Articles, Blog

Larry Brilliant Speaks About Bird Flu at Google

October 15, 2019


LARRY BRILLIANT: Pandemics
usually have two waves. The entire pandemic usually
lasts three years. There’s usually two
separate waves. The first wave is more
mild than the second. The first one, historically– if you look at the three
pandemics in the last hundred years– ’57, ’68, and 1918– 1918, which ultimately killed
probably 100 million people, the first wave was probably
15 million. The second wave was 85. And ’57 and ’68, which probably
killed about a million people each, as opposed
to every other year, where 600,000 are killed. So there’s a pandemic
every year. But the fluologists reserve
the term pandemic for something extraordinary. And 600,000 people dying of flu
every year is ordinary. So it’s a semantic issue. The people who die don’t care
that they died in an epidemic or a pandemic. I imagine for them, they didn’t
have much difference. The clue to any epidemic– the single most important
factor– is the density of
susceptibles. It’s not the number of people
who are susceptible. It’s the number of people who
are susceptible per cubic meter or per hundred people,
because without that, the disease can’t be transmitted. The first wave finds 100%
of people susceptible. And in that, normal influenza,
has a u-shaped death curve, where you get lots of deaths for
children under the age of one and lots of deaths for
people over the age of 70. Swine flu, or the Spanish
influenza, had a w-shaped death curve, where the maximum
number of people who died per capita were 20 to 40 year olds
who were perfectly healthy. And that’s one of the reasons
people got so scared, because it killed young, healthy people
more than it killed feeble or old people. And the reason was for a
phenomenon called cytokine storm, where the cytokine cells
are part of the immune system, and the pandemic flu
specifically infected the cytokine system. And it used the body’s immune
system against itself. So young, healthy people who
have strong cytokine systems, had more things to use against
themselves, and they died. In fact, one of the theories
is that people who have HIV/AIDS will not die
from bird flu. That was another error
in the movie. They showed people who had
HIV/AIDS, immune compromised, as being more likely to die. They’re actually probably less
likely to die, if the virus looks like– AUDIENCE: That’s quite
a [INAUDIBLE] LARRY BRILLIANT: No,
I don’t think– the question is this. If the disease uses the immune
system against itself, then people who are immune
compromised really have a survival strategy against
the virus. The question is not whether
that will actually happen. That’s pretty clear. The real issue is, what kind
of a virus will we get when the virus mutates? Will it look like
the 1918 virus? Will it look like the 1957 or
’68 virus or like the average virus we had last
year, the H3N2? But last year, six 600,000
people died and we didn’t notice. That’s a little bit of the
reason you find so much hyperbole in the whole question
of pandemic flu, because a lot of public health
people are saying, “Oh, goody. We have something that’s going
to frighten rich people. Let’s use it as a chance to
build up the public health system,” which, for
me, is a perfectly legitimate thing to do. I agree with them. So everybody’s scared now
about pandemic bird flu. Let’s use this as an opportunity
to build a surveillance system for
ordinary, garden-variety flu that kills 600,000
people a year. I don’t think that’s what
informed the movie makers, but that’s the argument you’ll
hear at every public health meeting. It’s exactly that. So yeah, I think it’s very
reasonable that people who have immune-compromised systems
are less likely to die from cytokine storm
because they don’t have cytokines, basically. They can die from
other things. AUDIENCE: Specific immune
cells [INAUDIBLE] LARRY BRILLIANT: It’s true. It’s true. We don’t know enough about
whether it’s cell mediated or whether it’s– we just don’t know, because we
haven’t had a disease that provoked cytokine storm in so
long, we didn’t have the tools to check it out. That’s exactly correct. Did anybody who’s just come
in see the movie? What did you think of it? MALE SPEAKER: I actually thought
it was horrible in terms of its production and
what it was trying to say. I thought its facts– just on
the little bit I know about avian flu, it’s facts seemed
incredibly sketchy, and therefore completely
sensationalistic, and therefore, a complete disservice
to what it was trying to do and
could’ve done. LARRY BRILLIANT: What do
you think was the most sensationalistic thing that
you think was wrong? MALE SPEAKER: I think the
component of the virus mutating again and again into
yet another, very-lethal strain was simply unrealistic. While it’s quite clear that
viruses can mutate, if the avian flu– or that particular
strain can mutate that fast, why not– and this is a genuine question
I have– why not the flu we all experience every so often? Why hasn’t that mutated time and
time again into something that’s incredibly deadly. LARRY BRILLIANT: Well, it does
mutate all the time. And of all viruses that we deal
with, it is the virus which mutates the very most.
