Background The health effect of PM_2.5 is a research hotspot in the field of environmental health, as well as a major social issue closely related to the physical health of residents.

Objective This study evaluates the effects of short-term exposure to PM_2.5 on the mortality of cardio-cerebrovascular diseases in Xi'an.

Methods Population deaths due to cardiovascular and cerebrovascular diseases, atmospheric pollutants (PM_2.5, PM₁₀, CO, SO₂, NO₂, and O₃), and meteorological data (temperature and relative humidity) were collected in Xi'an from January 1, 2014 to December 31, 2018. Correlation analysis of the variables was conducted. Generalized additive model based on time series was used to analyze the exposure-response relationship between the concentration of PM_2.5 and cardio-cerebrovascular disease deaths after controlling confounding factors such as time trend and day-of-the-week effect and having average daily temperature and relative humidity as covariates. On the basis of the single lag effect of PM_2.5 (lag0-lag7), the cumulative effect of PM_2.5 and other pollutants on the deaths due to cardio-cerebrovascular diseases at lag02 was estimated. Meanwhile, the effects of PM_2.5 were stratified by gender and age (0-44, 45-59, 60-74, 75-89, and ≥ 90 years).

Results During the study period, the average daily concentration of PM_2.5 was (67.0±55.4) μg·m^-3, and the average number of daily deaths from cardio-cerebrovascular diseases was 70.17±18.53. PM_2.5 was positively correlated with PM₁₀, SO₂, NO₂, and CO (r=0.88, 0.64, 0.68, and 0.79, P < 0.01), and negatively correlated with O₃, temperature, and humidity (r=-0.42, -0.51, and -0.05, P < 0.05). For every 10μg·m^-3 increase in PM_2.5 concentration, the excess risks (ER) at lag0-lag2 were 0.32% (95% CI:0.12%-0.52%), 0.28% (95% CI:0.07%-0.49%), and 0.31% (95% CI:0.09%-0.52%), respectively, while the effects at lag3-lag7 were not significant. At lag02, the ER of the PM_2.5 single pollutant model was 0.45% (95% CI:0.19%-0.70%), and the ERs of PM_2.5 combined with SO₂, NO₂, and O₃ were 0.38% (95% CI:0.10%-0.66%), 0.42% (95% CI:0.14%-0.70%), 0.45% (95% CI:0.20%-0.71%), respectively. The Z test results found that the cumulative lag effect of the multi-pollutant model was comparable to that of the single pollutant model, without significant differences (P>0.05). The ERs of PM_2.5 at lag02 were 0.47% (95% CI:0.17%-0.78%) and 0.42% (95% CI:0.10%-0.74%) for men and women (P < 0.05), and 0.41% (95% CI:0.01%-0.82%), 0.43% (95% CI:0.11%-0.75%), and 0.99% (95% CI:0.26%-1.74%) for the 60-74, 75-89, and ≥ 90 years groups (P < 0.05), respectively. However, the Z test results showed no differences between the sex groups or among the age groups (P>0.05).

Conclusion PM_2.5 may increase the risk of deaths from cardio-cerebrovascular diseases in Xi'an from 2014 to 2018.