Background Asthma in bakery workers caused by exposure to wheat flour proteins is an important occupational health problem. Until recently, gravimetric dust measurements were the only available technique for quantitative exposure assessment in bakeries. However, it is questionable whether dust levels are a good exposure parameter or only give a crude approximation of the actual flour allergen concentration. Objective In the present study we have investigated a method to measure wheat flour antigens with immunochemical methods. Methods Wheat flour antigens were measured in 449 personal dust samples taken in bakeries, using enzyme-linked immunosorbent assay (ELISA) inhibition and an anti-wheat IgG4 serum pool. Western-blotting was performed to compare the wheat flour proteins detected by IgE and IgG4. Results Electrophoresis and immunoblotting showed that many wheat flour proteins can bind IgG4 and IgE, but also a reasonable similarity in major allergens detected by our IgG4-serum pool and IgE-positive sera. Inhibition tests showed some crossreactivity with some cereal species, but not with other ingredients used in bakeries. In bakeries, large differences in personal airborne flour levels were found between occupational titles. For several groups clear differences in wheat antigen exposure levels existed, where no differences in dust exposure levels could be found. The relationship between dust and wheat antigen exposure varied considerably, depending on the specific bakery occupation, the size of the bakery, and the type of product produced by the bakery. This study also shows that personal sampling of wheat antigens is possible on a large scale and can be used for epidemiological field studies. Conclusion Measurement of airborne wheat antigens in bakeries is a more specific and sensitive measurement tool than measuring dust samples, and will probably be essential for epidemiologic field studies focusing on exposure-response relationships.