Background: The association between smoking and asthma or wheeze has been extensively studied in cross sectional studies, but evidence from large prospective cohort studies on the incidence of asthma during adolescence is scarce.Methods: We report data from a cohort study in two German cities, Dresden and Munich. The study population (n = 2936) was first studied in 1995/6 at age 9--11 years as part of phase II of the International Study of Asthma and Allergies in Childhood (ISAAC II) and followed up in 2002/3. At baseline the parents completed a questionnaire and children underwent clinical examination and blood sampling. At follow up the young adults completed questionnaires on respiratory health, living, and exposure conditions. Incidence risk ratios (IRR) were calculated and adjusted for potential confounders using a modified Poisson regression approach.Results: The adjusted IRR for incident wheeze for active smokers compared with non-smokers was 2.30 (95% confidence interval (CI) 1.88 to 2.82). The adjusted IRR was slightly higher for incident wheeze without a cold (2.76, 95% CI 1.99 to 3.84) and the incidence of diagnosed asthma (2.56, 95% CI 1.55 to 4.21). Analysis of duration and intensity of active smoking indicated dose dependent associations. Stratified analyses showed that the risk of incident wheeze without a cold in atopic smokers increased with decreasing plasma \textgreeka1-antitrypsin levels at baseline (1.64, 95% CI 1.22 to 2.20 per interquartile range).Conclusions: Active smoking is an important risk factor for the incidence of asthma during adolescence. Relatively lower plasma levels of \textgreeka1-antitrypsin, although well above currently accepted thresholds, may increase susceptibility to respiratory disease among atopic smokers.