The neurotoxin in tap water

How lead enters drinking water through aging pipes and fixtures, why there is no safe level for children, and practical steps to reduce exposure

Lead has no safe level in the human body, and drinking water is one of the most common — and overlooked — exposure routes. The problem is not what comes out of the treatment plant. It is what leaches into your water on the way to your tap from aging pipes, lead solder in older homes, and brass fixtures. Children are especially vulnerable: their developing brains absorb 4 to 5 times more ingested lead than adults do, and even very low exposures permanently lower IQ and alter behavior [1][2]. For adults, accumulated lead exposure is now linked to cardiovascular disease at blood levels that were once considered harmless [3].

How Lead Gets Into Your Water

Water utilities treat and test at the plant, but lead typically enters at the point of use. The main culprits are:

Lead service lines. An estimated 9 to 12 million lead service lines remain in use across the United States, connecting water mains to homes and apartment buildings. These were standard installation until 1986. When water is corrosive — low pH, soft, or low in mineral content — it dissolves lead from the pipe walls. Older cities (Chicago, Detroit, Newark) are particularly affected.

Lead solder. Homes built before 1986 often have copper pipes joined with lead-tin solder. The EPA banned lead solder in new plumbing that year, but millions of older homes still contain it. Solder at joints can leach heavily, especially in newly installed plumbing or when water chemistry changes.

Brass fixtures and faucets. Until 2014, "lead-free" legally meant containing up to 8% lead. Faucets, valves, and meters made before 2014 can contribute meaningful lead loads, especially if water sits in the fixture overnight.

Water that sits. Lead leaches primarily when water is stagnant — sitting in pipes for hours while you sleep or are at work. The first water drawn from a tap in the morning is typically highest in lead.

Who Is Most Exposed

Households in cities built before 1940, renters in older apartment buildings, and families in lower-income neighborhoods bear disproportionate risk. The Flint, Michigan crisis was a sharp demonstration of what happens when corrosion control fails: after the city switched water sources without proper treatment, the percentage of children with elevated blood lead levels nearly doubled — from 2.4% to 4.9% citywide, with hotspots reaching much higher [4].

What Lead Does in the Body

Lead has no known biological role. Once absorbed, it mimics calcium and is incorporated into bones, where it accumulates over decades. From bone stores it can be rereleased during pregnancy, lactation, menopause, and osteoporosis — meaning past exposures can harm future generations.

In the brain. Lead crosses the blood-brain barrier and disrupts calcium-dependent neurotransmitter signaling. It inhibits N-methyl-D-aspartate (NMDA) receptors, which are essential for memory formation and learning. It also impairs the enzyme delta-aminolevulinic acid dehydratase (ALAD), disrupting heme synthesis and affecting the red blood cells that carry oxygen to the brain.

In the cardiovascular system. Lead increases oxidative stress in blood vessel walls, promotes arterial stiffness, disrupts the renin-angiotensin system, and displaces calcium in smooth muscle cells — all mechanisms that raise blood pressure and accelerate atherosclerosis [3].

In the kidneys. Chronic low-level lead exposure damages the proximal tubule of the kidney, contributing to chronic kidney disease and gout via impaired uric acid excretion [5].

The Dose-Response Curve

What is unusual about lead — and alarming — is that the harm per unit of exposure is actually greater at lower blood levels than at higher ones. A major pooled analysis of 1,333 children across seven countries found that IQ declined more steeply as blood lead rose from 1 to 7.5 µg/dL than it did from 7.5 to 30 µg/dL [2]. This supralinear (nonlinear) dose-response means that reducing exposure in children who already have very low blood lead still matters — there is no exposure level below which protection stops.

The CDC has lowered its blood lead reference value twice in recent years, from 10 µg/dL (in 2012) to 5 µg/dL, and then to 3.5 µg/dL in 2021. Each reduction reflects accumulating evidence of harm at lower levels. The current scientific consensus is that there is no safe blood lead level for children.

