Risk factors for atopic and non-atopic asthma in school-aged children from high, and low and middle income countries

BackgroundIt is well established that there are different asthma phenotypes, but whereas determinants of atopic asthma are well studied, little is known about non-atopic asthma. We compared risk factors for atopy, atopic asthma (AA) in atopics, and non-atopic asthma (NAA) in non-atopics, in children in a wide variety of countries. MethodsUsing four studies, across 23 countries, we assessed asthma status and atopy (skin prick tests) for children aged 6-17, plus risk factors from housing, heating, pets, family, diet, and air-quality categories. Using mixed effects logistic regression models we assessed risk factors over 4 pathways: Pathway 1: non-atopic non-asthma to NAA; Pathway 2: non-atopic non-asthma to atopy (no asthma); Pathway 3: atopic non-asthma to AA; Pathway 4: non-atopic non-asthma to AA. We compared the log odds of risk factors between pathways using Pearsons correlation coefficient. ResultsOur final sample of 32741 children comprised 67% with neither atopy nor asthma, 15% with atopy but without asthma, 8% with AA and 10% with NAA. Risk factors were similar between Pathway 1 and Pathway 3 (Pearsons correlation = 0.81, 95% confidence interval = [0.68, 0.94]). In contrast, risk factors differed between Pathway 2 and Pathway 3 (-0.06, [-0.29, 0.17]). DiscussionThese findings indicate that although atopy increases the risk of asthma, the risk factors for subsequently developing asthma are generally the same in those with and without atopy. This raises important questions about the role of atopy in asthma, particularly whether it is an inherent part of the aetiological process or is coincidental. Key messagesO_ST_ABSWhat is already known on this topicC_ST_ABSIt is well established that there are different phenotypes of asthma but little is known about risk factors for non-atopic asthma. What this study addsUsing a novel approach, we found that lifestyle and environmental risk factors for developing asthma are generally similar in atopic children and non-atopic children but the risk factors for atopy are quite different. How this study might affect research, practice or policyOur findings suggest that atopy and asthma may be coincidental in a large proportion of children who are defined as having atopic asthma. This has important implications for our understanding of the causes and mechanisms of different asthma phenotypes, and therefore prevention and treatment of asthma.