Reasons for patient reluctance to take preventive medications for migraine: Results of the OVERCOME (US) study

Abstract

Objectives/Background

This study was undertaken to evaluate patient reasons for nonadoption of migraine‐preventive medications. Despite clear recommendations and eligibility criteria for migraine‐preventive treatment by the American Headache Society and the availability of these treatments, many people with migraine are not taking appropriate preventive medications. Many are not seeking medical care in the first place, but even among those who are seeking medical care and have a diagnosis of migraine, the uptake of preventive medications remains low.

Methods

The OVERCOME (Observational Survey of the Epidemiology, Treatment, and Care of Migraine) study is an observational, longitudinal web‐based survey conducted in more than 60,000 adults with migraine in the United States (US). The current analysis, a secondary post hoc analysis of the 2018–2020 baseline cross‐sectional surveys, evaluated medication use in participants. In particular, the analysis investigates why some participants have never taken prescription medication for migraine prevention and examines how this group differs from those who are taking preventive medication, specifically in terms of disease severity and other patient‐reported outcomes.

Results

Our findings revealed that among OVERCOME (US) participants who met criteria for migraine (n = 59,001), only approximately half (51.3%) had sought medical care for migraine in the previous 12 months, approximately one third (36.3%) had sought care and received a migraine diagnosis, and only 10% of participants had sought care, received a diagnosis of migraine, and were currently taking prescription medications for migraine prevention. Furthermore, among those who were eligible for migraine‐preventive medication based on their headache frequency and associated disability (n = 22,249), 65.3% indicated they had never taken a preventive medication for migraine. The reasons for this were mostly medication‐related (25.5% stated they were concerned about side effects, 23.3% said they did not like taking prescription medication, and 20.8% stated that their other medications worked well enough); however, there were also other reasons related to stigma, access, and communication with the health care provider that were noted by participants.

Conclusion

This study highlights an important need for patient education, especially as many of these individuals who had never taken medications to prevent migraine reported experiencing ≥15 monthly headache days (25.3%), severe interictal burden (43.3%), and severe migraine‐related disability (53.1%). We believe that these results may be of interest to health care providers who see people with migraine and help them better understand and anticipate their patients' educational needs regarding migraine prevention.

Plain Language Summary

This survey‐based study aimed to understand why many people with migraine are not taking preventive medications despite being eligible for them. Our results showed that patients who did not take preventive medications had concerns about side effects, disliked prescription medications in general, and were satisfied with other treatments. We found that issues related to stigma, access, and communication with health care providers also played a role.

Abbreviations

AHS

American Headache Society

AMPP

American Migraine Prevalence and Prevention

CGRP

calcitonin gene‐related peptide

MHD

monthly headache days

MIBS‐4

Migraine Interictal Burden Scale‐4

MIDAS

Migraine Disability Assessment Scale

OTC

over‐the‐counter

SMD

standardized mean difference

INTRODUCTION

Migraine is a disabling neurological disease that affects approximately 16% of Americans, and 1 in every 5 US women when measured over a 3‐month period. ¹ According to the World Health Organization, migraine ranks as the second leading cause of years lived with disability and the top cause of disability among women aged 15–49 years. ² This burden is further highlighted by the OVERCOME (Observational Survey of the Epidemiology, Treatment, and Care of Migraine) study in the United States (US), which reported that almost half of participants (42.4%) had at least moderate disability due to migraine. Additionally, the survey highlighted those individuals with migraine frequently face stigma due to their migraine that significantly impacts their daily lives. ³ , ⁴

Although use of preventive medications is associated with lowered number and frequency of headaches, ⁵ adherence to older, nonspecific medications used for migraine prevention (e.g., antiseizure, antihypertensive, antidepressant medications) is often low. ⁶ Migraine‐specific preventive medications that block calcitonin gene‐related peptide (CGRP) have become available in the last several years. ⁷ These have been shown to be efficacious, safe, and well tolerated, and are considered a first‐line option for migraine prevention. However, despite clear recommendations by the American Headache Society (AHS) and the International Headache Society regarding eligible patient populations, many people with migraine are not traversing barriers to medical care for migraine and/or are not taking recommended medications for migraine prevention. ⁷ , ⁸ , ⁹ According to recent results from the CaMEO‐I (Chronic Migraine Epidemiology and Outcomes‐International) study, fewer than 15% of people with migraine who were in need of medical care based on their level of disability due to migraine traversed all barriers to care (e.g., consulting a specialist, receiving a diagnosis, and reporting use of treatment for migraine). ¹⁰ In addition, acute medication overuse for headache is common, ¹¹ , ¹² and can complicate headache management, as it is associated with increased risk of migraine progression. ¹² , ¹³

Currently, there are few population‐based studies assessing why individuals do not utilize medications for migraine prevention despite their availability and clear recommendation by professional organizations. The OVERCOME (US) study is a longitudinal, multicohort, web‐based survey that annually recruited a demographically representative sample of adults with migraine in the United States. ³ The study collected information around care‐seeking, diagnosis, and treatment in migraine, and included measures assessing adoption of medications for migraine prevention. The current analysis aims to provide insight into the reasons why some treatment‐eligible individuals are not taking prescription medications for migraine prevention and to dive deeper into the disease characteristics, headache burden, and acute medication usage of participants in this group. These data may help clinicians caring for people with migraine better anticipate and proactively address patient concerns around preventive medication use in those at risk for declining preventive treatment.

