Headache 2008 the Authors Journal compilation 2008 American Headache Society

ISSN 0017-8748 doi: 10.1111/j.1526-4610.2008.01156.x Published by Wiley Periodicals, Inc.

Research Submission
Analysis of Headache Management in a Busy Emergency Room in the United States
Soma Sahai-Srivastava, MD; Prakash Desai, DO; Ling Zheng, MD

Objectives.To analyze the demographics, diagnosis, and treatment patterns in patients with headache-seeking treatment in one of the busiest emergency rooms (ER) of an academic medical center in the USA. Background.The past decade has seen tremendous improvement in acute and preventive management of headaches. However, there are very few data on how headache patients are managed by ER doctors. Methods.Retrospective analysis of 100 charts chosen at random for patients with a discharge diagnosis of headache (according to ICD-9 codes) from the University of Southern California + Los Angeles County Hospital ER. Results.The majority of patients were female (74%) and Hispanic (76%) with an age range of 15-68 years. The most common ER diagnoses were migraine (42%) and headache not otherwise specied (headache NOS 42%). Fifty-one percent of patients received a head computerized tomography; 9% received a lumbar puncture. Medications most frequently used for acute treatment, whether migraine or headache NOS, were narcotics (25%), followed by antiemetics (24%), nonsteroidal anti-inammatory drugs (19%), and acetaminophen (17%). Only 5% of migraine patients received migraine-specic triptans in the ER (2% overall). Thirty-one percent of migraineurs were given a prescription for a triptan upon discharge from the ER (17% of all patients). Eighteen percent of patients were admitted to the hospital with secondary headache. The nal diagnosis in the ER matched the diagnosis of the neurologist in 79% of cases with a moderate degree of agreement. Conclusion.Narcotics remain the medications most often chosen for treatment of all acute headaches (including migraine) in the ER. There is very little use of migraine-specic medications in the ER. In addition, neurology consults are underutilized even in an academic setting in the ER. The data suggest a lack of clear standards of care for diagnosis and treatment of headache, especially migraines. Specic guidelines for headache management should be established keeping in mind the unique setting of the ER. Key words: migraine, headache, treatment (Headache 2008;48:931-938)

Headache is a common problem encountered in emergency rooms (ER), accounting for 1-4% of all ER visits, or 2.8 million visits per year in the United States.1,2 It is the fth most common clinical condition leading to ER visits.2 In 2004, there were
From the University of Southern CaliforniaNeurology, Los Angeles, CA, USA. Address all correspondence to S. Sahai-Srivastava, LAC + USC Medical CenterNeurology, 1200 North State St., Room 5640, Los Angeles, CA 90033, USA. Accepted for publication April 7, 2008.

approximately 110 million ER visits in the USA and 2-3% were for headaches.3 One study showed that the vast majority of ER visits for headache are benign primary headaches with approximately 4% of patients having a secondary cause.4 Migraine, one of the more common primary headaches, accounts for up to 800,000 of headache patients in the ER per year.5 Migraineurs are 4 times more likely to visit an ER than nonmigraineurs.6 One study in which 1200 migraine patients were surveyed showed that 1 out of 5 migraineurs had visited the ER for headache in the
Conict of Interest: None

931

932 previous year.7 The largest study conducted on the diagnostic and treatment practices for headache among ER physicians was published by Goldstein et al in 2006.4 Standards of care or guidelines for routine management of migraine are established, but guidelines for headache management specically for ER physicians have not been determined.8 In addition, it is not clear if practices differ in a county hospital ER setting, in which a large volume of patients are likely to be evaluated. It is in this setting that standards of diagnosis care may be most readily established and modied. To begin to address these issues, we conducted a pilot study of headache diagnosis and treatment practices for the Los Angeles County, and University of Southern California (LAC + USC) ER, which is one of the busiest in the country, and is utilized predominantly by immigrant, lower-income, and uninsured patients who do not otherwise have access to health care. Our objectives were to study the demographics, diagnostic workup including neurology consult, treatment offered, and discharge instructions provided to headache patients in the ER.

