Abstract
Background:
Tropospheric ozone (O₃) is increasingly becoming the dominant urban air pollutant in China, posing significant public health risks that are exacerbated by meteorological conditions. A clear understanding of how O₃-related health effects are modified by atmospheric factors is crucial for targeted risk mitigation.
Methods:
This ten-year time-series study (2013–2022) was conducted in Taiyuan, China. We analyzed data on daily O₃ concentrations, meteorological factors, and all-cause and cause-specific mortality. The analysis employed Generalized Additive Models (GAMs) to assess the lagged effects of O₃ exposure on mortality and to investigate the interactions between O₃ and key atmospheric determinants, including temperature, sunshine duration, and season.
Results:
The study revealed distinct patterns of O₃-related mortality risk modified by meteorological conditions. The 10-year average daily O₃ concentration was 92.92 μg/m3. O₃ exposure significantly contributed to all-cause, respiratory, and circulatory mortality with lagged effects. While atmospheric pressure, sunshine duration, temperature, and season all influenced the O₃-mortality relationship, the effect was primarily modified through significant interactions with sunshine duration, season, and temperature. These interactive health risks were more pronounced among females and the older adults.
Conclusion:
Our study provides strong evidence that O3 increases the risk of all-cause, respiratory and circulatory mortality in the population. In addition, there were interactions between meteorological factors and O3, primarily involving sunshine duration, season and temperature.
1 Introduction
Air pollution stands as the leading global environmental health risk, accounting for approximately 6.7 million deaths annually and significantly exacerbating public health crises, particularly as anthropogenic ozone (O₃) pollution continues its upward trajectory, further intensifying its role as a critical environmental determinant of the global disease burden (1, 2). In Chinese cities, O3 pollution has become the leading air quality concern in many urban areas (3, 4). Presently, the country faces an unparalleled crisis of O3 pollution, which is even more severe than in other parts of the globe (3). O3 concentrations in China are projected to continue rising through 2050 (5). Therefore, studying the health effects of O3 exposure is essential in public health research.
Ambient O3 pollution continues to pose a significant global environmental health hazard (6). Acute exposure drove a 94% rise in premature deaths in China during 2013–2018, and high-O₃ events boost all-cause mortality (3, 7). Nationwide cohort studies in China have shown that long-term exposure to ozone significantly increases the risk of premature death and reduces life expectancy. It harms the circulatory, cardiovascular, respiratory, and neurological systems, with delayed effects seen in higher non-accidental mortality (2, 6, 7).
Ground O3 levels are influenced by both anthropogenic and meteorological factors, with atmospheric parameters serving as key drivers of surface O3 formation in Chinese urban areas throughout the year. Multivariable regression highlighted varying impacts from hydrometeorological variables [precipitation (PE), thermal radiation (TE), relative humidity (RH), photoperiod duration (SD), barometric pressure (AP), and wind velocity (WS)], particularly thermal radiation and photoperiod duration, which are key catalysts in O3 photochemical processes (8). Another study found that RH, among meteorological factors, is the primary driver of changes in O3 concentration (9). The SD, TE difference between day and night, and extreme high and low TE are all associated with an increased risk of all-cause death among residents (10–12). Additionally, research indicates that both cold and extreme heat influence cardiovascular mortality (13); lower AP values were notably associated with the occurrence of pulmonary embolism (14), while RH and TE were associated with the mortality risk of diabetes mellitus (15). A cohort study found a notable modifying effect of TE on the relationship between mortality and O3 (16, 17). Extreme heat and O3 significantly increase daily hospitalization rates for older patients with coronary heart disease, and their synergistic interaction exhibits a dose–response relationship in exacerbating cardiovascular morbidity (18). Currently, there is no relevant research on the joint effects of other meteorological factors and O3 on the population’s mortality risk. As the capital of Shanxi Province, Taiyuan’s air pollution primarily stems from coal smoke, which may differ from the general patterns. Monthly urban air quality reports from China’s Ministry of Ecology and Environment show a sustained annual increase in O3, the primary monthly pollutant in Taiyuan, between 2015 and 2023. The association between O3 and all-cause mortality risk in the Taiyuan population remains unclear. Additionally, it is uncertain how O3 interacts with meteorological factors to influence the risk of all-cause mortality in this population. Therefore, we have conducted relevant research to provide strong evidence for controlling O3 levels and reducing air pollution in China and globally.
