. The plateau and decline in CT utilization seen after 2009 and in certain age groups such as those aged 0–1 is likely due to the growing awareness of CT overutilization and radiation risks. Further work is necessary to uncover the factors that ultimately contribute to the utilization of CTs in the pediatric ER. This may begin with physician education on the risks of CT utilization to prevent unnecessary scans from taking place. REFERENCES 1. Broder J, Fordham LA, Warshauer DM. Increasing utilization of computed tomography in the pediatric emergency department, 2000-2006. Emerg Radiol 2007; 14(4):227–232. 2. Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. The New England journal of medicine 2007; 357(22): 2277–2284. 3. Broder JS. CT utilization: the emergency department perspective. Pediatr Radiol 2008; 38(Suppl 4):S664–S669. 4. Lee J, Kirschner J, Pawa S, et al. Computed tomography use in the adult emergency department of an academic urban hospital from 2001 to 2007. Ann Emerg Med 2010; 56(6):591–596. 5. Boone JM, Brunberg JA. Computed tomography use in a tertiary care university hospital. J Am Coll Radiol 2008; 5(2):132–138. 6. Sosna J, Slasky BS, Bar-Ziv J. Computed tomography in the emergency department. Am J Emerg Med 1997; 15(3):244–247. 7. Wang MY, Kim KA, Griffith PM, et al. Injuries from falls in the pediatric population: an analysis of 729 cases. J Pediatr Surg 2001; 36(10):1528–1534. 8. Hall JR, Reyes HM, Horvat M, et al. The mortality of childhood falls. J Trauma 1989; 29(9):1273–1275. 9. Kraus JF, Fife D, Cox P, et al. Incidence, severity, and external causes of pediatric brain injury. Am J Dis Child 1986; 140(7):687–693. 10. Mathers LJ, Weiss HB. Incidence and characteristics of fall-related emergency department visits. Acad Emerg Med 1998; 5(11):1064–1070. 11. Lehnert BE, Bree RL. Analysis of appropriateness of outpatient CT and MRI referred from primary care clinics at an academic medical center: how critical is the need for improved decision support? J Am Coll Radiol 2010; 7(3):192–197. 12. Flynn TW, Smith B, Chou R. Appropriate use of diagnostic imaging in low back pain: a reminder that unnecessary imaging may do as much harm as good. J Orthop Sports Phys Ther 2011; 41(11):838–846. 13. Berrington de Gonzalez A, Mahesh M, Kim KP, et al. Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 2009; 169(22):2071–2077. 14. Fazel R, Krumholz HM, Wang Y, et al. Exposure to low-dose ionizing radiation from medical imaging procedures. N Engl J Med 2009; 361(9): 849–857. 15. Frush DP, Donnelly LF, Rosen NS. Computed tomography and radiation risks: what pediatric health care providers should know. Pediatrics 2003; 112(4):951–957.
6
Academic Radiology, Vol -, No -, - 2015
16. Karlsson P, Holmberg E, Lundell M, et al. Intracranial tumors after exposure to ionizing radiation during infancy: a pooled analysis of two Swedish cohorts of 28,008 infants with skin hemangioma. Radiat Res 1998; 150(3): 357–364. 17. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet 2012; 380(9840):499–505. 18. Smith-Bindman R, Miglioretti DL, Larson EB. Rising use of diagnostic medical imaging in a large integrated health system. Health Aff (Millwood) 2008; 27(6):1491–1502. 19. Eaton DK, Kann L, Kinchen S, et al. Youth risk behavior surveillance— United States, 2011. MMWR Surveill Summ 2012; 61(4):1–162. 20. Larson DB, Johnson LW, Schnell BM, et al. Rising use of CT in child visits to the emergency department in the United States, 1995-2008. Radiology 2011; 259(3):793–801. 21. Roudsari B, Moore DS, Jarvik JG. Trend in the utilization of CT for adolescents admitted to an adult level I trauma center. J Am Coll Radiol 2010; 7(10):796–801. 22. Blackwell CD, Gorelick M, Holmes JF, et al. Pediatric head trauma: changes in use of computed tomography in emergency departments in the United States over time. Annals of Emergency Medicine 2007; 49(3): 320–324. 23. Adelgais KM, Browne L, Holsti M, et al. Cervical spine computed tomography utilization in pediatric trauma patients. J Pediatr Surg 2014; 49(2): 333–337. 24. Tsze DS, Asnis LM, Merchant RC, et al. Increasing computed tomography use for patients with appendicitis and discrepancies in pain management between adults and children: an analysis of the NHAMCS. Annals of Emergency Medicine 2012; 59(5):395–403. 25. Hoshiko S, Smith D, Fan C, et al. Trends in CT scan rates in children and pregnant women: teaching, private, public and nonprofit facilities. Pediatr Radiol 2014; 44(5):522–528. 