It’s why it’s so hard to predict what vaccine
to make every year. So when you guys keep your
regular flu vaccine, it’s actually got three antigens,
not one. And it’s got three because when
people formulate it in June for an influenza season
that begins in October, they’re sort of going like this,
“Let’s see, there’s H3N2 and there’s H4N1, and I’ll put
a little bit of H5N6 in just because they don’t know which
one is going to be the predominant strain. So I think that part
is correct, that it mutates all the time. The question you’re raising,
though, is right. At the end of that movie, they
showed it mutating into a more lethal virus that killed
100% of people. Killing 100% of people is a
very bad idea for a virus. It’s just a poor survival
strategy for the virus. You don’t want to
anthropomorphize too much for viruses, but they do
want to survive. And that’s why Ebola is
never anything to really worry about. Why worry about a disease
that kills 100% of people who get it? It’s not going to go very far. It’s not going to become
a pandemic. You’re never going to see a
pandemic of Ebola, even though people are terrified
of it, because it kills 95% of people. The best survival strategy
is a virus that kills 10%, 15% of people. Maybe 30 on the upside. But that’s probably what you’re looking at in a pandemic. So I think you’re right
about that. AUDIENCE: I think what you just
said, there’s a fallacy in what you said in that it
assumes that the virus gets multiple chances to
kill all people. There’s not time for there to
be evolutionary feedback if everyone’s dead– if all Homo sapiens species
are dead, right? A virus could mutate which does
that, and so then the virus would be dead, but
we’d be dead too. So for our purposes, your argument’s not really relevant. No, but [INAUDIBLE] be able to [INAUDIBLE]. LARRY BRILLIANT: Well,
can we pursue that a little bit again? Say that again, why
it’s not relevant. AUDIENCE: You’re making an
argument that it wouldn’t be adaptive, from an evolutionary
standpoint, for a virus to kill all of its hosts, but
maybe it’s not adaptive because the virus is dead after
all the hosts are dead, but it doesn’t mean
it can’t happen. LARRY BRILLIANT:
Oh, of course. AUDIENCE: In fact, the evolution
only happens after the fact, so to speak. So there’s really nothing to
stop the virus from doing that, other than the fact that
the spreading has to work out just right in terms of
how fast you did and how virulent it is. LARRY BRILLIANT: Yeah, I think
that’s correct, except– [SOMEONE SNEEZES] Anybody who sneezes
in this room is a grave, personal risk. But I just think teleologically,
the fact that viruses like that do
not exist now– viruses that have a propensity
for becoming both lethal at the 100% level and spreading
fast, they don’t exist because they’ve died out. It doesn’t mean they can’t– AUDIENCE: As have their hosts. LARRY BRILLIANT: That’s right. It doesn’t mean that
it can’t happen. It doesn’t mean that
it can’t happen. But if you look at those things
which have caused pandemics or bad epidemics,
they have the same characteristic, which is they
kill 5% to 25% of people, and you sneeze once and everybody
around you gets it. AUDIENCE: I would purport that
the answer to his question might be that mutations which
are sufficient to cause humans to no longer be immune are quite
likely, but mutations which cause a virus to go
from being not readily transmissible to being readily
transmissible through coughing are unlikely. So that was to answer his
earlier question. LARRY BRILLIANT: So let’s do
a couple of basics here. Almost all influenza
are zoonoses. They’re a disease of birds. Almost all human influenza are
primarily a disease of birds. And almost all pandemics have
arisen as bird flu. Almost all. I can’t think of a single
exception if the word pandemic is being used to apply
to influenza. So every year, the migratory
ducks have adapted to their viruses so they live
symbiotically with them. And then, every once in a while,
there’s a reassortment or genetic shift or drift
that makes the virus fit for humans. And it can be fit for humans in
a way that only people who are exposed to large amounts of
duck shit or duck blood or goose blood can get it. Or it can be fit and mutate so
that it affects humans, and it goes human to human. We’ve had about 125 deaths
from bird flu. We’ve had, as far as I know,
three cases of human-to-human transmission. We’ve had no cases of
human-to-human-to-human transmission. We don’t know why that is. So who else saw the movie? You saw it? AUDIENCE: Yeah. LARRY BRILLIANT: What
did you think of it? AUDIENCE: Well, it looked like
they basically used the premise of the flu just as a
vehicle for telling another disaster story with all
kinds of drama. But the actual– the filming of the initial
transmission of it as very, actually, well done, in terms
of helping people visualize it, and “Oh, that’s how that
virus can spread so easily and so quickly.” But the details about
the flu– I have a son who’s in biology
AP class, and he’s doing his research project on
the avian flu. And so he was pointing out,
“OK, no, that’s wrong. That’s wrong. That’s wrong.” So we had that
side commentary going on at the time.” LARRY BRILLIANT: I’ve got a list
of about ten things that were wrong. But in general, it was
less wrong than I expected it to be. Laurie Garrett, who’s coming
Sunday night to speak to our group, was the senior
advisor on it. And when she saw it, she had
her name removed as senior advisor because she thought
it was too sensational. And Mike Osterholm, who with
Laurie Garrett, did the Foreign Affairs article,
if you saw that. He also thought it was
too sensational. But I’ll tell you, there are
things in it that were scary but are really real. Those scenes of massive bodies
being buried en masse in unnamed graves, that’ll
