On Wednesday the World Health Organization said it still could not assess the danger from the death of a 28-year-old technician at Russia’s Siberian anti-plague institute. Five days earlier it had used a binding international health treaty to ask Moscow what killed her.
Darya Shipilova died early on Oct. 2 in the district hospital in Shelekhov, a town outside Irkutsk. Rospotrebnadzor, Russia’s public health watchdog, says she had “pneumonia of unknown etiology” and that tests found “no microorganisms associated with the patient’s professional activities.” On Wednesday the agency said there was no risk of an epidemic and that medical monitoring had been completed for “over 90%” of her contacts, CNN reported. Asked about a second employee reported to be ill, Dmitry Peskov, the Kremlin spokesman, said that “the vast majority of what the media is reporting regarding this situation is false information,” Al Jazeera reported.
Russia has said what Ms. Shipilova did not die of. It has not said what she did die of. On the wider questions the public record is firmer. Labs that handle dangerous pathogens report hundreds of releases a year, and almost none of them make anyone sick. Serious accidents are rare, and in the Soviet Union and Russia several stayed hidden for weeks or years. The treaty, the International Health Regulations, requires Russia to answer the W.H.O. within 24 hours. It gives the agency no way to make Russia answer in full.
What is known
Lyudi Baikala, an independent Siberian outlet, reported that Ms. Shipilova was last at work on Sept. 25, according to Radio Free Europe/Radio Liberty. She was admitted to the Shelekhov hospital on Sept. 29 and was put on a ventilator, The Moscow Times reported in the first accounts of her death. The W.H.O. gives Oct. 2 as the date of her death. Russian officials later told the agency that her illness had looked like an ordinary acute respiratory viral infection, Healthbeat reported.
Two years earlier she had finished a residency in clinical laboratory diagnostics at Irkutsk State Medical University, with honors. A co-author listed as D.N. Shipilova of the institute appears on a 2026 study, presented at the All-Russian Congress on Infectious Diseases, which found the institute’s plague strains highly susceptible to antibiotics. A colleague told Lyudi Baikala she was “deeply anxious about ever infecting anyone if something happened,” Business Insider reported.
The response on the day she died was not routine. Anna Popova, the head of Rospotrebnadzor, flew to Irkutsk, and local television linked the trip to a suspected especially dangerous infection. More than 60 institute employees were kept inside the building from Oct. 2 to wait for test results. “They’re sleeping there on the floor and on chairs,” a relative of one employee told RFE/RL. “There are, of course, no beds.”

The Shelekhov hospital was sealed, and quarantine was reported at two Irkutsk hospitals and a maternity hospital. An aluminum plant in Shelekhov ordered its workers to wear masks. The Moscow Times reported that the police had opened a criminal case over safety violations resulting in death. Rospotrebnadzor said it followed 197 contacts. Two had Covid-19 and two had rhinovirus.
Her mother, Svetlana Shipilova, is 64 and works in health care. “They just let my daughter die,” she told the Russian outlet Glubina, The Independent reported. Four days after the death the family still had no official cause. They were in quarantine and had not been able to bury her.
Two accounts that don’t fit
Two versions of how she fell ill have circulated, and Russian officials have confirmed neither. In the first, reported by Lyudi Baikala, she broke a tube of plague bacteria at work on Sept. 25, and The Moscow Times reported that she had told medical staff so. In the second, which the BBC described, she fell ill after a field trip to Buryatia, on the Mongolian border, where plague is endemic in wild animals. Alexei Tsydenov, the head of Buryatia, wrote on Oct. 2 that she had died of plague. He also said she had not been to his region. The post was later edited to say “possibly,” Thomas Jeffries wrote in The Conversation.
Rospotrebnadzor says no accident involving pathogens was recorded at the institute. It also says Ms. Shipilova had been “vaccinated and had immunity to the dangerous infections she was working with,” Euronews reported.
Dr. Nahid Bhadelia, the founding director of Boston University’s Center on Emerging Infectious Diseases, told Business Insider that for the tube story to be true, a lab built to stop each of these failures would have had to suffer all of them.
