Management of migraine in the emergency department: Findings from the 2010-2017 National Hospital Ambulatory Medical Care Surveys

a b s t r a c t

Objective: The study objective was to describe trends in the medical management of migraine in the emergency department (ED) using the 2010–2017 National Hospital Ambulatory Medical Care Survey datasets. Methods: Using the 2010-2017 NHAMCS datasets, we analyzed visits with a discharge diagnosis of migraine. Drug prescription frequencies between years were compared with the Rao-Scott chi-squared test. Adjusted odds ratios of opioid administration from 2010 to 2017 were calculated using weighted multivariable logistic re- gression with sex, age, race/ethnicity, pain-score, primary expected source of payment, and year as predictor var- iables.

Results: Our analysis captured 1846 ED visits with a diagnosis of migraine from 2010 to 2017, representing a weighted average of 1.2 million US ED visits per year. Parenteral opioids were prescribed in 49% (95% CI: 40, 58) of visits in 2010 and 28% (95% CI: 15, 45) of visits in 2017 (p = 0.03). From 2010 to 2017, there was a 10% yearly decrease in Opioid prescriptions. Metoclopramide and ketorolac were prescribed more frequently in years 2015 through 2017 than in 2010. Increased opioid administration was associated with female sex, older age, white race, higher pain score, and having Medicare or private insurance as the primary expected source of payment for all years.

Conclusion: Opioid administration for migraine in EDs across the US declined 10% annually between 2010 and 2017, demonstrating improved adherence to migraine guidelines recommending against opioids. We identified several factors associated with opioid administration for migraine, identifying groups at higher risk for unneces- sary opioids in the ED setting.

(C) 2021

  1. Introduction
    1. Migraine in the emergency department

Headache disorders are relatively common, with a prevalence of 47% worldwide [1]. Headache or pain in the head is the fourth most common chief complaint in emergency departments (EDs) nationwide and ac- counts for 3.1% of all visits [2]. To manage headaches properly, ED phy- sicians must differentiate between more common benign Primary headache disorders and rarer life-threatening secondary causes of

* Corresponding author at: Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, EC-10 Cleveland Clinic, 9501 Euclid Ave, Cleveland, OH 44195, United States of America.

E-mail address: [email protected] (P.R. Wang).

headache, including infection and trauma [3]. One of the more common primary headache disorders is migraine, which affects 15% of Americans, leads to 1.2 million ED visits annually, and has an estimated indirect impact of $17.2 billion a year on the US economy [3-5].

    1. Medical management of migraine in the ED

Current guidelines for the management of migraine in the ED recom- mend a wide variety of pharmacotherapies. First-line agents include do- pamine antagonist anti-emetics, Non-steroidal anti-inflammatory drugs (NSAIDs), and triptans [3,6-8]. Most guidelines do not recommend opi- oids because they are associated with longer ED stays, increased ED re- cidivism, and decreased sensitivity to more effective migraine treatments [9,10]. However, an analysis of the 2010 National Hospital Ambulatory Medical Care Survey demonstrated that 49%


0735-6757/(C) 2021

of ED patients in the United States diagnosed with migraine received opioids, similar to the frequency found in an analysis of the 1998 NHAMCS [11,12]. Given the increased clinical awareness of the negative effects of opioid use and the increased adoption of Prescription Drug Monitoring Programs (PDMPs) in the past decade, the question remained whether current guidelines and opioid restrictions have im- pacted the medical management of acute migraine in the ED over time [13-15].

    1. Objectives

The primary objective of this study was to compare medication uti- lization for acute migraine management in the ED between the time pe- riods of 2010 and 2015-2017. Secondary objectives were to compare trends in the frequency of IV opioid use for ED management of migraine between 2010 and 2017 and to identify risk factors associated with IV opioid use in the management of migraine.

  1. Methods
    1. Study design

This retrospective descriptive study analyzed data from the 2010-2017 NHAMCS ED datasets. The NHAMCS is a national probability sample program conducted by the National Center for Health Statistics (NCHS), part of the Centers for Disease Control and Prevention. The NHAMCS uses a staged sample design composed of geographic primary sampling units, hospitals within sampling units, and patient visits within emergency service areas. Since the NHAMCS considers estimates based on fewer than 30 visits unreliable, the 2015-2017 NHAMCS datasets were combined to facilitate comparison with 2010 NHAMCS data for less frequently prescribed drugs such as oral opioid analgesics [16]. The NHAMCS is approved by the Ethics Review Board of NCHS, with waivers of the requirements to obtain informed consent of patients and patient authorization for release of medical record data by health care providers [16].

    1. Selection of participants

The 2010-2015 NHAMCS datasets use the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes while the 2016-2017 datasets use the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes. Visits with a discharge diagnosis of migraine headache (ICD-9-CM 346; ICD-10-CM G43-) were eligible for study inclusion.