happen. The number of deaths
will exceed the ability to make coffins. That’s always happened
in a big pandemic. That’s just real. They made it look more gruesome
than it is, but that’s real. They used the wrong masks. They didn’t use the
right masks. And when you saw the
epidemiologist go in and touch the bodies and then go out and
cry and take the gloved hands and put in their face– AUDIENCE: Not a good idea. LARRY BRILLIANT: –there might
have been a preferred way to handle stuff like that. But the realities
are the numbers. If there is a pandemic and it
achieves escape velocity so that it goes human to human,
there’s nothing to stop it. We don’t have a vaccine. We will not have a vaccine. That part of the movie
was right. The best we can do is six
months after the virus mutates, that we can
have a vaccine. Much more likely, it will
be eight months. And even then, we won’t have
enough vaccine for everybody in the world. The president’s plan has $90
million in it for global surveillance. So we’re not going to find
the first cases. That’s not enough money to
find the first cases. And there’s almost no money in
the plan for the states and counties to have preparedness. So if you talk to Mike Leavitt,
who’s the Secretary of Health, he will tell you
quite candidly, if there is a pandemic, everybody’s
on their own. He says that openly. Everybody is on their own. Every state’s on their own. Every county’s on their own. So that part was all accurate. The numbers that they
talked about– 30 million to 150 million–
that’s pretty much what most experts think would
be the number of people who would die. One to two billion people
are likely to get sick. Those are the numbers. And if you think about it,
a new virus in a naive population, what’s going to stop
people from getting it? AUDIENCE: So I didn’t see the
film, but I guess the question that I would have is, what can
and should people do to decrease the likelihood
that something like that can spread? LARRY BRILLIANT: Well,
there’s lots. Social distancing, which is the
whole category of things that includes hand washing,
masks, closing schools, closing churches, doing
telecommuting, not going to work, but figuring out at the
workplace ways to be able to carry out work without actually
congregating in crowded places. Those were all very
useful strategies. They don’t have any guarantee of
protecting any individual, but they certainly have
a societal effect. The way in which you organize
your life, whether you stockpile food or you keep
antivirals available, what kind of virus it is and which
antivirals do or don’t work is going to be critical. You’ve all heard, I think, that
Tamiflu doesn’t work, or that Amantadine doesn’t work
against the current flu. But the truth is a lot more
subtle than that. The outbreak of H5N1 in Turkey
is sensitive to Amantadine, which is the cheapest
antiviral. You can get it at
any drugstore. It won’t work against this
current crop of regular H3N2 flu, but it worked
fine against the H5N1 Turkish strain. It doesn’t work against
the Chinese strain. AUDIENCE: This was
a bird strain? LARRY BRILLIANT: Yeah. Influenza viruses are named by
two surface proteins, the H and the N. The H is
hemagglutinin. The N is neurominidase. The H is the safe cracker
that gets the virus into the human cell. The N, or the neurominidase, is
the escape artist that gets the virus out of the human cell
after it’s replicated. So because they have different
roles to play in transmission, then you have different
strategies in making antivirals against them. So the current garden variety
of flu this year happens to be H3N2. And the worrying
strain is H5N1. It doesn’t matter what
these things are. We don’t really understand the
differences very much between an N1 or an N3, except H5 and
H7 tend to be more lethal. H3 tends to not be
quite so lethal. AUDIENCE: And I assume this
is a class thing. H5 isn’t a specific protein
[INAUDIBLE]. They’re not all identical. LARRY BRILLIANT: They’re
not all identical. That’s right. If you think that the genetic
equivalent would be alleles, these are expressions
of alleles. So you can get strains of H5N1
where the proteins look to be the same, but they have
different immunogenicity. Amantadine, which is the
antiviral that’s used for most flus doesn’t act on either the
H or the N. It acts on the core and it’s called
an M2 antiviral. And all the other antivirals
you hear about, which are Relenza and Tamiflu, they’re
neurominidase inhibitors. They work on the exit
part of the virus. What other strategies
might you have? We don’t have a vaccine. We have three extremely
good antivirals. We don’t know whether the strain
that will arise will be sensitive or not, or resistant
to those antivirus. And then you have social
distancing, which are hygiene and the way that we organize
ourselves as a community and as a world. And other than that, we
don’t have a lot of arrows in our quiver. We can have restrictions
on travel. That’s part of social
distancing. We can organize our company so
that people don’t come to work but work remotely. But a pandemic literally means
everywhere all at once. If there is a new influenza– H5N1 strain in China today– it will be everywhere in the
world within three weeks. And it’ll be some places
within 72 hours. So that part of the
movie was right. The weird stuff was
over the top. And I think it took away
a lot from the movie. First of all, what’s
the status today? We really have a lot of
epizootics, which just means disease of animals. We don’t have any pandemic. We have very little
person-to-person spread. Even that’s controversial. We have a steady increase in
the number of human cases. We have a case fatality
rate of about 50%– about 125 deaths, about twice
that many cases– overall, from the
very beginning. We’re at Category 3. WHO has a series of categories
