Abstract
Methods: SMILE was an observational study carried out in France among office-based general practitioners (GPs) and neurologists from November 2005 to July 2006 to assess the determinants of prescription of migraine preventive therapy in primary care medicine. A total of 1467 GPs and 83 neurologists were included, treating 5417 and 248 migraine sufferers, respectively.
Results: The main factors leading physicians to deem a patient eligible for preventive treatment were perceived medication overuse and frequency of headaches, and secondarily, severity of headaches and functional impact. On the other hand, patient satisfaction with the acute treatment of attacks and triptan use, and secondarily, a long migraine history were found to influence patient eligibility negatively.
Discussion/conclusion: Noticeably, psychiatric disorders (anxiety, stress) did not appear, aside from somatic factors, among the determinants that significantly influence physicians' judgment about the option of establishing a preventive treatment. However, they are important features of migraine condition and should be listed among the factors guiding choices about migraine preventive therapy.
Introduction
Migraine is no longer considered just an episodic condition but rather a chronic disease state that can start early in a patient's lifetime and last until late in life (1). Therefore, a comprehensive treatment plan includes acute attack treatment to relieve pain and impairment as well as long-term preventive therapy to reduce attack frequency, severity and duration (2). However, epidemiologic studies carried out in the United States (AMS-II, AMPP) and Europe (FRAMIG 2000) indicate that only a small minority of migraine patients (3–13%) receive preventive therapy (3–6), although preventive medication would be indicated for a much larger proportion of patients according to migraine management guidelines (6).
Migraine is heterogeneous in frequency, duration and disabling effects. Some migraine sufferers have fewer than one attack per month while others have one or more attacks per week (7). Guidelines drawn up by the US Headache Consortium stipulate that prevention should be considered for those patients whose migraine has a substantial impact on their lives—that is, those patients whose recurring migraines interfere with their daily routines despite acute treatment and who have frequent headaches (8). French guidelines similarly specify that institution of preventive treatment should be guided by the frequency and severity of migraine attacks, impact on family, social and professional life, and the use of more than six to eight acute treatments in a month even if they are effective—with the aim of preventing medication overuse (9). More recently, the World Health Organization stated that prophylactic therapy is indicated when attacks cause disability on two or more days per month and optimized acute therapy does not prevent this, and when the patient is willing to take daily medication. Other indications for prophylaxis are for risk of frequent use/overuse of acute therapy even when it is effective and for children with frequent absences from school (9). Finally, the goals of migraine preventive therapy are to (i) reduce attack frequency, severity and duration; (ii) improve responsiveness to treatment of acute attacks; (iii) improve function and reduce disability (8); and (iv) reduce the occurrence of aura.
Although several studies have investigated the use of migraine preventive treatments (see above), to the best of our knowledge only one (10), published only in German, investigated the actual determinants leading physicians to prescribe preventive therapy in everyday clinical practice. The objectives of the SMILE study, which was carried out in France in 2005 to 2006 among office-based general practitioners (GPs) and neurologists, were to compare the characteristics of patients prescribed or not prescribed a preventive therapy and to determine the main factors that determine the prescription of migraine preventive therapy in primary care medicine.
Methods
SMILE was an office-based, observational study carried out in France from November 2005 to July 2006 among GPs and neurologists by the TNS Healthcare Polling Institute. Four thousand physicians were recruited at random by mail or telephone from a physician database. Each physician agreeing to participate was to include, from the start date of the study, the first four patients consulting for migraine who were aged 18 years or older and not receiving ongoing preventive treatment. Only patients meeting the diagnostic criteria for strict or probable migraine according to the 2004 International Classification of Headache Disorders (ICHD-II) were included in the study (11).
Physicians completed a questionnaire that queried about personal demography, professional activity and the criteria commonly used for instituting preventive treatment. In addition, for each migraine patient included, physicians completed a patient sheet that captured information about the physician's perceptions of the following: migraine characteristics (frequency of attacks, severity of attacks, functional impact), patients' anxiety, existence of stress as a possible factor in triggering attacks, patients' satisfaction with the acute treatment of attacks, possible medication overuse, the patient's need for a preventive treatment, and about the treatment prescribed at the end of the consultation. Severity of migraine attacks, patients' anxiety and functional impact were evaluated subjectively by the physicians on scales of 0 to 10.