METHODS

Study design and population

OVERCOME (US) is a prospective, longitudinal, multicohort, web‐based survey of adults with migraine in the United States and has been previously described. ³ , ⁴ , ¹⁴ The current analysis is a secondary post hoc analysis of the pooled 2018–2020 baseline cross‐sectional cohorts, focusing on participants who were candidates for migraine‐preventive treatment based on criteria outlined in the 2021 AHS consensus statement. ¹⁵ As depicted in the participant flow in Figure 1, the full cohort was established first as a representative US population sample before identifying those with migraine who were further surveyed about beliefs, behaviors, and outcomes. All participants voluntarily provided electronic informed consent. The study was approved by the Sterling Institutional Review Board (IRB ID #6425–001).

FIGURE 1.

Consort diagram of the full OVERCOME population. This has been previously published. ^{4 a}Targeted sampling to represent the US adult population in terms of key demographic characteristics (age, sex, race, and geography) were applied. ^bConsists of the following: disqualified before consent (n = 2), did not consent (n = 56,559). ^cConsists of the following: did not complete screener (n = 79,014), did not pass data check/inconsistent response (n = 1526), did not meet inclusion criteria (n = 1374). ^dConsists of the following: no/don't remember headache in past 12 months (n = 152,484), did not meet migraine criteria (n = 49,878), over quota (n = 16,254). ^eConsists of the following: did not complete survey (n = 9589), disqualified (n = 25). ICHD‐3, International Classification of Headache Disorders, 3rd edition.

Individuals who reported having had ≥1 headache(s) in the previous 12 months and met the International Classification of Headache Disorders, 3rd edition ¹⁶ criteria for migraine assessed via the validated American Migraine Study/American Migraine Prevalence and Prevention (AMPP) study migraine diagnostic questionnaire ¹⁷ continued to complete the full migraine survey. Every participant was required to answer every question in the survey before moving on to the next, eliminating the possibility of missing data.

A power calculation was performed a priori and was based on precision of percentage of participants with a certain attribute (e.g., percent eligible for preventive migraine treatment). It was estimated that inclusion of 40,000 people with migraine would provide a precision of at least 0.75%. Initially, the planned sample size for OVERCOME was n = 40,000; however, another cohort of ~20,000 participants was added later in an effort to also include individuals on newly available medications for migraine.

Preventive medication eligibility was based on the 2021 AHS Consensus Statement ¹⁵ for determining when preventive medication should be offered, as was defined in our analysis as participants with the following: 6 or more monthly headache days (MHD) regardless of disability, or 4–5 MHD with moderate–severe disability, or 3 MHD with severe disability.

Participants were also asked to indicate whether they had ever taken prescription medication for migraine prevention. Preventive medications included antidepressants (amitriptyline, desvenlafaxine, doxepin, duloxetine, escitalopram oxalate, fluoxetine, imipramine, nortriptyline, paroxetine, sertraline, venlafaxine), antiseizure medications (divalproex, gabapentin, pregabalin, topiramate, valproic acid, zonisamide), cardiovascular medications (atenolol, candesartan, lisinopril, metoprolol, nadolol, nifedipine, propranolol, timolol, verapamil), CGRP monoclonal antibodies (erenumab, fremanezumab, galcanezumab, eptinezumab), gepants (rimegepant for prevention, atogepant), and neurotoxins (abobotulinumtoxinA, onabotulinumtoxinA). Based on their response, they were grouped into Ever‐Takers or Never‐Takers. Only individuals who indicated using any of these treatments for migraine or severe headache were included in the Ever‐Taker group; those who indicated using these treatments for other health care conditions or not using these treatments at all were classified as a Never‐Takers. Never‐Takers were asked to select any reason that applied from a list of reasons as to why they had never taken preventive medications for migraine. Based on the nature of the reasons and the authors' expert opinion, the reasons were grouped into five thematic categories: communication, access, stigma, medication‐related, and other. Questionnaire‐listed response options and categories are detailed in Table 1.

TABLE 1.

OVERCOME (US) survey response options and assigned categories for the question of why an individual had never taken preventive medication for migraine.

