Evidence
The research is organized around four themes. Each describes what researchers examined, what the evidence suggests, and where it is limited or mixed, with links to the Studies library.
How to read this page
Findings describe statistical associations observed across groups of children. They do not predict the outcome for any individual child, and they do not show that lead exposure guarantees a particular result. Where studies adjusted for other factors, this page says so.
For background, the Agency for Toxic Substances and Disease Registry concluded that neurological effects, particularly cognitive deficits in children, occur among the lowest blood lead concentrations studied and that a clear exposure threshold for neurological effects has not been identified (ATSDR, 2020).
Some of the 24 articles analyze the same cohort, so agreement among them is not the same as independent replication. See Understanding the Evidence for details on study design, confounding, and association versus causation.
Themes
Theme 1
Early cognitive effects and the beginning of a developmental trajectory.
What researchers examined
Whether children with higher blood lead concentrations score differently on tests of intelligence and related cognitive skills, and whether differences appear at low exposure levels.
What the evidence suggests
Researchers reported that higher blood lead in childhood was associated with lower intelligence scores, including at concentrations below those once treated as the level of concern. Canfield et al. (2003) measured blood lead repeatedly from infancy to age five and reported that higher lifetime-average exposure was associated with lower IQ after adjustment for maternal intelligence, home environment, and other potential confounders. The estimated difference was steepest between about 1 and 10 µg/dL. Jusko et al. (2008) followed the same Rochester cohort to age six and reported that children with lifetime averages of 5 to 9.9 µg/dL scored lower on full-scale IQ than children below 5 µg/dL.
Lanphear et al. (2005) pooled data from 1,333 children in seven prospective cohorts and reported an inverse association between blood lead and IQ, with a larger estimated decrease per unit at lower concentrations. A 2019 published erratum corrected data from one contributing cohort in Lanphear et al. (2005); the corrected analysis retained the same overall inverse, nonlinear conclusion. A systematic review and meta-analysis, Galiciolli et al. (2022), included 27 studies (7 in the meta-analysis) and reported an inverse relationship across diverse countries, although effect estimates varied. The ATSDR toxicological profile (2020) likewise concluded that no clear threshold for neurological effects has been identified.
Findings extend beyond a single IQ score. Surkan et al. (2007) studied 534 children aged 6 to 10 and reported poorer neuropsychological performance, particularly verbal intelligence and vocabulary, at blood lead below 10 µg/dL. Chiodo et al. (2004) reported associations with attention, executive function, visual-motor integration, fine-motor ability, reaction time, and social behavior. The evidence is consistent with early differences marking the beginning of a longer trajectory, as language, learning, and attention become more important over time. That is an interpretation of the pattern; these studies did not test the full trajectory directly.
Limitations and conflicting evidence
Canfield et al. (2003) and Jusko et al. (2008) analyze the same Rochester cohort, so the later paper is a follow-up rather than an independent replication, and pooled analyses partly overlap with individual studies. Canfield, Kreher, et al. (2003) found that several executive-function measures were initially associated with blood lead but some relationships weakened once IQ was included; the authors described evidence for a uniquely specific executive-function deficit as relatively weak. All of these studies are observational, and effect sizes differ between populations.
Studies cited in this theme
Theme 2
Behavioral and emotional regulation across development.
What researchers examined
Whether childhood lead exposure is associated with behavior problems, attention and impulse control, and emotional symptoms, and whether associations extend into adolescence.
What the evidence suggests
Sciarillo et al. (1992) reported that greater exposure in children aged two to five was associated with higher internalizing, externalizing, and total behavior-problem scores, and the association remained after considering maternal depressive symptoms. In a nationally representative sample of 4,704 U.S. children, Braun et al. (2006) reported greater odds of an ADHD-related outcome with higher blood lead after adjustment for several demographic and environmental variables.
Nigg et al. (2008) reported that blood lead was more consistently associated with hyperactive-impulsive than inattentive symptoms, and that weaker response inhibition statistically mediated part of the association. Nigg et al. (2010), with 236 children, again reported an association with ADHD-combined presentation; hyperactivity-impulsivity findings were more consistent, while inattention-related findings varied by reporter. Liu et al. (2014) followed 1,341 Chinese preschool children and reported that higher exposure was associated with higher teacher-reported emotional and behavioral problem scores, including emotional reactivity and anxiety-related measures. Hoover et al. (2026) reported that greater childhood blood lead in a lower-exposure contemporary cohort was associated with elevated depressive symptoms at age twelve on some measures.
In a dataset of 13,580 children, MacDonald Gibson et al. (2022) reported that homes served by unregulated private wells had higher average blood lead, and that private-well reliance was associated with greater risk of juvenile delinquency, including serious delinquency after age fourteen. Because attention, impulse control, and emotional regulation affect how children engage in school, the evidence suggests cognitive and behavioral findings may interact rather than act as separate pathways.
