THE USE OF DISEASE SUB-GROUP CLASSIFICATION WITHIN THE BROAD SPECTRUM OF CHILDHOOD ASTHMA AND ATOPY WHEN EXAMINING ENVIRONMENTAL DETERMINANTS.

Abstract

The clinical syndrome of asthma encompasses several subgroups that reflect airway obstruction of differing aetiologies. We discuss epidemiologic approaches that may be used to reduce disease misclassification with regard to allergen-induced airway obstruction and demonstrate these approaches using two recent analyses. First, we examine infant synthetic bedding and subsequent child wheeze. While house dust mite (HDM) sensitisation is only weakly associated with asthma (RR = 1.7), it is strongly associated with frequent wheeze (more than 12 episodes compared to no wheeze over the past year) (RR = 19.6). Thus, a focus on severe symptoms such as frequent wheeze should reduce disease misclassification in studies examining HDM-related airway disease. We examined a 1988 birth cohort in Tasmania (n = 863, 78% traced) with data available on infant bedding at one month of age and respiratory symptoms at age seven. The associations between synthetic bedding and asthma were not strong. However, strong associations emerged when the frequent wheeze was examined. We found synthetic pillow use at one month of age was associated with frequent wheeze at age seven (ARR 2.5 (1.2–5.5)), independent of childhood exposure. Current synthetic pillow and quilt use was strongly associated with frequent wheeze. Among children with asthma, the age of onset of asthma occurred earlier (p = 0.03) if infant synthetic bedding was used. This shift was particularly evident among children with frequent wheeze. Thus in this cohort, synthetic bedding was strongly and consistently associated with frequent wheeze. These associations would not have been evident if only the global outcome of asthma was considered. Second, we examine the use of mutually exclusive allergen-specific classifications within the broad category of atopy. Our aim was to examine the relative importance of family size on sensitisation to two different allergens:- ryegrass and HDM. An eight-year follow-up birth cohort study of children born 1988–1989 was conducted. The follow-up sample consisted of 498 children residing in Northern Tasmania in 1997 (84% of eligible). Outcome measures included skin prick test reaction to nine aeroallergens. Family size was defined as sibling number in 1997. Children with a positive skin prick test to (SPT) either Der p or Der f HDM but not ryegrass were classified as HDM-exclusive (n = 84). Children with a positive SPT to ryegrass but not HDM were classified as ryegrass-exclusive (n = 43). Family size was associated with reduced ryegrass-exclusive sensitisation (AOR 0.57 (0.39, 0.84) per increase in sibling number) but not HDM-exclusive sensitisation (AOR 0.97 (0.77, 1.23)). The difference in the effect of family size on these sensitisation outcomes was significant (p = 0.02). Thus, large family size was strongly associated with reduced sensitisation for ryegrass allergens but not HDM allergens. Further work is required to assess whether these findings reflect differences between these allergens with regard to the level or timing of early life exposure. These two cohort analyses demonstrate that careful sub-group definition within the broad spectrum of asthma or atopy may lead to new insights on the environmental determinants of atopic disease in childhood.