About & Methodology

Literature-review methodology

This project used a targeted narrative literature review to examine the long-term developmental consequences associated with childhood lead exposure. Sources were identified primarily through PubMed, with ATSDR Toxicological Profiles and PubChem used for toxicological and chemical background and AACT/ClinicalTrials.gov used to identify relevant clinical-trial information. Studies were selected for their relevance to the research question and to provide evidence across cognitive, behavioral, academic, neurological, treatment, and adult developmental outcomes. The literature was then synthesized thematically rather than summarized article by article.

This is a narrative review with a targeted search. It is not a systematic review, it does not follow PRISMA reporting, and it does not report a count of records screened or a formal database search date. It was designed to synthesize representative and highly relevant evidence, not to cover every published lead study.

• 24 peer-reviewed journal articles form the main research set, listed in the Studies library.
• The ATSDR Toxicological Profile for Lead (ATSDR, 2020) is used as an authoritative toxicological background source. It is not counted among the 24 articles.
• Some publications analyze the same underlying cohort at different ages or for different outcomes, so the 24 articles should not be read as 24 fully independent cohorts.

Thematic synthesis approach

Rather than combining results statistically, the review groups articles by the developmental question they address and compares what they report. The four themes on the Evidence page are early cognitive effects as the beginning of a developmental trajectory; behavioral and emotional regulation; translation into academic and functional outcomes; and persistence into adulthood together with the limits of reversibility. Each theme notes where findings agree, where they conflict, and where evidence is weaker.

Resources consulted

• PubMed: The primary source for locating the peer-reviewed biomedical literature in the research set. Study cards link to the relevant PubMed records; this site does not reproduce PubMed abstracts.
• ATSDR Toxicological Profiles: Authoritative toxicological background and synthesis on lead, published by the Agency for Toxic Substances and Disease Registry.
• PubChem: Chemical and toxicological background information on lead. PubChem records can contain material from multiple contributing sources, each with its own attribution or reuse terms.
• AACT / ClinicalTrials.gov: Trial-registry context relevant to intervention research.

These resources serve different roles. The 24 journal articles were located primarily through PubMed; the other resources provided background and trial-registry context, and no claim is made that each article came from all four.

Source attribution and independence

ATSDR background: Agency for Toxic Substances and Disease Registry. (2020). Toxicological profile for lead. U.S. Department of Health and Human Services. The profile is available free from ATSDR. View the ATSDR profile.

PubChem: Kim, S., Chen, J., Cheng, T., et al. (2025). PubChem 2025 update. Nucleic Acids Research, 53(D1), D1516–D1525. PubChem citation guidance.

AACT acknowledgment: Aggregate Analysis of ClinicalTrials.gov (AACT) Database. Clinical Trials Transformation Initiative (CTTI). AACT. Accessed October 2026.

Independence notice: Lead Across Development is an independent educational project. It is not affiliated with, sponsored by, or endorsed by CDC, ATSDR, HHS, NIH, NLM, PubMed, PubChem, CTTI/AACT, ClinicalTrials.gov, any cited journal or publisher, or any study author. References to those resources identify sources only. Article titles and bibliographic facts are provided for citation and navigation; summaries on this site are original paraphrases and do not reproduce full abstracts, figures, tables, or article text.

Understanding the Evidence

Most long-term lead studies are observational because deliberately exposing children to lead would be unethical. Researchers instead measure exposure that occurred in daily life and follow children over time, often with repeated measurements and statistical adjustment.

Observational versus experimental evidence

In an experiment, participants are assigned to conditions by chance, which makes groups comparable. In an observational study, groups can differ in other ways. The one randomized trial in this review, the Treatment of Lead-Exposed Children trial (Rogan et al., 2001; Dietrich et al., 2004), tested a treatment given after exposure; it did not assign exposure.

