Abstract
This study presents the first nation-wide survey of migraine in Austria. A sample of 997 Austrian ≥ 15 years old were interviewed personally (face-to-face) in a random sample in the whole country. Diagnosis of migraine was based on the International Headache Society (IHS) classification. Of the Austrian adult population 10.2% were identified to suffer from IHS migraine, 5.6% from migraine without aura, 2.3% from migraine with aura and 2.3% from borderline migraine. Another 8.5% have possible migraine. Other primary headaches were reported in 30.7%. Sex, age, working status and region were found to be the main demographic influencing factors. Further influences were stress, spinal column problems or weather changes. The most used acute medications were over-the-counter drugs, doctor attendance rate was very low. Working people with migraine dropped out of work 14 days per year, which adds up to 6.8 million working days per year. This remains a substantial economic factor. The findings indicate that migraine sufferers in Austria need to be more informed about their illness and what to do against it, especially encouraging doctor visits.
Introduction
Population-based epidemiological studies about headaches, especially migraine, have been carried out in many countries. Within Europe Austria is one of the ‘white points’ on the landscape with no official data reported. In 1990, a prevalence of 12% was estimated for Austria based on a limited sample from clinicians in particular areas of the country. The purpose of this paper is to report on a community-based study of the 1-year prevalence of migraine in a sample of the Austrian adult population using the operational diagnostic criteria of the International Headache Society (IHS) (1). Of main interest was the 1-year prevalence and frequency of migraine, socio–demographic factors, social impact, doctor attendance, and the treatment behaviour.
Subjects and methods
The study was conducted by personal face-to-face interviews among Austrians inhabitants ≥ 15 years old, representing all of Austria. The sample represents 6.65 million Austrian adults and was drawn at random from the Austrian population in August/September 2001. In this study a standardized questionnaire and a diagnostic algorithm applied by lay interviewers were used as the method of data collection. The algorithm was based on the diagnostic criteria of the IHS. The algorithm classified headache sufferers into four groups: (i) IHS migraine, fulfilling criteria for migraine without and/or with aura; (ii) borderline migraine, where either the duration of attacks was between 2 h and 4 h or there was photophobia or phonophobia (this group is part of the migrainous disorders group, defined by HIS code 1.7); (iii) possible migraine, where at least two IHS criteria were missing; and (iv) non-migraine headache. According to the HIS criteria, groups (i) and (ii) are included in the migraine group (corresponding codes 1.1–1.7 in the HIS classification, except 1.5, which concerns childhood periodic symptoms which are precursors to or associated with migraine, and 1.6, which includes status migrainosus and migrainous infarction). The survey included a structured headache interview with detailed questions about headache history including headache frequency, duration, location, character of pain, intensity and accompanying symptoms, precipitating and provoking factors and headache-related disability. For disability which occurred during paid work or homemaking, subjects were asked ‘How strongly are you affected in routine daily activities due to your most severe type of headache and what are the consequences?’ (‘cancellations of social activities’, ‘reductions in work performance’, ‘discontinuation of vacation’, ‘lost time from work’). Percentage disability was reported according to category, beginning with ‘none/little’ up to ‘mild’, ‘moderate’ and ‘severe’.
Patients were also asked about their doctor attendance, use of prescription and over-the-counter (OTC) medication. The lay interviewers were especially trained by the authors (C.L. and A.B.) until they were familiar with the interview methodology and showed a good knowledge of this particular neurological disease. We explained migraine criteria according to the IHS standards and taught them the background of each question, explaining when and how they should restate insufficiently answered questions. This was done primarily by asking questions in mock interviews. Migraine occurrence was asked about for the last 12 months, which can be taken as representative of the population penetration in general. It is important to point out that subjects with coexisting migraine and another primary headache (e.g. tension-type headache) were included in the diagnostic group of migraine. The validity of all diagnosis of migraine by lay interviewers was computed and compared with that of two authors (C.L. and U.B.). Validity and reliability were given by the proof of representativeness, which showed that the survey sample matched the population structure (for further details see Table 1).