2 Materials and methods
2.1 Data collection
We collected air pollution, meteorological, and cause-of-death monitoring data from January 1, 2013, to December 31, 2022, in Taiyuan City. The air quality measurements consist of the daily average concentrations of PM2.5, PM10, SO2, NO2, and CO over 24 h, as well as the maximum 8-h average concentration of O3 (MDA8 O₃). The climatological records comprise a 24-h mean ambient temperature (°C), humidity (%), wind velocity (m/s), sunshine duration (h), rainfall accumulation (mm), and surface pressure (hPa). Comprehensive mortality figures for all causes among Taiyuan residents were sourced from the National Health Protection Information System of the Chinese Center for Disease Control and Prevention, covering January 1, 2013, to December 31, 2022. Historical meteorological data were obtained from the National Meteorological Science Data Center.1 Daily air pollutant data were sourced from the National Air Quality Real-time Publishing Platform.2 Missing values were addressed using linear interpolation. In this study, the historical observed data had a missing rate of less than 0.1%; therefore, the impact of imputation on the overall results was considered negligible.
According to the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10), the daily all-cause mortality data were classified as due to tumors (C00-D48), respiratory system diseases (J00-J99), circulatory system diseases (I00-I99), and nervous system diseases (G00-G99) (19). Mortality from all causes, as well as fatalities resulting from tumors, diseases of the respiratory system, circulatory system, and nervous system, were categorized according to gender and age, defining individuals under 65 years as young and those aged 65 years and older as older adults; 24-h daily average atmospheric pressure and temperature were both dichotomized according to the median and were operationally characterized as depressed barometric pressure, elevated low barometric pressure, high barometric pressure, low temperature and high temperature in respective order. The average duration of sunshine in China is 6.39 h, based on which we divided the sunshine duration into short and long sunshine durations (20). The warm season encompasses the months of May through September, while the cold season comprises the remaining months from October to April (21). Days with temperatures at or above the 97.5 percentile are designated as “extreme heat”; all others are categorized as “non-extreme heat” (22).
2.2 Statistical analyses
This study used the mean, standard deviation, maximum, minimum, and quartiles to describe all-cause mortality, meteorological factors, and O3 in Taiyuan City from 2013 to 2022. The relationship between O3 and meteorological factors was analyzed using Pearson’s correlation and corresponding heat maps. We conducted quantitative modelling to establish the concentration-mortality associations between ambient O3 exposure and population-wide mortality outcomes, deaths due to respiratory diseases, circulatory diseases, neurological diseases, and tumors, respectively, using generalized additive models (GAMs), which controlled for long-term trends, “the day of week (DOW)” effect, and environmental determinants. Dose–response associations were subjected to stratified analytical evaluation incorporating gender-specific and age-cohort variables. We conducted lag (lag0-lag3) and cumulative lag (lag01-lag03) analyses to adjust for possible lagged impacts. Based on the results of the correlation analysis, the effects of meteorological factors on the relationships between O3 and all-cause deaths, deaths due to respiratory diseases, circulatory diseases, neurological diseases, and tumors were analyzed using GAMs with hierarchical parameters. The lagged and cumulative lagged effects of O3 were also investigated. An overdispersed Poisson regression model was employed. The models are as follow Equation 1:
X represents the concentration of O3; Mk represents dichotomized meteorological factors; β1(X) represents the effect of O3 when the meteorological factor is the reference category; β2(Mk) represents the effect when the meteorological factor is in another category; β3(X: Mk) represents the interaction effect of O3 and meteorological factors; is the non-parametric spline function of other variables including death date, meteorological factors, and other air pollutants. Natural cubic splines with 3 degrees of freedom were used to smooth meteorological factors and air pollutants, while natural cubic splines with 7 degrees of freedom were employed to control for long-term temporal trends. Wt(DOW) is a dummy variable for the day of the week. The effect of O3 when the meteorological factor is another category is β1+β3, and the odds ratio (OR) and corresponding 95% confidence interval are calculated. We used the “mgcv” package from R 4.0.2 software. A p < 0.05 was considered statistically significant. Model specification was selected based on the Akaike Information Criterion for the quasi-Poisson model (Q-AIC), while variable selection was consistent with previous studies (23, 24).
3 Results
3.1 Description of mortality, O3d and meteorological factors in Taiyuan
Table 1 shows the mortality rate in Taiyuan from 2013 to 2022. The results show that the all-cause mortality rate (7.72–13.79%), respiratory disease mortality rate (7.71–12.99%), circulatory disease mortality rate (7.65–14.50%), nervous system disease mortality rate (5.99–18.13%), and tumor mortality rate (8.22–12.33%) in Taiyuan from 2013 to 2022 exhibit an increasing trend. Table 2 included the number of daily deaths, O3 and meteorological factors, with 20.84, 12.15 and 3.82 deaths per day from circulatory diseases, tumors and respiratory diseases, respectively, being the top three. MDA8 O₃, temperature, relative humidity, precipitation, barometric pressure, wind speed and sunshine duration are 92.92 μg/m3, 11.37 °C, 57.13%, 2.59 mm, 927.24 hPa, 1.95 m/s and 7.11 h during the period from 2013 to 2022, respectively. Figure 1 illustrates the temporal changes in O3, meteorological factors, and all-cause deaths over time in Taiyuan from 2013 to 2022.