Practical Steps to Reduce Exposure

Test your water. The only way to know your tap water lead level is to test it. Many local utilities offer free testing kits, or you can use a certified lab. Test both first-draw water (after overnight sitting) and flushed water to distinguish service line lead from fixture lead.

Flush before use. Run cold water for 30 to 60 seconds — or until it feels noticeably colder — before drinking or cooking, especially after water has sat unused for several hours. This clears the stagnant water from your household plumbing, though it does not clear lead from a street-level lead service line.

Use a certified filter. Point-of-use filters certified to NSF/ANSI Standard 53 for lead removal are highly effective at the tap. Reverse osmosis and solid carbon block filters certified to this standard reliably reduce lead to well below detection limits. Make sure the filter you choose is certified specifically for lead, not just for taste and odor [5].

Use cold water for drinking and cooking. Lead leaches more readily into hot water. Never use hot tap water for infant formula, cooking, or drinking.

Consider your home's age. If your home was built before 1986, assume solder may be a source. If built before 1940, you may have a lead service line. Contact your water utility — many are now required to maintain maps of service line materials.

Cross-reference: See our water filtration page for filter selection guidance, and our heavy metal detox page for more on testing and body burden reduction.

Evidence Review

Neurodevelopmental Harm: The IQ Literature

The foundational study establishing harm below regulatory thresholds is Canfield et al. (2003), published in the New England Journal of Medicine [1]. The Rochester Longitudinal Study followed 172 children from birth, measuring blood lead at seven time points from ages 6 months to 5 years and testing IQ at ages 3 and 5. In a linear model, each 10 µg/dL increase in lifetime average blood lead was associated with a 4.6-point IQ decrease after adjustment for maternal IQ and home environment. In a nonlinear model examining the effect within the sub-10 µg/dL range, IQ declined 7.4 points as blood lead rose from just 1 to 10 µg/dL — demonstrating that the greatest harm per unit of exposure occurs at the lowest concentrations.

This supralinear dose-response was subsequently confirmed in a much larger international pooled analysis by Lanphear et al. (2005) [2], drawing on 1,333 children from seven longitudinal cohort studies in the United States, Australia, the United Kingdom, Mexico, and Yugoslavia. IQ declined approximately 2 points per doubling of blood lead concentration, and the slope was steepest below 7.5 µg/dL — well within the range of most US children at the time. No threshold was observed; effects were detectable at the lowest measured exposures. This finding — that there is no floor below which protection stops — is the scientific basis for the current CDC and WHO positions that no safe blood lead level exists for children.

Cardiovascular Mortality in Adults

The landmark study on adult cardiovascular risk is Lanphear et al. (2018) in Lancet Public Health [3]. This prospective cohort study followed 14,289 US adults aged 20 and older from NHANES III (1988–1994) for a median of 19.3 years, linking blood lead measurements to the National Death Index. The geometric mean blood lead at baseline was 2.71 µg/dL — representing an average American adult of that era. Moving from the 10th to the 90th percentile of blood lead (1.0 to 6.7 µg/dL) was associated with:

All-cause mortality hazard ratio: 1.37 (95% CI 1.17–1.60)
Cardiovascular disease mortality HR: 1.70 (95% CI 1.30–2.22)
Ischemic heart disease mortality HR: 2.08 (95% CI 1.52–2.85)

The population-attributable fraction analysis estimated that approximately 18% of all cardiovascular deaths in the US at that time — roughly 412,000 deaths per year — were attributable to lead exposure. These numbers are strikingly large for a toxin whose effects were largely considered a solved problem by the 1990s. The implication is that eliminating lead from the environment entirely would have a cardiovascular benefit comparable to large-scale reductions in smoking or hypertension treatment.

