Avian influenza is “evolving in ways we haven’t seen before,” says Martha Nelson, a computational biologist and staff scientist researching pathogen evolution at the National Institutes of Health — one of many scientists who have been monitoring the global H5N1 outbreak. Bird flu “is adapting to mammals, and it continues to show new tricks,” Nelson tells The Verge. The virus is spreading widely in domestic and wild animals, while exact transmission routes remain unclear. Confirmed human cases are rising, particularly among farmworkers.
More than two years into the US outbreak, we’re stuck with H5N1 for the long haul. The risk that it mutates to spread readily from person to person and that we could find ourselves in the middle of another pandemic is entrenched. As the Trump administration hacks away at scientific institutions and rapid federal changes impede the flow of information, the threat looms especially large. To mitigate those chances, animal agriculture, wildlife management, trade policy, and even cat owners may have to adapt to manage the virus on multiple fronts — indefinitely.
It’s like “watching a train wreck in slow motion.”
The Centers for Disease Control and Prevention (CDC), along with many virologists and epidemiologists outside the agency, continue to describe the threat that the virus poses to the public as low. Yet the likelihood of an H5N1 human pandemic “is growing,” says Nelson.
The vast majority of the 70 confirmed human cases in the US have been mild. But since last November, at least four people have been hospitalized with H5N1 in North America (three in the US and one in Canada). In January, one person died in Louisiana after contracting the virus from a backyard poultry flock.
Until recently, Nelson and other bird flu experts held out hope that, with some basic interventions, the H5N1 outbreak among cows and poultry would burn itself out and that cases among wild animals might fade away, as they did during a brief 2014-2015 outbreak. But the latest events prove that isn’t likely. “It’s hard to imagine a scenario where it’s no longer a pandemic threat,” Nelson says.
Since the virus was first detected in cows in March 2024, almost 1,000 dairy herds have been infected. Despite that spread, scientists saw a silver lining: for nearly the whole year, all of those cases were infections of a viral genotype called B3.13, believed to have entered the cattle population from a single spillover event in Texas, wherein a cow caught the virus from a wild bird. “We thought this was a one-off: one bird to one cow, and we wouldn’t see that again,” says Peter Halfmann, a virologist at the University of Wisconsin-Madison’s Influenza Research Institute.
Yet the more severe human cases are concurrent with the spread of a recently mutated, potentially more dangerous version of the virus called the D1.1 genotype. D1.1. is now circulating among wild birds and poultry, and it has spilled over into dairy cows at least twice in 2025, according milk testing data from the Agriculture Department. With D1.1, Halfmann explains that the threshold for cross-species transfer is “much lower than we previously thought.”
“There is absolutely no chance of containment, and we now have to think about mitigation and monitoring.”
Bird flu infections have also been confirmed among rats and mice near farms. Many other wild mammals, including fox, deer, and skunks are testing positive for the virus, but rodents are a particular concern given their notorious propensity to infest human dwellings and act as vectors of disease, says Meghan Davis, an epidemiologist and veterinarian at Johns Hopkins University. “When you have one of these classic reservoir hosts with highly pathogenic avian influenza, it gives you pause,” she says. “If this host could set up to be a reservoir, what implications would that have for our control strategies?”
It’s not yet known if mice or rats transmit the virus onward or are simply dead ends. However, research into 2023 H5N1 outbreaks among marine mammals and lab experiments in ferrets suggest that H5N1 can spread from mammal to mammal in certain cases. And at least one recent study of rats found the virus replicates in their respiratory tract, priming the rodents to shed it.
The good news is that there’s still been no confirmed instances of human-to-human transmission, and no human case clusters without clear ties to farmwork. These two things would signal a significant uptick in pandemic risk, Nelson says, and there’s no guarantee either will occur.
The bad news is that there’s ample opportunity for H5N1 to make that leap, says Seema Lakdawala, an epidemiologist studying avian influenza at Emory University. According to her, with multiple species acting as wild reservoirs, eradication is effectively impossible. “I think it’s endemic,” she says. And a December 2024 study, published in Science, concluded that just a single mutation could make the currently circulating H5N1 virus readily transmissible between humans. “What we’re allowing this virus to do is like 1,000 shots on goal,” Lakdawala says.
H5N1 has been around for decades. The virus first evolved in waterfowl like ducks and geese, but it has long been transmissible among many other types of birds including seabirds and songbirds. It was initially isolated following an outbreak in Scottish chickens in 1959. The first human infections and deaths occurred during a 1997 outbreak in Hong Kong. In the years since, public health authorities have contained periodic outbreaks as they’ve happened, and the worst case hasn’t materialized yet.
“If you look at the last 80 years, we’ve never seen anything like this with H5.”
Now, though, “we’re in a whole new ballgame,” says Michael T. Osterholm, an epidemiologist and director of the Center for Infectious Disease Research and Policy at the University of Minnesota. “If you look at the last 80 years, we’ve never seen anything like this with H5.”
Natasha Bagdasarian, an epidemiologist and the chief medical executive in Michigan, has been at the forefront of the state’s much-lauded multipronged response to the H5N1 outbreak. Last spring, she was hopeful that the spread could be stamped out through thoughtful communication with farmworkers and disease-monitoring strategies in just a handful of states. Less than a year on, despite the fact that new cases in Michigan dairies have stalled, that optimism is gone. “At this point, I would say there is absolutely no chance of containment, and we now have to think about mitigation and monitoring.”