Figure 1
Table 1
| Years | Overall deaths | Respiratory system disease | Circulation system disease | Nervous system disease | Tumor |
|---|---|---|---|---|---|
| 2013 | 12,504 (7.72) | 1,075 (7.71) | 5,820 (7.65) | 107 (5.99) | 3,647 (8.22) |
| 2014 | 13,179 (8.14) | 1,383 (9.92) | 5,986 (7.87) | 115 (6.44) | 3,705 (8.35) |
| 2015 | 12,445 (7.69) | 1,377 (9.87) | 5,521 (7.26) | 94 (5.26) | 3,624 (8.17) |
| 2016 | 15,673 (9.68) | 1,455 (10.43) | 7,261 (9.54) | 151 (8.45) | 4,406 (9.93) |
| 2017 | 15,725 (9.71) | 1,539 (11.04) | 7,473 (9.82) | 154 (8.62) | 4,393 (9.90) |
| 2018 | 18,256 (11.27) | 1,688 (12.10) | 8,477 (11.14) | 187 (10.46) | 4,948 (11.15) |
| 2019 | 14,654 (9.05) | 1,256 (9.01) | 6,422 (8.44) | 169 (9.46) | 4,270 (9.62) |
| 2020 | 18,628 (11.50) | 1,144 (8.20) | 9,112 (11.97) | 229 (12.81) | 5,033 (11.34) |
| 2021 | 18,550 (11.45) | 1,217 (8.73) | 8,988 (11.81) | 257 (14.38) | 4,877 (10.99) |
| 2022 | 22,325 (13.79) | 1811 (12.99) | 11,031 (14.50) | 324 (18.13) | 5,472 (12.33) |
Annual deaths in Taiyuan from 2013 to 2022 [n (%)].
Table 2
| Variables | Mean | SD | Min | P25 | P50 | P75 | Max |
|---|---|---|---|---|---|---|---|
| MDA8 O₃ (μg/m3) | 92.92 | 55.15 | 3.69 | 51.29 | 82.86 | 125.71 | 371.71 |
| Tem (°C) | 11.37 | 10.58 | −14.90 | 1.80 | 12.30 | 21.00 | 29.60 |
| Hum (%) | 57.13 | 18.93 | 12.00 | 42.00 | 57.00 | 72.00 | 100.00 |
| Pre (mm) | 2.59 | 2.61 | 0.00 | 0.00 | 2.44 | 4.44 | 9.95 |
| BP (hPa) | 927.24 | 7.00 | 911.40 | 921.20 | 927.60 | 932.60 | 948.90 |
| WS (m/s) | 1.95 | 0.99 | 0.30 | 1.20 | 1.70 | 2.40 | 7.20 |
| SSD (h) | 7.11 | 3.87 | 0.00 | 4.70 | 8.10 | 10.00 | 13.80 |
| Overall deaths | 44.34 | 15.57 | 3 | 35 | 43 | 52 | 271 |
| Respiratory system disease | 3.82 | 2.97 | 0 | 2 | 3 | 5 | 59 |
| Circulation system disease | 20.84 | 8.79 | 0 | 15 | 20 | 25 | 141 |
| Nervous system disease | 0.49 | 0.74 | 0 | 0 | 0 | 1 | 6 |
| Tumor | 12.15 | 4.15 | 1 | 9 | 12 | 15 | 40 |
MDA8 O₃, meteorological factors and death toll in Taiyuan from 2013 to 2022.
3.2 Analysis of the correlation between O3 and meteorological factors in Taiyuan
Our study used Pearson’s correlation to analyze the correlation between O3 and meteorological factors. Statistical analyses revealed ambient O3 concentrations demonstrated statistically significant positive associations with ambient thermal metrics and sunshine duration (Pearson’s r coefficients = 0.607, 0.282) while exhibiting inverse correlations with atmospheric pressure measurements (r = −0.552), but not strongly correlated with relative humidity, wind speed, and precipitation, as detailed in Figure 2.
