Nonclinical Factors Associated With Primary Care
Physicians’ Ordering Patterns of Magnetic
Resonance Imaging/Computed Tomography
for Headache
1
Glen R. Couchman, MD, Samuel N. Forjuoh, MD, MPH, DrPH, Mohammad H. Rajab, PhD, MPH
Charles D. Phillips, PhD, MPH, Jihnhee Yu, PhD
Rationale and Objectives. The purpose of ordering an imaging test for headache, which is one of the most common rea-
sons for patient consultation with a primary care physician (PCP), include medical indications as well as patient anxiety
and medico-legal concerns. The impact of nonclinical factors on PCPs’ ordering patterns of magnetic resonance imaging/
computed tomography for patients presenting with a headache was examined.
Materials and Methods. A retrospective record review of all patients treated for headache as their primary complaint by
a PCP from a multispecialty group practice associated with an 180,000-member Health Maintenance Organization during
calendar year 2000 was examined. The practice included 18 clinics. Logistic regression was used to estimate the associa-
tion between physician ordering patterns of imaging tests and patient, physician, and clinic setting characteristics.
Results. Of the total study subjects (n = 4,372), a magnetic resonance imaging/computed tomography imaging test was
ordered for 5.3%. At their first encounter, patient’s gender, site of care, and insurance status under the health care sys-
tem’s Health Maintenance Organization had statistically significant associations with the likelihood of imaging test orders.
Female physicians were less likely to order an imaging test (OR = 0.65; 95% confidence interval, 0.39 –1.08). For subse-
quent encounters, however, site of care was the only nonclinical factor that had a consistent association with the likeli-
hood of having a magnetic resonance imaging/computed tomography ordered.
Conclusion. These data show that a number of nonclinical factors may have been associated with having an imaging test
ordered by a PCP when patients presented with a headache. These factors were more significant when the patient was be-
ing seen for the first time with this complaint.
Key Words. Primary care; headache; health disparities; magnetic resonance imaging (MRI); computed tomography (CT).
©
AUR, 2004
Acad Radiol 2004; 11:735–740
1
From the Departments of Family & Community Medicine (G.R.C., S.N.F.) and Biostatistics (M.H.R., J.Y.), Scott & White Hospital and Clinic, Texas A & M
University System Health Science Center, College of Medicine, Century Square, 1402 West Ave H, Temple, TX 76504; and the School of Rural Public Health
(S.N.F., M.H.R., C.D.P.), Texas A & M University System Health Science Center, College Station, TX. Received 27 January 2004; revision requested February
25; revision received and accepted March 22. Supported by a grant (Principal Investigator, Glen R. Couchman, MD) from the Health Services Research Pro-
gram, a collaborative research program supported by the School of Rural Public Health at the Texas A & M University System Health Science Center, the
Scott and White Foundation, and the Scott and White Health Plan in Temple, TX. Address correspondence to G.R.C.
©
AUR, 2004
doi:10.1016/j.acra.2004.03.050
735