H5N1 bird flu was circulating in dairy cows for four months before it was detected, USDA scientists say

Bird flu was probably circulating in dairy cows for at least four months before it was confirmed to be the highly pathogenic H5N1 virus, according to a new analysis of genomic data by scientists at the US Department of Agriculture’s Animal Disease Center.

The research also found infected cattle that had no apparent connections, suggesting that “there are affected herds that have not yet been identified,” the study said.

It adds to a growing pile of evidence suggesting that the H5N1 virus had a head start in the US dairy industry for months before it came to the attention of scientists and government regulators.

The USDA’s study was published as a preprint, ahead of peer review, on the BioRXIV server on Wednesday.

It follows a similar analysis by an independent international group of almost two dozen evolutionary and molecular biologists who quickly analyzed raw genome sequences uploaded by the government to a server maintained by the National Library of Medicine. Despite the lack of critical background information on those samples, that group came to nearly the same conclusion as the USDA: that the virus had crossed over from wild birds to cows between mid-November and mid-January, which means it was circulating for months before anyone knew.

The USDA officially confirmed the presence of the H5N1 virus on March 25 in dairy cows in Texas. Since then, at least three dozen infected herds have been reported across nine states. At least one farmworker who was in contact with infected cows in Texas also tested positive for H5N1, the second human case of this type of flu ever reported in the US. The worker was given antiviral medication and has recovered.

Tests of milk from retail stores have shown inert remants of the virus in about 1 in 5 samples, the US Food and Drug Administration reported last week, suggesting that the infection had become widespread. Further testing by the FDA has confirmed that the virus in the samples of pasteurized dairy products was not active and couldn’t make anyone sick, but experts have strongly advised against the consumption of raw milk.

“We could have done a much better job” catching H5N1 in dairy cows, said Dr. Michael Worobey, head of the Department of Ecology and Evolutionary Biology at the University of Arizona who was part of the group of biologists that conducted the independent analysis. He studies how pandemics start.

Worobey said that as soon as the cows were noticeably sick with something mysterious, instead of testing for specific viruses and bacteria, ideally, a lab would have used a technique called metagenomic sequencing, which reads all the genetic material in a sample and uses computers help pick out relevant information.

“If that had been done, that would have revealed H5N1 in January and then even beyond that,” he said.

If we’re serious about preventing outbreaks in animals that could lead to human pandemics, Worobey said, regulators need to change their approach.

“We have to get out of the mindset of waiting for that tip of the iceberg of sick animals or sick humans to be noticed.” Instead, he said, animals need to be routinely tested with “modern techniques” for identifiying emerging pathogens.

The new study gives the USDA’s account of how bird flu seemed to spread so quickly to herds across the US. Samples collected between March 7 and April 8 found very similar H5N1 viruses in 26 herds in eight states and six poultry flocks in three states, suggesting that the virus had crossed over in a single spillover event between wild birds and cows.

The study says “production veterinarians” first noticed cows who weren’t eating well with changes to their milk production and quality in late January.

H5N1 has been devastating wild and domestic bird populations in the US since 2022, and it has infected a growing number of mammals.

In addition to the movement between cattle and wild birds, the study found evidence that infected cattle had passed the virus to domestic poultry flocks “through multiple transmission routes.” The researchers also found that the virus involved in the current outbreak had infected a wild animal – a raccoon – and cats that lived near the cows on dairy farms.

Interestingly, the virus sequenced from the infected farmworker had key differences from the cow genomes. The USDA scientists concluded that the differences might mean they were missing samples from the animals the farmworker had come into contact with or could have been due to the evolution of the virus from host to host.

Worobey says the study means H5N1 is “now seemingly well-entrenched in the dairy cattle population in the country” and might be something we have to deal with for years to come.

Although it’s by no means clear that this virus will change in the right combination of ways to launch a human pandemic, he said, allowing a virus to get a foothold in a population of domesticated animals puts everyone at risk.

“It does add one more species – a very important species – that didn’t have influenza A virus circulating in it before to the list of species where these viruses can have the opportunity to find that right combination that allows them to wreak havoc in the human population, not just animals,” he said.