Limitations and conflicting evidence
These studies do not show that lead exposure causes ADHD or any other behavioral disorder. Findings are heterogeneous: Liu et al. (2014) noted differences between parent and teacher reports and between outcomes, and Hoover et al. (2026) reported less consistent associations for anxiety and some other measures. Delinquency is shaped by many social and contextual factors, and the private-well finding links a household water source to blood lead and later outcomes. Not every longitudinal analysis finds robust behavioral or psychiatric associations after extensive adjustment (see Theme 4).
Studies cited in this theme
Theme 3
Translation into academic and functional outcomes.
What researchers examined
Whether early blood lead is associated with later school performance and educational attainment, which are outcomes with direct importance for children's lives.
What the evidence suggests
Chandramouli et al. (2009) measured blood lead at about 30 months among participants in the Avon Longitudinal Study of Parents and Children and, after adjustment for confounding factors, reported that greater exposure was associated with poorer reading, writing, and spelling at ages seven to eight. Higher exposure was also associated with hyperactivity and antisocial behavior. Because lead was measured years before the school outcomes, later difficulties cannot explain the earlier measurement, although confounding remains possible.
Large administrative datasets point the same way. Amato et al. (2012) analyzed 3,757 fourth-grade students whose blood lead was measured before age three and reported poorer performance in all five standardized-test domains after adjustment for poverty, race and ethnicity, English-language-learner status, attendance, disciplinary history, and other factors. Zhang et al. (2013), using Detroit public-school records, reported greater adjusted odds of scoring below proficient in mathematics, science, and reading.
Needleman et al. (1990) reexamined 132 young adults about eleven years after childhood assessment and reported that higher dentin lead was associated with greater risk of high-school dropout and reading disability, poorer class standing, and more absenteeism. These studies differ in place, period, exposure measure, and outcome, yet each links early exposure with some measure of later educational functioning. The evidence is consistent with a cumulative rather than deterministic pathway, in which early differences in cognition and behavior influence foundational skills and later opportunities.
Limitations and conflicting evidence
Educational performance depends on school quality, household resources, teacher support, chronic stress, neighborhood conditions, nutrition, and more. Statistical adjustment reduces but cannot eliminate these influences, and Zhang et al. (2013) noted that residual socioeconomic confounding remained possible. In Needleman et al. (1990), substantial loss to follow-up occurred and participants not reexamined differed from those who were, which may introduce selection bias. Historical exposure conditions also limit direct generalization to contemporary children.
Studies cited in this theme
Theme 4
Persistence into adulthood and limits of reversibility.
What researchers examined
Whether associations remain measurable after childhood and adolescence, and whether lowering blood lead after substantial exposure restores developmental outcomes.
What the evidence suggests
Cecil et al. (2008) used structural MRI in 157 young adults from the Cincinnati Lead Study and reported that greater childhood exposure was associated with reduced adult gray-matter volume in several regions, including frontal areas and the anterior cingulate cortex. Using the Dunedin birth cohort, Reuben et al. (2017) reported that higher blood lead at age 11 was associated with lower IQ and socioeconomic status at age 38, and with declines in both since childhood, after adjustment for maternal intelligence, childhood intelligence, and childhood socioeconomic status. Reuben et al. (2019) reported that, in the same cohort, higher childhood blood lead was associated with greater general psychopathology from ages 18 to 38 and with greater neuroticism and lower agreeableness and conscientiousness.
Not all adult findings held after adjustment. In the Port Pirie cohort, McFarlane et al. (2013) reported that associations with anxiety-related and substance-use measures among females were no longer significant after adjustment for childhood covariates, particularly home stimulation, and found no comparable robust associations among males. The contrast with Reuben et al. (2019) illustrates why long-term associations call for cautious interpretation: results can depend on cohort, measurement, and which confounders are measured.
On reversibility, the Treatment of Lead-Exposed Children trial randomized 780 children with blood lead of 20 to 44 µg/dL to succimer or placebo. Rogan et al. (2001) reported that succimer lowered blood lead but gave no statistically significant advantage in measured cognitive, behavioral, motor, or neuropsychological outcomes after three years, and Dietrich et al. (2004) reported no meaningful advantage at about age seven. The findings suggest that lowering a blood measure after exposure has occurred does not necessarily reverse earlier developmental differences. This does not mean reducing lead hazards is ineffective; it distinguishes preventing exposure from treating it afterward.
Limitations and conflicting evidence
Adult evidence is smaller than the childhood literature, and several adult papers come from the same few cohorts: Reuben et al. (2017, 2019) share the Dunedin cohort, and Rogan et al. (2001) and Dietrich et al. (2004) share one trial. Many classic cohorts began when average blood lead was higher than today, so generalization to current children is uncertain. Needleman et al. (1990) had substantial attrition. Findings across domains also differ in strength between cohorts and statistical models.