ST. JOSEPH, Missouri - On a sunny November afternoon in this historic city, birthplace of the Pony Express and death spot of Jesse James, Lauranda Mignery watched her son Kadin, 2, dig in their front yard. As he played, she scolded him for putting his fingers in his mouth.
In explanation, she pointed to the peeling paint on her old house. Kadin, she said, has been diagnosed with lead poisoning.
He has lots of company: Within 15 blocks of his house, at least 120 small children have been poisoned since 2010, making the neighborhood among the most toxic in Missouri, Reuters found as part of an analysis of childhood lead testing results across the country. In St. Joseph, even a local pediatrician's children were poisoned.
Last year, the city of Flint, Michigan, burst into the world spotlight after its children were exposed to lead in drinking water and some were poisoned. In the year after Flint switched to corrosive river water that leached lead from old pipes, 5 percent of the children screened there had high blood lead levels.
Flint is no aberration. In fact, it doesn't even rank among the most dangerous lead hotspots in America.
In all, Reuters found nearly 3,000 areas with recently recorded lead poisoning rates at least double those in Flint during the peak of that city's contamination crisis. And more than 1,100 of these communities had a rate of elevated blood tests at least four times higher.
The poisoned places on this map stretch from Warren, Pennsylvania, a town on the Allegheny River where 36 percent of children tested had high lead levels, to a zip code on Goat Island, Texas, where a quarter of tests showed poisoning. In some pockets of Baltimore, Cleveland and Philadelphia, where lead poisoning has spanned generations, the rate of elevated tests over the last decade was 40 to 50 percent.
Like Flint, many of these localities are plagued by legacy lead: crumbling paint, plumbing, or industrial waste left behind. Unlike Flint, many have received little attention or funding to combat poisoning.
To identify these locations, Reuters examined neighborhood-level blood testing results, most of which have not been previously disclosed. The data, obtained from state health departments and the U.S. Centers for Disease Control and Prevention, tracks poisoning rates among children tested in each location.
The resulting portrait provides a granular look at places where decades-long U.S. efforts to stamp out lead poisoning have fallen short.
"The disparities you've found between different areas have stark implications," said Dr. Helen Egger, chair of Child and Adolescent Psychiatry at NYU Langone Medical Center's Child Study Center. "Where lead poisoning remains common, many children will have developmental delays and start out behind all the rest."
In children up to age 6, the CDC threshold for an elevated blood lead level is 5 micrograms per deciliter. Any child who tests high warrants a public health response, the agency says; even a slight elevation can reduce IQ and stunt development.
Nationwide, the CDC estimates that 2.5 percent of small children have elevated levels. In the communities identified by this analysis, a far higher rate of children who got tested had lead poisoning. In most cases, the local data covers a 5- or 10-year period through 2015.
Reporters visited several of the trouble spots: a neighborhood with many rundown homes in South Bend, Indiana; a rural mining town in Missouri's Lead Belt; the economically depressed North Side of Milwaukee. In each location, it was easy to find people whose lives have been impacted by lead exposure. While poverty remains a potent predictor of lead poisoning, the victims span the American spectrum - poor and rich, rural and urban, black and white.
WATER KITS: Milwaukee residents wait in line for water filtration kits under a pilot program run by the Sixteenth Street Clinic, United Way and the city.
Photo credit: REUTERS/Darren Hauck
MAPPING LEAD HAZARDS
Most U.S. states disclose data on the percentage of child blood tests that show elevated levels of lead. Yet this data, often for statewide or county-wide populations, is too broad to identify neighborhoods where children face the greatest risk.
Instead, Reuters sought testing data at the neighborhood level, in census tracts or zip code areas, submitting records requests to all 50 states.
U.S. census tracts are small county subdivisions that average about 4,000 residents apiece. Zip codes have average populations of 7,500. In each area, a relatively small number of children are screened for lead poisoning each year.
Reuters found 2,606 census tracts, and another 278 zip code areas, with a prevalence of lead poisoning at least twice Flint's rate.
