Limited agreement between written and video asthma symptom questionnaires

Abstract

The prevalence of asthma remains difficult to determine with precision with no absolute or 'gold' standard for diagnosis. A recently developed video questionnaire for epidemiological studies with less reliance on understanding written questions provides another tool for determining prevalence and severity of asthma. This report from the International Study of Asthma and Allergies in Childhood (ISAAC) examines the agreement between the ISAAC video questionnaires on respiratory symptoms and reported asthma. Between December 1993 and April 1995, 4952 children aged 13-14 years in two Canadian communities completed sequentially the ISAAC written and video questionnaires at school. The agreement between responses to the two questionnaires for reported wheeze ever, current wheeze, wheeze on exercise, and nocturnal wheeze (the latter three questions relating to symptoms in the last 12 months), and to any combination of the latter three questions was examined in the full sample and in those reporting diagnosed asthma, using concordance and kappa coefficients as measures of agreement. The prevalences of wheeze ever, current wheeze, wheeze on exercise, and nocturnal wheeze were significantly lower based on responses to the video questionnaire compared with the written questionnaire in both regions in the full sample and in those labeled as having asthma. Although concordance between video and written questionnaires always exceeded 60% and often exceeded 70% for related questions, agreement measured by the kappa statistic for each question was only fair to moderate (κ = 0.22-0.51). We conclude that the video questionnaire yields lower reported prevalence rates for asthma symptoms, and that there is limited agreement between responses to the two questionnaires that is not explained by issues of language, culture, or literacy. (C) 2000 Wiley-Liss, Inc.