She needs to have broken a test tube, have a breach in her personal protective equipment, not reported the incident, not gotten the antibiotic prophylaxis, just gone home under no one’s observation, shown up at a local district hospital, and exposed 200 people
Two specialists interviewed by Radio Svoboda, RFE/RL’s Russian service, doubted the tube version for other reasons. Alisa Litovka, a microbiologist, said the rumor’s wording gave it away. There is one bacterium, Yersinia pestis, and the form the disease takes depends on how it enters the body. “There’s no ‘pneumonic’ plague in a test tube,” she said. Andrei Aleksandrin, an epidemiologist who founded Ukraine’s National Expert Group on Infection Control, said modern labs use sealed plastic tubes that are practically impossible to break. Svetlana Shipilova told Glubina the same thing: “Even if you fling it against a wall.”
Neither expert ruled out an infection at work. Mr. Aleksandrin said the dates, from her last day in the lab to her death a week later, fit plague’s incubation period of one to seven days. Ms. Litovka said the vaccine used in Russia gives weak, unreliable protection against the pneumonic form. A single negative test can also miss the bacterium if the sample is poor or the patient has already had antibiotics, she said, and she called the official wording a brush-off.
Kevin Esvelt, an M.I.T. biologist who works on biosecurity, argued on X that the scale of Russia’s response was itself the clue.
Russia’s five-hospital quarantine, FSB involvement, and reported hazmat use in response to the death of a single researcher at a weapons-lineage Irkutsk lab are inconsistent with a lab-acquired plague infection, which would have been treated with prophylactic antibiotics.
1/4
On Reddit, under a story about Ms. Shipilova’s mother, a reader came down on one of the explanations Dr. Esvelt raised.
I’m still going with the Multi-Drug Resistant explanation. It fits more explanations for why someone knowledgeable would seek treatment but not get it.
The public record does not support a drug-resistant strain, though it does not rule one out either. The only published evidence on the institute’s strains, the 2026 study with a D.N. Shipilova among its authors, found them susceptible to antibiotics. Dr. Jeffries, a microbiologist at Western Sydney University, wrote that there was “no evidence the strain alleged to have killed Shipilova could not be treated by antibiotics.” The quarantine was real, and it shows that on Oct. 2 officials thought something dangerous was involved. It does not show what. “If it’s plague, breathe a sigh of relief, because we know what to do about plague,” Dr. Jennifer Nuzzo, director of the Pandemic Center at Brown University’s School of Public Health, told Time.
The institute
The Irkutsk Anti-Plague Research Institute of Siberia and the Far East was created on June 5, 1934, by an order of the Soviet Council of Labor and Defense, as an operational body for fighting plague, according to its own history. Its founder, the bacteriologist Aleksei Skorodumov, set up plague stations across Siberia and the Far East. He was arrested in 1937 on fabricated charges, executed in 1939 and rehabilitated in 1955.
Besides plague, the institute now works on cholera, brucellosis, tularemia, anthrax and arboviral infections, and it has 95 researchers. Plague persists in wild animals in the region. If Ms. Shipilova’s death is confirmed as plague, Euronews reported, it would be Russia’s first verified infection since 2016.
The anti-plague system also has another history. Andrew Weber, who ran the Pentagon’s nuclear, chemical and biological defense programs under President Barack Obama, told Healthbeat that the Soviet anti-plague institutes did real public health work but also supplied the Soviet bioweapons program. “I’m not quite reassured yet by official Russian statements given the history of lying and covering up,” he said.
On Reddit, one reader treated the link as settled.
And we know this lab in a remote corner of Siberia was founded in the Soviet era to do bio-weapon research.
The institute’s published history does not say that. It records a founding in 1934 as an anti-plague body in a region where the disease is endemic. The connection Mr. Weber described was between the Soviet anti-plague system and the weapons program, not between this institute and weapons work today. RFE/RL reported that there was no public evidence linking the Irkutsk lab to weapons research, and Healthbeat said it had seen none.
How often labs fail
The fullest public count of lab mishaps comes from the United States. The Federal Select Agent Program regulates 230 registered entities. In 2024 it received 273 reports of releases, nine of losses and none of theft, according to its annual figures. None of the releases caused illness, death or transmission among workers or into the community. In 2023 there were 215 release reports. One exposure made a worker sick, and the illness did not spread.
Most countries publish nothing like this, and no international body inspects such labs. The Global BioLabs project, run by Gregory Koblentz of George Mason University and Filippa Lentzos of King’s College London, counted 59 Biosafety Level 4 labs, the kind built for the most dangerous pathogens, operating, under construction or planned in 2021. Its 2023 report counted 69 in 27 countries, 51 of them operating. About 75 percent of the operating labs are in city centers, Vox reported. Only one of those 27 countries has comprehensive national legislation on research that can make pathogens more dangerous.
Dr. Koblentz told Vox that the risk is greatest where a country is opening its first such lab.
