Sage Journals: Discover world-class research

Abstract

Migraine is a very common neurological disorder worldwide. Its prevalence is lower in developing countries. There were no data concerning the general population in Benin. We aimed to determine the prevalence of migraine in a rural community of Benin. A cross-sectional study was conducted in a rural area of Abomey from February to April 2003 and included 1113 persons selected by a two-stage survey. A case was defined according to International Headache Society criteria (1988). Of the 1113 persons, 37 had migraine. The lifetime prevalence of migraine was 3.3% (male 2.2%, female 4.0%). The peak prevalence was found in persons in the second decade of life. Higher levels of education were associated with migraine. Migraine without aura was the more frequent form (67.5%). The most frequent triggers of migraine were annoyance (73.0%), exposure to the sun (73.0%), heat (65.0%) and anxiety (51.4%). The low prevalence rate of migraine in Benin confirmed the results of the few available African studies. The disease is severe and occurs in a young population and could lead to a high socio-economical burden.

Keywords

Prevalence epidemiology migraine IHS criteria rural community Benin

Introduction

Migraine is one of the most common types of primary headache. It is considered to be a benign disease with a heavy impact on the quality of life and activities of patients.

Its diagnosis is based on history-taking and on the criteria defined by the International Headache Society (IHS) (1). Prevalence rates vary from 3.1 to 26.0% in Europe and the USA (2–7); it is lower in developing countries. The few studies in Africa found the prevalence to range from 3.0 to 7.2% (8–11). The study of Ogunyemi (12) among Nigerian students found a higher prevalence of 16.4%, but the study used the Ad Hoc Committee criteria.

In Benin, only one study has been conducted to estimate migraine prevalence among workers (unskilled workers, 59%) of five companies at Cotonou and found 8.9% prevalence (13). There are no data concerning the general or rural community in Benin.

The aim of this study was to evaluate the prevalence of migraine in a rural community in Benin as well as its clinical and demographic characteristics and triggering factors.

Methodology

Setting

This study was conducted from 20 February to end of April 2003 at Gbêcon-Hounli, one of the subdivisions of Abomey, a historical city in the centre of Benin, which is located some 140 km north of Cotonou, the capital city. The city is famous for its numerous palaces and their paintings relating to the royal lineage. The city has seven subdivisions over a 200-km² area and its population is estimated to 80 000 persons, with Gbêcon-Hounli accounting for 16 590 inhabitants [8904 (53.7%) women] (14).

Type of study

This was a cross-sectional study including all residents > 15 years old.

Sampling

This was a door-to-door, face-to-face study. Two of the four districts of Gbêcon-Hounli were selected by simple random sampling: the districts of Agnangnan and Zassa with, respectively, 1159 and 890 inhabitants (people > 15 years old in each location 703 and 489) were selected (14).

Sample size

The sample size was computed for an expected prevalence of 8.9% (average prevalence of migraine in a previous study among workers in Benin (13)) with a precision of results of 0.018 and α risk of 5%. The minimal number of subjects was 1001.

Diagnostic criteria

Migraine diagnosis was based on the IHS criteria of 1988 (1). The questionnaire has not been previously validated in our population. It listed criteria of migraine with and without aura and was the same as that used by Henry et al. (6), which has established validity (15,16). This questionnaire was translated and back translated by Beninese neurologists.

Data collection

The study was conducted during the season of rural activities. Therefore, there was very little population movement, which increased the reliability of the study. Interviewers went to houses early in the morning and late in the evening after the day's activities. The six interviewers were medical students trained before the study on the questionnaire and diagnostic criteria for migraine, and selected for their ability to talk the local language (fon) in order to ensure good communication with the interviewees. A standardized data collection sheet was used. It included, in addition to the questionnaire based on IHS migraine diagnostic criteria, all other sociodemographic information (age, sex, profession, level of education, matrimonial status) and all the clinical characteristics (intensity, frequency, duration and type of migraine) as well as triggering factors.

Subjects were asked to mention spontaneously triggers for their headaches or indicate them among those listed (exposure to sun, stress, eating habits, annoyance, sleep disturbance, physical activity, heat, changes in weather conditions and temperature, alcohol, menstruation, anxiety). Data were collected by face-to-face interview. Level of education was defined by illiterate (never went to school and could not read or write), elementary, or college and upwards. We asked people for the frequency of recurrent headaches in the last year.

Subjects considered to have migraine by screening questionnaire secondarily underwent a full neurological examination conducted by a neurologist at the local dispensary to confirm the diagnosis of migraine.

