Let’s imagine two very simple infection scenarios. They are analogous to typical interactions that occur in bars, clubs and other social events.
In the first scenario one infected person is talking to another uninfected person.
The probability of the person being infected will look like;
We’ve set the infection probability as being 50% per 10 minutes of exposure. The x axis is time in minutes. The reason for its curve is for each additional 10 minute time unit an additional 50% is added.
Note: The time lengths are arbitrary, and chosen for illustrative purposes. The graph tends towards 1, with 100% certainty corresponding to the person being definitely infected.
So after 50 minutes the probability of the person being infected is 0.96875. We can think of this statistically as being 0.96875 people infected.
Next, let’s imagine the same infected person but talking to 5 different uninfected people for 10 minutes, each time separately. The first portion of the graph 0.5 would be repeated 5 times, and this time we would get 2.5 people infected. More mingling dramatically increases the level of disease exposure.
There’s a very clear and obvious relation to the amount of mixing that occurs, and the number of people that become infected.
Similiar ratios should exist for all social groups. We can imagine that social events with lots of mixing would result in several times as many infected people, with the details depending on other more readily measured variables, such as individual age, and event duration length.
DISCUSSION
As measured by attack rate, many of the most infectious events show a high degree of mingling. I’ll add some examples here soon.
This is very important because the role of avoiding mingling in infectious events is not communicated in public advice, and it needs to be.