If this is their first lab, in some cases, they don’t have very strong national legal and regulatory mechanisms for biosafety and biosecurity. They might not have the human capital
He said a voluntary international group of biosafety regulators had gone defunct. Ideally, he said, labs would be audited by a body modeled on the International Atomic Energy Agency. None exists.
The Russian record
In 1979 an anthrax outbreak killed people in the Soviet city of Sverdlovsk. Soviet officials blamed contaminated meat and repeated that account to the U.S. National Academy of Sciences in 1988. In 1992, President Boris Yeltsin acknowledged that military bioweapons research had caused it.
Twenty-five years later the problem was delayed notification. On May 5, 2004, Antonina Presnyakova, a researcher at Vector, the former Soviet weapons laboratory near Novosibirsk, stuck herself with a syringe while working with Ebola. She died on May 19. Vector officials did not notify the W.H.O. until mid-May, so the agency’s experts could not advise on her care early, CIDRAP News reported at the time, citing The New York Times.
Accidents with real consequences have happened elsewhere. In 2004, SARS escaped from a Beijing virology institute, and nine people were infected, one of whom died. In 2019 a vaccine factory in Lanzhou, China, used expired disinfectant for nearly a month, and its exhaust carried Brucella bacteria into the city. The local health commission called it an accident in a report in September 2020. By that November, 10,528 people had tested positive for antibodies, The Global Times reported.
In an essay for OpenMind magazine, the science writer Wendy Orent went through smallpox escapes in Britain in 1966, 1973 and 1978 and the Soviet accidents. She concluded that no one knows how often labs leak.
A veil of secrecy seems to surround virtually all investigations of leaks involving dangerous pathogens.
Dr. Ashish Jha, who coordinated the White House Covid response under President Biden, made the same point about Russia to Healthbeat. “We’ve been here before,” he said. “It took years to become public.”
What the rules require
The International Health Regulations were last rewritten in 2005, after SARS. They bind 196 states, Russia among them. Article 6 requires a government to notify the W.H.O. within 24 hours of assessing an event that may be a public health emergency of international concern. Article 9 lets the agency act on reports from other sources, including the press. Under Article 10, when the W.H.O. asks a government to verify such a report, the government has 24 hours to acknowledge the request and 24 hours to supply the public health information it has.
The W.H.O. picked up media reports of the death on Oct. 2 and contacted Russia on Oct. 3. Russia’s substantive answer came on Tuesday, Oct. 6, ABC News reported. It has not been made public whether Moscow acknowledged the request within 24 hours. Russia told the agency that no plague case had been recorded in the Irkutsk region and that observation of all contacts was complete.
Dr. Tedros Adhanom Ghebreyesus, the W.H.O. director general, laid out the exchange on X, along with the questions Russia had not answered.
Since the first media reports, @WHO has been actively following up on the reported death of a laboratory worker at the Irkutsk Antiplague Research Institute in Irkutsk oblast, Russian Federation. The individual died on 2 October 2026 after being hospitalised with severe…
At a briefing in Geneva on Wednesday, Dr. Maria Van Kerkhove, who directs the W.H.O.’s epidemic and pandemic management department, said the agency had asked what tests were run on Ms. Shipilova and about 200 other people, what material she worked with, what safety measures the institute had and whether and how she had been vaccinated, Swissinfo reported.
So clearly, we would like more information to understand which laboratory tests were conducted, which pathogens were evaluated, and if this individual didn’t die from plague, then how did she die?
The Kremlin said it had kept the W.H.O. informed “in accordance with established practice.” Writing on the international law site Diplomacy and Law, Edmarverson A. Santos concluded that “the evidence currently available is insufficient to prove a Russian breach of Articles 6, 7, or 10.” He wrote that the W.H.O. had received no formal Article 6 notification by Oct. 6, and that the agency thought Russia might still have been assessing the event.
Where the treaty stops
The regulations carry no fines, and they give the W.H.O. no power to inspect a lab or a hospital. Under Article 10 the agency must offer to help. If a government refuses, the agency may share what it knows with other governments when the risk justifies it. Under the 2005 text, the only regular check on compliance was the director general’s annual report to the World Health Assembly, Gian Luca Burci, a former W.H.O. legal counsel, wrote in EJIL: Talk!
Amendments that took effect on Sept. 19, 2025 created a committee of the states that are party to the regulations. Mr. Burci wrote that the committee “is not really a compliance mechanism but rather a facilitative and consultative body.” Eleven states rejected the amendments, among them the United States, in July 2025. Russia was not on his list.
