The dynamics of bovine tuberculosis in possum populations, and its eradication or control by culling or vaccination

1. A simple model is used to investigate the feasibility of eradicating bovine tuberculosis (Mycobacterium bovis Karlson and Lessel) from possum (Trichosurus vulpecula Kerr) populations by culling, or by distributing a (hypothetical) non-transmissible vaccine; and of reducing the prevalence of tuberculosis to a defined level within a fixed time period. 2. The model suggests that tuberculosis could be eradicated from a possum population by reducing its density to 43% of the (no-disease) carrying capacity, which would require culling possums at a rate of 9.2% per year; or by maintaining 40% of the population in a vaccine-immune state, which would require vaccinating possums at a rate of 15% year(-1). 3. The model suggests that the prevalence of tuberculosis could be reduced to 10% of its pre-control level within 5 years by culling possums at a rate of 12% year(-1), or by vaccinating possums at a rate of 13% year(-1). 4. Taken together, these results suggest that culling possums to maintain a low population density may be the most efficient method of preventing tuberculosis from establishing in an uninfected population, but vaccination could be as or more efficient as a means of removing tuberculosis from an infected population. Both of these statements are sensitive to changes in parameter values. 5. The culling rate needed to achieve eradication of tuberculosis is sensitive to the maximum population growth rate, r, and to the logistic shape parameter, theta. It is insensitive to whether density-dependent population regulation acts through the birth or death rates. 6. The disease dynamics are sensitive to the shape of the contact rate function, which is determined by the relative importance of mating, den sharing (possibly sequentially), aerosols and marking, and other mechanisms in the transmission of tuberculosis. 7. All results are sensitive to the increased population mortality due to disease, alpha, which is difficult to estimate. 8. The results have been calculated under the assumption that immigration of possums from other areas does not occur. If it did it could change the results significantly.