The Peter Attia Drive - #106 - Amesh Adalja, M.D.： Comparing COVID-19 to past pandemics, preparing for the future, and reasons for optimism

00:00:00.000 Hey everyone, welcome to the drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:15.480 my website and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:19.800 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:24.760 wellness full stop. And we've assembled a great team of analysts to make this happen.

00:00:28.880 If you enjoy this podcast, we've created a membership program that brings you far more

00:00:33.280 in-depth content. If you want to take your knowledge of this space to the next level at

00:00:37.320 the end of this episode, I'll explain what those benefits are. Or if you want to learn more now,

00:00:41.740 head over to peteratiyahmd.com forward slash subscribe. Now, without further delay, here's

00:00:48.080 today's episode. Welcome back to another special episode of the COVID-19 series of the drive.

00:00:55.160 Joining me today is Dr. Amesh Adalja. Amesh is a senior scholar at the Johns Hopkins Center for

00:01:01.400 Health Security. He has long been focused on pandemic preparedness and emerging work on

00:01:08.020 infectious disease, biosecurity, et cetera, for many years long before this coronavirus entered

00:01:13.340 our consciousness. I wanted to talk with Amesh for a couple of reasons. One, I'd heard him on

00:01:18.460 other interviews, particularly the one with Sam Harris. Also, I'd seen him in interviews,

00:01:22.200 read some of his work, and just found him to be a very thoughtful guy who could put the

00:01:27.100 coronavirus pandemic in the context of all the previous pandemics, and not just the ones that

00:01:32.820 are immediately in our recollective memories, such as SARS, MERS, H1N1, but even going back a little

00:01:39.540 bit further than that. This is a very brief discussion. He was incredibly busy. We were very

00:01:43.900 fortunate to get 40 minutes of his time between other interviews, but we do cover quite a bit of

00:01:48.940 ground and certainly everything I was hoping to speak with, which is a bit of a more clear history

00:01:54.160 of when this virus likely emerged, how it came here, and what we know about it relative to other

00:01:59.940 coronaviruses. We talk a lot about what the plan forward should be, and we end with really what he

00:02:06.820 is most optimistic about. So though this interview is relatively short for the standards of interviews

00:02:11.320 that I do, like I said, it's about 40 minutes. I think we cover a lot, and there's no question I would

00:02:16.160 like to have Amesh back. I do think it is really just an inevitability that we will once again

00:02:21.120 face a pandemic, and whether it's a coronavirus or another virus is probably less the point here,

00:02:27.060 but unquestionably, there are things that we could do better the next time we're faced with this. So

00:02:32.280 without further delay, please enjoy my conversation with Dr. Amesh Adalja.

00:02:36.780 Amesh, thank you so much for making time to speak this afternoon. I know you're incredibly busy,

00:02:46.980 and we were lucky to sort of catch you in between interviews and television appearances. So thanks

00:02:51.640 very much. I'll just sort of jump right to it. I reached out to you because I was very interested

00:02:57.620 in your perspective on the historical similarities and differences between this pandemic and previous

00:03:05.660 ones. Maybe just briefly tell us a little bit about your background and why it is that you're

00:03:10.180 not just someone who came to be interested in this in the past three months. So I'm an infectious

00:03:15.100 disease emergency medicine and critical care physician who's focused his entire career basically on the

00:03:20.680 issue of pandemic preparedness, pandemic prediction, infectious disease and national security, emerging

00:03:26.460 infectious disease. Even from the time when I was a trainee, that's basically all I focused. And I work at a

00:03:31.860 think tank that's devoted to this issue. And I've been there basically full time since 2010 or so, but

00:03:37.160 I've been there since 2008 working on this. I published on pandemic prediction on what the

00:03:42.220 characteristics of certain pathogens that cause pandemics would be on H1N1, on Ebola, on agents of

00:03:49.500 bioterrorism. So I really focused and tried to niche myself into this aspect of infectious disease in

00:03:55.140 medicine. So this is an outbreak that had been on my radar before it hit headlines. And I've been

00:04:00.580 following it almost the way people follow sports teams that they like. Tell me a little bit about

00:04:05.340 what you thought in call it December when we, at least in the West, saw the first report of the

00:04:10.940 case in China. What was your intuition at the time? At that time, I was trying to take what the

00:04:15.280 Chinese were saying at face value, meaning that they had learned from the lessons of SARS about

00:04:18.860 transparency and were going to do this in a different manner. And they did produce the virus

00:04:23.680 sequence very quickly that allowed diagnostic tests and vaccine development to start. And we were able to

00:04:29.180 identify a novel coronavirus. But initially, a lot of the reports were saying this is something that was

00:04:34.860 animal to human, really tied to wet markets in Wuhan, that there wasn't evidence of human to human

00:04:39.740 spread. There were no deaths yet. But I did think that 41 patients getting something from one animal at one

00:04:45.960 market really seemed very odd to me, that that seemed to be too much. So I was a little bit skeptical that

00:04:52.500 this was just an animal to human event, unless there was something different going on, or there were

00:04:56.220 multiple animals that were infected and infected multiple people. But as soon as I saw the first

00:05:00.580 paper, and that was the Lancet paper, where they showed the first case got ill on December 1st,

00:05:05.900 and he had no contact with the market, that told me that we were dealing with a transmissible

00:05:10.020 human to human respiratory virus, and that this was going to be a pathogen that was going to spread

00:05:16.060 and not be containable, and that we were going to have to get ready. I had a lot of questions about

00:05:20.020 the case fatality ratio, the hospitalization rate, which I still have. But I knew at that point,

00:05:24.680 when you had a virus that spreads efficiently between human to humans through the respiratory

00:05:28.180 route, you really have to prepare for this being everywhere, especially when you know that this

00:05:32.380 had a head start at least spreading since mid-November in China, and nobody knew about it until late

00:05:37.520 December. So that gave a virus a very big head start and could have been anywhere by the time we

00:05:43.080 actually knew it was, as soon as we discovered that it actually existed.

00:05:46.620 Based on that, do you think it's possible that this virus was in the United States, potentially even

00:05:51.520 within an individual prior to December 31st?

