Hybrid DC Roundtable: The CDC Response to the Spread of H5N1 Avian Flu in Dairy Cows

Wednesday, May 1, 2024
Aaron Josefczyk/REUTERS
Nirav Shah

Principal Deputy Director, Centers for Disease Control and Prevention


Bloomberg Chair in Global Health; Senior Fellow for International Economics, Law, and Development; and Director of the Global Health Program, Council on Foreign Relations

Roundtable Series on Global Health, Economics, and Development

An outbreak of H5N1 avian influenza that was detected for the first time in a milking herd of cattle in Texas one month ago has now infected thirty-three herds in eight states and at least one farm worker, spurring alarm among some experts that human-to-human transmission could be next. Please join us for a discussion with Dr. Nirav D. Shah, Principal Deputy Director of the U.S Centers for Disease Control and Prevention, on the U.S. response to this avian flu outbreak and on how the CDC and its U.S. government counterparts are applying lessons from COVID-19 to respond to the potential threat.

Thomas Bollyky  00:00

Welcome. I'm Tom Bollyky, the director of the Global Health Program here at the Council on Foreign Relations. This is an event on the CDC response to the spread of H5N1, the avian influenza, in dairy cows.


Now, H5N1 avian influenza is not new. We've been tracking this virus since it was first identified in 1996 and we've been monitoring it for almost that long. It's spread among migratory birds and in poultry. Yet the discovery of this virus in March in dairy cows in a Texas Panhandle farm, and its subsequent spread to 36 herds in nine states, has alarmed some experts and some members of the public. Why? There are at least four reasons: First, in poultry, H5N1 avian influenza has been lethal and the numbers in people aren't that much better. Of the close to 900 people who have been confirmed to have been infected since 1996, a little over half, 52%, have died as a result of that infection.


Now, most of those cases are prior to 2015. And we frankly have no idea what the true number of infections were so that rate may well be lower. But we know enough to know that this virus is serious. In the current outbreak, I should also emphasize that we only have one confirmed infection and a dairy worker identified on April 1. But there have been many anecdotal reports of other farm workers with conjunctivitis and mild respiratory symptoms; the same symptoms that the confirmed case had. The second reason people are concerned is that the spread of this virus to mammals may mean that it will spread to other mammals, and in doing so may change in ways that make it easier to infect humans and ultimately transmit from humans to humans. We already know that there are numerous farm cats who have been infected and died from consuming the raw milk of infected cattle during this outbreak.


The third reason for concern is that the dairy farm context for an outbreak of this virus is novel. We don't know much about how it is spreading. We don't know much about the human animal interface in this context, the commercial context, and the details of the dairy worker population, many of whom are low wage workers, and non-English speaking [there's a great op-ed today by Erin Sorrell in the New York Times on these points]. That creates complications for testing and surveillance.

The fourth reason is that the COVID-19 experience is still quite fresh in the minds of the public, policymakers and politicians. We are starting from a fundamentally different place in terms of public support for measures such as masking, testing, isolation, and vaccination than we were in 2020. There is one similarity, of course, with COVID-19 and its emergence which is that we are once again in a presidential election year.


We have the perfect person to help us understand this current situation. Dr. Nirav Shah is the principal deputy director at the US Centers for Disease Control Prevention. Prior to this role, he had served as the director of the Maine Center for Disease Control and Prevention, leading the state response during COVID-19. In that context, Dr. Shah was well regarded for his open and transparent public engagement. And it is in that same spirit that he is here today.


I welcome you, and I'm grateful for the work that you and your colleagues are doing at this current time and grateful for you to make time to be here.


Dr. Shah and I will discuss the current situation. But the title for this event is purposeful. We want to focus mostly on what the US and CDC’s responses to this outbreak are. What is the agency currently doing? What does the road look like going forward? What will the triggers be for escalating its response? And what are we learning from this latest outbreak that will inform the US government response to future emerging infections and outbreaks? We have a large audience, especially online, and we will try to reserve as much time as possible for Q&A. Today's session is on the record. The video will be posted on


And with that, let me turn it to Dr. Shah. I would be grateful if you could tell us a bit about the current update on the outbreak, how widespread infections among cows are and within other animal species? What do we know about how it's spreading? And the same questions in terms of animal to human transmission.


Nirav Shah:

Great. Well, first, good afternoon, everyone. Thank you all for the invitation to join and thank you to each of you for taking time out of your days to join us today. Thanks to Tom and our colleagues at the CFR for the invitation and for hosting the event. As Tom mentioned, today's discussion is on the record. That said, I'm really going to be focused on what CDC’s responses, though, I acknowledge that there will be other questions, and I'm happy to discuss those. I want to note that there are multiple other agencies that are involved with this response: the USDA, the FDA, other colleagues within HHS, and so where appropriate, I may punt some of those questions to them. But I can at least sketch out the broad strokes of what's happening. The other provision right up at the top is what we are dealing with and talking about today is a bullet train. And as with any major outbreak, it is moving at the speed of a bullet train. What we'll be talking about is at best a snapshot of that very fast-moving train. So, all the information that I'll be sharing and conveying is current as of this minute. But if you happen to be watching this video tonight, let alone next week, let alone next month, please recognize that what we're talking about is only current as of the current minute right now.


