5 Case study: a typical floor in an office building
A supervisor is responsible for an office with 45 employees. She considers a few scenarios to mitigate the possible transmission of COVID-19 in the office once employees return to work. While the prevalence of COVID-19 infections in the population is rare, there is always a chance that an infected, but asymptomatic employee comes to the office. Her objective in this eventuality is to minimize transmission in the workplace and prevent an outbreak.
In order to have a point of comparison, she first specifies a baseline scenario in which interactions between employees average to 6 daily, with 95% of interactions including at most 5 individuals, and no special measures taken with respect to distancing or personal protection, other than the usual distance of 1m at which employees usually interact in the workplace. Since this is a work environment, direct contact between individuals is rare. The supervisor also specifies two possible options for mitigation measures (see Table below).
| Baseline | Option 1 | Option 2 | |
|---|---|---|---|
| Daily interactions per person | 6 | 4 | 4 |
| Upper group size for 95% of interactions | 5 | 4 | 3 |
| Social distancing | 1m | 2m | 2m |
| Personal protection | (none) | (none) | cloth masks |
The supervisor compares these three situations using the app. From the evidence she has on mitigation measures (see Figure), she knows that imposing social distancing of 2m should reduce the chance of transmission by a factor of about 2. One of her options also includes cloth masks. They are similar to surgical masks, which by themselves reduce transmission by a factor of at least 2. She makes the conservative assumption that together, 2m distancing and cloth masks should reduce transmission by a factor of 4. These three scenarios correspond to the following sets of input parameters:
Looking at the summary tables, the baseline case shows an Reff value of 5.3. This value is much higher than 1, and as expected there is a high chance of an runaway outbreak (87%), with a large number (100+), expected to be infected even in the first week. The mitigation measures in Option 1 reduce Reff significantly, from 5.3 to 1.5. However there is still a chance of an uncontained outbreak (33%), but the expected number of individuals infected in the first week is dramatically reduced, from 100+ to about 5, which might give enough time to intercept the outbreak through self-assessment of symptoms before it is too late.
In Option 2, a further reduction in interaction sizes, and the addition of cloth masks, brings Reff below 1. This reduces the chance of an uncontained outbreak to 0%, with a expected number infected in the first two weeks below 1.5. Even if all infected individuals remain asymptomatic, and undetected, the expected number infected remains small.