Figure 2
3.3 All-cause mortality risk and stratified analysis in Taiyuan
Figure 3 shows the death risk of O3 from lag0 to lag03 and the death risk after stratification by gender and age, respectively. The analytical findings demonstrate that ambient O3 exposure constitutes a non-negligible risk factor for elevated mortality hazards across all causes, mortality from respiratory diseases and mortality from circulatory diseases. In the all-cause mortality risk for the whole population, men and older population, the risk of death for lag2_O3 was 1.000254 (95% CI: 1.000048–1.000460), 1.000286 (95% CI: 1.000016–1.000556) and 1.000300 (95% CI: 1.000064–1.000537); the risk of death for lag3_O3 was 1.000205 (95% CI: 1.000002–1.000408), 1.000312 (95% CI: 1.000046–1.000578) and 1.000270 (95% CI: 1.0000374–1.000504); the risks of death for lag02_O3 were 1.000109 (95% CI: 1.000015–1.000202), 1.000128 (95% CI: 1.000006–1.000251) and 1.000136 (95% CI: 1.0000282–1.000243); the risks of death for lag03_O3 were 1.0000918 (95% CI: 1.000019–1.000164), 1.000118 (95% CI: 1.0000228–1.000213) and 1.000117 (95% CI: 1.0000337–1.000201), respectively (Figure 3A). The risk of death from respiratory diseases in the whole population, men and the older population, was 1.000711 (95% CI: 1.000167–1.001254), 1.001189 (95% CI: 1.000332–1.002046) and 1.000831 (95% CI: 1.000263–1.001399) for lag3_O3, respectively (Figure 3B). In the risk of death due to circulatory disease for the whole population, women, men and the older population, the risk of death for lag2_O3 was 1.000399 (95% CI: 1.000136–1.000663), 1.000408 (95% CI: 1.000088–1.000728), 1.000386 (95% CI: 1.000019–1.000752), and 1.000434 (95% CI: 1.000139–1.000729), respectively. The overall mortality risk due to circulatory disease across the population indicated that the risk associated with lag3_O3 was 1.000264 (95% CI: 1.000003–1.000524). For lag01_O3, the mortality risk from circulatory disease was found to be 1.000253 (95% CI: 1.000016–1.000489) for the general population, with values of 1.000209 (95% CI: 1.000018–1.000400) explicitly noted for males and older adults. In the risk of death due to circulatory diseases in the whole population, male and older population, the risk of death for lag02_O3 was 1.000163 (95% CI: 1.000044–1.000283), 1.000204 (95% CI: 1.000038–1.000370), and 1.000193 (95% CI: 1.000059–1.000327), and for lag03_O3 was 1.000132 (95% CI: 1.000039–1.000225), 1.000167 (95% CI: 1.0000374–1.000296) and 1.000153 (95% CI: 1.000048–1.000257) (Figure 3C). O3 had no significant effect on the population’s risk of death from neurological diseases and death from tumors (Figures 3D,E).
Figure 3
3.4 Analysis of the interaction between O3 and meteorological factors in Taiyuan
Our study further utilized GAM with stratification parameters to analyze the mortality risk associated with O3 and the interaction between O3 and meteorological factors. Figure 4 illustrates the impact of O3 on the risk of all-cause mortality across the entire population, including men, women, and the older adults, in relation to various meteorological factors, with O3 also interacting with hours of sunlight. O3 is more likely to interact with seasons and temperature in women and the older adults. Figure 5 illustrates the impact of O3 on the risk of death from respiratory diseases in the entire population, as well as in men, women, youth, and the older adults, considering various meteorological factors. O3 interacts with the duration, season, and intensity of sunlight across the whole population, including women and the older adults. Figure 6 shows the impact of O3 on the risk of death from circulatory diseases in the entire population, men, women and the older adults, with different meteorological factors, with O3 interacting with sunshine duration in men, while O3 interacts with both season and temperature in the population as a whole, and with season in the older adults. Figure 7 shows the impact of O3 on the risk of death from neurological diseases in the whole population, males, females, youth and the older adults, with different meteorological factors, with O3 interacting with sunshine duration in the whole population, males and youth; and O3 interacting with atmospheric pressure in the whole population and females, and also with season in females. Figure 8 illustrates the effect of O3 on the risk of death from tumor in males, considering various meteorological factors. O3 interacts with sunshine duration for the entire population, including men and young people, as well as with the presence or absence of weather extremes for the entire population, women, and the older adults.