26. Oh HY, Kim EY, Kim JE, et al. Trends of CT use in the pediatric emergency department in a tertiary academic hospital of Korea during 2001–2010. Korean J Radiol 2012; 13(6):771–775. 27. Lee CI, Haims AH, Monico EP, et al. Diagnostic CT scans: assessment of patient, physician, and radiologist awareness of radiation dose and possible risks. Radiology 2004; 231(2):393–398. 28. Babl FE, Lyttle MD, Bressan S, et al. A prospective observational study to assess the diagnostic accuracy of clinical decision rules for children presenting to emergency departments after head injuries (protocol): the Australasian Paediatric Head Injury Rules Study (APHIRST). BMC Pediatr 2014; 14(1):148. 29. Hess EP, Wyatt KD, Kharbanda AB, et al. Effectiveness of the head CT choice decision aid in parents of children with minor head trauma: study protocol for a multicenter randomized trial. Trials 2014; 15(1):253. 30. Bressan S, Romanato S, Mion T, et al. Implementation of adapted PECARN decision rule for children with minor head injury in the pediatric emergency department. Acad Emerg Med 2012; 19(7):801–807. 31. Hershkovitz Y, Zoarets I, Stepansky A, et al. Computed tomography is not justified in every pediatric blunt trauma patient with a suspicious mechanism of injury. Am J Emerg Med 2014; 32(7):697–699. 32. Schachar JL, Zampolin RL, Miller TS, et al. External validation of the New Orleans Criteria (NOC), the Canadian CT Head Rule (CCHR) and the National Emergency X-Radiography Utilization Study II (NEXUS II) for CT scanning in pediatric patients with minor head injury in a non-trauma center. Pediatr Radiol 2011; 41(8):971–979. 33. Cooper RJ. NHAMCS: does it hold up to scrutiny? Annals of Emergency Medicine 2012; 60(6):722–725. 34. McCaig LF, Burt CW, Schappert SM, et al. NHAMCS: does it hold up to scrutiny? Annals of Emergency Medicine 2013; 62(5):549–551.
Trends in CT Utilization for Pediatric Fall Patients in US Emergency Departments Varun Shahi, BS, Waleed Brinjikji, MD, Harry J. Cloft, MD, PhD, Kristen B. Thomas, MD, David F. Kallmes, MD Rationale and Objectives: Falls are a common cause of emergency department (ED) visits in the United States. We evaluated trends in computed tomography (CT) utilization for pediatric fall victims in the United States from 2001 to 2010. Materials and Methods: Using the National Hospital Ambulatory Medical Care Survey from 2001 to 2010, we identified all visits of pediatric (aged <18 years) patients presenting to EDs after falls. This database surveys approximately 500 EDs per year for 4 weeks providing national estimates on ED resource utilization and outcomes. We studied trends in CT utilization and proportion of visits with life-threatening conditions after falls. We also studied the association between CT utilization rates and demographic characteristics and admission status. Results: A total of 9763 unweighted observations for a total of 32,432,686 pediatric fall patients were seen in US EDs from 2001 to 2010. The proportion of pediatric fall patients receiving CT increased from 5.3% in 2001 to a peak of 16.6% in 2009 and decreased to 11.3% in 2010, whereas the proportion of pediatric fall patients with life-threatening conditions fluctuated between 1.2% and 3.3% during this period. In multivariate logistic regression analysis, each increasing year was independently associated with CT utilization (odds ratio [OR], 1.15; 95% confidence interval [CI], 1.14–1.16). Patients aged 0–1 years had higher odds of CT utilization than patients aged 13–17 years (OR, 2.27; 95% CI, 2.26–2.27). Conclusions: There was a twofold increase in CT utilization among pediatric fall visits from 2001 to 2010. When controlling for demographic and clinical variables, increasing year was independently associated with CT utilization. These findings suggest that CT may be overutilized among pediatric fall patients. Key Words: Computed tomography; emergency department; pediatric; falls; utilization. ªAUR, 2015
C
omputed tomography (CT) utilization in the pediatric population has seen a rapid increase over the past decade in pediatric emergency departments (EDs) (1). It is estimated that at least 4 million CT scans are conducted each year on children in the United States (2). Diagnostic capability, wider availability, and prompt results make CT a preferred method of imaging in EDs across the United States (2–6). Falls account for the most ED visits in the pediatric population with relatively high rates of morbidity and mortality (7–10). Recent studies indicate that nearly a quarter of all CTs ordered are inappropriate, thus subjecting patients to unnecessary radiation exposure (11–14). This is particularly concerning in the pediatric population, placing them at a
