Contact tracing is a mechanism for the control of outbreaks. It aims to break transmission chains by tracing and isolating people exposed to an infectious individual. In this research work, we investigate how the performance of contact tracing is affected by periodic contact patterns, which are present in events that favor the propagation of diseases and events that aim to control them.
To answer the research question, we constructed a mathematical model that incorporates periodicity for each type of contact tracing: backward - which identifies the infectors, forward - which identifies the infectees, and full - which identifies both. The models incorporate the non-recursive and recursive variants of contact tracing. In the non-recursive variant, the process stops after the contacts of one case are identified, whereas the recursive variant follows a chain of infections.
We reformulate the question as a set of optimization problems and solve it for the nonrecursive backward contact tracing scenario. Our results indicate that periodic contact rates positively affect the effectiveness of backward contact tracing. The existence of periodicity increases the probability of extinction of the epidemic and minimizes the effective reproduction number. The models and optimization problems we propose here can be employed in future research to explore more complex scenarios and other types of contact tracing.
NOTE
This is a fragment of my master thesis. Suggested citation:
Agudelo, S. (2021) Simulation of periodic contact patterns on the effect of contact tracing (Master's Thesis, Technical University of Munich, Munich, Germany).