Open Access

Migraine strikes study: factors in patients’ decision to treat early

The Journal of Headache and Pain200910:91

https://doi.org/10.1007/s10194-008-0091-1

Received: 17 October 2008

Accepted: 5 December 2008

Published: 10 January 2009

Abstract

To describe factors associated with early treatment of migraine, and to examine reasons patients do not treat early, this cross-sectional observational study email-recruited migraineurs ≥18-years-old who were currently prescribed acute migraine medication. Within 24 h of migraine resolution, eligible patients completed an online migraine strikes questionnaire which addressed pain severity, associated symptoms, and other variables including reasons for not treating early. Results reported were descriptive. Among 1,044 evaluable patients, early treatment was significantly associated with several factors such as leisure activity at onset (OR 1.32, = 0.010), photophobia (OR 1.39, P = 0.013), diagnosis of migraine with aura (OR 1.36, P = 0.004), and other factors. Among 840 patients who reported wanting to treat earlier desire to reserve medication for a severe migraine was the most common reason given for not doing so (51.2%). Overcoming these factors may facilitate earlier migraine treatment.

Keywords

MigraineTriptansEarly treatmentHeadache

Introduction

Migraine is a disabling disorder which affects about 30 million people in the United States [1]. Much research has been reported on the risk factors, triggers, premonitory symptoms, and concomitant symptoms associated with migraine attacks, such as nausea, photophobia, and visual disturbances [27]. Clinical and observational data from patients taking triptans have consistently demonstrated the benefits of treating early (i.e., at the onset of pain) [813], but despite this evidence, 69% of migraineurs report that they would not treat their migraine pain as soon as they feel pain [14]. In addition to issues regarding efficacy and potential side effects of triptans, other factors may affect patients’ ability or willingness to treat early. This study investigated other such external environmental factors which may influence patients’ decision to treat early, and factors which may prevent patients from treating a migraine as early as they desire.

Methods

Study design

This cross-sectional observational survey study [migraine strikes study (MIST)] was conducted from November to December 2006. A central Institutional Review Board (IRB) (ESSEX IRB, Lebanon, NJ) approved the study protocol, survey instrument, and the statement of informed consent. Consumer Health Sciences International (CHS), a contract research organization, provided each patient with full and adequate information about the objective and procedures of the study. Compensation was provided by the Lightspeed Consumer Panel to participants in the form of ceremonial “points,” which patients were eligible to redeem at the panel’s website for various retail products such as online gift certificates, music downloads, and kitchen appliances.

Patients and procedures

To be eligible for the study, patients had to be at least 18-years-old and fluent in English. Patients had to have been diagnosed with migraine (with or without aura) by a physician, and had to have a current prescription for acute migraine medication(s). Patients were contacted via the internet to participate in the study. In an initial invitation email circular, patients were informed about the purpose and design of the study. If patients experienced a migraine attack within 7 days after they received the invitation email, they were invited to report their migraine experience within 24 h of its resolution. Patients accessed a password-protected web link attached in the initial email circular. Those who gave consent to participate were allowed to proceed to the migraine strikes questionnaire (Electronic supplementary Appendix 1).

In the migraine strikes questionnaire, participants reported circumstances of the migraine attack that occurred within the past 24 h, such as the time and location in which the migraine occurred, and details about any therapy the patient used to treat the migraine. Information about migraine and other headache history was also collected. Consumer Health Sciences International fielded the study by setting up data programming for the migraine strikes questionnaire, tracking and monitoring patient’s compliance with completing the questionnaire, and sending email reminders and thank you notes to participants. Consumer Health Sciences International collected the data, which was accessible only to research team members. Upon stripping the data of all patient-identifying information, CHS provided the sponsor (Merck and Co., Inc.) with a cumulative data base in SAS format, which contained all information on headache experience for the migraine strikes questionnaire.

