The research team, led by Xiuqiao Wang and colleagues at Sichuan University’s West China School of Public Health together with the Sichuan Center for Disease Control and Prevention, assembled an unusually rich dataset spanning 2014 to 2019. Daily meteorological records, ground-level PM2.5 concentrations, and cardiorespiratory mortality counts were collected for all 130 counties of the Sichuan Basin, a densely populated topographic bowl in southwestern China where stagnant air and temperature extremes frequently coincide. Over the six-year window, the investigators recorded a total of 1,744,063 deaths from cardiovascular and respiratory causes, providing the statistical power needed to dissect interactions that smaller studies often miss.

At the methodological heart of the study sits the distributed lag nonlinear model, or DLNM, a statistical framework that has become the workhorse of environmental epidemiology. Unlike simpler regression approaches, a DLNM can simultaneously capture how a risk factor’s effect varies in intensity with the magnitude of exposure and how that effect unfolds across time, from the day of exposure to weeks afterward. This dual flexibility matters because temperature-related mortality follows strikingly different temporal signatures depending on whether the exposure is heat or cold. The researchers applied this framework county by county before pooling results, allowing local climate and pollution profiles to shape the estimates rather than imposing a single basin-wide curve.

What emerged from the temperature modeling was a classic U-shaped association between ambient temperature and cardiorespiratory mortality risk, with deaths climbing at both thermal extremes and reaching a minimum at moderate temperatures. But the two arms of the U behaved very differently in time. Low-temperature exposure produced a persistent, lagged association: elevated mortality risk lingered for roughly two weeks, with effects still detectable at lags of 13 to 14 days after the cold exposure occurred. High-temperature exposure, by contrast, acted with brutal immediacy. The highest heat-related risk appeared at lag 0, on the very day of exposure, and faded quickly thereafter. This asymmetry has practical consequences, because it means cold-related deaths accumulate gradually and can be partially anticipated, while heat-related deaths strike before public health systems can fully mobilize.

The study’s most consequential innovation lay in how it handled joint exposure. Rather than treating temperature and PM2.5 as separate predictors whose effects might modify one another in a conventional statistical sense, the team constructed discrete joint exposure events, combinations such as high temperature with high PM2.5 or low temperature with high PM2.5, and estimated the mortality risk, the attributable fraction of deaths, and the degree of additive interaction for each combination. This event-based framing mirrors how people actually experience the environment: not as abstract continuous variables but as days that are simultaneously hot and smoky or cold and hazy.

The headline finding was stark. High temperature combined with high PM2.5 produced a relative risk of 1.259 for cardiorespiratory mortality, with a 95 percent confidence interval of 1.191 to 1.330, and an attributable fraction of 20.56 percent, meaning roughly one in five cardiorespiratory deaths occurring under those joint conditions could be attributed to the combined exposure. No other exposure combination in the analysis reached that level of attributable burden. In other words, the worst days in the Sichuan Basin, those that paired extreme heat with heavy particulate pollution, were responsible for a disproportionate share of the region’s cardiovascular and respiratory death toll.

Crucially, the researchers went beyond risk estimation to test whether the two hazards interacted additively, using the relative excess risk due to interaction, or RERI, a standard epidemiological measure of synergism on the additive scale. For high temperature and PM2.5, the RERI was 0.195 with a 95 percent confidence interval of 0.109 to 0.282, a statistically significant positive value indicating that the joint effect of heat and particulate pollution exceeded the sum of their individual effects. The biological plausibility of such synergy is well grounded. Heat stress increases cardiac output, dehydration, and blood viscosity while triggering systemic inflammation, whereas inhaled fine particles penetrate deep into the alveoli, provoke oxidative stress, and can translocate into the bloodstream, promoting thrombosis and vascular dysfunction. When both stressors act on the same day, the physiological burden on a compromised heart or lung can compound in ways that neither stressor alone would produce.

Instructively, the same synergy did not appear on the cold side of the temperature spectrum. The analysis detected no significant additive interaction between low temperature and PM2.5, even though joint exposure events involving cold and particulate pollution were still associated with elevated mortality risk and attributable burden compared with baseline conditions. This asymmetry suggests that the mechanisms by which cold harms the cardiorespiratory system, including vasoconstriction, increased blood pressure, and heightened blood coagulability, may overlap substantially with the pathways activated by particulate inhalation, leaving little room for the two to amplify each other further. Heat, by contrast, appears to stress the body through channels that pollution then exploits, or vice versa, producing genuine excess risk beyond simple addition.

The findings arrive at a moment when climate change and urban air pollution are converging on the same populations. The Sichuan Basin is particularly vulnerable: its bowl-shaped topography traps air masses, its winters bring temperature inversions that concentrate pollutants near the surface, and its summers deliver intense heat and humidity. Previous work by some of the same authors had separately mapped temperature-mortality and PM2.5-mortality associations across the basin; the new study closes the loop by showing that the intersection of the two hazards is where the largest burden concentrates. The results align with a growing international literature, including multicountry analyses of extreme temperatures and cardiovascular mortality and studies of heat-related mortality across hundreds of cities, that have identified air pollution as a consistent modifier of temperature-related death risk.

For public health authorities, the message is that single-hazard warning systems are no longer sufficient. The authors argue that future prevention and control strategies should explicitly target joint exposure to both high and low temperatures and PM2.5 pollution, integrating air quality management with heat and cold response plans. Practically, that could mean issuing compound alerts when heat and pollution thresholds are crossed together, prioritizing cooling centers and clean-air shelters in the same neighborhoods, and timing interventions such as hospital surge planning around forecasted compound events. As global warming pushes more regions toward hotter summers while industrializing areas continue to battle particulate pollution, the Sichuan Basin’s experience offers a quantified preview of a compound hazard that much of the world is only beginning to confront, and a reminder that the deadliest air is often not simply hot or dirty, but both at once.

Subject of Research: Joint short-term exposure to ambient temperature and PM2.5 and its effects on cardiorespiratory mortality and disease burden in the Sichuan Basin, China

Article Title: Short-term joint exposure to temperature and PM2.5 and its effects on cardiorespiratory mortality and disease burden: a multicounty study in the Sichuan Basin, China

Article References: Short-term joint exposure to temperature and PM2.5 and its effects on cardiorespiratory mortality and disease burden: a multicounty study in the Sichuan Basin, China. (n.d.). https://doi.org/10.1007/s11869-026-02121-2

Image Credits: AI Generated

DOI: 10.1007/s11869-026-02121-2

Keywords: PM2.5, temperature, cardiorespiratory mortality, joint exposure, Sichuan Basin, air pollution, distributed lag nonlinear model, attributable fraction, additive interaction, heat stress, public health, environmental epidemiology