Acad Radiol 2015; -:1–6 From the Mayo Medical School, College of Medicine, Mayo Clinic, Rochester, Minnesota (V.S.); and Department of Radiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (W.B., H.J.C., K.B.T., D.F.K.). Received November 16, 2014; accepted February 18, 2015. Conflicts of Interest: The authors declare that they have no conflicts and financial disclosures to report. Address correspondence to: W.B. e-mail: [email protected] ªAUR, 2015 http://dx.doi.org/10.1016/j.acra.2015.02.016
potential risk for cancer, a risk that is cumulative over the patient’s lifetime (15–17). Furthermore, unnecessary imaging is a significant contributor to the rising cost of health care in the United States (18). Because of these issues of safety and inappropriate use and cost, we undertook a study to evaluate CT utilization trends in the pediatric population and study the associated demographic and clinical features.
MATERIAL AND METHODS Patient Population
This was a cross-sectional study using the ED component of the National Hospital Ambulatory Medical Care Survey (NHAMCS) from 2001 to 2010. The NHAMCS is a nationally representative survey of hospital outpatient and ED encounters in the United States administered by the National Center for Health Statistics, Centers for Disease Control and Prevention. Data are collected by trained interviewers who survey EDs using patient record forms. Patient record forms are obtained for a systematic random sample of patient visits during a randomly assigned 4-week reporting period. Data are collected on demographic characteristics of patients, payer 1
Academic Radiology, Vol -, No -, - 2015
SHAHI ET AL
status, patient complaints and diagnoses, diagnostic and screening services, procedures, medication therapy, cause of injury, and other variables. Further details regarding the data collection process and estimation procedures are provided at http://www.cdc.gov/nchs/ahcd/ahcd_questionnaires.htm# methodology. NHAMCS uses a complex design to sample ED visits in nonfederal hospitals. A weight is applied to each ED visit to create a nationwide estimate. Because this is based on probabilistic sampling of select visits, the values in our study should be interpreted as estimates rather than true values. In our data set, the mean weight per patient visit was 3345 (standard deviation, 2808) and the median was 2480 (interquartile range [IQR], 1462–4243). Thus, each visit in the database was estimated to reflect, on average, 3345 nationwide ED visits. Patients aged <18 years who presented to the ED with one or more reasons for visit or mechanism of injury were queried from the database and included in our study. A text search for the word ‘‘fall’’ in the reason for visit and mechanism of injury columns was performed. Any patients aged <18 years with a reason for visit or mechanism of injury of ‘‘fall’’ were included. We abstracted the following information on each patient: gender, age, ethnicity (white, black, Hispanic, and other), insurance status (Medicare, private, Medicaid, and self-pay), life-threatening injury (yes and no), arrival method (ambulance and no ambulance), and whether the patient was admitted. Our patients were broken down into the following groups depending on age: 0–1, 2–5, 6–12, and 13–17 years. Outcomes
We were particularly interested in the utilization of CT in the pediatric population from 2001 to 2010 and the correlation between the diagnosis of life-threatening injuries and the ratio of ED visits in which CTs were used. Through the utilization of International Classification of Diseases, Ninth Revision codes, we classified the following conditions under life-threatening injuries: cardiac tamponade (423.3), tension pneumothorax (512.0), liver lacerations (864.xx), spine fractures (805.x–806.x), skull fracture (800.x–802.xx, 803.x– 804.xx), spleen lacerations (865.xx), and intracranial hemorrhage (852.xx–843.xx). We studied the variables associated with CT utilization between 2001 and 2010 and the relationship between CT utilization, stratified by each age group, and the ratio of patients presenting with life-threatening injuries during this time. The NHAMCS has changed its classification and the coding of CT scans during the study period between 2001 and 2010. Particularly, magnetic resonance imaging (MRIs) and CTs were coded together from 2001 to 2004, and we resolved this conflict by making the assumption that CTs and MRIs for all pediatric patients were CTs. This assumption was based on the fact that from 2006 to 2010, 96.1% of cross-sectional imaging performed among patients in our database were CTexaminations and 3.9% were MRIs. During the period between 2007 and 2010, CTs were further classified as ‘‘CT head’’ and ‘‘CT other than head’’. 2