    1. Study variables and outcomes

Hospital staff or Census Bureau field representatives completed a patient record form for each sampled visit according to information ob- tained from the medical record. Patient characteristics, visit characteris- tics, primary expected source of payment, diagnoses, and medications were recorded by these trained abstractors. Both discharge medications and medications administered in the ED were included.

The NHAMCS classifies drugs using Multum’s Lexicon Drug Data- base. Each medication prescribed or administered was assigned one drug-identifying code and up to four drug categories. When possible, we classified Medication classes similarly to prior analyses of NHAMCS ED-diagnosed migraine visits to facilitate comparisons between this analysis and previous studies [12]. Drugs in Multum category 060 (nar- cotic analgesics) are classified as parenteral opioids.

    1. Statistical analysis

Data were analyzed using SPSS version 27’s complex samples mod- ule to account for the NHAMCS patient visit stratification, clustering,

and weighting. Categorical variables are reported with unweighted count, weighted count, and weighted frequency. Continuous variables were converted into categorical variables. Age was split into three cate- gorical variables: 0-18, 19-50, and > 50. Pain scores were categorized as 0 (no pain), 1-4 (mild pain), 5-7 (moderate pain), and 8-10 (severe pain). Statistical significance of frequencies between groups was assessed using the Rao-Scott chi-squared test. Adjusted odds ratios of opioid administration from 2010 to 2017 were calculated using multi- variable logistic regression using sex, age, race/ethnicity, pain-score, pri- mary expected source of payment, and year as predictor variables. Weighted count estimates are rounded to the nearest 1000 as per NHAMCS recommendations.

  1. Results

A total of 92,173 visits were identified in the NHAMCS database in 2010 and 2015-2017. 873 visits had a discharge diagnosis of migraine, representing an estimated 5,046,000 (95% CI: 4,502,000-5,589,000) mi- graine visits. Demographics are summarized in Table 1. The majority of the ED-diagnosed migraine patients were young (<50 years old) white females presenting with pain rated at a score of 8 to 10 out of 10, which is consistent with the typical demographics of ED migraine patients [11,12]. For migraine visits, the most common medications used in 2015-2017 were anti-emetics, which were prescribed in 73.9% of visits (95% CI: 68.3, 78.8), followed by NSAIDs, which were administered in 49.6% of visits (95% CI: 44.0, 55.2). Additional migraine medications ad- ministered during ED visits are summarized in Table 2. IV opioids were administered in 26.6% of visits (95% CI: 21.0, 33.3) while oral opioid an- algesics were prescribed in 12.6% of visits (95% CI: 9.3, 16.8).

The frequency of parenteral opioid administration in the manage- ment of ED-diagnosed migraine decreased from 49.0% of visits in 2010 to 27.5% of visits in 2017 (p = 0.03) (Fig. 1). Between 2010 and 2017, the year with the lowest opioid administration was 2016, with opioids being prescribed in 23.8% (95% CI: 17.8, 31.0) of visits.

Both IV opioids and oral opioid analgesics were administered less often in 2015-2017 than in 2010 (Table 3). Hydromorphone was the most commonly prescribed parenteral opioid in 2010 and in 2015-2017, but its frequency of use fell from 24.9% (95% CI: 19.0,

32.0) in 2010 to 12.4% (95% CI: 9.3, 16.4) from 2015 to 2017

(p < 0.001). Morphine was the second most commonly prescribed IV opioid from 2015 to 2017 (6.2%, 95%: CI 3.8, 9.9) and was prescribed

at similar frequencies as in 2010 (7.0%, 95% CI: 4.4, 11.1). Other paren- teral opioids such as nalbuphine, butorphanol, and meperidine were rarely administered from 2015 to 2017; estimates on their administra- tion from 2015 to 2017 are based on fewer than 30 visits in the NHAMCS.

The use of guideline-recommended agents such as ketorolac and anti-emetics increased. Ketorolac administration during ED visits in- creased from 33.3% in 2010 to 41.5% in 2015-2017 (p = 0.03) (Table 3). The most common anti-emetic prescribed was ondansetron, which was administered in 23.3% (95% CI: 17.2, 30.6) of visits in 2010

and 28.4% (95% CI: 23.3, 34.0) of visits from 2015 to 2017. The use of

metoclopramide increased from 16.8% in 2010 to 27.0% from 2015 to 2017 (p = 0.02). Other medications commonly used in the management of ED-diagnosed migraine include benzodiazepines, butalbital-containing medications, triptans, acetaminophen, and diphenhydramine. acetaminophen use tripled from 3.1% (95% CI: 1.4, 6.5) in 2010 to 10.3% (95% CI: 7.3, 14.2) from 2015 to 2017.