00:05:54.920 This is a little bit of a controversy. I do know that with past novel coronaviruses that have been

00:05:59.360 discovered, like HKU1, which isn't one that people think about, but it is one that was

00:06:02.960 discovered post-SARS, it was everywhere as soon as they found it. This one, it doesn't appear that,

00:06:09.820 at least that there wasn't widespread presence of it before maybe January. However, I don't rule out

00:06:15.320 the possibility that it could have been mixed in our flu and cold season, maybe a sporadic case here

00:06:19.440 were there that was mild, that didn't get diagnosed, but it doesn't appear, at least from

00:06:23.640 the phylogenetics, the genetics of the virus that we're seeing from both New York and

00:06:27.620 Washington, that this was around before that. But I do think it's going to be important to go back and

00:06:32.660 look at bank samples and look at people to see if there were cases. I don't think there were a lot of

00:06:37.380 them. I think we would have noticed if there were a lot of them, but I think there may have been sporadic

00:06:40.680 cases that were mixed into flu and cold season. But it's an open question, and I think it's a good

00:06:45.260 hypothesis, and I think it's something that deserves a lot more attention. At what point,

00:06:49.940 Amesh, were you becoming convinced that this was going to enter the U.S. in a manner that was going

00:06:55.220 to pose real difficulty for the country? So I knew it was going to enter the United States almost from 1.00

00:07:00.420 the onset, as soon as we knew that this was transmitting between human to human, that this

00:07:04.100 wasn't going to be containable, just like when H1N1 appeared in 2009 and was found in Mexico, that we

00:07:09.800 knew that this wasn't going to be something that would spare the United States. What I wasn't quite

00:07:15.080 attuned to was how difficult it would be to contain in the United States, because I, like many of us,

00:07:20.980 believe that our diagnostic testing and our case finding and our contact tracing would have been

00:07:25.480 really much better than it was. But the fact was, we didn't know who had this, who didn't have it. Our

00:07:31.280 testing made it much harder to actually do that, and we weren't even testing mild cases, and those mild

00:07:37.380 cases are arguably more contagious than the severe cases. So that let this slip from something that was

00:07:42.600 potentially controllable and wouldn't have been a problem or a major problem, not putting cities

00:07:48.360 like New York City under the stress that they're under, to one that became completely unmanageable

00:07:52.920 because we basically were allowing this virus to have about two months of unabated spread in the

00:07:57.740 United States. And that's something that most of us did not think would happen because we thought

00:08:02.720 that we were much more resilient to these types of infections than we really were. And I think that's,

00:08:07.800 it didn't have to be that way, but that's basically how it turned out.

00:08:11.440 When you go back and look at the history of not just the other coronaviruses, which have gotten a

00:08:16.640 lot of airtime lately, of course, most people, even if they don't remember the ins and outs of SARS and

00:08:21.740 MERS, they're certainly familiar enough. But when you go back even further and look at some of the

00:08:27.500 flus that occurred in Asia, Hong Kong in the 1950s, 60s, and of course, going back even further to the

00:08:34.500 Spanish flu of 1918 to 1920, what are some of the similarities that you see with this novel

00:08:42.720 coronavirus? And of course, I'll contrast that in a moment with what the differences are. And the

00:08:46.840 point of this exercise is less about just abstract history, but more to understand what we can learn.

00:08:53.180 Sure. So if you go back to 1957 and 1968, these were pandemics that were marked by the emergence of

00:08:59.440 a novel flu virus that spread around the world very rapidly. And if you look at the United States

00:09:05.000 experience of 1957 and 1968, about 100,000 people died from that, which is a substantial number because

00:09:11.000 on record right now, the worst flu season we've had outside of 57 and 68 has really been 2017,

00:09:17.540 2018, where about 80,000 people or so may have died. So these were severe outbreaks. And flu has a lot

00:09:24.220 of similarities with coronavirus, but there are some differences. So one thing is, you know,

00:09:29.400 they both are transmitted through the respiratory route. They both have symptoms that are included

00:09:34.540 coughing and sneezing and sore throat and muscle aches and pains and fever. So they have a lot of

00:09:38.660 overlap clinically. And I think that because of the way they spread and their symptoms, you can look at

00:09:43.720 their spread and there's a lot of analogies that you can draw. But what I would say is with flu is

00:09:49.040 sometimes it's difficult because people have already taken flu and put it into their own risk

00:09:53.500 of daily life, that they know that there's going to be maybe 30,000 to 50,000, 60,000 people who die

00:09:59.760 every year. So even during a pandemic like 1957 and 1968, most people still carried on because

00:10:05.400 this was still a flu virus. And the same thing happened to an extent with 2009 H1N1, especially

00:10:10.520 when we realized that its case fatality ratio was actually less than seasonal flu, even though it

00:10:15.080 infected 61 million Americans, led to a lot of hospitalizations and put hospitals into distress.

00:10:20.800 It wasn't that deadly. So people kind of took it in stride. This is something that's a little bit

00:10:26.620 different because it's on top of flu. It's additive to flu. I think that's one of my pet peeves of the

00:10:31.520 media coverage is sometimes they try to compare it to flu. But you remember that we're already going

00:10:35.320 to have 40,000 to 50,000 deaths from flu. And this is on top of that. I do think that influenza gets a

00:10:40.680 short shrift by people because it's something that they take for granted and don't realize the burden of

00:10:45.080 infection that it has. And that influenza still, even today, remains the biggest pandemic threat

00:10:50.180 that we face. If you think about some of the influenza viruses like avian influenza, they have

00:10:55.300 case fatality ratios of about 65%. That's too much for the world to bear. So if one of those, like H7N9,

00:11:02.060 became able to efficiently transmit from humans, that's a totally different type of pandemic that

00:11:06.360 we're talking about than what we're dealing with today. And I think that's important. And maybe that's

00:11:10.340 why I seem a little bit more optimistic than most people are a little bit more measured when I'm

00:11:14.380 talking about this, because in my mind, I'm thinking about avian influenza and what that kind of a

00:11:18.960 pandemic would represent. And I used to say that this is kind of a trial run because it's only, it's less