Distribution of sample population by sex, age, socio-economic groups and Austrian regions and prevalence of primary headache (%)
All data were processed using SPSS Data Entry II. Statistical analyses were performed using SPSS Base System for Windows 6.1. χ2 test with a 5% level of significance and 95% confidence limits for prevalence rates were calculated.
Results
Sample
The starting sample was 1200 households, the sample reached was 997 persons interviewed, representing an 83% response rate, which corresponds to international requirements. Demographic characteristics are listed in Table 1. The proportion of females was 53%, slightly higher than for males with 47%; mean age was 43.2 years.
Prevalence of primary headaches and regional distribution
Of the 997 subjects 482 (49.4%) suffered from primary headaches at least some time within a year. Women (54.6%) suffered more than men (43.6%), younger people < 30 years (59.8%) more than older> 50 years (39.5%), and working people (54.1%) more than non-workers (43.1%, who were mainly retired already). Prevalence in the capital Vienna was higher (59.7%) than elsewhere, i.e. in smaller towns or the countryside (for details see Table 1).
Migraine prevalence, demographic distribution, age at onset
After classifying the various headaches, we found a 1-year prevalence of 10.2% of the Austrian adult population suffering from IHS migraine (Table 2). Migraine without aura was found in 5.6%, migraine with aura in 2.3%, and again 2.3% suffered from borderline migraine. Another 8.5% was found to have possible migraine and 30.7% non-migraine headache. In both females and males the prevalence of migraine varied by age (Table 3). The main sufferers among men were the young < 30 years old, among women it was the age group between 30 and 49 years who suffered most.
Prevalence rate of International Headache Society migraine, possible migraine and other primary headache in the Austrian population (n = 997)
Prevalence of migraine without aura, migraine with aura and borderline migraine by age (n = 997)
Looking at the demographic distribution of various migraine types (Table 4), it was found that the main sufferers were females with a male/female ratio of 1:2.5, middle-aged working people (P < 0005) and the population in Vienna and West Austria (P < 0025). Age at onset for both sexes and intensity of migraine are shown in Table 5. For both sexes the preponderance of onset was in the second decade, with a female preponderance in all decades, mean age at onset was 17.8 years (female 18.9%/male 14.6%).
Migraine prevalence, demographic distribution
Age at onset of International Headache Society migraine
Frequency, duration and pain characteristics and intensity of IHS migraine (Table 6)
Frequency, duration and intensity of International Headache Society migraine
The main frequency of occurrence was once per month (44.9%), but an additional 20% suffered as much as once per week. Another 19.7% had their attacks once per quarter and only 13.0% once per year. Altogether this makes 13.4 days out of 3 months on which migraine occurred. The attacks usually lasted 3 h (34.3%) up to 1 day (38.8%). More than a quarter suffered more than 1 day (26.9%). Of patients who described their pain as pulsating (78.1%), in 22.2% pain was associated with tension. Pain intensity was moderate to severe in 86.5% (88.2% of males and 85.7% of females), the intensity did not differ significantly by gender.
Distribution of pain character, location and accompanying symptoms and location (Table 7)
Pain quality and distribution of accompanying symptoms
The main complaints were sensitivity to light and noise (78.8%), pulsating pain (78.1%), pain aggravation by routine physical activity (71.8%), and nausea (53.7%). More than half of the affected people had a unilateral pain (51.6%) and nearly half of the sufferers had to lie down during an attack (47.2%). There were no significant differences found in these complaints between sexes.