that are a little bit like hurricane categories, where
you start off with normal emergence of a new virus. You don’t get any human
cases at all. Then Category 2, you get human
cases, but you get no spread. And then, this pandemic alert
Stage 3, which is where we are right now, there’s no or very
limited human-to-human spread. And then, Stage 4, when you
start getting sustained human-to-human spread. If you hear WHO declares
Category 4, the world as we know it will change. People will not get
on airplanes. Transportation stocks
will go down. The just-in-time inventory
systems of all companies will be challenged, because people
will change their behavior. We haven’t really had that since
1968 did we get to a Category 4. And Category 5 is just like
a Category 5 hurricane. AUDIENCE: This 50% death
rate, is that accurate? Or I assume the cases that
actually get reported are more likely to be the serious ones. Because if I come down with
a cold-like thing after I [INAUDIBLE] chicken
[INAUDIBLE] LARRY BRILLIANT: I don’t know
exactly what kind of human-chicken relationships
you have, so I can’t– but I do worry. I think that’s exactly right. I think that deaths all get
reported, and light cases don’t get reported,
and asymptomatic cases don’t get reported. And I would imagine that the
death rate will go down to somewhere between 4% and
10%, ultimately. It may already be that, and we
just don’t find the light and asymptomatic cases. And that’s exactly right. And as I said, I don’t think 50%
death rate would be a good survival strategy
for the virus. If the virus is having a
strategic retreat, as we are, at google.org, trying to figure
out what case fatality rate it should have, I would
recommend four to 10, rather than 50%. So this is another way
of looking at that. We’re sort of here right
now, where we have this extended Phase 3. These are the four scenarios
that we have been wrestling with. What would it be? Could it be just a little
dribble, or could it be something really sensational? And every company that has
gone through a pandemic preparedness plan has used some
variant on these four scenarios to make its plans. Part of the problem that we have
is an avian problem in the sense that we have two
species of birds that we have to deal with. We have Chicken Littles
and we have ostriches. You’ll see on television last
night were a lot of Chicken Littles, people who’re
saying the sky is falling, the sky is falling. We have just as much trouble
with people who say there’s no problem at all. They stick their head in
the sand and they say it’ll all blow over. I think both of those
communities are causing us trouble. But if I could, let me make
the case for each of those different birds. Let me pretend I’m the lawyer. I’m making the case for
Chicken Little. The argument would be,
well, we have more and more human cases. We now have 125 deaths. More countries are affected. We’re seeing more deaths
every day. Multiple species are infected. You may not know this, but 25%
of all the tigers in Vietnam have died of bird flu. AUDIENCE: 25% of
all the tigers? LARRY BRILLIANT: All the tigers
have died of bird flu. We don’t know if it’s because
they ate infected chickens or there has been tiger-to-tiger
transmissions. It’s very hard to get tigers to
sit still for interviews, but we’ve tried. So we’ve lost tigers. We’ve lost civet cats. We’ve lost mice. We’ve lost regular cast. We
haven’t lost any dogs yet. We’ve lost pigs. So any time a virus begins to
show the ability to affect multiple species, that’s
not a good sign. That kind of pleomorphism
suggests that the virus is able to be fit for lots
of different species. AUDIENCE: Primates? LARRY BRILLIANT: All primates. Birds, you know, yeah. AUDIENCE: Like monkeys,
chimpanzees. LARRY BRILLIANT: It
has definitely affected monkeys, yes. And humans, of course, which,
when last I heard, most of us are still primates. The argument for Chicken Little
would include that the timing is right for
a pandemic. Pandemics seem to occur
every 50 or 80 years. We don’t have vaccines. We don’t have antivirals. And globalization has made
it more likely– not less likely– that there would be a pandemic
because in the 1918 eighteen pandemic, the virus went
around the world four times in one year. And there were no airplanes. Now with airplanes, it could
move around the world four times in about a week. Just-in-time inventory systems
are precariously perched in the event of a pandemic. Well, there’s also the case for
the ostriches, who say, don’t worry. There’s nothing here. No long-term,
human-to-human-to-human transmission documented. We overreacted before– Y2K. We overreacted with swine flu. We particularly overreacted
to swine flu. We wound up killing more people
with the vaccine than we did because of the disease. Why would you trust these people
who are bringing you bird flu now? We have better medical care. We can take care of people
in case they get sick. We understand how to deal with
these very bad pneumonia symptoms. Most people who die of
influenza die of pneumonia. So another strategy to survive
is to get yourself vaccinated against pneumonia. Right now we have a
very good vaccine. I recommend anybody who’s
thinking of it– young, old– get yourself vaccinated
against pneumonia. 40% of all the deaths from
influenza are actually deaths from pneumonia. Pneumonia becomes an
opportunistic disease. AUDIENCE: Is that true
of bird flu? LARRY BRILLIANT: Yes,
as far as we know. We don’t know which bird
flu is going to become the bird flu. The one we’re looking at right
now– the one in birds– has the characteristic of
affecting the deep cells, deep in the parenchyma of the lung. And that, in fact, may even
be more likely to cause secondary pneumonia. AUDIENCE: But the question
was the existing deaths. Were those pneumonia? LARRY BRILLIANT: Yeah,