Statistical methods

To assess the factors associated with early treatment patterns, questions evaluated included pain severity and associated symptoms, activities at onset of migraine, timing and selection of medication, physician-level variables, and co-morbid headache diagnoses. Logistic regression was used to identify factors associated with early treatment; adjustments were made for covariates. Chi-square and Student’s t tests were used for univariate analysis, where appropriate, and a binary logistic regression model using backward elimination was used for multivariate analyses. The dependent variable was “Treat early” (defined as treating within 1 h of headache onset), and candidate independent variables were those determined to be clinically relevant and/or which had a trend toward significance (P < 0.1) on univariate analysis.

To assess reasons for not treating early even when desiring to do so, data were collected for questions about whether the attack was treated when desired (“Did you want to take your medication earlier than you actually did?”), factors that may have prevented earlier treatment (“What prevented you from taking the medication sooner?”), physician specialty (“Are you currently seeing any of the following physicians for your headaches?”), physician education regarding early treatment (“Has your physician ever instructed or advised you to take your migraine medications ‘early’ when the pain is still ‘mild’?”) and satisfaction with pain relief (“How satisfied were you with the pain relief you achieved with the medication you took for this migraine attack?”).

Descriptive statistics were used to examine factors potentially hindering earlier treatment and Chi-square tests were used to examine univariate associations.

Results

Early treatment

Table 1 shows patient demographics and baseline characteristics. Of the 2,001 migraineurs who participated (mean age 42.7, 74.2% female), 1,856 (92.8%) reported at least one associated symptom (nausea, vomiting, photophobia, or phonophobia) at headache onset during their attack, providing supporting evidence that participants were correct in their ascertainment of migraine. Of the 1,752 participants who provided information on treatment timing, 1,044 treated their migraine early (i.e., within 1 h of headache onset) (59.6%).
Table 1

Patient demographics and baseline characteristics

Baseline characteristics

N = 2,001

Mean age, years (SD)

42.7 (11.7)

Female (%)

74.2

Associated symptoms (%)

 Photophobia

81.0

 Phonophobia

77.7

 Nausea

58.4

 Vomiting

11.7

 At least one of the above

92.8

Initial treatment chosen (%)

 None

0.0

 Over-the-counter (OTC)

12.6

 Triptan

62.3

 Other prescription

25.1

Age at first migraine, years (SD)

23.2 (11.3)

Age at diagnosis, years (SD)

27.1 (11.0)

Migraine type (%)

 With aura

44.1

 Without aura

55.9

General migraine severity (%)

 Mild

2.9

 Moderate

42.1

 Severe

55.0

Migraines per month (%)

 ≤2

34.8

 3–4

25.0

 5–8

19.5

 ≥9

20.7

Employment status (%)

 Full-time

48.2

 Part-time

12.4

 Student or unemployed

39.4

Education level (%)

 High school or less

17.5

 Some college

38.8

 College or graduate degree

43.7

Prescription drug coverage (%)

 Yes

87.4

 No

12.6

Other headache type

 Tension-type

79.4

 Cluster

36.4

 Chronic daily (>15/month)

33.6

 Sinus

66.6

SD standard deviation

Univariate analysis (Table 2) showed that factors significantly positively associated with treating early included increasing age, male gender, doing something for fun/leisure at onset, presence of photophobia, diagnosis of migraine with aura, physician advice about early treatment, and regular adherence to physician’s instructions. Patients were significantly less likely to treat early if they were at work or school at the time of migraine onset, or if they had co-morbid tension-type and/or chronic daily headache.
Table 2

Univariate analysis: early treatment

Demographic characteristics

Treat early (%)

N = 1,752

Age

 <30, n = 293

52.9

 30–39, n = 469

56.3

 40–49, n = 515

59.8

 50–59, n = 339

64.3

 ≥60, n = 136

72.8

 P-value

<0.001

Gender

 Female, n = 1,291

58.0

 Male, n = 461

64.0

 P-value

0.025

Employment status

 Full-time, n = 847

60.1

 Part-time, n = 216

55.1

 Student or unemployed, n = 689

60.4

 P-value

0.353

Education level

 High school degree or less, n = 305

57.1

 Some college, n = 681

59.5

 College graduate, n = 766

60.7

 P-value

0.544

Prescription drug insurance

 Yes, n = 1,551

60.2

 No, n = 201

54.7

 P-value

0.135

When your migraine began, were you…?

