[Epidemiology of headache]

Previous classifications of headache have lacked accuracy. In 1988, the Headache Classification Committee of the International Headache Society (IHS) introduced operational diagnostic criteria for all headache disorders. Since then, epidemiological headache research has been substantially made easily. Although the use of uniform and specific diagnostic criteria may reduce diagnostic variability, several other methodologic problems remain: episodicity, coexisting headaches, measurement techniques, sample source, may explain the differences among studies. All these difficulties overlap and explain the wide range of overall prevalence from 35p.100 to 100 p. 100. Information on migraine epidemiology are available: a one-year prevalence of migraine in America and Western Europe in adults is about 10 à 15 p.100. In France, Henry et al. found a prevalence of 12.1 p.100 (6.1 p.100 for men and 17.6 p.100 for women) for migraine. The aura symptoms are extremely difficult to describe, explaining the heterogeneity of prevalence for migraine with aura. Sex and age distribution as well as the relationship with sociodemographic factors and associated pathologies (stroke, epilepsia.) are now well-known. Epidemiology of tension-type headaches has received less attention. This headache varies both in frequency and severity; this fact make epidemiologic studies more difficult. Cluster headache is better defined by IHS criteria, but its prevalence remains debated. Different methodological approaches could explain these difficulties, rather real differencies in frequency among various population. Cluster Headache prevalence varies from 0.006 p.100 in China to 0.07 p.100 of the population in the republic of San Marino and 0.24 p.100 in the general population of the United states. In conclusion, the development of IHS criteria has made possible the progress of epidemiologic research. Additional studies are required to better define the risk factors and conditions that act as co-morbidity factors.