Bird Flu H5N1—What We Know So Far About Its Spread To Cows

Avian influenza has been spreading for at least four years, causing widespread outbreaks and killing millions of birds. Then, it was found in clusters of sea lions, farmed mink, cats, and other mammals.

This past month, the threat of human spread grew higher as the Highly Pathogenic Avian Influenza A H5N1 was found to have spread from dairy to dairy and from dairies to poultry farms. It likely began earlier than that and was not detected because we don’t do thorough surveillance.

In the U.S., only two human cases have been reported so far. The first, in 2022, was in a poultry worker in Colorado.

Since 1997, 909 cases of HPAI A(H5N1) have been reported from 23 countries, mainly after poultry exposure. More than 50% died. This high fatality rate is likely exaggerated because we don’t know how many people were actually infected. Notably, there was no sustained person-to-person transmission.

In March 2024, a dairy worker in Texas was found to be infected with HPAI A(H5N1)—aka, H5N1 bird flu. His only symptom was mild conjunctivitis, but he was treated with Oseltamivir (Tamiflu). Infections in dairy herds have now been reported from multiple states. In some cases in Michigan and Idaho, transfers of cows from Texas appeared to be the origin. There has been a spillover into a various mammals, mostly wild foxes and skunks. Infected cats on farms have shown neurologic symptoms like seizures, and a number have died.

Bird flu has also been found at one of the nation’s top egg producers, a large poultry farm in Michigan. Appropriate biosecurity measures have been taken there, including having workers change clothes, leaving work clothes and boots on the farm, spraying shoes and car tires before leaving, and cleaning all the trucks.

Transmission

We’ve learned that the viral load in milk from infected cows is especially high. This suggests that infection might be spread through milking machines or when the milking rooms are power washed, aerosolizing infected milk. This is why workers should wear protective gear.

Another huge problem is that cattle in the U.S. are fed poultry “litter,” a euphemism for leftover feathers, bird droppings, spilled seed, and the occasional dead mouse or rat mixed in. There is supposed to be a 15-day period before cattle are slaughtered when the broiler litter is not to be used. Because of this two-week “weaning period” before human consumption, dairy cows are not supposed to be fed bird litter.

Cow manure—based on poultry litter—is widely spread on fields as a fertilizer, including on organic farms. So, our whole agricultural supply chain could be at risk of spreading infection. At least with limited testing, it doesn’t look like cows are shedding virus thru their stool.

There are other possible routes of transmission. For example, house flies and blow flies can transmit H5N1 and other avian influenza viruses. During an HPAI (H5N1) epidemic in Thailand in 2005, blood-engorged mosquitoes collected at poultry farms tested positive for H5N1 by reverse transcription-PCR. Windborne spread could explain up to 24% of the transmission over distances up to 25 km in the 2003 outbreak.

There is now apparently spread of infection from cows back to birds. Grackles, blackbirds, and chickens all showed mutations from mammals, suggesting this route.

Gaps In Response

The government’s response to this potential biothreat has been reactive and limited.

The strongest initial criticism is that, despite calls for data and sample sharing as a foundation of pandemic preparedness, the USDA has not fully shared their data. When it finally shared genetic sequences with GISAID, the international database widely used by scientists since 2008, the USDA did not say where the samples originated nor include critical timing data.

Another problem is that the testing of cows is limited to dairy cows before they are moved to a different state. We don’t know if beef cattle are infected.

Surprisingly, the government apparently has had no power to restrict the movement of cattle or to require testing or reporting of infected herds or workers. A number of farmers are reluctant to allow testing. We know how voluntary reporting and restrictions will work…

However, on April 24, the USDA’s Animal and Plant Health Inspection Service announced that dairy cows would have to be tested before interstate transport and that labs and vets would have to report positive tests in cows. APHIS also announced finding H5N1 in a lung tissue sample from an asymptomatic dairy cow in an affected herd.

While there was one recently reported case of conjunctivitis in a person, there is likely to be considerable underreporting. There have been anecdotal reports of other symptomatic workers with fever, cough, and lethargy—and who do not want to be tested or seen by doctors. Many workers are likely undocumented and have no trust in the government. Dairy owners won’t want to risk prices falling or calls to cull their herds. Cattle veterinarian Dr. Barb Petersen said there has been underreporting and fear. She told Bovine Veterinarian’s journalist Rhonda Brooks, “But every dairy that I’ve worked with has – with the exception of one – had sick human beings at the same time they had sick cows.” She also reported infections in people with no direct contact with dairy cows. “I’m talking owners and feeders who don’t usually touch cows.”

Yet a group of veterinarians called the American Association of Bovine Practitioners proposed referring to the virus as ‘Bovine Influenza A Virus’ rather than bird flu in cattle because they “believe it is important for the public to understand the difference to maintain confidence in the safety and accessibility of beef and dairy products for consumers.”

Criticisms Of Response

There has been considerable criticism of the slow response from federal agencies. For example, Michael Osterholm, director of the University of Minnesota’s Center for Infectious Disease Research and Policy, told The Telegraph, “The USDA is primarily focused on protecting the animal production industry, and I think that’s a short-term goal, but a long-term mistake.”

Eric Topol, MD, director of the Scripps Research Translational Institute, tweeted, “I’m not worried about H5N1 transmission to humans (yet) but the @USDA’s pathetic lack of transparency, how long it took for the genomes to be released, lack of testing asymptomatic cattle……

All detracts from “confidence.”

Rick Bright, PhD, a virologist, immunologist, and CEO of the Pandemic Prevention Institute at The Rockefeller Foundation, has echoed those concerns, adding that he is going to “pause my milk consumption while waiting for data.”

Criticism has also been directed towards the lack of surveillance, particularly in asymptomatic animals, beef cattle, and farm workers. The farm hands are generally young and healthy and unlikely to present for care at a hospital, where most surveillance is focused.

Safety Of Dairy Products

Ill cows’ milk looks different—it is thicker and discolored. It is supposed to be discarded, yet it has clearly found its way into the commercial markets.

While the FDA has tried to assure the public that milk is safe because of pasteurization and that infection from milk is of “no concern,” critics note that the agency has not offered proof through testing.

There have been citizen science studies in which researchers bought 150 milk products at grocery stores in the Midwest, processed in 10 states. They found fragments of virus by PCR testing in 58 of the samples. They were unable to grow the virus from these samples. We are still waiting to see the results of FDA testing to see if pastuerization does kill the virus as they expect it to.

H5N1 news is very rapidly changing and hard to keep up with. Brennan Klein, Moritz Kraemer, and SV Scarpino have posted a handy timeline that they will frequently update.

A bit of good news to close with: The USDA is monitoring feral swine. They have not detected H5N1 so far. Pigs becoming infected is a major concern because mutations of the flu virus in pigs are most likely to be able to infect people as well.