June 2008 epileptics, tricyclic antidepressants, and others. The narcotics category included traditional narcotics (meperidine, fentanyl, morphine) as well as combination medications in which at least one of the medications was a narcotic (acetaminophen/hydrocodone, tramadol, and acetaminophen/codeine). The others category included all medications not tting the above categories. Specic medications administered in the ER for acute treatment and prescribed upon discharge were compared for migraine vs other headache types. This was analyzed with regard to medication categories as well as specic medications used. Migraine diagnoses included migrainous infarction and retinal migraine. The other headache types included a nal diagnosis in ER of nonspecic headache, cluster headache, sinus headache, tension headache, subarachnoid hemorrhage (SAH), stroke, and vertebrobasilar insufciency (VBI). Analyses excluded patients discharged from the ER (with or without a prescription) who were hospitalized. We also assessed the number of medications given to each patient and compared the total number of medications used to treat migraine vs other headache types. Furthermore, the ER notes were reviewed by a neurologist who formulated a diagnosis based on the available documentation by using the International Headache Society criteria for headache. This diagnosis was then compared with the ER physicians diagnosis and analyzed for the degree of consistency between the two. Statistical Analysis.Descriptive analysis was performed with mean (SD) for continuous variables and cross tabulation for categorical variables. Baseline characteristics were compared between the migraine and other headache groups using independent test for continuous variables, and chi-square tests or Fisher exact test for categorical variables. The number of medications used in ER and upon discharge was compared between the migraine and other headache groups using Fisher exact tests. The degree of interrater reliability (ER physician vs neurologist) was estimated using Cohens Kappa coefcient. As suggested by Landis and Koch, 9 the agreement was deemed poor (Kappa < 0), slight (Kappa = 0-0.20), fair (Kappa = 0.21-0.40), moderate (Kappa = 0.41-0.60), substantial (Kappa = 0.61-0.80), or almost perfect (Kappa =

MATERIALS AND METHODS Study Design.This study was designed as a retrospective chart review of 100 randomly selected patients who visited the ER at LAC + USC Medical Center with a chief complaint of headache during the period of January 1 to June 25, 2006. In the year 2006, 161,487 total ER patient visits were recorded, illustrating the high volume of patients seeking treatment at this facility. The study was approved by the Institutional Review Board of LAC + USC Medical Center. Clinical Chart Review.The selection of 100 patient ER charts was based on ICD-9 codes for a discharge diagnosis of headache. A data form was established to include demographics, chief complaint, presentation of symptoms, history, diagnosis made by the ER physician, specic medications administered in the ER, procedures and studies performed in the ER, medications prescribed upon discharge from the ER, follow-up provided to patients upon discharge from the ER, hospitalization, and consults. Medications administered were also grouped into the following categories: narcotics, nonsteroidal anti-inammatory drugs (NSAIDs), antiemetics, ergots, triptans, anti-

Headache
Table 1.Demographics of Patients Visiting ER for Migraines and Other Headaches

933 males (74% female vs 26% male). The majority of patients were female and Hispanic (58%). Overall, 42% of patients reviewed were diagnosed with migraine by the ER physicians. The mean age was 35 years for patients diagnosed with migraine (n = 42) and was 40 years for nonmigraine headaches (n = 58). Of the patients diagnosed with migraines, 81% were female and 19% male. Nonmigraine patients were 69% female and 31% male. Table 1 compares demographic variables between migraine patients with nonmigraine patients and no statistically signicant difference was found between these 2 groups. Diagnosis and Treatment.Diagnoses made by the ER physicians and clinical tests conducted are illustrated in Table 2. The most common diagnoses were headache not otherwise specied (headache NOS [42%]) and migraine headache (42%). When we were unable to categorize headache as tension or migraine due to lack of sufcient headache history, we used the category of headache NOS. Included in the migraine headache category were migrainous infarction (2%) and retinal migraine (1%). Cluster headache was evident in 6%, tension headache 1%, sinus headache 1%, and all others 8%. All others included the diagnosis of stroke (5%), SAH (2%), and VBI (1%). Computerized tomography (CT) scan of the head was performed for 51%, lumbar puncture was performed on 9% and a neurologist consulted on 17% of all headache patients. Only 7% of migraine

Migraine HA (n = 42)

Other HA (n = 58)

P value*

Age, years Sex, n (%) Female Male Race, n (%) Hispanic White Black Asian

35.6 (13.8) 34 (81) 8 (19) 29 (69) 9 (21.4) 4 (9.6) 0

40.5 (13.6) 40 (69) 18 (31) 47 4 6 1 (81) (7) (10.3) (1.7)

.08 .18

.14

*P value from independent test for continuous variables or chi-square test for categorical variables. Mean (SD). Fisher exact test. ER = emergency room; HA = headache.

0.81-1.00). All statistical testing was performed at a 2-sided 5% level of signicance by using Statistical Analysis System version 8.0 (SAS Institute, Cary, NC, USA).

