Validity of information on atopic disease and other illness in young children reported by parents in a prospective birth cohort study

BACKGROUND The longitudinal birth cohort study is the preferred design for studies of childhood health, particularly atopic disease. Still, prospective data collection depends on recollection of the medical history since the previous visit representing a potential recall-bias. We aimed to ascertain the quality of information on atopic disease and other health symptoms reported by parental interview in a closely monitored birth cohort study. Possible bias from symptom severity and socioeconomics were sought. METHODS Copenhagen study on Asthma in Childhood (COPSAC) is a clinical birth cohort study of 411 children born of asthmatic mothers from 1999 to 2001. Child health is monitored at six-monthly visits with particular emphasis on atopic symptoms and infections. Data from the first three study years on 260 children was compared with records from their family practitioner as an external reference. RESULTS A total of 6134 medical events were reported at the COPSAC interviews. Additional 586 medical events were recorded by family practitioners but not reported at the interview. There were no missed events related to asthma, eczema or allergy. Respiratory, infectious and skin related symptoms showed completeness above 90%, other diseases showed lower completeness around 77%. There was no meaningful influence from concurrent asthma or socioeconomics. CONCLUSIONS The COPSAC study exhibited full sensitivity to the main study objectives, atopic disease, and high sensitivity to respiratory, infectious and skin related illness. Our findings support the validity of parental interviews in longitudinal cohort studies investigating atopic disease and illness in childhood.