 At work or school?

  Yes, n = 248

52.0

  No, n = 1,504

60.8

  P-value

0.009

 Doing something for fun or leisure?

  Yes, n = 676

63.3

  No, n = 1,076

57.3

  P-value

0.012

 In a private place?

  Yes, n = 1,256

60.4

  No, n = 496

57.7

  P-value

0.302

 A critical member of the task at hand?

  Yes, n = 779

58.3

  No, n = 973

60.6

  P-value

0.318

 Doing something as part of your daily routine?

  Yes, n = 1,316

59.6

  No, n = 436

59.6

  P-value

0.963

Associated symptoms

 Nausea

  Yes, n = 1,042

60.4

  No, n = 710

58.5

  P-value

0.423

 Vomiting

  Yes, n = 212

57.6

  No, n = 1,540

59.9

  P-value

0.518

 Photophobia

  Yes, n = 1,428

61.1

  No, n = 324

53.1

  P-value

0.008

 Phonophobia

  Yes, n = 1,364

60.1

  No, n = 388

57.7

  P-value

0.398

Medication

 First medication used

  Triptan, n = 1,091

60.1

  Non-triptan prescription, n = 440

60.2

  OTC, n = 221

55.7

  P-value

0.444

Migraine characteristics

 Migraine type

  With aura, n = 781

63.8

  Without aura, n = 971

56.2

  P-value

0.001

 Migraines per month

  ≤2, n = 593

59.4

  3–4, n = 439

59.7

  5–8, n = 352

61.7

  ≥9, n = 368

52.9

  P-value

0.782

 General migraine severity

  Mild, n = 40

60.0

  Moderate, n = 714

59.8

  Severe, n = 464

59.4

  P-value

0.986

 Presence of prodromal symptoms

  Yes, n = 942

58.6

  No, n = 717

61.5

  Do not know, n = 93

54.8

  P-value

0.309

Other headache type

 Tension-type

  Yes, n = 1,384

57.7

  No, n = 368

66.9

  P-value

0.001

 Cluster

  Yes, n = 646

60.4

  No, n = 1,106

59.1

  P-value

0.610

 Chronic daily (>15/month)

  Yes, n = 583

54.6

  No, n = 1,169

62.1

  P-value

0.002

 Sinus

  Yes, n = 1,153

58.5

  No, n = 599

61.8

  P-value

0.180

Physician-related variable

 Physician specialty

  Primary care, n = 1,260

750 (59.5)

  Specialist, n = 403

245 (60.8)

  Others, n = 89

49 (55.1)

  P-value

0.605

 Physician advisement regarding treat early

  Yes, n = 1,447

889 (61.4)

  No, n = 305

155 (50.8)

  P-value

<0.001

 Regular adherence to physician’s instructions

  Rarely/sometimes, n = 343

146 (42.6)

  Most of the time/always, n = 1,361

871 (64.0)

  P-value

<0.001

After adjusting for age, gender, and pain severity, multivariate analysis showed that early treatment was significantly positively associated with leisure activity at onset [odds ratio (OR) 1.32, P = 0.010], photophobia (OR 1.39, P = 0.013), diagnosis of migraine with aura (OR 1.36, P = 0.004), prior early treatment instruction by a physician (OR 1.75, < 0.001), regular adherence to physicians’ instructions (OR 2.31, P < 0.001). Early treatment was significantly negatively associated with co-morbid tension and/or chronic daily headaches (OR 0.68, P = 0.004 and OR 0.78, P = 0.022, respectively) (Table 3). Being at work or school at onset did not retain a significant association with early treatment in the multivariate analysis.
Table 3

Multivariate analysis: early treatment

Variable

Adjusted OR

95% CI

P-value

Age (continuous)

1.02

1.01–1.03

<0.001

Female

0.81

0.64–1.02

0.071

Pain severity at onset (none, mild, moderate, severe)