A patient's eligibility for preventive treatment was determined following a discussion between the doctor and patient, on the basis of the physician's perception of these criteria: estimated number of headache days/month, overuse of acute treatments, severity of headaches, patient anxiety, stress as a triggering factor for migraine, functional impact of migraine and patient satisfaction with the acute treatment of migraine.
Independently of the physician, each patient completed a questionnaire that queried socio-demography, the presence of ICHD-II migraine diagnostic criteria, migraine history, medications used for the treatment of migraine attacks and their effectiveness.
Quantitative data are expressed as the mean and standard deviation (SD) and compared between groups using the Student's t-test. Qualitative data are expressed by the numbers and relative proportions in the corresponding categories and compared between groups using the Z test. The factors significantly associated with physicians' estimation of patients' need for a preventive treatment have been identified by regression analysis. Statistical results were considered significant at p < .05.
Results
Of the 4052 physicians recruited, 1618 agreed to participate in the study (41% and 29% of GPs and neurologists invited, respectively), treating a total of 6109 potential study patients. After exclusion of 329 patients for incomplete files (one or more missing questionnaires), 17 patients for returning blank questionnaires and 346 patients for not satisfying inclusion criteria (age <18 years: 4 patients; non-migraine sufferers according to ICHD-II criteria: 124 patients; receiving an ongoing preventive treatment for migraine: 218 patients), a total of 5417 patients consulting 1550 physicians (1467 GPs and 83 neurologists) remained for analysis.
Mean age of GPs (49.0 ± 7.4 years), age-class distribution and regional distribution were comparable with national averages, whereas the male-to-female ratio was slightly higher than the national average (82% vs. 76%; p < .01). Mean age of neurologists (47.0 ± 8.3 years), age-class distribution, sex ratio and regional distribution were also generally comparable with national averages. The mean number of migraine patients seen in a month was 15.1 ± 11.8 for GPs and 25.2 ± 22.2 for neurologists (p < .01).
Demographic and migraine characteristics of study patients
SD = standard deviation. *Significantly different from patients consulting a neurologist (p < .01)
Basic data with stratification by GPs versus neurologists
GPs = general practitioners.
Comparative characteristics of migraine patients deemed eligible/ineligible for preventive treatment
SD = standard deviation. aEpisodic/chronic migraine: >15/≥15 headache days/month, respectively. bTriptans, aspirin-metoclopramide, ergotamine derivatives. cParacetamol, non-steroidal anti-inflammatory drugs, opioids, aspirin. *Significantly different from patients found ineligible for preventive treatment (p < .01).
The percentage of probable migraine among those deemed eligible for preventive treatment seems to be very low because for GP's it is difficult to distinguish between tension-type-headache and probable migraine. As when we say “preventive therapy” for GP's it means “severe patients” and GP's deemed eligible only severe patients.
Comparative characteristics of migraine patients deemed eligible/ineligible for preventive treatment, according to physician perception
SD = standard deviation. *Significantly different from patients found not eligible to a preventive treatment (p < .01)
Factors significantly influencing physician opinion about patient need for preventive treatment of migraine
At the end of the consultation, a preventive treatment was offered to 99% of eligible patients: 94% of these accepted it, whereas 5% refused. More patients consulting a neurologist accepted the preventive treatment (97%; p < .05 vs. GPs) and fewer refused it (1%; p < .01 vs. GPs). Finally, a preventive treatment was prescribed (new or repeat prescription) to 89% of eligible patients. Of the patients who were prescribed a preventive treatment, 84% were perceived by the physicians as anxious (a perceived anxiety score of ≥4 on the scale of 0–10) and 70% were perceived as under stress. The mean score of perceived functional impact was 6.2 ± 1.9 on the scale of 0 to 10.
Discussion
The objectives of this study were to analyze the modalities of prescription of migraine preventive treatments in the primary care setting by comparing the characteristics of migraine patients found eligible and those of patients found ineligible for preventive treatment and determining the factors conducive to the prescription of a preventive treatment. Results indicate that patients deemed eligible for preventive treatment were more often women, with a higher frequency of headaches (more patients with chronic migraine), who used more units of nonspecific medications to treat their migraine attacks and experienced poor treatment effectiveness. Furthermore, according to physician perceptions, eligible patients were those with more frequent and severe headaches, who were anxious and/or stressed, who experienced greater functional impact, who were most dissatisfied with their acute treatment and overused acute treatments for attacks. A hierarchical classification of the factors leading physicians to consider a patient eligible to receive preventive treatment indicates that the primary factor was perceived medication overuse, whereas patient satisfaction with the acute treatment of attacks and triptan use were found to negatively influence patient eligibility.