Here's what's going on at a high level from a sit-rep perspective. In late March, March 25 to be exact, USDA announced that there had been detections of H5N1 among cattle, initially just a few herds in the northern Texas panhandle region. This was notable as Tom outlined for a few reasons, one of which is that this is the first time that H5N1 had been detected in cows. It was detected in mammals up until that point, but the first time in cattle, that is notable for two interrelated reasons. One, because being a food animal raises the likelihood of disruption. Secondly, there is a much tighter human cow interface than with say foxes or sea lions, which had been the other mammals in which H5N1 had been detected up until that point. So as a result of that, the apparatuses of the US government swung into gear to try to get a better sense of what was happening. Soon thereafter, we became aware of reports of workers on one particular farm in northern Texas that had been experiencing symptoms. We worked with the state of Texas health department, and were able to obtain testing of the employees on that farm.


A couple of days later, on April 1, the CDC announced that there had indeed been a human case that had been detected as a result of us going out and looking for it. Subsequent to that time, additional herds have tested positive and as of this morning, there have been 36 total herds in nine separate states that have tested positive, the bulk of those herds have been in Texas and in New Mexico. But we've also seen cases in herds, ranging from as far afield as Michigan, North Carolina, and most recently in Colorado. We can talk more in the Q&A about theories as to how H5N1 is spreading among cattle. But for right now, we have detected it in these 36 herds. We anticipate subsequent herds that we find to test positive, particularly because of changes that USDA announced last week around requirements for testing cattle before they move in interstate commerce. That's the big picture as to what's happening on the animal front.


As I mentioned, on the human front, there has been one case, a case in a dairy worker in northern Texas. This marks just the second case of H5N1 in United States, the first happening two years ago, and an individual from Colorado. That just gives you a sense of how rare these particular cases are. That said, the team at the CDC has taken a very close look at the genetic makeup of the virus that was recovered from this individual case in Texas and compared it against the other viruses that are known to be circulating around the globe, as well as in the United States. We do that for a few reasons. One of which is to determine whether there are any salient mutations that would enhance the likelihood that this particular virus could jump from human to human. Reassuringly, we have not detected any such mutations.


The other reason we take out our magnifying glass and take a closer look at the genome is to understand how the current armamentarium of vaccines and therapeutics may be effective against the virus that's recovered. There too, the analysis was reassuring, which is to say the building blocks of the vaccines that we already have, as well as the medications that we already have appear to retain their effectiveness even against this way to strain of virus. For those reasons, CDCs overall risk assessment of this virus to the general public has not changed. It remains a low risk to the general public at this time. However, within any risk assessment, we always know that there are individuals on that spectrum for whom the risk is higher. So, another area of intense CDC focus has been on those individuals who are at a higher risk, namely dairy farm workers, but more generally, livestock workers as the first category. The other category is individuals who drink or consume raw milk, which is a smaller group, but one that we are particularly focused on right now.


That's where things stand on the sit-rep perspective. Tom, I'll turn it back to you.


Thomas Bollyky:

Great, thank you so much. That's a really helpful overview. I'm going to focus mostly for this initial conversation on the animal to human transmission given your remit. When you mentioned that we've had one confirmed infection, can you say a bit about how many workers have been tested so far? What are we doing in terms of farms where infections have been identified? Are we monitoring those workers? And if so, for how long are they being monitored presently?


Nirav Shah:

Regarding the one who, as I mentioned, was confirmed on April 1 from Texas, that individual had mild symptoms. Their symptoms consisted almost entirely of redness in their eye, conjunctivitis, and thankfully, they did not have the more classical respiratory symptoms that are often associated with influenza. That said, that is a risk, let me be very, very clear and a risk that we are on the lookout for. I can talk later about the surveillance systems that we have that help us scan for that. But as it relates to this individual, their symptoms were thankfully mild and resolved after several days. They were provided one of the medications that I mentioned, as were all of the household contacts of that individual, all of whom were hand delivered this medication by the local health department on the scene or by someone who was affiliated with that health department. In terms of the testing architecture, I think it's important to note how testing for H5N1 is arrayed as a multistage process.


It starts with somebody going to their local health clinic and saying I've got a cough, fever and muscle aches and so on and so forth. Any clinician would take that as a possibility that influenza generally may be in the picture, generally not H5N1, but just the large family of flu. In most offices, most clinicians in the country can do a rapid or very quick test for influenza. That happens thousands of times a day across the US. If that test is negative, then the clinician moves on to other potential diagnoses. The CDC does not even get word of that as it is such a common occurrence. It's no different from maybe taking your kid and wondering if they've got strep or red eye. Those tests happen every day, and if they're negative, the clinician just moves on.


However, if that test is positive for influenza, generally, a succession of events is set into motion, particularly if that individual is a farm worker. Eventually, those tests, if they are of sufficient concern, find their way to the CDC: if they tested positive at the doctor's office, tested positive at the state laboratory, were confirmed at the state laboratory. Then those specimens come to CDC. I say all that because that's the tip of the iceberg that comes to CDC. Thus far, we have received 23 such tests for final confirmation, and all but one of those was negative, which is the one case that we're aware of.


But the public health response does not stop with just testing. It entails much more than testing. It entails, as Tom noted, active monitoring or monitoring of individuals who have been exposed. It's not just enough that you're exposed that gets you a test. It's that you're exposed and you're starting to exhibit some degree of symptoms. Up until that point, public health is very engaged. One way we are engaged is again with actively monitoring folks for symptoms. Right now, there are approximately 150 people across farms across the United States that are being actively or passively monitored, most often by their state health department, not by the CDC, but by their state health department.