Statistical Analysis
We derived annual estimates of total ED visits related to falls in our pediatric population which were calculated using sample design and weighting variables from the NHAMCS. Demographic variables were independently studied to look for a relationship between CT utilization using a multivariate logistic regression analysis and included gender, age, year (modeled as a continuous variable), ethnicity, hospital admission, insurance status, arrival method to emergency room (ambulance vs. no ambulance), and the presence of a lifethreatening condition. Furthermore, Student t test was used for comparison of continuous variables, whereas chi-squared testing was used for comparison of categorical variables. JMP 10.0 (SAS Institute Inc., NC, USA) was used for statistical analysis (www.jmp.com). RESULTS Patient Population
A total of 9763 unweighted observations for a total of 32,432,686 pediatric fall patients were seen in US EDs from 2001 to 2010. The number of fall patients fluctuated between a minimum of 2,876,799 patients in 2002 and a maximum of 3,665,240 patients in 2010. Median patient age was 7 years (IQR, 3–12); 4,716,880 patients (14.4%) were 0–1 years; 9,238,272 patients (28.3%) were 2–5 years; 11,379,168 patients (34.8%) were 6–12 years; 7,324,462 patients (22.4%) were 13–17 years; 13,192,248 patients (40.4%) were female; 22,407,019 patients (68.6%) were white; 16,719,694 patients (52.8%) had private insurance; 1,310,740 patients (6.8%) arrived to the ER in an ambulance; and 549,714 patients (1.7%) were admitted to the hospital. These data are summarized in Table 1. Trends in CT Utilization and Hospital Admission
Figure 1 summarizes trends in hospital admission rates and CT utilization of pediatric fall patients and by age group. The proportion of pediatric fall patients receiving CT increased from 5.3% in 2001 to a peak of 16.6% in 2009 and decreased to 11.3% in 2010, whereas the proportion of pediatric fall patients with life-threatening conditions fluctuated between 1.2% and 3.3% during this period. The proportion of patients aged 0–1 years receiving CT after a fall increased from 9.4% in 2001 to 27.6% in 2008 and trended back down to 12.2% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 1.0% and 4.5% during this period. The proportion of fall patients aged 2–5 years receiving CT for fall increased from 3.2% in 2001 to 12.2% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 0.3% and 2.6% during this period. The proportion of fall patients aged 6–12 receiving CT increased from 21.1% in 2001 to 48.1% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 0.8% and 4.1% during this period. The
Academic Radiology, Vol -, No -, - 2015
CT UTILIZATION TRENDS FOR PEDIATRIC FALLS
TABLE 1. Univariate Analysis of Variables Associated with Computed Tomography Imaging
Attribute Age (years) 0–1 2–5 6–12 13–17 Gender Female Male Insurance status Private Medicaid Self-pay Ethnicity White Black Hispanic Other Admission No Yes Life-threatening injury No Yes Arrival method No ambulance Ambulance
Patients Presenting with Fall, N (%) 4,716,880 (14.4) 9,238,272 (28.3) 11,379,168 (34.8) 7,324,462 (22.4)
Presenting with Fall Who Are Also Receiving CT Imaging, N (%) 684,887 (14.5) 804,689 (8.7) 873,404 (7.7) 683,347 (9.3)
Odds Ratio (95% Confidence Interval) 1.56 (1.55–1.56) 0.93 (0.92–0.93) 0.81 (0.81–0.81) Ref
13,192,248 (40.4) 19,466,534 (59.6)
1,260,621 (9.6) 1,785,706 (9.2)
0.96 (0.95–0.96)
16,719,694 (52.8) 10,281,238 (32.5) 2,711,403 (8.6)
1,618,127 (9.7) 953,315 (9.3) 204,450 (7.5)
Ref 0.95 (0.95–0.96) 0.76 (0.76–0.76)
22,407,019 (68.6) 5,631,363 (17.2) 3,397,591 (10.4) 1,222,809 (3.7)
2,094,860 (9.4) 479,026 (8.5) 348,655 (10.3) 123,786 (10.1)
Ref 0.90 (0.90–0.90) 1.11 (1.10–1.11) 1.09 (1.08–1.10)
32,109,068 (98.3) 549,714 (1.7)
2,884,419 (9.0) 161,908 (29.5)
4.23 (4.21–4.26)
31843166 (97.5) 815616 (2.5)
2763996 (8.7) 282331 (34.6)
5.57 (5.54–5.60)
17983261 (93.2) 1310740 (6.8)
1437118 (8.0) 263712 (20.1)
2.90 (2.89–2.91)
Figure 1. All pediatric patients. CT, computed tomography.
