How close is a new bird flu (H5N1) pandemic?

Probably not around the corner but getting uncomfortably closer.

What is a flu pandemic?

We are all familiar with the annual winter flu season. Even though circulating flu viruses mutate from year to year, existing population immunity from previous infection and vaccination limits the size of the annual wave and the severity of infections. Nonetheless, seasonal flu remains a significant problem. In Europe, about 10% of the population is infected every year and between 15,000 and 70,000 people die. Global estimates are about 400,000 annual deaths from seasonal flu. The elderly, the very young and those with other health conditions are most at risk of severe illness.

Sometimes a strain of flu arises that is so different to previous strains that most people have little pre-existing immunity to it, and, as long as it can transmit between people, the conditions are set for a flu pandemic. A pandemic flu strain is often due to spillover from flu strains circulating in animals. For instance in 2009’s swine flu H1N1 pandemic, it was a mix of pig, bird and human flu viruses that likely developed in pigs (which can be infected by all three) before jumping over to humans. The 1957 H2N2 pandemic arose from a mix of bird and human virus. Because far fewer people have any immunity to pandemic strains, attack rates are much higher - anywhere from 15% - 40% of the population can be infected by the time the pandemic is over. This means that even if the severity is not necessarily higher than seasonal flu, the population impact is much higher simply because so many more people get sick. This is what happened for instance in 1957, where a million people are estimated to have died. When the pandemic strain is more severe as well as infecting many more people, the consequences can be devastating as in the 1918 flu pandemic, where an estimated 50-100 million people died worldwide.

Flu pandemics can disproportionately affect younger adults and children, simply because older adults have been exposed to more strains of flu in their lifetimes. This means they might have some protection from severe disease if they were infected with a related strain years before. This is exactly what happened in the 1918 and 2009 flu pandemics.

Why did people worry about bird flu (H5N1)?

Because flu viruses continue to circulate in animals, evolve relatively rapidly and can jump between species, flu pandemics in humans will continue to occur at sporadic intervals for the forseeable future - we just can’t know exactly when they will hit. Scientists around the world have been worried about a new bird flu (H5N1) pandemic for years - partly because it was circulating in wild bird populations, particularly in Asia, and partly because in rare cases when people were infected the reported case fatality rates were really high (as high as 50%). While this case fatality rate is a significant overestimate since non-serious infections are generally unlikely to be caught, it is still frightening because even a mortality rate of a few percent can lead to horrific consequences. The mortality rate of the devastating 1918 flu was somewhere between 1-2%.

So, before Covid hit in 2020, the state of play was that scientists knew that there would be a new flu pandemic at some point and were most worried about the potential for one arising from wild bird populations first infecting poultry and then finding a route to humans. However, outbreaks in domestic bird populations were still sporadic, geographically concentrated in Asia, with very little spillover to other species, and, in the rare case a person was infected (normally through contact with infected poultry), little or no human-to-human transmission.

Global spread of bird flu (H5N1) in animals since 2020

In 2021, a new strain of H5N1 emerged in wild birds which spread rapidly around the world, helped by migratory wild birds. Some wild bird populations might never recover. By 2022, domestic poultry flocks were infected around the world, leading to the culling or death from infection of over 130 million domestic birds across the US, Europe, Asia and Africa.

Since 2022, H5N1 has infected over 50 species of mammals - mostly through eating infected wild birds, although some mass death events in seals and sea lions have raised worries that it might also be spreading from mammal to mammal. This matters because the most likely way H5N1 adapts to humans is by finding relevant evolutionary jumps in other mammals first. However, at least people were not in regular contact with these wild mammals and so human spillover events were rare. But that was about to change.

By the end of 2022, there was a big outbreak in farmed mink in Spain, with some evidence of mink-to-mink transmission. The mink got pretty sick and all of them were culled. Then, in early 2024, H5N1 popped up in dairy herds in two US states. The concerning thing is how quickly it then spread - confirmed in over 500 dairy herds across 15 states by mid November 2024.

The virus’s move into high volume animal farming, particularly cows, is concerning for two reasons: firstly, it means the virus comes into contact far more often with humans (mainly farm workers) and secondly, large cow herds mean it has a greater opportunity to adapt to mammals and develop mammal-to-mammal transmission.

Worryingly, the virus has been detected in 1 in 5 samples of milk for sale in the US. This means that it is likely more widespread in herds than confirmed testing has shown. We do know that pasteurisation of milk inactivates the virus. However, raw milk might be problematic. While it’s not known whether humans could catch flu by drinking infected raw milk, studies had shown that cats and mice definitely can.

Are humans getting infected?

The short answer is yes, but not in great numbers. There were a handful of infections confirmed in humans across various countries in 2022 and 2023 - all from exposure to infected birds (mainly poultry). In 2024, almost all cases have been in the US with 58 confirmed infections in people. 56 of these were in farm workers at either poultry (21) or cattle farms (35), with zero evidence of human-to-human transmission and all 56 farm-related cases were mild infections.

