More than 1,100 laboratory incidents involving bacteria, viruses and toxins that pose significant or bioterror risks to people and agriculture were reported to federal regulators during 2008 through 2012, government reports obtained by USA TODAY show.
More than half these incidents were serious enough that lab workers received medical evaluations or treatment, according to the reports. In five incidents, investigations confirmed that laboratory workers had been infected or sickened; all recovered.
In two other incidents, animals were inadvertently infected with contagious diseases that would have posed significant threats to livestock industries if they had spread. One case involved the infection of two animals with hog cholera, a dangerous virus eradicated from the USA in 1978. In another incident, a cow in a disease-free herd next to a research facility studying the bacteria that cause brucellosis, became infected due to practices that violated federal regulations, resulting in regulators suspending the research and ordering a $425,000 fine, records show.
But the names of the labs that had mishaps or made mistakes, as well as most information about all of the incidents, must be kept secret because of federal bioterrorism laws, according to the U.S. Department of Agriculture, which regulates the labs and co-authored the annual lab incident reports with the Centers for Disease Control and Prevention.
The issue of lab safety and security has come under increased scrutiny by Congress in recent weeks after a series of high-profile lab blunders at prestigious government labs involving anthrax, bird flu and smallpox virus. On Friday, a CDC investigation revealed how a rushed laboratory scientist had been using sloppy practices when a specimen of a mild bird flu virus was unwittingly contaminated with a deadly strain before being shipped to other labs.
Earlier this summer, other researchers at CDC potentially exposed dozens of agency staff to live anthrax because of mistakes; nobody was sickened. Meanwhile, at the National Institutes of Health, long-forgotten vials of deadly smallpox virus were discovered in a cold-storage room where they weren't supposed to be.
The new lab incident data indicate mishaps occur regularly at the more than 1,000 labs operated by 324 government, university and private organizations across the country that are registered with the Federal Select Agent Program. The program is jointly run by the USDA and the CDC, which are required by law to annually submit short reports with incident data to Congress.
The reports, released by CDC in response to a request from USA TODAY, contain few details beyond a count of incidents by categories, such as incidents involving bites or scratches from infected animals, needle sticks, failures of personal protection equipment, spills or specimen packages that temporarily went missing after they were shipped. No thefts were reported.
Data for incidents reported in 2013 is not yet finalized, CDC said. In 2012, lab regulators received 247 reports of potential releases of dangerous pathogens. They also received 247 reports in 2011. There were 275 reports in 2010; 243 in 2009; and 116 in 2008. The reports come from regulated select agent research labs as well as clinical or diagnostic labs that are exempted from registration with federal officials but still must report incidents if they identify a select agent.
"More than 200 incidents of loss or release of bioweapons agents from U.S. laboratories are reported each year. This works out to more than four per week," said Richard Ebright, a biosafety expert at Rutgers university in New Jersey, who testified before Congress last month at a hearing about CDC's lab mistakes.
The only thing unusual about the CDC's recent anthrax and bird flu lab incidents, Ebright said, is that the public found out about them. "The 2014 CDC anthrax event became known to the public only because the number of persons requiring medical evaluation was too high to conceal," he said.
CDC officials were unavailable for interviews and officials with the select agent program declined to provide additional information. The USDA said in a statement Friday that "all of the information is protected under the Public Health Security and Bioterrorism Preparedness and Response Act of 2002."
Such secrecy is a barrier to improving lab safety, said Gigi Kwik Gronvall of the UPMC Center for Health Security in Baltimore, an independent think tank that studies policy issues relating to biosecurity issues, epidemics and disasters.
"We need to move to something more like what they do in aviation, where you have no-fault reporting but the events are described so you get a better sense of what actually happened and how the system can be fixed," said Gronvall, an immunologist by training and an associate professor at the University of Pittsburgh School of Medicine.
Gronvall notes that even with redundant systems in high-security labs, there have been lab incidents resulting in the spread of disease to people and animals outside the labs.
She said a lab accident is considered by many scientists to be the most likely source of the re-emergence in 1977 of an H1N1 flu strain that had disappeared in 1957 because the genetic makeup of the strain hadn't changed as it should have over those decades. A 2009 article in the New England Journal of Medicine noted the 1977 strain was so similar to the one that disappeared that it suggests it had been "preserved" and that the re-emergence was "probably an accidental release from a laboratory source."
"People understand that mistakes will happen," Gronvall said. "But you want it to be captured, you want it to be learned from, you want there to be a record of how it was dealt with. That's something I think should happen with biosafety."
In 2012, CDC staff published an article in the journal Applied Biosafety on select agent theft, loss and releases from 2004 through 2010, documenting 727 reported incidents, 11 lab-acquired infections and one loss of a specimen in transit among more than 3,400 approved shipments.
The article noted that the number of reports received by CDC likely underestimates the true number of suspected losses and releases. Still, the data "indicate that the risk of exposure to [select agents] managed by US laboratories to the general population is low." The number of reports submitted rose annually during the period, from just 16 in 2004 to 269 reports in 2010, the article said. It's unclear why the numbers in the journal article differ slightly from those in the select agent reports to Congress reviewed by USA TODAY.
The newly released reports give limited information about the handful of incidents where there was occupational illness or an animal becoming unexpectedly infected.
• 2012: Two workers in different select agent facilities showed signs of infection with the bacteria that causes Q fever, a select agent that primarily causes illness in livestock but can also sicken people with sudden or chronic symptoms including high fevers, headaches, nausea and vomiting. While most people recover, some will experience serious illness and complications, including pneumonia, hepatitis and an inflammation of heart tissue. The report says both of the lab workers returned to full work status, but their cases were not being counted as confirmed laboratory acquired infections because each may have been infected outside their labs. One also worked with vaccine strains of the bacteria that aren't counted as select agents; the other served as a large-animal vet outside of the lab work, the report says.