PEDIATRICS Vol. 122 No. 5 November 2008, pp. e1001-e1005 (doi:10.1542/peds.2008-1159)
ARTICLE |
Neuroimaging-Use Trends in Nonacute Pediatric Headache Before and After Clinical Practice Parameters
a Sections of Neurology
b Medical Research
c Department of Radiology, Children's Mercy Hospitals and Clinics, Kansas City, Missouri
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVES. The objective of this study was to determine trends in diagnostic neuroimaging-use rates in nonacute pediatric headache before and after publication of clinical practice guidelines.
METHODS. Retrospective, cross-sectional analysis was conducted of neuroimaging rates for 725 children and adolescents who were aged 3 to 18 years with nonacute headache and normal neurologic examination and were evaluated in a single pediatric neurology clinic during study years 1992, 1996, 2000, and 2004. Following recommendations of current practice parameters, patients with conditions that justify consideration for neuroimaging (eg, progressive headache, abnormal neurologic examination) were excluded from this analysis. We recorded the origin of any neuroimaging request at the time of the clinic visit and any abnormal neuroimaging findings that led to major clinical consequences.
RESULTS. Overall, the mean rate of neuroimaging for patients with nonacute headache was 45%. Use rates remained steady during the 13-year study period (range: 41%–47%). The majority of neuroimaging studies were ordered originally by primary care providers. The proportion of neuroimaging studies that were ordered by primary care providers increased significantly from 1992 to 2004.
CONCLUSIONS. In the evaluation of patients who had nonacute pediatric headache and were referred to a child neurology clinic, neuroimaging-use rates remained stable during the past decade. An increasing proportion of neuroimaging studies are ordered by primary care providers. The influence of evidence-based medicine on medical decision-making may be partly responsible for curbing increases in neuroimaging overuse. The perceived value of neuroimaging by physicians and consumers deserves ongoing study.
Key Words: Diagnostic procedures • headache • neuroimaging • practice parameters/guidelines • practice-based research
Abbreviations: CT—computed tomography
Imaging enhances the practice of medicine by providing supplemental anatomic information in the evaluation of certain signs and symptoms. Indications for diagnostic neuroimaging in any neurologic condition are determined by multiple clinical criteria, including symptom severity, chronological course, physical findings, and pertinent risk factors. Recommendations for the use of imaging are established in clinical practice guidelines on the basis of the principles of evidence-based medicine.1 In actual practice, physicians combine their individual expertise with evidence-based guidelines in decision-making on behalf of their individual patients.2,3
Headache is a relatively common symptom in children and adolescents. Headache pathogenesis is complex with many potential causes, but clinicians can generally distinguish primary headache (ie, pain without evidence of a serious underlying illness) from secondary headache (ie, pain resulting from a serious pathologic condition). Although most pediatric patients who complain of headache do not have a serious underlying disorder, neuroimaging can be a valuable diagnostic tool for conditions for which headache is coupled with other worrisome signs and symptoms.4 Practice parameters for the evaluation of children and adolescents with recurrent headaches, published in 2002, recommend that diagnostic neuroimaging be considered for children with an abnormal neurologic examination or other historical features that suggest neurologic dysfunction.5 These recommendations agreed with the previous practice parameters for the evaluation of headache in adults, published in 1994 and 2000, but emphasized that obtaining a neuroimaging study on a routine basis is not indicated for children with recurrent headache and a normal neurologic examination.4,6
Nevertheless, both overuse and underuse of neuroimaging have been topics of debate during the past 2 decades, when health care costs, malpractice liability, consumer expectations, and the number of neuroimaging facilities all have dramatically increased. The aim of this study was to assess diagnostic neuroimaging-use trends in a nonacute pediatric headache population before and after the publication of clinical practice parameters. We hypothesized that neuroimaging-use rates increased from 1992 to 2004.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Design and Setting
We performed a retrospective, cross-sectional analysis of children and adolescents who were aged 3 to 18 years and had headache symptoms of sufficient severity to be referred to a pediatric neurology clinic for additional evaluation. We chose 1992, 1996, 2000, and 2004 as study years to ascertain a representative patient sample during (1) an era when both computed tomography (CT) and MRI were commonly used in medical practice and (2) a period before and after the publication of clinical practice guidelines.4–6 We obtained a listing of all medical charts with the appropriate International Classification of Diseases, Ninth Revision diagnosis codes (346.0, 346.1, 346.2, and 784.0) for headaches in each study year. To avoid seasonal bias (eg, an increase in stress-related headache during certain times of the school year), we sampled up to 50 consecutive qualifying new-patient charts from each quarter of each study year beginning the first day of January, April, July, and October.
