Simulating the effect of quarantine on the spread of the 1918-19 flu in central Canada.

Quarantine is often proposed and sometimes used to control the spread of infectious diseases through a human population. Yet there is usually little or no information on the effectiveness of attempting to quarantine humans that is not of an anecdotal or conjectural nature. This paper describes how a compartmental model for the geographic spread of infectious diseases can be used to address the potential effectiveness of human quarantine. The model is applied to data from the historical record in central Canada around the time of the 1918-19 influenza epidemic. Information on the daily mobility patterns of individuals engaged in the fur trade throughout the region prior to, during, and immediately after the epidemic are used to determine whether rates of travel were affected by informal quarantine policies imposed by community leaders. The model is then used to assess the impact of observed differences in travel on the spread of the epidemic. Results show that when mobility rates are very low, as in this region, quarantine practices must be highly effective before they alter disease patterns significantly. Simulation results suggest, though, that effectiveness varies depending on when the limitation on travel between communities is implemented and how long it lasts, and that a policy of introducing quarantine at the earliest possible time may not always lead to the greatest reduction in cases of a disease.