To explore environmental impacts such as the meteorological and air pollution factors on respiratory diseases and to provide scientific basis for the prevention of these diseases in Zunyi City, distributed lag non-linear model together with generalized linear and additive models are applied to analyse the exposure-response relationship between environmental factors and respiratory diseases from 2012 to 2016 in Zunyi City. Results show that the changes in respiratory diseases are mainly consistent with the local long-term climatic conditions, and the impact of climate change is dominant. Winter and spring are the peak periods with high respiratory diseases number, and during the Start of Autumn and Stopping the heat periods, the respiratory diseases number is the lowest, indicating that local climatic conditions have positive climatic effects on patients with respiratory diseases during this time period. The impact of temperature on respiratory disease is mainly low-temperature lagged effect with the patients increasing by 31.6% (95%CI: 4.4%~65.8%) if the temperature changes by 1°C.The pressure mainly has the high-pressure lagged effect on respiratory diseases, while the relative humidity has both immediate and lagged effects in lower humidity. The number of respiratory diseases is significantly higher under cold and hot uncomfortable levels than that of comfort levels. PM_(2.5), SO_2, and NO_2 mainly show immediate effects on respiratory diseases, while the CO had the highest risk if lagged for four days. The exposure-response relationship between respiratory diseases and PM_(2.5) shows an monotonously linear distribution, while that of SO_2, NO_2, and CO are J-type distribution, and the synergistic effects between low temperature and high concentration NO_2 or low humidity and high concentration SO_2 both have significant impact on respiratory diseases. The accuracy of annual and seasonal regression equations is over 75% (except for summer equation), and the seasonal equations' prediction effect is better than that of the annual equation.