Abstract: Objective To analyze the influence of PM_2.5 on different causes of death in Jiamusi City from 2015 to 2021. Methods The cause of death monitoring data,meteorological data and air pollutant monitoring data from 2015 to 2021 were collected,the Poisson generalized addition model was used to control the influence of confounding factors such as meteorological factors,time trend and other pollutants,and to analyze the influence of PM_2.5 exposure on population mortality risk. Results From 2015 to 2021,the total number of deaths in Jiamusi City was 108 800,and the average annual concentration of PM_2.5 was 29.910 μg/m³. There was a positive correlation between PM_2.5 concentration and death of all diseases（all P<0.05）. The change of PM_2.5 concentration had a lag effect on the risk of death from diseases,and the cumulative lag of 5 days had the greatest effect on total death and death from circulatory diseases,with ER values of 0.053%（95%CI：0.021-0.086） and 0.078%（95%CI：0.035-0.120）,respectively. The 2-day lag concentration had the greatest effect on death from respiratory diseases,with ER value of 0.079%（95%CI：0.004-0.154）. Men and older adults ≥65 were more sensitive,with ER values of 0.061%（95%CI：0.024-0.099） and 0.054%（95%CI：0.024-0.085）,respectively. The pollutant model showed that when PM_2.5 concentration increased by 1μg/m³,the risk of total death,the risk of death from circulatory diseases,and the risk of death from respiratory diseases increased by 0.050%（95%CI：0.020-0.090）,0.080%（95%CI：0.040-0.120） and 0.080%（95%CI：0.010-0.015）. Conclusion The exposure of PM_2.5 in the atmosphere in Jiamusi City from 2015 to 2021 is closely related to the increased risk of death of the population in Jiamusi City,especially for men and the elderly population.

Key words: Fine particulate matters, Cause of death monitoring, Generalized additive model, Time series analysis