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
4 Discussion
The results of our study show that the number of deaths in Taiyuan City from 2013 to 2022 exhibits a trend of slow increase year by year, which is consistent with the trend of steady increase in O3 in China as one of the air pollutants (1). Therefore, exploring the relationship between O3 and mortality risk in Taiyuan City is essential. Although the average daily O3 concentration in Taiyuan from 2013 to 2022 is 92.92 μg/m3, which is lower than China’s primary limit (100 μg/m3) and secondary limit (160 μg/m3), the time series shows periodic fluctuations. There are more periods when the O3 level is higher than China’s primary and secondary limits. We suggest that paying attention to O3 pollution levels in Taiyuan is essential to establish an early warning system and strengthen O3 prevention and control. Our study reveals significant correlations among O₃ levels, temperature, sunshine duration, and barometric pressure, likely driven by year-round influences of these meteorological factors as primary contributors to ground-level O₃ formation. Our findings align with prior studies, indicating that climatic parameters such as precipitation, humidity, and wind speed variably affect O₃ levels, with temperature/solar radiation, and relative humidity identified as key determinants in separate analyses (8, 9). Therefore, it is crucial to mitigate O3-related health risks by investigating the influence of meteorological factors on O3 levels and population mortality.
It is generally accepted that O3 has a lagged effect on population mortality. Our results show that O3 has a lag in the risk of all-cause mortality and in the risk of death due to respiratory disease, with gender- and age-stratified results showing that the lag is still present in males and older age groups. An epidemiological study found that a 10 μg/m3 increase in 24-h lagged O3 exposure correlates with a statistically significant 1.38% increase in population mortality risk (25). A longitudinal study linked a 10 μg/m3 rise in 3-day cumulative O3 exposure to a 0.24% increase in population mortality (26). The results show that for every 10 μg/m3 increase in 2-day lagged ambient O₃ exposure, the risk of all-cause mortality increases by 0.0254% (p < 0.05), with comparable risk magnitudes showing non-significant variation. Furthermore, the analytical framework quantified a statistically significant 0.0711% increase in respiratory disease-specific mortality per 10 μg/m3 rise in 3-day cumulative O₃ exposure (p < 0.01), consistent with prior correlational studies identifying lag3 O₃ as the exposure window with the highest respiratory mortality risk (27). A peer-reviewed study reported a statistically significant 0.09% elevation in respiratory disease mortality per 10 μg/m3 24-h lagged O3 exposure (p < 0.05). In comparison, a comparable epidemiological investigation demonstrated a 0.78% increase in respiratory mortality burden with 3-day cumulative O3 exposure (p < 0.01) (2, 27), which was similar to our study. This may be because O3 entering the respiratory system can exacerbate a series of responses, such as oxidative stress, inflammation, and lung injury, ultimately leading to worsening respiratory disease and death (28–30). Our analysis revealed a lagged association between ambient O3 exposure and circulatory disease mortality, persisting across gender and age subgroups. The highest risk occurred at lag 2 O3 exposure, with a 0.0399% increase in circulatory mortality per 1 μg/m3 rise (p < 0.05). Stratified analyses showed stronger female susceptibility (0.0408% vs. 0.0386% in males) and greater older adults vulnerability (0.0434% increase). A prior study reporting 0.11% mortality elevation per 10 μg/m3 lag01 O3 exposure yielded lower risk estimates than our findings (2). Still, our study spanned a considerably more extended period; the results are likely to be more realistic. It also noted that females and the older adults were more susceptible to O3 exposure. This is mainly because O3 enters the circulatory system through the blood-oxygen barrier after entering the lungs via the respiratory tract, causing vascular inflammation, oxidative damage, and vascular endothelial dysfunction (31, 32).
Meteorological factors play a crucial role in the risk of death among O3-affected populations. Our study reveals that meteorological factors contribute to the risk of O3-induced mortality from all-cause, respiratory, circulatory, and neurological diseases, including sunshine duration, season, temperature, barometric pressure, and extreme heat, with sunshine duration, season, and temperature being the primary factors. Nonetheless, the impact of O3 on tumor mortality risk was unaffected by meteorological factors. Currently, there are inconsistent findings on the effect of O3 on tumor mortality, with one study showing no significant correlation between O3 and the risk of death due to malignant tumors (33). In contrast, another study showed that O3 increases lung cancer mortality and that warm and cold seasons play an important role in this effect (34). Research on the interplay between O3 and meteorological factors in relation to tumor mortality risk remains sparse, indicating a need for further investigation. Our study, after categorizing participants by gender and age, revealed that the duration of sunshine influences the risk of mortality from tumors associated with O3 in males.
In summary, sunshine duration emerged as the primary modifier of O3-related mortality risk, with significant interactions observed between O3 exposure and meteorological factors, including barometric pressure, season, temperature, and extreme heat. Our findings align with studies demonstrating temperature-O3 synergies in ischemic heart disease pathogenesis (35) and elevated warm-season O3-associated mortality (17, 36). In general, long sunshine duration, high temperature and warm season contribute to O3’s increased mortality risk.