Similarly, ibuprofen use roughly doubled from 5.3% (95% CI: 3.1, 8.9)

in 2010 to 10.4% (95% CI: 7.4, 14.4) from 2015 to 2017. Other medica- tions recommended for management of acute migraine such as dihy- droergotamine and sodium valproate were prescribed very rarely and are not included in this analysis .

Using multivariable logistic regression analysis, we identified several factors independently associated with opioid administration in ED- diagnosed migraine visits from 2010 to 2017 (Table 4). We found that

Table 1

Demographics and visit characteristics of US ED migraine visits from 2010 and 2015–2017.




Raw count

Weighted Count Visits x 103, (95% CI)

Percent (95% CI)

Raw Count

Weighted Count Visits x 103, (95% CI)

Percent (95% CI)



1145 (931, 1359)


3900 (3400, 4400)

Patient sex



916 (734, 1099)

80% (71.9, 86.3)


3245 (2790, 3699)

83.2% (79.0, 86.7)



229 (145, 313)

20% (13.7, 28.1)


656 (492, 819)

16.8% (13.3, 21.0)

Age (years)



71 (24, 117)

6.2% (3.3, 11.2)


424 (278, 570)

10.9% (7.8, 14.9)



953 (789, 1118)

83.3% (77.9, 87.5)


2758 (2350, 3165)

70.7% (65.7, 75.3)



121 (73, 169)

10.6% (7.1, 15.4)


719 (523, 914)

18.4% (14.4, 23.3)


Non-Hispanic White


928 (754, 1101)

81.0% (74.2, 86.4)


2597 (2197, 2997)

66.6% (60.7, 72.0)

Non-Hispanic Black


104 (55, 154)

9.1% (5.4, 14.9)


656 (453, 858)

16.8 (12.7, 21.9)



98 (46, 150)

8.6% (5.3, 13.6)


565 (401, 729)

14.5% (11.0, 18.9)

Non-Hispanic Other


15 (0,32)

1.3% (0.4, 4.0)


82 (23, 141)

2.1 (1.0, 4.2)

Pain Score



37 (3, 72)

3.6% (1.4, 9.1)


201 (104, 299)

6.3% (4.0, 9.9)

1 to 4


40 (5, 74)

3.8% (1.6, 8.9)


262 (153, 372)

8.2% (5.5, 12.1)

5 to 7


208 (123, 294)

19.9% (14.1, 27.3)


799 (586, 1012)

25.1% (20.2, 30.7)

8 to 10


764 (604, 924)

72.8% (65.0, 79.3)


1920 (1570, 2270)

60.3% (54.2, 66.1)

Primary Expected Source of Payment

Private Insurance


434 (328, 539)

49.8% (41.4, 58.3)


1386 (1146, 1626)

35.5% (31.1, 40.2)

Medicaid/CHIP/State Program


325 (229, 422)

37.4% (29.4, 46.1)


1259 (1015, 1502)

32.3% (27.7, 37.2)



111 (56, 167)

12.8% (7.9, 20.0)


472 (319, 624)

12.1% (7.2, 16.6)

* Estimate based on raw count <30 visits, which is considered unreliable by NHAMCS.

females were more likely to receive IV opioids than males (aOR: 1.53, 95% CI: 1.04, 2.24). Patients over the age of 50 were more likely to re- ceive opioids than patients between 19 and 50 (aOR: 1.65, 95% CI: 1.03, 2.63). Non-Hispanic white patients were more likely to receive opioids than black patients (aOR: 1.67, 95% CI: 1.01, 2.77). Patients reporting a severe (8 to 10) pain score had higher odds of receiving opi- oids than patients reporting mild (1 to 4) pain (aOR: 4.42, 95% CI: 1.92, 10.18). Relative to patients whose primary expected source of payment was either Medicaid, CHIP, or a state program, those with Medicare or private insurance were more likely to receive opioids (aOR: 1.48, 95% CI: 1.03, 2.12 and aOR: 1.60, 95% CI: 1.02, 2.52, respectively). Finally, the likelihood of IV opioid administration decreased by 10% each year (aOR: 0.90, 95% CI: 0.81, 0.99).