Ethical considerations

The study was authorized by the state health authorities. Subjects were informed before being included in the study and oral consent was obtained.

Data analysis

Statistical analysis was performed by using Epi-Info 6.04C (Ecole Nationale de Santé Publique, Paris, France, and CDC, Atlanta, GA, USA).

Usual descriptive analysis was used: means and standard deviations, proportions, and confidence intervals (CI). χ² test or Fisher's exact test was used to compare the different proportions. P < 0.05 was considered to be statistically significant.

Results

Of 1192 eligible persons, 58 refused to participate, 21 were not seen, 1113 responded to interviewers. The participation rate was 93.4%. The mean age was 36.8 years (± 17.8) and ranged from 15 to 80 years. Four hundred and seven were male (36.6%) and 706 female (63.4%).

The first column of Table 1 summarizes the sociodemographic characteristics of the sample.

Table 1.

Lifetime prevalence of migraine in a rural community in Benin, 2003

Characteristics of sample	Total n (%)	Migraine case n (%) [95% CI]	P
Sex
Male	407 (36.6)	9 (2.2) [1.1, 4.2]	0.110
Female	706 (63.4)	28 (4.0) [2.7, 5.7]
Age (years)
< 20	171 (15.4)	11 (6.4) [3.3, 11.2]	0.003
20–29	323 (29.0)	15 (4.6) [2.6, 7.5]	0.003
≥ 30	619 (55.6)	11 (1.8) [0.9, 3.2]
Level of schooling
Illiterate	613 (55.1)	17 (2.8) [1.6, 4.4]	0.010
Elementary	350 (31.4)	9 (2.6) [1.2, 4.8]	0.010
College and upwards	150 (13.5)	11 (7.3) [3.7, 12.7]
Matrimonial status
Single	277 (24.9)	13 (4.7) [2.5, 7.9]	0.300
Married	757 (68.0)	22 (2.9) [1.8, 4.4]	0.300
Widowed or divorced	79 (7.1)	2 (2.5) [0.3, 8.8]

Among the 1113 interviewees, 49 were screened as migraine suffererers and 37 cases of migraine were confirmed. The lifetime prevalence of migraine was 3.3% (95% CI 2.4, 4.6).

Table 1 shows the prevalence of migraine according to sex, age, matrimonial status and level of schooling.

Table 2 summarizes the clinical characteristics of migraine in our patients.

Table 2.

Clinical characteristics of migraine, Benin 2003

Characteristics	Number	Percentage
Pain level
Mild	9	24.3
Strong	22	59.5
Extreme	6	12.6
Monthly frequency of occurrences
0	11	29.7
1–4	23	62.2
5–9	1	2.7
> 10	2	5.4
Mean daily amount of sleep
< 5 h	3	8.0
≥ 5 h	34	92.0
Types of migraine
Without aura	25	67.6
With aura	12	32.4
Types of aura*
Visual aura	7	50.0
Sensory aura	4	28.6
Aphasic	2	14.3
Motor	1	7.1

The types of aura are not mutually exclusive: one migraineur had visual aura and aphasic, another had sensory and visual aura.

For 16 patients (43.2%), the age at onset was 15 years.

Of the 37 patients, 15 thought that their headaches resulted from one or several factors. Six (16.2%) often recognized those factors, whereas eight (21.6%) recognized them sometimes.

Table 3 shows the most frequent triggering factors.

Table 3.

Main triggering factors of migraine in a rural community, Benin 2003

Triggering factors	Frequency (%)
Annoyance	73.0
Exposure to sun	73.0
Physical activity	65.0
Heat	65.0
Anxiety	51.4
Sleep disturbance	51.4
Menstruation (in women)	14.3
Eating habits	2.7

