In contrast to respiratory allergies, the epidemiology of food allergy has been little studied, and there is no strong evidence for an increasing incidence, either among infants and children or in adults. Neither are there any studies showing regional differences in prevalence. On the contrary, studies in Estonia, Iceland and Sweden indicate a similar prevalence during the first 2 years of life, both in verified food allergy and reported food intolerance. This is despite a low prevalence of respiratory allergies in the two former countries and a high prevalence in Sweden. The major problem with such epidemiological studies lies in the fact that there are no simple diagnostic criteria to verify the diagnosis. So far IgE determinations have been the only available diagnostic test, and their value is limited by poor sensitivity and the fact that at best they would only verify a small proportion of food intolerance, i.e. that caused by IgE-mediated reactions. A diagnosis of food allergy/intolerance must be based on a double-blind placebo-controlled food challenge, and not on the patient's or doctor's impression. More studies are required from different regions in order to identify similarities and differences in the patterns of food allergy. In particular, there is a need for properly conducted epidemiological studies in adults. Such studies should be interdisciplinary, as the cultural and social perceptions of food allergy and food intolerance would be expected to have a major impact on prevalence, perhaps even more than medical factors.