almost always. It’s almost always pneumonia. It is interesting that
you do get a viremia. We have retrieved virus
from the bloodstream. We’ve retrieved virus from
feces and from the gut. That’s unusual. You don’t usually expect
to find in influenza. So this is a different
kind of a virus. It seems to go in more and
different places than we would’ve expected. The biggest argument against
this being a really big thing to worry about is there’s
no World War II. You don’t have young people in
trenches who have very good immune systems to get
cytokine storm. There’s no wartime censorship
to hide epidemics. So having said that, remember
the Spanish flu? Who knows why it was called
the Spanish Flu? Because it appeared first
in Spain, right? Wrong. Where did it first appear? Does anybody know? Minnesota. Why is it called the
Spanish flu? Because Spain was the only
country that was not under military censorship because of
World War I, and it had a free press, so the first reports
that came were from Spain. So you had an epidemic of
reports, not an epidemic of the disease. So Spain forever got tainted
with it being the Spanish flu. No good deed goes unpunished. Because the Spaniards had an
open and free press, they’ve been tagged with this. AUDIENCE: Wait, now,
what do you mean it was from Minnesota? It wasn’t a Southeast
Asia-originiated– LARRY BRILLIANT: Not at all. Not at all. AUDIENCE: So it came
from birds– LARRY BRILLIANT: 1918. We don’t know where
it came from. We just know where it
was first found. People are in denial and
they have fatalism. What can I do? If it happens, it happens. So I’m giving you the legal
case for the ostriches. And I think what we really need
to find is a different kind of bird. We need to be really
practical. We need to take a look at the
truth of bird flu, understand what really causes it, what our
real risks are, and how we should react to it. We shouldn’t get carried away,
either, by the kind of craziness that you worry about
it too much, nor should we just pretend it’s not out there,
because it’s something in the middle. So in an effort to do
that, in October– and I’ve got copies here of
Harvard, Business, Review, which reports a meeting that we
had in November, where we got all the top fluologists in
the country together, and we actually had them come
to San Mateo. And we asked them, we did
surveys, we did workshops, we did retreats, we tried to figure
out what was going on in Vietnam at the time and
Indonesia, and we asked people, what do you think
is the most likely scenario going forward? And I’ll show you exactly what
we asked them, and then you can read the report
if you want. But we asked the leading experts
in the world what were their estimates of the
likelihood of the disease, the severity, the mortality,
morbidity, when it’s likely to occur. What were the best
interventions? How likely is it that we’d
have vaccines, again, for humans and for poultry. How likely is it we’d
have antivirals? What good would social
distancing do? How do you deal with business
continuity problems? And how do you communicate
risk? So those were the variables. And we have a long
study of it. And we published it in several
different places, and you can find it. This is our model. It’s a very simple model. All the engineers will grasp it
in seconds and the rest of will spend our lifetime
trying to figure out what the hell it is. But we just tried to watch all
the different variables to know what causes disease, what
affects rate of spread, what affects absenteeism, how can
you influence the number of people who are sick, the
efficacy of medical care, looking at social costs, health
care costs, as well as non-health-care economic
costs. You don’t care about
any of that. But we then asked people, what’s
the likelihood that we’re going to actually
get a human-to-human-t o-human-efficient virus, one
that’s capable of being propagated through at least
two epidemiological generations? And the experts, the median, was
that 15% said that within the next three years, we
would have a pandemic. Now, it’s not 100%. It’s not 80%. But it’s not zero. So 15% of the experts felt that
within the next three years, we’d be facing the
kind of an epidemic that you saw in that film. without the hype. It’s a data point. It’s not nothing. It’s not everybody. And then we said, when do you
think it’s likely to happen, and what do you think the
chances that it’ll happen at different points in time so that
we could do a cognitive map of when it was likely to
occur, at least what was in the minds of people there? And you can see that
almost everybody– up to 90 some odd percent– believed that we will have a
pandemic in our children or our grandchildren’s lifetime. That sort of what you’re
looking at. 15% said we’ll see it in
the next three years. Almost everybody said we’d
see it in our kids or our grandchildren’s lifetime. And you’ll hear people say it’s
not a question of it, it’s a question of when. If there were a pandemic, how
many people would get sick? And what’s your best case and
what’s your worst case? And the experts who were there
said the best case was 650 million people would get sick. And the worst case
was 2.8 billion. So that’s a big range. But in no case is it
a nice number. What about deaths? Between 25 million and
165 million will die. So 15% of the experts felt that
within the next three years, 25 to 165 million
people will die. And 90% felt that within
our children or our grandchildren’s lifetimes,
25 to 165 million people would die. That’s what motivates me to
be serious about bird flu. My lifetime is closer to the
end than it is to the beginning, but my kids and my
grandchildren, that 90% of my colleagues think they will have
to experience that is a reason for me to be concerned. And if that happened, how many