The study was performed in a single-institution, regional children's hospital that serves a large metropolitan area. During the study period, the child neurology clinic was composed of 4 to 6 child neurologists, a nurse practitioner, and a pediatrician with special interest in pediatric headache. Two of the child neurologists worked continuously in the clinic throughout the entire study period. The clinic was the main resource for child neurology in the study metropolitan region, and there were no other child neurologists or pediatric headache specialists in private practice in the area. All caregivers in the clinic were salaried by the institution. The patient population comprised a mixed combination of Midwest urban, suburban, and rural children divided along all socioeconomic, racial, and health insurance groups. This study was approved by the institutional review board.
Inclusion Criteria
Patient charts were included in the study when the referral was made for outpatient evaluation of recurrent, nonprogressive headache with no recorded evidence of serious underlying medical conditions or remarkable abnormalities on physical examination.
Exclusion Criteria
According the recommendations of the current clinical practice parameter, we identified a priori conditions and risk factors for which "neuroimaging should be considered."5 Specifically, patients were excluded from the study when they had historical evidence of progressive headache, meningitis, stroke, vascular malformation, neurocutaneous syndrome, hydrocephalus, traumatic brain injury, seizures, a significant neurodevelopmental disorder, brain tumor, other cancer, or major illness. Patients were excluded when abnormal neurologic examination findings were documented (eg, focal deficits, signs of increased intracranial pressure, alteration of consciousness, macrocephaly [head circumference >98th percentile], and microcephaly [head circumference <2nd percentile]).
Data Collection
No patients were contacted or interviewed by data gatherers. We recorded the origin of the neuroimaging request at the time of the clinic visit. Historical, physical examination and neuroimaging data were abstracted from the medical chart. Categorically recorded (normal/abnormal) historical data included height, weight, blood pressure, and diet, as well as significant past medical, family, school and social histories; review of systems; and general physical and neurologic examinations. Other recorded data included age, gender, current medications, and family history of migraine, as well as a synopsis of headache chronology, location, severity, and associated symptoms.
In a separate data file, diagnostic neuroimaging data were recorded as CT, MRI, both, or neither. We further analyzed the frequency and types of neuroimaging findings but focused on study years 2000 and 2004 because of easier access to electronic medical charts and picture archive and communication systems neuroimaging. Neuroimaging reports by the neuroradiologist or pediatric radiologist were recorded as "normal," "abnormal," or "remarkable." Abnormal neuroimaging results were defined as pathologic neuroanatomic findings that resulted in a new primary clinical diagnosis, prognosis, or significant change in the patient's clinical management. Remarkable neuroimaging findings were any comments in the summary of the neuroradiology report of unexpected, probable incidental findings that the study child neurologists and pediatric neuroradiologist believed to be unrelated to the presenting headache symptoms.
Charts were abstracted by a medical student and child neurology fellow with close supervision by the senior child neurologist. Abstracted charts of all patients for whom there was a historical question of progressive headache or possible physical examination abnormality were independently reviewed by 2 child neurologists who were blinded to any neuroimaging data. Any abnormal or remarkable neuroimaging findings were reviewed by the pediatric neuroradiologist.
Major Outcome Variables
We assessed the diagnostic neuroimaging-use rates from the 4 study years during the 13-year period and the proportion of normal, remarkable, and abnormal neuroimaging results.
Statistical Analysis
Statistical analyses were conducted by using SPSS 12 (SPSS Inc, Chicago, IL). Median values and associated ranges were reported for nonnormal data. We used 95% confidence intervals where appropriate to summarize the uncertainty associated with sampling error.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Patient Characteristics
Fifty-one percent of the patients were male. The mean age of patients was 10.2 years for boys and 11.4 years for girls. All of the patients in this study had normal weight, blood pressure, and diet. For all patients, the general physical and neurologic examinations, as well as significant past medical, school and social histories, and review of systems, were essentially unremarkable. A family history of migraine was recorded for 74% of the patients’ families, whereas no family headache history was present for 21.5% and no information was available for 4.5% of the individual medical charts. Patient scheduling data that were available for 2004 showed that the mean appointment wait time was 55 days (range: 28–103 days). The number of charts for study year 1992 was lower than other study years as a result of a smaller number of available archived paper charts of patients who matched the study criteria.
