Increase in the prevalence of allergen skin sensitization in successive birth cohorts

Background: Little is known about longitudinal trends in the prevalence of allergen skin sensitization in the general population. Objective: We sought to measure the change in prevalence of allergen skin sensitization over a 9-year period in a cohort of adults and hence to determine whether cross-sectional differences in prevalence between age groups are due to an aging or cohort effect. Methods: In 1991 and 2000, we measured skin sensitization, defined its a wheal diameter of 3 mm or larger than that elicited by a saline control, to Dermatophagoides pteronyssinus, cat fur, mixed grass, Aspergillus funtigatus, and Cladosporium herbarum in a cohort of 1339 adults from Nottingham aged between 18 and 71 years in 1991. Subjects were divided into six 9-year successive birth cohorts, and the effects of birth cohort and the within-subject change from 1991 to 2000 were analyzed in a generalized estimating equation logit model. Results: The unadjusted prevalence of sensitization to any allergen was 30.5% in 1991 and 31.8% in 2000. In cross-sectional analyses the prevalence of sensitization decreased with increasing age at both surveys (risk ratio, 2.15; 95% confidence interval [CI], 1.45-3.17 for 18- to 26-year-old patients relative to 63- to 70-year-old patients in the 1991 survey). In longitudinal analysis there was no within-subject change in sensitization from 1991 to 2000 (adjusted odds ratio, 1.07; 95% Cl, 0.97-1.19), but there was a significant cohort effect (adjusted odds ratio per successive 9 year cohort, 1.27; 95% CI, 1.18-1.37). Conclusion: The cross-sectional decrease in allergen sensitization with age in the general population arises predominantly from a secular increase in sensitization prevalence with successive birth cohorts and not to a loss of sensitization within subjects over time. As a result of this cohort effect, the prevalence of allergic sensitization has increased in this general adult population sample.