Discussion

Our study is the first of its kind conducted among the general population in Benin and one of the few in West Africa using migraine criteria set by the IHS. Identification of migraine sufferers was done using a questionnaire based on the IHS criteria. We used a questionnaire already used in other epidemiological surveys and with a sensitivity > 80% (15,16). The questionnaire had not been validated in Benin before the study, but had been widely used in various settings. The participation rate (93.4%) and random selection were excellent. The training of medical students was carefully done and door-to-door interviews are considered an excellent research tool in epidemiological studies. In our sample the proportion of women was higher than that in other studies (8), but was close to that of women in the rural population in our country. It can be explained by the rural exodus (men gravitate to urban areas), and the season (more men went about their activities early). The lifetime prevalence of migraine in our study was 3.3%, very close to the 3.0% from a study of Tekle-Haimanot in an Ethiopian community (9). It was slightly lower than the 5% of Dent's study within a Tanzanian community (8). These two studies, like ours, were held in rural communities and used IHS criteria. However, these two studies asked about migraine in the last year, whereas our study looked over the entire life of the interviewees. A study in Canada by O'Brien et al. (17) found some discrepancy between data collected concerning 1 year and lifetime prevalence (14.6% vs. 17.1%). All other studies on this topic in Africa and other areas have used different criteria, leading to very diverse results (Table 4). However, in Nigeria, Ogunyemi (12) found a high prevalence (16.7%) among students. The prevalence of migraine seems higher in developed than in developing countries. A recent literature review (18) including the main general population studies found that the worldwide estimated prevalence was 11% (1–25%), and was 5% in Africa (3–7%), 9% in Asia (1–22%), 15% in Europe (10–25%), 13% in North America (9–16%) and 9% in Central and Latin America (5–14%). Another study by Stewart et al. (5) in a heterogeneous American sample confirmed those tendencies. Migraine was found to be more prevalent among Caucasian women (20.4%) than in Black women (16.2%), and Americans of Asian origin showed even a lower prevalence (9.2%). The authors ascribed the variations to environmental factors (mostly food intake) or genetics. Stewart et al. (19), using a meta-analysis of 23 studies, observed that in a linear regression model, 70.6% of the variation could be explained by sex and age. According to Dent et al. (8), the age pyramid combined with cultural factors can explain the lower prevalence in Africa. The authors claimed that the socioeconomical weight of infectious diseases makes it culturally unacceptable to mention headaches.

Table 4.

Some migraine prevalence studies among populations

Year	Authors (references)	Country	Migraine definition	Timeline	Population	Age of interviewees, years	Prevalence		Overall prevalence, %
M	F
1982	Osuntokun et al. (10)	Nigeria	AHC	–	Rural community	All	4.6	8.8	6.7
1983	Levy (23)	Zimbabwe	AHC	1 year	Rural community	> 5	3.5	8.9	4.2
1984	Ogunyemi (18)	Nigeria	AHC	Whole life	Students	16–42	15.6	19.8	16.7
1988	Longe and Osuntokun (24)	Nigeria	–	–	Rural community	All			6.3
1992	Osuntokun et al. (11)	Nigeria	AHC	–	Urban community	All	5.0	5.6	5.3
1993	Romdhane Attia et al. (25)	Tunisia	–	–	–	0–100	2.5	4.5	3.4
1995	Matuja et al. (26)	Tanzania	AHC	2 years	Urban community	20–40	–	7.2
1995	Tekle-Haimanot et al. (9)	Ethiopia	IHS	1 year	Rural community	> 20	1.7	4.2	3.0
2004	Dent et al. (8)	Tanzania	IHS	1 year	Rural community	> 11	2.5	7.0	5.0
2003	Our Study	Benin	IHS	Lifetime	Rural community	> 15	2.2	4.0	3.3
1997	Sakai and Igarashi (27)	Japan	IHS	1 year	General population	≥ 15	3.5	13.0	8.4
1986	Cheng et al. (28)	China	–	Lifetime	Community	–	–	0.7
1988	Zhao et al. (29)	China	AHC	Lifetime	Community	–	2.4	11.3	6.9
2003	Lampl et al. (30)	Austria	IHS	1 year	National	> 15	6.1	13.8	10.2
2005	Celik et al. (31)	Turkey	IHS	Lifetime	Rural community	> 15	9.3	29.3	19.9
1992	Henry et al. (6)	France	HIS +probable	Lifetime	National	5–65	6.1	17.6	12.1
2005	Lanteri-Minet et al. (32)	France	HIS +probable	Lifetime	National	≥ 18	6.3	15.7	11.2
2000	Bank and Marton (33)	Hungary	IHS	1 year	General population	15–80	2.7	6.9	9.6
2000	Roncolato et al. (34)	Italy	IHS +probable	–	General population	–	5.0	15.8	11.6
1994	O'Brien et al. (17)	Canada	IHS	Lifetime	General population	> 18	7.8	24.9	17.1
1996	Stewart et al. (5)	USA	IHS	1 year	General population	18–65	8.9	19.0	14.0
2005	Morillo et al. (35)	Brazil	IHS	1 year	Urban community	≥ 15	7.8	17.4	12.6

AHC, Ad Hoc Committee Criteria; IHS, International Headache Society.

