Why “Winging It” on Surveillance Won’t Work for Bird Flu: A Conversation with Dr. Barney Graham


Editor’s note (2/12/2025): Since the beginning of 2024, the CDC has reported at least 68 human cases of bird flu in the U.S., including one death in an individual over age 65 with underlying conditions. The H5N1 virus has also spread beyond wild birds to include cats, racoons, sea mammals, and cows, with virus being detected in raw milk from infected cows. While the current risk of human-to-human avian flu transmission appears low, a dairy worker in Nevada was infected with a new strain of the virus this month, demonstrating that the pathogen continues to mutate, and prompting experts to call for increased surveillance.
With California declaring a state of emergency over concerns about the spread of H5N1 influenza (bird flu) and amid reports of “worrisome” mutations found in the bird flu virus that recently infected a hospitalized teen in Canada, we discussed the possibility of a bird flu pandemic with immunologist and virologist Barney Graham, MD, PhD.
Dr. Graham won the 2021 Albert B. Sabin Gold Medal for his pioneering leadership in vaccine design and pandemic preparedness as deputy director of the Vaccine Research Center and chief of the Viral Pathogenesis Laboratory at the National Institute of Allergy and Infectious Diseases (NIAID) at the National Institutes of Health. Now a professor of medicine and microbiology, biochemistry and immunology at the Morehouse School of Medicine in Atlanta and the founding director of their new David Satcher Global Health Equity Institute, he is keenly focused on growing global public health leaders and global preparedness for pandemics because, as he says, “viruses don’t respect borders.”
Why is influenza always mentioned as one of the next possible pandemics?
The influenza virus caused one of the great pandemics of the modern era. The 1918 influenza epidemic killed upwards of 50 million people, and that was when the population was much smaller. So, it made a huge impact and probably caused even more deaths than World War I.
Influenza virus has the potential to cause pandemics like that because of the way it replicates. The virus has a segmented genome, so it can make small and large changes in its surface proteins which are the targets of protective antibodies. We saw smaller influenza pandemics again in 1958, 1967, and a scare in 1975 and then another one in 2009.
So even with anti-virals and vaccines, we’re still at risk of an influenza pandemic?
Influenza has adapted very well to humans and animals and has ways of evading immunity either by “drifting” — one mutation at a time — or by “shifting”, which means you’ve replaced an entire hemagglutinin or neuraminidase by reassortment of its segmented genome. (Editor’s note: Hemagglutinins are glycoproteins present on viruses that bind to receptors on cells, which leads to infection. Neuraminidases are proteins with enzymatic activity and help influenza viruses enter and exit a cell. Different subtypes of influenza viruses are designated by H and N numbers.)
Bird flu has been around for more than 25 years, and isn’t easily transmitted from human to human. Should we be concerned about bird flu becoming a pandemic?
I think we need to be vigilant, and we need to be prepared. There are 18 different influenza hemagglutinins and 9 different neuraminidase molecules that combine to make a large number of possible influenza virus subtypes carried by migratory birds all over the world. These include the typical seasonal viruses humans are often infected with (e.g, H1N1 or “Swine flu” and H3N2). Migratory birds also carry viruses like H5N1, or viruses with H7 or H10 combined with a variety of neuraminidase subtypes that can infect domestic birds and sometimes infect humans. Even though it’s relatively infrequent, there have been hundreds of humans infected with H5N1 influenza virus over these last 25 years. And unfortunately, the mortality rate is quite high, in the 40 to 50 % range.
It would be devastating if H5N1 influenza did adapt to grow better in humans. Right now, it prefers a sialic acid receptor with an alpha 2,3 linkage instead of an alpha 2,6 linkage more common in humans. However, it only takes a few mutations in the hemagglutinin to improve binding to that different type of sialic acid, where it would more easily infect humans.
