Back

"Now you see me": detecting asymptomatic infectious individuals in the population

Iseli, A.; Dobay, A.; Breitling, L. P.

2023-05-14 epidemiology
10.1101/2023.05.08.23289685 medRxiv
Show abstract

Quarantine is an effective countermeasure to stop or slow the spread of an emerging infectious disease when no other preventive measures are available to protect the population. However, when the disease results in a proportion of asymptomatic infections, the spread dynamics are affected, and quarantine efficiency is impaired. Here, we introduce an extended susceptible-infected-recovered (SIR) model to study the effects of asymptomatic individuals at the onset of an emerging infectious disease when no vaccination is yet available and/or when a vaccine is available but only a subset of the population can be vaccinated due to limited supply or the unwillingness of susceptible individuals to receive an injection. These aspects have been indirectly incorporated into the model using a time-dependent vaccination rate. With this model, we confirm that, in the case of a missing vaccine, quarantine is effective in stopping the spread of an infectious disease, but its efficiency can be substantially reduced in the presence of individuals developing asymptomatic infection. Moreover, we show that vaccination is effective only if available early during the epidemic and if the vaccination rate is sufficiently high. By applying this model to Zurich and all of Switzerland in case of the COVID-19 pandemic, we found that the following two strategies have similar outcomes: either placing infectious individuals into quarantine when no vaccine is available or dropping quarantine measures but administering a vaccine at a daily rate of 1%, starting no later than 105 days after the onset of the epidemic. Beyond this time period, a vaccination campaign will have no effect in stopping the spread of the disease if 25% of the susceptible population is asymptomatic. We also found that the option of deploying a vaccination campaign was more effective for all of Switzerland than for only the city of Zurich.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.