00:11:23.180 than 1% case fatality rate, maybe as low as 0.3 case fatality rate. But we didn't do that great of a job

00:11:28.940 with it in terms of diagnostic testing and hospital capacity and personal protective equipment. So that's

00:11:33.580 magnified this. And that's the human factor that's magnified what the virus could do. And I think

00:11:37.580 that's an important point to make. And that's how I would kind of leave it is that you have certain

00:11:42.260 ways to think about this using flu as an analogy. But I do think that I'm a little more worried about

00:11:47.240 our pandemic resiliency based on how badly we've handled a 1% case fatality rate pandemic virus,

00:11:54.200 where you've got cascading decisions by governors in states and countries all around the world that

00:12:00.000 really have magnified the damage that the virus has done. One other question I want to ask before we

00:12:05.000 leave the historical, which is the sort of ebbing and flowing recurrences that occurred back in some

00:12:12.060 of these other pandemics, where it's easy to sort of think of them as this was the impact on the

00:12:17.360 United States, but not realizing, well, actually it had a different impact on this city versus that

00:12:21.620 city. And they were not just different in terms of the strain they put on the healthcare system or

00:12:26.620 even the mortality rate, but even temporally, they could be separated by quite a period of time.

00:12:30.220 Is that something that also applies here in your opinion?

00:12:34.980 I do think so. We have this tendency to think of a pandemic as a homogenous 0.99

00:12:38.300 wave over the whole world or even a country, but it's actually many small outbreaks together. And

00:12:44.340 everybody's on a different timescale based upon when the virus was introduced into that area,

00:12:48.720 what their population density is, what their hospital capacity is, what demographic got infected.

00:12:53.800 So you aren't going to see synchronous outbreaks. They're going to be a little bit staggered.

00:12:57.680 And there's going to be differences on, for example, when did somebody do social distancing?

00:13:01.460 When did someone not do social distancing? How do they vary between this state and that state?

00:13:05.040 That's also going to impact the trajectory of the outbreak. You might see places like I'm sitting

00:13:10.140 here in Pittsburgh right now, which hasn't had a bad experience with this virus yet, but we've had

00:13:15.360 the opportunity to learn from New York and Seattle and San Francisco. And that's gauged the way that

00:13:21.220 we've dealt with testing and hospital capacity. So our outbreak is going to be different because we've

00:13:25.520 learned from them and we don't have the population density issue. So I do think there is going to be

00:13:29.720 ebbing and flowing, especially with social distancing varying and maybe lifting in certain

00:13:34.300 places of the country and not lifting in other places. And you may find that this virus is going

00:13:38.860 to have some degree of seasonality. I don't think it will have complete full seasonality because

00:13:43.180 there's so many people that are susceptible to it, but you may see this come back in successive

00:13:47.920 waves. And remember, there are four other coronaviruses that circulate every year and cause

00:13:51.340 about 25% of our colds. This is something that I suspect will be the fifth one that does that. And

00:13:56.900 it looks like it has an intermediate severity. It's not as severe as SARS, for example, which has a

00:14:02.020 case fatality ratio of about 10%, but it doesn't seem to be as mild as the other four coronaviruses.

00:14:07.020 So I do think you're going to see ebbing and flowing, especially as social distancing changes

00:14:10.540 across the country until we have a vaccine. And is it your view that this will now be a fifth

00:14:16.860 coronavirus that will fit into the mix and it's never really going to go away? In the way that

00:14:21.720 at least SARS and MERS, because of their severity, they don't really factor into that recurrent cycle

00:14:28.020 we see every year. I do think this is going to be the fifth seasonal coronavirus. And I would say

00:14:32.660 with SARS and MERS, it's not just that they were more severe, which they are, it's that they're poorly

00:14:37.240 transmissible from human to human. They are mostly zoonotic or meaning animal to human transmission. So SARS

00:14:42.860 from palm civet cats, Middle East respiratory syndrome emerged from camels. So that's something that's really

00:14:48.680 limited to their spread. It's only those individuals who are in contact with those animals that are really

00:14:53.060 at risk. And when you look at their outbreaks, they're very specialized. They're happening in healthcare

00:14:57.160 facilities with poor infection control, and it doesn't really sustain itself in the human population.

00:15:01.960 Whereas if you look at the other four coronaviruses, the ones that cause common colds, they are ubiquitous.

00:15:07.320 They transmit very easy. They have a mild spectrum of illness, which allows people to go about their

00:15:11.980 daily life and spread. And this new novel coronavirus does appear to be more like them in

00:15:16.760 terms of their transmissibility. So that's why I think that this will be the fifth seasonal coronavirus

00:15:20.860 until there's a vaccine. Do you think a vaccine is going to be specific to this coronavirus? Or do you

00:15:27.380 think that it will be more geared towards all coronaviruses to cover not just this one, but perhaps

00:15:34.600 others that will potentially emerge, a SARS-CoV-3 that's five years away?

00:15:38.840 Right now, vaccine development is premised on making something specific to the specific

00:15:44.000 immunogenic protein that the immune system recognizes for this virus. So it will be specific

00:15:48.860 to that. But I do suspect you might see some cross-reactivity between the vaccine for this

00:15:53.760 SARS-CoV-2 and other related coronaviruses. Maybe the SARS-CoV-2 clusters in something called the beta

00:15:59.520 coronaviruses. Maybe the vaccine will work against all beta coronaviruses, but it would be great if it

00:16:03.760 worked against all coronaviruses and we had a pan-coronavirus vaccine. We might get something like that

00:16:08.460 because coronaviruses are different than, for example, influenza, which has been very hard to

00:16:12.600 make a universal flu vaccine. The coronaviruses in general tend to be much more stable, even though

00:16:18.140 there's some diversity among them. And we might have cross-protection, which would be useful to

00:16:21.940 take other threats like Middle East Respiratory Syndrome and the original SARS off the table as well.

00:16:27.080 When H1N1 hit, you sort of alluded to this briefly, the case fatality rate was initially deemed to be

00:16:33.000 much higher. It was only once we appreciated how prevalent it was that the case fatality rate

00:16:38.160 came so far down. I believe in the end, it's less than 0.1%, correct?