RESULTS Demographics.The age range of patients in this study was 15-68 years. Of the 100 charts reviewed, 76% of patients were Hispanic, 13% white, 10% black, and 1% Asian. There was a signicantly higher number of females presenting with headache compared with

Table 2.ER Diagnosis and Number of Patients With Each Diagnosis Who Were Hospitalized, Received LP, Head CT, and Neurology Consult

ER diagnosis

All

Hospitalized

LP performed

CT performed

Neurology consulted

Migraine HA* Headache NOS Cluster HA Tension HA Sinus HA All others Total

42 42 6 1 1 8 100

(42) (42) (6) (1) (1) (1)

3 (7) 8 (19) 0 0 0 7 (87.5) 18 (18%)

3 (7.7) 6 (14.3) 0 0 0 0 9 (9%)

16 25 2 0 0 8

(38) (59) (33.3)

(100)

3 (7.7) 7 (16.7) 0 0 0 7 (87.5) 17 (17%)

51 (51%)

*Includes migrainous infarction and retinal migraine. No. (%). Includes stroke, subarachnoid hemorrhage, vertebrobasilar insufciency. CT = computerized tomography; ER = emergency room; HA = headache; LP = lumbar puncture; NOS = not otherwise specied.

934 patients were seen by a neurologist in the ER. Eighteen percent of all headache patients seen were hospitalized. Only 4.8% of migraine patients received triptans in the ER. However, 31% were discharged with a prescription for triptans (Table 3a). Ten percent of migraine patients were discharged with prescriptions for migraine prophylaxis (topiramate, valproic acid, and nortriptyline). Narcotics were given to 25% of all headache patients in the ER for acute relief (26% of migraine patients). Twenty percent of patients were discharged with narcotics (28% of migraine patients). Vicodin (acetaminophen/hydrocodone) was commonly used, with 14% of migraine patients and 12% of nonmigraine patients receiving this drug (Table 3b). Upon discharge, 25% of migraineurs and 18.6% of nonmigraineurs received a prescription for Vicodin. Only one patient out of 42 migraine patients received ergots in the ER (Table 3c). Acetaminophen was used more frequently for nonmigraineurs (27.6%) than migraineurs, and also given more frequently to nonmigraineurs on discharge (21%). Table 4 compares the number of medications received by migraine patients with nonmigraineurs. Most patients received a single agent in the ER for pain control. It was surprising that 18 (43%) migraine patients did not receive any medication in the ER and 8 (22%) did not receive any medication upon discharge.A total of 36 headache patients did not receive any medication while in the ER. The migraine group did not differ from the headache NOS group with regard to the number of medications used in ER, or given upon discharge (P = .77, .36, respectively). Table 5 shows that the ER physicians and neurologists agreed upon a diagnosis of migraine in 38 cases. A diagnosis of migraine was made by the neurologist for an additional 17 patients who had received another diagnosis by the ER physician. Cohens Kappa coefcient test revealed a Kappa value of 0.59, indicating a moderate agreement between the ER physician and neurologist regarding diagnosis of migraine. We reviewed the discharge follow-up appointments provided by the ER to headache patients. Upon discharge from ER, follow-up appointments were given to 40% of all headache patients for primary care

June 2008 clinics and 16% patients for neurology clinic. Among the migraineurs, one-third were given neurology clinic and another third were given primary care clinic appointment for follow-up. It was surprising that 26% of headache patients and 33% of migraineurs received no follow-up appointments, while 15% of all headache patients and approximately one-third of migraineurs had follow-up with a neurologist.

DISCUSSION This study shows that the majority of ER visits for headaches at the LAC + USC during the 6-month chart review period were for migraine. The nding of female preponderance among headache patients in the ER is consistent with published data.10,11 The majority of the patients in our study were of Hispanic heritage. This nding differs from data reported in one of the largest retrospective analyses reported in 2002, in which review of 24,000 ER charts for headache diagnoses revealed a predominance of young white females.5 The procedures performed in the LAC + USC ER included an imaging rate of 51% (CT scans) and lumbar puncture rate of 9%. This was higher than reported for national rates, which have been reported at 14% and 2.4%, respectively.4 To our knowledge, no previous study has evaluated the percentage of neurology consults called by the ER physicians. Only 17% patients were seen by specialist neurologists in the ER at LAC + USC, with a moderate degree of agreement between ER physicians and neurologists regarding migraine diagnosis. Our study showed that 42% of patients were discharged from the ER without a proper and specic diagnosis. This is similar to observations made by other ER studies on headache patients. We would like to emphasize that lack of adequate diagnosis leads to improper and inadequate treatment of headache patients in the ER. Neurology consult in the ER can be useful in the effective management of headaches. The underdiagnosis and misdiagnosis of migraine headaches by the ER have been reported in previous studies, and all the current advances in understanding the pathogenesis and management of headaches have had no apparent impact in the ER. In one study in 2001, 67% of patients discharged with a diagnosis of nonmigraine were determined to have migraine.12