would get sick in the US and how many would die in the US? Yes? AUDIENCE: What are the
quoted experts? LARRY BRILLIANT: Oh, well, we
had two classes of experts. I’m sorry about that. We had epidemiologists who had
worked with influenza for 20 years or more, who had been in
field investigations, who understood the biology
and virology. And we had epidemiologists who
had worked in cholera or worked in polio and didn’t
really have specific expertise but would be better informed
than the average but not that close to it. Sometimes, being too close
to it, as you know, is a deficiency. Too much expertise is, as I
think we’ve seen in Google, sometimes a formula for
loss of creativity. If that were the case, how many
would get sick in the US and how many would die? And the experts– you can see the ranger here. Here, it’s almost the
same, isn’t it, the two classes of experts? AUDIENCE: Actually, this
is kind of surprising. Why are the– oh, so this isn’t deaths. LARRY BRILLIANT: These
are just cases. And through it all, you’ll see
that there’s sort of an implicit expectation that
the case fatality rate will be 4% to 10%. How many will die in the US? Here you see some variability. But people believe– the experts, the ones
who deal with flu– think that as many as five
million and as few as 300,000 people will die. And this is over a very
short period of time. That’s actually catastrophic
to the US. You think of a million people
dying in six months from any disease, you think of bird
flu on CNN 24-7. We’re not, as a society– you get that many people who
have scratches on their knees, and we slow down
and get scared. And here’s the case
fatality rate. You can see that most people
here are looking at case fatality rates in the three
to five to seven range. It doesn’t matter which group
said what, but that’s sort of the range that people
think that this thing will end up at. Now here are some really
sad expectations. If it happens, do you think we
will have enough effective human vaccine to vaccinate
even one third of the population? Only 1% thought we would. Nobody thinks we will
have vaccine. What about antivirals. The same. nobody thinks we will have
sufficient antivirals to treat one third of the world’s
population. Yes? AUDIENCE: Is that because the
vaccine just can’t be produced quickly enough after
the thing starts? Or is it because people
are insufficient? LARRY BRILLIANT: Well,
it’s both. And you know that Kleiner
Perkins now has a fund where they are trying to put $200
million into a fund which, by the way, the fund was stimulated
by this meeting. John Doerr attended
the meeting. Beth Seidenberg attended
this meeting. And that fund is intended to
try to find the gaps in our ability to respond
to a pandemic. And obviously, one of the
biggest gaps is you can’t make a vaccine until you first
have the antigen, which is the virus. You don’t know what the antigen
is going to be until the virus mutates. So the stopwatch doesn’t start
until the virus is there. And you can’t get it done with
current technology in any less than six months. So what they’re trying
to do is to get it done in two months. That’s just a production
of it. That’s not the distribution
of it. Here’s the real question. In the middle of a pandemic,
when you’ve got no airplanes flying, and you’ve got no trains
going, no boats going, how the hell are you going to
ship this stuff all over? You’re going to have to
manufacture it in multiple different sites all
over the world. So you’ve got a logistic
conundrum. AUDIENCE: Well, once you have
enough, you can vaccinate the cargo plane pilots
and what not. LARRY BRILLIANT: Yeah,
absolutely. Absolutely. But you have all these issues
that add complexity to complexity. And you’ve got to make
sure that it works. And people have to believe
that it works, which is another issue. Yes? AUDIENCE: [INAUDIBLE] LARRY BRILLIANT: Well, there’s a
couple of things that people can do to protect themselves
that may not have much of a role in stopping the pandemic. And some of these are
speculative, and I’m not recommending them until the
science gets a little clearer. How many of you take
Lipitor or any cholesterol-lowering drug? You don’t have to tell me
if you don’t want to. But if you do, that seems to
be a very effective way to reduce deaths from bird flu. What? Makes no sense at
all, does it? Well, maybe it does. Maybe the way that Lipitor and
other cholesterol-lowering pharmaceuticals work
is they reduce– they’re anti-inflammatory
agents. That’s maybe how they work. There’s a 40% to 50% reduction
in deaths from influenza for people who are taking Lipitor
or any of the other cholesterol-lowering drugs–
any of the statins. I can’t explain that. That doesn’t make any sense. But if you talk to the people
at CDC, all the epidemiologists are taking
statins now. So some people are interested
anecdotally in that. I think more than anecdotally,
taking an immunization against pneumonia– I have– it’s unlikely to hurt you, the
side effects are so low. It’s very likely that if you
were stuck in a pandemic, that’s the most likely
cause of death. It’s not the only cause of
death, but if 40% of all the deaths from previous influenza
outbreaks are due to pneumonia, if you’re wealthy,
if you can afford it, if you have access to health care– AUDIENCE: Well, this is
bacterial pneumonia. LARRY BRILLIANT: It’s both
bacterial as well as H. flu. It’s Haemophilus influenzae
primarily, but it’s– AUDIENCE: So you can’t
just a Z-Pak? LARRY BRILLIANT: You can’t
just take what? AUDIENCE: I mean, if it’s
bacterial, you could just take a Z-Pak or something
[INAUDIBLE] LARRY BRILLIANT: Yeah, I
mean, you can take– if you have bird flu and you get
a bacterial pneumonia, you can certainly treat the
bacterial pneumonia with antibiotics. Absolutely. AUDIENCE: [INAUDIBLE] LARRY BRILLIANT: But I mean,
why wouldn’t you want to be immunized against it if