  1. Discussion
    1. Non-opioid medical management of migraine

In our analysis of the NHAMCS data set, we found that the frequency of administration of several non-opiate medications increased from the 2010 study period to the 2015-2017 study period. Anti-emetics were prescribed in 73.9% of visits in 2015-2017, and the most commonly pre- scribed anti-emetic was ondansetron, administered in 28.4% of visits. Ondansetron and other selective 5-HT3 receptor antagonists are not

recommended in current guidelines for acute treatment of migraines and have not been studied for migraine treatment in clinical trials. We hypothesize that ondansetron is being given primarily to manage migraine-associated nausea, not as an anti-migraine agent, and its frequency of administration may be a byproduct of its frequent use in the overall population of ED patients [17]. Preliminary results of ondansetron trials in the primary management of migraine are promising in Pediatric populations [18,19]. We also showed that metoclopramide was used more frequently from 2015 to 2017 relative to 2010, while promethazine was prescribed less. Promethazine received a black box warning due to tissue injury with IV administration in 2009, which might explain its decreased use [20]. Ketorolac was also used more often, which may reflect increased clinical appreciation of its efficacy and safety [21]. Other non-opioid analgesics such as ibupro- fen and acetaminophen were also used more often, perhaps as an

Opioid Prescription Frequency by Year

Percent of migraine visits in which an parenteral opiod was administered




Table 2

Frequency of use of medication classes to treat migraine in US EDs from 2015 to 2017.

Raw count

Weighted count Visits X 103, (95% CI)

Percent (95% CI)



2882 (2440, 3323)

73.9% (68.3, 78.8)



1933 (1611, 2256)

49.6% (44.0, 55.2)



1613 (1324, 1902)

41.4% (36.1, 46.8)

IV Opioid


1037 (750, 1324)

26.6% (21.0, 33.3)

Opioid analgesic


490 (337, 643)

12.6% (9.3, 16.8)



400 (258, 452)

10.3% (7.3, 14.2)



355 (240, 471)

9.1% (6.7, 12.4)



332 (221, 444)

8.5% (6.2, 11.7)



303 (187, 420)

7.8% (5.4, 11.1)





Fig. 1. Parenteral opioid prescribing frequencies in the ED management of migraine from 2010 to 2017.

Table 3

Frequency of use of medications to treat migraine in US EDs from 2015 to 2017 versus 2010

analysis by Ruzek et al. demonstrated reduced Opioid prescriptions be- tween 1990 and 2000 and 2014 [12,22]. Our study corroborates the findings of Ruzek and colleagues and demonstrates decreased opioid

Frequency of use in 2010 (95% CI)

Frequency of use from 2015 to 2017

(95% CI)


administration in the management of migraine between 2010 and 2017 using a national representative probability survey.

Guidelines generally recommend against opioids for management of

IV Opioids 49.0% (40.3, 57.8) 26.6% (21.0, 33.0) <0.01

Hydromorphone 24.9% (19.0, 32.0) 12.4% (9.3, 16.4) <0.01

migraine for several reasons [6,7]. Multiple retrospective studies have linked opioid administration to longer stays and repeated visits to the

Morphine 7.0% (4.4, 11.1)

Nalbuphine 10.3% (4.9, 20.2)?

6.2% (3.8, 9.9)

2.1% (0.7, 6.1)?



ED [9,23-25]. Furthermore, opioid administration has been associated

Butorphanol 1.4% (0.6, 3.3)? 1.4% (0.6, 3.3)? 0.98

Meperidine 7.0% (3.9, 12.2)? 1.1% (0.5, 2.4)? <0.01

with increased likelihood of medication overuse and treatment- resistant headache [10,26]. While opioids have been demonstrated to

Opioid Analgesics 18.9% (14.1, 24.8) 12.6% (9.3, 16.8) 0.05

be more effective than placebo, randomized controlled trials have

Hydrocodone 13% (9, 18.4) 4.0 (2.6, 6.2)

Oxycodone 4.9% (2.7, 8.9)? 4.9 (2.7, 8.8)?



shown that other agents such as prochlorperazine with diphenhydra-

Codeine 0.6% (0.1, 3.8)? 4.4 (2.7, 7.0)? 0.02

NSAIDS 38.1% (32.3, 44.3) 49.6% (44.0, 55.2) <0.01


33.3% (27.7, 39.3) 41.5% (36.3, 46.9)



5.3% (3.1, 8.9)? 10.4% (7.4, 14.4)



77.3% (70.2, 83.1) 73.9% (68.3, 78.8)



23.3% (17.2, 30.6)

28.4% (23.3, 34.0)



16.8% (11.8, 23.4)

27.0% (22.7, 31.8)



33.0% (24.0, 43.4)

18.2% (14.0, 23.3)



14.8% (9.9, 21.5)

12.5% (9.5, 16.4)



3.5% (1.8, 6.5)?

9.1% (6.7, 12.4)



Triptans 8.8% (5.6, 13.6) 8.5 (6.2, 11.7) 0.92

Acetaminophen 3.1% (1.4, 6.5) 10.3% (7.3, 14.2) <0.01

Diphenhydramine 22.5% (16.2, 30.4) 41.4% (36.1, 46.8) <0.01

Dexamethasone 2.3% (0.9, 5.5)? 4.5% (2.6, 7.6)? 0.20

* Estimate based on raw count <30 visits, which is considered unreliable by NHAMCS.

alternative to opioids to manage pain in migraine patients. Other med- ications occasionally utilized for acute migraine management such as dihydroergotamine and valproate were rarely used.

