Background: Hen's egg food allergy is frequent in childhood and phenotypically heterogeneous. Some children can tolerate extensively heated egg. We investigated whether individual relative responses could differentiate children who tolerate baked egg. Methods: Reactivities to raw, pasteurized or hard‐boiled egg (E), egg white (EW), and egg yolk (EY) fractions were tested by skin prick test (SPT) in 54 egg‐allergic children. IgE‐sensitization to EW and EY was determined by ImmunoCAP and IgE‐binding to EW and 8 EW proteins and to EY and 4 EY sub‐fractions by ELISA. Population heterogeneity was assessed by hierarchical ascending classification upon individual variations of reactivity and links between classifications and clinical features by analyzing the contingency tables. Results: All children had positive SPT to raw E and raw EW and 72% to raw EY. Heating decreased SPT‐reactivity for some children, pasteurization being less effective than hard‐boiling. Children were classed into three classes from relative SPT‐reactivity to raw fractions, two from variations of SPT‐reactivity with each thermal processing or EW/EY ratio of sensitization, and four from their sensitization pattern.Classifications according to heating were found independent of each other. SPT variations with hard‐boiling, IgE‐sensitization (ratio or pattern) were linked to allowance by the physicians of egg in baked products.Conclusions: Egg‐allergic children were often both sensitized to EY and EW, and heterogeneous patterns of relative responses were evidenced. Irrespective of age and level of sensitization, a low EW/EY ratio or SPT getting null with hard‐boiling was found in children allowed to eat baked egg.