Articles, Blog

COVID-19 Update 8: This is probably the most important picture of the whole coronavirus epidemic

March 13, 2020

I hear a lot of people say, “Oh, COVID-19
that’s just as dangerous as influenza.” Well, it’s not. It’s way more infectious
and it’s way more lethal. Let me show you why. A patient with influenza will infect,
on average, between 1.4 and 1.8 other individuals, someone who is infected with
SARS-CoV-2, who has COVID-19 the disease, will infect between two
and 3.1 other individuals. So we’re looking at an
exponential growth curve here. Every three to four days, the number
of infected individuals doubles. Here are the total cases excluding
mainland China from worldometer. We can see that today, it’s March 11th,
we have 38,170 cases outside China. And four days ago – one, two,
three, four, we had 21,399. So roughly a doubling time of
four days, maybe even a bit more. So it seems like the disease spreads a
little slower than initially described. Yeah. We see a number of three days on let me show you what you can do in order to assess the likely
scenario that’s going to occur in your area, in your city, in your state,
based on the example of Austria. That’s the country I live in. It’s a small country of
8.8 million inhabitants. What’s the situation here? So in Austria, we had 29 cases on
March 4th and 157 cases on March 9th. Let’s calculate the doubling time. So March 4th to March
9th that’s five days. We had 157 cases on March 9th so 29 that’s
the number on March 4th times two to the power of X. X is the number of doubling intervals. Let’s solve that for X. X is equal to 2.4 for doubling intervals. So there are 2.44 doubling
intervals in these five days. Now what do we need to do in order
to calculate the doubling time? Well, we need to divide five by
2.44 and that’s equal to 2.1 days. That’s the doubling interval in Austria. So that’s a bit scary because it’s
actually quite significantly lower than in the rest of the world outside China. So I took the number of cases that we
had on March 9th, and I extrapolated the development of cases taking a
doubling interval of three days. In order to see how many cases we’re
likely going to see in the immediate future, if the doubling interval is not
prolonged through social distancing and other public health interventions. So as you can see on this graph in 45
days, we could see 4 million Austrians being infected at a doubling
interval of three days. And that’s when we would reach herd
immunity, at which point the epidemic would likely peak and the number
of cases would go down again. But obviously that would only happen if
we wouldn’t do anything against it if we wouldn’t implement any measures like
social distancing, avoiding crowds, reducing unnecessary travel, and so forth. So that’s the worst case
scenario, if you will. So if we don’t pay attention in Austria,
then in 18 days we would exceed the number of cases that we currently see in
Italy, which was 9,172 on March 10th. So obviously the situation in Italy, which
is a neighboring country of Austria is pretty bad. And if I tell people that number, they’re
actually quite scared and it’s a bit of a wake up call to them. What you can do next is to look at how
many severe cases or critical cases you’re likely to encounter in your area. We know that 81% of cases of COVID-19 are
mild, 14% are severe, and 5% are critical. But critical likely need a bed
in the intensive care unit. So what I would do next is to look at the
number of hospital beds that we have in Austria, and you can do
the same for your area. And we can see that in Austria,
we have 67,000 hospital beds. But what’s more important is to look at
how many intensive care unit beds we have in Austria. So I found this article, and this article
says that we have 23.4 beds in the intensive care units
per 100,000 inhabitants. So we have 23.4 ICU beds per 100,000 in a
country that has 8.8 million inhabitants. How many beds are those? That’s 23.4 times 88 at equals 2059 beds. But obviously these
beds are not all empty. There are patients in them. So if we assumed, and that’s already a
pretty high assumption, that we could free up 50% of these beds
for COVID-19 patients. Then we would have 1000 beds available. Now, if we assume a doubling interval of
three days, then in 21 days we would reach 1000 critically ill COVID-19 patients. This means that in 21 days, this capacity
of 1000 free beds, 50% of the entire ICU capacity would be reached. After this time, there will be no space
for critically ill patients with COVID-19. Let’s look at the incubation period. When are these people
getting sick actually? Well, as we’ve seen in a previous video,
the incubation period of COVID-19 is 5.2 days on average. So let’s put that into this graph. So the patients who won’t get a bed here
will actually be infected somewhere around here. So this time period that we’re in right
now is super critical if we want to implement efficient
public health measures. So in order to persuade your public
health authorities to act, and in order to persuade the population to act, you would
need to take the numbers of your state of your city and do a similar calculation. I would also present this curve here,
which in my opinion, is probably the most important graph of the entire epidemic. Now I want to give credit to Carl
Bergstrom for creating this graph. Many people have actually contributed to
this graph and it’s been widely shared on Twitter, and the reason why I think it’s
so, so important is that it drives home the point elegantly of why we need to act
and why we all need to work together in order to spread out the epidemic. We absolutely need to
avoid this spike here. We all need to work together to
prolong the doubling interval. Here you see the capacity of the health
care system, if we don’t work together to create a curve that looks more like this,
we will arrive at way higher mortality rates than have been
described in the literature. On the other hand, if we manage to extend
the curve through measures like hand washing, teleworking, limiting large
gatherings, minimizing travel, no shaking hands, if we do all that, then we will be
able to extend the curve to look something like this without overstretching
the healthcare system too much.


  • Reply Jan Olsen March 12, 2020 at 5:12 pm

    Yes, BUT. Only if each individual remains infectious for 45 days, which is unlikely. Once the infected recover and no longer are infectious they will no longer infect or, presumable get reinfected. Therefore, it unlikely that the exponential growth will last 45 days.

  • Reply G4MERSCF March 12, 2020 at 5:16 pm

    Hello from Mexico, I'm from Guadalajara and it's disgusting how people minimise COVID-19 saying that it's nothing dangerous, the medical budget of this year it's less expensive than the last year, and people are saying that it's unnecessary used protection on mouth (in Spanish it calls "mascarillas" i don't know how to say it on English) or washing hands, I think that a Lot of People will die in México, and really I'm scared about what will happen here.
    But thanks for your videos, good luck.

  • Reply Maria Bernardes March 12, 2020 at 7:06 pm

    Thank you for the valuable information and support! I am Portuguese do you authorize me to share your presentation here in my country via internet more specifically Facebook? Media is not informing correctly….

  • Reply Hiroshi loves You March 13, 2020 at 8:53 am

    It seems that picture became famous overnight

  • Leave a Reply