Studying the prevalence of headaches at age extremes is of important clinical relevance. Pediatric studies inform us about determinants of incident disease; studies of elderly populations inform us about the long-term consequences of headaches, as well as about determinants of headache remission. As with other subspecialties of headache research, research on pediatric headache is an evolving field. However, although substantial advances have been achieved in understanding headaches in adolescents, knowledge of early childhood headaches is not as advanced conceptually. This review provides a theoretical framework for our current understanding, then summarize the results of a large, ongoing, epidemiological study in pre-adolescent children. It is clear that both in adolescents and in pre-adolescents, migraine is frequent. Diagnostic criteria for migraine and chronic migraine are certainly over-restrictive for young children. Migraine often lasts less than 1 hour in young children. A vulnerable population at risk of migraine progression also exists, likely reflecting increased biological predisposition, but also early life exposures. Indeed, it seems that even prenatal exposures of certain substances may increase the risk of migraine progression. Of relevance is the frequency of headaches within a family. Finally, migraine seems to be associated with behavioral hyperactivity, but is not comorbid with attention-deficit disorder and hyperactivity.