Background: Very few studies have examined whether a long-term beneficial effect of physical activity on lung function can be influenced by living in polluted urban areas.Objective: We assessed whether annual average residential concentrations of nitrogen dioxide (NO2) and particulate matter with aerodynamic diameters < 2.5 mu m (PM2.5) and < 10 mu m (PM10) modify the effect of physical activity on lung function among never- (N = 2801) and current (N = 1719) smokers in the multi-center European Community Respiratory Health Survey.Methods: Associations between repeated assessments (at 27-57 and 39-67 years) of being physically active (physical activity: >= 2 times and >= 1 h per week) and forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC) were evaluated using adjusted mixed linear regression models. Models were conducted separately for never-and current smokers and stratified by residential long-term NO2, PM2.5 mass and PM10 mass concentrations (<= 75th percentile (low/medium) versus > 75th percentile (high)).Results: Among current smokers, physical activity and lung function were positively associated regardless of air pollution levels. Among never-smokers, physical activity was associated with lung function in areas with low/medium NO2, PM2.5 mass and PM10 mass concentrations (e.g. mean difference in FVC between active and non-active subjects was 43.0 mL (13.6, 72.5), 49.5 mL (20.1, 78.8) and 49.7 mL (18.6, 80.7), respectively), but these associations were attenuated in high air pollution areas. Only the interaction term of physical activity and PM10 mass for FEV1 among never-smokers was significant (p-value = 0.03).Conclusions: Physical activity has beneficial effects on adult lung function in current smokers, irrespective of residential air pollution levels in Western Europe. Trends among never-smokers living in high air pollution areas are less clear.