    1. Opioid use in the management of migraine

We found that IV opioids were used less often in 2015-2017 than in 2010 and were 10% less likely to be prescribed each subsequent year during ED visits with migraine discharge diagnoses. Friedman et al. found similar frequencies of overall opioid prescription between 1998 and 2010 using the NHAMCS data, but a multi-hospital retrospective

Table 4

Factors associated with IV opioid prescription in the ED management of migraine from 2010 to 2017: multivariable logistic regression analysis.

Variable Adjusted odds ratio (95% CI)

Patient Sex

Female 1.53 (1.04, 2.24)

Male 1 (reference)


0-18 0.44 (0.23, 0.86)

19-50 1 (reference)

>50 1.65 (1.03, 2.63)


Non-Hispanic White 1.67 (1.01, 2.77)

Non-Hispanic Black 1 (reference)

Hispanic 1.30 (0.71, 2.41)

Non-Hispanic Other 1.30 (0.45, 3.79) Pain Score

0 0.79 (0.25, 2.49)

1 to 4 1 (reference)

5 to 7 1.95, (0.76, 5.03)

8 to 10 4.42 (1.92, 10.18)

Primary Expected Source of Payment

Medicaid/CHIP/State Program 1 (reference)

Medicare 1.48 (1.03, 2.12)

Private Insurance 1.60 (1.02, 2.52)

Survey Year (one year increment) 0.90 (0.81, 0.99)

mine are significantly more efficacious than opioids in the management of migraine [27,28].

While this analysis cannot identify the specific sources of the ob- served decrease in opioid administration, we hypothesize that the cause is multifactorial. Many states have established PDMPs, which are state-maintained databases that track the prescription of controlled substances.13 PDMPs existed prior to 2010, but it was not until 2012 that the first comprehensive mandate to access PDMPs before prescribing opioids was established, in Kentucky. Twenty-two other states had adopted similar mandates by 2015 [29]. PDMPs with mandates are more efficacious at preventing unnecessary opioid prescriptions than PDMPs without mandates. The increased adoption of mandates and ease of use of PDMPs could be responsible for the observed decreased in opioid administration [14,30,31]. Furthermore, while migraine guide- lines have recommended against opioids in the management of acute migraine since 1999, there is often a lag between publication of guide- lines and physician adherence to them [6,32]. Especially in the ED, the number of pathologies that physicians must know how to diagnose and manage may delay knowledge of and adherence to guidelines. Therefore, physicians could have opted to use narcotics that they knew would reduce pain rather than other more efficacious anti- migraine agents. Additionally, increased physician and public aware- ness of the opioid epidemic might have contributed to the decrease in opioid prescription [33].

    1. Factors associated with opioid administration in migraine management

We identified several factors associated with opioid administration between 2010 and 2017. As expected, initial pain score was associated with IV opioid administration. Since opioids are indicated for pain man- agement, physicians might be electing to control patients‘ pain at the expense of worse outcomes [23-25,34]. We found that older patients (>50) were more likely to receive opioids than younger patients. This is inconsistent with the literature, which indicates that older patients are less likely to receive IV opioids, possibly due to increased risk of side effects in the elderly [35,36]. White patients were more likely to re- ceive opioids than black patients, in contrast to the findings of a previ- ous analysis of outpatient office-based visits for migraines [37]. These apparent racial discrepancies in opioid prescribing for migraines in EDs require further study. Similar to prior analyses, we observed that patients whose expected source of payment was either private insur- ance or Medicare were more likely to receive IV opioids than patients whose expected payment source was Medicaid, CHIP, or a state pro- gram [38,39]. We were surprised to find that females had greater odds of receiving opioids than males. Prior studies have not demonstrated a consistent relationship. Friedman et al. found that female migraine pa- tients were more likely to receive opioids (aOR: 1.67, 95% CI: 0.53, 5.29) using the 2010 NHAMCS dataset, although the difference was not significant [12]. Their study might have been underpowered to as- sess the actual association between sex and opioid prescription [12,34,38,39]. Women have higher rates of prescription opioid use and are more likely to fill opioid prescriptions, but the reason women are

more likely to receive opioids in the acute management of migraine is unknown and requires further investigation [40,41].