00:16:43.320 Yes. Yeah, definitely.

00:16:44.780 What is your real assessment now? And again, you can only be speculating at this point,

00:16:49.460 I understand. But what do you think is the true case fatality rate of SARS-CoV-2 specifically?

00:16:56.180 And if you want to answer that, by the way, in terms of, I think that the CFR is going to be this for

00:17:01.220 people over 60 and this for people under 60, and it blends out to this, I mean, answer it any way

00:17:05.660 you see fit. So the CFR has been really hard to calculate because we have a severity bias because

00:17:10.840 testing has been so heterogeneous around countries. So what I do is try to look at a place that's

00:17:15.700 tested extensively and use that as kind of a barometer. And right now, it used to be South

00:17:20.340 Korea that tested, and now it seems to be Germany is doing the best testing. And you're also looking at

00:17:25.020 modeling studies. So some of the modeling studies from Imperial College put the case fatality ratio at

00:17:28.920 0.66%. Germany looks like it's at a 0.3 something percent.

00:17:34.360 Yeah, I think 0.37 this morning. Yeah.

00:17:37.060 Yeah, doing antibody testing to try and understand. So I do think it's probably in that range,

00:17:41.680 probably in the 0.3 to 0.66. I say that with some confidence, but it may drift lower or higher

00:17:48.080 depending upon how much severity bias is in the samples. And it's very hard. We probably won't

00:17:51.940 truly know until we do retrospective studies looking at antibodies to understand how prevalent it is.

00:17:57.440 And there's differences amongst that because that's an average number. If you're above 80,

00:18:01.440 your case fatality ratio may be as high as 15%. If you're eight years old, your case fatality ratio

00:18:06.760 may be 0%. So I think it's important to remember that these are average numbers and it's not

00:18:10.960 every person carries that risk. Some people will have much, much higher risks and some people will

00:18:15.820 have lower risks. And I think that's sometimes lost in nuance when you try to come up with one number.

00:18:20.220 Absolutely. Yeah. Blending that is, you know, there's lots of glib examples of how

00:18:24.580 you can drown in an average of three inches of water. Of course, if you're walking across a river

00:18:29.720 that has vacillating depth, because you're again, kind of a veteran of this, what do you think was

00:18:35.720 sort of the over slash underdoing of some of the predictions that came out in this sort of gen one

00:18:43.600 models that showed up in sort of February where they were saying, look, this is something that is

00:18:48.500 going to infect 200 million Americans. It's going to kill two to 4 million Americans. Do you think

00:18:55.560 that that type of modeling historically has ever shown to be accurate? Or do you think that, yes,

00:19:02.840 that was accurate and it's the measures that are in place that are going to hopefully prevent that

00:19:07.880 from happening? Because it seems less likely now that we're heading in the direction of those types

00:19:14.340 of doomsday scenarios. But again, it's hard to know how much of that is in response to the measures that

00:19:19.820 have been enacted versus predictions that were predicated on poorly understood things, including

00:19:25.900 what the are not was. I think that models you have to realize have assumptions built into them and you

00:19:31.980 have to look at those assumptions because just a small difference in the assumption can lead to a big

00:19:36.560 change on the end of it. And what I would think, at least from my understanding of many of the

00:19:41.160 models, is that the hospitalization rate was probably overstated. Because we know, for example, that the

00:19:47.940 diagnosis that we're talking about in any given city or town are likely understated by a factor of at least

00:19:52.640 maybe 10. And I can say that from my own practice when I order the test and don't order the test. There's many

00:19:56.960 patients I think have the disease and I don't order the test. So there clearly is a severity bias in who gets

00:20:02.100 tested. And then I think you see this idea of 15 to 20% getting admitted to the hospital. I think that

00:20:08.600 that doesn't necessarily mean everybody of the people that get infected, maybe that's 40% of the

00:20:13.960 population over time, that 15 to 20% of those individuals get admitted. It's more like of the

00:20:20.400 40% of the population that get infected, the ones who go to an emergency department, 15 to 20% get

00:20:26.200 admitted. And maybe the real hospitalization rate is 5%. If you look at, for example, Westchester

00:20:31.440 County's data, which I haven't looked at lately, but the last time I looked, they did a lot of heavy

00:20:35.780 testing in that part of New York State, because they had that outbreak in New Rochelle. And their

00:20:40.460 hospitalization rate was around 5%, a little bit less than 5% the last time I calculated it.

00:20:45.680 And that changes, that gives you a major change from going down from 20% to 5% in terms of what your

00:20:51.760 ventilator needs are, what your ICU bed capacity needs are, and what the case fatality ratio is going to be,

00:20:57.360 if you look at what the hospitalization rate is. And I think that that's, at least one of my

00:21:02.340 criticisms of some of the models, the hospitalization rate was set too high, and that they were taking

00:21:07.260 too big of a fraction, using the wrong denominator, I think, to come up with what their case fatality

00:21:12.740 ratios are, what their ICU bed needs would be, and what their mechanical ventilation needs would be.

00:21:16.740 And that happens, because models have lots of assumption in them, and they should be used as tools,

00:21:20.700 they're not ironclad. And I think that sometimes that gets lost in the press reporting of them,

00:21:25.300 they get looked at as if they are the truth, and they need to be revised. And when you have real

00:21:29.680 data from real hospitals and real people, it should supplant what you're using with the model.

00:21:34.760 So if your model is not matching reality, then I think you need to either change the assumptions

00:21:38.320 on the model or actually look at reality. These are tools, and all models are going to be wrong.

00:21:43.260 Some of them are going to be useful, and some of them are not going to be useful.

00:21:46.660 I agree. I think it's, unfortunately, the press sometimes views corrections of models as a sign of

00:21:52.120 weakness as opposed to a necessary part of the evolution of utilizing the tool as you described

00:21:56.380 it. We've talked a little bit about this being a dress rehearsal for what is coming unquestionably

00:22:03.560 at some point, very likely in our lifetimes, which is another pandemic, another virus, and potentially

00:22:10.180 one that could be much more devastating. You used an example of viruses that are typically

00:22:16.080 transmitted only from animals to humans that can potentially be much more catastrophic.

