PM 2.5 ), ozone (O 3 ), nitrogen dioxide (NO 2 ) and sulphur dioxide (SO 2 ) -collected in 46 municipalities in Portugal in 2010 -2015, were retrieved from Portuguese databases [4,5]. General linear models (GLM) were used to analyze time variations and differences between regions for the environmental exposures and for mortality rates. Backward regression models (BRM) were used to assess the relationship between levels of air pollution and mortality from main causes. Results: GLM results show a downward tendency for the air concentration of PM10, NO 2 and O 3 , with different patterns in urban, suburban and rural region. Concomitantly, mortality from main causes has remained fairly constant, except for diseases of the respiratory tract, which has increased, and is higher in the rural areas. Additionally, BRM identified significant associations between rates of mortality from diabetes, diseases of the circulatory system and malignant tumors, with PM10 and NO 2 air concentrations. The model provides a rough estimate of 15 deaths from main causes that could have been avoided per 10000 inhabitants with a 10mg/m 3 reduction of the atmospheric level of PM10 and NO 2 . Discussion and conclusions: significant relationship between mortality rate by main causes and atmospheric concentrations of relevant pollutants has been suggested, providing an estimate of how many lives could be spared by tangible reduction of pollution level.