Trends in Neuroimaging Use
Of 725 total charts of patients with nonacute headache analyzed, 325 (45%) had at least 1 neuroimaging study. Neuroimaging-use rates ranged between 41% and 47% without discernible trends throughout the 4 main study years of the 13-year period. As in previous studies of neuroimaging for nonacute headache, no patient had abnormal neuroimaging findings that led to a neurosurgical procedure or other major change in diagnosis or clinical outcome. In this study, the majority (75.7%) of neuroimaging tests were ordered by primary care providers. The proportion of test orders that originated from primary care providers increased significantly during the course of this study (odds ratio: 1.40 [95% confidence interval: 1.10–1.78]; P = .006; Table 1).
|
Neuroimaging Results
Of the 400 patient charts selected for evaluation in years 2000 and 2004, 185 (46%) had neuroimaging studies. CT was performed for 51%, MRI was performed for 38%, and 11% of the patients had both CT and MRI studies. Of the 185 patients with neuroimaging studies, 145 (78.4%) of the studies were normal, 40 (21.6%) had remarkable findings, and none was abnormal.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This retrospective, cross-sectional study of neuroimaging use in a stable Midwestern pediatric population with nonacute headache found relatively consistent neuroimaging rates during a 13-year period. These neuroimaging-use data were collected during an era that is noteworthy for routine access to modern diagnostic neuroimaging, the emergence of clinical practice parameters, and marked growth in the number of CT and MRI systems.4–8 The evidence gathered in this study rejects the original study hypothesis that use of diagnostic neuroimaging procedures is increasing. Nonetheless, on the basis of the recommendations of published practice parameters, it is evident that neuroimaging overuse is persistent in the clinical evaluation of nonacute headache.
Although the design of this study used the current practice parameters as a benchmark control, it could not be determined from these observational data whether referring physicians were aware or unaware of the 2002 practice guidelines or which "aware" physicians either followed or ignored the guidelines for any number of plausible reasons (Table 2). Nevertheless, this retrospective study reveals compelling data that deserves additional practice-based comparative effectiveness research. Furthermore, this study did not assess the outcomes of patients with progressive headache or those with abnormal neurologic examination findings for which neuroimaging should be strongly considered. According to the published practice guidelines and our exclusion criteria, all patients who were assessed in this study lacked apparent clinical indications for neuroimaging. As in previous neuroimaging studies for nonacute pediatric headache without additional abnormal signs, this study found no clear pathologic results that led to a new primary clinical diagnosis, surgery, or other major change in patient treatment or prognosis.9–15 Thus, the findings in this study corroborate the conclusions and recommendations of the current clinical practice parameter.5 It is known that primary headache disorders, such as migraine and tension-type headaches, account for the vast majority of pediatric headaches. Neuroimaging is helpful in the diagnosis of secondary headache as a result of serious underlying illness that is reliably accompanied by a shorter symptom interval and other abnormal clinical features by the time of presentation or reevaluation.16,17 It is possible that evidence-based clinical practice parameters have played an important role in educating physicians and in helping to curtail an otherwise increasing trend to use neuroimaging. Future studies will be needed to demonstrate clearly the influences of such practice guidelines.
|
Nonetheless, surveys that were performed after the publication of adult headache practice guidelines indicated that the majority of adult neurologists still use neuroimaging in evaluating severe migraine with a normal neurologic examination.3,4 There are many plausible reasons for overuse of diagnostic tests. In the clinical evaluation of nonacute headache, high rates of neuroimaging studies could result from a combination of factors such as increasing consumer expectations, perceived need for diagnostic certainty, time constraints in primary care practice, proliferation of imaging technology, fear of liability, and misunderstanding of appropriateness guidelines (Table 2).18–20 Individual physicians may find legitimate reasons to disregard practice guidelines after weighing anticipated benefits against potential risks, harms, and costs in individual patient circumstances.
Limitations to this study are mostly related to its retrospective, cross-sectional study design. Multiple variables may influence individual physician judgment and decision analysis around a point of time in a study. These study data were gathered in the sole venue for child neurology consultation in a diverse community, but there was no analysis of cost, patient income, insurance status, patient expenditure, educational status, geographic origin, clinic access time, or appointment time; however, overuse of neuroimaging has been reported in child neurology clinics in other geographic regions, and it is unlikely that the findings reported here are unique to the study hospital in the Midwestern United States.21
These study results cannot project an estimate of neuroimaging-use rates in the larger population of patients who have nonacute headache in primary care practice and are not referred to a local child neurology clinic. Nevertheless, this study shows the odds that primary care provider–ordered neuroimaging for nonacute headache rose from 
2-to-1 in 1992 to 7-to-1 in 2004 (Table 1). These findings strongly suggest a trend of increasing neuroimaging use in the primary care setting. Additional study is needed to evaluate the awareness of practice parameters among primary care and specialist providers as well as the utility of such practice guidelines in various clinical practice settings.