00:22:22.420 But if those viruses ever figure out how to go human to human, they spread much more.

00:22:27.880 What is your view on the role of local versus central government in dealing with that? Can't imagine

00:22:37.540 there isn't a role for both, but I feel personally very confused when I try to, you sometimes play the

00:22:43.820 game. If I were czar for a day, what would I do different? But the reality of it is, it's easy

00:22:47.580 to play armchair quarterback. I'm not really sure I know what the federal government should be doing

00:22:52.420 in this situation versus the state governments and local governments.

00:22:57.420 So this is something that's pretty unique to the United States because we have a system of

00:23:00.560 a federal government with states and locals having most of the power. And it's especially true in

00:23:05.660 public health because most of the public health powers are vested at the local and state level.

00:23:09.040 And the federal government is more of a coordinator. Even the CDC can't actually get involved in

00:23:14.300 something unless they're asked by a state. So you will often see differences and heterogeneity

00:23:19.340 between recommendations from maybe one county to the next or even, and definitely from one state to

00:23:24.080 the next. And sometimes that can be confusing. I do think that I'm generally supportive of local

00:23:29.380 health departments being the ones running it because they actually know their community and they know

00:23:32.680 their capacities. They know where their gaps are and they're able to really be on the ground with the

00:23:38.060 people and able to do great things when it comes to stopping an outbreak. But often what we find

00:23:43.520 is local health departments aren't appropriately resourced. And you've got one person doing four

00:23:47.000 different roles in a small town's health department, and that can be very constraining. And what you need

00:23:51.880 to really do is have those local health departments actually operating the way that they should be

00:23:56.520 and thought of as part of the whole pandemic response apparatus. Whereas many people think of just

00:24:02.020 the CDC, the NIH, and parts of the Health and Human Services Department as the main

00:24:06.040 pandemic apparatus, but it's actually the local health departments that do all that case finding

00:24:10.620 and isolation and talk to the public and deal with hospital capacity levels. It's the local health

00:24:15.640 departments. So I can't overstate how important local health departments are. And I do think that

00:24:20.500 that system works well, but you do need to have federal leadership to kind of guide the nation as a

00:24:25.760 whole on what to expect and what's going on. And I think that's sort of been missing a lot during this

00:24:30.260 outbreak compared to other outbreaks. And because of that, you've seen governors at state health

00:24:35.280 departments take on roles that they usually haven't taken, where they've deferred much more to federal

00:24:39.300 experts, not the powers, but the guidance and looking to them to set the tone. And I think now

00:24:45.080 we're finding governors basically stepping into that role for the most part, and even mayors in some

00:24:49.620 places. And I think that can be confusing to a member of the general public, because they don't

00:24:53.320 know who to believe, especially if there's conflicting information. You have a press that's constantly

00:24:57.340 trying to pit one governor against another governor against the federal government. And that makes it much

00:25:01.540 harder. But I do think that when the process works well with local, state, and federal government all

00:25:06.020 in step, all doing the appropriate roles, I do think it works pretty well. You have a locally managed,

00:25:12.140 federally coordinated response, which I think is the best way to think about how it would be ideally

00:25:16.740 done. Is it safe to say that testing is really something that like, if you go back in time and maybe

00:25:22.060 change one thing, if we're sitting here and it's January 12th, and we now have the sequence of this virus,

00:25:27.700 would that have potentially been one of the more important things for the federal government for

00:25:32.460 that centralized piece of government to have put in place? The CDC could have said, look, we're going

00:25:37.660 to make this the highest priority. Because it strikes me as that's something very difficult to be done

00:25:42.380 in a decentralized manner. Right. So what happened was the CDC put out guidance on who should be tested,

00:25:48.440 which basically was taken as gospel by the state health departments. And that included

00:25:51.900 only people that had traveled to China in the last 14 days, as well as someone had had to

00:25:57.620 have lower respiratory tract symptoms, you couldn't have just had a sore throat, you had

00:26:02.320 to have evidence that maybe you had pneumonia to be tested. So we weren't testing mild cases,

00:26:06.300 and we weren't testing people that hadn't come to China. That was a federal decision. And I think

00:26:10.700 that could have been done better and allowed much more latitude. Because you can remember that first

00:26:14.820 case in California that didn't have traveled to China, they the hospital had to actually fight

00:26:18.840 to get that test run. And there were many cases like that all over the country. And if you look at

00:26:24.360 New York's epidemiology, their introduction of the virus was not from China, it was from Europe.

00:26:29.600 And that slips through the type of testing algorithm. So I do think that there could have

00:26:34.500 been at the beginning, an idea that somebody could have said, this is a respiratory virus,

00:26:38.060 it has many overlapping symptoms with common colds and flus, you should think about this in your

00:26:42.240 patients. And we are going to allow testing to be done. If you have certain risk factors for this,

00:26:47.920 and they shouldn't just be restricted to you having severe disease or having traveled to China,

00:26:52.880 that would have changed the way that the general public and clinicians would have thought about

00:26:56.420 this. The other thing is, is that there were bureaucratic wrangles that paradoxically,

00:27:00.560 once a public health emergency was declared, they were unable to make diagnostic tests as freely

00:27:05.380 available as would have been if there wasn't a public health emergency made. So for example,

00:27:08.980 you had university labs and big commercial labs not being able to make a test, even the CDC's test

00:27:13.120 had to go through FDA emergency use authorization before it could be distributed to the states.

00:27:17.300 So there were a lot of bureaucratic hiccups that created a problem that compounded the testing

00:27:22.240 protocol with the scarcity of tests and a delay in getting testing kits everywhere. And then

00:27:26.860 we still have shortages of reagents and nasal swabs, and we're still not to where we need to be with

00:27:31.920 testing. And not to get too far ahead of ourselves, but do you get the impression that the response to

00:27:38.300 this is serious enough that it will now be taken more seriously to have kind of that type of emergency

00:27:45.000 response ready five years from now when people have long forgotten about this?