Our study provides scientific evidence for mitigating O3-related mortality risks by integrating meteorological factors into O3 mortality risk assessments, with a particular emphasis on sunshine duration as a key modifier. While findings support integrating meteorological data into O3 mitigation strategies, limitations persist—notably, the existing literature underscores that extreme heat amplifies O3-associated cardiovascular mortality in populations under 65, a gap warranting further investigation (28, 37). The frequency of extreme heat events in the dataset of this study was limited; thus, further expansion of the dataset is required to explore the role of extreme heat events in ozone-related mortality risk. Future work will expand the dataset to further investigate these interactions. This study provides a robust scientific foundation for informing O3 control strategies and mitigating population-level mortality risks associated with O3 exposure. Nevertheless, it is worth noting that this study has other limitations. Firstly, the collection of pollutant data relies on fixed monitoring stations, which may introduce exposure measurement bias; this is an inherent limitation in most studies (23, 38).
5 Conclusion
Our study provides strong evidence that O3 increases the risk of all-cause, respiratory and circulatory mortality in the population. In addition, there were interactions between meteorological factors and O3, primarily involving sunshine duration, season, and temperature.
Statements
Data availability statement
The data analyzed in this study is subject to the following licenses/restrictions: the datasets generated and/or analyzed during the current study are not publicly available due to data ownership by a third-party institution, but are available from the corresponding author upon reasonable request. Requests to access these datasets should be directed to ZZ, zzh1973@sxmu.edu.cn.
Author contributions
NC: Writing – original draft, Visualization, Data curation, Validation, Conceptualization, Methodology, Writing – review & editing, Formal analysis, Software. XY: Writing – review & editing, Methodology, Data curation, Investigation. YC: Writing – original draft, Formal analysis, Writing – review & editing. LZ: Writing – review & editing, Data curation, Supervision, Funding acquisition. SG: Data curation, Writing – review & editing, Funding acquisition, Supervision. RL: Writing – review & editing, Supervision, Data curation. GZ: Data curation, Supervision, Writing – review & editing, Visualization. LM: Methodology, Data curation, Writing – review & editing. ZZ: Resources, Project administration, Funding acquisition, Supervision, Methodology, Conceptualization, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. The work was supported by the National Natural Science Foundation of China (no. 82273595); Shanxi Province Higher Education “Billion Project” Science and Technology Guidance Project (BYBLD005); the Horizontal Project of Shanxi Centre for Disease Control and Prevention (No. 2023049); the Open Fund from China CDC Key Laboratory of Environment and Population Health, National Institute of Environmental Health, Chinese Center for Disease Control and Prevention (no. 2023-CKL-02); the Startup Foundation for Doctors of Shanxi Province (SD2219) and Startup Foundation for Doctors of Shanxi Medical University (XD2123).
Acknowledgments
The authors would like to thank all participants and investigators, as well as the Chinese Centre for Disease Control and Prevention, for providing the data.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1.
WHO. Sustainable development goal indicator 3.9.1: Mortality attributed to air pollution. Geneva: WHO (2025).
2.
ChenCLiTSunQShiWHeMZWangJet al. Short-term exposure to ozone and cause-specific mortality risks and thresholds in China: evidence from nationally representative data, 2013-2018. Environ Int. (2023) 171:107666. doi: 10.1016/j.envint.2022.107666,
3.
YanDJinZZhouYLiMZhangZWangTet al. Anthropogenically and meteorologically modulated summertime ozone trends and their health implications since China's clean air actions. Environ Pollut. (2024) 343:123234. doi: 10.1016/j.envpol.2023.123234,
4.
MengXWangWShiSZhuSWangPChenRet al. Evaluating the spatiotemporal ozone characteristics with high-resolution predictions in mainland China, 2013–2019. Environ Pollut. (2022) 299:118865. doi: 10.1016/j.envpol.2022.118865,
5.
ChenKFioreAMChenRJiangLJonesBSchneiderAet al. Future ozone-related acute excess mortality under climate and population change scenarios in China: a modeling study. PLoS Med. (2018) 15:e1002598. doi: 10.1371/journal.pmed.1002598,
6.
PengMZhangFYuanYYangZWangKWangYet al. Long-term ozone exposure and all-cause mortality: cohort evidence in China and global heterogeneity by region. Ecotoxicol Environ Saf. (2024) 270:115843. doi: 10.1016/j.ecoenv.2023.115843,
7.
