London's clean air zone restored children's lung growth, and four years later the deficit was gone

Children’s lungs are supposed to get bigger every year. That growth is measurable, it follows a predictable curve, and the size a pair of lungs reaches in adolescence is roughly the size they will be for life. Damage done to the curve does not get made up later.

Which is why the London finding is unusual. Wood and colleagues, writing in The Lancet Public Health, recruited 3,414 children aged 6 to 9 from 84 primary schools before the Ultra Low Emission Zone existed, split between central London and Luton, a comparison town with no clean air zone. At the start, baseline adjusted FEV1 was lower in London than Luton, by 38 milliliters.

Then the charge came in, the traffic changed, and the London children’s lungs grew faster. After 4 years, mean FEV1 reached parity across sites.

What a natural experiment is buying you

Nobody can randomize a child to breathe cleaner air. Clean air zones are public health policy interventions used to reduce traffic-related air pollution in urban areas, but evidence of their health benefits is limited, largely for that reason: the studies that would settle it cannot be run.

A natural experiment is the workaround. The policy does the assigning, and the researchers supply the control group. Luton is a similar English town of similar size and diversity with no clean air zone, so it stands in for what London would plausibly have done otherwise. The children were similar across the two sites in terms of age, height, and weight.

That design is stronger than the before-and-after studies this field usually produces, and weaker than a trial. It cannot rule out everything that differs between two towns over five years, and the study ran through a pandemic that emptied roads everywhere.

The number, and what it is worth

Over the follow-up period, children’s FEV1 growth increased by 10 mL/year more in London than in Luton, 233 against 223.

Ten milliliters a year sounds like nothing. Compounded across four years in a child whose total lung volume is around two liters, it is the difference between staying behind and catching up, and that is what happened: 2,283 mL in London against 2,282 mL in Luton at the end, from a 38 mL deficit at the start.

The proportion of children with clinically impaired lung function fell from 14% to 9% in London, and from 9% to 7% in Luton. Both improved. London improved more, from a worse starting point.

Why the pollution measurement matters more than the policy date

The weakest version of this study would compare London before and after a date. This one models what each child actually breathed, at their own address.

Annual residential exposures were estimated at each child’s home at 20 square meter resolution using a validated dispersion model, for nitrogen dioxide and for fine particulates. At baseline, London children’s modeled annual exposure to NO2, the pollutant most reflective of exhaust emissions, ran nearly 19 micrograms per cubic meter higher than Luton’s.

Then it fell, and it fell faster in London: decreases of 3.77 micrograms per cubic meter per year against Luton’s 1.77. Exposure dropped faster in the place where lung growth rose faster. That alignment is what turns a coincidence of timing into an argument about cause.

What it does not show

Air pollution is a mixture of solid particles and gases in the air, from car emissions, chemicals from factories, dust, pollen and mold spores, and a low emission zone touches only the first of those.

FEV1 is also one measure. It captures how much air a child can force out in a second, which is the standard index of lung function and not the whole of respiratory health. Asthma symptoms, infections and long-term disease risk are different outcomes that this analysis does not report.

And two towns are two towns. The researchers matched what they could and modeled exposure carefully, but a natural experiment inherits every difference between its arms that nobody thought to measure.

What a parent can and cannot do about it

Almost nothing in this study is a personal choice, and that is the point of it.

The children did not change their behavior. Their parents did not move house. A policy changed which vehicles could drive into central London, the air at their home addresses got cleaner, and their lungs grew faster. People with heart or lung disease, older adults and children are at greater risk from air pollution, and this is one of the few pieces of evidence showing what happens when that risk is reduced at the level it is actually created.

For a family, the practical reading is narrow: this supports living and schooling away from the heaviest traffic where that is possible, which for most people it is not. The wider reading is the one the authors draw, that the evidence supports wider implementation of clean air zones as a public health intervention.