Smoking and new-onset asthma in a prospective study on italian adults

Background: The existence of a cause-effect relation between active smoking and new-onset asthma in adults, although supported by several studies, has not been proven yet. Aims: To prospectively study asthma incidence as a function of smoking habits in the Italian adult population. Methods: A population-based cohort of 5,241 non-asthmatics was enrolled in Verona and Sassari in 1998–2000. The cohort was contacted again in 2007–2009 within the Gene-Environment Interactions in Respiratory Diseases study, and 3,187 subjects (60.8%) answered a screening questionnaire on smoking habits and respiratory disorders. The relation between smoking habits and self-reported new-onset asthma, defined as asthma attacks/use of medicines for asthma, was investigated by a multivariable logistic model. Results: During follow-up 145 new cases of asthma were observed, yielding a cumulative incidence of 4.6% (95% CI 3.9–5.4); cumulative incidence of asthma did not significantly differ among never smokers (76/1,666=4.6%), ex-smokers (30/554=5.4%) and current smokers (39/883=4.4%) (p=0.641). In multivariable analysis the most important risk factor for asthma onset was allergic rhinitis (OR=4.00, 95% CI 3.68–4.35). Compared to never smokers, the risk of asthma onset was slightly increased in ex-smokers (OR=1.28, 1.09–1.49) but not in current smokers (OR=1.01, 0.66–1.53). Current smoking became a significant predictor only when both new-onset wheezing and new-onset asthma were considered as outcome (OR=2.03, 1.35–3.05). Conclusions: In this prospective study current smoking was not a risk factor for new-onset asthma, unless new-onset wheezing was also considered. The increase in asthma incidence among ex-smokers was likely due to reverse causation.