proportion of fall patients aged 13–17 receiving CT increased from 6.5% in 2001 to 14.5% in 2010, whereas the proportion of fall patients with life-threatening conditions fluctuated between 2.1% and 5.7% during this period. These data are summarized in Figures 2–5. Variables Associated with CT Utilization
Univariate analysis of variables associated with CT utilization is summarized in Table 1. In multivariate logistic
Figure 2. Patients aged <2 years. CT, computed tomography.
regression analysis, each increasing year was independently associated with CT utilization (odds ratio [OR], 1.15; 95% confidence interval [CI], 1.14–1.16). On the adjusted analysis, the odds of CT utilization in 2010 compared to 2001 was 2.21 (95% CI, 2.20–2.22). Patients aged 0–1 years had higher odds of CT utilization than patients aged 13–17 years (OR, 2.27; 95% CI, 2.26–2.27), and the same was true for patients aged 2–5 years (OR, 1.01; 95% CI, 1.01–1.02). Patients aged 6–12 years had lower odds of 3
Academic Radiology, Vol -, No -, - 2015
SHAHI ET AL
TABLE 2. Multivariate Analysis
Attribute
Figure 3. Patients aged 2–5 years. CT, computed tomography.
Age (years) 0–1 2–5 6–12 13–17 Male versus female Insurance status Private Medicaid Self-pay Ethnicity White Black Hispanic Other Arrival in ambulance Admitted versus not admitted Year (unit change) Year (over range) Life-threatening condition
Odds Ratio (95% Confidence Interval)* 2.27 (2.26–2.27) 1.01 (1.01–1.02) 0.92 (0.92–0.93) Ref 0.86 (0.86–0.86) Ref 0.90 (0.89–0.90) 0.55 (0.55–0.55) Ref 0.98 (0.98–0.99) 1.14 (1.14–1.15) 1.14 (1.13–1.15) 2.30 (2.29–2.31) 3.25 (3.22–3.28) 1.15 (1.14–1.16) 2.21 (2.20–2.22) 3.78 (3.76–3.80)
Analysis adjusted for age, gender, insurance status, ethnicity, arrival in ambulance, hospital admission, year, and presence of lifethreatening condition. *All P values <.0001 unless otherwise indicated. Figure 4. Patients aged 6–12 years. CT, computed tomography.
DISCUSSION
Figure 5. Patients aged 13–18 Years. CT, computed tomography.