Outside of cultural and racial variations, those differences could be explained by (20):

Sampling methods vary from study to study

Data collection through phone survey and self questionnaire is less reliable

The diagnostic criteria for migraine are not the same in all studies

Greater tolerance to pain in rural communities and in African culture than in Western countries

Diseases are perceived as of magico-religious significance

Low medical accessibility

Headaches are not considered as a major health problem compared with infectious diseases and poverty

The triggering factors were not the same in Africa, where exposure to sun and heat is often reported

Possible genetic factors (5).

The prevalence of migraine among the women of Gbêcon-Houégbo was higher than among men, in accordance with other literature data (6,21), but the difference was negligible. This could be explained by the low number of migraine patients in our study. Other reasons could also explain it:

- a high proportion of women in our sample (different from other studies on migraine)

- tolerance to pain is probably higher in women than in men

- the education level is lower in women than in men

- most rural female migraine sufferers are from a low socioeconomic background and are less educated.

Subjects < 20 years old had a higher prevalence of 6.4% and then as age increased the prevalence of migraine surprisingly decreased. Henry et al. (6) found a peak in prevalence in subjects aged 30–40 years. Similar results were obtained by Bigal et al. (21). His study of > 145 335 people living in three large American cities found a high prevalence in the age range 30–39 years regardless of sex or type of migraine (IHS or probable). In a rural community in Ethiopia (9) the authors find a high prevalence in the fourth decade, and in Tanzania Dent et al. (8) find a high prevalence in the fourth decade in women and in the third decade in men. The difference from our study can be explained by the lifetime difference. In our study people were questioned regarding lifetime headache. Elderly migraineurs who had had mild or moderate migraine in their youth and rare occurrences could have forgotten their headaches. Abu-Arafeh and Russel (22) noted a prevalence of 10.6% among school children aged 9–15 years. The same study also observed that migraine occurrence was higher among boys < 12 years and higher among girls > 12 years old.

The prevalence of migraine differed also by level of schooling in our study. Migraine prevalence was higher among people with secondary level schooling. Level of schooling can be a possible confounding factor. Among workers in Cotonou in Benin, Adoukonou (13) find a high prevalence in senior executives. Henry et al. (6) observed a high prevalence among teachers and nurses, whereas low-level workers had a lower rate of prevalence. The majority of individuals had at least one attack per month. This is quite high, and it is possible that the study was not able to capture the milder form of migraine, which could have an impact on the overall prevalence.

Of the migraine patients of Gbêcon-Houégbo, 75.7% had more often severe to incapacitating occurrences of migraine. That number is comparable to the 76.6% of Dent et al.'s study (8) in a rural community in Tanzania. The prevalence of aura was also comparable to other studies in rural communities in Africa. In summary, even if the prevalence of migraine is low in rural Benin, it affects young people and is severe with frequent and incapaciting attacks. This leads probably to a high burden on society.

References

Headache Classification Committee of the International Headache Society Classification and diagnostic criteria for headache disorder, cranial neuralgias and facial pain. Cephalalgia 1988; (Suppl. 7): 1–96.

Stewart

Shechter

Rasmussen

. Migraine prevalence: a review of population-based studies. Neurology 1994; 44(Suppl. 4): S17–23.

Rasmussen

. Epidemiology of headache. Cephalalgia 2001; 21: 774–7.

Silberstein

Lipton

. Epidemiology of migraine. Neuroepidemiology 1993; 12: 179–94.

Stewart

Lipton

Liberman

. Variation in migraine prevalence by race. Neurology 1996; 47: 52–9.

Henry

Michel

Brochet

Dartiques

Tison

Salamon

. A national survey of migraine in France: prevalence and clinical features in adults. Cephalalgia 1992; 12: 229–37.

Göbel

Peterson-Braun

Soyka

. The epidemiology of headache in Germany: a nationwide survey of a representative sample on the basis of the headache classification of the International Headache Society. Cephalalgia 1994; 14: 97–106.

Dent

Spiss

Helbok

Matuja

WBP

Scheunemann

Schmutzhard

. Prevalence of migraine in a rural area in South Tanzania: a door-to-door survey. Cephalalgia 2004; 24: 960–6.

Tekle-Haimanot

Seraw

Forsgren

Ekbom

Ekstedt

. Migraine, chronic tension-type headache, and cluster headache in a Ethiopian rural community. Cephalalgia 1995; 15: 482–8.

10.

Osuntokun

Schoenberg

Nottidge

Adeuja

Kale

Adeyefa

. Research protocol for measuring the prevalence of neurological disorders in developing countries. Results of a pilot study in Nigeria. Neuroepidemiology 1982; 1: 143–53.

11.

Osuntokun

Adeuja

Nottidge

Bademosi

Olumide

Ige

. Prevalence of headache and migrainous headache in Nigerian Africans: a community-based study. East Afr Med J 1992; 69: 196–9.