It hasn’t happened, and I’m hoping that it’s unlikely to happen. But now H5N1 is not just in birds, which have mostly the alpha 2,3 sialic acid, but it’s in mammals that have a mixture. The more viruses have a chance to adapt to mammalian host, the more likely it is to adapt to humans. Therefore, we should be vigilant and strive for better surveillance and better readiness.
Many are calling for improved surveillance of bird flu cases and potential hot spots for the disease in the US. In Britain, they are now testing most farm workers and hospital patients who test negative for some of the other influenzas. Do you think we need that type of testing here in the US?
I’m not involved in the surveillance part of this work, so I don’t know exactly what’s happening right now. Our capabilities for surveillance are extraordinary because of our ability to do high-throughput sequencing and sequencing of large amounts of material. We have the capability to do exquisite testing and exquisite surveillance, as people witnessed during the coronavirus pandemic.
It’s a matter of willpower and organization and coordination between agencies, between the USDA, the NIH, the CDC, and others.
So, if needed, could we ramp up very quickly and have a pretty efficient system for identifying and then targeting bird flu?
Well, I think we’ve already reached the point where we need to ramp up and have intense surveillance.
Is there something about this bird flu outbreak that is different compared to other influenza outbreaks?
The difference in this outbreak that we’ve known about — at least since May 2024 — is that it’s in cattle. Nobody expected cattle to be infected by an influenza A type virus, in this case H5N1, which is a virus typically found in birds because it prefers binding to the alpha 2,3 sialic acid receptor. It hadn’t been seen before in cattle, and cattle are mammals. The cattle are infecting other mammals that are around farms, like domestic cats and occasionally humans.
The virus right now seems to be restricted mostly to the mammary tissue of the cattle. Farm workers have been infected, especially in their eyes. There was a recent case in Canada in a teenager where there was severe illness, and a worker exposed to birds in this country who got H5N1 and died.
What hasn’t happened yet is that H5N1 isn’t spreading from human to human. Until the virus learns how to spread from human to human, it doesn’t pose much of a risk for a pandemic. But that could happen quickly. If it happens and it starts spreading, the situation could change quickly, as we saw during the coronavirus pandemic.
I think now that it’s in mammals, it’s adapting in different ways than it would if it’s just growing in birds. We don’t really know what to expect in terms of how the virus is going to adapt in this mammalian tissue over time.
That definitely helps explain why there isn’t a lot of agreement on what to do, because if this hasn’t occurred before, it means all of this is new for everybody.
Yes. We know now that it’s in our milk supply. You can find some of the genetic remnants of the virus. The virus is not alive in our milk supply because it’s pasteurized milk. But the virus remnants can be found there.
It reinforces the idea of how important pasteurization is, because pasteurization will kill the virus. If you’re drinking raw milk, you’re going to be exposed to live virus, at least to some extent. I don’t think you want to be exposing yourself to live H5N1 on purpose.
The other thing I’m worried about is, we know that this virus can infect alpacas and llamas, which means it can probably infect camels. If it’s infecting these types of animals and it gets into places in low-income countries, or even Saudi Arabia, which is a relatively high-income country, where people drink a lot of raw milk, I’m concerned about what that could lead to.
How easy will it be to produce vaccines for H5N1?
We already have mechanisms to make influenza vaccines. We’ve had those since the 1940s, actually. We can make effective influenza vaccines, especially when they’re matched to the exact virus. The seasonal vaccines that we have are based on prediction of what will be the next virus and not always matched exactly to the virus that’s circulating.
But if you can match the vaccine to the virus, the influenza vaccines that we already know how to make can work. The problem with those vaccines is that most of them are made in eggs. And the world supply of eggs to rapidly make influenza vaccines is probably around a billion. But there’s seven billion people on Earth.
So, we don’t really have the capacity to scale up influenza vaccines for everyone. And the process takes four to five months to make the vaccine. It’s not as fast as some of the modern technologies we have.