00:27:49.140 I hope so. I think this is something that's going to leave a mark on society. This hasn't happened in

00:27:53.740 modern times, not during the 68 or 57 pandemics. And it did happen in 1918, but that's not in

00:27:58.920 anybody's living memory anymore. So I do think that this is something that people will remember,

00:28:03.580 and they'll remember the costs that they personally had to incur filing for unemployment for the first

00:28:07.800 time. All of that type of stuff is going to hopefully push the public to demand that pandemic

00:28:13.900 preparedness be taken seriously so this doesn't happen again, and that this should be a

00:28:17.760 priority. This should be something that's in a candidate's campaign literature. This is what I

00:28:21.980 think about pandemic preparedness, and it always should have been. But we've gone through this cycle

00:28:25.700 for a long, long time. You can think about anthrax in 2001, bird flu scares in 2005,

00:28:31.440 the H1N1 pandemic in 2009, Ebola in 2013, 2014, Zika right after that. We've had multiple types of

00:28:39.480 episodes, and you get this cycle where everybody runs to fund this reactively, and then it goes from

00:28:45.640 the headlines, and no one remembers it. And then the same cycle happens. They cut positions at the

00:28:51.060 National Security Council when there's nothing going on. They do lots of things that make us

00:28:54.740 less resilient to pandemics, not realizing that this is a perpetual threat, just like

00:28:58.220 any other national security concern, that you have to be prepared for this at all times. And you have to

00:29:03.080 actually think about it that way, and fund it that way, and have the proper personnel, even between

00:29:07.100 pandemics and between outbreaks. But you are optimistic that this time, I mean, just based on the

00:29:12.520 economic consequences of this, even if not one more person were to die in the United States,

00:29:17.560 which means, frankly, let's be clear, if not one more person died in the United States as of today,

00:29:22.780 this would not be a major source of mortality. This would still be a rounding error compared to

00:29:27.760 influenza. But you're just saying the economic consequences of this have been so severe that

00:29:32.800 you're optimistic that we're not going to walk away from this one in 18 months and sort of forget about

00:29:37.340 it and do all the wrong stuff all over again. Yeah, I do suspect we're going to have

00:29:41.420 many more deaths, probably closer to that 60,000. I do think that the fact that this outbreak touched

00:29:47.120 people personally in a way that Ebola did not, the way that Zika did not, the way even that H1N1

00:29:52.000 did not. H1N1 actually engendered complacency because only about 12,000 Americans died, and then

00:29:56.960 people said we all overreacted to H1N1. So I think this is actually something that every American is

00:30:03.080 feeling right now because of the economic shutdown, the stay-at-home orders, all of that, the fact that they

00:30:08.420 had to adjust their entire life, this is something that's been extremely disruptive. And I think that

00:30:13.000 hopefully the public remembers that when they vote, and when they ask their policymakers about

00:30:17.360 what their plans are for the future, that pandemic preparedness becomes something that is a platform

00:30:21.800 issue now. Based on where we are, what do you think is the right strategy in, for example, a place like

00:30:29.700 New York versus a place like, you could pick any city you like, whether it be Pittsburgh, Houston,

00:30:34.660 cities where it's been nowhere near that. I mean, how would you start to think about changing any of

00:30:41.220 the policy, or is the answer until we have more testing, we can't make any more decisions?

00:30:47.500 I do think based on modeling, if these assumptions are valid, and what's going on in the ground in

00:30:53.100 hospitals, you can start to see this heterogeneity across the country. And not every place is going to

00:30:57.960 be New York City. Not every place has that population density or hospitals that are at the brink all the

00:31:02.720 time. So I do think that there are places where you can start to think about relaxing some of the

00:31:08.160 social distancing recommendations, as well as the economic shutdowns, especially things like

00:31:12.340 elective surgeries at hospitals and opening clinics at hospitals. I think that already needs to happen,

00:31:17.460 especially so in places where they're not inundated, because you're going to get other

00:31:21.140 health consequences that are not captured by the models, which are really measurable and will pay for

00:31:25.940 down the road. I do think you can look right now at the governor's lists of what is an essential or a

00:31:31.420 life-sustaining business and what is not, and look at who they're granting exemptions to and try to be

00:31:35.200 a little bit broader about that. Think about looking to see what your school system is like,

00:31:39.240 and can you open schools in a safe manner right now based on what the conditions are in your area?

00:31:43.700 Because even the whole closure of schools was very controversial and not supported by everybody

00:31:47.900 in my field. So there are things that you can do. And I do think it's not going to be one size

00:31:52.180 fits all. It's going to be dependent on what's going on locally, how much transmission do you have,

00:31:56.360 what is the antibody status of your population, what's your hospital capacity,

00:32:00.200 and what is your ability to do diagnostic testing? Do you have the new rapid test

00:32:04.220 available in many different places? All of that can help condition how we get back to normal or a

00:32:09.460 new normal, because I do think that things like mass gatherings are going to be very hard to have

00:32:13.840 for some time until we have a vaccine, because I think a mass gathering is kind of can put a town

00:32:18.500 over the edge if they get multiple episodes of transmission at a mass gathering. But I do think

00:32:22.720 that we can start taking steps, and I hope that we start doing it, because this cost is something

00:32:26.700 that is measurable, and it increases every day. And I do think that there are going to be consequences

00:32:31.520 that are not captured by our models, which are really only focused on coronavirus.

00:32:36.320 Yeah, it's an interesting point you raised there with respect to mass gatherings. I mean,

00:32:38.840 I think the German data, which just came out basically last night, would suggest that it's really the mass

00:32:45.960 gatherings that are disproportionately driving the spread versus two people having dinner at a restaurant.

00:32:52.100 Is it your view that there's really something quite devastating about concerts and live sporting

00:32:58.260 events that is not necessarily captured in going to the grocery store?