Headache

935

Table 3.(a) Migraine-Specic Medications Used in ER for Migraine and Other Headache, (b) Nonspecic Medications Used by the ER, (c) Other Medications Used by the ER

Treatment given in ER Migraine HA* (n = 42) Other HA (n = 58)

Discharge medications given by ER Migraine HA (n = 39) Other HA (n = 43)

(a)

Triptans Sumatriptan Rizatriptan Antiepileptics Topiramate Valproic acid Tricyclics Amitriptyline Nortriptyline

2 (4.8) 0 0 0 0 0 Migraine HA (n = 42)

0 0 0 0 0 0 Other HA (n = 58)

10 (25.6) 2 (5.2) 2 (5.2) 1 (2.6) 0 1 (2.6) Migraine HA (n = 39)

2 (4.6) 0 0 0 2 (4.6) 0 Other HA (n = 43)

(b)

NSAIDs Ibuprofen Ketorolac Naproxen Narcotics Acetaminophen-hydrocodone Meperidine Fentanyl Morphine Tramadol Codeine (T#3) Antiemetics Promethazine Prochlorperazine Metoclopramide

5 (12) 6 (14.3) 0 6 1 0 4 0 1 (14.3) (2.4) (9.5) (2.4)

7 (12) 1 (1.7) 0 7 2 1 5 1 0 (12) (3.5) (1.7) (8.6) (1.7)

4 (10.3) 0 1 (2.6) 10 (25.6) 0 0 0 0 1 (2.6) 0 1 (2.6) 1 (2.6) Migraine HA (n = 39)

12 (27.9) 0 1 (2.3) 8 (18.6) 0 0 0 0 1 (2.3) 1 (2.3) 2 (4.6) 1 (2.3) Other HA (n = 43)

1 (2.4) 5 (12) 2 (4.8) Migraine HA (n = 42)

7 (12) 10 (17.2) 0 Other HA (n = 58)

(c)

Others Acetaminophen Fioricet Midrin Excedrin Cafergot

1 (2.4) 0 0 0 1 (2.4)

16 (27.6) 0 0 0 0

1 1 2 2 1

(2.6) (2.6) (5.2) (5.2) (2.6)

9 (21) 1 (2.3) 0 0 0

*Includes migrainous infarction and retinal migraine. Includes headache NOS, cluster HA, tension HA, sinus HA, stroke, subarachnoid hemorrhage, and vertebrobasilar insufciency. No. (%). ER = emergency room; HA = headache; NOS = not otherwise specied; NSAID = nonsteroidal anti-inammatory drug.

936
Table 4.Number of Medications Used in ER to Treat Headache, and Given on Discharge

June 2008

No. of medications

Total

Migraine HA*

Other HA

P value

Treatment in ER, n (%) 0 1 2 3 4 Total Rx upon D/C, n (%) 0 1 2 3 4 Total

36 44 13 4 3 100 21 45 14 1 1 82

(36) (44) (13) (4) (3) (100) (25.6) (55) (17) (1.2) (1.2) (100)

18 16 5 2 1 42 8 24 5 1 1 39

(43) (38) (12) (4.5) (2.5) (100) (22.2) (58.3) (14) (3) (3) (100)

18 28 8 2 2 58 13 21 9 0 0 43

(31) (48) (14) (3.5) (3.5) (100) (30) (49) (21)

0.77

0.36

(100)

*Includes migrainous infarction and retinal migraine. P value obtained from Fisher exact test. D/C = discharge; ER = emergency room; HA = headache; Rx = prescription.