CT utilization than patients aged 13–17 years (OR, 0.92; 95% CI, 0.92–0.93). Male patients had lower odds of CT utilization than females (OR, 0.86; 95% CI, 0.86–0.86). Medicaid patients (OR, 0.90; 95% CI, 0.89–0.90) and self-pay patients (OR, 0.55; 95% CI = 0.55–0.55) had lower odds of CT utilization than private patients. Patients arriving in an ambulance had higher odds of CT utilization than patients arriving through self-transport (OR, 2.30; 95% CI, 2.29–2.31). Patients with life-threatening conditions had higher odds of CT utilization than those without (OR, 3.78; 95% CI, 3.76–3.80). These data are summarized in Table 2. 4
Our study demonstrated that CTutilization among pediatric fall patients doubled between 2001 and 2010, although the prevalence of life-threatening conditions remained relatively stable. On our adjusted analysis, each incremental year saw an increase in odds of getting a CT (OR, 1.15). Additionally, we found that the 0–1 year age group had the highest likelihood of getting a CT. This may be due to the fact that physical examination findings in this age group may be less reliable thus leading physicians to depend on CT findings to rule out life-threatening conditions. The 13–17 years age group had the second highest likelihood of getting CT in 2010, perhaps because this group is at high risk of conducting ‘‘risky’’ behaviors (19). Overall, increased utilization of CT imaging in the pediatric population is concerning, especially in the context of wider campaigns to reduce pediatric imaging and radiation exposure. A number of previously published studies have demonstrated increased utilization of CT imaging in the pediatric population (1,20–22). Roudsari et al. (21) used a trauma registry containing 4589 pediatric patients from 1996 to 2006 and found that utilization of repetitive head CT studies for adolescents admitted to an adult level 1 trauma center increased during this period from 2% to 19%. In that study, the greatest increase in imaging utilization occurred in the adolescent age group, similar to our study, with an incident risk ratio of 1.16. Another study used data from the
Academic Radiology, Vol -, No -, - 2015
NHAMCS database from 1995 to 2003 to show that the use of CT in pediatric patients with acute closed head trauma increased from 12.8% to 22.4% during this period (22). Similarly, a study by Adelgais et al. (23) used patient data from a pediatric trauma registry from 2002 to 2011 to show that cervical spine injury in pediatric trauma patients remained stable; however, rate of cervical spine CT, especially in patients originally assessed at general EDs, increased significantly during this period from 6.8% to 42.0%. Increased CT utilization among pediatric patients has been reported for a number of nontraumatic conditions as well. Tsze et al. (24) demonstrated that CT utilization for right lower quadrant pain in pediatric patients increased from 0% to 59.8% from 1996 to 2006 although the frequency of appendicitis diagnosis in children did not see an appreciable increase during this period. One interesting finding from our study was that CTutilization in patients aged <13 years was lower in 2010 than in 2008–2009. Other studies of pediatric CT utilization have demonstrated a similar decline in pediatric CT utilization in the latter half of the decade. For example, Hoshiko et al. (25) showed that CT scans in certain California EDs in children and pregnant women increased initially in 2005 then began to decline in 2008. An article published by Brenner et al. (2) in 2007, and those similar to it, highlighted the harmful effects of radiation exposure from CT use. Thus, this likely suggests that the decline seen in both California EDs noted previously and in our study after 2009 is probably not a random fluctuation but rather a real change in practice. The decline seen may be partly attributed to ‘‘Image Gently’’ campaign (2008) and the growing emphasis on ALARA radiation concept: As Low As Reasonably Achievable both highlighting using as little radiation as possible because of the adverse effects of radiation exposure. Similarly, another study used the Gachon University Gil Hospital Emergency Center Database to show that the utilization of CTs in pediatric ED increased from 2001–2006, a trend that was more pronounced in adolescents, and then declined in 2006 (26). Pediatric radiation exposure from CT is especially harmful because they are potentially at a higher risk of cancer which is cumulative over patient’s lifetime (15). This is complicated by the fact that many ER physicians fail to realize the potential risk of cancer associated with radiation from CTuse. Previous research indicated that before the 2007 article by Brenner et al., only 9% of ER physicians recognized an increased risk of cancer associated with CT scans—a number that is perhaps, quite different now (2,27). A number of studies have examined the role of decision tools in an attempt to curb imaging utilization among pediatric trauma patients. Recently, randomized control trials by Hess et al. and Babl et al. have been initiated to study the effectiveness of decision aids and rules to assist in deciding whether a CT on a child is necessary (28,29). Furthermore, a study conducted by Bressan et al. (30) demonstrates that the validated Pediatric Emergency Care Applied Research Network rule is effective in identifying children
CT UTILIZATION TRENDS FOR PEDIATRIC FALLS
who are at a low risk for traumatic brain injury and thus are not in need for an emergent CT. Additionally, Hershkovitz et al. (31) also advocate a decrease in CT utilization in cases of pediatric intra-abdominal injury because they describe that serial physical examinations are more reliable in diagnosing intra-abdominal injury and only two patients from their cohort of 42 patients required a change in management based off CT results alone. Schachar et al. (32) found that use of clinical decision aids resulted in a 10.9% reduction in CT scans in pediatric patients with head trauma. Limitations
