Migraine is the leading cause of recurrent headaches in children and adolescents [1] [2] . According to the literature, 9.1% of children and adolescents suffered from migraine. It can have significant impacts on the quality of life and education of the child or adolescent [3] - [5] . In Madagascar, the annual prevalence of migraine in the general population was estimated at 19% [6] . In our practice in Madagascar, data on migraine in adolescents are scarce. Knowledge of these data would improve the care of adolescents suffering from this pathology in our context. Our objective was to determine the annual prevalence and characteristics of migraine sufferers in adolescents and Young adults in our population.

Ethical standards applicable to this type of study were upheld, with informed consent obtained from both parents and participating students.

Study design: A cross-sectional survey was carried out in a high school “Notre Dame high school” in the city of Mahajanga. Registered in this high school 416 students with 151 students in the second class, 115 students in the first class, and 150 students in the final class. The city of Mahajanga is a city located 560 km northwest of the capital. It has an area of approximately 57 km² and is an urban, coastal district. With approximately 258,068 inhabitants in 2020.

Recruitment: The survey was carried out in January 2021. The interviewers were registrar were supervised by a neurologist during the survey. Was chosen as interviewer for all the registrars of the neurology department of the Mahajanga university hospital.

Participants: We surveyed without matching high school students at the Notre dame school. The number of study subjects was calculated by simple random sampling with an expected prevalence of 10%, a confidence limit of 5%, and a calculation factor of 3 giving 415 subjects necessary (this approach allowed us to standardize participant recruitment across classes of varying sizes, thereby increasing the study sample and ensuring a more representative population) for a confidence interval of 95%. This expected prevalence of 10% was chosen using as a reference the overall prevalence of migraine in children and adolescents, which is estimated at 9.1%. The choice of survey subject was made exhaustively by all the students were present during the study period. We considered as the case all subjects had at least one episode of headache during the past year. The diagnosis of the headaches was made by a neurologist by reporting the characteristics of the headaches collected on the criteria of the International Classification of Headache Disease version 3. Cases of probable migraine were considered migraine. The severity of headache attacks was assessed by a visual analogue scale included in the questionnaire. Attack scores of 0 to 3 were considered with mild attack, 4 to 6 moderate, 7 to 8 severe and 9 to 10 very severe. The need for preventing treatment was defined by an attack frequency of more than 3 attacks per month.

Data collection and analysis: Used for the survey structured questionnaires that were self-completed under supervision by pupils within the student’s classroom. The questionnaire details the headache characteristics with a multiple-choice question. A question on a significant event of the year 2019 in Madagascar was added to evaluate the recall. A pre-test of the questionnaire was carried out on 30 students beforehand to assess their understanding of the questionnaire with a completion rate of 100%. Each interviewer was trained to give the same explanation for each paragraph of the questionnaire. Crude and specific annual prevalence of headache and migraine was analyzed.

At the end of the study, 415 subjects were surveyed. Women represented 55.71% of our study population. The average age of our population was 16.91 with an extreme of 14 to 24 years. Subjects under 18 years old represented 69.88% of survey subjects and those aged 18 and over 30.12% ( Table 1 ). The second, first, and final classes represented respectively 36.4%, 27.9%, and 35.9% of the subjects of the survey. The response rate to the survey was 100%. The annual prevalence of headache was 89.4% (95% CI, 86.2 - 92.1). The crude prevalence of migraine was 19.3% (95% CI, 15.6 - 23.5). Among females, the prevalence was 24.24% (95% CI, 19.1 - 30.0), and among males, 13.04% (95% CI, 8.7 - 18.6). In subjects under 18 years old, the prevalence of migraine was found at 18.96%, and in subjects aged 18 and over at 20%. Migraine without aura represented 57.50% of migraine cases. Ten percent of migraine sufferers had more than 3 migraine attacks per month. Migraine attack was severe in 52.5% and very severe seizures in 27.50% ( Table 2 ). The triggering factors found were lack of sleep in 76.3%, stress in 50%, excess sleep in 21.3%, strong emotions in 21.3%, and unspecified factors in 18.8%. The most relieve factor were sleep at 70%, rest at 46.25%, and relaxation at 41.3%.

The annual prevalence of migraine was found among high school students higher than our estimate. At the end of the study, 89.40% of the students had headaches during the past year. According to the literature, students would have more headache attacks, and this is because of the presence of several factors favoring headaches such as stress, skipping meals, physical exertion, fatigue, noise, sunlight, and menstruation [7] [8] .

The other studies carried out among students of the same age group also showed the same findings. The study carried out among adolescents in a school in Porto Alegre in Brazil by LM Barea et al. reported an annual prevalence of headaches among adolescents was 83% [9] . For Antoniuk et al. the annual prevalence of headaches in a school was 93.5% [10] . In schools in Karachi in a study by Sarwer Jameel Siddiqui et al., the annual prevalence of headaches among schoolchildren was 85.5% [11] . We found a high annual prevalence of migraine among students, at 19.30%. In the literature, the prevalence of migraine in children and adolescents would be 5% to 10%, but this prevalence varies according to the methodology and the age of the subjects included [2] [12] . In schools, a higher prevalence of migraine was found in children and adolescents ranging from 5.2% to 21.7% [4] . The same result has been reported in African and occidental studies [13] - [16] . This prevalence varied according to age and gender. In our population, women were the most affected with an annual prevalence of 24.4% and 13% for the male gender. It increased with age with a prevalence of 19% for those under 18% and 20% for those aged 18 and over. These results were comparable to those described in the literature with a greater prevalence of migraine in young subjects and females [2] [4] [17] [18] .

In our study, severe migraine attacks were found in the majority of cases with infrequent attacks. Ten percent of subjects could indicate preventative treatment [19] .

Strenthg and limits: These results allow us to have an overview of the prevalence of migraine in adolescents in our population. In accordance with ICHD-3 recommendations, cases of probable migraine were included in the migraine analysis. This approach is commonly adopted in epidemiologic studies to capture a broader spectrum of the disorder. This study was carried out in a single center which may limit the generalization of these results even if results are similar to the data found in the general population in Madagascar. Disability data were not collected and should be explored in future work.

This high prevalence of headaches and migraine in adolescents and young adults suggests that this is a public health problem because our population is mainly composed of young subjects. We suggest a study in several regions to complete the data to know the impact of the disease at the national level.