How the EPA has left Americans exposed to lead in drinking water

May 4, 2020

The toxin has been buried for years. Across the country, lead pipes are still carrying water into millions of homes, more than 30 years after they were banned. They're tucked underground, out of sight and, for most Americans, out of mind, relics of an earlier time. But these aging conduits are still a risk for tens of millions of people. A new data analysis by APM Reports shows that those pipes may be leaching significantly more lead into Americans' tap water than government monitoring has revealed.

The Environmental Protection Agency — charged with ensuring the nation has clean air and water — has allowed utilities to use a testing method that doesn't detect the highest concentrations of lead from these water pipes, a deficiency the agency has long known about.

Scientists at the EPA have spent a decade urging the government to require more rigorous testing methods. But in its first major revision of lead-in-water regulations, made public in October, the EPA ignored years of research by its scientists.

The agency instead sided with water utilities in choosing to preserve its misleading test standards, an APM Reports investigation has found. And now the Trump administration, amid a global pandemic, is pushing to finalize the revised regulations this summer.

"This rule does absolutely nothing to address all the deficiencies we've known about for the last 10 years," said one EPA researcher, who requested anonymity. "It's an amazing house of cards that's not supported by the data. ... From a scientific perspective, we're writing a rule that is ass-backwards."

It's indisputable that there's no safe amount of lead for humans. The toxin is especially dangerous for children; even small amounts can inhibit brain development and intellectual ability. Congress banned the use of lead pipes in 1986 but allowed those already in the ground to remain. Three decades later, an estimated 15 to 22 million Americans still cook with and drink tap water entering their homes through lead pipes, known as "service lines."

Instead of replacing all the lead service lines, the government has attempted to monitor and limit lead contamination in water, principally through the EPA's Lead and Copper Rule. The nearly 30-year-old regulation lays out treatment standards that depend on regular testing. Lead is colorless and odorless when it's dissolved in water. The only way to detect it — and confirm that treatment works — is by testing water from the tap.

APM Reports spent months investigating how the EPA monitors lead levels in drinking water and the process and people behind the rule's revision. Reporters interviewed 17 current and former EPA scientists and experts, obtained internal memos, and analyzed extensive lead testing data for the first time. The investigation found:

  • The EPA's limit on lead in water — its "action level" — isn't based on what's best for human health. An internal analysis, obtained by APM Reports, estimated that the limit would likely need to be 70 percent lower to prevent lead poisoning among young children.
  • The EPA has known since 2011 that its testing method doesn't adequately measure how much lead can be released by lead pipes. Consequently, the EPA's water experts proposed changes to the Lead and Copper Rule — never publicly disclosed — that would have forced utilities to conduct more rigorous tests.
  • Instead of enacting tougher requirements, the EPA dismissed them and ceded the process to an ad hoc advisory board with heavy representation from water utilities. Ultimately, instead of revising its testing method to incorporate its scientists' research, the EPA decided to leave the testing standards largely unchanged.
  • In recent years, utilities in Chicago and the state of Michigan used more rigorous testing methods in thousands of homes with lead service lines. APM Reports analyzed the results and found lead levels two times higher on average than the results from EPA's standard procedure.

Now, with the coronavirus pandemic occupying the nation's attention, the Trump administration is rushing to finalize its revision of the Lead and Copper Rule. The EPA plans to condense a scientific review that would normally take months into just six weeks. That follows efforts by the EPA during the pandemic to roll back fuel emissions standards and limit the use of some health data in crafting new regulations.

In an email response to questions from APM Reports, EPA spokesperson Corry Schiermeyer wrote, "EPA's proposed changes better protect public health through a holistic approach that focuses on targeting the communities most impacted by lead in drinking water, strengthening treatment requirements, removing lead service lines, increasing sampling reliability, improving risk communication, and protecting children in schools and child care facilities."

While they noted some improvements, current and former agency scientists told APM Reports that the fundamentals of the EPA's new proposal are as weak as, or worse than, what is now in place.

Unless changes are made in the next few months, the Trump administration plans to enact a rule that agency scientists say could leave Americans unknowingly exposed to harmful levels of lead for years to come.


Hidden lead in drinking water

In 2011, a team of EPA researchers in Chicago made a startling discovery: Millions of Americans were likely consuming more lead in their water than anyone had known before.

The group was led by Miguel Del Toral, who had worked for EPA's Midwest office in Chicago for nearly two decades as a regulations manager for the drinking water program. He was responsible for writing rules and helping utilities meet them. "It was challenging, but it really was a rewarding kind of work," Del Toral said. "You were trying to figure out ways to protect people."