Another study concluded that neurology consult was necessary in the ER for a correct diagnosis of headache and, subsequently, prompt management.13 A puzzling nding of our study is that 43% of migraineurs and 31% of nonmigraineurs did not receive any acute treatment in the ER. Although we considered that this could be due to lack of documentation, it is unusual that both physicians and nurses would omit such information during documentation practices. It is possible that the lack of treatment reects spontaneous resolution of headaches during the long wait for treatment in a busy county facility ER. More patients were discharges with a triptan pre-

Table 5.Agreement of Migraine Diagnosis between ER Physician and Neurologist

Neurologist + Total

ER physician

+ Total

38 17 55

4 41 45

42 58 100

ER = emergency room.

scription (31%) than were treated with triptans in the ER (4.8%). We postulate that some of our patients go to the ER when they need prescription rells instead of using primary care services due to lack of insurance; this may explain the above paradox. Ten percent of our patients were discharged from ER with a prescription for preventative treatment. Since there is no continuity of care or follow-up for these patients by their ER physicians, and ER physicians may not have the time to adequately explain drug interactions and adverse reaction, and moreover do not know how patients responded to the new treatment, it is advisable that they focus their attention on adequate acute treatment. The data in this study also indicate that an oral narcotic (Vicodin) was the most frequently used medication for both migraine and nonmigraine headaches by the ER physicians. The preference of oral narcotics over a parenteral route for headache treatment in the ER has not been previously reported. In 2002, Vinson reported that 45% migraineurs received opioids for acute treatment in the ER and intramuscular meperidine was the most commonly administered (70%). This was similar to the nding reported by Barton a decade earlier (1993) that intramuscular meperidine was the treatment of choice for migraine

Headache in the ER. Our study conrms that narcotics are still commonly used for migraine and nonmigraine headache treatment even in the ER of an academic medical center. The problems with opioids are wellknown and include addiction potential, variable relief, and adverse event prole. Another disadvantage of opioids is their long duration of action and resulting sedation, which may render a patient unable to function for 24 hours or longer. This treatment pattern of migraines by the ER can result in dependence on narcotics and the expectations of receiving such medications on a subsequent visit to the ER. It also deprives patients an opportunity to receive migraine-specic treatment which they can use for self-treatment at home. This is especially important for a county hospital population, which often has no other source of receiving medical care. The effectiveness of migraine-specic therapy for acute attacks is now well established.14 Triptans, ergot derivatives, and dopamine antagonists (antiemetics that are considered migraine-specic include prochlorperazine and promethazine) have been shown to be superior to placebo and NSAIDs for the treatment of acute migraines. A multicenter, randomized, doubleblind, placebo-controlled study15 involving 12 emergency rooms in the United States compared subcutaneous sumatriptan with placebo and found signicant headache relief (75% vs 35%), more rapid onset of headache relief (43 vs 66 minutes), and reduced time to discharge from the ER (60 vs 96 minutes). Metoclopramide has comparable pain relief to sumatriptan at 2 and 24 hours.16 Another antiemetic, intravenous (IV) chlorpromazine, is also effective in decreasing the pain and associated symptoms of migraine.17 The US Headache Consortium guidelines recommend using migraine-specic medications to treat acute moderate to severe migraine attacks and preventive medications for frequent and disabling migraines.8 However, our study indicates that ER physicians are not using these. There could be many reasons for this practice. Patients given medications intravenously or intramuscularly may be more likely to be discharged sooner and need less observation. Other explanations include patient expectation of being treated with a shot, and physician bias including familiarity and comfort level with narcotics.

937 Since headache is such a commonly seen condition by the ER, guidelines should be established for their management, tailored to the specic setting in which the hospital operates. Diagnostic workup for headache patients with nonfocal neurological examination should generally be limited. Patients with abnormal fundus exam or symptoms and signs of raised intracranial pressure require further workup (imaging, lumbar puncture), neurological evaluation, and possibly admission. Criteria for hospitalization will vary based on the availability of imaging studies and specialist care. Sometimes, patients are admitted to the hospital due to inability to perform MRI scans in a short time, and when the ER is concerned about patient reliability or long wait time for follow-up appointment with a neurologist. In our opinion, migraine patients who present to the ER should be offered a rapidly acting migrainespecic acute agent.There are many options available, including subcutaneous sumatriptan, intranasal sumatriptan zolmitriptan, ergotamine, or parenteral (IV/ intramuscular[IM]/subcutaneous[SC])dihydroergotamine. Patients who have nausea and vomiting should be offered a parenteral antinausea agent. This is a special point of emphasis, since there is delayed gastric emptying during a migraine attack. Patients with status migrainosus (migraine for longer than 3 days) or those who have intractable vomiting and dehydration requiring parenteral uids should be hospitalized. Migraine is a chronic condition, and therefore it is very important to use the opportunity presented by an ER visit to educate patients about headache prevention and treatment. Even with the time constraints of emergency medicine, preprinted educational materials should be utilized and nursing staff trained to provide quick counseling. Upon discharge, the patient may be prescribed the same migraine-specic medication that they responded to in the ER. Every headache patient should be discharged with an appointment to the primary care doctor, and those with either chronic daily headaches or frequent migraines to the neurologist for follow-up care. Patients who are sent home with a prescription for preventative medication are likely having frequent migraine, and should also be given follow-up with neurologist. Studies have shown that headache

