Background Many clinical and epidemiological studies have measured the prevalence of IgE sensitization using skin tests and/or serum-specific IgE. Most of them have been done in only one country using a battery of selected allergens relevant to that country. In multi-centre studies, the number of tested allergens is often limited by the cost. It is therefore difficult to compare prevalence of sensitized subjects between studies.
Objective To define the number and the type of allergen that should be tested in order to characterize a person as sensitized.
Method Subjects were selected from the European Community Respiratory Health Survey I. All subjects underwent skin prick tests to nine of the most common allergens. In addition, two local allergens were tested in some centres.
Result Using nine allergens, 35.6% of the 11 355 subjects were sensitized. The prevalence of sensitization increased with the number of tested allergens. Seven allergens enabled the identification of almost all sensitized subjects, adding another one inducing, in most countries, an increase of prevalence under 0.5%. Adding one local allergen to the battery of tests increased the overall estimated prevalence by only 1%. This increase was not seen in Ireland and was less marked in the United Kingdom (0.3%) but was greater in France (2.6%), Australia (2.5%) and Belgium (1.9%).
Conclusion Seven selected allergens (Dermatophagoides pteronyssinus, cat, grass, birch, olive pollen, Alternaria and Cladosporium) allow the identification of almost all sensitized subjects in epidemiologic studies. Inclusion of local allergen should be considered in a standard panel for international studies.