Alison Young shares 3 lessons from past lab accidents:
Lab workers can be infected without a recognized accident
Initial local media reports said Shipilova broke a test tube before becoming ill, but Russian authorities have since told the WHO that no biosecurity incidents occurred in the laboratory.
Studies of lab-acquired infections have documented for decades that when workers become infected, they often can’t recall having any kind of exposure incident.
As far back as the 1950s, studies found that just 16% of lab worker infections involved recognized accidents.
[…]
Infected workers may not realize their illness came from their lab
Initial Russian media reports, which government authorities dispute, suggested that Shipilova died from pneumonic plague. This lung infection can be lethal if not treated within 24 hours of initial symptoms, which include shortness of breath and coughing.
It remains unknown what killed Shipilova or whether her lab work played a role. Before she died with severe pneumonia, her illness seemed like a typical respiratory infection, Russian health officials told the WHO earlier this month. Tests didn’t detect any microorganisms associated with her work, Russian authorities have said.
[…]
When a tularemia researcher at the prestigious U.S. Army Medical Research Institute of Infectious Diseases — often referred to as USAMRIID — started feeling ill with a fever, chills, and headache in 2009, she never alerted the lab’s medical team because she assumed she had just caught something from her children, who also were sick.
[…]
That same year University of Chicago plague researcher Malcolm Casadaban died after becoming infected with a weakened form of the Yersinia pestis bacteria he was studying. This particular strain was not known to cause lab infections or fatalities.
[…]
The public rarely learns the details of laboratory accidents
As I report in my book, “Pandora’s Gamble,” there is no universal, mandatory system for reporting lab accidents and lab-associated infections. Not in the U.S. or worldwide.
There also isn’t any comprehensive system to analyze biosafety breaches and share lessons learned so that safety can be improved. This makes it difficult to know how often accidents happen and how often workers become infected.
The 2025 systematic review of published reports during 2000 to 2024 identified 712 laboratory-acquired infections worldwide, including 276 infections and eight deaths involving research facilities. The researchers cautioned that the findings are likely an underestimate because biosafety lapses occur more frequently than are reported.
[…]
For example, after the critical lab waste sterilization plant serving USAMRIID at the Army’s Fort Detrick in Maryland suffered a catastrophic leak in 2018, the government assured the public that no safety threat occurred.
Only through months of digging into the government records behind the official statements was I able to report that a leaking tank had spewed thousands of gallons of unsterilized lab wastewater into an area with an open storm drain that emptied into a public waterway. While no illnesses were connected to the event, community leaders expressed concern about how they’d been kept in the dark about the extent of the incident.
USAMRIID used to publish some statistics about its lab incidents online. But the data was removed in 2022, around the time I was asking officials why more recent data hadn’t been posted. At that time, USAMRIID said the removal was part of a website update. A spokesperson for USAMRIID did not answer questions last week about why the institute hasn’t resumed posting the statistics.
Russia has its own history of concealing laboratory accidents. After dozens were killed by an accidental anthrax release from a Soviet biological weapons facility in Sverdlovsk in 1979, Soviet authorities blamed contaminated meat. It wasn’t until 1992 that the government acknowledged the outbreak was caused by military bioweapons research.