Cleaner air, longer lives: What does science say?

In a different context, Ella Kissi-Debrah, a bright nine-year-old from suburban Britain, died due to her 27th asthma attack triggered by pollution. She used to live near a notoriously congested and polluted roadway. She had her first attack at the age of seven. 

Scientists emphasise that personal stories like Ella’s resonate more deeply than statistics, helping to expose the human toll of air pollution. Similar accounts of Kristina from Utah and Monu and Aamya have drawn sympathy, yet the conflict between development and clean air persists. But what does science say about the causal relation between air pollution and life expectancy? 

How do we know air pollution truly causes harm?

Air pollution is as old as fire itself. Early humans likely suffered from smoke in enclosed and poorly ventilated caves, though its causes were not understood. As the society evolved and experienced the process of modernisation, there emerged orgainsed political system and then there was the industrial revolution (1750-1914). But the industrial revolution also caused widespread air pollution.

For many profit-seekers, however, air pollution was viewed as a natural consequence of wealth generation and often termed as “smells like money to me”. Official recognition came much later. The US enacted the Air Pollution Control Act (1955), the Clean Air Act (1963), and the Air Quality Act (1967). The UK introduced its Clean Air Act in 1956 and established a national monitoring system by 1961.

In 1970, the US strengthened these efforts by creating the Environmental Protection Agency (EPA) and setting national air quality standards. India followed with the Air (Prevention and Control of Pollution) Act of 1981, which created pollution control boards to regulate emissions.

Communities and policymakers often ask how we know that air pollution truly causes harm – a question that science needs to answer carefully. Epidemiology provides this evidence by collecting data, designing observational studies, and applying statistical analysis to link exposures with health outcomes. Since direct experiments on humans are impossible, researchers rely on health data collected during naturally occurring pollution events, such as temperature inversions that trap pollutants in the air. 

Cross-sectional studies, for instance, offer a snapshot of disease burden at a given point in time. Outcomes of interest are typically modelled using generalised linear models: logistic regression for binary outcomes (yielding odds ratios) or linear models for continuous outcomes (yielding mean differences). However, such studies cannot determine whether exposure preceded disease.

Air pollution and increased mortality risk

The foundation for regulatory monitoring of air pollution was laid by the Harvard Six Cities Study, a long-term cohort investigation that followed participants over time. Using survival analysis, it demonstrated that chronic exposure to air pollution was strongly associated with mortality from lung cancer and cardiopulmonary diseases. 

Case-control studies, in contrast, begin with individuals who already have a disease (cases) and compare them with those who do not (controls), looking retrospectively at prior exposures. To reduce bias, cases and controls must come from comparable backgrounds. While such studies demonstrate statistical associations, they still fall short of establishing causality.

Case-crossover studies examine health outcomes during pollution events, such as temperature inversions, by comparing individuals’ exposure in a “hazard period” (just before death or hospitalisation) with their exposure during control periods when there is no pollution. This within-person design eliminates confounding from fixed traits like genetics or socio-economic status, making it a powerful quasi-experimental approach. Using US Medicare data from 2000 to 2012, Harvard researchers showed that exposure to PM2.5 and warm-season ozone was significantly associated with increased mortality risk.

Can statistical association alone prove causality?

No statistical association alone can prove causality. However, if air pollution truly harms health, its reduction should yield improvements. Such abrupt changes can be modelled as interruptions in statistical analysis. Interruptions can be interventions or policy shifts or even seemingly minor events like workers’ strike at a plant. 

The quasi-experimental interrupted time-series (ITS) approach is a powerful method for analysing such events. By controlling for background, time trends, seasonality, and weather, ITS can reveal whether mortality counts related to different diseases including influenza/pneumonia, cardiovascular and other respiratory diseases can deviate from what would have been expected in the absence of the interruption.

For example, a copper smelter strike across the Southwest US reduced sulphate aerosols and lowered mortality by 2–4 per cent, while the temporary closure of the Geneva steel plant in Utah Valley was linked to a sharp decline in hospital admissions, both captured through ITS analysis.

A related approach, regression discontinuity, exploits sharp policy-driven contrasts in exposure. Under China’s Huai River policy, cities only on north of the Huai river received free or highly subsidized coal for heating, creating a clear north-south spatial divide in pollution levels. Analysis revealed that a 42 µg/m³ increase in PM2.5 north of the river reduced life expectancy by 3.1 years, while no such effect was observed to the south.