With a virus like rabies, wildlife managers will sometimes disperse an oral vaccine through food to control local outbreaks among animals. With H5N1, Andrew Pekosz, a virologist at Johns Hopkins University, says that’s not a viable option. It’s in too many different species across too large of an area. Even if we had some readily dispersible bait that all of these animals were attracted to, there is no oral H5N1 vaccine to use. In lieu of that far-fetched hypothetical, Pekosz says mitigation means doubling down on reducing risk at the interfaces between wildlife, domestic animals, and humans.
Backyard poultry and pets — especially cats — are vulnerable to the virus and worrying potential transmission sources to the public, says Davis. Health officials have confirmed dozens of infections in domestic cats, most recently in New York City.
In February, the CDC published, deleted, then rereleased an unsettling report describing two instances where indoor cats in the households of dairy workers fell ill with bird flu. Out of five cats, three got sick and two died. Some people in the households, including those who had no direct exposure to farm animals or raw milk, also showed symptoms of illness. Despite layers of safety precautions, H5N1 found its way in and potentially spread from cow to human, human to cat, clothes to cat, cat to cat, or cat to human. The web of possible transmissions and infections remains unclear and unconfirmed.
Davis advocates for pets to be included in viral surveillance efforts. In the interim, pet owners should keep track of H5N1 reports in their area, be sure to always supervise pets outside, and avoid feeding animals raw milk or meat products. Those with backyard chickens or other fowl should implement hygiene measures, in accordance with federal and local guidelines.
Yet, for now, the biggest locus of prevention over the long term remains farms. Michigan and other states, like California — where the virus has hit about 70 percent of dairies — should commit to increased and constant surveillance on farms, coupled with immediate interventions upon detection, Lakdawala says, adding, “But I think it’s going to be really hard. I don’t know if the farmers are ready for something like that.”
“It’s going to be really hard. I don’t know if the farmers are ready for something like that.”
The USDA began requiring states to submit samples of raw milk for H5N1 testing last December, but that doesn’t identify individual infected cows or stop the virus from spreading among an entire herd or farm. To reduce overall human exposure to H5N1, and minimize the chances the virus makes a significant jump from cows to humans or spills back into wildlife, we need to be testing individual cows on affected farms, Lakdawala says. Such a regimen would also help scientists better understand the virus, offering a window into the number of sick cows and how the flu is spreading. The current prevailing belief is that most transmission between cows occurs via contaminated milking equipment, yet it’s important to monitor for evidence of other forms of spread, she explains, because that would mean a significant change in the virus and a rise in risk.
Biosecurity efforts on farms should also be ramped up and sustained, Pekosz says. The types of flock containment measures that have been deployed for poultry need to be strengthened and translated to dairy cattle, he explains. “For most of the poultry industry, there is no such thing as free-range chickens.” On many farms, birds are held in closed facilities, and all people and equipment entering undergo a disinfection process: changing clothes and sanitizing boots. “A cow isn’t a bird, but the same principles can be applied,” he says.
Many states have already advised or mandated farms to institute new biosecurity protocols, including sanitizing vehicles and equipment, preventing contact between wildlife and farmed animals, restricting animal transport, and recommending protective equipment like masks and face shields to workers. Yet major challenges remain, says Tim Boring, director of the Michigan Department of Agriculture and Rural Development. “The prospect of completely segregating out wildlife from dairy farms would certainly be a daunting task,” he says, adding that it would also be an expensive one.
More wildlife testing could help farms defray costs by catching new viral mutations and trends of local spread even earlier, Pekosz says. With sufficient wildlife monitoring, farmers could hold off on the most intensive defense strategies unless and until nearby wild animals test positive.
Though a wildlife vaccination campaign isn’t feasible, the USDA is investigating the use of avian flu vaccines in poultry and cows. Despite some countries already vaccinating poultry, the US has so far avoided the practice as many trade partners ban imports of vaccinated birds. Moving forward, global trade agreements will likely require renegotiation.
It’s a good sign that the US is seriously considering vaccines, Halfmann says, but he warns that any vaccination effort will require sustained investment to be effective. “We can’t have one season where we vaccinate and then stop. This is a long-term endeavor that we’re going to have to go through that’s going to cost money and require buy-in from farmers. It’s a lot of hurdles.”
“It’s hard to imagine a scenario where it’s no longer a pandemic threat.”
An animal vaccine would need to be regularly assessed and updated to keep pace with viral mutations, just like the seasonal human flu has historically been, he explains. Given that the Food and Drug Administration is now helmed by prominent anti-vaxxer Robert F. Kennedy Jr. and that the agency recently canceled a meeting for the seasonal flu vaccine selection process without explanation, Halfmann is skeptical that the federal government is up to the task.
He’s not the only one. While the virus is evolving quickly, so are US regulatory agencies. Lakdawala says there are fewer updates on federal efforts around H5N1. Until recently, Davis participated in regular small group briefings with the CDC, where governmental and nongovernmental experts would share data and insights on the virus. Now, she says those meetings have completely stopped. “What’s been most concerning to me is the lack of communication,” she says. “The backbone of outbreak response is communication.”
Without up-to-date information, virologists, epidemiologists, and state public health officials are left in the dark, with little to guide their science or policy. At the same time, research institutions are facing real and threatened major budget cuts and are enacting extensive layoffs.
“We are flirting with a massive global pandemic that could be equal to or greater than COVID-19,” Nelson says; yet, all the knowledge and tools we built during the pandemic aren’t being harnessed. It feels, she adds, like “watching a train wreck in slow motion.”