Precipitating factors
We found three main triggers for individual migraine attacks (Fig. 1). In 60.4% stress was reported, followed by spinal column problems (51.1%) and a change in weather (42.2%); 31.1% attributed their migraine attacks to bad lifestyle, such as too little sleep; alcohol (29.8%) and smoking (28.1%) were also reported; 31.2% mentioned bright light, e.g. from TV or PC monitors, and 29.9% of the women IHS migraine sufferers, menstruation. Certain foods were reported in 17.7%.
Precipitating factors: International Headache Society migraine (n = 101), male (n = 29), female (n = 71). ▪, Female; hatched, male; □, total.
Migraine-related disability
About half of the migraine sufferers (46.8%) had a reduced ability to function during attacks, with one-fifth (19,8%) reporting severe disability (Fig. 2). The consequences of attack-related disability were seen in cancellations of social activities (46.2%), 24.3% reductions in work performance (24.3% including household), and discontinuation of vacation (7.7%). Moderate to severe migraine attacks also caused lost time from work. People suffered from attacks on 13.4 days in 3 months on average in our population. This adds up to 53.6 days per year. Males suffered on 33.2 days per year, whereas female on> 62.4 days. On average, individuals with migraine experienced 23.2 days of lost work time per year due to reduced productivity from attacks. In individuals with migraine who were in paid employment missed 3.5 days in 3 months on average, i.e. 14 days a year.
Migraine-related affects on routine daily activities (household work and non-working activities).
Consultation rate and medication use
Of the migraine sufferers, 39.6% had never visited a doctor, 30.8% had once visited a doctor but a long time ago, 16.8% visited a doctor irregularly and only 12.8% went to a doctor regularly and got treatment. They mainly went to their general practitioner (23.8%), only seldom to a neurologist (7.9%) and rarely to an internist (3.0%). There were no significant gender differences in consultation frequencies. Over 84% of migraine sufferers took some medication for their headache, regularly (43.9%) or occasionally (40.4%). OTC drugs were the most common medication, with 58.6% regularly using OTC, 2.9% ergots and 0.8% triptans. Only 9.2% of migraine sufferers used a preventive medicine (e.g. β blockers, flunaricine) for at least a period of 3 months, 18% may have tried such a medication for only a few weeks.
Discussion
The first prevalence study using the IHS criteria to diagnose migraine was performed and published in 1991 (2). Since then, estimates of the prevalence of migraine from most European countries have been reported to range from 1% or 2% to about 35% (3). Comparing Europe and North America the 1-year prevalence of migraine is about 10–12% (4). The present study is the first epidemiological survey of migraine in the Austrian general population. The 1-year period prevalence of a primary headache disorder in this sample was found to be 49.4%. The 1-year prevalence of migraine in our study was estimated in total as 10.2%, 13.8% in females and 6.1% in males, which is in the range of comparable previous community-based studies in Europe. For example, there is no significant difference between the prevalence in Austria and France (t = 0.92). Among the sufferers from IHS migraine are 71% females, with a male/female ratio of 1:2.5. Half of them are between 30 and 49 years old (50%), three-quarters are working (72%), one-quarter live in Vienna (26%) and one-quarter in West Austria–Alpine region (27%). In our population migraine starts very early in life, mainly before 19 years, and partly between 20 and 29 years. After 30 years it hardly ever starts. For each one-third of the sufferers their migraine attacks last up to 12 h, 1 day or longer than 1 day, with a peak at 1 day. Almost seven out of 10 migraine sufferers (67.3%) have the attacks at least monthly. The main complaints are pulsating pain, sensitivity to light and noise, a worsening pain when moving, and nausea. The reasons for migraine, as reported by the subjects, are mainly stress, spinal column problems or change in weather. Half of the migraine sufferers are affected very much or rather much in their social activities or in their job, leading to 4 days illness and 8 days reduced performance in their job within 3 months. Besides the personal harm, this entails substantial economic cost.