The test results allow for local analysis, pinpointing neighborhoods whose lead poisoning problems may be obscured in broader surveys.
For example: Across Maryland, 2 percent of childhood lead tests were high in recent years, just a small fraction of the rate in the worst-affected Baltimore tracts. In Flint, while 5 percent of children citywide recently tested with high blood lead levels, the highest rate has been in the downtown zip code, where about 11 percent tested high from 2005 to 2015.
"I hope this data spurs questions from the public to community leaders who can make changes," said epidemiologist Robert Walker, co-chair of the CDC's Lead Content Work Group, which analyzes lead poisoning nationwide. "I would think that it would turn some heads."
The findings, Walker said, will help inform the public about risks in their own neighborhoods and allow health officials to seek lead abatement grants in the most dangerous spots.
There isn't much federal help available. Congress recently directed $170 million in aid to Flint. That's 10 times the CDC's budget for assisting states with lead poisoning this year.
Looking for lead
Since the 1970s, U.S. efforts to eradicate childhood lead poisoning have made remarkable progress. But the advances have been uneven. Legacy lead - in paint, plumbing, yards, well-water or even playgrounds - means that kids in many neighborhoods remain at a disproportionately high risk of poisoning.
Reuters conducted a nationwide analysis of childhood blood lead testing data at the neighborhood level. This map will allow you to see what the local prevalence of elevated lead tests has been in the 21 states where we obtained census tract or zip code level data. The CDC estimates that nationwide, around 2.5% of children ages 0-6 have an elevated lead level, defined as 5 micrograms/deciliter or higher. Among small children tested in Flint, Michigan during the peak of that city's lead contamination crisis, 5% had elevated levels, or double the average. In many neighborhoods - census tracts or zip code areas - across the country, a far higher rate of children have tested high in recent years.
, drop down to the map, enter city or town name or zip code.
Map credits and notes:
Graphic: Charlie Szymanski, Christine Chan, Matt Weber
Source: State agencies or CDC. Some states did not include data for census tracts and zip codes if the testing numbers were small, usually below five, to protect patient privacy. These areas appear as if no one was tested.
Data by census tract: CO, IN, MA, MD, MN, MO, NY, WI, PA, OH
Data by zip code: CA, IL, MI, NM, OK, RI, SC, TX, WV, CT, IA
Years: Data for 2012: CA, 2015: IL, NH; 2005-2011: TX; 2005-2015: IN, MA, MD, PA, NY, OH, MI, OK, RI, WV, CT; 2006-2015: NM, IA; 2010-2014: CO, WI; 2010-2015: AL, AZ, DE, GA, KY, LA, MO, NC, NJ, NV, OR, SC, TN, VT; 2011-2015: MN; 2014 DC; 2010-2012: FL, KS, WA; 2010-2011: VA.
Result types: TX, WC, CT, IA, MO, MN only included confirmed results, MA included estimated confirmed results and all other states included a mix of confirmed and unconfirmed test results.
* All the states included children tested under the age of six except Indiana, which included children under seven.
* California could only provide results for 200 zip codes with the highest number of children under age 6 years who had lead levels at or above 4.5 micrograms per deciliter and where at least 500 children were tested.
* New York data from the CDC is generally limited to the period between 2005-2007, though there are a few records from 2015.
* Iowa was only able to provide data on numbers of children who tested above 10 micrograms per deciliter.
* Massachusetts uses confirmed and unconfirmed results to estimate the number of children with elevated lead levels.
* Wisconsin's numbers represent the percentage of children who tested with blood lead levels between 5 to 9.9 micrograms per deciliter. In addition, Wisconsin offered separate figures for children with highly elevated levels above 10 micrograms per deciliter, which are not included here. In some of the most heavily impacted census tracts in the state, 10 percent or more of blood tests came back at a 10 or higher.