    1. Limitations

Our study has several limitations, most of which are due to the nature of the NHAMCS dataset and have been noted in the literature [16]. NHAMCS data are abstracted from patient records by hospital staff or field representatives, and this process is subject to abstraction error. However, the variables we analyzed (patient demographics, medica- tions, etc.) are discrete and not subject to interpretation, making abstrac- tion errors less likely. Furthermore, NHAMCS migraine diagnosis data has shown consistency with other large databases such as the National Health Interview Survey2. Medication data from the National Ambula- tory Medical Care Survey, which utilizes nearly identical data collection methods as the NHAMCS, have also been validated [42]. The NHAMCS dataset also does not include information on medication dosages, reason for administration, time of administration, or route of administration. Additionally, since the NHAMCS is a cross-sectional survey that tracks visit characteristics without patient identifiers, it is impossible analyze treatment of individual patients over time and patients with multiple ED visits. The lack of this feature is not important when analyzing indi- vidual ED visits, which was the main goal of our study.

Our inclusion criteria of ICD-9 or ICD-10 diagnosis of migraine un- doubtedly failed to include many migraine patients who received non- specific headache codes, as migraine is commonly underdiagnosed in the ED [43]. Since management of tension-type and Cluster headaches differs from management of migraine, we chose to maximize specificity for migraine patients rather than include all headache patients. This ap- proach best captures how ED clinicians are managing patients who they suspect have migraine, which was the ultimate goal of our study [12]. There may be differences in management between patients with a diag- nosis of migraine and misdiagnosed migraine patients, but the NHAMCS dataset is not capable of identifying those differences.

The number of diagnoses per ED visit that could be recorded by ab- stractors increased from three in 2010-2013 to five in 2014-2017. Like- wise, the number of medications that could be recorded increased from 8 in 2010 to 12 in 2012 to 30 in 2014. Even though this increase in ab- stractable medications may reflect as increased frequencies in drug ad- ministration from 2014 to 2017 relative to earlier years, we felt it was best to include all medications available for each year because NHAMCS does not record medications in order of administration or clin- ical significance. Given one of our original hypotheses was that opioid administration in the ED management of migraine would decrease from 2010 to 2017, we did not want to artificially limit the number of medications administered per visit and increase our chance of making a type I error. This could also explain the increase in opioid prescription in 2014 in Table 4 and Fig. 1. Therefore, any trends that show an in- creased frequency of drug administration between 2010 and 2015-2017 should be viewed with this limitation in mind. Furthermore, eight or more medications were administered in only 3.6% (95% CI: 2.4, 5.4) of all migraine visits from 2010 to 2013, indicating that increasing the number of abstractable medications would have a small effect on frequency of use.

We included all five diagnoses from 2014 to 2017 for similar reasons. The diagnoses are not ordered in clinical significance or likelihood; therefore, we included all patients with a recorded migraine diagnosis to increase the power and sample size of our study. Furthermore, mov- ing from recording three diagnoses to five diagnoses did not result in a significant change in the weighted number of migraine diagnoses be- tween 2010 and 2013 and 2014-2017 (p = 0.72).

    1. Conclusion

In conclusion, the use of guideline-recommended therapies such as metoclopramide and ketorolac increased during ED visits for migraines

while the use of IV opioids decreased significantly based on analysis of the NHAMCS dataset from 2010 through 2017.


No funding was provided for this project.

Credit author statement

PRW, BLU, and MPP contributed to the conceptualization of the study. PRW, RL, BLU, and MPP contributed to the methodology. PRW and RL contributed to software development. PRW conducted formal data anal- yses. Data curation was conducted by PRW and RL. PRW drafted the original manuscript, and all authors contributed to draft review and editing. PRW and RL contributed to data visualization. Supervision was conducted by BLU and MPP.

Declaration of Competing Interest

The authors, to the best of our knowledge, have no conflicts of inter- est, financial or otherwise, to declare related to this manuscript.