Our study has limitations. The NHAMCS data set used is based on probabilistic sampling of select visits in select EDs. It is possible that the national estimates generated may not accurately represent true values. A number of investigators have raised concerns regarding the accuracy of NHAMCS data and the methodology of data collection (33,34). NHAMCS chart abstraction is performed by a hospitaldesignated employee or a Census Bureau field representative who is not required to have medical background or training. There is no independent review of data collection and no assessment of interobserver agreement on data collection. It is not clear which parts of the medical record are available for data abstractors as well. Thus, it is clear that the data collection process for NHAMCS is far from ideal (33). However, the NHAMCS does have a number of strengths. The NHAMCS includes broad spectrum of hospitals including small community hospitals and large academic medical centers thus giving a complete representation of health care utilization. Validity of the NHAMCS has been confirmed by corroborating findings from this database to other, large data samples (34). Given the limitations and advantages of the NHAMCS, findings from this study should be corroborated with those of single and multi-institutional ED registries. One major limitation of the data collection process in this setting is that pediatric injuries are thought to have substantial seasonal variability (more injuries in warmer months than cooler months). However, data collection at centers is performed only over 4 weeks and at different times at different centers. Thus, we cannot exclude the possibility that data collection was more likely to occur during warmer months than cooler months from one year to another. An assumption was made that life-threatening injuries mostly require CT imaging to show that doubling of CT utilization in pediatric patients between 2001 and 2010 is not likely attributed to an increase in life-threatening injuries during this time. The presence of a short list of ‘‘life-threatening injuries’’ was used as a surrogate for trends in severe injuries over the period. This is not a comprehensive list and there are certainly many indications for imaging beyond the perceived presence of a life-threatening injury. We did not adjust for fall type or fall severity (ie, fall from height vs. fall from chair, etc.). Finally, our study did not include patient outcomes after admission or arrival to the ED. 5
SHAHI ET AL
CONCLUSIONS Our data indicated a significantly greater increase in CT utilization than an increase in life-threatening injuries from 2001 to 2010. Even when controlling for injury severity, the odds of CTutilization from 2001 to 2010 was 2.2 in the pediatric population. The plateau and decline in CT utilization seen after 2009 and in certain age groups such as those aged 0–1 is likely due to the growing awareness of CT overutilization and radiation risks. Further work is necessary to uncover the factors that ultimately contribute to the utilization of CTs in the pediatric ER. This may begin with physician education on the risks of CT utilization to prevent unnecessary scans from taking place. REFERENCES 1. Broder J, Fordham LA, Warshauer DM. Increasing utilization of computed tomography in the pediatric emergency department, 2000-2006. Emerg Radiol 2007; 14(4):227–232. 2. Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. The New England journal of medicine 2007; 357(22): 2277–2284. 3. Broder JS. CT utilization: the emergency department perspective. Pediatr Radiol 2008; 38(Suppl 4):S664–S669. 4. Lee J, Kirschner J, Pawa S, et al. Computed tomography use in the adult emergency department of an academic urban hospital from 2001 to 2007. Ann Emerg Med 2010; 56(6):591–596. 5. Boone JM, Brunberg JA. Computed tomography use in a tertiary care university hospital. J Am Coll Radiol 2008; 5(2):132–138. 6. Sosna J, Slasky BS, Bar-Ziv J. Computed tomography in the emergency department. Am J Emerg Med 1997; 15(3):244–247. 7. Wang MY, Kim KA, Griffith PM, et al. Injuries from falls in the pediatric population: an analysis of 729 cases. J Pediatr Surg 2001; 36(10):1528–1534. 8. Hall JR, Reyes HM, Horvat M, et al. The mortality of childhood falls. J Trauma 1989; 29(9):1273–1275. 9. Kraus JF, Fife D, Cox P, et al. Incidence, severity, and external causes of pediatric brain injury. Am J Dis Child 1986; 140(7):687–693. 10. Mathers LJ, Weiss HB. Incidence and characteristics of fall-related emergency department visits. Acad Emerg Med 1998; 5(11):1064–1070. 11. Lehnert BE, Bree RL. Analysis of appropriateness of outpatient CT and MRI referred from primary care clinics at an academic medical center: how critical is the need for improved decision support? J Am Coll Radiol 2010; 7(3):192–197. 12. Flynn TW, Smith B, Chou R. Appropriate use of diagnostic imaging in low back pain: a reminder that unnecessary imaging may do as much harm as good. J Orthop Sports Phys Ther 2011; 41(11):838–846. 13. Berrington de Gonzalez A, Mahesh M, Kim KP, et al. Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 2009; 169(22):2071–2077. 14. Fazel R, Krumholz HM, Wang Y, et al. Exposure to low-dose ionizing radiation from medical imaging procedures. N Engl J Med 2009; 361(9): 849–857. 15. Frush DP, Donnelly LF, Rosen NS. Computed tomography and radiation risks: what pediatric health care providers should know. Pediatrics 2003; 112(4):951–957.