938 repeaters make up to 10% of the ER patients that present with headaches, and can account for up to 50% of the ER visits for headaches.18,19 In this era of overcrowded ERs and long wait times, these are many small measures that can be taken to potentially decrease headache repeaters in the ER. The limitations of our study include its retrospective nature and relatively small sample size. We were not able to assess what level of patient education was conducted in the ER nor were we able to assess efcacy of treatment, medication side effects, or rebound headache. We conclude that national guidelines for migraine management may not be useful in all emergency rooms due to a variety of constraints that they operate under. Specic guidelines for headache management should be established keeping in mind the unique setting of the ER. Our study demonstrates the uniqueness of each ER and in doing so should help toward establishing general guidelines taking into account the county population. Acknowledgment: We thank Dr. Wendy Gilmore for critical reading of the manuscript.

June 2008
7. Lipton RB, Elston Lafata J, Moon C, Leotta C, Kolodner K, Poisson L. The medical care utilization and costs associated with migraine headache. J Gen Intern Med. 2004;19:1005-1012. 8. Silberstein S. Practice parameter: Evidence-based guidelines for migraine headache (an evidencebased review). Neurology. 2000;55:754-762. 9. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33: 159-174. 10. Lipton RB, Bigal ME, Diamond M, Freitag F, Reed ML, Stewart WF. Migraine prevalence, disease burden, and the need for preventative therapy. Neurology. 2007;68:343-349. 11. Stewart WF, Lipton RB, Celentano DD, Reed ML. Prevalence of migraine in the United States. JAMA. 1992;267:64-69. 12. Maizels M. Headache evaluation and treatment by primary care physicians in an emergency department in the era of triptans. Arch Intern Med. 2001;161:1969-1973. 13. Agostoni E., Santoro P, Frigerio R, Frigo M, Beghi E, Ferrarese C. Management of headache in the emergency room. Neurol Sci. 2004;25:S187S189. 14. Diamond S, Bigal ME, Silberstein S, Loder E, Reed M, Lipton RB. Patterns of diagnosis and acute and preventative treatment for migraine in the United States: Results from the American migraine prevalence and prevention study. Headache. 2007; 47:355-363. 15. Akpunonu B. Subcutaneous sumatriptan for treatment of acute migraine patients admitted to the emergency department: A multi-center study. Ann Emerg Med. 1995;25:464-469. 16. Friedman BW, Corbo J, Lipton RB. Metoclopramide vs. sumatriptan for ED treatment of migraines. Neurology. 2005;64:463-468. 17. Bigal ME, Bordini CA, Speciali JG. IV chlorpromazine in the emergency department treatment of migraines: A randomized controlled trial. J Emerg Med. 2002;23:141-148. 18. Maizels M. Health resource utilization of the emergency department headache repeater. Headache. 2002;42:747-753. 19. Freitag FG, Kozma CM, Slaton T, Osterhaus JT, Barron R. Characterization and prediction of emergency department use in chronic daily headache patients. Headache. 2005;45:891-898.

REFERENCES
1. Barton C. Evaluation and treatment of headache patients in the emergency department: A survey. Headache. 1993;34:91-94. 2. McCaig LF, Burt CW. National Hospital Ambulatory Medical Survey: 2003 Emergency department summary. Advance data from vital and health statistics. No. 358. Hyattsville, MD: National Center for Health Statistics; 2005. 3. Centers for Disease Control and Prevention. Advance data from vital and health statistics. June 2006; No.372. 4. Goldstein JN, Camargo CA Jr, Pelletier AJ, Edlow JA. Headache in United States emergency departments: Demographics, work-up, and frequency of pathological diagnoses. Cephalalgia. 2006;26:684-690. 5. Vinson D. Treatment patterns of isolated benign headache in US emergency departments. Ann Emerg Med. 2002;39:215-222. 6. Clouse J, Osterhaus JT, Barron R. Healthcare resource use and costs associated with migraine in a managed healthcare setting. Ann Pharmac. 1994; 28:659-664.