The nationwide map constructed through this analysis has empty spaces: The available data includes 21 states, home to around 61 percent of the U.S. population. Health departments in some states didn't possess the data or respond to records requests. Others wouldn't share it, saying they weren't required to, or citing patient privacy laws.
Even with these gaps, the data shows that despite broad national progress in curbing poisoning, lead hazards continue to imperil many communities.
Since the heavy metal was phased out from paint and gasoline in the late 1970s, children's average blood lead levels have dropped by more than 90 percent. That success story masks a sober reality in neighborhoods where risk abatement has failed.
"The national mean doesn't mean anything for a kid who lives in a place where the risks are much higher," said Dr. Egger.
TINY PLUMES OF DUST
St. Joseph, Missouri, is filled with old homes that for a century featured lead paint and plumbing. From 2010 to 2015, more than 15 percent of children tested in seven census tracts here had elevated lead levels - well beyond the Missouri average of 5 percent.
Dr. Cynthia Brownfield's family lives in a Victorian home on Museum Hill, overlooking City Hall. Built in the 1880s, it has restored tiger-stripe oak floors and an antique clawfoot bathtub.
As a pediatrician, Brownfield treats lead-poisoned children. A decade ago, her children were among them. Soon after moving into the home in 2006, her two youngest daughters tested high.
"It was dramatically shocking to get a call from the health department," said Brownfield, who had sought to make her home safe.
Inspectors found the home's old windows released tiny plumes of lead dust. There was lead in the tiles near the fireplace, the original stenciled wallpaper and the bathtub. The family did extensive work to fix the hazards.
DOCTOR'S CASE HITS HOME: Dr. Cynthia Brownfield, a pediatrician, once learned that two daughters, including Emma, 11, tested with high lead levels. She blames lead paint in their house, built in 1883 in the Museum Hill Historic District in St. Joseph, Missouri. The family fixed the hazard.
Photo credit: REUTERS/Whitney Curtis
Residents take pride in preserving old homes, said St. Joseph's community development manager, Gerald McCush. But many aren't aware children can be poisoned during renovations. Others ignore explicit warnings.
McCush, a certified lead inspector, says his office told one family that sanding paint off their walls was poisoning their son. "The dad said we were full of baloney," he said. "He wasn't going to stop working."
The Mignery family's situation helps explain why problems persist.
When Mignery moved into an old, affordable home, the mother of four was aware of dangers like street crime. She didn't know the neighborhood also had alarming rates of lead poisoning. Over the past five years, 20 percent of tested children in the census tract showed poisoning.
At son Kadin's one-year doctor visit, Mignery was told his lead levels were so high that, without quick intervention, he would need to be hospitalized. An inspector visited the home and found the culprit: old peeling lead paint.
The family could only afford a partial fix. "It wasn't easy having to repaint all the rooms downstairs," she said. "We want to do the outside here, too."
Children in at least 4 million U.S. households are exposed to high levels of lead, the CDC says.
`MESSED UP IN THE HEAD'
Lead continues to plague many industrial cities in the Midwest.
In Milwaukee last year, 11.5 percent of children tested had elevated lead. In a half dozen depressed North Side census tracts, about a third of tests showed poisoning from 2010 to 2014. One in 10 was highly elevated, at 10 micrograms per deciliter or more.
Childhood poisoning victim Brandon, now 20, lives with his mother DeeDee in an old two-story home. The family agreed to meet a reporter but asked that their last names not be published, citing the stigma attached to lead poisoning.
LEAD'S TOLL: DeeDee, whose son Brandon, now 20, was hospitalized for severe lead poisoning as a child from paint chips in his then-home in Milwaukee. Lead's impact, she said, has carried over in her son's life.
Photo credit: REUTERS/Darren Hauck
Across the street is the old rental house where, as a baby, Brandon was exposed to peeling lead paint. Health records show that before age 2, his levels reached nearly 10 times the current CDC threshold. He was hospitalized and received chelation treatment. The drugs remove heavy metals from the body and help prevent further damage, but once a child is exposed, the impact can be irreversible.