Deterioration of body organs and systems

Advanced statistical approaches like difference-in-differences allow stable cross-sectional differences to be controlled for as if by design to strengthen causal interpretations of increased life expectancy against reduction in air pollution. Many of these techniques have emerged from different disciplines. 

Like Quasi-control groups replicate natural experiments but have control groups who are not exposed, and instrumental variables are largely derived from econometrics where certain variables are associated with treatment (here air pollution) but not with outcome (mortality). This approach further reinforces causality. Epidemiologists also use Bradford Hill criterion to establish causation. Furthermore, to establish this association, a deep dive into the inherent biological mechanism is imperative.

Frontline studies on animal models and cell lines have established that air pollution is associated with deterioration of several body organs and systems. The heart is affected through altered cardiac autonomic function, increased arrhythmia risk, coronary artery disease, and myocardial infraction. The lungs experience oxidative stress, inflammation, infection, and cancer. 

In the blood, particulate accumulation causes platelet aggregation, disrupts fibrinolysis, and accelerates thrombolysis. Vascular effects include endothelial dysfunction, vasoconstriction, hypertension, atherosclerosis progression, and plaque vulnerability. Air pollution also triggers systemic inflammation and oxidative stress, increases inflammatory cytokines, and suppresses growth factors, which impairs tissue injury repair. Finally, the nervous system is disrupted.

Air pollution as a subject of inquiry across disciplines

The picture remains incomplete: air pollution has emerged as a subject of inquiry in all disciplines, from ecology to economics where air pollution is an allocation problem in which polluters retain profits while externalising costs onto exposed populations. 

Social scientists highlight its inequitable burden: poorer communities often lack the resources for prevention or treatment, while ethnic minorities in the US and Europe, particularly native and Hispanic populations, face disproportionate exposure and limited social protection. These communities are also likely to be less informed and aware of air pollution. 

These inequalities are reflected in personal stories – Ella’s family, lacking awareness and mobility, remained in a polluted neighbourhood, whereas Kristina’s family in Utah could relocate and secure timely medical care.

Debates over air pollution often reflect conflicting positions. Polluters prefer a three-way stand – first, there is mostly no air pollution, second, if it is there, it’s not solely because of them, and third, even if it is, there are too many uncertainties to do anything about it. 

Economists and governments often hold the Kuznets curve mentality: pollution will initially rise with progress, then stabilize, and finally decline. Scientists frequently limit themselves to reporting data and outcomes in lieu of publication. 

But evidence underscores urgency of action

Yet the evidence underscores the urgency of action. Meeting the World Health Organisation’s global PM2.5 standard would add 1.8 years to average life expectancy, while the US EPA estimated that a $65-billion investment in cleaner air in 2020 yielded benefits of nearly $2 trillion. 

While lifelong smokers inhale fine particles in kilograms and lose on average a decade of life, exposure equivalent to a small pouch of fine particles in lifetime due to air pollution can still reduce life expectancy by nearly three years – demonstrating the severe health impact even at lower doses. 

It is true that we will not be dead, just as we often perceive in an apocalyptic pollution scenario. But we would do well to remember the last lines from T. S. Eliot’s The Hollow Man: “This is the way the world ends…, not with a bang but a whimper”

Post read questions

How does air pollution function as both a scientific and social problem, particularly in the context of environmental justice?

Socioeconomic status and ethnicity intersect to influence exposure to and protection from air pollution. Do you agree?

How do biological mechanisms linking air pollution to diseases of the heart, lungs, blood, vasculature, and nervous system strengthen the case for causality?

Why can statistical associations alone not establish causality in air pollution research, and how do quasi-experimental designs attempt to address this limitation?

(Case studies, statistics and details of analysis are summarised from the book, Particles of Truth: A Story of Discovery, Controversy, and the Fight for Healthy Air (2025) by C. A.Pope III and  D. W. Dockery.) (Dr. Arunangshu Das is the Principal Project Scientist at the Centre for Atmospheric Sciences, Indian Institute of Technology, Delhi.)

Share your thoughts and ideas on UPSC Special articles with ashiya.parveen@indianexpress.com.

Subscribe to our UPSC newsletter and stay updated with the news cues from the past week.

Stay updated with the latest UPSC articles by joining our Telegram channel – IndianExpress UPSC Hub, and follow us on Instagram and X.