In comparison with other prevalence data of migraine reported in previous general population surveys in Europe, our 1-year prevalence rate of 10.2% is within the range of other European countries (Table 8). In contrast, the overall prevalence of migraine headache in Germany (8), 32% in women and 22% in men, is in the high range of comparable studies. This high rate is the result of adding migraine IHS code 1.1–1.6 (11.3%) and IHS code 1.7 (16.2%). Furthermore this result can be explained by the direct implementation of the IHS classification and is comparable to similarly high prevalences in Switzerland (16) or Sweden (15). The result of a similarly structured study from France (7), also carried out by a national public opinion poll agency, showed a prevalence in the previous few years of 12.1%, quite similar to the results from the Danish study (1). However, these surveys used different time periods (1 year vs. life-time prevalence), as well as the reports from The Netherlands (11), UK (17) and Spain (14). As was previously pointed out (3), it is important to assess the period of time under consideration when comparing the prevalence and distribution of a disorder with an episodic nature.
Estimates of migraine prevalence in Europe using the International Headache Society diagnostic criteria
In our study group 86.5% recorded their migraine as moderate to severe, 92% had a reduced quality of life according to their headache, about three-quarters of migraine sufferers had a reduced ability to function during attacks. Comparison with studies conducted in other European countries shows similar results (2, 10). Working people report migraine attacks on 9.8 days in 3 months, i.e. 39 days per year. This leads to a drop-out rate at work with males of 12 days and with females of 15.6 days per year. In total, 14 days per year are lost per paid employment due to migraine. If we consider the prevalence to be 12.8% in the working population (including household) and a working force of 3.8 million Austrians, then we have 486 000 working who suffer from migraine. They drop out 14 days per year on average. It can be estimated that in Austria 6.8 million working days are lost annually as a result of migraine. In a Danish population-based study it was found that, considering all subjects with paid work, 5% had been absent from work 1–7 days in the previous years (18). The extent of work absence due to migraine found in our study is lower than reported above.
Furthermore, it was very surprising that more than one-third had never visited a doctor for their migraine attacks and another one-third only visited a doctor long ago. Only 16.8% of the migraine sufferers consulted a doctor regularly, mainly the general practitioner. This is an extremely low rate compared with others (18). This means that 87% were not under doctor control. The reasons are either that the pain did not occur so often (13.4 days in 3 months) and did not last very long (on average 1 day) or that most patients treated their attacks themselves. Similar explanations are given in a German prevalence study (8). Another plausible explanation is that people will not go to a doctor because of headache in the fear of being interpreted as ‘psychic’. This consequently leads to a high rate of self-medication. In our population 58.6% regularly used OTC, most of them a combination of different ingredients, followed by ergots (2.9%) and triptans (0.8%). Acute medications were regularly used by one-third, whereas preventive medicines were used only by a quarter of sufferers; 15.7% did not take medication. Compared with others (10), this is a significantly high proportion of patients.
Limitations of our study could be the fact that the structured interview was not performed by a physician to diagnose the headaches. Our results apply to all forms of migraine. Migraine with aura was recorded in 2.3%, but it is well known that a description of neurological aura symptoms is difficult to assess in an interview not done by a neurologist. Our study supports the view that, due to the IHS criteria of migraine prevalence, rates do not differ greatly within Europe when standard criteria are used. The reported variation could be due to the variation in sample size, difference in study design and the method of data collection (19).
We conclude that the prevalence rate in Austria is comparable to that observed in other European countries, and that most migraine sufferers have never received a medical diagnosis. We also emphasize that further studies are necessary to elucidate the satisfaction with treatment and overall effectiveness of the medical care system in Austria.
Footnotes
Acknowledgements
We would like to thank Birthe Rasmussen for her advice, comments and constructive suggestions in preparing the study and on the manuscript draft. We are particularly indebted to statistician Gabriele Kaplitza from the IMAS Institute Linz for her valuable assistance in collecting data and for fruitful discussions and comments on the manuscript. This study was supported by grants from AstraZeneca, GlaxoSmithKline, MSD and Pfizer.