Confounding

A confounder is a factor related to both exposure and outcome. Socioeconomic conditions, housing quality, nutrition, parental education and intelligence, home stimulation, neighborhood conditions, school quality, and co-occurring toxic exposures may all influence both lead exposure and development. Many of the strongest studies reduce this problem through prospective exposure measurement, repeated assessment, statistical adjustment, or longitudinal follow-up, but residual confounding cannot be completely eliminated. The phrase “after adjustment for measured confounders” means only measured factors were accounted for.

Association versus causation

An association means two things tend to occur together; causation means one produces the other. Observational findings, however consistent, cannot automatically be interpreted as proof of causation, which is why this site uses wording such as “was associated with” and “the evidence is consistent with.”

Overlapping study cohorts

The number of published articles exceeds the number of independent populations. Examples supported by the review:

• Canfield et al. (2003) and Jusko et al. (2008) are successive analyses of the Rochester cohort.
• Reuben et al. (2017) and Reuben et al. (2019) analyze different adult outcomes in the same Dunedin cohort.
• Rogan et al. (2001) and Dietrich et al. (2004) report different follow-up periods of the Treatment of Lead-Exposed Children trial.
• Cecil et al. (2008) draws on the Cincinnati Lead Study.
• Pooled analyses and meta-analyses, such as Lanphear et al. (2005) and Galiciolli et al. (2022), may overlap with individual publications.

Publications are not equivalent to independent datasets, so repeated analyses of a valuable cohort should not be mistaken for independent replication.

The knowledge-translation gap

This website does not claim to fill an experimental scientific gap. Its purpose is to address a knowledge-translation problem: evidence is spread across epidemiology, toxicology, psychology, education, neuroimaging, psychiatry, and clinical research, and each study usually addresses one part of the developmental trajectory. This makes it difficult for nonspecialists to understand the overall picture, and isolated findings are easy to misread. For example, an association with ADHD can be mistaken for proof that lead inevitably causes ADHD, and a trial showing no benefit from chelation can be mistaken for evidence that reducing exposure is unimportant.

Translating this literature into a clear, source-linked, developmentally organized resource helps students, educators, families, and the public see what current research supports and what remains uncertain, without overstating certainty.

Limitations of the literature

• Observational designs limit conclusions about cause.
• Residual confounding may remain because lead exposure is embedded in broader patterns of housing inequality, poverty, nutrition, and educational opportunity.
• Overlapping cohorts reduce the independence of the evidence, and publications are not equivalent to independent datasets.
• Historical versus modern exposure levels: many informative cohorts began when average childhood blood lead was substantially higher than today, so generalization to current children is uncertain.
• Different exposure measures and time points: studies used repeated blood lead, single measurements, tooth dentin lead, surveillance records, or environmental proxies. Blood lead mainly reflects recent exposure rather than lifetime burden.
• Fewer independent studies extend into adulthood, and several adult papers come from the same few cohorts.
• Heterogeneous behavioral findings: associations differ across measures, reporters, and cohorts.
• Selection and attrition in some longitudinal cohorts, such as the loss to follow-up reported by Needleman et al. (1990), can introduce bias.

An example of why caution is needed

Reuben et al. (2019) reported associations between childhood blood lead and adult psychopathology and personality in the Dunedin cohort. In the Port Pirie cohort, McFarlane et al. (2013) reported that several apparent adult mental-health associations among females were no longer significant after adjustment for childhood covariates, particularly home stimulation. This does not mean one study is right and the other wrong; long-term outcomes can depend on cohort characteristics, measurement strategy, confounder selection, and social context.

AI assistance and editorial responsibility

AI-assisted drafting and coding tools, including OpenAI ChatGPT and Anthropic Claude, were used during development of portions of this website. They were used as tools for drafting, organization, code generation, editing, and source-checking; the project is published under the editorial responsibility of its author.

Because AI systems can produce inaccurate or incomplete information, the scientific claims, bibliographic details, and study links used for publication were checked against source records, and readers are encouraged to consult the linked original sources when making academic or health-related judgments. The website does not present AI-generated material as a substitute for medical or other professional advice.