The public health response doesn't just entail that, as you as you heard me allude to, we're also where possible on the farm, providing guidance to farm owners providing guidance to people who are exposed, and where appropriate, providing them medication to reduce the likelihood that they would go on to develop disease if they've actually been exposed. A lot is going on, but I wanted to walk through what the cadence and the rhythm of the public health responses in these situations has been.


Thomas Bollyky:

Can you say more about those 150 people being monitored? How are they selected? Are they people on those farms? How do we triage that?


Nirav Shah:

Great question, Tom. As to the selection criteria, what gets you in the door for monitoring is that you've been exposed. What constitutes exposure? That's a question I get a lot from epidemiologists at state health departments. We had a worker, hypothetically, who walked by the cows 10 times in and out of the barn. Is that an exposure? There are all these ways to answer or ask these questions. But the default position that I share with state health departments is it's better to over monitor.


By monitoring, what we're really looking for is a daily check in with the exposed worker. Sometimes it happens by text message, sometimes it happens by a phone call. In some instances, it might happen from a daily visit from a county health department to see if they've developed symptoms, to see how they're feeling, to see if anyone in their household has symptoms. If the answer to any of those is a maybe, then immediately the testing apparatus has sprung into being. That’s the nature of the monitoring, that's active monitoring. In some cases, it's more passive, which I don't love but that's the system we've got. I'd love if everything was more active. But passive monitoring is the same thing, but the default is flipped. In passive monitoring, the baseline is let us know if you have symptoms. As you can tell from my comments, I'm more of a proponent of active monitoring, because that daily interaction with the health department is useful.


Thomas Bollyky:

As I mentioned in the introduction, this is a different context, of course, and you mentioned the challenge around dairy workers and farm workers. It's a different commercial context as well from poultry, where we have systems set up to address the economic incentives for farms to cooperate, and much more established pathways for this. What is CDC doing to bridge this divide, potentially leverage other agencies, either state or local, that address occupational issues? How is this being done?


Nirav Shah:

Great, that's very helpful. We have been working with a number of different agencies, and I'd be remiss if I didn't note that we've worked very closely in particular with USDA. Undersecretary Jennifer Moffitt and I are on the phone multiple times a day on these issues, as well as our colleagues within CDC: NIOSH, for example, but in addition to that OSHA, the Department of Labor, and then their relevant counterparts at the state level. Let me say that working with farms requires a deft touch. Farm owners may be wholly skeptical of this entire enterprise, and I think we at CDC have to respect that view. If you look at this from the perspective of a farm owner, this may not be the situation that we at the CDC think it is.


From a farm workers perspective, some of their cattle, approximately 10% of their herd have been ill, many of the cattle were transiently ill, they were taken out of milk production, but about seven days or so later, they were able to resume. And for the most part, we've only seen this one case. I say that not too agree with that perspective, but to acknowledge that from a CDC perspective, we've got to start our advocacy from that vantage point. We can't just discount their rationale, no one's no one's ever convinced of that. We've all been at dinner parties with that person that treats every interaction like a high school debate that has never been successful at changing anyone's mind. So, I've been approaching these conversations with farm owners from the vantage point where I see what they're saying, I see what they're seeing. Let me see if I can present an alternative perspective. That said, we have a long way to go with respect to getting on to farms. That's a top priority for us at the CDC, because we'd like more insight into ways we can better answer some of these tough questions. What constitutes an exposure? What are the risk factors that make someone more susceptible to illness rather than less susceptible? What do we actually know about the utility of our testing regimen? We'd like to have more data and that's a top priority for us right now.


Thomas Bollyky:

Let me pivot to countermeasures. You mentioned that countermeasures are potentially available to address this crisis. Can you tell us a little bit about their suitability, and our current stocks and our ability to scale them up? Is this for diagnostics, are we talking point of care diagnostics for antivirals, what kind of stocks do we have of the antivirals that might make a difference for this vaccine? Is it a single dose or a two dose? What kind of supplies do we have presently?



Nirav Shah:
Great. Here too, I want to give a big shout out to my colleagues at ASPR, which is an operational division within HHS. I work very closely again on a daily basis with their head of infectious disease and response Dr. David Boucher, as well as their administrator. We are on the phone all the time and they really are the primary actor with respect to countermeasures. CDC is a is a big player, but ASPR is the principal there.


Let's start with medical countermeasures, particularly antivirals. There are a number of antivirals, although the one that a lot of folks have heard of is one called Tamiflu. The US government has for years been stockpiling it at the federal level, and then many states have their own stockpile. There are many millions of courses available. The number is known, but it's not a number that is shareable more publicly, but there are many millions of courses of Tamiflu that are available. And as I mentioned a moment ago, one of the first things that CDC scientists did when we recovered the virus and isolated it from this patient in Texas, was to see how it performed against Tamiflu (a key question as you can imagine). Thankfully, the news there was reassuring. Tamiflu appears to retain its effectiveness. That said, the effectiveness of Tamiflu is something we all wish was better than what it is. That's where we stand with respect to medications.


Now, the other the other area that that folks are focused on right now is vaccines. I want to state once again, we are not at a point where vaccination is something that we are moving out with. We can talk about escalation triggers, as you hinted at Tom, but I don't want anyone to think that because I am discussing vaccine planning that we are moving out with vaccine planning, just to be very clear. That said, the way that we do vaccines for influenza in partnership with ASPR is that CDC creates what are called candidate vaccine virus (CVV). These are functionally the building blocks, the components, that could if needed, and we are not there, but that could if needed be taken and then scaled up from a manufacturing perspective.