When he and his team launched their study in Chicago, the EPA's revision of the Lead and Copper rule had begun, and Del Toral had signed up to help. The agency hadn't done much research on whether its sampling method was still adequately detecting high levels of lead in homes with lead service lines. Del Toral wanted an answer.

Lead is a dangerous toxin, fatal in large doses. By the time President Richard Nixon created the EPA in 1970, it was clear that even small amounts of lead could be harmful — especially to kids, whose brains and bodies are still growing.

Chronic, low-level exposure can cause a "wearing down" of intellectual ability in kids, said Dr. Bruce Lanphear, a professor and clinician scientist at the BC Children's Hospital Research Institute in Vancouver, who studies the effects of lead on kids.

It can be subtle at first — a few lost IQ points, increased risk of attention deficit hyperactivity disorder or difficulty focusing in class. Kids who struggle in school are less likely to graduate and go to college. Prolonged exposure can have more severe effects. There's a growing body of research that associates lead exposure with lack of impulse control and criminal behavior. So far scientists haven't found a threshold of lead exposure in kids at which there's no risk. Beginning in the early 1970s, the government ordered manufacturers to begin phasing out lead in everyday products, starting with gasoline and paint.

Water that travels through a lead pipe will almost certainly contain some level of lead, though it can fluctuate daily. The only way to ensure that drinking water is completely free of lead is to remove lead pipes.

When Congress banned the installation of new pipes in 1986, lawmakers didn't do anything about the 10.2 million lead service lines that were connecting individual homes to water mains. In some cities, they'd been in the ground for more than a century. The cost of removing them would have been steep: at least a thousand dollars per household. Today, it can cost as much as $12,000.

In the late 1980s and early 1990s, as the EPA was exploring ways to control lead, the agency found reason to believe the cost of replacing pipes was justified. "It's the same thing as with abating lead paint," said Ronnie Levin, a retired EPA scientist who worked on lead issues at the time. "You have significant one-time costs and an infinite number of years of benefits."

But water utilities weren't convinced. Representatives from across the country said pipe replacements would be more trouble than they were worth. And many utilities argued it wasn't their responsibility to remove the whole pipe. Many local governments — backed up by court rulings — have deemed the section of service lines that run under private property to be the responsibility of homeowners.

In 1991, the EPA tried for a compromise in the form of the Lead and Copper Rule, the same regulations set to be revised this summer.

The rule allows utilities to replace lead service lines only as a last resort. At first, the utility can simply monitor and treat the water to prevent it from picking up lead as it travels through the pipes.

The key to this approach is testing. Utilities are required to sample water from the kitchen tap regularly in customers' homes. If the lead levels exceed 15 parts per billion in more than 10 percent of tested homes — referred to as the "action level" — the utility has to be more vigilant. To start, that means testing more often and increasing anti-corrosion measures, like chemical treatments that can build up into a protective coating inside the pipes. If the lead levels still don't come down, only then must the utility consider pipe replacements.

Scientists and policy experts told APM Reports the flaws in the rule are significant, starting with the lead action level. That limit doesn't reflect what's safe to drink. "How much lead is good for you is zero," Levin said. "That's it."

The EPA arrived at the 15 parts per billion limit because it was the lead level the agency deemed achievable for most utilities, said Jeff Cohen, who helped design the system for the EPA more than three decades ago.

In other words, human health wasn't the top priority.

Utilities also were protected by the rule's lenient testing requirements. Even in the nation's largest cities, the EPA requires utilities to only sample a minimum of 50 homes with lead pipes as little as once every three years, unless certain lead levels are exceeded. Then, tests have to be run more frequently, as often as every six months.

Lead levels reported in drinking water have dropped significantly since the Lead and Copper Rule went into effect. The EPA used to claim this was proof that the rule was working. But there's evidence to support another possibility: There are still high lead levels in drinking water, but the Lead and Copper Rule's lax testing methods simply aren't catching them.

The rule says that when gathering samples, utilities must fill a bottle with the first liter out of the tap after water has remained unused in the plumbing for six hours.

But that doesn't collect the water that's exposed to the most lead. Here's why: The first liter out of the tap is usually the water that's been sitting in the home's plumbing, not the lead service line. Depending on the house, you may need to sample the fourth, fifth, or sixth liter out of the tap to obtain a sample of water that had been sitting in the lead pipe.

