Human agency and infection rates: implications for social distancing during epidemics

Christopher Bronk Ramsey

For air-borne virus infections, the strong relationship between the probability of picking up an infection and passing it on is critical to evaluation of distancing strategies. In order properly to take account of this it is necessary to consider the very wide range of levels of interaction that individuals have through choices they make, and the jobs that they do. A very simple model of infection propagation demonstrates that it is much more efficient, when considering interventions to reduce the growth rate of an epidemic to below 1, to target interventions at those activities that result in the very highest interaction levels, and specifically for people who are engaged in those on a regular basis. There is a non-linearity in the system which implies that limiting upper levels of interaction will work more effectively than reducing interactions across the population as a whole. If we can equate lost interaction time with societal cost there is something like a 50% reduction in the cost in taking this approach.