Scaling analysis of COVID-19 spreading based on Belgian hospitalization data

Bart Smeets; Rodrigo Watte; Herman Ramon

In Belgium, policies to contain the spreading of COVID-19 have proceeded in multiple phases. Initially, in phase I, strong quarantine measures were imposed on detected and suspected individuals who traveled from at-risk regions. In case of confirmed infections, their recent history of contacts was traced back and these individuals were quarantined as well as tested for COVID-19. Phase II included fast testing and quarantine of all individuals that exhibit symptoms. In phase III, drastic societal containment strategies are enacted. Regardless of symptoms, individuals are to minimize any social interactions. Due to restricted testing capacity, tests are only performed on individuals that exhibit severe symptoms. An important consequence of this strategy is that the number of confirmed cases can be heavily biased by shifting testing capacity and testing priorities. As an alternative, we propose that the accumulated number of hospitalized individuals is a good indicator for the number of actual COVID-19, albeit with a shift in time. This temporal shift is roughly equal to the combination of the mean incubation time (≈ 5 days) and the average time from the onset of symptoms to hospitalization (≈ 2 days) [6] .