How much lead? Depends how you test This is a simplified diagram of a home's plumbing. Current EPA lead regulations require that utilities test the first liter of water drawn from a faucet after water has sat in the plumbing unused for at least six hours. This is called first-liter sampling. Lead regulations have a problem, though. First-liter sampling doesn't test water that has been in prolonged contact with lead pipes.

When you turn on a faucet, the first liter of water has been sitting in the pipes closest to the tap. In this example, those are pipes made of polyethylene. Liters three and four out of this tap have been sitting lower down in the home's plumbing, in copper pipes. Determining exactly which liters were sitting in the service line can be difficult, but in this example, lead is more likely to be found between the fifth and tenth liter.

Illustration by Andrea Edstrom

Would testing water that had been in contact with the lead service line the longest show a higher level of lead?

Utilities around the country had collected data that provided some insight, but the EPA hadn't attempted to answer that question since the Lead and Copper Rule had taken effect 20 years before.

"I'd talked to our managers and said, 'We need data,'" said Del Toral. "If this is going to be it — the final revision [of the Lead and Copper Rule], which was the plan — we need to know whether our sampling is actually capturing the lead."

Del Toral and two other EPA researchers conducted three rounds of testing — in spring, summer and fall — at 32 households in Chicago, sampling up to 20 liters of water at a time.

What they found was alarming: Lead levels in the first liter, which was all the EPA asked for, were a fraction of what was found closer to the service line. In some homes, samples from inside the lead pipe contained levels more than four times higher.

The study was small, but the findings were similar to data collected by water utilities in other cities, including Madison, Wisconsin, and Washington, D.C. In their report, Del Toral and his co-authors concluded that sampling under the Lead and Copper Rule "systematically misses high lead levels and potential human exposure" from lead service lines.

The peer-reviewed research is now well known. Since it was published nearly seven years ago, the paper has been cited 57 times by researchers worldwide.

Because of that work, EPA officials have known that the Lead and Copper Rule doesn't accurately measure lead in drinking water and that some Americans may be unknowingly exposed to unhealthy levels.

"They know full well using the first-liter sample will miss or underestimate what is actually there in systems with lead service lines," Del Toral told APM Reports recently. "It just blows my mind that they would do that again."

The revision that wasn't

A year after Del Toral's team completed its research, the EPA was poised to make the Lead and Copper Rule much tougher.

By 2012, a workgroup of chemists, physical scientists, and policy experts from across the EPA had developed draft options for a new regulation and began circulating them inside the agency.

The details of these policy proposals have never been reported. But internal documents and interviews with people who worked on the draft show the agency moving toward a stricter rule than what's on the books now and stricter than what the EPA is rushing to complete in a few months.

The most significant change would have been to the testing methodology.

So Close, Yet So Costly In cities on the Great Lakes, water pipes are crumbling and poor people are paying the price. Read the story.

Aligned with Del Toral's findings in Chicago, the workgroup recommended that utilities be forced to take an additional sample beyond the first liter, aiming for water inside the lead service line, because that sample could reveal the highest risk. More utilities would then be forced to upgrade their treatments and eventually replace lead pipes running into customers' homes. As a result, there would be less lead in the system, less risk of exposure, and less risk that people would be harmed by simply drinking their tap water.

"We wanted it to be as protective as possible," said one retired EPA field officer who worked on the draft options and wants to remain anonymous. "It was really important to do the right thing with lead service lines."

Federal health officials had waited years for the EPA to implement tighter regulation of lead pipes. Researchers had found that lead at increasingly lower levels was a health risk to kids, and the EPA's 15 parts per billion level wouldn't protect human health.

"[The EPA's rule] makes no sense in terms of serious protection of public health," Christopher Portier, the former director of the CDC's environmental health division, told APM Reports. "A child drinking 15 parts per billion of lead in their drinking water on a regular basis was going to exceed our [CDC lead limit] of 5 micrograms per deciliter in the blood. So to start with, they have a problem with that 15 parts per billion."

At the very least, Portier argues, the EPA should cut its action level in half, to 7.5 parts per billion. The American Academy of Pediatrics goes further, recommending a cap of 1 part per billion for water that kids drink in schools.

An internal EPA analysis — obtained by APM Reports — estimated that to prevent a child under age 7 from tipping into what the CDC considers lead poisoning, their drinking water would have to be limited to 4.7 parts per billion each day, on average. That's less than a third of EPA's current limit.

