Comparing intervention measures in a model of a disease outbreak on a university campus.

A number of theoretical models have been developed in recent years modelling epidemic spread in educational settings such as universities, often as part of efforts to inform re-opening strategies during the COVID-19 pandemic. However, these studies have had differing conclusions as to the most effective non-pharmaceutical interventions. They also largely assumed permanent acquired immunity, meaning we have less understanding of how disease dynamics will play out when immunity wanes. Here, we complement these studies by developing and analysing a general stochastic simulation model of disease spread on a university campus where we allow immunity to wane, exploring the effectiveness of different interventions. We find that the two most effective interventions to limit the severity of a disease outbreak are reducing extra-household mixing and surveillance testing backed-up by a moderate isolation period. We find that contact tracing only has a limited effect, while reducing class sizes only has much effect if extra-household mixing is already low. We identify a range of measures that can not only limit an outbreak but prevent it entirely, and also comment on the variation in measures of severity that emerge from our stochastic simulations. We hope that our model may help in designing effective strategies for universities in future disease outbreaks.