We started thinking how a common airborne disease spread to the population, with pathogens that can be transmitted via infectious droplets or through touching contaminated surfaces. If we consider a regular day life (for many of us at least) we start the day with commuting by train or subway, then we work with colleagues for some hours in an office or factory, and eventually we gather with friends in a bar. We expose ourselves to hundreds of other people. People we don’t necessarily know and we have no idea how to get back to if needed.
If one of them was infected, we could have been infected too, and so our further connections. Few infected people that are roaming freely can potentially reach all the citizens of a city through second and third degree connections.
Central government cannot lock down the whole city just because of those few. Actually they can but it will come at considerable cost for economy and people comfort.
However, as each of those interactions carry different intrinsic risks, a probabilistic model that takes into account transmission factors (mainly proximity and length of exposure) is needed to take actions only over people that have high risk of being infected.
A probabilistic model will also enable the public health service to better prioritize who should be tested first and who should adopt less restrictive measures. We believe this is a key factor for success as resources available to do a test are scarce, and a massive test campaign didn’t prove to produce a real impact yet.
Moreover, the boundaries between people with high-risk profiles (that requires urgent action) and low-risk profiles can be adjusted according to the current capability of making tests, with a better and clearer return on the investment.