Lead is different from almost every other drinking-water contaminant, in two ways that matter before any number makes sense. First, there is no safe level of lead for a child - the CDC and EPA are explicit about this. Second, lead usually does not come from the source or the treatment plant. It comes from the pipe between the water main and your tap, and from the plumbing inside your own home. That means the water leaving a treatment plant can be clean while the water reaching your glass is not.
So the national question - how much lead is in America's tap water - has a national answer and a personal one, and they are not the same. We pulled every lead result in the EPA's Lead and Copper Rule dataset to answer the national one honestly. Here is what more than 260,000 readings across nearly 64,000 water systems actually show.
Of the 63,737 US water systems with lead data, about 4,800 - roughly 1 in 13, or 7.5% - have recorded a 90th-percentile lead level above the EPA's 15 parts-per-billion action level at least once. Most exceedances are modest: the median over-the-line system sits at 27 ppb. The very worst readings come almost entirely from tiny systems - mobile-home parks, small water associations, a college, a care home - not big cities. But big cities are on the list too, including Atlanta at 89 ppb. And the trend is slowly improving: the share of samples over the action level fell from 3.7% in 2016 to under 2.5% in 2024-2025.
The number people cite - 15 ppb - is not a health limit. It is an action level, and the distinction is the single most misunderstood thing about lead in water.
Under the EPA's Lead and Copper Rule, a system takes samples from high-risk homes (older houses with lead service lines or lead solder) and looks at the 90th-percentile result. If that figure is above 15 ppb, the system must take action: corrosion-control treatment, public education, and eventually lead-pipe replacement. If it is below 15 ppb, the system passes - but "passes" does not mean "safe." The CDC's position is that no blood lead level in a child is known to be without harm, so the 15 ppb trigger is a regulatory and engineering threshold, not a line below which the water is healthy.
This is why a system's 90th percentile can look fine while an individual home is high. The action level describes the system; your tap describes your risk.
Across the 63,737 systems in the data:
A note on data quality, because it matters for trust: the raw dataset contains a handful of physically impossible readings - a "90th-percentile" of 35,000 ppb, for instance, which is a unit-entry error, not a real result. We excluded the 48 readings above 1,000 ppb (a level implausible for a system-wide 90th percentile; even Flint's system figure peaked around 27 ppb) so the ranking reflects real contamination, not typos. Everything below is from the credible set.
Sort the credible data by peak lead and the top of the list is not Flint or Newark. It is small systems - the kind that serve a few dozen to a few thousand people, often with aging lead plumbing and almost no budget to fix it:
| Water system | State | Served | Peak lead | | --- | --- | --- | --- | | Cottonwood Water Association | AR | 820 | 984 ppb | | Ohman Family Living at Blossom | OH | 87 | 924 ppb | | Muncy Borough Water Company | PA | 2,900 | 890 ppb | | Pleasant Valley | MD | 139 | 796 ppb | | Elderberry Mobile Home Park | PA | 60 | 640 ppb | | Covered Bridge Apartments | ME | 25 | 624 ppb | | Paul Smith's College | NY | 1,100 | 610 ppb |
These are mobile-home parks, apartment complexes, a college, small-town associations. It is a consistent pattern in the data and a badly under-covered one: the households at highest risk of severe lead exposure are often on the smallest systems, which get the least attention and have the fewest resources for corrosion control or pipe replacement. If you are on a small rural or private-adjacent system, the odds that no one has looked closely at your lead are higher, not lower.
Severe readings cluster in small systems, but exceedances are not a small-town-only problem. Several major utilities have posted 90th-percentile results over the action level:
| Water system | State | Served | Lead | | --- | --- | --- | --- | | Atlanta | GA | 1,089,893 | 89 ppb | | Newark Water Department | NJ | 294,274 | 57 ppb | | Pittsburgh Water & Sewer Authority | PA | 520,000 | 22 ppb | | Providence | RI | 333,142 | 22 ppb | | Portland Water Bureau | OR | 666,200 | 21 ppb |
Newark's lead crisis was national news; Pittsburgh's and Portland's are well documented. Atlanta at 89 ppb - roughly six times the action level, in a system serving more than a million people - is the kind of figure that gets lost when coverage focuses on a single famous city. Lead is an old-infrastructure problem, and old infrastructure is everywhere.