Brandon, who is easily excited, was at turns cheerful and mournful during an interview. He never finished high school and hasn't held a job. He has cognitive impairment, Attention Deficit Hyperactivity Disorder and outbursts of anger. He was recently arrested after a dispute with a convenience store clerk over soda pop, and is now on probation.
"Ever since I caught the lead, I've been messed up in the head. I can't control my anger or feelings," Brandon said. "I could have been better than I am."
A mile away, in Milwaukee's census tract 88, Isaiah Martin, 18 months old, recently ingested old paint in a family home and on the porch outside, where he loved to watch a neighbor's dog. Isaiah's initial lead test, in June, showed a level four times higher than the CDC threshold.
"As a first-time mom, I freaked out," said Isaiah's mother, Shantrice, who moved her family to an apartment in another neighborhood.
Milwaukee's health department inspected Isaiah's old home and has monitored his lead testing. The city's poisoning rates remain high, but are improving as public health officials and outreach groups combat the problem.
Federal law requires owners of homes built before 1978, when lead paint was banned, to disclose hazards to tenants or buyers. Pamphlets and warning statements, however, can't make the dwellings safe. Laws that can require owners to remediate lead from properties vary across the country.
Milwaukee frequently takes legal action against "slumlords" who rent dangerous properties, said Health Commissioner Bevan Baker. He also said childhood blood screening has more than doubled in the city since the late 1990s.
The city still has 135,000 prewar dwellings with lead paint, and 70,000 with lead water service lines. Most of its poisoning occurs in a few zip codes, where Baker says $50 million has been spent to protect children.
"We need to do more," said Baker.
On a recent evening, hundreds of people lined up in a cold drizzle at Kosciuszko Park on Milwaukee's south side, where poisoning rates are also high. The Sixteenth Street Clinic, a non-profit, was giving away tap-water filters.
Among them was Rebeca Velazquez, a mother of five. Her 5-year-old daughter was recently diagnosed with lead poisoning. "I don't want to take a risk with her," Velazquez said, motioning to her youngest child, age 4 months.
TOXIC IN BALTIMORE, CLEVELAND, PHILADELPHIA
Like Brandon, many children with lead poisoning fall into a vicious cycle: Cognitive deficits breed poor school performance, high dropout rates, few job opportunities, and brushes with the law.
Last year, one man mired in that cycle met a notorious end. At 25, Baltimore resident Freddie Gray sustained a fatal spinal cord injury in a police van, setting off months of tension in the city and fueling the national debate over policing in black communities.
In the 1990s, starting at age 2, Gray lived in a row house in Baltimore's Sandtown-Winchester area tainted with old lead paint, according to a 2008 lawsuit filed by Gray and his siblings against the property's landlord. He was exposed and suffered developmental problems, the legal filings say. The case was settled for an undisclosed amount.
Gray's former home sits in an area where children's lead exposure has persisted at shocking levels, testing data shows.
In several nearby census tracts, elevated lead levels were found in between 25 percent and 40 percent of children tested from 2005 to 2015.
Cleveland has similar problems. In the city's east side St. Clair-Superior area, nearly half of kids tested in the last decade had elevated lead. The state health department refused to provide census tract testing data; the news agency obtained the information from the CDC.
"Cleveland is my home, so it's deeply personal every time we see new numbers on lead exposure in our neighborhoods," said U.S. Senator Sherrod Brown of Ohio. Brown has pressed federal and state agencies to increase childhood blood testing rates and fund more lead abatement efforts.
Pennsylvania has a dubious distinction. The state contains the most individual census tracts - 1,100 in all - where at least 10 percent of childhood lead tests were elevated over the last decade. In 49 different tracts, from inner city Philadelphia to capital Harrisburg, at least 40 percent of children tested had high lead.
Those figures are disturbing - but not surprising - to health officials in the state.
"I believe that," said Dr. Loren Robinson, Pennsylvania's deputy secretary for health promotion and disease prevention. "Beyond the history of industry, our state has some of the oldest homes in the country."
