Rabies in urban foxes (Vulpes vulpes) in Britain: the use of a spatial stochastic simulation model to examine the pattern of spread and evaluate the efficacy of different control régimes.

The threat of rabies being reintroduced into Britain is probably greater now than at any time over the last 60 years. This threat is reviewed with particular regard to the problems that would be posed should rabies be introduced to the high-density fox populations found in many cities in southern England. Computer models can provide a valuable means of understanding the pattern of rabies spread in fox populations and the likely problems of control, so the construction of previous rabies models was reviewed. None were found to be suitable for analysing the particular problems posed by high-density, spatially heterogeneous, urban fox populations. Therefore, a new spatial stochastic simulation model was produced, based on demographic and other data collected during a long-term study on the urban fox population in Bristol, and fox density data collected from a number of cities in southern England. The simulation model was used to analyse the effects of spatial heterogeneity in the fox population on the pattern of rabies spread. Simulations were then used to evaluate the effects of: (i) varying levels of fox control; (ii) changing the size of the control zone; (iii) the onset of the rabies epizooty at different times of the year: and (iv) delay before the commencement of control on the chances of containing the disease. These simulations were run for four cities (Bournemouth and Poole, Bristol, Leicester and the West Midlands conurbation) with different mean fox population densities. It was found that the variance in the monthly velocity of the rabies front was greater for heterogeneous fox populations. In cities with high fox densities, low or moderate levels of control were unsuccessful in containing the disease, but these urban areas had the highest rates of success with the highest levels of control. A three-month delay in the commencement of a rabies control campaign on average reduced the chance of successfully controlling the disease by 10-20%, although this was higher in lower-density fox populations. Rabies outbreaks in the dispersal period were on average 10% less likely to be contained. Increasing the size of the control zone increased the chances of successfully containing the disease, although this effect was density dependent, so the effect was less in low-density fox populations. These results are discussed in relation to the current rabies contingency plans for British urban areas.