Which clinical subgroups within the spectrum of child asthma are attributable to atopy?

Study objectives: The contribution of atopy to childhood asthma has been debated. We aimed to examine the relationship between atopy and asthma, taking into account differences in respiratory symptoms and disease severity. Design: A cross-sectional asthma survey involving the following: (1) a population sample of 758 (81% of eligible) school children aged 8 to 10 years from randomly selected schools in the Australian Capital Territory in 1999, and (2) a hospital-based sample of 78 (70% of eligible) children attending the hospital for asthma. Skin-prick test results to 10 common aeroallergens were available on 722 children and 77 children, respectively. Baseline spirometry was obtained on a subset of school children (n = 515, 78% of eligible). Results: The association between atopy and wheeze by wheeze frequency over the past year was as follows: no episodes (odds ratio [OR], 1.00 [reference]), 1 to 3 episodes (OR, 3.27; 95% confidence interval [CI], 2.15 to 4.97), 4 to 12 episodes (OR, 3.44; 95% CI, 1.75 to 6.75), and > 12 episodes (OR, 8.70; 95% CI, 3.07 to 24.55), with a higher population attributable fraction (PAF) for > 12 episodes (75%) than 1 to 3 episodes (49%). Atopy was moderately related to asthma ever (OR, 2.09; 95% CI, 1.52 to 2.85; PAF, 33%) but strongly related to 1999 hospital attendance for asthma (OR, 16.95; 95% CI, 6.76 to 42.48; PAF, 89%). Adjustment for child age, gas heater use, and maternal smoking near the child did not materially alter these findings. Conclusions: The clinical features of frequent wheeze or hospital asthma attendance are largely attributable to atopy, but infrequent wheeze or a history of asthma ever are not. Atopic children are overrepresented in the severe range of the asthma spectrum.