Study On The Health Risk Of Air Pollution And Extreme Temperature On Cardio-Cerebrovascular Disease Mortality Among Residents

Background and purposeEpidemiological evidence has shown that cardio-cerebrovascular diseases were the foremost causes of death worldwide and posed a severe threat to human health,while increasingly prominent environmental issues and climate change resulting in air pollution and extreme temperature events might have important implications for cardio-cerebrovascular diseases.Lishui District,Nanjing,the first entitled a“Healthy Jiangsu practice demonstration area”in Jiangsu Province was selected as the study area.It is a burgeoning Chinese city to visualize the conception of“Healthy China”,and the local government has been in the process of government interventions on abating air pollution.The study focused on the characteristics of air pollutant concentration changes,the trends of average temperature changes and heat or cold effects,the occurrence of extreme temperature events such as heat waves and cold spells,the magnitude of changes in temperature variability,and the mechanism of interaction between air pollution and extreme temperature on the occurrence and development of cardio-cerebrovascular diseases in Lishui District from 2015 to 2019.In terms of developing and optimizing the synergistic management pathways for air pollution management and climate change response,this study provided informative insights on the significance and magnitude of its relationship to reduce the health risks of residents in Lishui District.Methods（1）In this study,the mortality data from cardio-cerebrovascular diseases（cardiovascular diseases,cerebrovascular diseases,and arteriosclerotic heart diseases）was collected from 2015 to2019.The matched air quality observations(PM:PM_2.5 and PM₁₀;gaseous pollutants:SO₂,NO₂,O₃,and CO)and meteorological observations（average temperature,maximum temperature,minimum temperature,and relative humidity）were also included.The distributions of air pollutant concentrations,meteorological observations,and mortalities were assessed by the K-S test.The correlation was calculated by Spearman’s rank correlation coefficients.（2）This study applied the generalized additive model（GAM）with a quasi-Poisson distribution to explore the association between short-term exposure to air pollution and the daily mortality risk for cardio-cerebrovascular diseases.Different lag structures were used,including single-day lag（lag0–lag7）and moving average lag（lag01–lag07）.The two-pollutant model was employed to validate the potential confounding effects of gaseous pollutants and the robustness of the single-pollutant model fit results.Akaike information criterion（AIC）was implemented for the degree of freedom selection.（3）This study fitted GAM with a penalized smoothing spline to evaluate the shape of exposure-response curves to visualize the effects of air pollutant concentrations on cardio-cerebrovascular disease mortality.The season（warm and cold）,age groups（0–64 years,65–74 years,and≥75 years）,and sex（male and female）were divided into different subgroups.Z test was conducted to observe whether the differences were statistically significant as a modifier.（4）The air temperature on the mortality risk was assessed in three dimensions:average temperature,extreme temperature,and temperature variability（TV）.Firstly,the minimum mortality temperature（MMT）was determined by the threshold model and set as the reference temperature.The"cross basis"function of average temperature and the lag effect was established in the distribution lag nonlinear model（DLNM）.3D and contour plots were made to observe the effect of the average temperature on the mortality risk under different lag days.This study defined the P10 and P90 of the annual average temperature distribution as the cold effect and the heat effect.The cold effect and the heat effect on the mortality risk were quantitatively analyzed under the reference of MMT.（5）The extreme weather was defined based on temperature threshold percentile（heat wave:P95-P99;cold spell:P1-P5）and duration（2-4 days）.DLNM was applied to analyze the main effects of extreme temperature events.The mortality risk of heat wave（cold spell）was estimated compared with non-heat wave（cold spell）using different definitions.（6）TV was constructed by the standard deviation of the minimum and maximum temperatures during different exposure days.DLNM was applied to explore the correlation between TV and mortality risk.The strongest mortality effect of TV was filtered.