Our manuscript explaining how field-data about pedestrian crowds can be coupled to viral transmission models to assess the risks of new Covid-19 infections has finally been published in Safety Science (see the Publications page for a freely accessible preprint version).
One of the main goals of the paper was to check whether close encounters between unrelated people in fairly busy streets were not a blind spot of epidemiological studies, granted that many Covid-19 patients (more than one third in France in March 2021) had no clue about where they had been infected. This question has bearing on public policies regarding face covering outdoors.
Using an original combination of empirical data collected in diverse daily-life situations and a gamut of simple viral transmission models, we came to the conclusion that the collective risk raised by fairly busy streets (assessed by the estimated number of new cases caused by a contagious person wandering for one hour on the premises) is far lower than that incurred at a street café, when no one is wearing a mask in any situation. Note that this collective risk (in terms of the number of new infections) differs from the subjective risk incurred by a person in the situation under study; the former is more relevant for policy-making while the latter is what matters most to a susceptible individual.
In our ranking of situations by the risks that they present, second to the street cafés come crowded outdoor markets, followed by train and metro stations; fairly busy streets come last. For situations with moving crowds, the parameter that matters most in this ranking is the density of the crowd, as one might have expected.