But these more stringent proposals were stalled by top officials in the Obama administration's EPA. Leaders at the EPA's Office of Water expressed concern that some recommendations might cost too much money. At one point, the workgroup of experts traveled to Washington, D.C., from across the country to make their case. Two people in the group described the meeting with top agency officials as tense and unusually heated.

"The people [in the workgroup] thought they would stand up for a more protective rule," said another EPA field officer, "but we didn't have a crystal ball. We didn't know it would become this way."

The meetings broke off. Gradually EPA leaders stopped responding to questions from the workgroup. But by 2013, with the draft revision seemingly dead, the Obama administration had won another term in office. And that meant the EPA's Office of Water was still on the hook for producing a new Lead and Copper Rule.

What the EPA did next has confounded scientists, activists and veteran staff ever since.


Scientists get sidelined

In late 2012, the EPA appointed Peter Grevatt, an agency veteran, to lead the drinking water section of the Office of Water. Toward the end of his first year on the job, at a public meeting in December 2013, Grevatt said the EPA needed to hear from more "diverse" voices and interests about the Lead and Copper Rule.

Those voices would largely come from the water industry, which had continued advocating for less restrictive regulations, according to interviews and internal documents obtained by APM Reports.

Grevatt, who declined an interview request through a spokesperson, asked the EPA's National Drinking Water Advisory Council (NDWAC), which exists to give stakeholders a voice on water regulations, to create a new workgroup to help revise the Lead and Copper Rule. Instead of allowing the EPA's in-house experts to stay on the job, Grevatt and the Office of Water gave utilities a substantial opening to rewrite the rule.

APM Reports reviewed thousands of pages of internal documents, drafts, technical memos and emails exchanged by members of the NDWAC. These records show that the workgroup largely adopted proposals weakening the Lead and Copper Rule, advocated by members from large metropolitan water utilities. Those proposals heavily influenced the rule that the Trump administration is now trying to finalize.

Utilities of all sizes have pushed the EPA to be more forgiving about lead pipes. They worry that more stringent regulations — without more federal money — would force them to replace more pipes than they could afford.

"You're not going to get all the lead service lines removed in any city in a year," said Andrea Holthouse Putz, a water quality manager for Chicago's utility. "How do you prioritize people who we know have elevated lead?"

The water industry's perspective came to dominate the workgroup for a few reasons. Composition was a big one: An EPA contractor assembled the group, recruiting three state-level regulators, two directors from local public health departments, five public health or environmental advocates, and five employees from water utilities.

Missing were the EPA's scientists and experts. They were invited to give presentations to the group about the Lead and Copper Rule, but none were voting members.

"So we could sit there during debates, hearing utter scientific nonsense," one EPA scientist told APM Reports. "Just because you represent a water utility doesn't mean that you've got an agenda to do the right thing."

EPA employees told APM Reports they noticed a lack of understanding among some panelists about how the complex rule worked. Two of the members of the workgroup — Chris Wiant of the Caring for Colorado Foundation and Tom Neltner, now a policy director for the Environmental Defense Fund — said they'd encountered the Lead and Copper Rule in passing before they were invited to join, but didn't consider themselves experts.

One of the few clean-water activists was Yanna Lambrinidou. She became interested in water issues in the early 2000s, after moving with her family to Washington, D.C. She wasn't aware her new city was in the midst of a water crisis until her pediatrician advised Lambrinidou to get her infant daughter's blood checked for lead poisoning.

In a preview of Flint, Michigan, the D.C. water authority had altered its water treatments and set off a chain reaction that kept lead levels above the federal limit for years. That galvanized Lambrinidou. By the time the NDWAC group got started, Lambrinidou had built a name for herself as an activist.

It was apparent to her that the EPA was no longer listening to its experts on the Lead and Copper Rule. "I felt that I had entered some reality TV game," she said. "EPA had the answers. But now it's going to put on this island a group of non-experts to see what ideas they can come up with."

Lambrinidou and a handful of others pushed the workgroup to adopt policies more like the ones EPA scientists had recommended in 2012, but they were continually outflanked. In a series of meetings between 2014 and 2015, the NDWAC accepted recommendations designed by utility employees working with one of the water industry's most prominent interest groups, the American Water Works Association, records show.

APM Reports reviewed hundreds of email messages exchanged by the utility employees, who were selected to represent these groups, and the American Water Works Association regulatory affairs director, Steve Via, who coached them throughout the process.