Questionnaire response options for “Why have you never taken a prescription medication to prevent or reduce the frequency, severity, or duration of your migraine or severe headaches?”	Category of answer
My migraines/severe headaches are not that bad/serious	Stigma
I did not know there were prescription medications that would prevent/reduce migraine/severe headache	Communication
The other medications I use work well enough	Medication related
I do not like taking prescription medications	Medication related
It costs too much/I do not want to spend the money	Access
I do not want to bother my health care provider	Communication
My doctor told me I did not need a prescription/told me to just use an “over‐the‐counter” medication	Communication
Another health condition prevented me from taking it	Medication related
I am concerned about the side effects of medication/the effects of using them too often	Medication related
I am concerned about it interacting with other medication, alcohol, or recreational drugs	Medication related
(If female) I have been pregnant, trying to become pregnant, or breastfeeding	Medication related
I do not have health insurance/my insurance does not cover medications for migraine/severe headaches	Access
I don't want anyone to think I'm just a complainer	Stigma
I don't want anyone to think I'm trying to get painkillers or opioids	Stigma
I don't want to be labeled as a person with migraine or headache	Stigma
Other reasons	Other

Variables and measures of interest

Sociodemographic, clinical, and migraine‐related characteristics

This analysis reports sociodemographic characteristics including age, sex at birth, race, ethnicity, college degree status, employment status, health insurance, and annual income for each population. Migraine‐related characteristics included years with migraine, monthly headache day frequency, average pain intensity (0–10 scale), and how headaches affect sleep, mood, driving, and exercise. Participants also reported whether they had received a medical diagnosis of cardiovascular comorbidities (including prediabetes, diabetes, high cholesterol, and hypertension; these were rated as “not present,” “1 condition present,” or “≥2 conditions present”) or any comorbidities considered a contraindication to triptan use (including aneurysm, angina, cerebral hemorrhage, claudication, myocardial infarction, stroke, transient ischemic attack, and blood clots in legs/lungs).

Migraine Disability Assessment Scale

Migraine‐related disability was assessed utilizing the five‐item Migraine Disability Assessment Scale (MIDAS). ¹⁸ MIDAS measures the number of days an individual missed or had reduced productivity at work, home, or social events over the previous 3 months due to headaches. MIDAS categories were converted into monthly estimates. The score categories utilized were as follows: 0–5 = little/none, 6–10 = mild, 11–20 = moderate, and ≥21 = severe.

Migraine Interictal Burden Scale‐4

Interictal burden of migraine (burden of migraine between attacks) during the previous 4 weeks was assessed with the four‐item Migraine Interictal Burden Scale‐4 (MIBS‐4). ¹⁹ Each item of the scale contains five responses ranging from “never” to “most or all of the time,” with a total score ranging from 0 to 12. The score categories were utilized as follows: 0 = none, 1–2 = mild, 3–4 = moderate, and ≥5 = severe.

Migraine‐Specific Quality of Life Questionnaire v2.1–Role‐Function Restrictive

Designed to establish how migraine affects daily functioning across three domains, ²⁰ , ²¹ the 7‐item Migraine‐Specific Quality of Life Questionnaire v2.1–Role‐Function Restrictive scale measures how migraine has affected social and work activities over the previous 4 weeks. Each item offers six responses from 1 (“none of the time”) to 6 (“all of the time”). Scores are transformed to a 0–100 scale, with higher scores indicating better role function.

Migraine‐related stigma

Migraine‐related stigma was assessed using the Migraine‐Related Stigma questionnaire. The questions were developed based on review of existing stigma scales and literature, ²² , ²³ experts, and people with migraine and were previously published. ⁴ , ²⁴

Seeking medical care for migraine

Participants were asked to report whether they sought medical care for migraine/severe headache (in‐person, phone, or e‐visit) in the past 12 months at any of the following locations: primary care (including family medicine or internist office/clinic), general neurology, headache specialist, pain specialist, emergency department at a hospital, urgent care center, or retail community/pharmacy clinic. Those with at least one visit were classified as “seeking care—yes” and others as “seeking care—no.” To determine the most specialized level of care sought, these response options were further grouped into “emergency department/urgent care,” “primary care,” “specialist” (neurologist, headache specialist, or pain specialist), or “other”. Consulting at a specialist was considered the most specialized level of care, followed by primary care and emergency department/urgent care.

Statistical analysis

The current analysis utilizes pooled data from the three OVERCOME (US) baseline cohorts fielded in 2018, 2019, and 2020. Sociodemographic characteristics, clinical variables, and migraine‐related outcomes were reported as percent for dichotomous, ordinal, and categorical variables and mean ± standard deviation (SD) for continuous variables.

To assess differences between the groups, standardized mean differences (SMDs) were calculated and classified as follows: SMD values of < 0.2 = no difference, 0.2–0.49 = small difference, 0.5–0.79 = moderate difference, ≥ 0.8 = large difference. ²⁵ SMDs are sample size independent and hence are an appropriate approach for comparing characteristics among groups with large sample size as opposed to p‐values, which may be statistically significant even for small differences that are not clinically meaningful.

SAS Enterprise Guide software version 7.15 (SAS Institute, Cary, NC, USA) was used for all analyses.

RESULTS

Diagnosis and preventive medication use among those seeking care for migraine

Among individuals in OVERCOME (US) who met International Classification of Headache Disorders, 3rd edition criteria for migraine (n = 59,001), approximately one half (51.3%) had sought medical care for migraine in the previous 12 months, approximately one third (36.3%) had sought care and had a migraine diagnosis, and one tenth (10%) had sought care, received a diagnosis of migraine, and were currently taking preventive medications for migraine (Figure 2).