DuHYanMLiuXZhongYBanJLuKet al. Effects of meteorological conditions and anthropogenic precursors on ground-level ozone concentrations in Chinese cities. Environ Health Perspect. (2024) 132:47012. doi: 10.1289/EHP13790,
8.
LiuPSongHWangTWangFLiXMiaoCet al. Environ Pollut. (2020) 262:114366. doi: 10.1016/j.envpol.2020.114366
9.
QianJLiaoHYangYLiKChenLZhuJ. Meteorological influences on daily variation and trend of summertime surface ozone over years of 2015–2020: quantification for cities in the Yangtze River Delta. Sci Total Environ. (2022) 834:155107. doi: 10.1016/j.scitotenv.2022.155107,
10.
FuYWangW. Association between provincial sunshine duration and mortality rates in China: panel data study. Heliyon. (2023) 9:e15862. doi: 10.1016/j.heliyon.2023.e15862,
11.
DingZGuoPXieFChuHLiKPuJet al. Impact of diurnal temperature range on mortality in a high plateau area in Southwest China: a time series analysis. Sci Total Environ. (2015) 526:358–65. doi: 10.1016/j.scitotenv.2015.05.012,
12.
YangXWangJZhangGYuZ. Spatiotemporal distribution and lag effect of extreme temperature exposure on mortality of residents in Jiangsu, China. Heliyon. (2024) 10:e30538. doi: 10.1016/j.heliyon.2024.e30538,
13.
ZhangHWangQZhangYet al. Modeling the impacts of ambient temperatures on cardiovascular mortality in yinchuan: Evidence from a northwestern city of China. Environmental Science and Pollution Research International. (2018) 25:6036–6043. doi: 10.1007/s11356-017-0920-3
14.
AksakalAKergetBCilGet al. Effect of atmospheric pressure changes on the development of pulmonary embolism: A retrospective analysis of 8 years of data. Annals of Saudi Medicine. (2023) 43:204–212. doi: 10.5144/0256-4947.2023.204
15.
WuHZhangXZhangTLiGXuLLiZet al. The relationship of short-term exposure to meteorological factors on diabetes mellitus mortality risk in Hefei, China: a time series analysis. Int Arch Occup Environ Health. (2024) 97:991–1005. doi: 10.1007/s00420-024-02102-x,
16.
ZhangYYinZLiSet al. Ambient PM2.5, ozone and mortality in Chinese older adults: A nationwide cohort analysis (2005-2018). Journal of Hazardous Materials, (2023) 454:131539. doi: 10.1016/j.jhazmat.2023.131539
17.
YanJWangXZhangJet al. Research on the spatial and temporal patterns of ozone concentration and population health effects in the central plains urban agglomeration from 2017 to 2020. PloS One. (2024) 19:e0303274. doi: 10.1371/journal.pone.0303274,
18.
HuangJYangZ. Correlation between air temperature, air pollutants, and the incidence of coronary heart disease in Liaoning province, China: a retrospective, observational analysis. Ann Palliat Med. (2021) 10:12412–9. doi: 10.21037/apm-21-3212,
19.
The International Statistical Classification of Diseases and Related Health Problems 10th Revision.
20.
JiLXuJWuDXieSTangNLSZhangY. Association of disease-predisposition polymorphisms of the melatonin receptors and sunshine duration in the global human populations. J Pineal Res. (2010) 48:133–41. doi: 10.1111/j.1600-079X.2009.00736.x,
21.
ChenYChenSZhangLKangWLinGYangQ. Association between ambient air pollutants and short-term mortality risks during 2015-2019 in Guangzhou, China. Front Public Health. (2024) 12:1359567. doi: 10.3389/fpubh.2024.1359567,
22.
GaoPWuYHeLWangLFuYChenJet al. Adverse short-term effects of ozone on cardiovascular mortalities modified by season and temperature: a time-series study. Front Public Health. (2023) 11:1182337. doi: 10.3389/fpubh.2023.1182337,
23.
ChenSZhouMLiuDLTongSXuZLiMet al. Mortality burden of diabetes attributable to high temperature and heatwave under climate change scenarios in China. NPJ Clim Atmos Sci. (2024) 7:289. doi: 10.1038/s41612-024-00839-3
24.
AjMReWSH. Climate change and human health: present and future risks. Lancet. (2006) 367:9513. doi: 10.1016/S0140-6736(06)68079-3
25.
QinLGuJLiangSFangFBaiWLiuXet al. Seasonal association between ambient ozone and mortality in Zhengzhou, China. Int J Biometeorol. (2017) 61:1003–10. doi: 10.1007/s00484-016-1279-8
26.