Via helped develop what he described as a "quid pro quo exchange" plan. Under this proposal, the utility members were authorized to commit to "more aggressive [lead service line] removal" under a new Lead and Copper Rule in exchange for a far more limited in-home testing requirement. (Asked about this email exchange, Via wrote to APM Reports, "I wish I had not used the term 'quid pro quo' in that 2014 email to describe the discussion. ... Those words are politically charged, and more importantly, they aren't the best words to describe what I put forth. ... AWWA has been advocating for a revised rule that strengthens protections today while we work for a future when lead is no longer in contact with the water we drink.")

In this 2014 email, AWWA's Steve Via explains that water utilities are not willing to pay for pipe replacements on private property.

EPA scientists told APM Reports the plan was dangerously thin on science. Water samples would only be taken from people who asked for them, and there would be no requirement to test in homes with lead service lines. Instead, utilities would keep tabs on water chemistry.

When the NDWAC panel took a final vote in 2015, the more lenient recommendations passed 14-1. The only dissent was Lambrinidou.

At least one major industry group was satisfied: When the American Water Works Association reported back to its members toward the end of the process, the news was mostly good. An internal memo obtained by APM Reports expressed confidence that the EPA would run with the plan that utilities had wanted.

"We expect EPA to follow the recommendations," the memo read, "particularly if they can be characterized as a broad-based consensus agreement."

The association also pointed out how far it had come, considering what the EPA's experts had originally wanted in a new lead regulation back in 2012. "EPA wanted to continue mandatory in-home tap sampling for lead and copper, shift in-home lead monitoring to only [the] highest possible risk locations ... [and] revise the sampling protocol to draw water from inside the lead service line."

That kind of robust sampling would have placed utilities in a "difficult position," the memo stated. More utilities might have found high levels of lead and become responsible for replacing more lead pipes than ever before. It was a scenario the association was happy to have avoided.

This 2015 briefing paper lays out the pros and cons for the water industry in what NDWAC had proposed. Reducing in-home testing had been a priority for utilities from the start.

Within months, the Lead and Copper Rule revisions were overshadowed by the poisoning of an entire city.

Flint and its aftermath

The crisis in Flint started in 2014, after city officials tried to save money by transitioning away from using Detroit's water and temporarily tapping into the nearby Flint River.

The river water was naturally more corrosive and would need to be treated. But utility managers made a disastrous move. They stopped the necessary treatments, and the protective coating inside the city's lead pipes melted away, carrying toxic doses of lead to residents' faucets.

The tragedy was, in some ways, an exceptional event due to negligence by city and state officials that's been well documented.

But one detail is often overlooked. Despite the incredibly high lead levels in Flint's drinking water — in some cases as high as 13,200 parts per billion — the Lead and Copper Rule requirements failed to flag the disaster.

Astonishingly, even amid the crisis, Flint officials still managed to produce lead test results below the federal action level for months. The city needed 90 percent of its sampling sites to come back below 15 parts per billion, and when they did, officials claimed the water was safe.

But some residents didn't trust Flint and sounded the alarm that their water was poisoned. LeeAnne Walters was one: After municipal tests found 397 parts per billion at her home, Walters reached out to Miguel Del Toral, who eventually blew the whistle on Flint within the EPA.

She also was among some activists who criticized the EPA's revision of the Lead and Copper Rule.

Walters — whose four children had developed ailments ranging from hair loss and rashes to kidney trouble during the crisis — testified before Congress in 2016 that if the country was to avoid another disaster, the EPA would have to make lead regulations much stronger.

The only way to do that, she said, was to ignore the NDWAC workgroup's recommendations.

"They are there to represent utilities and protect utilities. Let's get that on the record right now," Walters said. "If what the NDWAC is suggesting is adopted by the EPA, it will — what happened in Flint will happen throughout the United States."

With Flint igniting national outrage, documents show that the EPA was struggling to turn the NDWAC revisions into a workable policy. There were huge gaps in the underlying science. The agency's enforcement officers also shredded NDWAC's proposals. The EPA let the revisions stall and decided to hand them off to the next administration.

In 2016, an EPA lawyer named Bob Kaplan became the acting administrator for the Midwestern field office charged with carrying out the emergency response in Flint. Kaplan decided to again raise serious concerns inside the agency about the inadequacies of the Lead and Copper Rule.

In 2017 Kaplan wrote a scathing memo to the head of the EPA's Office of Water. He claimed the rule had hindered the response in Flint and made it impossible to assure people their water was safe.

The Lead and Copper Rule was designed to ensure levels would stay at or below 15 parts per billion. But that level didn't necessarily mean the water was safe. "This action level has been used by many as a default measure of safety," Kaplan wrote. "However, there is no evidence to support [that] conclusion."

