Is Lead Poisoning a Missing Link in the Fight Against Malnutrition?

Lead exposure is increasingly recognized as a major global health problem, particularly in low- and middle-income countries (LMICs). Much of the research and policy attention to date has focused on its effects on cognition, schooling, and productivity. But much less is known about how lead interacts with another core development priority: child nutrition. Does lead exposure increase the odds of malnutrition or are malnourished children disproportionately vulnerable to the negative impacts of lead exposure? While the biological pathways are relatively better documented, the empirical evidence on the magnitude of this relationship remains surprisingly scant.

But this link matters. Stunting is among the most widely used indicators of child well-being, shaping priorities and financing in global health and development. If lead exposure meaningfully worsens nutritional outcomes—or if malnutrition significantly increases the detriments caused by lead exposure—then environmental policy and nutrition policy are more tightly linked than is currently understood.

Why might lead exposure and stunting be correlated?

Research points to plausible biological pathways through which lead exposure might impair growth and through which malnutrition may increase both lead absorption and its physiological damage.

For starters, lead competes with calcium for absorption in the body, disrupting bone formation and skeletal growth. It also interferes with growth hormones and can impair appetite and nutrient uptake. When dietary calcium is low, the body absorbs more lead and retains it longer. Dietary calcium sources are particularly expensive and often under-represented in diets across LMICs. Similarly, iron-deficient children may absorb substantially more lead through shared transport pathways in the intestine. Iron deficiency anemia is common in undernourished populations.

Because of these mechanisms, two children exposed to the same environmental lead burden can experience vastly different internal doses and harms depending on their nutritional status. And in environments where children already face marginal diets and repeated infections, even small physiological disruptions can lead to measurable deficits in height-for-age (HAZ) or weight-for-age (WAZ). These disruptions may also increase lead absorption and later-life detriments to learning and cognitive development.

Despite these mechanisms, much of the existing human evidence has focused on outcomes adjacent to nutrition rather than nutrition itself. Several quasi-experimental studies estimate the causal effects of lead exposure on birthweight (which is a predictor of later stunting) in the United States, Mexico, and Bangladesh. Other studies also document impacts on infant health or cognitive development. But very little research has directly examined standard anthropometric measures such as stunting or wasting.

Animal studies provide clear experimental evidence. Controlled experiments show that lead exposure reduces body weight in mice and pigs relative to untreated controls. These findings reinforce the plausibility of an effect on growth, but translating results from laboratory settings to children living in complex, resource-constrained environments is far from straightforward.

What can we learn from observational data?

In the absence of causal evidence, large observational datasets provide a useful starting point. But only a handful of nationally or regionally representative surveys include both biomarker data on lead exposure and standard measures of child nutrition. Three stand out.

1. The 1999 India Demographic and Health Survey, which includes blood lead measurements for children in Delhi and Mumbai

2. The 2018 Georgia Multiple Indicator Cluster Survey (MICS)

3. Mexico’s 2023 ENSANUT survey

All three measure nutritional status using internationally comparable indicators such as HAZ and WAZ. The left panel of Figure 1 below plots the correlation between lead exposure and HAZ in these data.

Plotting simple correlations between blood lead levels and nutritional outcomes across these settings yields a striking pattern. First, there is an overall log-linear relationship. The shape of this relationship is strikingly similar to that between blood lead and IQ (right panel of Figure 1). This makes sense if we think there are similar underlying biological mechanisms driving the relationship between lead exposure and stunting as there are between lead exposure and IQ.

Figure 1. Height-for-age has a similar relationship to blood lead as IQ does

Image

Image

Note: Data on height-for-age z-score (HAZ) and blood lead are from the Georgia 2018 Multiple Indicator Cluster Survey (MICS), India 1999 Demographic and Health Survey (covering only Delhi and Mumbai), and Mexico 2023 ENSANUT survey. Stunting is defined as being two standard deviations below the mean height-for-age z-score (HAZ). The figure on the right is reproduced from Canfield et al. (2003).

However, looking at the individual countries—presented in Figure 2 below—we also see some contextual differences. In India, higher blood lead levels are strongly correlated with worse nutritional status for both girls and boys. In Georgia there is an effect only for girls, and in Mexico we see no correlation. There are multiple possible explanations here. One obvious difference across these contexts is exposure intensity. Average blood lead levels in the Indian sample are roughly twice as high as those observed in Georgia or Mexico. This suggests that the correlation between lead and nutritional status may be nonlinear, becoming visible only once exposure exceeds a certain threshold. Alternatively, differences in diet quality, health systems, or co-exposures could mediate the relationship. Girls in Georgia may face systematically poorer diets or healthcare access than boys, making nutritional status more sensitive to lead exposure at the margin.

Clearly, these correlations are not causal and may well reflect confounding by poverty, housing quality, or environmental conditions that jointly determine lead exposure and nutrition, as well as reverse causality. Still, the fact that strong associations appear in at least one high-exposure context underscores the need for more systematic evidence.

Figure 2. Children’s blood lead correlates with stunting in some but not all settings

Image

Note: Data on height-for-age z-score (HAZ) and blood lead are from the Georgia 2018 Multiple Indicator Cluster Survey (MICS), India 1999 Demographic and Health Survey (covering only Delhi and Mumbai), and Mexico 2023 ENSANUT survey. Stunting is defined as being two standard deviations below the mean height-for-age z-score (HAZ). The figure on the right is reproduced from Canfield et al. (2003).

Longitudinal studies show early lead exposure predicts later growth

In the absence of experimental or quasi-experimental studies, we can also look at cohort or longitudinal studies, in which children’s blood lead is measured at one point in time, and the same children are then tracked over time and their height-for-age measured at a later date. There are two such studies from low- and middle-income countries.

A study in Mexico by Afeiche et al. (2012) tested blood lead in children aged one- and two-years old and correlated this with height at age four. Another study in Benin by Ahmadi et al. (2022) tested blood lead at age one and then height at ages four to six. Afeiche et al. compare two groups, above and below median blood lead (that have a mean difference of around 4 μg/dL), adjusting for maternal height, education level, age, and child birth length. They find a difference in height aged four of around -0.8cm (or around -0.2 standard deviations in height-for-age). Ahmadi et al. compare the top and bottom quintiles in blood lead (a mean difference of around 8 μg/dL), and see a difference in height-for-age of around -0.11 standard deviations at age four, and -0.13 at age six, after adjusting for maternal education and wealth, and child sex, birth weight, and iron deficiency at age one.

How big are these effects? The average child in most low- and middle-income countries has a blood lead level of around 5 μg/dL, so these are very common levels of exposure. The threshold for stunting is two standard deviations below the mean, so these effects are not large enough to cause stunting alone, but might contribute 5 to 10 percent.

What this means for policy

The evidence to date is suggestive but incomplete. Observational data hint at strong relationships in high-exposure settings. Biological mechanisms are well established. What is needed now is synthesis and causality: a meta-analysis to establish stylized facts, and targeted quasi-experimental work to pin down causal effects.

Malnutrition remains one of the most persistent development challenges. At the same time, lead exposure is increasingly recognized as a widespread but solvable environmental hazard. If lead meaningfully contributes to child malnutrition, then investments in environmental remediation could deliver nutrition gains.

Conversely, nutrition programs that ignore environmental toxins may be less effective than expected. Iron and calcium supplementation, for example, may interact with lead exposure in ways that blunt or amplify their effects. Understanding these interactions is essential for designing integrated interventions.

Bringing lead into the nutrition conversation could open new avenues for policy coordination and funding. Just as the education community has begun to grapple with the hidden costs of lead exposure, the nutrition community may soon need to do the same.