The state's health department has partnered with schools, daycare centers and nonprofits to remove lead from properties, and is working on drafting new municipal codes to ensure rental properties are free of lead hazards, Robinson said.
`FUNDING DRIED UP'
Even in some of the highest risk areas around the country, many small children go untested for lead, Reuters reported in June. The gaps make tracking poisoned children more daunting.
In South Bend, Indiana, where health officials face a cash crunch, lead testing is in sharp decline even as existing data points to a serious problem.
In one tract there, 31 percent of small children tested from 2005 to 2015 had high levels - more than six times Flint's rate last year. The area, 1.5 miles southwest of the University of Notre Dame's idyllic college campus, is home to about 250 children. Filled with old dwellings, it has one of the highest poverty rates in town.
Dr. Luis Galup, a pathologist and the county health officer, said funds to tackle the problem in South Bend have dwindled. "We are the lowest of the low in terms of public health funding," he said.
Mayor Pete Buttigieg said the evidence that poisoning is widespread in parts of South Bend could cause public health administrators to seek more state funds or reallocate resources.
ONCE HOSPITALIZED: Edward Brown Jr., once stricken with lead poisoning so severe he was hospitalized, feeds his sister Jewel, 1, at their home in South Bend, Indiana. His mother, Victoria Marshall, said Edward's health has improved but she remains worried about lead's lasting impact.
Photo credit: REUTERS/Joshua Lott
"It's an eye-opener," he said of the Reuters study. "As a community with lots of low income residents and lots of old housing, we're vulnerable ... The county health department does everything they can just to keep up with child immunizations and restaurant inspections."
The county, with around 265,000 residents, has two nurses and one environmental inspector tasked with lead poisoning prevention. Thinly spread, they conduct home inspections only when a child's lead levels reach double the CDC's elevated threshold.
Finding those children is getting harder.
Housing and Urban Development grants that paid for South Bend lead testing ended in May. For years, the local Women, Infants and Children program, or WIC, conducted hundreds of childhood blood lead tests annually. That testing, which has stopped, relied on outside funds from HUD and others.
South Bend pediatricians and the local Head Start program still order screenings, but many children go untested.
"I bet there are hardly any tests being done now," said WIC program director Sue Taylor. "The funding dried up."
Edward Brown Jr., 2, was first tested last year. He'd been living with his mother, Victoria Marshall, in a central South Bend home. An inspector found lead paint inside and contaminated soil outside.
Marshall says Edward's blood lead reached 90 micrograms per deciliter. Levels that high can be life threatening, provoking seizures or coma. Edward's blood levels have receded since he was hospitalized for a week.
Now in a new home, Edward danced around and shared applesauce with his baby sister. He has met many of the typical developmental milestones for his age. Still, Marshall worries.
"He's got a lot of energy. Some people say he might have ADHD," she said. One-in-five cases of ADHD may be linked to lead poisoning, a recent American Academy of Pediatrics report concluded.
IRON COUNTY INDUSTRY: The Doe Run Company's Resource Recycling Facility, in Boss, Missouri, located in Iron County, an area with a high percentage of children with elevated lead levels. The company said it takes steps to avoid lead problems.
Photo credit: REUTERS/Whitney Curtis
In Missouri, it's a six hour drive southeast from St. Joseph to rural Viburnum in Iron County, situated in a mining district known as the Lead Belt. Viburnum's tract had the sharpest rate of elevated childhood tests in the state, or 30 percent since 2010.
"Really? I didn't know about it," said Viburnum Mayor Johnny Setzer.
Mark Yingling, an executive in charge of health and safety at The Doe Run Company, which operates nearby mining and smelting works, said the high rates were "news to me."
Doe Run deploys safety measures, Yingling said, such as washing all of its trucks that may come into contact with lead, and containing tailings and emissions. Lead workers shower and change outfits at the end of their shifts, to avoid tracking toxins into their homes.