0.89

0.77–1.02

0.096

Leisure activity at onset

1.32

1.07–1.62

0.010

Photophobia

1.39

1.07–1.81

0.013

Migraine with aura (vs. all others)

1.36

1.10–1.67

0.004

Tension headaches

0.68

0.53–0.89

0.004

Chronic daily headaches

0.78

0.63–0.96

0.022

Physician early treatment instruction

1.75

1.33–2.30

<0.001

Regular adherence to physician’s instructions (most of the time/always vs. rarely/sometimes)

2.31

1.80–2.96

<0.001

OR odds ratio

Reasons for not treating earlier

Of the 2,001 migraineurs who completed the survey, 840 (42.0%) reported willingness to treat their migraine earlier. Among the 764 patients who wanted to treat earlier and who also provided information on treatment timing, 379 (49.6%) treated within 1 h of headache onset, compared with 67.3% of patients who did not want to treat earlier (P < 0.001). Of those who desired earlier treatment, the median time to treatment was 60 min, compared with 20 min among patients who did not want to treat earlier (P < 0.001) (Fig. 1). The most commonly stated reason for not treating when wanted was the desire to reserve medication for a severe migraine (51.2%). Other reasons included not having medication on hand (34.9%), health plan quantity limit on drugs (27.0%), inability to swallow because of nausea (18.7%), not having access to fluid (17.3%), and not having privacy (13.5%) (Fig. 2).
Fig. 1

Time from headache onset to first dose of medication.* < 0.001 for difference between medians

Fig. 2

Factors preventing early treatment among those who sought it (= 840)

Desire to treat earlier was not significantly associated with physician education regarding early treatment; 82.1% among those desiring to treat early reported having been educated about early treatment by their physician versus 80.3% among those who did not want to treat earlier (P = 0.293). Likewise, between-group differences were not significant based on physician specialty (71.4 vs. 70.8% seeing a primary care physician, P = 0.760; 22.7 vs. 22.1% seeing a headache specialist, P = 0.750; and 5.8 vs. 7.1% seeing other or no physician, P = 0.273). Desire to treat earlier was associated with dissatisfaction with pain relief (18.6 vs. 14.3%, P = 0.015).

Discussion

Early treatment of migraine has been shown to be particularly beneficial compared with treating >1 h after onset of pain, but it is known that not all patients take advantage of this benefit. To elucidate the possible reasons, which might then indicate paths forward in terms of helping patients maximize the benefit of migraine treatment, data from this questionnaire were assessed to answer two main questions about early treatment among a sample population of migraineurs. First, the analysis assessed what proportion of patients treated their headache within 1 h of onset; second, among those who did not treat their migraine as early as they wanted to, the data were evaluated to help determine why not. Within 24 h of resolution of a headache, patients reported the circumstances of the migraine attack, including their physical location and time when the migraine occurred, their activities at migraine onset, and medication use. Patients also provided information on their personal migraine history and its treatment, presence of migraine-associated symptoms, and current headache frequency and severity.

The analysis found that more than half (approximately 60%) of patients treated their headache within an hour of its onset; thus, about 40% of patients waited more than an hour after the onset of pain to take therapy for the migraine. These results are in accord with those of Gallagher et al. [15], who also reported that prescription medications were delayed in 40% of migraine episodes, and Foley et al. [14], who reported that migraine medications were delayed by 49% of patients. These findings clearly indicate a need to address potential reasons for delays in treatment, in order to ensure that as many patients as possible are able to optimize their use of treatment.

To obviate unwanted delays, it is useful to understand specific reasons that patients, who may indeed know that early treatment is beneficial, nevertheless may not be taking their medication as early as possible. About 42% of patients indicated that they wanted to treat their migraine earlier than they actually did. Of those who expressed desire for earlier treatment, only about half treated their most recent migraine within an hour. The desire to reserve medication for a severe migraine was the most common reason given for delaying treatment, reported in fully half the patients who wanted to treat earlier. The next most common reason, reported in one-third of patients, was not having the medication on hand, and about one quarter of the patients said that their health plan quantity limit on drugs prevented them from treating as early as they wanted; additional reasons included nausea interfering with the ability to swallow, lack of access to fluid, and lack of privacy.

