The FBI used flawed scientific methods to investigate the 2001 anthrax attacks that killed five people and sickened 17 others, federal auditors said Friday in a report sure to fuel skepticism over the FBI's conclusion that Army biodefense researcher Bruce Ivins was the sole perpetrator.
The 77-page report from the Government Accountability Office says the FBI's research, including novel microbial forensic tests, did not provide a full understanding of how bacteria change in their natural environment and in a laboratory. This failure to grasp the reason for genetic mutations that were used to differentiate between samples of anthrax bacteria was a "key scientific gap" in the investigation, the report says.
The GAO also found a lack of rigorous controls over sampling procedures and a failure to cite the degree of uncertainty in measurement tools used to identify genetic markers.
"Although the complexity and novelty of the scientific methods at the time of the FBI's investigation made it challenging for the FBI to adequately address all these problems, the agency could have improved its approach," the report said.
The GAO didn't take a position whether Ivins, who worked at Fort Detrick in Frederick, Maryland, made and mailed the anthrax-filled envelopes.
This is the second government report to find fault with the FBI's methods. The National Research Council said in 2011 that federal investigators overstated the scientific case against Ivins but that the evidence was consistent with the FBI's conclusions.
The genetic markers were the basis for investigators' conclusion that the parent material of the Ames strain of anthrax spores used in the attacks came from a flask labeled RMR-1029 that was created and solely maintained by Ivins.
The new report doesn't shake the FBI's confidence in its decision to formally close in early 2010 the investigation it dubbed Amerithrax, spokesman Christopher Allen said.
"Our conclusions weren't based solely on the science but the full evidence before us," he wrote in an email. Investigators have acknowledged that the evidence is all circumstantial.
The FBI said in a written response to the report that in the past decade it has helped develop a national microbial forensics laboratory to conduct "validated and accredited" investigations.
Ivins died in July 2008 of an apparently intentional Tylenol overdose as the Justice Department prepared to indict him for the attacks. He had denied involvement, and his lawyer and some colleagues have maintained he was an innocent man hounded to self-destruction.
The GAO report was requested by a trio of congressional members led by Rep. Rush Holt, D-New Jersey, whose district includes the Princeton mailbox used in the attacks. Holt, who is retiring at the end of the year, said Congress should demand a comprehensive, independent review of the investigation to ensure that lessons have been learned.
"The GAO report confirms what I have often said—that the FBI's definitive conclusions about the accuracy of their scientific findings in the Amerithrax case are not, in fact, definitive," Holt said in a written statement.
Ivins' attorney, Paul F. Kemp, said he hopes Holt's plea doesn't fall on deaf ears.
"There's no evidence he did it," Kemp said.
Five people died in October and November 2001 from anthrax inhalation or exposure linked to the letters. They were a Florida photo editor, two postal workers in Washington, a hospital employee in New York City and a 94-year-old woman in Oxford, Connecticut. Seventeen others were sickened.
Postal facilities, U.S. Capitol buildings and private offices were shut for inspection and cleaned by workers in hazardous materials suits from Florida to New York and elsewhere.
Explore further: Science agency to review FBI's anthrax inquiry