The concern is that two recent US confirmed infections were in people with no known exposure to infected animals (one a child in California, the other a man in Missouri who was ill enough to need hospital). It has been confirmed this week that the California child’s strain was closely related to the strain circulating in US cattle. Alongside that, Canada recently reported a severe case in a child, also with no known exposure to infected animals. The Canadian child’s strain is genetically distinct from that infecting US cattle, so that is not a possible route of transmission.

What does this all mean?

It looks as if people infected through direct contact with infected animals are primarily infected through their eyes - this leads to distinctive conjunctivitis symptoms and might also explain the mildness of the disease (it mostly isn’t - thankfully - getting into their lungs).

It is also worrying to have three recent cases with no known source of exposure. This certainly doesn’t mean that human-to-human transmission has occurred, but as Prof Alan McNally says, unexplained transmission of any non-seasonal (i.e. new or unusual) virus is worrying.

The other worrying thing is the strain infecting the severely ill Canadian child because it has picked up a key new mutation that might help the virus attack the lungs - and make it easier to spread between people (because being in the lungs makes it more likely to spread through exhaled air). Essentially, there a few necessary mutational steps the virus needs to take to become pandemic, and this is one of them. Thankfully, the virus is still not good at transmitting between people (no one else in contact with the child has been infected) - and so hopefully this particular strain is a dead end. However, it is not good news that the virus has found mutations in this direction and might do so again.

Finally, recent research has shown that a single mutation in the US cattle H5N1 strain is enough to switch it to be more specific to humans than to birds. The good news is that this mutation has not yet been found in any of the sequenced cases, and, on its own, this mutation is not sufficient for rapid human-to-human spread. But it has been found in a mouse H5N1 sequence and it does mean that the number of mutational steps the virus needs to become a pandemic threat are perhaps fewer than hoped.

What about H5N1 vaccines?

There are already vaccines for H5 type flu, but very few are tailored to the circulating H5N1 strains (a new vaccine based on the latest strain is being produced now in the US). Of course, a strain that takes off in humans will be genetically different to current strains and will likely need an updated vaccine to be fully effective.

We know a lot about flu viruses and vaccines, and should be able to make an excellent vaccine quickly. But this is not a get out of jail free card - manufacturing, and then administering, tens or hundreds of millions of doses nationally and globally will take months, during which many millions could get infected.

What next?

Essentially, it’s a numbers game. The more mammals - and people - are infected, the higher the chance that the H5N1 virus will take the mutational steps it needs to spread easily between people. The conditions are as conducive to an emerging H5N1 pandemic as they’ve ever been. As Seema Lakdawala, a flu researcher, told Science Magazine “If H5 is ever going to be a pandemic, it’s going to be now”.

This means that surveillance and preparedness are crucial. Canada is testing its cattle farms, restricting livestock imports from the US and monitoring its wastewater for bird flu strains. The UK is actively testing poultry farms and culling birds where there are outbreaks (as there have been this week). There have been no confirmed infections in any UK cattle so far. The UK is also buying access to existing H5 vaccines in case a pandemic takes off, although these are a stopgap measure for use while pandemic specific vaccines are developed and deployed.

Given where the virus spreading, surveillance and action are particularly crucial in the US. The Centers for Disease Control (CDC) is testing cattle herds, people with suspected bird flu, waste water, and doing genetic testing as part of its routine annual flu surveillance. However, many have criticised the US response for slowness and a lack of transparency. For instance, mandatory nationwide testing of raw milk is only starting this month despite months of infections in dairy herds. Vanity Fair reported that private enterprise is conflicting with public health, delaying a coordinated response. When Donald Trump takes office in January 2025, we can expect reduced funding and power for the CDC, and infectious disease research in general, perhaps at the worst possible time.

Summary and Further Reading

Essentially, we can’t say how close a bird flu pandemic is because there is simply a lot of chance and unpredictability involved. At the moment, there is no evidence that the virus is spreading between people and the virus still has some genetic hoops to jump through to get to that stage. However, we can say that we are a lot closer to one than we have been in decades - uncomfortably close. In a world of rising vaccine scepticism, disinformation and populism, how confident can we be in an effective, coordinated, response?

Finally here are some useful and accesssible articles about the current situation if you want to read more.

Why hasn’t the bird flu pandemic started?, Kai Kupferschmidt, Science Magazine, 5 Dec 2024

Under certain conditions, a single H5N1 flu virus mutation could lead to spread among people, University of Minnesota, 6 December 2024.

H5N1 influenza: some troubling preliminary signals, Adam Kucharski, 27 Nov 2024

Why a teenager’s bird-flu infection is ringing alarm bells for scientists, Heidi Ledford, Nature, 20 November 2024

A Bird Flu Vaccine Might Come Too Late to Save Us from H5N1, Maggie Fox, Scientific American, 30 October 2024

What is the best way to stop a pandemic? Adam Kucharski, 22 July 2024

How fatal is H5N1 influenza? Adam Kucharski, 24 June 2024