An SIQ delay differential equations model for disease control via isolation (Q2311901)

This paper considers and adaptation of a SIS (susceptible-infectious-susceptible) model under the assumption that a fraction of the hosts which remain infectious for a certain amount of time without recovering are detected and isolated, also for a fixed amount of time. The resulting model is called a SIQ (susceptible-infectious-quarantined) model. In this regard, throughout the paper ``quarantine'' is to be understood as ``isolation of infectious individuals'' rather than ``isolation of possibly infected individuals which are not yet symptomatic'', which is another commonly accepted meaning. The main focus is on how isolation, even imperfectly implemented, can successfully tame a (generic) disease outbreak. In this sense, precise values for the required response capabilities required are determined. It is proved that in order to have a chance to tame the outbreak, one must be able to identify and quarantine a sufficiently large fraction of infectious individuals. If this condition is met, then the reaction should also be quick enough, in the sense that the identification time (which appears as a delay in the SIQ model) should be less than a certain critical value. If these critical conditions are not met, then it is shown that the disease becomes endemic, and the authors provide an estimation of the fraction of infectious individuals at the endemic equilibrium. It is also investigated how the isolation time affects the outcome of the epidemic, being shown that as the isolation time increases, the endemic equilibrium loses its stability through a Hopf bifurcation and it undergoes a cascade of Hopf bifurcation if this isolation time increases further. That is, larger isolation times can lead to oscillations in the fraction of infectious hosts, and this fraction can actually rise above the corresponding fraction obtained for shorter isolation times. Further, a possible model extension which includes latency time is also discussed. The mathematical approach is based on the theory of (infinite-dimensional, due to the presence of the delay in the SIQ model) dynamical systems. In this regard, local, linear analyses are coupled with an analysis of the respective stable and unstable manifolds, more geometric in its nature, for further clarifications of the dynamical picture.