it was a protective– it’s not a good epidemiological
strategy. It’s not going to impact the
way that the epidemic goes, but for personal health, it’s
just something to talk to your doctor about. AUDIENCE: Actually, it would
probably help with the transmission. LARRY BRILLIANT: It might
increase transmission because you are healthy, worried sick,
and you’re walking around. Yes? AUDIENCE: Do you recommend that
for children, as well? [INAUDIBLE] LARRY BRILLIANT: Yes. I personally do, and my kids
have been vaccinated, but you’ve got to talk to your
doctor about stuff like that. So what are the most important
consequences of a worst-case outbreak other than morbidity
and mortality? Well, you can figure out what
most of these things are. Commerce disruption,
health care disruption, food shortages. You saw pictures in the film of
people fighting over food, people willing, almost, to get
violent over scarce food. I think that’s probably
realistic. I think you do see stuff like
that happening in food shortage situations. I’d be surprised if we didn’t
see stuff like that. It looked staged and badly done,
but that didn’t seem unrealistic to me. AUDIENCE: [INAUDIBLE] during the
Black Death and during the Spanish flu, social
order and social institutions were preserved. LARRY BRILLIANT: That’s
not completely true. Actually, In each case, they
gave rise to new social institutions which destroyed
the old ones. The guild system arose in
response to the Black Death. It didn’t survive the Black
Death, for example. I can’t think of an epidemic
that didn’t change the way society was organized– a big one. So I would say that it’s more
likely that it’ll change society than that these
institutions will survive. It depends. If you’ve got 200 million people
dying in the world, it’s going to change
a lot of things. All right, so here’s
the summary. 15% of experts and 60% of
non-flu experts think that there will be a pandemic
within three years. Nearly 90% think there will be
a pandemic in our children or our grandchildren’s lifetimes. They think, as a group, that
there would be 650 million to 2.8 billion people
would get sick. 3% to 7% of those people
who got sick would die. We won’t have vaccines. We won’t have antivirals. And the non-medical social
costs of a pandemic are estimated to be somewhere
between $1 trillion and $3 trillion worldwide. So I’ve got a lot more stuff on
the biology and physiology, but that’s sort of where we are
with what experts, who’ve been studying this all
their life, think. They’re going to be wrong. We just don’t know in what
direction or how much they’ll be wrong. But that’s the state
of the art right now, what people think. AUDIENCE: So another
Chinese [INAUDIBLE] vaccinate all their chickens. Will that help? LARRY BRILLIANT: It’ll either
help or it’ll hurt. It’s hard to know. I would say most people think
it’ll hurt, because it will create the likelihood of a
surviving virus being immune to Tamiflu. AUDIENCE: [INAUDIBLE] LARRY BRILLIANT: Yeah, because
what the Chinese have done, first of all, is they’ve ground
up Tamiflu and fed it to their chickens. Now they’re trying to use a
vaccine on top of that. Do you know how many chickens
there are in China? Here’s an interesting
statistic. Anybody here 30 years old? AUDIENCE: 35. LARRY BRILLIANT: So 30 years
ago, there were a billion people in China. Now there’s 1.35
billion people. That’s a growth rate of 35%. 30 years ago, there were 12
million chickens in China. Today there are 15 billion. Just wrap your mind
around that. So that’s sort of the problem. The Chinese government has said
they will vaccinate 15 billion chickens. They’re not going to be able to vaccinate 15 billion chickens. Besides, 15 billion is
the wrong number. It’s 15 billion that are
produced every year. The life of a chicken
is two months. So you vaccinate all of them
today, there’s going to be a new crop tomorrow. You’re going to have to keep
on vaccinating like that. AUDIENCE: So one more thing. So how about trying to change
the people’s behavior with respect to birds to lower the
risk of cross-species– and therefore, more
opportunities for the viruses to make [INAUDIBLE] That would destroy the chicken
economy in most of the world, I guess, but maybe that’s
maybe the price you pay. LARRY BRILLIANT: I think one of
the biggest problems is you can’t expect peasants and poor
farmers to give up the protein for their children in order to
stop something that they can’t see and will help people who
live 10,000 miles away. That’s just not fair. It’s not reasonable. We wouldn’t behave like
that if we were put in a similar situation. I watch these American pundits
who get mad because the people aren’t willing to cull
their birds. Well, cull is a very
soft word. But they’re really saying kill
their livelihood and deprive their families of protein. So it’s kind of hard to– so we put in place for FAO a
chicken exchange program. Any farmer who delivers a sick
chicken can have back, no questions asked, 1.25 healthy,
vaccinated chickens. And 1.25 because we didn’t want
to make it three chickens for every sick chicken. That would incentivize people
to make their chickens sick. So if you have a sick chicken
and you deliver it to an FAO office, you’ll get back 1.25
healthy, vaccinated chickens. Just enough of an incentive to
make it worth your while, but not too much of an incentive
to cause you to get your chicken sick. And that seems to be a good,
practical strategy. AUDIENCE: Will that help? LARRY BRILLIANT: Yes. Yes, it’ll help. In 1997, there was
an outbreak of– they said 1996 in the movie, I
think, but it was really 1997. In 1997, there was an outbreak
of bird flu in Hong Kong. And in response to that
outbreak, they killed 1.5 million chickens in 72 hours. It stopped the outbreak
in its tracks. And the woman who ran that, Dr.