For making mRNA vaccines, which is the type of vaccine used in COVID-19 and is probably our fastest vaccine approach, vaccines can be made in a matter of a few weeks and then scaled up rapidly. But right now, we haven’t proven that mRNA influenza vaccines could work on a large scale. We know that they can make antibody responses that we think will protect. But because we already have licensed influenza vaccines, it hasn’t been tested broadly for actual efficacy in field testing. So, in that sense, we’re not completely prepared to use some of the newer technologies for vaccines if we had a large outbreak. But we do know how to make influenza vaccines.
Even the vaccines that were made in 2005, when there was another scare about avian influenza, even those vaccines that are not perfectly matched to the current strain could have some level of protection against the current virus. But there’s not very many doses of those vaccines available.
Ever since COVID-19, there has been increased skepticism around vaccines. What effect would this have in another pandemic, and how would we convince people that a vaccine would be useful or necessary?
There’s a lot of vaccine hesitancy now. There’s always been some vaccine hesitancy going back to 1798 when Edward Jenner introduced this idea of vaccination by using cowpox to “vaccinate” against smallpox. But the track record of vaccines speaks for itself. Albert Sabin, for instance, developed a polio vaccine, a live attenuated polio vaccine that helped us almost eliminate polio, and backed up with the Salk vaccine, which is still being used. Those vaccines for polio have almost eliminated polio except in places where there’s ongoing war and conflict. The track record of vaccines, if you look at the data, is indisputable.
The Expanded Program for Immunization (EPI) from the World Health Organization just published a 50-year review of its program, going back to 1974 when it started. It estimates that over 150 million lives have been saved with just the handful of vaccines that are used in the EPI program. Not to mention the effectiveness of other vaccines that EPI doesn’t necessarily use.
During the COVID outbreak, it’s estimated that as many as three million American lives were saved. And if you count excess deaths just in 2021, it’s estimated that up to 19 million lives were saved. So, if you look at all that evidence, the effectiveness of vaccines is beyond reproach.
The sad part about vaccine hesitancy is that, in the people who didn’t take vaccines for COVID after they were available, it’s estimated that there were more than 200,000 unnecessary deaths in the United States. If you look at the data, it’s clear that vaccines improve overall health.
Vaccines do have side effects and that’s also indisputable, but the consequences of vaccine side effects and the consequences of uncontrolled infections are not even comparable.
If you were advising the U.S. government at this point, what would you say our next steps should be for bird flu?
What I’d like to see happen is that surveillance intensity is increased. I think that the general population should be aware, but I don’t think they should worry about this. We’ve had H5N1 circulating around the world for 25 years, and it has infected hundreds of people, but not thousands and not millions. It’s still a relatively rare infection and it doesn’t spread easily from human to human. I think it’s good to be aware of it. I think it’s good to not drink raw milk. But I don’t think the general public should worry about it as a cause of a pandemic in the near future. That should be the concern of public health officials.
I do think we need to do some work on our vaccine supply and not necessarily to stockpile vaccines, but to be prepared to use some of the more modern technologies to scale up if needed, and to use the old technologies which we know work but also be prepared by proving that the new technology vaccines work for influenza. I think that needs to be proven in humans so that we could depend on them in the case of a new outbreak.
That sounds very hopeful.
Something people should keep in mind is that we have such extraordinary technologies now. The things that we can do and measure with such precision and understand in such detail is unlike anything we had even 10 or 15 years ago. We have the technology to deal with this in a proactive way.
We have the technology to deal with most emerging threats. Not all, because some problems still need breakthroughs. Therefore, we have to maintain investment in basic research. Having the basic research, knowing the atomic level structures, knowing the replication process and tropism and things that drive mutation, these kind of basic research questions need to be known ahead of time in order to prepare for potential outbreaks and respond to actual outbreaks. Supporting basic research is an underappreciated investment that pays great dividends.
Recommended for you

We make vaccines more accessible, enable innovation and expand immunization across the globe.