We have a library of the CVVs that we are continually testing. Probably the other first thing that our team did was take the virus that was isolated from the patient in Texas and run it against the performance of these two candidate vaccine viruses. And there too, the news was reassuring.


That leaves us in a strong position of if we were to need to scale up the manufacture of these vaccines. It still takes time, which takes me to my third point, which is that there are already some of these vaccines on the shelves. We wouldn't be starting from ground zero if we needed to scale up. We already have the blueprint. It would be largely a process of taking the blueprint and putting it into place.

To preempt a question that folks may be saying, why don't you just do that now? Right? You got the vaccine building blocks, you just told us they seem to work? Why doesn't the US government just go out and make 300 million vaccines and just have them on the shelves? Well, it's a great question that I want to address really briefly. The first is that the virus is always changing. If we made 300 million doses, the virus could shift in a month from now as we've seen with COVID. We have to be careful about over supplying because we may be producing something that is soon out of date.


The other reason is that the vaccine production system in the US for novel influenzas, which this is, runs headlong into the production system for seasonal influenza, the flu shot that I hope everyone gets on a yearly basis. It's my little plug for public health messaging. The tradeoff between moving out and producing a novel influenza vaccine compromise or takes away from our capacity to produce the seasonal flu shot. It’s a very difficult policy choice as to when you flip the switch. When do you pause production of the seasonal flu shot, which is happening right now, and shift over to the production of a novel flu shot?


If you get that wrong, if you overestimate the threat of the novel flu, you're shortchanging the availability of vaccine come the fall. Even in an average year, the regular flu can hospitalize up to 60 million people and can kill over 75,000. This is a trade off with very real consequences and it's one that we take very seriously.


Thomas Bollyky

Okay, I want to talk you for about another 10 minutes, and then I'll turn to Q&A just so that people know what's coming. I want to ask one more question about our current response and then pivot to the road ahead. In terms of the current response, how do you think the lessons of COVID-19 for the CDC as an agency have influenced how the agency has been going about its response?


Nirav Shah

If you all would permit me to just wax philosophical for a quick second, because I want to share how I think about these things and how I believe we've tried to pivot at CDC. There is this thought exercise that I think is best attributed to Jared Diamond, from an essay from 20 plus years ago preceding his book Collapse. The essay is about Easter Island, and at least Jared Diamond's hypothesis as to why Easter Island’s entire community collapsed. Diamond suggests that it had to do with the destruction of all the fauna, and of all the flora on the island, particularly the trees. He posits that by removing all the trees off the island so they could have logs with which to roll the giant rocks that constitute their statues, that precipitated an ecological collapse. Now, that hypothesis has been hotly debated and that's not my purview.


But in this essay, Diamond poses a fascinating question that animates a lot of my thinking, which is, what were the people who cut down the last tree on Easter Island saying when they cut that last tree down? Were they saying, ‘no big deal, technology, we'll come up with a solution?’ Were they saying ‘God will provide?’ Were they saying ‘this is going to be fine, there's not a problem here.’


That thought exercise is always in my mind when I'm thinking of big decisions because it's easy to make a decision, but it's really difficult to make a decision anticipating what the unintended consequences would be. One thing that we've tried to do at CDC much more deliberately is to think through the unintended consequences of decisions that we may make. There are folks who have said, ‘Hey, why don't you just do X?’ I understand that, and I respect that question, but we are always thinking of what the potential unintended consequences might be, in the same way that those Easter Islanders probably were not thinking about that when they cut down that last tree. That mindset of what are we saying right now? What mistake might we be committing that people 10 years from now we're going to say, what were they thinking? What was CDC doing when it did X? Trying to think a lot more long term is one very concrete thing.


The other, more tactical, is with respect to communications. One area that public health I don't think did what it should have done writ large during COVID, was to signal that what we are doing is fundamentally a scientific enterprise. That as a result of that, what we know today will necessarily be different from what we know week in a month from now and to signal that at the outset. I think we gave people a sense during COVID that what we knew was settled, and we often used words to that effect. I myself was guilty of doing that, rather than saying this is based on what we know today, but what you the American public are seeing is public health and action. You are seeing science in action. As a result of that, a corollary of that is that what we know today will change and thus, our recommendations may change as a result of that. Saying that now is a lot more useful than trying to explain why something has changed after the fact, and I hope you've seen CDC be more circumspect in the way that we've talked about the future of what we might know.


Thomas Bollyky

I think that's a fabulous anecdote, but what it assumes on Jared Diamond's part was that the same was one of commission. That it was an act to cut down the tree as opposed to the population of trees falling below a certain level where it couldn't sustain itself, a sin of omission. I wonder with that in mind, there are many areas in the COVID pandemic where we committed sins of omission and were slow to respond. Specifically, in relation to this, how quickly, if this were virus were to evolve to transmit efficiently, how quickly would a pandemic ensue? What is the ramp of this look like? What is that risk of omission?


Nirav Shah:

Yes, that's exactly my question and one that we internally, within the CDC and across the US government, we talk about a lot. The risk here of something going from one or two sporadic cases to becoming something of international concern, it's not insignificant. Let me talk about why that is, but then also at the end provide some factors that are reassuring.


