Comparison of responses to an asthma symptom questionnaire (ISAAC core questions) completed by adolescents and their parents. SCARPOL-Team. Swiss Study on Childhood Allergy and Respiratory Symptoms with respect to Air Pollution

The primary objective of the study was to determine the impact of the identity of the respondent (parents versus adolescents) on prevalence estimates of asthma symptoms in Swiss adolescents. In addition, factors influencing agreement between parents' and adolescents' responses to the same questions were analysed. One thousand three hundred and seventy-four (78.4%) adolescents, aged 14 years, self-completed a questionnaire at school based on the International Study of Asthma and Allergy in Childhood (ISAAC) core questions on wheezing and asthma. The same questions were incorporated into a questionnaire to be completed by the parents at home. The adolescents' self-reported prevalence rates of current asthma symptoms and "asthma ever" were significantly higher than those obtained from the parental questionnaires. 856 (62.6%) parental questionnaires were filled in by parents without the help of the adolescents, 460 (37.4%) were completed by parents and adolescents and 51 (3.7%) were completed by the adolescents without the parents. Prevalence rates were higher when parents and adolescents completed the questionnaire jointly than when questionnaires were completed by parents alone. The level of agreement between parental and self-completed questionnaires was moderate to low (kappa coefficients 0.22-0.68). Agreement between parental and adolescents reports of asthma symptoms was best when questionnaires were completed jointly by parents and adolescents, when the adolescent was a girl, when a family history of asthma was recorded, when the adolescent was a non-smoker, and when the parental education was high. We conclude that the higher reporting of prevalence rates of current asthma symptoms by adolescents compared to reporting by their parents demonstrates the need to take the respondent to a questionnaire into account when comparisons are made between prevalence studies. The results also suggest that factors related to the family milieu influence symptom reporting.