FIGURE 2.

Seeking care, diagnosis, and use of preventive treatment for migraine by monthly headache day group among OVERCOME (US) participants who met International Classification of Headache Disorders, 3rd edition criteria (N = 59,001). [Color figure can be viewed at wileyonlinelibrary.com]

Seeking care, diagnosis, and taking preventive medication for migraine was generally higher among those with higher MHD. Among participants experiencing ≥15 headache days per month, 69.7% had sought care in the previous 12 months, 56.1% had sought care and received a migraine diagnosis, and 19.5% had sought care, received a diagnosis, and were taking preventive treatment for migraine. By contrast, of those with 4–7 headache days per month, 58.2% indicated that they had sought care for migraine, 43.0% had sought care and been diagnosed, and 11.4% had sought care, been diagnosed, and were taking preventive medication for migraine.

Demographics of migraine preventive‐eligible participants

To understand why some people with migraine do not take preventive medications, those who were eligible for preventive medications (Figure 1; n = 22,249) were divided into individuals who indicated ever having taken preventive medications (n = 7715, 34.7%) and those who had not (n = 14,534, 65.3%), hereafter referred to as “Ever‐Takers” and “Never‐Takers,” respectively (Figure 1).

Between the Ever‐Taker and Never‐Taker groups, demographics were similar (SMD < 0.2), including age, years with migraine, sex at birth, ethnicity, race, education, presence of health insurance, and annual household income (Table 2). Cardiovascular comorbidities were higher among Ever‐Takers (33.5% of Ever‐Takers reported two or more cardiovascular comorbidities compared to 23.3% of Never‐Takers, SMD = 0.23), and more Ever‐Takers (28.2%) had a comorbidity that contraindicated triptan use compared to Never‐Takers (15.0%, SMD = 0.32).

TABLE 2.

Demographic and clinical characteristics of participants who have ever taken preventive medication for migraine (Ever‐Takers), participants who have never taken preventive medication (Never‐Takers), and all migraine‐preventive treatment‐eligible participants.

Characteristic	Ever‐Takers, n = 7715	Never‐Takers, n = 14,534	Migraine treatment‐eligible, n = 22,249	SMD (Ever‐Takers vs. Never‐Takers)
Age, years, mean (SD)	42.8 (13.2)	40.3 (14.3)	41.2 (14.0)	0.18
Years with migraine, mean (SD)	20.6 (14.5)	18.4 (14.0)	19.1 (14.2)	0.16
Sex at birth, n (%)
Male	1604 (20.8)	2994 (20.6)	4598 (20.7)	0.00
Female	6111 (79.2)	11,540 (79.4)	17,651 (79.3)	0.00
Ethnicity, n (%)
Non‐Hispanic	6740 (87.4)	12,533 (86.2)	19,273 (86.6)	0.04
Hispanic	752 (9.7)	1464 (10.1)	2216 (10.0)	0.01
Prefer not to answer	223 (2.9)	536 (3.7)	760 (3.4)	0.04
Race, n (%)
Black or African American	522 (6.8)	938 (6.5)	1460 (6.6)	0.01
White or Caucasian	6343 (82.2)	11,953 (82.2)	18,296 (82.2)	—
Asian or Asian American	136 (1.8)	276 (1.9)	412 (1.9)	0.01
Others a	680 (8.8)	1253 (8.6)	1933 (8.7)	0.01
Prefer not to answer	34 (0.4)	114 (0.8)	148 (0.7)	0.05
Annual household income, n (%)
<$50, 000	3858 (49.9)	8095 (55.6)	11,953 (53.8)	0.11
$50, 000 to $99,999	2514 (32.6)	4276 (29.4)	6790 (30.5)	0.07
≥$100,000	1170 (15.2)	1813 (12.4)	2983 (13.3)	0.08
Prefer not to answer	173 (2.2)	350 (2.4)	523 (2.4)	0.01
Education, n (%)
High school or less	1590 (20.6)	3799 (26.1)	5389 (24.2)	0.13
Some college	3379 (43.8)	6412 (44.1)	9791 (44.0)	0.01
Degree	2691 (34.9)	4269 (29.4)	6960 (31.3)	0.12
Prefer not to answer	55 (0.7)	54 (0.4)	109 (0.5)	0.04
Employment status, n (%)
Employed (full/part‐time)	3975 (51.6)	7672 (52.8)	11,647 (52.3)	0.02
Retired or homemaker	1688 (21.9)	3108 (21.4)	4796 (21.5)	0.01
Other	1981 (25.7)	3601 (24.8)	5582 (25.1)	0.02
Prefer not to answer	71 (0.9)	153 (1.1)	224 (1.0)	0.02
Health insurance, n (%)
Yes b	4720 (61.2)	9959 (68.5)	14,679 (66.0)	0.15
No	1642 (21.3)	3214 (22.1)	4856 (21.8)	0.02
Prefer not to answer	1353 (17.5)	1361 (9.4)	2714 (12.2)	0.24
Number of cardiovascular comorbidities, n (%) c
None	3172 (41.1)	7995 (55.0)	11,167 (50.2)	0.28
1	1960 (25.4)	3147 (21.7)	5107 (23.0)	0.09
2+	2583 (33.5)	3392 (23.3)	5975 (26.9)	0.23
Presence of comorbidity that contraindicates triptan use, n (%) d
No	5540 (71.8)	12,352 (85.0)	17,892 (80.4)	0.32
Yes	2175 (28.2)	2182 (15.0)	4357 (19.6)	0.32

Note: SMD was used to compare characteristics among Never‐Takers and Ever‐Takers.