12.

Ogunyemi

. Prevalence of headache among Nigerian university students. Headache 1984; 24: 127–30.

13.

Adoukonou

TTA

. Etude épidémiologique de la migraine en milieu professionnel au Bénin: à propos d'une enquête en milieu de travail à Cotonou. 2001, Medical Thesis, School of Medicine. Cotonou, p. 122.

14.

Institut National de la Statistique et de l'Analyse Economique. Recensement général de la population et de l'habitation, Bénin, Février 2002. Synthèse des résultats. Direction des Etudes Démographiques. Cotonou, Bénin, Octobre 2003; 1–36.

15.

Dartigues

Henry

Michel

Lindoulsi

. Epidemiology of headache by a validated computerized questionnaire. In: Olesen

(ed.) Headache classification and epidemiology. New York: Raven Press, 1994, p.247–57.

16.

Rasmussen

Jensen

Glesen

. Questionnaire versus clinical interview in the diagnosis of headache. Headache 1991; 33: 403–12.

17.

O'Brien

Goeree

Streiner

. Prevalence of migraine headache in Canada: a population-based survey. Int J Epidemiol 1994; 23: 1020–6.

18.

Stovner

Hagen

Jensen

Katsarava

Lipton

Scher

. The global burden of headache: a documentation of prevalence and disability worldwide. Cephalalgia 2007; 27: 193–210.

19.

Stewart

Simon

Shechter

Lipton

. Population variation in migraine prevalence: a meta-analysis. J Clin Epidemiol 1995; 48: 269–80.

20.

Tekle-Haimanot

. Burden of headache in Africa. J Headache Pain 2003; 4: S47–54.

21.

Bigal

Liberman

Lipton

. Age-dependent prevalence and clinical features of migraine. Neurology 2006; 67: 246–51.

22.

Abu-Arafeh

Russel

. Prevalence of headache and migraine in school children. BMJ 1994; 309: 765–9.

23.

Levy

. An epidemiological study of headache in an urban population in Zimbabwe. Headache 1983; 23: 2–9.

24.

Longe

Osuntokun

. Prevalence of neurological disorders in Udo, a rural community in Southern Nigeria. Trop Geogr Med 1988; 65: 36–40.

25.

Attia Romdhane

Ben Hamida

Mrabet

Larnaout

Samoud

Ben Hamda

. Prevalence study of neurologic disorders in Keliba (Tunisia). Neuroepidemiology 1993; 12: 285–99.

26.

Matuja

WBP

Meteza

IBH

Rwiza

. Headache in a non-clinical population in Dar es Salaam, Tanzania. A community-based study. Headache 1995; 35: 273–6.

27.

Sakai

Igarashi

. Prevalence of migraine in Japan: a nation wide survey. Cephalalgia 1997; 17: 15–22.

28.

Cheng

Ziegler

Dai

Chandra

Schoenberg

. A prevalence survey of incapaciting headache in the People's Republic of China. Neurology 1986; 36: 831–4.

29.

Zhao

Tsay

Cheng

Wong

Yao

. Epidemiology of migraine: a survey in 21 provinces of the people's Republic of China, 1985. Headache 1988; 28: 558–65.

30.

Lampl

Buzath

Baumhackl

Kingler

. One-year prevalence of migraine in Austria: a nationwide survey. Cephalalgia 2003; 4: 280–6.

31.

Celik

Ekuklu

Tokuç

Utku

. Migraine prevalence and some related factors in Turkey. Headache 2005; 45: 32–6.

32.

Lanteri-Minet

Valade

Géraud

Chautard

Lucas

. Migraine and probable-migraine—results of FRAMIG 3, a French survey carried out according to the IHS classification. Cephalalgia 2005; 25: 1146–58.

33.

Bank

Marton

. Hungarian migraine epidemiology. Headache 2000; 40: 164–9.

34.

Roncolato

Fabbri

Recctria

Cavazzuti

Visona

Brignoli

Medea

. An epidemiological study to assess migraine prevalence in a sample of Italian population presenting to their GPs. Eur Neurol 2000; 43: 102–6.

35.

Morillo

Alarcon

Aranaga

Aulet

Chapman

Conterno

. Prevalence of migraine in Latin America. Headache 2005; 45: 103–4.

Prevalence of Migraine in a Rural Community in South Benin

Abstract

Keywords

Introduction

Methodology

Setting

Type of study

Sampling

Sample size

Diagnostic criteria

Data collection

Ethical considerations

Data analysis

Results

Discussion

References