Headaches and Neuroimaging: High Utilization and Costs Despite Guidelines

While most headaches are attributable to benign conditions, patients and physicians are often concerned about intracranial pathologic conditions. However, the yield of significant abnormalities on neuroimaging in patients with chronic headaches is 1% to 3%.^1-3 Given the comparable yield in patients without headaches, multiple guidelines have recommended against routine headache neuroimaging,^4-6 and efforts to improve the efficiency of health care utilization, such as the Choosing Wisely campaign (ABIM [American Board of Internal Medicine] Foundation; http://www.choosingwisely.org), have identified these tests as a target. However, little is known about recent headache neuroimaging utilization, associated expenditures, and temporal trends in the United States.

Institutional review board exemption was obtained from the University of Michigan Health System. We utilized the National Ambulatory Medical Care Survey (NAMCS), a nationally representative survey that uses a 3-stage sampling design (geographic regions, physician practices stratified within specialties, and patient visits within practices) to characterize all outpatient office-based care in the United States. We analyzed all headache visits for patients 18 years or older identified using the Healthcare Cost and Utilization Project (HCUP) Single-level Clinical Classification System (CCS) (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 339.xx, 784.0x, 346.xx, and 307.81). For migraine, ICD-9-CM codes 346.xx were used.

To characterize recent headache neuroimaging utilization, the proportion of headache visits with computed tomography (CT) or magnetic resonance imaging (MRI) ordered from 2007 through 2010 was estimated using descriptive statistics for multiple visit categories: all headache visits, all migraine visits, and visits with a primary diagnosis of headache or migraine. Neuroimaging use (CT or MRI) was directly entered onto the NAMCS survey instrument by physicians or their staff. Neuroimaging payments were determined using the Medicare physician fee schedule. To evaluate for trends over time, we identified headache neuroimaging utilization in years where these tests were directly abstracted onto the NAMCS survey instrument (1995-2000 and 2005-2010) in all headache visits. Survey weights were applied for all analyses.

Of all visits, 88% were by patients younger than 65 years and 78% were by female patients. Most visits were to primary care physicians (54.8%), followed by neurologists (20%), other specialists (12.9%), and nonprimary care generalists (12.4%). Over 4 years, a total of 51.1 million headache visits were identified, including 25.4 million migraine visits. Neuroimaging was obtained in 12.4% (95% CI, 10.5-14.7) of all headache visits and 9.8% (95% CI, 7.4-12.9) of migraine visits (Table). Headache neuroimaging utilization was higher if the headache or migraine diagnosis was listed as the primary diagnosis for the visit. Total neuroimaging expenditures were estimated at $3.9 billion over 4 years, including $1.5 billion from migraine visits. Between 1995 and 2010, neuroimaging utilization increased from 5.1% (95% CI, 2.7%-7.5%) to 14.7% (95% CI, 9.4%-20.0%) of all annual headache visits (Figure) (P < .001 for trend).

In the United States, neuroimaging is frequently ordered during outpatient headache visits (12%), contributes substantial cost (nearly $1 billion in annual costs), and is increasing over time. Since 2000, multiple guidelines have recommended against routine neuroimaging in patients with headaches because a serious intracranial pathologic condition is an uncommon cause.^4-6 Consequently, the magnitude of per-visit neuroimaging use found in this study suggests considerable overuse. Perhaps guidelines have not curbed utilization because patients, as opposed to health care providers, may be the primary drivers of utilization. If so, efforts such as the Choosing Wisely campaign, which seeks to empower patients with knowledge regarding unwarranted testing, may be more effective than guidelines alone. Requiring preauthorization of these costly tests and/or value-based insurance designs that shift the cost burden for costly, low-yield tests to patients are alternative strategies. Given that headache neuroimaging is common, costly, and likely substantially overused, interventions to curb utilization of these tests have the potential to substantially reduce health care expenditures while improving guideline concordance. Therefore, optimizing headache neuroimaging practices should be a major national priority.

Corresponding Author: Brian C. Callaghan, MD, MS, Department of Neurology, University of Michigan Health System, 109 Zina Pitcher Pl, 4021 BSRB, Ann Arbor, MI 48104 (bcallagh@med.umich.edu).

Published Online: March 17, 2014. doi:10.1001/jamainternmed.2014.173.

Author Contributions: Dr Burke had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Callaghan, Kerber, Pace, Burke.

Acquisition of data: Pace, Burke.

Analysis and interpretation of data: Callaghan, Kerber, Pace, Skolarus, Burke.

Drafting of the manuscript: Callaghan, Pace, Burke.

Critical revision of the manuscript for important intellectual content: Callaghan, Kerber, Pace, Skolarus.

Statistical analysis: Callaghan, Burke.

Administrative, technical, and material support: Burke.

Study supervision: Pace.

Conflict of Interest Disclosures: None reported.

Funding/Support: Dr Callaghan is supported by the Katherine Rayner Program, the Taubman Medical Institute, and grant K23NS079417 from the National Institutes of Health, National Institute of Neurological Disorders and Strokes. Dr Kerber is supported by grant R18 HS017690 from the Agency for Healthcare Research and Quality. Dr Skolarus is supported by grant K23NS073685 from the National Institutes of Health, National Institute of Neurological Disorders and Stroke. Dr Burke is supported by grant K08NS082597 from the National Institutes of Health, National Institute of Neurological Disorders and Strokes.

Role of the Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Wade Cooper, MD, University of Michigan, contributed in critical reviewing the manuscript. He received no compensation for his contribution.