Keep Your Distance

The number of people infected with the novel coronavirus is increasing rapidly, as each infected person potentially spreads the virus to a number of other people. The fewer people infected, the better. How can we achieve this? Our simulation shows that social distancing is key.

– by Dilay Avci, Michael Kreil, Hakan Tanriverdi and Maximilian Zierer

Deutsche Version: Warum nur Abstand halten hilft

March 20th the German Federal State of Bavaria issued comprehensive stay-at-home-orders. By now all German Federal States have followed suit. Public life has been restricted to a great extent. These measures, which affect our everyday lives severely, aim to preventthe novel virus from spreading exponentially. “Social distancing” has become the catchphrase in the fight against the coronavirus pandemic. It means avoiding crowded spaces and gatherings of people and reducing physical and social contacts to a minimum. Social distancing is vital in order to slow down the pace at which the virus is spreading.

To understand how rapidly a virus is spreading, epidemiologists mainly focus on a number called R0 - the reproduction number - which allows them to predict how many people will get infected by one infected person. As for the novel corona virus, the public Robert-Koch-Institute assumes a basic reproductive number of 2.4 to 3.3, which means that each infected person will infect three more persons unless countermeasures are in place. To slow down the pace at which the pandemic spreads, every effort has to be made to lower the reproduction number. The fewer people infected at the same time, the better our health system can cope with the challenge. Even small changes will make a big difference as you can see when moving the slider in the below simulation up and down.

The smaller the reproduction number, the better

  • Total of people infected: 30 %
  • Maximum of sick people at the same time: 94%

Assuming a reproduction number of 3, without any counter measures such as social distancing in place, we will experience a very steep rise of the curve. The virus will spread more rapidly with more people falling ill at the same time, causing our hospitals and medical staff to be overwhelmed and no longer being able to provide adequate medical treatment. The smaller the reproduction number, the fewer people will fall ill at the same time. It’s this scenario to which the German Association for Epidemiology has been trying to draw our attention to recently: “The huge risk of an uncontrolled outbreak with so many people getting sick at the same time and needing intensive care treatment is that our health system and medical facilities would be overwhelmed within a very short time”.

In order to slow down the rate at which the virus spreads, we have to try to lower the reproduction number as far as possible, ideally to a value of 1 or smaller. How can we achieve this?

Experts recommend social contacts to be restricted to a minimum and keeping one’s distance. With this in mind, mass events were cancelled, schools and universities closed and stay-at-home-orders were issued. The simulation below shows what might happen without appropriate counter measures, based on an assumed population of 300 individuals.

People do not keep their distance – simulation without social distancing

Healthy: 0 %
Infected: 0 %
Recovered: 0 %

Each grey ball () represents one healthy person; each red ball () one infected person. The balls are moving around freely. When two balls bump into each other, the virus is transmitted, infecting another healthy person. A process that is gaining speed rapidly without practicing “social distancing”. Only after a certain period, an infected person will recover and become immune against the virus (blue ball: ), so that it can no longer infect other people. This simulation is inspired by the Washington Post.

Conclusion: Within a short period, the majority of the simulated population will be infected with the virus simultaneously. Projecting this scenario onto real life implies overburdening our health system, while social distancing decreases the rate at which infections rise considerably.

People keeping their distance – simulation with social distancing

Healthy: 0 %
Infected: 0 %
Recovered: 0 %

In this simulation only a fifth of the population is moving freely, the rest is observing stay-at-home-orders and practicing social distancing. While the numbers vary each time you restart the simulation, there is one consistent development to be seen: By strictly observing social distancing the virus expands at a significantly lower rate than without practicing social distancing – a picture in which our health system could cope much better with the situation in real life.

“What is most important is that we manage to avoid all people becoming sick at the same time”, concludes Uta Merle, a professor at the University of Heidelberg. “That is why all of us, every single person, should try to behave reasonably. The virus is in our world, it will remain and it will continue to circulate. The more we are able to extend the period of time in which people get infected, the better our hospitals will be able to cope with the challenge”.

About the project:

“Keep your distance“ is a project by BR Data.

Visualization model of a “crowd“: : For this visualization we consider a crowd of about 300 people shown as dots. Initially there would be one of the plenty dots infected moving freely across the room. Whenever two dots might cross their paths,an infection can take place causing another dot to change color from grey to red. Six seconds later the first person represented by the (red) dot would recover and therefore again change its color to blue. At this stage, the same person (blue dot) would be immune to the virus. The visualization is inspired by the Washington Post.

Visualization model as a graph: : In order to visualize the spreading of the virus on a societal level we use the so-called SEIR model (as in “Susceptible, Exposed, Infectious, Recovered”). This model is assuming that people who are unprotected and volatile towards infection are very likely to get infected and afterwards recover due to their natural immunity system. For this, we take the numbers which are also being used by the German association for Epidemiology within this statement (PDF). The incubation period as in the time it takes between the first infection and the actual outbreak of the corresponding disease is being set up by the model as 5.5 days. Based on this example the duration, within which an infection can take place before the respective person would go into quarantine, would be set up as 3 days.

Original version published March 20, 2020. English version published March 27, 2020.

Reporters: Dilay Avci, Michael Kreil, Hakan Tanriverdi and Maximilian Zierer
Editors: Uli Köppen, Miriam Stumpfe
Graphics: Michael Kreil
Development: Michael Kreil, Steffen Kühne
Translation: Ulrike Ecker, Claus Zingler