（7）The nonparametric bivariate response surface model was developed for qualitative evaluation of the interaction between air pollutant concentrations and air temperature.The average temperature was divided into three strata,defined as the extreme high-temperature strata（T₁≥P95）,medium temperature strata（P5≤T₂<P95）,and extreme low-temperature strata（T₃<P5）,respectively.The interaction term of air pollutant concentrations and the average temperature was introduced into GAM.This study quantitatively assessed the mortality risk due to air pollution in different temperature strata.Z test was conducted to observe whether the differences between the estimates of the effects of temperature strata,genders,and age groups were statistically significant.Results（1）The average annual concentrations of PM_2.5,PM₁₀,SO₂,NO₂,and CO presented a decreasing trend year by year,showing higher concentrations in the cold season than in the warm season.In contrast,O₃ has increased year with higher concentrations in the warm season than in the cold season.The air temperature showed"M"type fluctuations.PM concentration and the cardio-cerebrovascular disease mortality were positively correlated.PM_2.5 and PM₁₀ were highly positively correlated,with a correlation coefficient of 0.85.（2）The strongest effect of PM occurred at lag0.Each 10μg/m³ increase in PM_2.5 and PM₁₀concentrations were associated with an increase of 1.33%（95%CI:0.08%,2.60%）and 1.12%（95%CI:0.43%,1.82%）in daily cardio-cerebrovascular disease mortality.For cumulative risk,the estimate of association was strongest at lag03 in PM_2.5.The seasonal stratification suggested that in the cool season,a 10μg/m³ increase in PM_2.5 and PM₁₀ concentrations were associated with a 1.68%（95%CI:0.23%,3.15%）and 1.14%（95%CI:0.33%,1.95%）increase in daily mortality.The age stratification suggested that the greatest effect of PM₁₀ occurred in the 65-74 years age group.The gender stratification suggested each 10μg/m³ increase in PM₁₀ concentration was associated with an increase of 1.31%（95%CI:0.38%,2.24%）and 1.34%（95%CI:0.38%,2.30%）for male and female,respectively.（3）The exposure-response curves were approximately nonlinear and had a non-monotonic function over the entire exposure range of the concentrations.The results of PM_2.5 and PM₁₀ exposure curves showed similar trends,but there was no evidence of a threshold.（4）The exposure-response curve between average temperature and mortality risk was U-shaped.The MMT for cardio-cerebrovascular diseases was 14.7°C.The 3D and contour plots suggested that the relationship between air temperature and mortality risk was fluctuating and non-linear.The heat effect from cardio-cerebrovascular and cerebrovascular diseases appeared and peaked at lag0 with relative risks（RR）of 1.35（95%CI:1.06,1.71）and 1.44（95%CI:1.10,1.88）.At lag2,the cold effect on mortality risk began to appear,with RRs of 1.12（95%CI:1.03,1.23）and 1.19（95%CI:1.05,1.35）.（5）Heat waves increase the mortality risk,particularly significant effect on cerebrovascular diseases.The higher the temperature threshold and the longer the duration set,the greater the RR presented.The most significant effect of cerebrovascular disease mortality was observed when the temperature was higher than P99 for at least 4 days,with the RR of 2.85（95%CI:1.02,7.92）.（6）The effect of TV on cerebrovascular diseases was most significant at lag07 with the RR of1.04（95%CI:1.00,1.08）.The mortality risk from TV on cardio-cerebrovascular diseases and atherosclerotic heart diseases reached its highest value at lag0 with RRs of 1.02（95%CI:0.99,1.04）and 1.02（95%CI:0.96,1.07）,respectively.（7）The bivariate response surface model suggested that the highest cardio-cerebrovascular disease mortality occurred at high temperatures and high concentrations of air pollutants.The temperature stratification model indicated that PM_2.5 and PM₁₀ mortality effects were apparent in the mid-temperature strata.The ERR of cardio-cerebrovascular diseases attributable to NO₂ and CO in extreme high-temperature strata was 27.89%（95%CI:4.16%,57.03%）and 6.18%（95%CI:0.94%,11.68%）,respectively,and there was an interaction effect.Effect values attributed to O₃ in people over 65 years old at the extreme high-temperature strata were statistically different from other age groups.Conclusion（1）There was a non-linear positive association between air pollutant concentrations and the cardio-cerebrovascular disease mortality risk in Lishui district,with the lagged effect being highest on the current day of exposure.（2）The effect of the temperature on the cardio-cerebrovascular disease mortality risk suggested a U-shaped correlation and lag effects.The unsuitable temperatures increased the cardio-cerebrovascular disease mortality risk.The heat effect was brief but violent,while the cold effect appeared with a lag and was weaker than the heat effect.The heat wave was associated with cardio-cerebrovascular disease mortality risk.The higher temperature threshold and the longer duration set,the greater mortality risk presented.Cerebrovascular disease mortality was more sensitive to TV.（3）High concentrations of air pollutants and high temperatures enhanced the cardio-cerebrovascular disease mortality risk.There was an interaction between air pollutants and extreme high-temperature strata.The seniors over 65 years of age might be more sensitive to air pollution in extreme high-temperature strata.