WASHINGTON (AP) -- Federal health officials knowingly used flawed data in a study that calmed public fears about lead in the District of Columbia's drinking water in 2004, according to a congressional investigation released Thursday.

The report by a House science and technology subcommittee admonishes the Centers for Disease Control and Prevention for the study's methodology and says the CDC has "failed its public health responsibilities" by refusing to withdraw the report.

District officials say the problem with high lead levels in drinking water has since been fixed.

A CDC official defended the federal agency, saying it reported as factually as it could in 2004, based on information it had. A second analysis -- with many more blood tests -- was later conducted.

"We have concluded that CDC's initial reports did not understate the magnitude of the problem," said Dr. Robin Ikeda, the CDC's deputy director over environmental health.

She was speaking at a congressional hearing Thursday chaired by Rep. Brad Miller, a North Carolina Democrat. Miller faulted the CDC's handling of other environmental health investigations, suggesting a pattern of downplaying problems that were later exposed as legitimate concerns.

The problem with Washington's drinking water began in 2000 when officials switched the disinfectant they used to purify the water. The switch was supposed to make the water cleaner. But the change also increased corrosion from the city's lead pipes, upping the amount of lead in the water.

City officials knew about the problem, but failed to quickly warn residents, according to the report.

In January 2004, The Washington Post exposed the issue. In response to a public outcry, D.C. sought the CDC's help in evaluating the impact of the high lead levels.

The CDC study was reassuring. It found that the high lead levels were not noticeably harming city residents. But the congressional investigation says the CDC study was based on "fundamentally flawed and incomplete data."

Among other things, the subcommittee's report said the CDC ignored the fact that thousands of blood lead tests were missing.

The CDC's Ikeda told the hearing that its 2004 report was based on the 2003 blood lead tests results it got at the time from the district. It recently collected all 2003 test results and did another analysis. The percentage of elevated tests was actually lower in the second analysis than in the first, according to Ikeda.

During Thursday's hearing, Marc Edwards, a Virginia Tech scientist who has studied the issue, called the CDC report "little more than a cheap publicity stunt" and a "historic violation of the public trust." A 2009 analysis he wrote prompted the congressional investigation.

The CDC has its defenders. In October, a group of public health officials and advocates sent a letter to Miller, saying they believe CDC acted quickly and appropriately in the D.C. investigation by advising pregnant women and young children against drinking unfiltered tap water. The agency also dispatched public health officers to provide filters to residents who needed them, the letter said.

"In our view, this exemplifies the quick response that was needed," said the letter signed by David Jacobs, research director for the National Center Healthy Housing, and more than a dozen others.

Copyright 2010 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.