Development and validation of a structured questionnaire for the diagnosis of oral allergy syndrome in subjects with seasonal allergic rhinitis during the UK birch pollen season

Background Birch pollen–associated oral allergy syndrome, also known as pollen–food syndrome (PFS), is the most common food allergy in adults in the United Kingdom. Because of its characteristic rapid onset of oro-pharyngeal symptoms associated with specific plant foods, it was hypothesized that a history-based questionnaire could accurately diagnose PFS in subjects with rhino-conjunctivitis symptoms in the UK springtime. Objective In this study of diagnostic accuracy, we aimed to validate a simple PFS diagnostic questionnaire and algorithm against a reference diagnostic test method (RTM) comprising diagnosis by expert evaluation of clinical history, skin prick tests and oral food challenge, in subjects reporting allergic rhinitis (AR) in the UK birch pollen season from March to May. Methods Participants were UK adults reporting symptoms of spring time-AR (hayfever). They self-completed a diagnostic questionnaire in addition to undergoing an RTM comprising clinical history, skin prick testing to foods and pollens and oral food challenge. Subjects who reported anaphylaxis were excluded on the basis that they required specialist referral. Results One hundred and twenty three subjects took part in the study. Data from 110 participants were analysed; of the 13 exclusions, four provided insufficient data and nine reported anaphylaxis such that they warranted specialist assessment. Fifty-two participants (47%) were diagnosed with PFS by the RTM in comparison with 51 (46%) by a diagnostic questionnaire and algorithm (P=1.000, McNemar's test). The diagnostic questionnaire and algorithm had a sensitivity of 0.90 (0.78–0.96), a specificity of 0.93 (0.82–0.97), a positive predictive value of 0.92 (0.80–0.97) and a negative predictive value of 0.91 (0.80–0.96) when measured against the RTM. Conclusion and Clinical Relevance The diagnostic questionnaire and algorithm is a practical and robust tool, which enables rapid identification, and therefore management, of individuals with PFS who experience rhino-conjunctivitis symptoms in the UK birch pollen season. Registered with CinicalTrials.Gov. registration number NCT00854958. Cite this as: I. J. Skypala, M. A. Calderon, A. R. Leeds, P. Emery, S.J Till and S. R. Durham, Clinical & Experimental Allergy, 2011 (41) 1001–1011.