Abbreviation: SMD, standardized mean difference.

^a “Others” included American Indian/Alaska Native, Native Hawaiian/Asian/Pacific Islander, and people who selected multiple options.

^b “Yes” included all health insurance categories: through employer, through spouse's employer, through State Health Exchange, plan purchased directly, Medicaid (MediCal for California residents), Medicare, Medicare Advantage, Veterans Administration or CHAMPUS, TRICARE, and through a parent or legal guardian's employer.

^c These included prediabetes, diabetes, high cholesterol, and hypertension.

^d These included aneurysm, angina, cerebral hemorrhage, claudication, myocardial infarction, stroke, transient ischemic attack, and blood clots in legs/lungs.

Reasons selected for never having taken preventive medication

Of the Never‐Takers (n = 14,534), 49.9% selected only one reason for never having taken preventive medication for migraine, 21.4% selected two reasons, and 28.7% selected ≥3 reasons for never having taken migraine‐preventive medication (Figure 3A).

FIGURE 3.

Reasons selected by participants for not taking preventive medications. (A) The number of reasons selected by participants who have never taken preventive medications (Never‐Takers). (B) Categorized breakdown of the reasons for not taking preventive treatment by the following categories: communication, access, stigma, medication related, and other. OTC, over‐the‐counter. [Color figure can be viewed at wileyonlinelibrary.com]

Selected reasons were grouped into five thematic categories: communication, access, stigma, medication‐related, and other (Figure 3B). Medication‐related reasons were most frequently selected, with the largest percentage of participants (25.5%) indicating “I am concerned about the side effects of medications or effects of using them too often,” 23.3% indicating “I do not like taking prescription medications,” and 20.8% indicating “the other medications I use work well enough.”

Medication access‐related reasons were the second most frequently selected category, with “it costs too much, or I do not want to spend the money” (18.6%) being the most frequently selected reason within that category. In the stigma category, the highest portion of participants indicated “I do not want anyone to think I am trying to get painkillers or opioids” (15.6%). For the communication category, “my doctor told me that I did not need a prescription or just told me to use over‐the‐counter (OTC) medications” was selected by 12.8% of participants.

Migraine‐related, care‐seeking, and treatment‐related characteristics in Ever‐Takers and Never‐Takers

Among all migraine treatment‐eligible participants (n = 22,249), more than one third (35.7%) were Never‐Takers who had ≥8 MHD (Figure 4A), almost one half (49.7%) were Never‐Takers who experienced moderate or severe migraine‐related disability (Figure 4B), and more than one third (39.3%) were Never‐Takers who experienced moderate or severe interictal burden (Figure 4C).

FIGURE 4.

Headache frequency, disability, and burden among participants eligible for preventive medication (N = 22,249). (A) Monthly headache days (MHD), (B) migraine‐related disability (assessed via Migraine Disability Assessment Scale [MIDAS]), and (C) migraine interictal burden (assessed via Migraine Interictal Burden Scale‐4 [MIBS‐4]) among participants who are eligible for preventive medication categorized by preventive medication use status. Never‐Takers = those who have never taken preventive medication for migraine. Ever‐Takers = those who have taken preventive medication for migraine. [Color figure can be viewed at wileyonlinelibrary.com]

Migraine‐related characteristics in Ever‐Takers and Never‐Takers

To better understand differences between the Ever‐Taker and Never‐Taker groups, we compared migraine‐related characteristics, care‐seeking, and other outcome variables between the two groups (Table 3). Generally, Ever‐Takers more frequently reported experiencing 15 or more MHD compared to Never‐Takers (35.1% vs. 25.3%, SMD = 0.21) and also reported higher headache pain severity (7.9 vs. 7.3, SMD = 0.38) and higher interictal burden as assessed by the MIBS‐4 (severe MIBS‐4: 65.2% vs. 43.3%, SMD = 0.45) compared to the Never‐Taker group. Similarly, migraine‐related disability (per MIDAS) was also higher in the Ever‐Taker group (68.7% had severe scores on MIDAS) compared to the Never‐Taker group; however, 53.1% of the Never‐Taker group also experienced severe disability (SMD = 0.32). Although some participants in the Ever‐Taker and Never‐Taker groups had a migraine diagnosis with or without other headache diagnoses, those in the Never‐Taker group were more likely to have a diagnosis of other headaches (23.4% vs. 11.1%, SMD = 0.33) or no diagnosis at all (18.9% vs. 1.6%, SMD = 0.60) compared to the Ever‐Taker group.