Mabel Cheng, is now the head of WHO’s pandemic
office for influenza. She’s terrific. And that kind of aggressive
and difficult and painful executive decisions
are needed. And what she did was
unbelievable. It was very, very difficult
to stop it. And we’re trying to build global
surveillance systems like the one that I’m working
on and that Google is incubating over in our building
called InSTEDD. We’re trying to build an early
warning system that can find outbreaks of novel diseases like
bird flu or SARS or Ebola early enough that the world
can respond to it. So we’re still working on it. AUDIENCE: So it sounds like a
pandemic is, in some sense, inevitable. But our optimum strategy is to
delay it as long as possible and hope that technology and
economic infrastructure advances to a way so that we
minimize the impact, I guess? LARRY BRILLIANT: Yeah, we call
it the speed bump strategy. I think that’s it. You’re trying to put
speed bumps in the path of the pandemic. I think that’s exactly right. AUDIENCE: [INAUDIBLE] you could also have more
than one [INAUDIBLE] LARRY BRILLIANT: You could have
more than one pandemic? AUDIENCE: I mean, if you were
able to catch one early, you could actually stop it, even
though it’s slightly [INAUDIBLE] LARRY BRILLIANT: Yeah, well,
here’s the real question. Harvard School of Public Health
published a study in December which assumed that
the virus did multiple mutations in multiple parts of
the world at the same time, which I thought was a pretty
dumb thing to assume, because it hasn’t done one. Why do you think, all of the
sudden, they’ll do seven or eight or nine, all spread
out all over the world. In that case, there’s really
nothing to do. You’re not going to be able
to find them all. You’re not going to be able
to respond to them all. But if the virus behaves the
way we think the 1918 virus behaved and the ’57 and ’68
viruses behaved, we will be able to find the
first outbreak. It’ll just be at what time
will we find it? When there’s one case? When there’s 100 cases? When there’s 1,000 cases? When it’s spread to other
countries or not? Two excellent studies done in
Science and in Nature by Ira Longini, who’s a modeler,
suggests that if we could find the first outbreak within 21
days and saturate the entire area around the outbreak with
Tamiflu or any other neurominidase inhibitor, we
could stop the whole thing. And that’s a bit of
a controversy. But certainly, it’s
worth trying. And you can slow it down. So it’s either speed bumps
or you stop it. AUDIENCE: But will we ever reach
a point in technology where we won’t have to worry
about the threat of a pandemic, because we will the
early warning systems, we’ll be able to quickly identify,
come up with antivirals, and then just stop it. LARRY BRILLIANT: The only virus
that I don’t think we have to worry about is smallpox,
because we’ve eradicated it. Until we eradicate polio, we
still have to worry about it. We’re this close to eradicating
polio, yet last month, there were six
cases in Minnesota. Just out of the blue. Viruses are hard to deal with. The only strategy against
a virus that works is eradicating it, in my opinion. And probably, influenza is the
hardest one to eradicate. AUDIENCE: But there’s a
difference between existing at a low level and us being able to
stop it from crossing those escape velocities,
as you called it. So will we ever get to the point
where, for any viral instance, we’ll be able to stop
it before it gets– well, basically, can we ever make
pandemics go away? LARRY BRILLIANT: I don’t
think that’s possible. AUDIENCE: Never? There’s always going
to be a threat? LARRY BRILLIANT: I’ve been
reminded because the way we’re planning in our google.org
planning mechanisms right now is that we’ve been challenged
to remember this example. Before you say never about
anything, in 1885, when immigrants poured into New
York City, as immigrants poured in from all over the
world, more food was consumed, milk was consumed, more milk
wagons delivered milk to people’s homes with horses
bringing the milk wagons. There were more horses, more
horse-driven carts, and more horse shit. So a very serious scientific
study was done calculating how much horse shit would be
deposited on the streets of New York from all the immigrants
and demanding all the new services to make
New York uninhabitable. And all the best scientists
in the world at that time predicted that by 1905, New York
would have too much horse shit produced to be able
to take it away. It’d be uninhabitable. And then what happened? Somebody Invented a car. So the challenge always
is, don’t get stuck in the horse shit. Invent a car, right? AUDIENCE: Yeah, sure. LARRY BRILLIANT: So that’s
sort of the game-changing invention that we would
need to have, the game-changing idea. I would never give up on the
possibility that we’ll have one, but absent such a
game-changing invention, it doesn’t look good that we
could stop pandemics. [INTERPOSING VOICES} That’s right. Now we have a whole new
kind of horse manure. So now you have to invent
a green horse. I just wanted to have this in
case people saw the movie, got scared, wanted to
talk about it. Many of you, I think, sent me
emails that are more concerned about bird flu than the
average person. Yes? AUDIENCE: [INAUDIBLE] very
personal question. I’m going on a tour of Turkey
for two weeks next Saturday. Should I go? LARRY BRILLIANT:
Should you go? Sure, why not? AUDIENCE: I thought you just
mentioned [INAUDIBLE] places where– LARRY BRILLIANT: I
mean, the odds of catching this right now– I think, what have there been,
28 cases in Turkey? And what’s the population
of Turkey, 60 million? Your odds are very low, and
you’re not going to be in the places where the bird flu was. And there isn’t any
more anyway. But my son, who goes off into
China and goes to the area of Qinghai, which is the bird
lake around which it is hypothesized that there’s a lot
of likelihood that bird flu might emerge. When he goes, I send him with a
little pocketful of Tamiflu and amantadine, and I say call
if there’s a bunch of people getting sick around you. No, I don’t think you should
change anything. I think it’s a middle course. You have to be conscious
and aware. You can’t get crazy
about this stuff. You can’t let it paralyze
you, but you can’t pretend it’s not there. AUDIENCE: [INAUDIBLE] LARRY BRILLIANT: Yeah. To me, this is the challenge
for any executive decision making under conditions
of uncertainty. And each one of us in our
lives have got to do it. It’s just really complicated
getting it in a 15-second sound bite on television. It’s actually not an
insurmountable amount of information to absorb
and think through. It’s just the world goes so
fast and we have so little time for this that we’d like to
pretend this one goes away. But you’ve got a small, but
non-zero probability, of something terrible happening. And there are some small, but
of unknown effectiveness, interventions that a prudent
person would take. And that’s sort of where
we are right now. It’s unsatisfying, but that’s
the truth of it. Thanks very much for
coming over here.

No Comments

Leave a Reply