The primary aim of the study was to assess the validity of the ISAAC core questions on rhinitis in a population of Swiss school children by comparing them to skin prick test results. Second, the positive predictive value in detecting atopy among children with rhinitis symptoms was determined. Third, agreement between parental reports of hay fever and rhinitis symptoms was evaluated, since earlier Swiss prevalence surveys had exclusively relied on reported hay fever.
Material and methods: Two thousand nine hundred and fifty-four (81.2%) parents of 7, 10 and 14-year old children filled in an exhaustive questionnaire which included the ISAAC core questions on rhinitis. Two thousand one hundred and twenty children also underwent skin prick testing against six common aeroallergens (grass mixture, birch, mugwort, D. pteronyssinus, cat and dog dander). The analysis is restricted to children with both questionnaire data and skin prick test results.
Results: Sensitization to any allergen was most strongly associated with reported hay fever (OR = 5.7, 95% CI 4.4-7.4), nose problems accompanied by itchy-watery eyes (OR = 4.4, 95% CI: 3.3-5.7), symptoms occurring only during pollen season (March through September) (OR = 4.9, 95% CI: 3.6-6.5) and a combination of these latter two symptoms (OR = 5.8, 95% CI: 4.1-8.1). The association was stronger for a sensitization to outdoor allergens than for indoor allergens. The specificity of the various questions was high, ranging from 77.5% to 97.6%, but the sensitivity was low (2.6% to 42.7%). The positive predictive value for atopy among children with symptoms was 63% for sneezing accompanied by itchy-watery eyes, 67% for symptoms occurring only during the pollen season and 70% for reported hay fever. However, agreement between reported rhinitis symptoms and hay fever was only moderate. About one third of the children with symptoms indicative of seasonal rhinitis did not report the label "hay fever".
Conclusions: We conclude from our analyses that the ISAAC core questions on rhinitis are highly specific and therefore useful in excluding atopy. In addition they have a high positive predictive value in detecting atopy among children with symptoms, but they are not helpful for detecting atopy in a general population of children (low sensitivity). To monitor time trends in the prevalence of allergic rhinitis in Switzerland, questions on rhinitis symptoms as well as on the diagnostic label "hay fever" have to be included in a questionnaire because they contain complementary information since under-diagnosis of allergic rhinitis is common.