Temporal changes in short-term PM2.5-cardiorespiratory mortality associations and implications for the health benefits of the clean air action in the Sichuan Basin, 2014-2022.
Chen J, Ou Y, et al. • International archives of occupational and environmental health • 2026
Temporal changes in short-term PM2.5-related mortality risk partially offset the health benefits achieved through reductions in PM2.5 concentrations during the later Clean Air Action period in the Sichuan Basin.
Key Findings
Results
A significant nonlinear temporal pattern in excess risk (ER) per 10 μg/m3 increase in PM2.5 was observed across the study period.
The nonlinear temporal pattern was statistically significant (P < 0.001)
The pattern was characterized by an increase during 2014-2016, a decline during 2017-2019, and an increase again during 2020-2022
ERs for the three periods were 0.7% (95% CI: 0.6%-0.9%), 0.4% (95% CI: 0.3%-0.6%), and 1.4% (95% CI: 1.2%-1.6%), respectively
The study used a multi-county time-series design in the Sichuan Basin, China, covering 2014-2022
Results
Declining PM2.5 concentrations following the Clean Air Action were associated with 52,831 fewer cardiorespiratory deaths.
This estimate was derived from scenario analyses quantifying mortality changes attributable to PM2.5 concentration reductions
The Clean Air Action (CAA) was the sustained air quality improvement policy under study
The study covered the Sichuan Basin, China, from 2014 to 2022
Results
Changes in the PM2.5-mortality exposure-response relationship (ER changes) were associated with an increase of 10,041 cardiorespiratory deaths.
This offset partially counteracted the 52,831 deaths averted by PM2.5 concentration reductions
The increase in risk was driven primarily by the later CAA period (2020-2022) when ER rose to 1.4% (95% CI: 1.2%-1.6%)
Scenario analyses were used to separate contributions of concentration changes versus exposure-response relationship changes to mortality outcomes
Results
The later-period increase in PM2.5-related mortality risk was consistently observed across sexes and was more pronounced among individuals aged 65 years and older.
Subgroup analyses by sex showed consistent findings in both males and females
The increase in ER during 2020-2022 was more pronounced in the older age group (≥65 years)
These findings suggest differential population vulnerability to PM2.5 exposure over time
Methods
The study applied an extended two-stage analytical framework with longitudinal meta-regression to assess annual trends in excess risk.
Stage one estimated county-specific PM2.5-mortality associations using time-series methods
Stage two used pooled analyses across counties
Longitudinal meta-regression was used to assess annual trends in ER
Period-specific analyses were performed across different stages of the CAA
The study covered multiple counties in the Sichuan Basin, China, from 2014 to 2022
Discussion
The authors highlight the importance of incorporating time-varying health risk and population vulnerability into evaluations of air quality policies.
Temporal increases in short-term PM2.5-related mortality risk partially offset health benefits from concentration reductions
The findings suggest that standard policy evaluations relying on fixed exposure-response relationships may overestimate net health benefits
The study period aligned with implementation of China's Clean Air Action
What This Means
This research suggests that while China's Clean Air Action successfully reduced fine particulate matter (PM2.5) air pollution in the Sichuan Basin between 2014 and 2022, the relationship between PM2.5 exposure and the risk of dying from heart or lung disease did not remain constant over time. Specifically, the risk associated with each unit increase in PM2.5 went up in the early period (2014-2016), then dropped (2017-2019), and then rose again to its highest level in the most recent period (2020-2022). This means that even as air quality improved, people became more sensitive to remaining PM2.5 pollution in the later years of the study.
When the researchers calculated the net health impact, they found that lower PM2.5 concentrations prevented approximately 52,831 cardiorespiratory deaths over the study period. However, the increase in how deadly each unit of PM2.5 became added back roughly 10,041 deaths, partially canceling out those gains. Older adults (aged 65 and above) showed a more pronounced increase in risk during the later period, suggesting this group is particularly vulnerable to changes in air pollution's health impact.
This research suggests that simply measuring how much air pollution decreases is not enough to fully understand the health benefits of clean air policies. The risk that a given level of pollution poses to human health can change over time — possibly due to shifts in population health, the chemical composition of PM2.5, or other factors — and these changes need to be accounted for when evaluating whether air quality policies are delivering their expected health benefits.
Chen J, Ou Y, Qian J, Zeng J, Li S, Wang J, et al.. (2026). Temporal changes in short-term PM2.5-cardiorespiratory mortality associations and implications for the health benefits of the clean air action in the Sichuan Basin, 2014-2022.. International archives of occupational and environmental health. https://doi.org/10.1007/s00420-026-02237-z