By Julian Lee (V)
With statewide stay-at-home orders currently issued in at least 42 states, we should take into consideration the factors that could compromise the effectiveness of this quarantine. Inspired by the simulations created by the Washington Post and the YouTube channel 3Blue1Brown, I wanted to further investigate how human behavior––specifically, visiting friends––can impact the spread of COVID-19 under a quarantine environment.
I created a simulated environment of 100 households, where only interactions between family members and one-on-one visits with friends can cause infections. The user can change various parameters, such as the average days between visiting friends, and observe how changing these variables affect the spread of the virus. The simulation can be found here.
The simulation suggests that in the world of social distancing, the frequency of visiting friends has a greater impact on the spread of the virus than the size of a person’s social network (simulation results shown at the end of the article). Someone who visits the same friend every other day spreads the virus faster than someone who visits one friend every four days in a ten-person social network. Based on the simulation, reducing the number of friend visits during quarantine by a factor of two could have an effect comparable to halving the infection rate of the virus.
Below are my findings from the simulation (100 simulations were run for each setting):
- Doubling the average time between friend visits from 2 days to 4 days caused the virus’ average spread to decrease from 51% to 29% of the population.
- Halving the infection rate for both friend visits (from 20% to 10%) and family members (from 40% to 20%) resulted in a similar reduction in virus’ spread from 51% to 27% of the population.
- Decreasing the number of friends (i.e. the social network) from 10 to 1 caused the virus’ average spread to decrease from 54% to 40% of the population.
While someone might think it is completely benign to visit just “one” friend every other day, such behavior by an entire population can still result in an exponential growth of the virus. For example, if someone infects the one friend they are visiting during quarantine, that friend would then infect their entire family, and these family members would infect their own friends.
This simulation helps to quantitatively demonstrate an obvious yet powerful fact about social distancing: to ensure that our quarantine proves effective, it is essential that we work towards minimizing the frequency of visiting others.