A pan-European perspective on food allergy : prevalence, predictors and patient profiles

• Main Author: Lyons, Sarah Anne
• Other Authors: Knulst, A.C., Ree, R. van, Le, T.T.M., Welsing, P.M.J.
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: Utrecht University 2021
• Subjects: food allergy; IgE sensitisation; crossreactivity; prevalence; predictors; epidemiology; diagnosis; Europe; Iceland
• Online Access: https://dspace.library.uu.nl/handle/1874/401598

The largest study to investigate food allergy across Europe reveals that food allergy is common: the prevalence ranges from 0.3% in Greek adults to as high as 6% in Polish children and Swiss adults, with prevalence estimates of 2 to 3% in the Netherlands. Geographical variations in prevalence and causative foods are strongly influenced by environmental factors. In Northern and Central Europe, cross-reactivity with birch pollen is largely responsible for the many fruit (apple, peach, kiwi), vegetable (carrot, celery) and tree nut (hazelnut) allergies. Frequent seafood consumption in Spain and Iceland corresponds with these countries’ relatively high prevalence of fish and shrimp allergy. Dog ownership in early childhood can protect against later food allergies, but other previously reported protective early-life exposures, such as attending day care or growing up on a farm, have limited impact. The ‘gold standard’ diagnostic test for food allergy, oral food challenge, is burdensome and costly. Accurate diagnostic alternatives are crucial. This thesis demonstrates that almost half of adults presenting with food adverse reactions can be correctly classified as non-allergic using only information directly available from patient history. Blood tests detecting allergy-related antibodies (IgE) can also decrease the need for challenges, but performance is variable per food and with age. Levels of IgE to specific allergenic proteins in hazelnut are shown not to distinguish between presence and absence of hazelnut allergy in Dutch adults, in contrast to previous findings in children. Estimating the risk of severe reactions is perhaps even more important than establishing food allergy. Statistical models combining information from patient history with results from blood and skin prick testing are found to accurately predict severity of tree nut allergy, potentially further reducing the demand for challenges in the future.