YinPChenRWangLMengXLiuCNiuYet al. Ambient ozone pollution and daily mortality: a nationwide study in 272 Chinese cities. Environ Health Perspect. (2017) 125:117006. doi: 10.1289/EHP1849,
27.
LiMDongHWangBZhaoWZare SakhvidiMJLiLet al. Association between ambient ozone pollution and mortality from a spectrum of causes in Guangzhou, China. Sci Total Environ. (2021) 754:142110. doi: 10.1016/j.scitotenv.2020.142110,
28.
ChoHYJedlickaAEChangFHMarzecJBauerAKKleebergerSR. Transcriptomics underlying pulmonary ozone pathogenesis regulated by inflammatory mediators in mice. Antioxidants. (2021) 10:1489. doi: 10.3390/antiox10091489
29.
WengJLiuQLiCFengYChangQXieMet al. TRPA1-PI3K/akt-OPA1-ferroptosis axis in ozone-induced bronchial epithelial cell and lung injury. Sci Total Environ. (2024) 918:170668. doi: 10.1016/j.scitotenv.2024.170668,
30.
TianYXuPWuXGongZYangXZhuHet al. Lung injuries induced by ozone exposure in female mice: potential roles of the gut and lung microbes. Environ Int. (2024) 183:108422. doi: 10.1016/j.envint.2024.108422,
31.
XiaYNiuYCaiJLiuCMengXChenRet al. Acute effects of personal ozone exposure on biomarkers of inflammation, oxidative stress, and mitochondrial oxidative damage - shanghai municipality, China, may-october 2016. China CDC Wkly. (2021) 3:954–8. doi: 10.46234/ccdcw2021.232,
32.
HunterRBairdBGarciaMBegayJGoitomSLucasSet al. Gestational ozone inhalation elicits maternal cardiac dysfunction and transcriptional changes to placental pericytes and endothelial cells. Toxicol Sci. (2023) 196:238–49. doi: 10.1093/toxsci/kfad092,
33.
XiaSYHuangDSJiaHZhaoYLiNMaoMQet al. Relationship between atmospheric pollutants and risk of death caused by cardiovascular and respiratory diseases and malignant tumors in Shenyang, China, from 2013 to 2016: an ecological research. Chin Med J. (2019) 132:2269–77. doi: 10.1097/CM9.0000000000000453,
34.
WangNMengersenKTongSKimlinMZhouMWangLet al. Short-term association between ambient air pollution and lung cancer mortality. Environ Res. (2019) 179:108748. doi: 10.1016/j.envres.2019.108748,
35.
GongXSunFWeiLZhangYXiaMGeMet al. Association of ozone and temperature with ischemic heart disease mortality risk: mediation and interaction analyses. Environ Sci Technol. (2024) 58:20378–88. doi: 10.1021/acs.est.4c05899,
36.
ZhaoZChuJRenJXuCXuXCaoYet al. Effect modification of heat-related mortality risk by air pollutants in Shandong, China. Am J Trop Med Hyg. (2024) 111:440–6. doi: 10.4269/ajtmh.23-0295,
37.
KhatanaSAMWernerRMGroeneveldPW. Association of extreme heat with all-cause mortality in the contiguous US, 2008-2017. JAMA Netw Open. (2022) 5:e2212957. doi: 10.1001/jamanetworkopen.2022.12957,
38.
LiuJDongHLiMWuYZhangCChenJet al. Projecting the excess mortality due to heatwave and its characteristics under climate change, population and adaptation scenarios. Int J Hyg Environ Health. (2023) 250:114157. doi: 10.1016/j.ijheh.2023.114157,
Summary
Keywords
generalized additive model, health risk assessment, interaction, meteorological factors, mortality, ozone
Citation
Cao N, Yang X, Chen Y, Zhao L, Guo S, Li R, Zhu G, Ma L and Zhang Z (2026) Mortality risk effects of ozone and meteorological factors: a 10-year time-series study. Front. Public Health 14:1742521. doi: 10.3389/fpubh.2026.1742521
Received
09 November 2025
Revised
25 January 2026
Accepted
31 January 2026
Published
18 February 2026
Volume
14 - 2026
Edited by
Xu Zhang, Anhui Medical University, China
Reviewed by
Long Cheng, First Affiliated Hospital of Anhui Medical University, China
Qiong Duan, First Affiliated Hospital of Anhui Medical University, China
Updates
Copyright
© 2026 Cao, Yang, Chen, Zhao, Guo, Li, Zhu, Ma and Zhang.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Zhihong Zhang, zzh1973@sxmu.edu.cn
†These authors have contributed equally to this work
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.