6
Academic Radiology, Vol -, No -, - 2015
16. Karlsson P, Holmberg E, Lundell M, et al. Intracranial tumors after exposure to ionizing radiation during infancy: a pooled analysis of two Swedish cohorts of 28,008 infants with skin hemangioma. Radiat Res 1998; 150(3): 357–364. 17. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet 2012; 380(9840):499–505. 18. Smith-Bindman R, Miglioretti DL, Larson EB. Rising use of diagnostic medical imaging in a large integrated health system. Health Aff (Millwood) 2008; 27(6):1491–1502. 19. Eaton DK, Kann L, Kinchen S, et al. Youth risk behavior surveillance— United States, 2011. MMWR Surveill Summ 2012; 61(4):1–162. 20. Larson DB, Johnson LW, Schnell BM, et al. Rising use of CT in child visits to the emergency department in the United States, 1995-2008. Radiology 2011; 259(3):793–801. 21. Roudsari B, Moore DS, Jarvik JG. Trend in the utilization of CT for adolescents admitted to an adult level I trauma center. J Am Coll Radiol 2010; 7(10):796–801. 22. Blackwell CD, Gorelick M, Holmes JF, et al. Pediatric head trauma: changes in use of computed tomography in emergency departments in the United States over time. Annals of Emergency Medicine 2007; 49(3): 320–324. 23. Adelgais KM, Browne L, Holsti M, et al. Cervical spine computed tomography utilization in pediatric trauma patients. J Pediatr Surg 2014; 49(2): 333–337. 24. Tsze DS, Asnis LM, Merchant RC, et al. Increasing computed tomography use for patients with appendicitis and discrepancies in pain management between adults and children: an analysis of the NHAMCS. Annals of Emergency Medicine 2012; 59(5):395–403. 25. Hoshiko S, Smith D, Fan C, et al. Trends in CT scan rates in children and pregnant women: teaching, private, public and nonprofit facilities. Pediatr Radiol 2014; 44(5):522–528. 26. Oh HY, Kim EY, Kim JE, et al. Trends of CT use in the pediatric emergency department in a tertiary academic hospital of Korea during 2001–2010. Korean J Radiol 2012; 13(6):771–775. 27. Lee CI, Haims AH, Monico EP, et al. Diagnostic CT scans: assessment of patient, physician, and radiologist awareness of radiation dose and possible risks. Radiology 2004; 231(2):393–398. 28. Babl FE, Lyttle MD, Bressan S, et al. A prospective observational study to assess the diagnostic accuracy of clinical decision rules for children presenting to emergency departments after head injuries (protocol): the Australasian Paediatric Head Injury Rules Study (APHIRST). BMC Pediatr 2014; 14(1):148. 29. Hess EP, Wyatt KD, Kharbanda AB, et al. Effectiveness of the head CT choice decision aid in parents of children with minor head trauma: study protocol for a multicenter randomized trial. Trials 2014; 15(1):253. 30. Bressan S, Romanato S, Mion T, et al. Implementation of adapted PECARN decision rule for children with minor head injury in the pediatric emergency department. Acad Emerg Med 2012; 19(7):801–807. 31. Hershkovitz Y, Zoarets I, Stepansky A, et al. Computed tomography is not justified in every pediatric blunt trauma patient with a suspicious mechanism of injury. Am J Emerg Med 2014; 32(7):697–699. 32. Schachar JL, Zampolin RL, Miller TS, et al. External validation of the New Orleans Criteria (NOC), the Canadian CT Head Rule (CCHR) and the National Emergency X-Radiography Utilization Study II (NEXUS II) for CT scanning in pediatric patients with minor head injury in a non-trauma center. Pediatr Radiol 2011; 41(8):971–979. 33. Cooper RJ. NHAMCS: does it hold up to scrutiny? Annals of Emergency Medicine 2012; 60(6):722–725. 34. McCaig LF, Burt CW, Schappert SM, et al. NHAMCS: does it hold up to scrutiny? Annals of Emergency Medicine 2013; 62(5):549–551.