• food allergy;
  • paediatric;
  • allergen skin test;
  • diagnosis;
  • cow milk;
  • peanut;
  • egg;
  • likelihood ratios;
  • ROC curves


The specificity of allergen skin prick testing to diagnose clinically relevant food allergy is controversial.


To determine the specificity of the allergen weal diameter to correctly identify children who react on formal open food challenges.


Over a 9-year period children referred to a tertiary allergy clinic for the evaluation of suspected food allergy were prospectively studied. Allergen skin prick testing to cow milk, egg white and peanut extracts (Dome-Hollister-Stier, Spokane, WA, USA) was undertaken using a lancet technique. All children underwent open food challenges to the relevant food(s) in a hospital clinic. Challenges were classified as positive, if objective signs were seen; negative, if the child could tolerate normal quantities of the food, daily, for one week; or inconclusive if none of the former criteria were met.


Five hundred and fifty-five challenges were undertaken in 467 children: 339 challenges to cow milk, 121 to egg, and 95 to peanut. Fifty-five percentage of challenges were positive, 37% negative, and 8% inconclusive. For each food it was possible to identify a skin weal diameter at, and above, which negative reactions did not occur: cow milk, 8 mm; egg, 7 mm; peanut, 8 mm. In contrast, positive reactions could occur with a skin wheal diameter of 0 mm.


In this high risk referral population it was possible to define skin weal diameters to egg, milk and peanut above which open oral food challenges were positive (100% specificity). By utilizing these measurements the need for formal food challenges can be reduced.