RESUMO
Long-term exposure to air pollution has adverse respiratory health effects. We investigated the cross-sectional relationship between residential exposure to air pollutants and the risk of suffering from chronic respiratory diseases in some Italian cities. In the BIGEPI project, we harmonised questionnaire data from two population-based studies conducted in 2007-2014. By combining self-reported diagnoses, symptoms and medication use, we identified cases of rhinitis (n = 965), asthma (n = 328), chronic bronchitis/chronic obstructive pulmonary disease (CB/COPD, n = 469), and controls (n = 2380) belonging to 13 cohorts from 8 Italian cities (Pavia, Turin, Verona, Terni, Pisa, Ancona, Palermo, Sassari). We derived mean residential concentrations of fine particulate matter (PM10, PM2.5), nitrogen dioxide (NO2), and summer ozone (O3) for the period 2013-2015 using spatiotemporal models at a 1 km resolution. We fitted logistic regression models with controls as reference category, a random-intercept for cohort, and adjusting for sex, age, education, BMI, smoking, and climate. Mean ± SD exposures were 28.7 ± 6.0 µg/m3 (PM10), 20.1 ± 5.6 µg/m3 (PM2.5), 27.2 ± 9.7 µg/m3 (NO2), and 70.8 ± 4.2 µg/m3 (summer O3). The concentrations of PM10, PM2.5, and NO2 were higher in Northern Italian cities. We found associations between PM exposure and rhinitis (PM10: OR 1.62, 95%CI: 1.19-2.20 and PM2.5: OR 1.80, 95%CI: 1.16-2.81, per 10 µg/m3) and between NO2 exposure and CB/COPD (OR 1.22, 95%CI: 1.07-1.38 per 10 µg/m3), whereas asthma was not related to environmental exposures. Results remained consistent using different adjustment sets, including bi-pollutant models, and after excluding subjects who had changed residential address in the last 5 years. We found novel evidence of association between long-term PM exposure and increased risk of rhinitis, the chronic respiratory disease with the highest prevalence in the general population. Exposure to NO2, a pollutant characterised by strong oxidative properties, seems to affect mainly CB/COPD.