On the non-insignificant side, we've all seen how a virus can spread around the globe, before public health has even had a chance to get its shoes on. That's a risk and one that we have to be mindful of. Indeed, at CDC, we think about the risks that we're facing, with respect to infectious disease threats, and pandemic influenza is at the top of that list. In particular, with avian influenza, it is particularly salient because of some of the factors I noted: we're now in a food animal and an animal where there's a distinct animal human interface. But what's also concerning and what would precipitate greater spread is if the virus were to find itself into the pig population that, for various biological reasons, are a perfect of vessel through which an even more virulent strain could emerge. Those are the factors that are of concern and why I'm here today and why the CDC is working on this 24/7 as well as with our other federal partners who I mentioned.


There are some reassuring factors along those lines, though. The first is, I think it's important to distinguish our flu architecture from our COVID architecture. We are where we are with respect to influenza because of 20 plus years of sustained investment, specifically around problems of this nature, specifically around pandemic influenza. You heard me tick through the testing platforms we have, the diagnostics, the countermeasures, the vaccines, the entirety of the base of knowledge that we have thousands of publications. All of that is because of decades of sustained investment that puts us in a better position. That said, we need more sustained investment so we can continue that posture.


We know more going into this than we did four years ago with coronavirus. The other thing is that although we have not, thus far, seen evidence of this particular strain suggestive of an enhanced ability of the virus to transmit between animals to humans, let alone among humans, that is to your earlier question about escalation points, something we are very much looking at.


Thomas Bollyky:

You anticipated my next question, which is about escalation points. We didn't get into details of diagnostics, but I understand that the analysis is still happening at regional and state labs, a situation that proved problematic in the COVID pandemic in terms of delays. You mentioned the availability of Tamiflu in stocks of that, but ideally, you preposition it to a point where you could be able to distribute it. You mentioned this difficult trade off on vaccines and when do we start producing? What are the escalating indicators that you are looking for, for CDC to partner with commercial diagnostics companies to make more suitable tests that could more quickly provide an answer to mobilize this distribution and more production of antivirals, to make this difficult call in favor of producing a vaccine that might make a difference for H5N1.


Nirav Shah

Let me start by going into some of the epidemiological pieces and then I'll go very tactically through each of the axes you mentioned. From an epidemiological perspective, what we're looking for are two layers of data. Some of them are what you would expect if we see more cases, if we see more severe cases. Those are not surprising to anybody and those are the things that we're constantly scanning for. There's also on the animal health side, we're thinking through what the progression of H5 into more and different mammals means. As I mentioned, starting about two years ago, H5 started moving its way through different mammal species: foxes, raccoons, sea lions, goats, bottlenose dolphins earlier in the week. We're thinking through what the implications of that are from a human health perspective.


There are also some nuances there around what the composition of any subsequent human cases look like. Epidemiologists look not just at the numbers, but the pattern. If we saw a cluster all in one place, that takes us down one strategy, maybe a Tamiflu-only strategy. If we saw what are called unlinked chains of transmission, cases popping up everywhere, but maybe only one or two subsequent cases thereafter, that's what's called a studded chain of transmission that takes us down a slightly different pathway. What activates all of our systems, however, make no mistake, is what we refer to as sustained human to human transmission. Rather than sporadic clustered cases, sustained cases, going from person to person to person rather than dying out after what's called one generation. That takes us to our highest level of activation.


What does that mean for some of the axes that you noted with respect to countermeasures like Tamiflu? Tom, you suggested prepositioning. I'm pleased to report that's already been done. Tamiflu has long lived at state health departments. And even more recently, our colleagues at ASPR have opened up the doors and said to states if you need more, let us know now. Likewise, with things like PPE, where we've said the same thing, and I'm looking at ways where we can more intensively protect workers by furnishing and equipping them with PPE.


With respect to diagnostics, I'm glad you mentioned commercial laboratories. A key part of our diagnostics array is not just the labs at the state health department and at the CDC, but it's the labs that each and every one of you knows. It's Quest, LabCorp and ARUP and Mayo and BRL, where you get your cholesterol checked, that's where people go to access health care. For years now, but most definitely in the aftermath of COVID, our office of laboratory science has had warm handoff relationships with the biggest diagnostic equipment manufacturers, Roche, Cepheid, QIAGEN, as well as with the laboratories themselves. Many of those labs actually have our test already and so if we needed to scale up, that's something that they could do in house shortly. I don't want to promise what it would be, but again, not starting from scratch.


Then finally on vaccines. The question of when to scale up vaccines uses some of the same epidemiological parameters that I mentioned. It's not simply the number of cases, although that is critical. It's the composition of those cases. If we saw a handful of cases in a very limited geographic area, maybe Tamiflu is the better strategy. It's faster to get to them, so on and so forth. However, if we saw a different complexion that takes us down a different pathway.


Thomas Bollyky:

I see some people queuing online already, people should do that in the audience as well. Turn up your placard and I will call you in the order that I see you. One last question. We talked about how COVID has influenced the agency's response. What do you think the early lessons of this response are for what CDC does in the future, either around influenza or other emerging threats?


Nirav Shah:

I think the lesson that I will take away and that I think my colleagues from CDC will take away is that from a coordination perspective, you can never be too coordinated. I'll that that similarly with respect to communications. An old adage in the crisis communications world is just at the time you are sick and tired of saying something, that is just the time in which the public is starting to internalize the message that you are providing. I don't think we did as a public health system did a great job of that during COVID, and one of the reasons I'm here today and on the record today, is to try to make sure we're doing a more consistent, forward-leaning job on that latter piece on communications.


