How do you measure a metal in a child's body? What biomarkers can and cannot tell us

When a family worries that their child may have been exposed to lead from old paint, or to arsenic from well water, the natural question is simple to ask and surprisingly hard to answer: how much is actually in the child's body?

Behind that question sits one of the central problems of environmental epidemiology. You cannot measure exposure directly the way you measure height. You measure a trace of it, captured in a tissue, at one moment in time, and then you try to reason backward to what the body has actually absorbed. Getting that step right is most of the work, and getting it wrong quietly distorts everything that follows.

This is an explainer about that step. It is educational, not a substitute for personal medical advice, and it does not recommend any test or course of action. The goal is to make the measurement itself legible.

Exposure, dose, and body burden are not the same thing

The first thing researchers separate is three ideas that everyday language blurs together.

Exposure is contact with a substance in the environment, the lead in the dust, the arsenic in the water. Dose is the amount that actually crosses into the body and reaches tissues. Body burden is the total amount stored in the body at a given moment, some of it circulating, some of it tucked away in bone or other tissue for years.

A child can have meaningful exposure and a modest internal dose, or a low recent exposure and a high body burden from the past. Because of this, the tissue a researcher chooses to sample, and the window of time it reflects, decides what the resulting number can honestly claim.

The body keeps different records in different tissues

A biomarker is a measurable signal in the body that stands in for something we cannot observe directly, in this case, exposure. Different tissues act like records kept on different timescales.

• Blood captures what is circulating right now. Blood lead is the standard measure for recent and ongoing lead exposure, but lead leaves the blood within weeks even as some of it moves into bone. Blood mercury similarly reflects recent intake. Blood is a snapshot, not a history.
• Urine reflects what the body is currently excreting, which makes it useful for metals like arsenic and cadmium. Urine also forces an important distinction called speciation: total urinary arsenic includes relatively harmless organic forms from seafood, so researchers measure the specific inorganic species and their metabolites to avoid mistaking a tuna sandwich for a contaminated well.
• Hair grows slowly and incorporates methylmercury as it forms, so segments of hair can reflect exposure over recent weeks to months. It is a longer window than blood, but it is also more vulnerable to external contamination and to differences in how hair is treated.
• Teeth are the most remarkable record of all. As a child's primary teeth mineralize, from the second trimester through the first year of life, they lay down growth layers in the dentin, including a visible line marking birth. Using laser ablation mass spectrometry, researchers can read those layers and estimate metal exposure in roughly weekly increments across the prenatal and early-infant window, long after the exposure occurred (Journal of Exposure Science and Environmental Epidemiology, 2024).
• Bone holds the longest memory. Most lead in the body is stored in bone over years, and in research settings it can be measured noninvasively with K-shell X-ray fluorescence, giving an estimate of cumulative rather than recent exposure.

No single one of these is the truth. Each answers a different question: what is in the child now, what is leaving, what built up over months, what happened before birth.

What a number actually means

Suppose a blood lead test comes back at a certain value. What is that compared against?

For lead, the United States Centers for Disease Control and Prevention uses a blood lead reference value. In 2021 it lowered that value to 3.5 micrograms per deciliter, defined as the 97.5th percentile of blood lead levels among United States children ages one to five in national survey data (CDC, 2021). It is essential to read that correctly. The reference value is a statistical flag, not a safety threshold. It marks the small percentage of children with the highest levels relative to their peers. The CDC is explicit that there is no known safe level of lead in children.

For substances taken in through food, agencies use a different yardstick. The Environmental Protection Agency's reference dose for methylmercury is 0.1 micrograms per kilogram of body weight per day, an estimate of daily intake unlikely to cause harm over a lifetime, with developmental neurotoxicity in children exposed before birth as the critical effect it is built to protect against (EPA).

Population biomonitoring puts any individual number in context. Through the National Health and Nutrition Examination Survey, the CDC measures hundreds of chemicals in blood and urine across a representative sample of the country, published in its National Report on Human Exposure to Environmental Chemicals. That report adds one caution that belongs at the center of this whole topic: the presence of a chemical in blood or urine is a measure of exposure, and "does not by itself mean that the chemical causes disease or an adverse effect" (CDC National Exposure Report).

Why the same exposure can leave different marks

Two children can drink from the same tap and end up with different internal doses. Several things drive that.

Half-life matters. A metal that clears the blood in weeks tells you about the recent past, while one that lodges in bone records years. The same external exposure looks different depending on which clock you read.

Form matters. As with arsenic, the chemical species changes both toxicity and where it shows up. A total measurement can mislead if it lumps benign and harmful forms together.

And the body itself varies. Children differ in how they absorb, process, and clear metals, including differences in genes that govern detoxification pathways. This is one reason researchers distinguish external exposure from internal dose so carefully, and it is part of what programs studying autism and the environment work to untangle when they ask whether an exposure shapes a developmental outcome.

Why measurement is the hard part of the science

This is not a side issue for a study of children's health. It is the load-bearing wall.

When exposure is measured with error, when a single blood draw is treated as a stable history, or when timing is ignored, the result is exposure misclassification. In most situations that kind of random error biases an association toward the null, meaning a real relationship between an exposure and an outcome looks weaker than it is, or disappears. A study can be large, well-funded, and carefully analyzed, and still mislead if the exposure on its left-hand side was measured poorly.

That is why so much methodological effort goes into the unglamorous parts: choosing the right tissue for the question, capturing the right time window, repeating measurements rather than trusting one, and matching the biomarker to the developmental period that actually matters. In studies of the developing brain, a tooth that records the second trimester can be worth more than a blood sample taken at age four, because it speaks to the window when the exposure could plausibly have acted.

What these numbers are for

A biomarker is a careful estimate, not a verdict. It tells you something measured, in one tissue, over one window of time, to be interpreted against a reference and alongside everything else known about the child and the environment. Read that way, these measurements are among the most useful tools in environmental health. Read as simple yes-or-no answers, they mislead.

For readers who want to see how these methods are applied in published research on environmental exposures and child development, the peer-reviewed publications are the place to start.