The rule desperately needed to be fixed, he wrote, emphasizing Del Toral's 2011 research in Chicago that showed that testing a sample from inside the lead service line resulted in much higher lead levels. He pointed out that the EPA's method of taking the first liter out of the tap was potentially misleading the public about the safety of its water.

The message was unmistakably clear: Even if we follow the Lead and Copper Rule to the letter, it won't protect us from lead.

"There is a perception ... that appropriate implementation of the existing [rule] will adequately protect the public from lead in drinking water," Kaplan wrote. "Based on lessons learned ... about the actual levels of lead in drinking water, we do not believe that will always be the case."

An excerpt from Bob Kaplan's memo to the Office of Water in December 2017, detailing the failings of the Lead and Copper Rule.

Kaplan declined an interview request from APM Reports. Though documents show Kaplan has tried repeatedly to draw the government's attention to the points in his memo, it's never been widely reported — and it didn't change the EPA's approach.

Mike Shapiro, a longtime agency official who was leading the Office of Water at the time and who received Kaplan's memo, said the document barely registered with officials when it was circulated in 2017. "Those are known concerns," Shapiro told APM Reports. "It wasn't like, 'Surprise! [The] fifth liter gives you higher numbers.' I don't think people reacted that it was a shocking result."

In other words, they'd already known.

Poison in the pipes

Del Toral's 2011 research in Chicago has since been confirmed by academic studies and data gathered by utilities around the country. They show that many Americans might be consuming far more lead in their water than the EPA-mandated testing would find.

Some of the most extensive sampling was done in the same city where Del Toral did his work.

From 2016 to 2019, the Chicago water department tested the first 10 liters drawn from the tap at 650 homes. According to an APM Reports analysis of that data, first-liter samples contained the least amount of lead. Water from deeper in the system, in some cases had two- or three-times higher lead concentrations than the federal limit.

More than half the sites in the study had at least one liter of water that tested above the federal limit. But under the Lead and Copper Rule, tests showed Chicago was in compliance, with a lead level of 9.1 parts per billion systemwide.

The lead in Chicago's water Over a four-year period, the Chicago water department tested the first 10 liters of water at 650 homes for lead. The first liter out of the tap — the sample required by the EPA's testing methods — had the lowest lead levels. And lead levels escalated with each successive liter tested. The EPA's testing method simply couldn't capture the amount of lead Chicagoans were actually consuming.

NOTE: Graph shows the 90th-percentile lead levels for each liter. SOURCE: City of Chicago

The testing data is on the water department's website, and stories by the Chicago Tribune in 2018 showed that the nation's third-largest city has a widespread lead problem.

The water department says it's working to help people whose water is contaminated. Mayor Lori Lightfoot, who has assigned a deputy to come up with a plan for removing lead service lines across the city, told WBEZ in a statement last year that it would likely be a "multi-billion-dollar program."

Michigan has uncovered similar lead problems.

After the Flint crisis, state lawmakers cracked down themselves rather than wait for the EPA to pass stricter laws. In 2019, Michigan's new lead rule took effect with testing requirements that resemble the ones Del Toral and other EPA scientists advocated for nearly a decade. Utilities now must collect an extra sample, which is more likely to include water resting inside the lead pipe. Michigan samples both the first and fifth liter out of the tap, and reports the higher of the two samples.

APM Reports is the first media outlet to analyze the statewide data. The results are remarkable.

What happened when Michigan changed its testing In 2019, Michigan began using more rigorous lead testing. That year, 134 water systems in the state began using first- and fifth-liter samples, to capture water more likely exposed to a lead service line. And the average lead level more than doubled, from less than 5 parts per billion to more than 10 ppb.

The fifth-liter samples had more lead An APM Reports analysis of more than 30,000 individual lead tests in Michigan over the last four years found that, on average, fifth-liter samples had two and a half times as much lead as first-liter samples.

And more utilities were above the EPA limit After Michigan began sampling the fifth and first liter, the number of water utilities above the EPA's action level doubled, from 16 in 2018 to 31 in 2019. This data lends credence to the argument that first-liter sampling is more likely to find lower lead levels and help utilities stay below the EPA's limit of 15 ppb.

SOURCE: Michigan Department of Environment, Great Lakes, and Energy

Journalists examined both the lead levels that Michigan utilities reported to EPA from 2010 to 2019 and the amount of lead found in samples collected between 2016 and 2019. On average, when utilities tested a fifth-liter sample, lead levels doubled or tripled.

