From 2001 to 2004, Washington, D.C., experienced what may have been the worst lead contamination of city water on record. Tens of thousands of homes had sky-high levels of lead at the tap, and in the worst cases, tap water contained enough lead to be classified as hazardous waste. Not that the Centers for Disease Control and Prevention, the government oversight agency for public health, was worried.
A 2004 CDC report found that water contamination "might have contributed a small increase in blood lead levels." The study has been influential. School officials in New York and Seattle have used the CDC report as justification for not aggressively responding to high levels of lead in their water, and other cities have cited the report to dispel concerns about lead in tap water.
But the results of thousands of blood tests that measured lead contamination in children were missing from the report, potentially skewing the findings and undermining public health. Further, the CDC discovered in 2007 that many young children living in D.C. homes with lead pipes were poisoned by drinking water and suffered ill effects. Parents wondered whether the water could have caused speech and balance problems, difficulty with learning, and hyperactivity. Yet the health agency did not publicize the new findings or alert public health authorities in D.C. or other federal agencies that regulate lead, such as the U.S. Environmental Protection Agency or Housing and Urban Development.
"This is a disaster of accountability from CDC's point of view," says John Rosen, a pediatrician and national expert on lead poisoning at Montefiore Medical Center in New York City. "This raises troubling questions about CDC's complicity in passing on dubious data -- and further questions about why CDC did not publicize the 2007 results more broadly."
CDC scientists and press representatives did not respond to requests for an explanation about why the results were not widely publicized. George Hawkins, director of the District Department of the Environment, in Washington, says he became aware of the 2007 study only on April 2 this year, when Salon showed him an abstract of the study. Scientists from other agencies, including EPA and HUD, also say they were never told about the results. "CDC never told us," says an EPA scientist, "and they never asked our help or any other water expert's help when they did their studies. That's a shame and a waste, because when it comes to lead in water, you need engineers, chemists and health people to figure it out." The scientists requested anonymity because they were not authorized to speak to the press.
Salon raised questions in 2006 about the influential 2004 CDC report of lead contamination in the D.C. area. New scrutiny of CDC's work has been sparked by a scientific study published in January that contradicts CDC's conclusion of minimal harm. Environmental engineer Marc Edwards of Virginia Tech, and pediatrician Dana Best of Children's National Medical Center in Washington, used Best's data for children's blood-lead levels and found a jump in high-level results among kids who were infants and toddlers from 2001 to 2004. The authors conclude that hundreds, possibly thousands, of children were adversely affected.
Edwards and Best raised further health concerns about the 40,000 Washington children who were either in the womb or using formula during the crisis, for whom health effects are expected to be the most severe. These children (now 4 to 9 years old) are at particularly high risk for future health and behavioral problems linked to the lead exposure.
In February, a D.C. resident filed a $200 million lawsuit against the D.C. water company, claiming that lead-contaminated tap water poisoned his twin sons as infants, causing them to have ongoing learning and behavioral problems. The D.C. Inspector General is investigating the reasons behind the apparently conflicting results of the two CDC reports.
Mary Jean Brown, head of the CDC Lead Poisoning Branch and principal author of the 2004 study, acknowledges that thousands of blood tests were lost and not included in the study. But she defends the paper's conclusion -- that children's exposure to lead and consequent harm was slight -- on the grounds that only low-level test results were lost.
In a written memo to Salon, explaining the missing data in the 2004 report, Brown writes that the issue became apparent in 2004. The D.C. Health Department had reported testing the blood of 15,755 children in 2002, 18,038 in 2004, but only 9,765 children in 2003. When questioned, D.C. Health Department staff attributed the gap, according to Brown, to a commercial laboratory that failed to submit some test results in the last quarter of 2003, a year when lead levels in the water were high. Specifically, the lab was said to have omitted results below 10 micrograms per deciliter. (This is CDC's current level of concern and a level that many pediatricians and public health experts classify as lead poisoning.)
Brown accepted this explanation. She states that the highest results in 2003 were reported accurately and that a comparison with blood-lead trends in the city suggests that the loss of low-level results did not introduce a bias in CDC's analysis.
"This is just a circular argument, and it doesn't wash," says a government scientist who requested anonymity. "When CDC learned the data was missing, someone could have called the lab and asked for it. If it was the lab's mistake, they would have sent the data," the scientist adds.
Edwards, who examined the data used by CDC for the 2004 report, says that numerous high results were also omitted. Since 2004, he has made a number of attempts to question CDC scientists about apparent problems with the 2004 study. In 2007, Edwards filed a complaint of scientific misconduct with the CDC, alleging that Brown must have known about serious flaws with the data but failed to acknowledge them when writing the 2004 CDC report.
In a message to James Stephens, CDC's associate director of science, he wrote, "Why is it that every child I have personal knowledge of, who had a strong chance of having elevated blood lead from water, is either deleted or otherwise misrepresented in the data that CDC has and used for this publication?" Edwards did not receive an answer. In March 2008, Stephens wrote to Edwards, informing him, "We have examined CDC's role in the study and have found no evidence of misconduct."
CDC scientist Jaime Raymond presented the 2007 results at the American Public Health Association's 2007 annual meeting in Washington. The study used data from a total of 22,981 children under 6 years old who lived in Washington from 1998 to 2006. Twenty-nine percent of the children lived in houses with lead water pipes and were more likely to have contaminated tap water; the rest lived in houses without lead pipes.
Starting in 2001, when the lead levels in the city's drinking water soared, this CDC study shows that the number of kids with high blood-lead levels (above the CDC line of concern) also increased -- and the problem is more severe for kids living in the lead pipe houses. The number of affected kids went down in 2004 as the lead in water decreased. Raymond and colleagues showed that the link between the water and the lead was strong and couldn't be explained by other factors.
"Why has CDC kept quiet about these results?" asks Yanna Lambrinidou, president of Parents for Nontoxic Alternatives, a D.C. activist group "It makes no sense if they are concerned about public health. Are they trying to cover up the harm that lead contaminated water caused in D.C.?" Raymond would not respond to any questions posed by Salon.
Bruce Lanphear, a pediatric epidemiologist at Simon Fraser University in Vancouver, who has studied lead effects on children, says, "It is critical to investigate how and why these earlier studies failed to show any increase in children's blood-lead levels."
The House Science and Technology Committee Investigations and Oversight Subcommittee is beginning an investigation into CDC's handling of the D.C. lead crisis. Subcommittee chair Brad Miller, D-N.C., wrote the CDC on March 13, requesting all "records that indicate possible, probable or actual forgery, fabrication or other intentional misrepresentation of data," concerning lead in the water.
"It would be easier to understand CDC's nonchalance about losing almost half the results for 2003 if its conclusions were consistent with what other scientists found," Miller says. "It's also difficult to understand why the loss of so much data didn't merit a caveat or even a footnote in CDC's report." He adds, "If the CDC tells parents that they shouldn't worry about their children's health, its evidence had better be rock solid. It's hard to win back lost trust."
Shares