U.S. childhood lead exposure declines overall, but racial and economic disparities persist

The good news arrived wrapped in a caveat. Over the past decade, blood lead levels in American children have fallen—a genuine public health victory built on decades of policy work to strip lead from gasoline, paint, and consumer products. But when researchers at Harvard's School of Public Health and Project TENDR dug into state-level data, they found a story the national numbers were hiding: children of color and children from low-income families are still being poisoned at higher rates than their white and wealthier peers.

The study, published in the American Journal of Public Health, analyzed blood lead trends from two sources: the National Health and Nutrition Examination Survey, which has tracked American health since 1971, and detailed records from seven states—Connecticut, Delaware, Illinois, Indiana, Massachusetts, Pennsylvania, and South Dakota—that break down lead exposure by race, ethnicity, and income. The national picture looked reassuring. The state data told a different story. Children living in older, poorly maintained housing and those from racial and ethnic minority backgrounds carried higher lead burdens, the result of exposure patterns that had accumulated across generations. The problem was not disappearing; it was concentrating.

Lead has no safe level in a child's blood. It crosses the developing brain's barrier and damages the neural architecture that governs learning, attention, and behavior. The consequences are not abstract—they show up in school performance, in the ability to sit still, in the capacity to regulate emotion. For decades, the primary source was lead paint in aging housing stock. But the researchers documented something newer and more diffuse: lead in imported spices, in traditional cosmetics and medicines, in cookware brought from other countries. These emerging pathways to exposure, they found, were hitting the same populations hardest—families of color, families without wealth, families with fewer resources to identify and avoid contamination.

Then, in October 2025, the Trump administration eliminated NHANES. The program, which had operated for fifty-four years as the nation's primary tool for tracking health trends and toxic exposures, was shut down through layoffs. The decision erased the mechanism by which disparities could be seen, measured, and addressed. The most recent NHANES data, collected from 2021 to 2023, had not captured enough children to reveal the gaps that state-level analysis now made visible. Without the national survey, those gaps would become invisible again.

Mary Jean Brown, the study's corresponding author and a former chief of lead poisoning prevention at the Centers for Disease Control and Prevention, put it plainly: the national success story was real, but it masked a local crisis. "At the local level, too many children still face higher exposures," she said. Aimin Chen, a professor at the University of Pennsylvania Medical School and a co-author, was more direct about what the data meant: "Lead is still a serious issue for some children." The word "some" carried the weight of a demographic fact—it was not random which children those were.

The researchers called for a suite of interventions. Restore NHANES and expand it. Increase blood lead screening, particularly for children in minority communities and other high-risk groups. Require federal, state, and local governments to analyze and share data quickly so that interventions could be targeted where they were needed most. Train clinicians and public health workers to communicate clearly with families about lead risks and reduction strategies. The recommendations were not radical—they were the basic infrastructure of a public health system that could see disparities and act on them.

What remained unsaid was the political reality: the tool for seeing had been dismantled. Without NHANES, the nation would lose the capacity to track whether the disparities were widening or narrowing, whether new sources of lead were spreading, whether the children being poisoned were the same children year after year. The data would still exist in seven states. It would still be incomplete. And for the vast majority of American children, there would be no systematic way to know whether they were safe.