Richer countries carry higher air pollution death burdens because they are older

The standard finding in environmental justice research is that poorer people breathe dirtier air. That is well documented and this study does not dispute it. What it does is separate two things that are usually merged: who is exposed, and who dies [s1].

Those turn out to run in different directions.

The design

The authors combined high-resolution estimates of secondary air pollutants with income data to investigate the relationship between global air pollution-attributable mortality and poverty, across urban and rural contexts and across pollutants [s1].

The distinction they draw is between exposure inequality — how much pollution different income groups breathe — and mortality inequality — how many deaths that exposure produces. Those diverge because attributable mortality is a product of exposure and susceptibility, and susceptibility is dominated by age. An 80-year-old and a 25-year-old breathing identical air do not face identical risk.

The counterintuitive result

High-income countries face higher air pollution-attributable mortality, owing to population aging — in contrast to the established exposure inequality pattern [s1].

This is not a claim that wealthy countries have dirtier air. They do not. It is a claim that attributable deaths are concentrated where old people are, and old people are concentrated in wealthy countries. Attributable mortality is a rate applied to a population, and demographic structure can dominate the exposure term.

Within countries, the pattern is more mixed. Affluent populations within most countries also face higher mortality burdens, but in several low-income countries this pattern reverses, driven by elevated exposures among impoverished rural populations [s1].

Rural exposure in low-income settings is largely a household-fuel story — solid fuel combustion for cooking and heating produces concentrations that dwarf what most urban ambient monitoring captures. Where that is the dominant source, the poorest are both the most exposed and the most burdened, and the aging effect does not offset it.

Where the burdens concentrate

South Asia and Africa exhibit the highest levels of vulnerability to coincident mortality and poverty, with populations living in peri-urban transition zones bearing disproportionate dual burdens [s1].

The peri-urban finding is the one with the clearest policy content. These are the zones outside formal city boundaries where industrial emissions, traffic, waste burning and household fuel use overlap, and where environmental regulation, monitoring and health infrastructure are all thinnest. They are also where a large share of the world's urban population growth is happening.

The authors describe a related mechanism in low-income countries: those living near wealthier urban centres also face elevated health burdens from air pollution — what they call a "cost of opportunity" when air pollutant regulations are weak [s1]. People move toward economic activity and absorb its emissions without the regulatory protection that the same activity carries in a high-income setting.

The study's conclusion is that tailored policy interventions are needed to mitigate amplified health risks in vulnerable communities [s1].

What control measures have delivered

A study published two days later quantifies what the intervention side can achieve, using the largest example available [s2].

Since 2006, China has undertaken sustained efforts to reverse its rise in atmospheric pollution, particularly fine particulate matter [s2]. Using satellite observations and a scenario-based method, the authors estimated the PM2.5-related excess mortality avoided by those control measures [s2].

By lowering PM2.5 concentrations to 2018–2019 pre-COVID levels, Chinese policies prevented approximately 655,000 excess deaths per year compared with a moderate mitigation scenario, and 1.28 million compared with no mitigation [s2]. Within about a decade, the policies prevented the annual loss of up to 19.2 million life-years and a life expectancy reduction of 1.2 years [s2].

The authors' conclusion is that continued reductions at this pace over the coming decades could essentially eliminate the health burden of air pollution [s2].

Reading the two together

The Chinese figures are counterfactual estimates, not observed deaths. They come from comparing a modelled world in which the policies happened against modelled worlds in which they did not, and their magnitude depends entirely on how the counterfactual scenarios were specified [s2]. "Moderate mitigation" and "no mitigation" are analytic constructs, and a different specification produces different avoided-death totals.

With that caveat, the pairing makes a coherent argument. The first study shows that the health burden of air pollution is unevenly distributed in ways that exposure maps alone do not reveal, and that the highest-vulnerability populations sit in South Asia, Africa and peri-urban transition zones [s1]. The second shows that concentration reductions at national scale produce mortality benefits large enough to be measured in life-expectancy years [s2].

The uncomfortable implication of putting them side by side is that the places with the largest remaining gains — peri-urban South Asia and Africa — are the places with the weakest regulatory and monitoring capacity to deliver a Chinese-style reduction.

What these methods cannot do

Both papers estimate deaths rather than count them. Air pollution mortality is not recorded on death certificates; it is inferred by applying exposure-response functions, derived from cohort studies conducted mostly in high-income countries, to modelled concentration surfaces. Every figure in both studies inherits the uncertainty of those functions and their transferability to populations where they were not estimated.

Satellite-derived PM2.5 also requires ground calibration, and calibration networks are sparsest exactly where the first study identifies the highest vulnerability. The mortality inequality result depends on income data that is coarse in many of those same places.

Neither study addresses indoor air directly, despite household solid fuel being a dominant exposure route in the rural low-income populations that drive the within-country reversal [s1].

What to watch

The measurement infrastructure is the bottleneck. Ground monitoring in African and South Asian peri-urban zones, and cohort studies that estimate exposure-response relationships in those populations rather than importing them, would do more to firm up these numbers than any refinement of the models.

Sources

• [s1] Inequality in air pollution-attributable mortality by income level between and within countries — Proceedings of the National Academy of Sciences, published 6 October 2025. https://doi.org/10.1073/pnas.2511394122
• [s2] Avoided mortality by particulate air pollution control measures in China — Science of the Total Environment, published online 8 October 2025. https://doi.org/10.1016/j.scitotenv.2025.180566