News · Brain & Mental Health
Toddlers' cognitive scores rose as New York's air cleaned up, across three birth groups from 1998 to 2020
Columbia researchers followed 1,109 New York mother and child pairs through two decades of falling pollution. As prenatal exposure dropped across successive groups, scores at ages one, two and three went up.
- This is an observational study of 1,109 New York City mother and child pairs, so it cannot prove cause.
- Prenatal exposure to three pollutants fell across groups enrolled between 1998 and 2020.
- Toddlers' scores on a standard development test rose over the same period.
- The link was clearest in the earliest group, which had the highest exposure and the steepest drop.
- The lead author says today's pollution levels still cannot be called safe for every child.
Most environmental health research asks what harm a pollutant does. This one asks the rarer and more useful question: what happened when the pollutant went away. New York City spent two decades cleaning up its air, and Columbia University researchers had been following pregnant women and their children across that whole stretch, testing the children at ages one, two and three with the same instrument each time.
Across 1,109 mother and child pairs, the pollution measurements fell and the toddlers’ scores rose.
What was measured before the children were born
Three pollutants, in the mother, during pregnancy. Fine particulate matter, the soot-sized fraction called PM2.5. Nitrogen dioxide, which mostly comes off traffic. And polycyclic aromatic hydrocarbons, produced whenever fuel burns incompletely.
These were not estimates from a city-wide map. The Columbia Center for Children’s Environmental Health has run birth cohort studies in the same neighborhoods since the late 1990s, which is what makes a comparison across twenty years possible at all. PM2.5 and nitrogen dioxide were available in the first two cohorts, covering 1998 to 2016, while the hydrocarbons were monitored in three cohorts running to 2020.
How toddlers’ thinking was scored
With the Bayley Scales of Infant and Toddler Development, which produce a Mental Development Index. A trained examiner sits with the child and works through a set of tasks: handling objects, responding to speech, solving small problems that suit the age.
It is not an IQ test and it does not predict a life. What it does well is put children of the same age on the same scale, using the same tasks, which is exactly what a study comparing groups twenty years apart needs.
What happened as the air cleaned up
Both things moved, in opposite directions. Across the cohorts, levels of prenatal exposure to the three pollutants decreased significantly, and there was a significant upward trend in scores.
Putting the samples together, the pollutants were significantly and inversely associated with scores at 1, 2, and 3 years, meaning higher exposure went with lower scores at every age tested.
The pattern within the groups is the part that makes the finding harder to dismiss. The clearest associations showed up in the larger initial cohort enrolled between 1998 and 2006, the group that experienced the highest exposure and the greatest decrease in levels of air pollutants. In the later cohort, enrolled from 2008, where the air was already cleaner and the fall smaller, the associations were weaker.
That is what an exposure-response relationship looks like when it turns up in the real world rather than in a model. It is also the sort of internal consistency that is hard to manufacture by accident.
Why cleaner air is a policy result, not a parenting one
An individual cannot do much about the nitrogen dioxide outside their window. What changed in New York was regulation: clean air and climate policies enacted over the study period, which the authors treat as the reason exposure fell.
That makes the practical audience for this finding a city council rather than a pregnant woman, and it is a rare case where an environmental study can say something about a policy that already happened rather than one being proposed. MedlinePlus makes the point that air pollution isn’t just outside, and that the air inside buildings can also be polluted and affect your health. The neurological end of that harm has been the hardest part to pin down.
What a study of 1,109 children cannot settle
Cause, first of all. Twenty years separate the earliest enrollment from the latest, and a great deal besides air quality changed in New York across them: schooling, screens, family income, how a pediatrician talks to a parent. The authors adjusted for demographic and other factors that contribute to neurodevelopment and changed over time, but adjustment handles the things you can measure.
The sample is also modest for a cohort study, and specific. These are children from particular neighborhoods in one city, tested before their fourth birthday, and early scores are a shaky guide to later ones.
The lead author, Frederica Perera, put the limit of the optimistic reading plainly. Despite the improvements in New York, she said, “We cannot conclude that contemporary levels of air pollution are safe for the neurodevelopment of all children.”
What the New York air result adds
Studies linking prenatal pollution to worse cognitive development already exist, including in these same New York cohorts. What had been missing was the other direction: evidence that when exposure drops, the outcome moves too.
That matters because it is the claim a clean-air rule actually makes. Showing harm establishes the case for acting; showing improvement after action is the part that usually goes unmeasured, because it requires somebody to have been running the same study through the whole period. Here somebody was.
People also ask
What did the study find?
Across the cohorts, levels of prenatal exposure to PM2.5, nitrogen dioxide and polycyclic aromatic hydrocarbons decreased significantly, and there was a significant upward trend in Mental Development Index scores among 1,109 children. Combining the samples, the pollutants were significantly and inversely associated with scores at 1, 2 and 3 years.
What is the Mental Development Index?
A score from the Bayley Scales of Infant and Toddler Development, a standard assessment in which a trained examiner watches a young child handle objects, respond to language and solve simple problems. It is used widely in research and in clinics.
Which pollutants were measured?
Fine particulate matter known as PM2.5, nitrogen dioxide, which mostly comes from traffic, and polycyclic aromatic hydrocarbons, which are produced by burning fuel. The first two were available in the first two cohorts and the third was tracked across all three.
Can this show that cleaner air caused the improvement?
No. It is observational, and a great deal besides air quality changed in New York between 1998 and 2020. The design compares successive groups rather than randomly assigning exposure.
Was the effect the same across groups?
No. The associations were clearest in the larger initial cohort enrolled between 1998 and 2006, which had the highest exposure and the greatest decrease in pollutant levels. Associations were weaker in the later cohort enrolled between 2008 and 2016.
What can a parent do about this?
Individual exposure is mostly set by where you live and what traffic runs past it, which is why this is a policy finding rather than a personal one. This is general information rather than medical advice.