And the number of utilities in the state that exceeded the 15 parts per billion limit nearly doubled, from 16 in 2018 to 31 in 2019.

Taken together, the data shows that water may be a significant source of lead exposure in other older American cities, as well. Utilities in Michigan and Chicago have found elevated lead levels simply because they gathered samples of water sitting inside or near lead service lines.

Similar testing in cities across the country would likely find elevated lead levels — if utilities would conduct such tests. But the EPA has never required them to do so.

It's a potential public health crisis that the EPA's current first-liter testing simply can't detect.

'Industry won'

The EPA finally unveiled its proposed revisions to the Lead and Copper Rule last October. For Miguel Del Toral, that was the day it finally sank in: All the research, the presentations, the endless rounds of policy negotiations had been wasted.

As he read, he realized just how much the agency had ignored his and other scientists' input.

"Shocked, upset, angry. It was raising my blood pressure as I went along," Del Toral said. "To just see that we've worked so hard on this, you know, banging our heads against the wall for I don't know how many years. And to have this come out?"

Del Toral and other scientists did note a few improvements. For one, the new rule would require utilities to create an inventory of local properties that still have lead service lines. Utilities that must replace lead service lines would have to remove the entire pipe, from the street up into the customer's home, instead of leaving portions behind.

But in other areas, the Trump administration's draft revision looks similar to the NDWAC workgroup proposal that favored industry.

The new rule would leave the lead action level at 15 parts per billion, a level that CDC and public health advocates say doesn't protect human health.

The draft does create a new, intermediate "trigger level" of 10 parts per billion. But this appears to be a toothless addition. If utilities find lead levels above 10 parts per billion, they're told to set up a plan with goals for replacing lead lines and follow through on it for two years. And if they don't replace the pipes, the utility is simply forced to choose from a list of public outreach activities.

The safest solution is still the same: Remove the pipes from the ground. But the new rule could slow that down, cutting the mandatory rate of replacement in half, from 7 percent of lead pipes to 3 percent annually for each utility with lead levels over the action limit.

EPA administrator Andrew Wheeler told reporters at the unveiling of the rule in October that the agency was eliminating loopholes that allowed utilities to get out of replacing pipes. He claimed that would more than make up the difference in replacement rates, but past and present EPA employees told APM Reports that the agency has provided almost no data to back up that claim.

"What we're focusing on is making sure that those communities, regardless of their ZIP code, regardless of whether or not [it] is an affluent community or struggling, underserved community," Wheeler said at the unveiling, "that those communities that have the worst lead in their pipes get those pipes replaced first." But looking at the EPA's proposal, researchers and activists have concluded the opposite.

"The most logical thing for utilities to do will be to go to richer, whiter neighborhoods where they know that the residents can pay for the private side replacement," said Karen Baehler, a public policy scholar-in-residence at American University. "It's very much built into the proposal."

Baehler recently analyzed a decade's worth of pipe replacement data in Washington, D.C. She found that black residents and people living in low-income neighborhoods were far less likely to have their lead service line replaced when the city offered to do the work, compared with people in wealthier, less diverse neighborhoods.

Baehler believes it had something to do with the cost. Many residents still had to pay more than $1,000. And an American Water Works Association survey found that most cities leave it to customers to get their side of the pipe replaced, regardless of whether they can handle the logistics or the cost.

The EPA knew that its proposal could leave low-income families at risk. The agency hired a consultant to analyze the fairness of the Lead and Copper Rule revisions before they were public. The consultant reported that children living in poor neighborhoods are more likely to have high levels of lead in their blood, and more likely to live in older homes with lead service lines. Under the EPA's proposal, they concluded, low-income families with lead service lines would be left with "disproportionately higher health risks."

EPA contractor Abt Associates warned in this 2019 policy review that relying on families to help pay for service line replacement in the new Lead and Copper Rule would lead to inequities.

Some cities have taken on this problem themselves. Washington, D.C., for instance, recently created a program to help families pay for replacement. But such programs aren't widespread.

Replacing all remaining lead service lines — which water policy experts have estimated could cost $30 to $44 billion — seems unlikely without significant investment from the federal government. Congress has shown no appetite for funding replacement projects on a grand scale, and even after Flint, few politicians on the national stage have proposed it.

There's one more avenue to get lead service lines replaced. If utilities find lead above the action level of 15 parts per billion, that would trigger more replacements.