There are numerous concerns about the overuse of neuroimaging. Although the risk of ionizing radiation that are associated with CT scanning is generally considered rather small when compared with the benefits of accurate diagnosis and treatment, recent studies suggested that physicians are not sufficiently aware of the potential long-term health risks that are associated with the radiation from CT imaging, which is highest in children.22–25 Sixty-two percent of the neuroimaging patients in this study had CT alone or CT plus MRI.
As in the application of many technologies, the value of neuroimaging cannot be judged solely on how it is used to diagnose disease in populations of patients. From formal evidence for nonacute headache, the practice of obtaining a neuroimaging study in an otherwise neurologically healthy individual should now seem to have little value; however, clinical studies that are used as a basis for practice parameters generally do not evaluate variables such as patient preference, physician perception of complexity, or medicolegal perspectives of health care standards.26,27 Furthermore, access to medical technology such as diagnostic imaging may have become a proxy for health care quality as measured by patient-perceived value. Reassurance for consumers and physicians might be evolving from the value of examination and clinical judgment into "imaging for problems imagined."28,29 Future studies of neuroimaging use could attempt to measure the perceived value of imaging procedures for various symptoms from the perspective of various health care stakeholders.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We found stable neuroimaging-use rates, rampant neuroimaging overuse, and a shift toward primary care neuroimaging in the evaluation of nonacute pediatric headache during the period 1992–2004. Such health care practice patterns may be influenced by evidence-based clinical practice guidelines as well as other value-based perceptions. Overuse of imaging in health care is a topic of concern for consumers, physicians, payers, and policy makers. The trends, benefits, costs, risks, and perceived value of modern imaging deserve ongoing analysis.
|
ACKNOWLEDGMENTS |
|---|
Support for this study was received from the Katharine B. Richardson Associates Endowment Fund.
We thank Gloria Womelduff, Amy Wolf, and Benjamin C. Williams for assistance with data retrieval.
|
FOOTNOTES |
|---|
Accepted Jul 21, 2008.
Address correspondence to William D. Graf, MD, Children's Mercy Hospitals and Clinics, 2401 Gillham Rd, Kansas City, MO 64108. E-mail: wdgraf{at}cmh.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
Statistical analysis was performed by Dr Simon.
| What's Known on This Subject Neuroimaging recommendations in the evaluation of children and adolescents with recurrent headache have been established in clinical practice guidelines. Little is known about how these guidelines influence neuroimaging use in contemporary clinical practice.
|
| What This Study Adds We found stable neuroimaging-use rates, rampant neuroimaging overuse, and a shift toward primary care neuroimaging in the evaluation of nonacute pediatric headache during 1992–2004. Neuroimaging practice patterns seem to be influenced by evidence-based clinical practice guidelines and other value-based perceptions.
|
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Rosenberg W, Donald A. Evidence based medicine: an approach to clinical problem-solving.
BMJ. 1995;310
(6987):1122
–1126
[Free Full Text] - Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn't.
BMJ. 1996;312
(7023):71
–72
[Free Full Text] - Lipton RB, Bigal ME, Rush SR, et al. Migraine practice patterns among neurologists.
Neurology. 2004;62
(11):1926
–1931
[Abstract/Free Full Text] - Silberstein SD. Practice parameter: evidence-based guidelines for migraine headache (an evidence-based review)—report of the Quality Standards Subcommittee of the American Academy of Neurology.
Neurology. 2000;55
(6):754
–762
[Free Full Text] - Lewis D, Ashwal S, Dahl G, et al. Practice parameter: evaluation of children and adolescents with recurrent headaches—report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.
Neurology. 2002;59
(4):490
–498
[Abstract/Free Full Text] - American Academy of Neurology, Quality Standards Subcommittee. Practice parameter: the utility of neuroimaging in the evaluation of headache in patients with normal neurologic examinations (summary statement).
Neurology. 1994;44
(7):1353
–1354
[Free Full Text] - Rao VM, Parker L, Levin DC, Sunshine J, Bushee G. Use trends and geographic variation in neuroimaging: nationwide Medicare data for 1993 and 1998.