The nearest precedent shows how far public pressure gets. In September 2019 the W.H.O. said Tanzania had not given it the information it needed to assess suspected Ebola cases, an unusually public rebuke, U.N. News reported. Tanzania had told the agency that both suspected cases tested negative. It did not say what the patients had, which is the same gap the W.H.O. now describes in Irkutsk.
The United States is pressing Moscow from outside the W.H.O., which it has left under the Trump administration. The State Department sent a démarche, a formal diplomatic request. “It’s incumbent upon Russia to obviously share more information with the world,” Secretary of State Marco Rubio said on Tuesday. “Maybe it’s bigger than we thought.” On Wednesday President Trump said the United States did not believe Ms. Shipilova had died from a bioweapon. “We’re going to find out soon enough, but we don’t think so,” he said. Lawrence Gostin, a global health law professor at Georgetown University, asked in an interview with CNN “how is it possible to trust the Kremlin is telling us the full and accurate information?”
On Hacker News, a reader argued that the answers probably already exist.
There are probably already clear answers to some of these questions, but that doesn’t mean the public will get them.
Specialists say the same about the first half of that. Dr. Nuzzo told Time that by now the authorities “should have a full sequence” of whatever infected Ms. Shipilova. Asked about a possible lab accident, Dr. Van Kerkhove said, “I know what you know from the media.”
The case against alarm
So far the record shows no spread. From 2019 to 2025, 10 countries reported 3,847 suspected human plague cases and 423 deaths to the W.H.O., a fatality rate of 11 percent, according to the agency. Most were in Congo and Madagascar. Pneumonic plague spreads between people through respiratory droplets and is always fatal without treatment. Antibiotics cure it if they are started early, and under W.H.O. guidance close contacts get a seven-day preventive course. The 2017 outbreak in Madagascar caused 2,417 cases and 209 deaths, then ended.
“This case does not cause undue concern,” Michael Osterholm, director of the University of Minnesota’s Center for Infectious Disease Research and Policy, told CIDRAP News. Gigi Gronvall of the Johns Hopkins Center for Health Security pointed to geography: “Siberia is not exactly Paris,” she told Wired. Sentinel Global Risks Watch, a team of forecasters, put the chance of more than 1,000 plague deaths worldwide in the next 12 months at 0.7 percent, within a range of 0.1 to 2 percent, Scott Alexander wrote on Astral Codex Ten. He disclosed that Sentinel is a grantee of his blog. He dismissed Polymarket’s 5 percent odds on the W.H.O. declaring a plague pandemic, writing that markets handle very small probabilities badly.
The C.D.C. said on Tuesday that it saw no broader threat to the United States.
CDC is aware of and closely monitoring reports of a suspected case of pneumonic plague in the Irkutsk region of Russia that resulted in a death. We are working across the U.S. government to assess developments as additional information becomes available.
While the circumstances…
On Hacker News, a reader asked whether air travel should be shut down to contain the outbreak. One reply rejected the idea.
It is also stoppable with contact tracing. As happened with, for example, SARS. Which is why something like 200 people have been tested.
The point about contact tracing matches W.H.O. and C.D.C. guidance, and Russia says testing of its 197 contacts found no dangerous pathogens. The same commenter put plague in Congo at about 1,000 cases a year. That is more than the W.H.O.’s figures show for all 10 reporting countries combined, which average about 550 suspected cases a year. The agency says countries that do not report may still have plague.
For Dr. Nuzzo, the larger worry is readiness. Comparing today with 2019 and early 2020, she told Time, “we are more vulnerable now than we were then.”
What comes next
The W.H.O. is still waiting for answers about the second employee. Abdirahman Mahamud, a W.H.O. official, said the request was based on a media report that the agency had checked and found credible. Swissinfo described the case as a possible death. Dr. Tedros described an employee with pneumonia of undetermined cause. No name or condition has been released, and CNN said neither local media nor local officials had confirmed a second case. On Tuesday Mr. Trump said he had a call scheduled with President Vladimir Putin about the incident. On Wednesday the Kremlin said no call had been arranged.
Questions about weapons work come under the Biological Weapons Convention, which has no verification system. Its states parties meet in Geneva from Dec. 14 to 16.
In Irkutsk, Svetlana Shipilova is still waiting to be told what killed her daughter. “I’m waiting for it to become official,” she told Glubina. “Not just something in the media; it has to be an official document.”

