Our nationwide population-based study estimated the 1-year prevalence of definite migraine as 16.4%, probable migraine as 12.4% and of pure TTH as 5.1%, probable TTH as 9.5% with ICHD-II criteria, constituting a total of 43.4% of the general population suffering from these two primary headache types. We had planned to reach 6,000 representative households and in the end, a total of 5,323 households were examined for headache. This excellent response rate of 89% probably reflects the conductance of the study directly by physicians face-to-face rather than sending a questionnaire. The prevalence of migraine was highest among 35–40-year-old women while there were no big differences in age groups among men and in TTH overall, as shown in Fig. .
The striking well-known female preponderance in patients with migraine which is also evident in our study is more consistent across studies than the overall prevalence figures of migraine [1
]. All of the studies reveal that migraine is [6
] two or three times more common in females than in males. Interestingly, the rates of the present study using ICHD-II criteria for migraine in adults aged 18–65 years (16.4%) as well as for migraine aura (21.5% in migraineurs) are identical with the previous largest Turkish nation-wide headache prevalence study with the participation of 2,007 households aged between 15 and 55 years [18
] with the ICHD-I criteria. Some studies indicate that the prevalence of headache and especially of migraine has been increasing during the last decades in Europe [6
]. Although our study showed no significant change in the migraine prevalence compared to the national study of 10 years ago from the present study, the male to female ratio was 1:3 in the present one while it was 1:2 in the previous one. Our study with more than the double sample size in comparison to the former one probably reflects the real gender difference. Although it is well-established that headache suffering, including migraine, was highly prevalent especially in younger women overall in the world, the differences of headache features between men and women were thoroughly investigated only in a few studies [24
]. Our study showed that women had a significantly longer attack duration, more nausea and more allodynia in comparison to men among other differences as seen in Table . A population-based study in the UK reported the mean headache duration of 28.4 h in men versus 36.7 h in women along with non-significant changes of attack frequency and pain intensity, similar to our results [5
]. Several hypotheses have been proposed to explain these differences, including fluctuations in sex hormones and receptor binding, genetic factors, differences in exposure to environmental stressors, as well as differences in response to stress and pain perception [24
On the contrary of the small changes in the migraine prevalence around the world, the prevalence of TTH is a matter of debate and has varied widely among studies. TTH is known as the most prevalent type of headache across all age groups worldwide [11
]. Nineteen studies have reported the TTH prevalence in Europe and the prevalence of current TTH among 66,000 adults was reported as 62.6%, and chronic TTH (i.e. on 15 days per month) occurred in 3.3%. Much lower figures (current TTH 15.9%, chronic TTH 0.9%) were found in the nine studies among almost 25,000 children and the youth showing the possible increase with age [6
]. The largest American study with telephone surveys reported a TTH prevalence of 38.3% [11
] and higher figures and lifetime prevalence around 80% were reported in Denmark [26
]. In our study, TTH prevalence is much lower than most of the other studies, even after the inclusion of cases with probable TTH, interestingly. Rare episodic form is the most frequent form of TTH and followed by frequent episodic form and lastly chronic TTH is the most infrequent form in both definite and probable TTH categories, in our study.
The wide variations in the estimated prevalence of TTH can result from the methodology, case definitions, sampling procedures, possible influence of the physician/investigators and the inclusion or exclusion of cases of infrequent episodic TTH and overlap with probable migraine. We applied the ICHD-II 2004 criteria very strictly, without allowing any influence of the physician. It is also highly likely that some unknown genetic factors besides variables such as environmental risk factors or culturally determined differences in symptom reporting may further explain this discrepancy. It is important to note that in this study, TTH was a diagnosis of exclusion and it was only diagnosed in headache sufferers if definite or probable migraine were not diagnosed according to ICHD-II criteria. Hence, this could be one of the important reasons that the TTH rate in our study is not as high as the previous study in our country [18
The difference of results between these two Turkish headache epidemiological studies can also be evaluated, considering the continuum hypothesis as a basis. The two ends of headache spectrum are TTH and migraine, both might evolve into other during time or from one attack to another. Mixed headache, so called TTH and migraine in the same individual, is accepted as the occurrence of spectrum of headache in the same individual [27
]. Both adolescent and adult studies have shown that headache might evolve into both ends of spectrum [28
]. Thus, the low prevalence of TTH might be the evidence of evolving of TTH into probable migraine/migraine by some external or internal modifiers such as socio-economic difficulties or hormonal changes.
