Background Recent evidence showed that 50% of chronic obstructive pulmonary disease (COPD) may be attributable to air pollution. We aimed to investigate the association between long-term air pollution exposure and COPD symptoms/diagnosis in an Italian epidemiological study. Methods A total of 14 420 adults living in Ancona, Pavia, Pisa, Sassari, Turin and Verona were investigated in 2005–2011. Data on risk factors and health outcomes were collected by questionnaires; mean annual concentrations of particulate matters (PM) like PM 10 and PM 2.5 as well as NO 2 and mean summer concentrations of O 3 (µg/m 3 ) at residential level with a 1 km resolution (period 2013–2015) were obtained by machine learning techniques. The relationship of pollutant exposure and COPD prevalence was assessed by logistic regression models (single pollutant) and principal component logistic regression models (multipollutant) adjusting for sex, age, education level, smoking habits, season of interview, and city-specific climatic index and including a random intercept for cohorts. Results A 10 µg/m 3 increase of PM 10 , PM 2.5 and NO 2 exposure was related to COPD diagnosis and symptoms (OR 1.31, 95% CI 1.03 to 1.65 for PM 2.5 ; OR 1.26, 95% CI 1.03 to 1.54 for PM 10 and OR 1.07, 95% CI 1.00 to 1.15 for NO 2 ) using a multipollutant approach. Similar results emerged for dyspnoea (OR 1.24, 95% CI 1.05 to 1.47 for PM 2.5 ; OR 1.21, 95% CI 1.05 to 1.39 for PM 10 and OR 1.06, 95% CI 1.01 to 1.11 for NO 2 ). Associations between COPD symptoms and summer O 3 were less clear. By multipollutant models, OR estimates were lower than those by single pollutant models. Conclusions Further evidence about the relationship between air pollution and respiratory effects in Italian adults was provided indicating PM as the main driver.