These results are in general agreement with the small amount of prior work that has previously been done in this area. In the aforementioned survey by Gallagher et al. [15], the authors reported that the most common reasons for delaying prescription medication use were medication access issues, limited numbers of pills, difficulties in functioning, and concerns about adverse events. Interestingly, Gallagher did not report on what we found to be the most commonly reported reason for delay: the desire to reserve medications for severe migraine. This commonly cited reason for treatment delay was also observed by Foley et al. [14]. Of patients who reported delayed treatment, 70% cited wanting to wait and see if the attack was really a migraine attack as a reason for the delay, and 46% cited wanting to take medication only for a severe attack.

Despite the observed high rate of reported migraine-associated symptoms at headache onset, we observed that patients with other co-morbid primary headaches were more likely to delay treatment. Epidemiologic research has shown that up to 90% of migraine patients also suffer from tension-type headache [16] and migraine and tension-type headaches share many important similarities [17]. To optimize migraine care, physicians should routinely educate their patients about early migraine symptoms, particularly for patients recently diagnosed with migraine, in addition to prescribing appropriate medications. Both Gallagher and Foley also found concern about medication side effects as a factor in treatment delay, reported by 15 and 37%, respectively. Our study did not directly explore fear of side effects as a potential reason for delaying treatment.

Some of the stated reasons for treatment delay can be addressed with practical adjustments, such as carrying medication when leaving home. Alternative formulations (e.g., rapidly dissolving tablet, nasal spray) can be considered in patients prone to migraine-associated nausea or who are frequently on the go and may not have continuous access to water or other liquids. However, solutions to the most commonly reported obstacles are not straightforward. For example, in patients with other types of headaches and/or chronic daily headache, the advantages and disadvantages of early treatment must be carefully weighed, bearing in mind that frequent use of certain acute migraine medications may predispose to medication overuse headache and chronic migraine [18].

Several factors act as limitations on the analysis of reasons for delayed treatment. It is possible that the retrospective desire to treat earlier may be influenced by severity of headache pain and/or any associated symptoms the patient may have experienced. Thus, the wish itself to treat earlier may not have emerged until after the first hour of migraine. Moreover, the survey does not account for the fact that reasons for not treating may vary by attack, within the same patient. For example, on two different occasions in which a migraine begins, a patient may have considerably easier access to medication in one situation versus the other. In addition, other practical considerations that could interfere with timely treatment may not have been captured by this survey.

In summary, a large proportion of migraine patients did not treat their headache within the first hour of headache onset. The study found that the most common reasons for the delay in treatment were (1) a wish to reserve the medication for a severe migraine, and (2) lack of easy access to the medication. Finding ways to overcome these and other practical obstacles preventing patients from medicating early may help a greater number of patients to relieve their migraines faster.

Notes

Abbreviations

MIST: 

Migraine strikes study

IRB: 

Institutional Review Board

CHS: 

Consumer Health Sciences International

OR: 

Odds ratio

SD: 

Standard deviation

Declarations

Acknowledgments

Merck and Co., Inc. funded the study. Drs. Golden and Hu are employees of Merck and Co., Inc. who may potentially own stock and/or hold stock options in the company. Dr. Evans was an employee of Merck during the conduct of this study and the writing of this manuscript.

Conflict of interest

Drs. Golden and Hu are employees of Merck and Co., Inc., and may own stock or stock options in the company. Dr. Evans was an employee of Merck during the conduct of this trial and the writing of this manuscript, and may own stock in the company.