Another alternative conceptual approach, the “severity model” of headache, considers a continuum of headache ranging from mild to severe forms with specific headache subtypes distinguished by level of severity rather than unique constellations of symptoms [30
Stovner and Colette [6
] compared the results from the studies using different methods of data collection and reported that only for migraine and headache in general could meaningful comparisons be made; in relation to TTH, there were too few studies available. Most questionnaire studies use somewhat modified criteria, whereas studies based on personal interviews seem to give somewhat higher prevalence than those using questionnaires. The ways the ICHD criteria are applied and the diagnoses included are also of great importance. The problem of multiple headache types occurring in the same patient may represent problems in headache epidemiologic studies. One diagnostic dilemma is the overlap between TTH and probable migraine. It is well-known in clinical practice that many patients have comorbid TTH and migraine, or in other words many migraineurs may experience headaches very similar to TTH in some of their attacks. Thus, the trend and thoughts of the physician could affect the diagnosis. Being aware of this, our study was based on the strict computerized application of ICHD-II criteria aiming to exclude the subjectivity of the conducting physicians. Furthermore, some individuals suffer from infrequent, not disturbing headaches and could not remember the exact profile. It is also known that subjects’ headache symptoms might change during a given period or they might even forget that they had experienced headache [29
]. All these factors pose difficulties in diagnosing headache in the population based epidemiological studies. This is particularly true for the probable headache diagnoses. Using ICHD-II criteria strictly, we showed that pure TTH is indeed rare. In ICHD-II, fulfilment of the diagnostic criteria for main groups of migraine and TTH or any of their subtypes, always trumps fulfilment of criteria for the probable diagnostic categories [19
Although many studies investigated the prevalence of migraine and TTH in Western Europe and North America, there are only a few studies carried out in Eastern Europe. In the Republic of Georgia, an eastern neighbour of our country, one-year prevalence was estimated to be 6.5% for migraine, 9.2% for probable migraine (all migraine 15.6%), 10.0% for TTH, 27.3% for probable TTH (all TTH 37.3%) in a community-based door-to-door survey, conducted by four medical residents [31
]. So they found a lower rate for migraine but a higher rate for TTH in comparison to our results. Another study from Croatia located also in the eastern bank of Europe reported a crude and lower prevalence of TTH as 21.2% [32
]. It is interesting to note that both of these studies also showed relatively low prevalence rates of TTH, like in our study. Whether these regional differences are real or mainly a result of differences in the methodology and conduction of the studies is uncertain.
The prevalence of chronic daily headache (≥15 headache days per month) was 3.3% in our nationwide study, similar to many studies worldwide [12
]. Interestingly, an unusually high prevalence of chronic headache with a rate of 7.6% was reported from Georgia associated with a low socioeconomic status [31
], showing variability of headache disorders, even in neighbours. Another population-based study from Far East of chronic daily headache in 3,377 participants reported a prevalence of 3.2% being higher in women (4.3%) than men (1.9%) similar to our results [35
]. A 2.1% prevalence rate for MOH in our total study population seems to be some higher than reported rates before [31
], however, recent studies reported higher rates of MOH in general population as in our study [34
]. A reason for high rate of MOH in our study population might be related with low rate of prophylactic medication use which is 4.9% among migraineurs.