TABLE 3.

Migraine‐related characteristics of participants who have ever taken preventive medication for migraine (Ever‐Takers), participants who have never taken preventive medication (Never‐Takers), and all migraine‐preventive treatment‐eligible participants.

Characteristic	Ever‐Takers, n = 7715	Never‐Takers, n = 14,534	Migraine treatment‐eligible, n = 22,249	SMD (Ever‐Takers vs. Never‐Takers)
Monthly headache days, n (%)
0–3	462 (6.0)	899 (6.1)	1361 (6.1)	0.00
4–7	2525 (32.7)	5698 (39.2)	8223 (37.0)	0.14
8–14	2023 (26.2)	4260 (29.3)	6283 (28.2)	0.07
≥15	2705 (35.1)	3677 (25.3)	6382 (28.7)	0.21
Headache pain severity, mean (SD)	7.9 (1.5)	7.3 (1.6)	7.5 (1.6)	0.38
MIBS‐4, n (%)
No interictal burden	746 (9.7)	3198 (22.0)	3944 (17.7)	0.34
Mild interictal burden	838 (10.9)	2602 (17.9)	3440 (15.5)	0.20
Moderate interictal burden	1100 (14.3)	2437 (16.8)	3537 (15.9)	0.07
Severe interictal burden	5031 (65.2)	6297 (43.3)	11,328 (50.9)	0.45
MIDAS, n (%)
Little or no disability	562 (7.3)	2244 (15.4)	2806 (12.6)	0.26
Mild disability	404 (5.2)	1234 (8.5)	1638 (7.4)	0.13
Moderate disability	1451 (18.8)	3342 (23.0)	4793 (21.5)	0.10
Severe disability	5298 (68.7)	7714 (53.1)	13,012 (58.5)	0.32
Migraine Dx class, n (%)
Migraine Dx only	1565 (20.3)	2241 (15.4)	3806 (17.1)	0.13
Other headache Dx only	859 (11.1)	3404 (23.4)	4263 (19.2)	0.33
No migraine or headache Dx	121 (1.6)	2744 (18.9)	2865 (12.9)	0.60
Migraine plus other headache Dx	5170 (67.0)	6145 (42.3)	11,315 (50.9)	0.51

Note: SMD was used to compare characteristics among Never‐Takers and Ever‐Takers.

Abbreviations: Dx, diagnosis; MIBS‐4, Migraine Interictal Burden Scale‐4, MIDAS, Migraine Disability Assessment Scale; SMD, standardized mean difference.

Ever‐Takers and Never‐Takers differed in their care‐seeking behavior, medication use, and experiencing migraine‐related stigma. Never‐Takers more frequently did not seek medical care for migraine compared to Ever‐Takers (44.0% vs. 16.2%, SMD = 0.63; Figure 5A). Conversely, a higher proportion of those in the Ever‐Taker group had seen a specialist in the previous 12 months compared to those in the Never‐Taker group (52.5% vs. 17.6%, SMD = 0.78). The proportion seeking primary care or emergency room/urgent care was similar between groups (SMD < 0.2).

FIGURE 5.

Level of care sought, stigma, and medication use among participants eligible for preventive medication (N = 22,249). (A) The most specialized level of care sought among emergency room/urgent care, primary care, and specialist. (B) Migraine‐related stigma. (C) Use of recommended acute treatment for migraine. (D) Acute medication overuse. (E) Over‐the‐counter medication overuse in participants who have never taken preventive medication (Never‐Takers) and those who have taken preventive medication (Ever‐Takers) for migraine. SMD, standardized mean difference. [Color figure can be viewed at wileyonlinelibrary.com]

Although use of only OTC pain medication was similar in both groups (SMD < 0.2; Figure 5E), those in the Ever‐Taker group were more likely to use recommended acute treatments for migraine (63.9% vs. 32.2%, SMD = 0.67; Figure 5C) but were also more likely to be overusing acute medication for migraine (45.1% vs. 31.6%, SMD = 0.28; Figure 5D). Additionally, a higher proportion of Ever‐Takers reported experiencing higher levels of migraine‐related stigma compared to Never‐Takers (experiencing stigma often or very often: 53.2% vs. 39.1%, SMD = 0.29; Figure 5B).

DISCUSSION

This analysis of OVERCOME (US), a population‐based study of nearly 60,000 individuals living with migraine, aimed to understand why some people who are eligible for migraine‐preventive medication have never taken one, despite clear recommendations by the AHS and International Headache Society. ⁷ , ⁸ , ⁹ This analysis highlights the unmet treatment need that exists in the use of preventive medications for migraine, with the goal of encouraging early conversations about potential reasons that may prevent a person from seeking appropriate management of their migraine.