Health officials haven't recently informed Doe Run of any local children with lead poisoning, he said. Iron County's health department said it isn't required to provide such information, and never has.
In fact, lead contamination in the community is pervasive. The Environmental Protection Agency has mandated so-called Superfund cleanups of two contaminated lead sites in and around Viburnum.
And Doe Run and other mining firms are currently under federal orders to clean up lead contamination from 150 properties in the region and test the soil at 250 more. The order, issued by the EPA in June, says these activities should "prioritize" properties where children have had high lead tests.
Among the spots the EPA ordered to be cleaned or tested: playgrounds, daycare yards or other places where small kids gather.
For some locals, poisoning is a cost of living in this mining town. They use a colloquial term for the syndrome - getting leaded. "You can get very leaded," said Antonin Bohac, a mechanic at the nearby Brushy Creek Mine. "But I make good money."
Bohac said he was nearly taken off the job a few years ago after he was poisoned himself - with a lead level four times higher than what federal officials consider healthy for adults.
LEAD WORKER: Antonin Bohac, a mechanic at the Brushy Creek Mine, with son Sawyer, 2, at the Bixby Country Store in Bixby, Missouri, as worker Joyce Chandler looks on. Bohac himself was diagnosed with high lead levels.
Photo credit: REUTERS/Whitney Curtis
How Reuters analyzed blood testing data
Reuters obtained data from 21 states to identify neighborhoods where testing showed the highest rates of small children with elevated levels of lead in their blood.
This data explored test results at the census tract or zip code level. Broader data surveys, such as the ones usually published by states or counties, cannot identify specific neighborhoods with a high rate of lead poisoning.
The neighborhood-level testing results helped us pinpoint areas where lead exposure is a high risk. The data has limits, however: The picture it paints is not complete. Testing rates vary significantly by city and state, and millions of children go untested. Variances in data from states can make comparisons challenging.
To identify high-risk areas, Reuters compared the rates of children who tested with high lead levels in each location to the rates in Flint, Michigan at the peak of that city's lead crisis in 2014-2015. Reuters only included census tracts and zip codes where at least 100 children were tested over a multi-year period, as small sample sizes can distort the extent of the problem.
Reuters defined elevated lead levels as any test result at or above 5 micrograms per deciliter, the Centers for Disease Control and Prevention threshold for an elevated test. The CDC lowered its threshold most recently in 2012 to reflect the medical consensus that even low levels of lead exposure can do permanent harm to children. Some states differ with the CDC in how they define high blood lead levels or lead poisoning.
At the neighborhood level, which often included only a small sampling of tested children, anomalies can cause an annual rate to spike or drop drastically.
For that reason, Reuters combined data over a broad period, up to 11 years for each census tract or zip code. Epidemiologists say this approach paints a more accurate picture.
Nationally, lead levels among children have declined steadily over recent decades. In any given community, the rate of children who tested with high lead levels over the last decade in aggregate is usually higher than the rate in the most recent testing year, reflecting the larger trend.
Though many of the troubled spots have seen some decline in lead poisoning rates over the period of analysis, they remain areas of concern with disproportionate risks.
Most states provided data covering 2005 through 2015. Some could only provide five years' worth of data or fewer. The differences are noted on our interactive map.
No child is counted more than once in the same year, but in a few states' data, a small number of children could have been counted more than once over several years.
In the data, some states counted only children whose test results were "confirmed," meaning blood was taken directly from a vein or that the child had two finger prick capillary tests over a short period that each came back high. Other states provided results for capillary tests that did not undergo a second test or finger prick.
Rates can also be influenced by different policies on who is targeted for testing. For instance, pediatricians in some areas may only test the children they view as having the highest risk of poisoning.
Off the Charts
By M.B. Pell and Joshua Schneyer
Data: M.B. Pell and Joshua Schneyer
Graphics: Christine Chan and Charles Szymanski
Design: Troy Dunkley
Video: Chris Dignam and Mike Wood
Edited by Ronnie Greene