Although migraine is a remarkably common cause of temporary disability worldwide, many migraine sufferers have never consulted a physician. While 47.0% of migraineurs had physician consult for their headache in 1998 [18
] in Turkey, this ratio has raised to 70.6% in 10 years. Though consultation rates have increased remarkably the underlying epidemiology of migraine remains stable over a decade in our country. Thus, our data support that there is no evidence of increasing prevalence of migraine with increased awareness. On the other hand, only one-third of the TTH patients had ever consulted a physician in 2008. Mostly consulted physicians were neurologists as seen in Fig. . Primary care physicians, who are supposed to be the first to consult for headache, were far less than neurologists in our country, reflecting the choice of the patients. A study from United States reported that 66.1% of migraineurs (68.1% in females and 57.3% in males) had ever consulted a physician [42
]. While in this American study 61% of migraineurs who never consulted reported severe headache, in our study 40% of migraineurs who never consulted had usually severe headache. Of migraineurs who never consulted, 47.6% had 4 or more attack frequency per month, 14.1% had more than 1.5 days average attack duration and 21.5% had more than 6 headache days per month whereas of those who ever consulted, 57.4% had 4 or more attack frequency per month, 28.4% had more than 1.5 days average attack duration and 36.4% had more than 6 headache days per month. These facts reflected that there were still some patients with significant impact of migraine who did not consult for their headaches.
Prevalence studies exploring the relation between socio-economic status (SES) and headache have shown some conflicting results. The present study revealed a negative correlation of migraine prevalence with educational status unrelated to gender and with socioeconomic status only in women. Higher prevalence with lower educational status/lower income was reported in some other studies [1
]. This contradicts the usual clinical perception that migraine is a disease of rich people. In previous studies done in Turkey, there was a positive correlation showing higher migraine prevalence with higher educational status [13
]. These studies are possibly reflecting that people with higher income/education are far more likely to consult a physician or volunteer to participate in a study. In three very large population based studies in United States, the decline of migraine prevalence with increased income or education has been explained by “social causation hypothesis” such as “factors with low socioeconomic status increase migraine prevalence” and “social selection hypothesis” such as “migraine-related dysfunction interferes with educational and occupational functioning leading to low income and low education” [42
]. A prospective study analysing the relation between SES and risk of headache in Norway showed that low SES was associated with increased risk of frequent and chronic headache at follow-up [47
]. Interestingly, the risk of frequent and chronic headache decreased with increasing individual income, but only among men [47
], showing again a gender difference of SES with migraine.
Every type of misdiagnosis is still very common both for migraine and TTH in our country as shown in Table . Furthermore, prophylactic medication usage was unexpectedly low (4.9%), even though neurologists were in charge for headache care for most of the patients. These points draw attention to the need of continuing education for headache management for physicians and for public to lift the real burden. The headache lectures and courses addressed mostly secondary headaches in the medical curriculum and seemed not be sufficient for appropriate management of primary headaches, taking the overall burden in daily life into account. Moreover, the optimal visit duration of headache patients should not be short. This is one of most limiting problems of Turkish neurologists who should examine huge numbers of patients every day.
There are some strong points of our study including face-to-face assessment of headaches by a specifically trained physician group with electronic database system, a large nation-wide sample size and a random population, strict application of the ICHD-II diagnostic criteria of the IHS excluding the subjectivity of the physician’s diagnosis. However, there is an unavoidable risk of the effect of the question style even with the same questions and with an electronic recording system. Due to the higher impact of migraine in clinical practice in our country [45
] it is possible that the physicians are more prone to handling the migraine patients than the TTH sufferers.
In conclusion, our study showed a 16.4% prevalence rate of migraine in Turkey, and it is similar or even higher than the well-established prevalence figures of migraine worldwide. Although there are still misdiagnoses, the rate of physician consults for migraine has remarkably increased to 70.6%, whereas the rate of migraineurs on prophylactic treatment is still lower than expected. Finally, the prevalence of TTH with strict application of the 2004 ICHD-II diagnostic criteria is very low in our study (5.1% for definite TTH and 9.5% for probable TTH), a finding which could reflect some unknown genetic, cultural, environmental factors or methodological differences in the study designs.