Among those who were eligible for migraine‐preventive medications, the majority (65.3%) had never taken one (Never‐Taker), despite meeting eligibility criteria due to their disease severity (25.3% experienced ≥15 MHD, 43.3% had severe scores on MIBS‐4, and 53.1% had severe scores on MIDAS). Despite this high migraine burden among Never‐Takers, they were less likely to seek care in general, particularly a specialist, for their headache. Instead, many used OTC medication for pain relief (88.4%) and thought that these medications worked well enough (20.8%). This finding substantiates an important unmet need in the management of migraine within this population. This is particularly important clinically, when it is not clear how a person defines that a medication may “work well.” For some, this may mean slight improvement of their migraine, whereas for others, this could mean headache freedom. It is therefore important for clinicians to educate their patients on the overall goals of the treatment and discuss with each patient in detail what the treatment expectations should be.

Our analysis showed also that Ever‐Takers, those who had ever taken preventive treatment for migraine, showed higher disease burden (35.1% had ≥15 MHD, 65.2% had severe scores on MIBS‐4, and 68.7% had severe scores on MIDAS) and similarly high use of OTC pain‐relieving medication (81.5%). This likely suggests that even though they had taken a preventive medication at some point in their migraine journey, their disease was still not optimally managed. The observed reliance on OTC medications and avoidance of preventive treatments may reflect a lack of understanding about the potential risk of medication overuse, progression to more frequent migraine headaches, and the benefits of preventive therapies. It is important to note, however, that preventive treatments considered in this analysis were both older, nonspecific migraine‐preventive treatments (e.g., antiseizure, antihypertensive, and antidepressant medications), as well as newer, migraine‐specific (CGRP‐targeting) preventive medications. At the time when the OVERCOME (US) baseline survey was fielded, CGRP‐targeted preventive medications were just becoming available, and thus many Ever‐Takers evaluated in this analysis may have been taking the older preventive medications, which are known to be less effective and have more side effects compared to CGRP‐targeting preventive treatments. This may have contributed to the high level of burden seen among Ever‐Takers in this analysis. Ever‐Takers were also more likely to have cardiovascular comorbidities. The reason for this observation could be that individuals with cardiovascular comorbidities were already seeking health care for their cardiovascular condition(s).

Further analysis of the Never‐Taker group showed that common reasons selected for never having taken a preventive medication were concerns about side effects, cost or lack of insurance, and fear of experiencing stigma. These barriers are consistent with previous studies, which found that people with migraine did not perceive the impact of their disease to be significant enough to seek care from a health care professional to access preventive medications for migraine ²⁶ or have a desire to be involved in shared decision‐making over medication use. ²⁷ Perceived migraine‐related stigma among people living with migraine has been well described ²⁸ , ²⁹ , ³⁰ and is known to be correlated with higher disability, higher interictal burden, and lower quality of life. ⁴ It plays a particularly important role in care‐seeking for migraine, as previous data from OVERCOME (US) has shown that those who perceive migraine‐related stigma more often are more likely to hesitate to seek care. ²⁶ An analysis among people without active migraine recently showed that many expressed stigmatizing attitudes toward those living with the disease, in particular that people with migraine should be able to easily treat their condition, have the condition as a result of their own unhealthy behavior, and/or use the condition to get out of family or social commitments. ³¹ , ³²

This high level of stigma in the general population underscores the fear of experiencing stigma that is reflected among Never‐Takers in this analysis who stated that they had never taken preventive medication because they did not want to be perceived as a complainer and did not want others to think that they were trying to get pain medication. It is critical that this public stigma toward people with migraine be addressed to make them feel safe and listened to when they seek care and consider appropriate management options for their migraine.

Overall, the population evaluated in this analysis represents a group that needs effective education about migraine and appropriate management options. Among Never‐Takers, 11.4% stated that they did not know prescription medications for migraine were available, and 12.8% noted that their doctor told them they did not need a prescription medication and should just use OTC pain relievers. The present analysis provides an opportunity for health care providers to identify and consider engaging this high‐need population. This could include preemptively addressing concerns of candidates for preventive migraine treatment, particularly medication side effects, cost, and fear of stigma. In addition, public education about migraine symptoms and available treatment options may help to increase the number of individuals who are seeking care for their headache symptoms.

Previous publications from the OVERCOME study and other population‐based surveys report that care‐seeking remains low among people with migraine, ¹⁴ who as a population have many unmet needs. ³ The Never‐Taker population represents a participant group of high unmet need and would benefit from the introduction of preventive medication with timely engagement with health care providers. Specifically, reducing their barriers will require targeted efforts to enhance access to migraine care and promote proactive migraine management. Learnings from other therapeutic areas, such as cardiology, have shown the benefit of adopting different measures to reach this population. Initiatives driven by pharmacist‐led community settings such as pharmacies ³³ or social setting like barbershops ³⁴ have a positive impact on patient care in these fields and contribute positively to the education of the patients. ³³ , ³⁴ This may be beneficial in the field of neurology to help patients with migraine understand the benefits and side effects of preventive medication. However, future research should evaluate the effectiveness of alternative engagement strategies in improving preventive medication uptake and outcomes within this population. Further studies could also dive into more detail on current versus past use of preventive medications among those who had reported ever using preventives. This has been explored with regard to CGRP‐targeted treatment in the OVERCOME (US) population. ³⁵