Yes, slowly. Because the Lead and Copper Rule data carries sampling dates, we can watch the exceedance rate move year over year. The share of lead samples above the action level has fallen fairly steadily:
That decline tracks a decade of corrosion-control improvements, lead-service-line replacement, and - more recently - federal money: the 2021 infrastructure law put billions toward replacing lead pipes, and the EPA's Lead and Copper Rule Improvements set a goal of removing nearly all lead service lines within a decade. Progress is real. But "no safe level" means a slowly falling exceedance rate still leaves millions of people drinking water that has passed a test it was never designed to call safe.
The national trend does not tell you about your house, because lead is a last-mile problem. Here is the practical translation:
Your system can pass the 15 ppb action level and your own tap can still be high, because the lead usually comes from your service line or your home's plumbing and solder - especially in houses built before the 1986 lead-plumbing ban. The only way to know your number is to test your own water, not your system's average.
If you find lead, the fixes are well understood. A filter certified to NSF/ANSI 53 for lead reduction removes it at the tap; reverse osmosis does too. Running the tap for 30-60 seconds after it has sat clears the water that has been in contact with lead plumbing. Never use hot tap water for drinking, cooking, or formula - it leaches more lead. For the specifics, see our guide to the best lead water filters, check whether your area has federal lead-pipe replacement funding, and read the lead in drinking water pillar. Our companion analyses cover the neighboring exposures: The State of PFAS in US Drinking Water covers the other contaminant driving filter demand, and The State of Lead Paint in US Homes tracks lead's other major route into the body – the dust from pre-1978 paint.
### How many US water systems have lead over the limit? Of the 63,737 systems in the EPA Lead and Copper Rule data, about 4,800 - roughly 1 in 13, or 7.5% - have recorded a 90th-percentile lead result above the 15 ppb action level at least once. The median over-the-line system is at 27 ppb, so most exceedances are modest, but a few hundred systems have topped 100 ppb.
### Is 15 ppb of lead safe? No. 15 ppb is an "action level," not a health limit - it is the point at which a water system must start taking corrective action, not a threshold below which water is safe. The CDC states there is no known safe level of lead for children. A system that passes the action level can still deliver harmful lead to an individual home.
### What US water system has the worst lead? In the credible EPA data, the highest 90th-percentile lead readings come from small systems - the Cottonwood Water Association in Arkansas (984 ppb), a care home in Ohio (924 ppb), and Muncy Borough in Pennsylvania (890 ppb) lead the list. The worst lead is concentrated in tiny systems with old plumbing, not big cities.
### Do big cities have lead in their water? Yes. Alongside the small systems, several major utilities have exceeded the action level, including Atlanta (89 ppb, serving over a million people), Newark (57 ppb), Pittsburgh, Providence, and Portland, Oregon. Lead is an old-infrastructure problem and old infrastructure is everywhere.
### Is lead in drinking water getting better? Slowly. The share of Lead and Copper Rule samples over the 15 ppb action level fell from 3.7% in 2016 to under 2.5% in 2024-2025, tracking corrosion-control upgrades and lead-service-line replacement funded in part by the 2021 infrastructure law. Because there is no safe level of lead, the improving trend still leaves a large exposed population.
### How do I find out if my water has lead? Because lead comes mostly from your service line and home plumbing, your system's average will not tell you your risk - you have to test your own tap. Look up your water system to see its reported lead data and history at the KnowYourExposure home page, then test at the tap and filter with an NSF/ANSI 53 certified device if you find lead.
Lead in American tap water is a smaller-headline, wider-footprint problem than the famous crises suggest: a slowly improving national average hiding thousands of small systems with severe contamination and millions of homes whose real risk lives in their own pipes. You cannot see it, taste it, or fix the national pipeline - but you can find out your own number and filter it. Start by looking up your water system.
This analysis draws on the EPA's Lead and Copper Rule monitoring – the 90th-percentile lead results public water systems are required to report – combined with SDWIS compliance records, aggregated and de-duplicated in our own database. Small systems, which carry the highest and most variable lead, are analyzed separately from large ones. Full sources, definitions, and our public corrections log are on our research page. Data as of August 2026.
Cite this analysis (APA):
> Know Your Exposure Research Team. (2026). The State of Lead in US Drinking Water (2026). KnowYourExposure. https://knowyourexposure.com/blog/state-of-lead-in-us-drinking-water-2026
Journalists and researchers are welcome to cite these figures with a link back. For a custom data pull or a quote, contact [email protected].