00:33:03.520 Yeah, a mass gathering brings people from wide geographic areas. If you think about,

00:33:08.240 I live in Pittsburgh, and there are Steeler fans that come from everywhere in the country to watch a

00:33:12.360 Steeler game, and then they go back to their hometowns. That's a way you can disperse things. And

00:33:17.040 sporting games are not people sitting quietly with just one person. They are social gatherings,

00:33:22.260 where people are yelling and screaming and eating and drinking, and all types of things that a virus

00:33:27.620 would look at as an easy way to get from one person to another. We know that when people shout and

00:33:31.900 scream, they make particles come out of their mouth. That can transmit. We've seen this at choir

00:33:36.040 practice, for example, with this coronavirus. Just think of that. It's at a football stadium, and you can

00:33:41.280 imagine how these types of things can transmit. I do think that mass gatherings, because of the density,

00:33:46.160 because of the fact that people come from different geographic regions, and then just

00:33:49.500 disperse, are a particular problem when it comes to communicable infectious diseases. And we see this

00:33:54.060 every year with, for example, religious pilgrimages to Saudi Arabia, where they make sure that you have

00:33:58.560 this type of vaccination. They have a whole division of the World Health Organization devoted to mass

00:34:03.180 gatherings, because we know what their role is in spreading infectious diseases. And I think that's going

00:34:07.220 to be something that's going to be a challenge to have until there is a vaccine.

00:34:11.600 And I want to go back to something you said about HKU1. Tell folks a little bit more about that

00:34:15.600 coronavirus and the concern it gives you. It's less of a concern than it gives me now,

00:34:20.420 but it's more of understanding what happens with coronaviruses. So go back to 2003, and SARS is the

00:34:27.140 first coronavirus that really hits the map as a pandemic threat, before they're thought of as

00:34:31.100 common cold viruses. Now everyone's on the lookout for any new coronavirus. And what they do is,

00:34:35.540 in Hong Kong, HKU stands for Hong Kong University, they find a novel coronavirus in some individuals

00:34:41.580 that had pneumonia. And they start looking, they find more. They actually look at bank samples that

00:34:46.240 were negative for SARS, positive for HKU. They look in other countries, they find HKU1. They actually

00:34:51.760 find it in Cleveland. In the proportion of patients that were hospitalized for coronavirus, HKU is

00:34:56.600 disproportionately found, even in patients who died or were on ventilators. And it was basically

00:35:01.120 everywhere you looked. That's interesting, because it kind of flew under the radar, because no one knew it

00:35:06.260 was there, because we do such a poor job at testing for respiratory viruses. Many times people go to

00:35:11.140 the doctor and they say, oh, you've got some virus. We don't know which one, but you're going to get

00:35:14.220 better. And that's even the case for pneumonias, because most people don't get a specific microbiologic

00:35:18.820 diagnosis of their pneumonia. So we really have this biological dark matter everywhere. And I initially

00:35:23.760 thought maybe this coronavirus was around, hidden in our cold and flu season. It clearly was hidden in

00:35:29.120 China's cold and flu season since at least November. But it doesn't appear, at least from what I've seen

00:35:33.540 now, that there was much burden of that in the United States prior to 2020, that this might have

00:35:38.560 been something that really only began in earnest in January. But I would not be surprised if you find

00:35:43.620 a bank sample that there were cases in December that were mixed in, just like with HKU1. But I do

00:35:48.660 think we would have noticed if there was a lot of these people getting really ill and ending up on

00:35:53.020 ventilators and were flu negative and negative for everything else. At least enough of them would have

00:35:58.340 raised some alarm bells, I would hope. But I do think it's an important lesson to think about with a

00:36:03.100 virus that can spread surreptitiously and you don't know about it because our diagnostic curiosity is

00:36:08.540 so bad for many infectious disease syndromes. Yeah, that's a nice way of putting it, poor

00:36:13.320 diagnostic curiosity. Have you been following the sort of natural experiment that's going on in

00:36:18.040 Sweden? Natural experiment meaning there's no randomization, but rather Sweden has sort of

00:36:22.260 elected to not shut down to the extent that other European and Scandinavian countries have.

00:36:28.180 Have you paid attention to the transmissibility in Sweden or do you have any comments on it?

00:36:32.780 I've looked a little bit at what Sweden's doing where they're trying to pursue a herd immunity

00:36:36.280 strategy. And I do think that you're going to see more cases there, which is what they're

00:36:40.260 aiming for. I'm just worried because their per capita ICU bed numbers are not very high. And that's

00:36:45.800 what we're really worried about is do you put an ICU into crisis? Do you have problems with ventilators?

00:36:50.320 And I think it will be really instructive to see if they can get through this because I know they

00:36:54.420 have a steep curve of infections, but it's all going to depend upon who's getting sick and how well

00:36:59.220 they can sequester their high risk groups, which I think is very daunting and challenging. I'm all

00:37:02.720 for cocooning the elderly and those with other medical conditions, but I know it's very challenging

00:37:06.960 because they have to interact with other people to get their food, to do other things. And that can

00:37:11.600 be really challenging. So I do think that everybody's eyes are on Sweden to see if this type of thing

00:37:15.700 can work. But I'm worried about how challenging it might be for them because there is that group of

00:37:21.020 people that are going to get hospitalized and could put a hospital into crisis.

00:37:23.880 Last thing I want to ask you is what role do you think masks are going to play as we start to

00:37:29.020 slowly ease restrictions in the coming months? Do you think that we kind of got off to a bad start

00:37:34.940 on understanding the potential benefit of an N95 mask for people in public who are otherwise at

00:37:40.660 relatively low risk? So this is controversial in my field and I tend to be someone who's not

00:37:45.780 someone supportive of masks by the public and especially not N95 masks, which I think are in short

00:37:51.320 supply, unclear whether the public can actually bear wearing them for a long period of time because

00:37:55.920 they're not comfortable to wear. And really what we saw in the beginning was a recommendation to not

00:38:02.180 wear masks for the general public because it wasn't going to protect you from getting infected to one

00:38:06.280 that's transitioned to wear a mask so that you don't infect other people. So I would say if someone 0.91

00:38:10.400 is sick, if they have a cough, if they have a fever, if they're sneezing or a sore throat, they should

00:38:15.660 be wearing a mask when they're out in public. Question is, are people who are asymptomatic, having no

00:38:20.580 symptoms, how transmissible are they if they don't wear a mask? And this is an open question, but the

00:38:27.240 CDC made a recommendation for people to use homemade masks in that event. And I'm not sure how well those

00:38:32.280 homemade masks prevent you from spreading it if you are one of those asymptomatic persons because they 1.00

00:38:36.560 don't even stop the coughs and sneezes very well based on some studies that have been published. So

00:38:41.060 I'm someone who doesn't necessarily think that these masks are going to be very beneficial and they

00:38:45.760 could be paradoxically negative because people may then refrain from washing their hands as much.