  1. Stovner LJ, Hagen K, Jensen R, et al. The global burden of headache: a documentation of headache prevalence and disability worldwide. Cephalalgia. 2007;27(3):193-210. https://doi.org/10.1111/j.1468-2982.2007.01288.x.
  2. Burch RC, Loder S, Loder E, et al. The prevalence and burden of migraine and Severe headache in the United States: updated statistics from government health surveil- lance studies. Headache. 2015;55(1):21-34. https://doi.org/10.1111/head.12482.
  3. Long BJ, Koyfman A. Benign headache management in the emergency department. J Emerg Med. 2018;54(4):458-68. https://doi.org/10.1016/j.jemermed.2017.12.023.
  4. Raval AD, Shah A. National trends in direct health care expenditures among US adults with migraine: 2004 to 2013. J Pain. 2017;18(1):96-107. https://doi.org/10. 1016/j.jpain.2016.10.005.
  5. Burch R, Rizzoli P, Loder E. The prevalence and impact of migraine and severe head- ache in the United States: figures and trends from government health studies. Head- ache. 2018;58(4):496-505. https://doi.org/10.1111/head.13281.
  6. Ducharme J. Canadian Association of Emergency Physicians Guidelines for the acute management of migraine headache. J Emerg Med. 1999;17(1):137-44. https://doi. org/10.1016/s0736-4679(98)00136-x.
  7. Orr SL, Friedman BW, Christie S, et al. Management of adults with acute migraine in the emergency department: the American Headache Society evidence assessment of parenteral pharmacotherapies. Headache. 2016;56(6):911-40. https://doi.org/10. 1111/head.12835.
  8. Gilmore B, Michael M. Treatment of Acute migraine headache. Am Fam Physician. 2011;83(3):271-80.
  9. Colman I, Rothney A, Wright SC, et al. Use of narcotic analgesics in the emergency department treatment of migraine headache. Neurology. 2004;62(10):1695-700. https://doi.org/10.1212/01.WNL.0000127304.91605.BA.
  10. Jakubowski M, Levy D, Goor-Aryeh I, et al. Terminating migraine with allodynia and ongoing central sensitization using parenteral administration of COX1/COX2 inhibi- tors. Headache. 2005;45(7):850-61. https://doi.org/10.1111/j.1526-4610.2005. 05153.x.
  11. Vinson DR. Treatment patterns of isolated benign headache in US emergency de- partments. Ann Emerg Med. 2002;39(3):215-22. https://doi.org/10.1067/mem. 2002.121400.
  12. Friedman BW, West J, Vinson DR, et al. Current management of migraine in US emergency departments: an analysis of the National Hospital Ambulatory Medical Care Survey. Cephalalgia. 2015;35(4):301-9. https://doi.org/10.1177/033310241 4539055.
  13. Bote SH. U.S. opioid epidemic: impact on public health and review of prescription drug monitoring programs (PDMPs). Online J Public Health Inform. 2019;11(2). https://doi.org/10.5210/ojphi.v11i2.10113.
  14. Bao Y, Wen K, Johnson P, et al. Assessing the impact of state policies for prescription drug monitoring programs on high-risk opioid prescriptions. Health Aff. 2018;37 (10):1596-604. https://doi.org/10.1377/hlthaff.2018.0512.
  15. Ayres I, Jalal A. The impact of prescription drug monitoring programs on U.S. opioid prescriptions. J Law Med Ethics. 2018;46(2):387-403. https://doi.org/10.1177/ 1073110518782948.
  16. McCaig LF, Burt CW. Understanding and interpreting the national hospital ambula- tory medical care survey: key questions and answers. Ann Emerg Med. 2012;60

(6). https://doi.org/10.1016/j.annemergmed.2012.07.010 716-721.e1.

  1. Malhotra S, Rana D, Patel V, et al. Drug utilization in emergency medicine depart- ment at a tertiary care teaching hospital: a prospective study. J Basic Clin Pharm. 2013;4(4):78-81. https://doi.org/10.4103/0976-0105.121650.
  2. Bachur RG, Monuteaux MC, Neuman MI. A comparison of acute Treatment regimens for migraine in the emergency department. Pediatrics. 2015;135(2):232-8. https:// doi.org/10.1542/peds.2014-2432.
  3. Talai A, Heilbrunn B. Ondansetron for acute migraine in the pediatric emergency de- partment. Pediatr Neurol. 2020;103:52-6. https://doi.org/10.1016/j.pediatrneurol. 2019.06.011.
  4. Onifer DJ, Butler FK, Gross KR, et al. Replacement of promethazine with ondansetron for treatment of opioid- and trauma-related nausea and vomiting in tactical combat casualty care. J Spec Oper Med. 2015;15(2):17-24.
  5. Taggart E, Doran S, Kokotillo A, et al. Ketorolac in the treatment of acute migraine: a systematic review. Headache. 2013;53(2):277-87. https://doi.org/10.1111/head. 12009.
  6. Ruzek M, Richman P, Eskin B, et al. ED treatment of migraine patients has changed. Am J Emerg Med. 2019;37(6):1069-72. https://doi.org/10.1016/j.ajem.2018.08.051.
  7. Young N, Silverman D, Bradford H, et al. Multicenter prevalence of opioid medication use as abortive therapy in the ED treatment of migraine headaches. Am J Emerg Med. 2017;35(12):1845-9. https://doi.org/10.1016/j.ajem.2017.06.015.
  8. Tornabene SV, Deutsch R, Davis DP, et al. Evaluating the use and timing of opioids for the treatment of migraine headaches in the emergency department. J Emerg Med. 2009;36(4):333-7. https://doi.org/10.1016/j.jemermed.2007.07.068.
  9. McCarthy LH, Cowan RP. Comparison of parenteral treatments of acute primary headache in a large academic emergency department cohort. Cephalalgia. 2015;35 (9):807-15. https://doi.org/10.1177/0333102414557703.
  10. Thorlund K, Sun-Edelstein C, Druyts E, et al. Risk of medication overuse headache across classes of treatments for acute migraine. J headache pain. 2016;17(1). https://doi.org/10.1186/s10194-016-0696-8.
  11. Kelley NE, Tepper DE. Rescue therapy for acute migraine, part 3: opioids, NSAIDs, steroids, and post-discharge medications. Headache. 2012;52(3):467-82. https:// doi.org/10.1111/j.1526-4610.2012.02097.x.
  12. Friedman BW, Irizarry E, Solorzano C, et al. Randomized study of IV prochlorperazine plus diphenhydramine vs IV hydromorphone for migraine. Neurology. 2017;89(20): 2075-82. https://doi.org/10.1212/WNL.0000000000004642.
  13. Haffajee RL, Jena AB, Weiner SG. Mandatory use of prescription drug monitoring programs. JAMA. 2015;313(9):891-2. https://doi.org/10.1001/jama.2014.18514.
  14. Bachhuber MA, Tuazon E, Nolan ML, et al. Impact of a prescription drug monitoring program use mandate on potentially problematic patterns of opioid analgesic pre- scriptions in New York City. Pharmacoepidemiol Drug Saf. 2019;28(5):734-9. https://doi.org/10.1002/pds.4766.
  15. Rhodes E, Wilson M, Robinson A, et al. The effectiveness of prescription drug moni- toring programs at reducing opioid-related harms and consequences: a systematic