The OVERCOME study has many strengths, including capturing a US representative population and inclusion based on the AMPP criteria allowing capture of difficult‐to‐track participants with an array of presentations. ³ The study also collected, where possible, previously validated assessments such as MIDAS, MIBS‐4, and the American Migraine Study/AMPP migraine diagnostic questionnaire. Despite this, population sampling techniques leave room for error, and the sample was approximated through quota sampling rather than random sampling. The survey was also internet‐based, rendering some potential participants unable to respond based on socioeconomic background or limited English fluency. The OVERCOME study seems to capture a population of higher reported MHD and higher reported disability compared to prior population‐based studies. ³⁶ This may be due to the accessibility of internet surveying, or because participants self‐reported throughout the survey, and were not validated by medical professionals.

Due to the limitations of the internet survey format, participants were unable to input responses not listed into survey responses. This means that some medications they may have been taking that were not listed in the survey as a response option will not have been captured. Furthermore, this analysis focused on prescription medication for migraine prevention and did not account for those who may have had sufficient effect using other modalities, such as injections, stimulators, behavioral treatments, or supplements. Also, other reasons for not taking preventive medication were not captured. It is also possible that some of these participants may have been offered preventive medications by their health care providers and declined. This level of information was not recorded in the survey format and may have provided further insight into the population. Furthermore, although item nonresponse was minimized by requiring individuals to respond to each question in the survey before moving on to the next question, selection bias could still arise from differential participation or dropout rates among individuals with varying characteristics or migraine severity, potentially affecting the final cohort's representativeness. Lastly, as baseline data were collected over 3 years (2018–2020), changes in the environment (e.g., contemporaneous CGRP uptake, COVID‐19 influences on care‐seeking) may have differed between the individual cohorts.

CONCLUSION

In summary, this study emphasizes the incredible unmet treatment need in migraine, especially among individuals who are eligible for preventive treatment based on their disease severity and disability. ⁸ , ¹⁰ , ³⁷ The current analysis characterizes individuals who are eligible for prescription medication for migraine prevention but have never taken one based on several reasons. This highlights that individuals' reasons for not taking migraine preventives are mostly due to concerns about side effects, cost or lack of insurance, and fear of experiencing stigma. Importantly, as shown recently, ³⁸ , ³⁹ never having taken a preventive treatment for migraine increased the risk of migraine progression over 1 year; hence, appropriate intervention in the form of effective recommended acute and preventive treatments for migraine can improve a person's quality of life if implemented early, ⁴⁰ , ⁴¹ as well as increasing work productivity and social activity. ⁴² , ⁴³ This is particularly important at a time in which effective and safe treatments for prevention are available and recommended first‐line. ⁷

AUTHOR CONTRIBUTIONS

Jessica Ailani: Conceptualization; formal analysis; writing – review and editing. Rose Okonkwo: Data curation; formal analysis; writing – original draft; writing – review and editing. Elizabeth Johnston: Conceptualization; writing – review and editing. Helen Hochstetler: Visualization; writing – review and editing. Betzaida Martinez: Formal analysis; visualization; writing – review and editing. Anthony Zagar: Data curation; formal analysis; validation; visualization; writing – original draft; writing – review and editing. Robert A. Nicholson: Conceptualization; data curation; visualization; writing – review and editing. Bert B. Vargas: Formal analysis; visualization; writing – review and editing. E. Jolanda Muenzel: Conceptualization; data curation; visualization; writing – original draft; writing – review and editing. Richard B. Lipton: Conceptualization; data curation; visualization; writing – review and editing.

CONFLICT OF INTEREST STATEMENT

Jessica Ailani: Consulting (honoraria): AbbVie, Aeon, Dr. Reddy, Eli‐Lilly, Lundbeck, Linpharma, Ipsen, Merz, Pfizer, Neurolief, Gore, Satsuma, Vectura Fertin, Tonix; clinical trials (grant to institution): Parema, Ipsen, Lundbeck, AbbVie, Pfizer, Merz; clinical trials (grants to principal investigator): Mi‐Helper; editorial boards/steering committees: SELF Magazine (medical editor). Rose Okonkwo, Elizabeth Johnston, Helen Hochstetler, Betzaida Martinez, Anthony Zagar, Robert A. Nicholson, Bert B. Vargas, and E. Jolanda Muenzel: Employee/shareholder: Eli Lilly and Company. Richard B. Lipton: Stock or options: Manistee Health; consultant, advisory board member, honoraria, or research support: AbbVie (Allergan), American Academy of Neurology, American Headache Society, Aeon, Amgen, Biohaven, Biovision, Boston, Dr. Reddy’s (Promius), Electrocore, Eli Lilly and Company, Equinox, GlaxoSmithKline, Grifols, Lundbeck, Merck, Pernix, Pfizer, Teva, Vector, Vedanta.