Analyzing symptom data in indoor air questionnaires for primary schools

Abstract

Questionnaires on symptoms and perceived quality of indoor environment are used to assess indoor environment problems, but mainly among adults. The aim of this article was to explore best ways to analyze and report such symptom data, as part of a project to develop a parent-administered indoor air questionnaire for primary school pupils. Indoor air questionnaire with 25 questions on child's symptoms in the last 4 weeks was sent to parents in five primary schools with indoor air problems and in five control schools. About 83% of parents (N=1470) in case schools and 82% (N=805) in control schools returned the questionnaire. In two schools, 351 (52%) parents answered the questionnaire twice with a 2-week interval. Based on prevalence of symptoms, their test-retest repeatability (ICC), and on principal component analysis (PCA), the number of symptoms was reduced to 17 and six symptoms scores were developed. Six variants of these six symptom scores were then formed and their ability to rank schools compared. Four symptom scores (respiratory, lower respiratory, eye, and general symptoms) analyzed dichotomized maintained sufficiently well the diversity of symptom data and captured the between-school differences in symptom prevalence, when compared to more complex and numerous scores.