Thomas Bollyky:

Terrific. At long last, I will turn it over to all of you. When you ask a question, please make it sound like a question, have it be short. We have a lot of people who want to speak, so I'm going to try to get through as many as we can. Please state your name and affiliation, and I'm going to take two at a time to start with David Lim and then Lena Sun.


Q: Thanks for taking my question. David Lim with Politico.


You mentioned how CDC has received 23 tests for reference testing from public health labs. First, how many tests have been run at the state public health lab level? How many tests haven't actually been referenced to CDC? Could you talk a little bit about some of the challenges that CDC has had getting onto farms and have you been working with USDA to get onto those farms? Has there been any pushback from USDA officials as to kind of those efforts to survey farms?


Thomas Bollyky 

We're going to take two at a time.


Q: This has to do with vaccinations. We know there are hundreds of thousands of prefilled syringes in the SNS, Dawn O’Connell has talked about it and I know that you've said that we're not at that activation stage. But given that we're not really testing people as much as we would like, and we're not testing cows as much as we would like, is there any thought given to maybe using those vaccinations in a targeted way against the folks most at high risk, so the farm workers and the processing plant workers, such truck drivers who track the milk, a limited pool?


Nirav Shah:

Very good, thank you for both of those questions. I'm going to take David's and I'm going to try to go in reverse order from memory, but let me know if I missed anything.


There has been no resistance on the part of our USDA colleagues from our efforts at CDC to get on the farms, zero whatsoever. They have been shoulder to shoulder with us, they share that same view that we need to take a One Health approach to this situation. The delineation between animal health and human health is in some sense artificial and they see our mission as intertwined and aligned with theirs. We've worked with farms, and I don't want to castigate them, because they themselves are facing a lot of challenges, as are the workers themselves. I've said this before on the record, which is that we need to be sensitive to the concerns of workers, workers who may be deeply skeptical of anybody remotely affiliated with the government, let alone the consequences of needing to be out of work and what the economic consequences of that may be to their family. We've had success in a lot of places. In many states, the local health departments have been trusted. Veterinarians who work on those farms are often even more trusted, so we've had success in getting on the farms. I don't want to discount that, but I'd like to see more of it.


With respect to the testing numbers, as I mentioned, David, a lot of the influenza testing happens in the doctor's office, it doesn't even get reported because it's such an everyday thing. In terms of the number of tests that have been run at the state health department level, I will have our comms folks follow up with you. I know it's more than 23. I don't know how many more, because most all of those are reflex to CDC, even if they are negative. Would we like to be doing more testing? Absolutely, we would. That said, we also assess what's going on, and this sort of dovetails with Lena’s question. It’s not just through reports from local health departments and the vets that we're assessing worker health, we're also looking at other metrics, namely through our syndromic surveillance system. This is a system that is constantly scanning the reasons why people are going to emergency departments, urgent care departments, as well as a separate system that's constantly looking at the tests that clinicians are ordering. Thankfully, we have not seen any concerning signals come out of either of those systems, even when we compare areas that have had positive herds against areas that haven't.


Lena, with respect to your question about the vaccines that we currently have in the stockpile. There's been a lot of discussion about the approach to those. I think some of the analysis that I mentioned earlier cross-applies there, which is we'd be looking for additional numbers, as well as the composition, pattern of transmission, and concerning signs from a virulence or a severity perspective. Reassuringly, we have not seen those, even adjusting for the fact that we haven't done as much testing at the animal or at the human level. That's because we have these auxiliary mechanisms that help us understand what may be happening.


If we were, and I'm speaking hypothetically, which my comms people tell me never ever to do, but I'm going to break that rule for a second to illustrate the point that if we were to see a significant uptake in individuals walking into emergency rooms in Region X of the country, who were reporting as being dairy workers, who had signs and symptoms consistent with influenza, conjunctivitis, things like that, that would that would be something we'd take a look at. But in and of itself, the query whether Tamiflu alone is at least an equivalent strategy, that's the conversation. It's not that we're averse to doing it, but there's a lot of analysis, and please note that ASPR is very much connected with us there.


Q: What’s a significant number?


Nirav Shah:


We usually do this statistically. We look at not just an increase, but an increase above an anticipated baseline. Flu season is low right now, and we have historical data on flu rates and all of these regions. Generally speaking, when you get one standard deviation above, people start looking, they start noticing, I start getting phone calls. When you hit one and a half, two standard deviations above that, I'm immediately calling the state health officials. When you get more than that, that's when we're really saying, we need to take action now, rather than waiting for more data to come in.


Thomas Bollyky: 

As I mentioned, we have quite a robust audience online so I'm going to try to switch back and forth. We have about 14 minutes left and many, many questions. Please limit yourself to one question and deliver it as quickly as you can. I'm going to call next on Rajeev and Amy, online.


Q: Great, great overview the situation. Nirav, I'm sure this wasn't intentional, but the way you discuss the trigger points for escalation of activity could give the impression that we're waiting for the virus to do something, and that we're not doing enough to prevent the virus from doing that, and that being adapting itself to humans and enabling human to human transmission. My question is, are we doing everything we can to limit that risk of human adaptation, specifically biosecurity in dairy plants, considering vaccinating cows? Are we doing things like serosurveys, looking at farm workers, that may not have any symptoms to report but may have evidence of prior infection that would suggest that there's more of a risk here than we currently appreciate? And by the way, this is Rajeev Venkayya, the CEO of Aerium Therapeutics.


