Parents may want to think twice before letting their toddlers crawl around the house.
A Duke study used data from all 100 counties in North Carolina to analyze the effect of lead exposure on test scores. The study, led by Marie Miranda, director of the Children’s Environmental Health Initiative, found that elementary school children performed worse on tests if they were exposed to small amounts of lead as infants.
The Centers for Disease Control and Prevention currently attempts to keep blood lead levels in children under 10 micrograms per deciliter, their “action level,” according to the CDC Web site. Miranda, associate professor of environmental sciences and policy at the Nicholas School of the Environment, focused on identifying the effects of lower blood levels—2 to 3 micrograms per deciliter—on End of Grade test scores.
The current CDC blood lead action level is found in about 500,000 children nationwide, Miranda said. But the level of focus in this study affects roughly 30 percent of children. She said the study’s findings aim to reduce the negative impact of such minimal levels of exposure to lead through effecting policy changes at the local, state and national level.
“That’s a lot of kids. It’s hard to ignore that many kids,” Miranda said.
The primary source of lead exposure is through deteriorating lead-based paint in older houses, she said. As the paint starts to degrade, it sheds a fine dust that can mix with household dust. Because infants tend to crawl on the floor and teethe on toys left on the ground, they have a higher level of exposure to dust, and thus to lead.
Researchers analyzed blood lead data from the North Carolina Childhood Lead Poisoning Prevention Program and education data from the North Carolina Education Research Data Center.
“One of the things that we specialize in is linking previously unrelated data sets,” Miranda said. “So, we were able to find a child that had been tested for lead as an infant and reach forward into the education data and find their End of Grade test scores in elementary school.”
Researchers controlled for the numerous variables that could affect EOG test scores, such as poverty level and parental education, by conducting an individual analysis of the impact of each. Miranda said these variables were found to negatively affect test scores.
Statistical analysis played a large role in the study. The primary technique used was quantile regression, said Jerome Reiter, a researcher in the study and associate professor of statistical science. Unlike linear regression, which analyzes the impact of certain variables on the mean, quantile regression focuses on the impact of these variables on different areas of the distribution.
“Quantile regression allows you to focus your analysis on the tails of distributions—the low ends or the high ends—as opposed to the middle,” Reiter said. “Often, that’s where the action is.”
This form of analysis led researchers to conclude that, in addition to being exposed to higher amounts of lead, children at the low end of the EOG score distribution are more heavily affected by this lead exposure than children at the high end. This same trend was evident in the variables of poverty level and parental education attainment.
Miranda said she hopes the findings will have an impact on future lead regulations.
“I think that the CDC blood lead action level needs to be reduced to five micrograms per deciliter,” she said. “And, I think that we need to have a concerted effort to create more protective housing environments for children.”