AJNR Am J Neuroradiol. 2001;22
(9):1643
–1649
[Abstract/Free Full Text] - IMV Medical Information Division. Benchmark Report CT 2006. Des Plaines, IL: IMV Medical Information Division; 2006
- Alehan FK. Value of neuroimaging in the evaluation of neurologically normal children with recurrent headache.
J Child Neurol. 2002;17
(11):807
–809
[Abstract/Free Full Text] - Maytal J, Bienkowski RS, Patel M, Eviatar L. The value of brain imaging in children with headaches.
Pediatrics. 1995;96
(3 pt 1):413
–416
[Abstract/Free Full Text] - Medina LS, Pinter JD, Zurakowski D, Davis RG, Kuban K, Barnes PD. Children with headache: clinical predictors of surgical space-occupying lesions and the role of neuroimaging.
Radiology. 1997;202
(3):819
–824
[Abstract/Free Full Text] - Dooley JM, Camfield PR, O'Neill M, Vohra A. The value of CT scans for children with headaches. Can J Neurol Sci. 1990;17 (3):309 –310[Web of Science][Medline]
- Wöber-Bingöl C, Wober C, Prayer D, et al. Magnetic resonance imaging for recurrent headache in childhood and adolescence. Headache. 1996;36 (2):83 –90[CrossRef][Web of Science][Medline]
- Chu ML, Shinnar S. Headaches in children younger than 7 years of age.
Arch Neurol. 1992;49
(1):79
–82
[Abstract/Free Full Text] - Lewis DW, Dorbad D. The utility of neuroimaging in the evaluation of children with migraine or chronic daily headache who have normal neurological examinations. Headache. 2000;40 (8):629 –632[CrossRef][Web of Science][Medline]
- Medina LS, Kuntz KM, Pomeroy SL. Children with headache suspected of having a brain tumor: a cost-effectiveness analysis of diagnostic strategies.
Pediatrics. 2001;108
(2):255
–263
[Abstract/Free Full Text] - Wilne SH, Ferris RC, Nathwani A, Kennedy CR. The presenting features of brain tumours: a review of 200 cases.
Arch Dis Child. 2006;91
(6):502
–506
[Abstract/Free Full Text] - Kassirer JP. Our stubborn quest for diagnostic certainty: a cause of excessive testing. N Engl J Med. 1989;320 (22):1489 –1491[Web of Science][Medline]
- Bodenheimer T. Primary care: will it survive?
N Engl J Med. 2006;355
(9):861
–864
[Free Full Text] - Amis ES. American College of Radiology standards, accreditation programs and appropriateness criteria.
AJR Am J Roentgenol. 2000;174
(2):307
–310
[Free Full Text] - Schor NF. Brain imaging and prophylactic therapy in children with migraine: recommendations versus reality. J Pediatr. 2003;143 (6):776 –779[CrossRef][Web of Science][Medline]
- Brenner D, Elliston C, Hall E, Berdon W. Estimated risks of radiation-induced fatal cancer from pediatric CT.
AJR Am J Roentgenol. 2001;176
(2):289
–296
[Abstract/Free Full Text] - Thomas KE, Parnell-Parmley JE, Haidar S, et al. Assessment of radiation dose awareness among pediatricians. Pediatr Radiol. 2006;36 (8):823 –832[CrossRef][Web of Science][Medline]
- Picano E. Sustainability of medical imaging.
BMJ. 2004;328
(7439):578
–580
[Free Full Text] - Hall EJ. Lessons we have learned from our children: cancer risks from diagnostic radiology. Pediatr Radiol. 2002;32 (10):700 –706[CrossRef][Web of Science][Medline]
- Brown MM, Brown GC, Sharma S. Evidence-Based to Value-Based Medicine. Chicago, IL: American Medical Association; 2005
- Rosoff AJ. Evidence-based medicine and the law: the courts confront clinical practice guidelines. J Health Polit Policy Law. 2001;26 (2):327 –368[Abstract]
- Illes J, Racine E. Imaging or imagining? A neuroethics challenge informed by genetics. Am J Bioeth. 2005;5 (2):5 –18[Medline]
- Howard L, Wessely S, Leese M, et al. Are investigations anxiolytic or anxiogenic? A randomised controlled trial of neuroimaging to provide reassurance in chronic daily headache.
J Neurol Neurosurg Psychiatry. 2005;76
(11):1558
–1564
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