But that requires effective testing. And the Lead and Copper Rule revisions would still require just one sample: the first liter out of the tap.

Del Toral's research, Kaplan's memo, and other data points indicate that even one additional sample from water that's been sitting in or near the service line would show higher levels of lead. But the EPA decided that was unnecessary.

"You know, it's like an ostrich putting its head in the sand," said Mae Wu, senior health director for the Natural Resources Defense Council. "[They] just don't want to know where the problems are."

Asked why the agency kept first-liter sampling in the revised rule, Schiermeyer, the EPA spokesperson, wrote that the EPA made a number of other, small changes to improve sampling — such as requiring use of wide-mouth bottles — and had specifically asked for feedback during the comment period about which liter to sample. "We are carefully considering all comments we received as part of finalizing the rule," she wrote.

Utilities didn't get everything they'd wanted in the EPA's proposed Lead and Copper Rule. But it was clear to Del Toral whose voices were listened to, and it wasn't the scientists'.

"Industry won," he said. "When you look at what we were trying to do, where we were trying to go and where we ended up in this proposal, it's pretty clear who the winners are. And it's not the people."

Rushing to the finish

After more than a decade of revising the Lead and Copper Rule in fits and starts, today the EPA is suddenly rushing to finish it.

The Trump administration quietly announced in March — as the nation was consumed with the coronavirus pandemic — that it plans to finalize the new Lead and Copper Rule this summer. That condenses the final stages of the review process, which normally would take months, into weeks.

After the draft was released in October, the EPA received nearly 80,000 public comments, some technical, related to testing standards, and others severely critical. Some of the harshest critics are former EPA employees, who were expecting a tougher regulation. "This rule should be as aggressive as taking lead out of gasoline and lead out of paint," Betsy Southerland, who resigned as a science director in the EPA's Office of Water in 2017, told APM Reports. "When will they revise this again? Another 20 years? I'll be dead."

The EPA's Science Advisory Board had expressed interest in reviewing the research and modeling that underpin the new rule. That review would normally take months. But the board is giving itself just six weeks, to match the EPA's timeline. If the board finds the scientific basis for the rule is unsound — and even if it doesn't — advocacy groups could sue to block the rule.

The science board plans to come up with its initial findings by May 11. "Under the best circumstances, it would be a quick review," said Michael Honeycutt, the board's chairman. "Now, it's a herculean task."

The global pandemic has upended the economy, but not the EPA. The agency has been pushing ahead with major regulatory actions amid the crisis, like repealing the Obama administration's new mileage standards for cars and putting new constraints on the health research the EPA can use to write rules going forward.

"It's the best time to do it," Del Toral said. "Nobody's going to pay attention. There's absolutely nothing else on the news."

Del Toral retired from the EPA this winter. He lives in Chicago's Pilsen neighborhood, close to where he grew up. Del Toral still follows the final stages of the rule he worked on for so many years. "The Lead and Copper Rule revisions have been the most frustrating time in my career," Del Toral said. "Honestly, even more so than Flint. Because it impacts people from now on."

The public health experts who spoke with APM Reports for this story said we might not know the full impact of the revised Lead and Copper Rule for years.

Children who've been exposed to lead often show few symptoms in the short term, unless they've been severely poisoned. The consequences emerge over time, across an entire community. The demand for special education programs in Flint schools is already rising. Without interventions in schools, research suggests the city will likely see fewer kids enrolled in gifted programs, and graduation rates may fall.

"All these kids are our kids," said Dr. Mona Hanna-Attisha, a pediatrician in Flint. "And I think that's the lens we need to use. If they were our kids, would we be OK with these lead levels in our water? And would we be OK with not even informing people that this is a risk?"

Lauren Rosenthal   @laurenthal
Will Craft   @craftworksxyz

Support for this project came from The Water Main, an initiative from American Public Media focused on building public will for clean, affordable, accessible water. Image at top by Paul Sancya | AP.

About the reporting
APM Reports analyzed thousands of individual lead tests and hundreds of system-wide lead levels to investigate the consequences of the EPA relying on first-liter sampling in its lead regulations. We gathered state-wide sampling data in Michigan, where a change in state law required utilities to test both the first and fifth liter. We also analyzed sampling data in Chicago, where the city tested each of the first 10 liters from 650 homes over a four-year period. APM Reports found that lead levels are sometimes two to three times higher when water from within the lead service line is tested. Data for Michigan and Chicago, as well as our analysis for both, can be found on Github.

Don't miss our next investigation

Enter your email below to receive notifications of new stories.