Thomas Bollyky 

Great, Amy Maxmen?


Q: Have you conducted serological studies on farm workers to see if maybe had been infected in the past?



Nirav Shah:

I'll take that question first. This is a question around doing serological antibody studies to see if there have been previous exposures. We're very interested in doing that. It of course, requires the buy-in of the farmers and most importantly, the buy-in of the workers themselves. It's not a simple swab, it’s a blood test. Folks are averse to blood tests in general, so on and so forth. However, CDC and our counterparts in the UK have done similar analyses which we've published. Thankfully, the exposure rates in poultry workers, which I acknowledged are distinct, have not been such that we've seen widespread evidence of prior infection. But we would like to be replicating that on the human health side.


Rajeev, I love that question because it's akin to the Easter Island question. Are we doing everything we can to think through the next steps? All of us, I think, would like to be leaning in as much as possible. Thankfully, we've had success and getting on to farms, to try to understand these things and implement some of the biosafety measures that you noted. USDA has been in the lead there and they've really taken they've taken a lot of effort to try to put those systems in place. That said, it's a concern that we have, particularly as we think about potential intermixing of cattle with pigs, that could be a turning point.


Thomas Bollyky 

I'm going back to the room. Steve Morrison and then Erin Sorrell.


Q: Thank you, Nirav. This is really rich and valuable and thank you for taking the time. I can imagine that we could have some negative things happen even before there's been a whole lot of mutation. In other words, you could see a drop of consumer confidence and buying patterns. You could see disruptions in the industry, through fairly modest sort of threats perceived. You could see this issue falling into partisan toxicity quickly, with a nice dose of disinformation. How are you preparing for those possibilities?


Q: Thank you so much. Considering that we know currently that the virus has gone from wild migratory birds to poultry to cows to cats and from cows back to poultry, is there any effort to look at co-localization of farms with farms that have cattle and swine that are mixing cattle and poultry, poultry and swine? Is there increased surveillance or active surveillance going on there, either in the animal populations, or the agricultural workers that are present in those targeted states and targeted regions?




Nirav Shah:


Let me start with the former question. I totally concur, there could be effects that are prior to the public health implications I mentioned. I really want to make sure, for the record and for everyone listening, I note that the preliminary data around milk safety from FDA are reassuring ie pasteurization is a good thing and it continues to be. Likewise with respect to some of the other dynamics, the overall percentage, for example of the milk supply that has been affected has thus far been quite small. Those are more USDA’s lanes, but they have those data and as you know how to get to those data.


That said, we are on the lookout for other non-public health risks, particularly with respect to misinformation. One reason I'm here today is among those, but we've also been trying to take more of the misinformation head on, pre-bunking, if you will. That was a strategy that I think was used in some places successfully during COVID vaccination and it's something that I and our communications team are trying to do more deliberately here. I concur with your point, I wanted to focus more on the CDC piece but you're absolutely right about the broader tapestry of potential harms.


With respect to other broader surveillance, completely concurred. You're correct as to the description of some of the transmission patterns that immediately raises concerns. We're now beyond the world of poultry which is where we had been, the first mink in Spain and then poultry. Now we're beyond that world, and that raises the bar in terms of what we need to be doing subsequently. Pigs are top of mind, but there are other farms, cross-apply the Easter Island hypothesis here. What are people going to be saying 10 years from now about us standing around here? Saying ‘why didn't they think to do X?’ And farm surveillance is top of mind there.


Thomas Bollyky: 

We're going back online. I'm going to call on Rick Bright and then Syra Madad.


Q: Great, thank you. First of all, thanks for the update. You mentioned waiting for people to come in with symptoms. There's a lot of asymptomatic transmission of this virus right now in cows and we're learning more and more about asymptomatic. It's really concerning that we're only waiting for people to show up in a doctor's office with symptoms, that we might be responding too lates. I'd like to know what work you're doing on that and also on the testing, because we can't wait for someone to get a test from a central lab or even one of these public health labs when you have to have Tamiflu in the first 24 to 36 hours after exposure. So what work is CDC doing to put a home testing into place and a test to treat system?


Q: Perfect, thanks so much. First, I just want to give CDC kudos for the technical reports that you all have been producing, I think they've been fabulous. At Health + Hospitals, I oversee the system wide special pathogens program and we ran a tabletop exercise with healthcare leaders last year on H5N1. Two things that have come out that I have questions on. One is can all be US hospitals be able to care for potential patients with H5N1 once we get to that point, or do we want to utilize these bio containment units that we have set up across the United States? And then secondly, if you can talk a little bit more about health care worker protection and health care worker monitoring, and some of those measures that go around it?


Nirav Shah:

Thanks, Rick, for that question. That's been on our mind as well, because a lot of the systems that I mentioned, syndromic surveillance, lab testing, do require someone to come in and that may be too late. That said, those systems have proven their utility and other circumstances as well, so I don't want to discount them, nor do I think you're doing that. That said, we'd like to have a broader access to testing. We've been working with farms to try to conduct more enhanced epidemiological surveillance to understand what the likelihood of conversion is, of someone who's been exposed to later test positive.