review. BMC Health Serv Res. 2019;19(1):784. https://doi.org/10.1186/s12913- 019-4642-8.

  1. Fischer F, Lange K, Klose K, et al. Barriers and strategies in guideline implementation

–a scoping review. Healthcare. 2016;4(3):36. https://doi.org/10.3390/healthcare40 30036.

  1. Cantrill SV, Brown MD, Carlisle RJ, et al. Clinical policy: critical issues in the prescrib- ing of opioids for adult patients in the emergency department. Ann Emerg Med. 2012;60(4):499-525. https://doi.org/10.1016/j.annemergmed.2012.06.013.
  2. Pletcher MJ, Kertesz SG, Kohn MA, et al. Trends in opioid prescribing by race/ethnic- ity for patients seeking care in US emergency departments. JAMA. 2008;299(1): 70-8. https://doi.org/10.1001/jama.2007.64.
  3. Platts-Mills TF, Esserman DA, Brown DL, et al. Older US emergency department pa- tients are less likely to receive pain medication than younger patients: results from a national survey. Ann Emerg Med. 2012;60(2):199-206. https://doi.org/10.1016/j. annemergmed.2011.09.014.
  4. Marra EM, Mazer-Amirshahi M, Mullins P, et al. Opioid administration and prescrib- ing in older adults in U.S. emergency departments (2005-2015). West J Emerg Med. 2018;19(4):678-88. https://doi.org/10.5811/westjem.2018.5.37853.
  5. Charleston L, Burke JF. Do racial/Ethnic disparities exist in recommendED migraine treatments in US ambulatory care? Cephalalgia. 2018;38(5):876-82. https://doi. org/10.1177/0333102417716933.
  6. Tomaszewski DM, Arbuckle C, Yang S, et al. Trends in opioid use in pediatric patients in US emergency departments from 2006 to 2015. JAMA Netw Open. 2018;1(8): e186161. https://doi.org/10.1001/jamanetworkopen.2018.6161.
  7. Ritsema TS, Kelen GD, Pronovost PJ, et al. The national trend in quality of emergency department pain management for long bone fractures. Acad Emerg Med. 2007;14

(2):163-9. https://doi.org/10.1197/j.aem.2006.08.015.

  1. Schieber LZ, Guy GP, Seth P, et al. Variation in adult outpatient opioid prescription dispensing by age and sex–United States, 2008-2018. MMWR Morb Mortal Wkly Rep. 2020;69(11):298-302. https://doi.org/10.15585/mmwr.mm6911a5.
  2. Serdarevic M, Striley CW, Cottler LB. sex differences in prescription opioid use. Curr Opin Psychiatry. 2017;30(4):238-46. https://doi.org/10.1097/YCO.000000000 0000337.
  3. Higashi A, Zhu S, Stafford RS, et al. National trends in ambulatory asthma treatment, 1997-2009. J Gen Intern Med. 2011;26:1465-70.
  4. Gupta MX, Silberstein SD, Young WB. Less is not more: underutilization of headache medications in a university hospital emergency department. Headache. 2007;47: 1125-33.