Canfield, RL, CR Henderson, DA Cory-Slectha, C Cox, TA Jusko and BP Lanphear. 2003. Intellectual Impairment in Children with Blood Lead Concentrations below 10 µg per Deciliter. New England Journal of Medicine 348:1517-1526.

Coverage in LA Times

This eye-opening study by Canfield et al. will force public health officials to reconsider existing thresholds for "safe" levels of lead in children. The current "level of concern" as determined by the Centers for Disease Control is 10 µg/dL.

Canfield et al. studied the relationship between levels of lead in children's blood and IQ. They found that even blood lead levels beneath 10 µg/dL were associated with lower IQs. Previous studies had not focused on children whose lifetime exposures to lead remained beneath 10 µg/dL.

Most unexpectedly, they discovered that as lead levels rise from 1 to 10 µg per deciliter, IQ drops by 7.4 points. But above 10 µg/dL, the drop for every additional 10 µg/dL was "only" 4.6 points. Thus declines per unit of lead are even greater with very low lead levels than declines at higher levels.

What did they do? Canfield et al. measured blood lead levels in a group of 172 children seven times between birth and 60 months of age. They obtained IQ scores using a standard test (Stanford-BInet Intelligence Scale) at two separate ages, 3 and 5 yrs old.

The children were all born between July 1994 and January 1995. Several screens were used to exclude children with conspicuous developmental abnormalities from the study, or because of low English proficiency of the parents, or other considerations that might confound the analysis.

Their multiple lead measurements allowed the scientists to estimate peak lead exposures experienced by the children and also chronic levels.

They also used statistical procedures to control for as series of measured variables that might confuse the interpretation, including mother's IQ, parental smoking status, iron levels, and a standard measurement of the home environment.

What did they find? Out of the 198 children who completed at least one assessment, 172 had complete data for all variables. These became the focus of the study.

As can be seen in the graph to the right, average blood lead levels were lowest at the age of 6 months, rose to a mild peak at age two years, and then declined.

Lead levels at different ages are shown in the graph to the right. The red dot shows the mean. The box around the dot runs from the 25th to the 75th percentile. The vertical lines extending above and below the boxes represent 1.5 times the interquartile range. Smaller symbols represent all individuals whose measurements fell above the vertical lines.

 

figure adapted from Canfield et al. 2003

At the ages of 3 and 5 years, most children in the study (57% and 56%, respectively) had peak blood lead concentrations beneath 10 µg/dL.

The relationship between blood lead level and IQ was very strong. Looking at the entire data set before adjusting for possible confounding variables (see above), every 1 µg/dL increase in average lifetime blood level was associated with a decrease in 0.87 IQ points.

After performing the statistical corrections for the possible confounders, Canfield et al. continued to find significant inverse relationship between lead and IQ, although the drop associated with 1 µg/dL decreased from 0.87 IQ points to 0.46 IQ points.

Canfield et al. performed a separate analysis based only upon those children with blood lead levels beneath 10 µg/dL. They found a stronger effect at within these data: even after adjusting for confounders, the decrease associated with 1 µg/dL was 1.37 IQ points.

The greater sensitivity to changes in lead levels beneath 10 µg/dL is evident in the graph to the left, which shows a curve that Canfield et al. fit statistically to their data. The slope of the curve is steeper beneath 10 µg/dL and then flattens out.

According to Canfield et al. this pattern indicates that "most of the damage experienced by children with higher lead levels actually is due largely to the initial IQ loss at blood lead concentrations of 10 µg/dL or less."

figure adapted from Canfield et al. 2003

What does it mean? These results indicate that intellectual damage from lead poisoning is occurring beneath the CDC "level of concern" of 10 µg/dL. It will be essential to revisit those standards and adjust public health recommendations accordingly.

Canfield et al. conclude by observing: "Our findings suggest that considerably more US children are adversely affected by environmental exposure to lead than previously estimated. Because there is no effective treatment for children with moderately elevated blood lead concentrations, the collective evidence argues for a shift toward primary prevention of lead exposure in contrast to the current, almost exclusive emphasis on the treatment of children with eleveated blood lead concentrations."