WASHINGTON — For two years, the nationwide BioWatch system, intended to protect Americans against a biological attack, operated with defective components that left it unable to detect lethal germs, according to scientists with direct knowledge of the matter.
The federal official who oversaw installation of the components was quietly shifted to a position with no responsibility for BioWatch, and the entire episode was kept out of public view.
The U.S. Department of Homeland Security, which oversees BioWatch, opened an internal investigation, whose status remains confidential.
In more than 30 cities, BioWatch samplers located atop buildings, in train stations and in other public places suck air through dry filters around the clock. Once a day, the filters are taken to public health laboratories to be analyzed for traces of smallpox, anthrax, plague and other pathogens.
Lab technicians extract genetic material from the filters and then use kits, called assays, to release fluorescent dyes into it. When a laser is shined through the mixture, the dyes are supposed to light up if one of the pathogens targeted by BioWatch is present.
The labs originally used a series of separate assays, each designed to detect a specific germ. In 2007, Homeland Security equipped most of the labs with new kits intended to screen for multiple pathogens at the same time.
The aim was to reduce personnel costs and enable faster detection of a biological attack, and thus a speedier response.
But the new components, called "multiplex" assays, triggered false alarms, a recurring problem with BioWatch since the system was put into operation nationwide in 2003.
After scientists at many of the labs voiced concerns, Homeland Security officials, in consultation with microbiologists from other federal agencies, ordered testing of the new assays.
The tests, conducted in secrecy at the Pacific Northwest National Laboratory in Washington state and the federal Centers for Disease Control and Prevention in Atlanta, found that the kits were unsuitable for BioWatch, scientists familiar with the matter said. They spoke on condition of anonymity, citing the sensitivity of the information.
The multiplex assays could not distinguish between the bacterium that causes tularemia, a potentially deadly condition also known as rabbit fever, and similar but benign organisms called "near neighbors" that are abundant in outdoor environments.
The original assays had exhibited the same problem. But the multiplex assays had an additional shortcoming, scientists said: They were found to be far less sensitive to the presence of actual pathogens than Homeland Security officials had presumed.
In late 2009, Homeland Security officials removed the new assays and returned to using kits that searched for pathogens one at a time.
Peter Boogaard, a Homeland Security spokesman, declined to respond to written questions about the matter. Jeffrey Stiefel, the department official responsible for installing the ill-fated assays, said he was not authorized to comment.
Some of the scientists familiar with BioWatch said the multiplex assays were put into use without adequate testing to validate their effectiveness.
The assays were designed at the CDC and the Lawrence Livermore National Laboratory and were built to Homeland Security's specifications by a private company, the scientists said.
Richard F. Meyer, a microbiologist who helped develop the multiplex assays while at the CDC and later supervised their installation as a contractor for Homeland Security, defended the kits.
Meyer said the original assays "were past their life cycle and in constant need of repair." Data collected by Livermore scientists, he said, "supported the use of the [new] technology."
Meyer acknowledged that he lost his contracting role with Homeland Security because of dissatisfaction over how the multiplex assays performed once installed.
"When you don't agree with those in charge you get pushed aside," he said in an email.
A spokesman for Livermore, Steve Wampler, declined to discuss the lab's role in developing assays for BioWatch.
The failure of the multiplex assays is one in a slew of problems that have beset BioWatch since President George W. Bush unveiled the system during his State of the Union address in January 2003.
Bush said BioWatch would "protect our people and our homeland" against a germ attack by terrorists. In subsequent years, presidential appointees in Homeland Security have repeatedly assured Congress that BioWatch was functioning effectively.
The Los Angeles Times reported in July that BioWatch has been unable to distinguish between dangerous and benign organisms, and that as of 2008, federal agencies had documented 56 false alarms.
In one of those incidents, during the 2008 Democratic National Convention in Denver, BioWatch units signaled the presence of the tularemia bacterium, triggering tense deliberations among local, state and federal officials over what steps should be taken to protect the public.
After follow-up tests found no traces of the germ at the convention site, officials decided not to take emergency measures, and that evening Barack Obama accepted his party's nomination for president on an outdoor stage, as scheduled, before a crowd of more than 80,000 people.
Not once have public health officials had enough confidence in a BioWatch alarm to evacuate an area, dispense antibiotics or take any other emergency action.
After considering the potential disruption from false alarms, federal aviation officials shelved plans to install air-sampling units inside the nation's major airports.
In response to The Times' reporting, congressional Republicans and a senior Democrat have written to Homeland Security Secretary Janet Napolitano seeking documents and explanations. Although Napolitano has not commented publicly, the department's chief medical officer, Dr. Alexander Garza, has staunchly defended BioWatch.
In a statement, Garza said in July that the system had never generated a false alarm. "The detection of commonly occurring environmental agents," he wrote, "is not a 'false positive.'" Asked to elaborate while appearing before a congressional panel Sept. 13, Garza said each detection by BioWatch was "a true positive."
The notion that such events — which Homeland Security calls BioWatch Actionable Results, or BARs — are not false alarms was earlier considered and rejected by a committee of experts appointed by the National Academy of Sciences.
In its report in October 2010, the committee said that "all BARs to date have been 'BAR false positives,' meaning they have signaled the potential occurrence of a terrorist attack when none has occurred."
The committee warned that "repeated false alarms may eventually create a sense of skepticism or complacency that could delay or hinder an appropriate response to a true bioterrorism event."
One of the committee members, Northern Arizona University geneticist Paul Keim, said in an interview that the detection of a benign organism could not be considered a "true positive."
"That's why we call them near neighbors," Keim said. "If they cause disease, we call them a pathogen."
Garza, in his recent congressional testimony, said that the existence of the near neighbors had come as a surprise to Homeland Security and that the department was now seeking "more specific assays."