Between 2003, when the nationwide network of air samplers was first deployed, and 2006, officials at the federally funded Lawrence Livermore National Laboratory filed five patent applications aimed at improving BioWatch's reliability.
"The existing methods for detecting" a release of disease-causing organisms into the environment were "inadequate," according to a patent application filed on behalf of Livermore scientists in December 2006.
The application cited a "higher than acceptable rate of false positive ... results," adding: "False positive results lead to confusion regarding whether [a pathogen] is actually present and whether protective measures should immediately be implemented."
The previously unpublicized documents contradict repeated assertions by the Department of Homeland Security, which oversees BioWatch, that the system has performed flawlessly. The department's chief medical officer, Alexander Garza, said last month that "there has never been a false positive result."
President George W. Bush unveiled BioWatch in his 2003 State of the Union address, saying it would "protect our people and our homeland" against a rogue state or terrorist group. Since then, the Bush and Obama administrations have spent about $1 billion to deploy detection equipment in more than 30 U.S. cities and at major spectator events, including Super Bowls and national political conventions.
As the Los Angeles Times reported last month, the system has been plagued by false alarms — at least 56 in all — that triggered tense, high-stakes deliberations over whether to order mass evacuations, distribute emergency medicines or shut down major venues.
In each case, health authorities decided, sometimes with great trepidation, to disregard BioWatch. No evidence of an intentional release of a pathogen has ever been found.
That history has caused state and local health officials to call the system ill-conceived or unworkable. Scientists at the U.S. Centers for Disease Control and Prevention told White House aides in November that they would not release emergency medications based solely on a BioWatch alert. And federal officials have given up plans to deploy the system in airports for fear of needlessly disrupting travel.
Nonetheless, Homeland Security officials have continued to defend BioWatch, insisting that it has been an "enormous success story," as a Bush administration appointee told Congress in 2007.
BioWatch uses vacuum-powered samplers that suck air through composite filters, which are removed daily and analyzed at public health labs for signs of anthrax, smallpox, plague or other deadly pathogens. Janet Napolitano, secretary of Homeland Security, is expected to decide soon whether to approve an automated "lab in a box" version that would cost an additional $3.1 billion over the next five years.
Most of the documented false alarms have involved the supposed detection of tularemia, a bacterium that can infect or kill humans in lower concentrations than anthrax or other agents.
On the closing day of the 2008 Democratic National Convention in Denver, for example, BioWatch signaled the presence of tularemia at the convention site, threatening to disrupt plans for Barack Obama's speech accepting his party's presidential nomination.
After six hours of deliberations and further testing, state health officials decided that the alert had been erroneous, and Obama gave his speech on schedule.
The reason for such false alarms is that BioWatch has been unable to distinguish between the tularemia bacterium and genetically similar organisms that pose no harm to humans, according to scientists with knowledge of the system.
Three of the five patent applications filed by the Livermore lab involved efforts to better distinguish between the DNA of tularemia and that of its close cousins.
"A more reliable method" of detecting tularemia, the applications said, "would reduce the occurrence of false positive ... results and provide decision makers with greater confidence in implementing appropriate countermeasures."
The other patent applications dealt with two other pathogens targeted by BioWatch — the germs that cause brucellosis and plague.
Patents were granted in response to all five applications.
A spokesman for the Livermore lab, Steve Wampler, referred questions about the patents to Homeland Security officials in Washington.
Citing "security concerns," a Homeland Security spokesman, Peter Boogaard, declined to say whether any of the patented innovations were incorporated into BioWatch.
Detection methods "are constantly evolving, and the BioWatch program aims to utilize the best, most reliable resources," Boogaard said.
One of the seven inventors named on the patent applications confirmed in an interview that the innovations were intended to improve BioWatch's performance.
The scientists were looking for "ways of increasing specificity" — a technical term for the system's ability to distinguish between different microorganisms — said biologist Gary Andersen, now based at the Lawrence Berkeley National Laboratory. The newer laboratory techniques, he said, targeted "unique regions" of the tularemia genome, in search of a more reliable signature.
Homeland Security officials say that none of BioWatch's alerts has been a "false positive" because the system detected some organism in the environment — even if it posed no danger to humans and had no connection to a biological attack. The department calls such incidents BioWatch Actionable Results, or BARs, rather than false alarms.
A National Academy of Sciences panel that examined BioWatch rejected that interpretation, labeling the incidents "BAR false positives."
Garza, Homeland Security's chief medical officer, said last month that determining what to do about a BioWatch alert is the responsibility of state and local officials.
"If BioWatch detects a potential threat, state and local officials as well as first responders have the ability to investigate the incident to the fullest and determine if there is a credible threat to the public," Garza wrote in response to the July 8 Los Angeles Times article.
In a letter to Napolitano on July 19, members of the House Energy and Commerce Committee requested documents on BioWatch and an explanation for Garza's claim that there had been no false alarms.
Some state and local officials believe BioWatch is so unreliable — and the potential for disruption so high — as to render the system useless.
In New York City, a senior official ordered the removal of automated BioWatch detectors, which were being tested in the subways in 2008, after learning of repeated false indications of a pathogen, according to those with knowledge of the matter, who spoke on condition of anonymity.
Richard Falkenrath, who served as a White House advisor on bio-weapons preparedness from 2001 to 2006 before becoming a counter-terrorism official for the New York Police Department, said he had concluded that it was unreasonable for city officials to have to interpret BioWatch alerts.
"We are not deploying these to do science projects. We are deploying these to respond when they go off," Falkenrath said at a biodefense conference in Baltimore two years ago. The remarks were videotaped.
A detection system was useful, Falkenrath told the audience, only if police could respond to an alarm immediately, without "debate on conference calls about whether we can trust that result."
"And if it doesn't meet that criteria, I don't want it in my city," he added.
Falkenrath, who left his NYPD position in the summer of 2010, noted that 5.5 million passengers ride New York's subways daily. In the event of a BioWatch alert, stations would have to be evacuated or quarantined. "Everyone in the whole world will see it going on," he said. "That is a serious, serious event."
Falkenrath said he questioned "whether the top-most federal officials, including my former boss, President Bush, appreciated what the nature of the early-warning system was... and what would happen if it went off."
One audience member at the conference, Dr. Albert J. Romanosky, medical director for the state of Maryland, was prompted to describe his own frustration with the system.
"We found out in Maryland that they had deployed BioWatch collectors indoors, at a facility that we knew nothing about," Romanosky said. "And then we got a positive result. On the conference call, it's a debate about what it means.... The federal officials on the call were saying, 'Oh, it doesn't mean anything. It's nothing; don't respond.'"
In his remarks, Romanosky did not identify the facility, and he did not answer questions from The Times seeking more information about the incident.
But a colleague, Jack DeBoy, who headed Maryland's public health lab at the time, said it was not the only BioWatch false alarm in his state.
"Quite frankly," he said, "we had a whole bunch of them."