Authors’ Affiliations

(1)
Outcomes Research Department, Global Human Health, Merck and Co., Inc.
(2)
Medical Communications, Merck Research Laboratories

References

  1. Lipton RB, Stewart WF, Diamond S, Diamond ML, Reed M (2001) Prevalence and burden of migraine in the United States: data from the American Migraine Study II. Headache 41:646–657View ArticlePubMedGoogle Scholar
  2. Kelman L (2005) Validation of the classification of migraine with aura (IHS A1.1) proposed in ICHD-2. Headache 45:1339–1344View ArticlePubMedGoogle Scholar
  3. Martin VT, Penzien DB, Houle TT, Andrew ME, Lofland KR (2005) The predictive value of abbreviated migraine diagnostic criteria. Headache 45:1102–1112View ArticlePubMedGoogle Scholar
  4. Mulleners WM, Aurora SK, Chronicle EP, Stewart R, Gopal S, Koehler PJ (2001) Self-reported photophobic symptoms in migraineurs and controls are reliable and predict diagnostic category accurately. Headache 41:31–39View ArticlePubMedGoogle Scholar
  5. Silberstein SD (1995) Migraine symptoms: results of a survey of self-reported migraineurs. Headache 35:387–396View ArticlePubMedGoogle Scholar
  6. Giffin NJ, Ruggiero L, Lipton RB, Silberstein SD, Tvedskov JF, Olesen J et al (2003) Premonitory symptoms in migraine. An electronic diary study. Neurology 60:935–940View ArticlePubMedGoogle Scholar
  7. Kelman L (2004) The premonitory symptoms (prodrome): a tertiary care study of 893 migraineurs. Headache 44:865–872View ArticlePubMedGoogle Scholar
  8. Mannix L (2008) A review of the 5HT1B/1D agonist rizatriptan: update on recent research and implications for the future. Expert Opin Pharmacother 9:1001–1011View ArticlePubMedGoogle Scholar
  9. Ng-Mak DS, Hu XH, Chen Y, Ma L, Solomon G (2007) Times to pain relief and pain freedom with rizatriptan 10 mg and other oral triptans. Int J Clin Pract 61:1091–1111PubMed CentralView ArticlePubMedGoogle Scholar
  10. Moschiano F, D’Amico D, Allais G, Rigamonti A, Melzi P, Schieroni F et al (2005) Early triptan intervention in migraine: an overview. Neurol Sci 26(Suppl 2):s108–s110View ArticlePubMedGoogle Scholar
  11. Hu XH, Raskin NH, Cowan R, Markson LE, Berger ML, On behalf of The United States Migraine Study Protocol (USMAP) Group (2002) Treatment of migraine with rizatriptan: when to take the medication. Headache 42:16–20View ArticlePubMedGoogle Scholar
  12. Cady RK, Sheftell F, Lipton RB, O’Quinn S, Jones M, Putnam DG et al (2000) Effect of early intervention with sumatriptan on migraine pain: retrospective analyses of data from three clinical trials. Clin Ther 22:1035–1048View ArticlePubMedGoogle Scholar
  13. Cady R, Martin V, Mauskop A, Rodgers A, Hustad CM, Ramsey KE et al (2006) Efficacy of rizatriptan 10 mg administered early in a migraine attack. Headache 46:914–924View ArticlePubMedGoogle Scholar
  14. Foley KA, Cady R, Martin V, Adelman J, Diamond M, Bell CF et al (2005) Treating early versus treating mild: timing of migraine prescription medications among patients with diagnosed migraine. Headache 45:538–545View ArticlePubMedGoogle Scholar
  15. Gallagher RM, Kunkel R (2003) Migraine medication attributes important for patient compliance: concerns about side effects may delay treatment. Headache 43:36–43View ArticlePubMedGoogle Scholar
  16. Lyngberg AC, Rasmussen BK, Jørgensen T, Jensen R (2005) Has the prevalence of migraine and tension-type headache changed over a 12-year period? A Danish population survey. Eur J Epidemiol 20:243–249View ArticlePubMedGoogle Scholar
  17. Cady R, Schreiber C, Farmer K, Sheftell F (2002) Primary headaches: a convergence hypothesis. Headache 42:204–216View ArticlePubMedGoogle Scholar
  18. Dodick D, Freitag F (2006) Evidence-based understanding of medication-overuse headache: clinical implications. Headache 46(Suppl 4):S202–S211View ArticlePubMedGoogle Scholar

Copyright

© Springer-Verlag 2008