The main differences observed between the groups of migraine patients deemed eligible versus those found ineligible for preventive treatment are in accordance with the guidelines currently in force (8,9,12). These guidelines stipulate that institution of preventive treatment should be guided mainly by the number of monthly migraine attacks, migraine-related functional disability and risk of overuse of acute treatments (12). French guidelines specify that preventive treatment should be instituted when the consumption of acute treatments has reached elevated levels (at least six to eight uses per month over the last three months), in order to prevent the risk of medication overuse (12,13). SMILE study patients found eligible for preventive treatment indeed exhibited increased consumption of acute treatments, especially nonspecific medications, with markedly reduced treatment effectiveness. Ineffectiveness of acute treatments may lead to medication overuse and ultimately to evolution from episodic to daily or near-daily headache (chronic migraine) by a mechanism that remains incompletely understood. The profile of the SMILE patients deemed eligible for preventive treatment of migraine thus appeared to correspond satisfactorily to the aims currently assigned to migraine preventive therapy.
The largest differences in physicians' perceptions between the groups of patients deemed eligible versus ineligible for preventive treatment concerned the frequency of headaches, patient satisfaction with the acute treatment of attacks and medication overuse (although significant between-group differences also existed for the perceived severity of headaches, anxiety, stress and functional impact). Among 17 factors that might have influenced physician opinion about the need to institute preventive treatment, perceived medication overuse ranked first, followed by the perceived number of headache days per month, while perceived severity of headaches and perceived functional impact of migraine appeared to play minor (but still significant) roles. Furthermore, the reasons given by physicians for considering study patients to be eligible for preventive treatment were, in decreasing order of frequency, the number of attacks per month, severity and duration of attacks, overuse of acute treatments and patient preference. It is striking to note that neither perceived patient anxiety and stress nor functional impact of migraine appeared among the factors cited by the physicians. Previous studies in the French general population found that 50.6% of migraine sufferers were anxious and/or depressive (14) and that 18–22% had grade III or IV disability (moderate or severe disability) according to the Migraine Disability Assessment (MIDAS) questionnaire (5,15). Similar results (22% of patients with moderate or severe disability) were obtained in the recent AMPP study (6). It therefore appears that, at least for French physicians, somatic factors predominate over psychological factors (anxiety, stress) and disability in the decision-making process leading to initiation of migraine preventive treatment.
Physicians cited patient preference as the last reason for prescribing a preventive treatment for migraine. This factor, mentioned in the guidelines (8), may deserve more consideration. Including the patient in the decision-making process is likely to increase patient involvement in therapy and therefore compliance with the prescribed treatment (16–18). Patient preference for the mode of treatment administration should be determined, and medications with side effects known to be unacceptable to the patient (such as weight gain for women or sedative effects) should be avoided (13,19).
This study has some limitations. Although it was conducted among a sample of GPs and neurologists recruited at random from a physician database, the physicians who agreed to participate were possibly those most involved in migraine management. Additionally, participating physicians were all aware of the objectives of the study, which may explain the unusually high proportion of consulting patients deemed eligible to receive preventive treatment for migraine (83%). According to data from the AMPP study, less than 40% of migraine sufferers in the general population are concerned about migraine prevention (6). Nevertheless, study results show consistent and significant differences between the patients deemed eligible and those deemed ineligible for preventive treatment, and these differences are in keeping with the current guidelines. One strength of the study is the great number of patients analyzed, which reinforces the validity of the results.
Conclusion
Migraine has long been recognized as associated with a characteristic set of psychiatric disorders, namely anxiety, depression, phobias and panic disorders, caused by the painfulness and unpredictability of attacks, and their uncontrollability if poorly treated. Therefore, physicians are influenced to prescribe a preventive treatment. Some factors like medication overuse, severity and number of attacks and impact of the disease are influential to initiate preventive treatment. On the contrary, the duration of the disease, the sex (male), the use and satisfaction of triptans are against the prescription of these preventive treatments.
Footnotes
Acknowledgements
This work was supported by a grant from Schwarz Pharma France.