We have data on that from prior studies in poultry, both in the US and the UK. The results are reassuring, but we'd like to have more data. That's one reason we're working with farms. One element of that would be home testing or point-of-care testing. I know that NIH has been working on that. I don't want to speak for them, but I do know that there's work underway though it’s not something that's available now.


Syra raises great points about health care capacity and infrastructure. With respect to the capacity of all hospitals to care for H5, it probably would not be all hospitals. Much as we do with other disaster related planning efforts, we are thinking of what a graduated system would look like. We're not there right now. Indeed, the cases of H5 that have occurred in the US at least have been mild. Some of the other severe ones that Tom noted have been just the tip of the iceberg. We're not in a position where we're rolling out what are called PTCs, tent hospitals or anything like that, but it's something that's part of our background planning. Likewise, with health care worker protection, that too would be contingent upon and critical to have the right PPE, so we've been working with state health departments to assess their PPE stocks. Thankfully, those are quite good.


I think you mentioned vaccines, again, as I mentioned, the key question with respect to vaccines is to go back to questions Steve mentioned. Having the vaccine is great, but what really matters is getting people to take it, turning the vaccine into vaccination. That is really the key.


Thomas Bollyky: 

We have about four minutes left and many questions, so I'm going to try to take four and do a lightning round to the room to online.


Q: Bruce Gellin from the Rockefeller Foundation. Thanks for coming and thanks for organizing. Should the situation arise where this switch to away from seasonal flu is the decision, who makes that call? And then the elated piece is what are the consequences downstream, since many of the manufacturers are supplying flu vaccine to other places in the US?


Q: Rochelle Lipsitz, International SOS. Can you talk a little bit about the collaboration you're doing with international countries and the WHO? Have there been any upticks in any other country?


Q: Sonya Stokes, I'm an emergency physician at Saint Francis Memorial Hospital. My question is about potential gaps in our syndromic surveillance. In outpatient clinical settings, we are typically using rapid influenza diagnostic tests for patients with flu-like illness. When practicing in rural communities and Mendocino and Sonoma counties, positive RADTs are not sent to DPH. They are immediately discarded unless these patients require hospitalization. Now it is known these tests are unable to distinguish between influenza A subtypes. Given we are seeing unseasonable cases of influenza A in multiple regions, including I have personally diagnosed four in the past four weeks, what is the likelihood we are missing cases of H5N1 and instead misdiagnosing flu A, especially in catchment areas where patients have high contact with cattle and poultry?


Q: Devi Sridhar, University of Edinburgh. The April 1 confirmed case, was that cow to human or was that poultry to human? Are you able to see that they're actually getting it from cattle?


Nirav Shah:

Taking those in reverse order. Debbie, that was cow to human based on the lack of exposures to any other poultry or tiny poultry. I'll come back to the diagnostics question.


Rochelle, on the question of international collaboration, I'd be remiss if one of the keys to our international collaboration, I didn't call out Dr. Gawande is here. He and his team at USAID are critical to that. CDC’s influenza division also has over two decades of bilateral relationships with influenza laboratories and clinicians in a number of states, including via the WHO. We’re on a number of panels where we're sharing a lot of the scientific information that I've discussed here today in real time. We have not seen significant increases in H5 activity that are above the norm for what we might see, based on where the country is located. There are periodic flare ups of H5, typically associated with poultry, and then in Vietnam, more cases of H7 that we've seen recently there. All of this is on our website, thankfully, but we keep very close tabs on these things. There have been recent cases in Cambodia, one in India, etc. So we're in very close contact because of the implications.


There was a question from Bruce around who makes the call, it's a collective decision. It's not only one single person. As you can imagine, there are myriad dynamics and tradeoffs here, and so it would be unfair, almost inhuman, to task one person with carrying the weight of that decision. It's a collective of folks, some of which are at CDC, some of which are broader at HHS, some of which are down the street here. Regarding misdiagnosis of potential H5 cases, that is a risk of misdiagnosis in any infectious disease situation, particularly in situations where the baseline prevalence is increasing. We don't have any evidence that that's happening based on the testing that we've done so far. But as we're thinking about the models that we undertake, that's always a consideration that we have in our sensitivity analysis. Again, I wish we had better data around testing. We're getting there, but it's a task right now.


Thomas Bollyky:

We have one minute left. What has your experience being a state official during COVID-19 brought you in this response that perhaps not all your CDC colleagues have, or those from other agencies that have been involved in this response?


Nirav Shah:

So you know, we don't have I'll be I'll get it done in a minute. We don't have a national public health system in the United States, but what we do have is a nationwide public health system and the effector arms of that public health system are state and local health departments: county, city and state health departments. We can talk about all we want to in Atlanta, but the job is done at the local level, and so having that perspective has been really critical as we think about policies from a public health lens.


Thomas Bollyky: 

Great. I want to thank Dr. Shah for joining us today and all of you. Apologies again for those questions I didn't get to. I also want to thank my colleagues for organizing this event. As I mentioned, we will post this video as fast as we can.

Top Stories on CFR

International Law

The high court’s decision could allow future U.S. presidents to commit grave abuses of power with impunity, with serious implications for U.S. foreign policy and national security.

Election 2024

The Ohio senator is Donald Trump’s choice as his running mate for the 2024 presidential election.

Election 2024

After a shooting that injured former President Donald Trump and killed a spectator at a campaign rally, leaders of both parties must unite behind efforts to calm and stabilize the political climate.