00:38:51.300 They may not social distance as much. They may contaminate other people with their mask if they 0.67

00:38:55.160 don't store it properly or wash it. So I have a lot of concerns about masks, but I think that this is a

00:39:00.020 decision that's going to be made on a political basis. And there is enough scientific controversy

00:39:04.460 that I think politicians may use it as a way to move forward in a way that allows us to open schools,

00:39:09.600 open businesses up if they have people wearing masks. But I'm not sure if we'll get much benefit from them.

00:39:14.360 But it is something that's going to be an object of controversy for some time in the field.

00:39:19.220 What are you most optimistic about today?

00:39:21.540 I would say that I'm most optimistic about the fact that we have seen plateauing in New York,

00:39:26.080 Seattle, California. We've heard about, for example, California actually taking ventilators

00:39:30.440 and giving them to other states. We've heard about Washington State dismantling their field

00:39:34.600 hospital that they made that did not see any patients. We've heard about Washington returning

00:39:39.000 ventilators, strategic national stockpile. All of that makes me very optimistic that we will be

00:39:44.340 able to meet the challenge of this virus without putting any of our hospitals into crisis,

00:39:48.400 that this is going to be a severe challenge for this country, but it's not something that

00:39:51.900 is going to break the country. And it's not going to be cataclysmic. So all of those types of things,

00:39:58.740 which are not well reported, the fact that ventilators are going back to the stockpile or

00:40:02.180 that field hospitals are closing. I think that makes me optimistic because there is a narrative

00:40:06.200 that's rightly focused on areas like New York and New Orleans and Chicago and Detroit.

00:40:10.880 But you also have to tell the good stories that there are places that were preparing for a surge

00:40:15.240 and now they're downsizing their nursing staff because less people came. So I think that that's

00:40:19.300 an important part to know that it's not going to be doom and gloom in every place and that we need to

00:40:22.860 help New York and New Orleans and Detroit and Chicago get through this. But not every city is going to

00:40:27.460 have that experience. We're going to learn from those experiences. And I think hopefully we'll get to

00:40:31.400 a better pandemic resiliency position after this. So I am generally more optimistic than most people in my field,

00:40:37.540 I think.

00:40:38.400 Mesh, thank you very much for all your insight today. I'd like to reserve the right to sort of

00:40:43.560 invite you back when the dust has settled and we have much more time to talk about what a true

00:40:47.520 preparedness strategy would look like. Because again, I do have a significant fear that as visceral

00:40:54.120 and palpable and disturbing as all of this is today, both in terms of the physical suffering,

00:40:59.960 but the fear, the economic devastation, that we're a species that is relatively hardwired to have

00:41:05.460 remarkable amnesia. I don't know. I think it would be an unmitigated disaster if two years from now

00:41:11.880 we're sitting here and this is not at all a topic of discussion and someone like you is not able to

00:41:18.580 command the type of audience of policymakers to do what's necessary. So anyway, I hope that we can

00:41:25.100 have that longer discussion when there are fewer fires burning, but when I think the stakes are equally

00:41:29.640 high. Thank you. Thank you for listening to this week's episode of The Drive. If you're interested

00:41:36.120 in diving deeper into any topics we discuss, we've created a membership program that allows us to

00:41:41.060 bring you more in-depth exclusive content without relying on paid ads. It's our goal to ensure members

00:41:46.680 get back much more than the price of the subscription. Now to that end, membership benefits include a bunch

00:41:52.400 of things. One, totally kick-ass comprehensive podcast show notes that detail every topic, paper,

00:41:58.220 person, thing we discuss on each episode. The word on the street is nobody's show notes rival these.

00:42:04.180 Monthly AMA episodes or ask me anything episodes, hearing these episodes completely.

00:42:09.280 Access to our private podcast feed that allows you to hear everything without having to listen to

00:42:14.560 spiels like this. The Qualies, which are a super short podcast, typically less than five minutes that we

00:42:20.420 release every Tuesday through Friday, highlighting the best questions, topics, and tactics discussed

00:42:25.020 on previous episodes of The Drive. This is a great way to catch up on previous episodes without having

00:42:30.600 to go back and necessarily listen to everyone. Steep discounts on products that I believe in,

00:42:36.240 but for which I'm not getting paid to endorse, and a whole bunch of other benefits that we continue

00:42:40.720 to trickle in as time goes on. If you want to learn more and access these member-only benefits,

00:42:45.320 you can head over to peteratiamd.com forward slash subscribe. You can find me on Twitter,

00:42:51.620 Instagram, and Facebook, all with the ID peteratiamd. You can also leave us a review on Apple

00:42:57.640 Podcasts or whatever podcast player you listen on. This podcast is for general informational purposes

00:43:03.820 only and does not constitute the practice of medicine, nursing, or other professional healthcare

00:43:08.180 services, including the giving of medical advice. No doctor-patient relationship is formed.

00:43:14.280 The use of this information and the materials linked to this podcast is at the user's own risk.

00:43:20.320 The content on this podcast is not intended to be a substitute for professional medical advice,

00:43:25.520 diagnosis, or treatment. Users should not disregard or delay in obtaining medical advice

00:43:31.620 from any medical condition they have, and they should seek the assistance of their healthcare

00:43:36.240 professionals for any such conditions. Finally, I take conflicts of interest very seriously. For all of my

00:43:43.080 disclosures and the companies I invest in or advise, please visit peteratiamd.com forward slash about

00:43:50.720 where I keep an up-to-date and active list of such companies.

00:44:13.080 Thank you.

00:44:20.720 Thank you.

00:44:22.720 Thank you.

The Peter Attia Drive - April 13, 2020

#106 - Amesh Adalja, M.D.： Comparing COVID-19 to past pandemics, preparing for the future, and reasons for optimism

Episode Stats

Summary

Transcript