The Flint Water Crisis as a Natural Experiment

The Flint crisis provides some of the strongest real-world evidence linking water infrastructure decisions directly to children's blood lead levels. Hanna-Attisha et al. (2016) [4] analyzed blood lead data for children under 5 in Flint before (2013) and after (2015) the city switched its water source from Detroit's treated Lake Huron water to the corrosive Flint River, without implementing corrosion control treatment. The proportion of children with blood lead at or above 5 µg/dL rose from 2.4% to 4.9% citywide (statistically significant; P<0.05). In the highest water-lead-risk neighborhoods, the increase was 6.6 percentage points. A comparison group outside Flint showed no change. The spatial analysis demonstrated a dose-response relationship: areas with the highest pipe corrosion scores showed the largest increases in children's blood lead. This peer-reviewed documentation triggered a federal public health emergency declaration and became the reference case for regulatory reform of the Lead and Copper Rule.

Pathways and Mitigation Evidence

Levallois et al. (2018) [5] synthesized the mechanistic and public health literature on drinking water lead pathways. The review confirms that the primary determinants of water lead levels are: (1) presence of lead service lines, (2) lead solder in household plumbing, (3) water chemistry (corrosive, low-pH, or low-mineral-content water dissolves lead more readily), and (4) contact time (stagnant water overnight has higher lead concentration than flushed water). Population toxicokinetic modeling reviewed in the paper shows a clear predictive relationship between water lead levels and children's blood lead, even at very low water lead concentrations.

Point-of-use filtration certified to NSF/ANSI Standard 53 is reviewed as reliable for lead removal. The EPA's Lead and Copper Rule requires utilities with lead service lines to optimize corrosion control, typically through orthophosphate dosing, which deposits a protective phosphate film inside lead pipes [6]. However, the rule's action level of 15 µg/L at the 90th percentile tap is a regulatory trigger for treatment changes, not a health-based target; the WHO and many public health researchers argue it is far too permissive given the evidence that harm occurs at much lower levels.

Limitations and Caveats

Most IQ studies in this literature measure blood lead at discrete time points rather than continuously, introducing exposure misclassification that likely underestimates the true effect. The cardiovascular studies are observational and thus subject to unmeasured confounding, though the large effect sizes, biological plausibility, consistency across populations, and dose-response relationships all support causal inference. Individual variation in lead absorption (influenced by calcium and iron status, genetics of ALAD enzyme activity) means some people are more susceptible than others. Children with iron deficiency absorb lead with particular efficiency — nutritional status is a meaningful modifier of risk.

The overall strength of evidence for neurodevelopmental harm at low blood lead levels is very high; this is one of the most extensively replicated findings in environmental health research. The cardiovascular evidence is strong and growing, having shifted from probable to probable-causal in the last decade. Evidence for kidney effects and cancer risk is more limited but consistent with plausible mechanisms.

References

Intellectual impairment in children with blood lead concentrations below 10 microg per deciliterCanfield RL, Henderson CR, Cory-Slechta DA, Cox C, Jusko TA, Lanphear BP. New England Journal of Medicine, 2003. PubMed 12700371 →
Low-level environmental lead exposure and children's intellectual function: an international pooled analysisLanphear BP, Hornung R, Khoury J, Yolton K, Baghurst P, Bellinger DC, et al.. Environmental Health Perspectives, 2005. PubMed 16002379 →
Low-level lead exposure and mortality in US adults: a population-based cohort studyLanphear BP, Rauch S, Auinger P, Allen RW, Hornung RW. Lancet Public Health, 2018. PubMed 29544878 →
Elevated Blood Lead Levels in Children Associated With the Flint Drinking Water Crisis: A Spatial Analysis of Risk and Public Health ResponseHanna-Attisha M, LaChance J, Sadler RC, Champney Schnepp A. American Journal of Public Health, 2016. PubMed 26691115 →
Public Health Consequences of Lead in Drinking WaterLevallois P, Barn P, Valcke M, Gauvin D, Kosatsky T. Current Environmental Health Reports, 2018. PubMed 29556976 →
Lead and Copper RuleU.S. Environmental Protection Agency. EPA, 2021. Source →