Appendix C. Evaluation of the potential efficacy of four management strategies aiming to control disease spread.
Methods
We evaluated the potential efficacy of four different management strategies aiming to control disease spread in the colonizing population, as a function of disease transmissibility, representing moderate (β = 0.5) and high (β = 1) infectiousness (regimes 2 and 3): halting further releases of deer once the disease was recognized; vaccinating all released deer; vaccinating all released deer and all wild-born young (at age two, before they establish home ranges) in a spatially-focused region near the area of introduction (specifically, within a 3.4 km by 3.4 km square centered on the release site, which is approximately two average home range diameters); and vaccinating all released deer and wild-born young in the entire population. Within the last strategy we simulated two scenarios: complete coverage—all young are vaccinated, and limited coverage—the probability of finding a fawn in the wild and vaccinating it was set to 0.7. Note that the status of fawns already infected by maternal transmission was not altered by vaccination.
Vaccination has been recognized as a possible means of controlling BTB in wildlife (Griffin and Mackintosh 2000), although the long-term effectiveness of the bacillus Calmette-Guerin (BCG) vaccine is unknown for many host species (de Lisle et al. 2002). Vaccination was assumed in the model to be 100% effective and provide life-long protection. This is beyond the effectiveness of the currently available vaccine (Griffin and Mackintosh 2000), but was simulated as a theoretical exercise to assess whether we may expect future reasonably effective vaccines to have the potential to control a BTB outbreak. Different assumptions about vaccine efficacy do not substantively affect the conceptual arguments illustrated by our simulations, but would reduce the clarity by introducing additional structure and parameters. All strategies began from the third year after initial release onwards, assuming a two-year delay to identify and respond to the disease in the wild population.
Results
In terms of our first management scenario we found that stopping further releases in the reintroduction program (after 20 individuals have been released over the first two years) had minimal effect, increasing the probability of disease extinction within 20 years by only 5–10% for any scenario (Fig. C1). When transmissibility was moderate (β = 0.5, Fig. C1) or high (β = 1), the disease expanded and persisted in the population in more than 70% (probability of disease extinction, pext = 0.3) and 90% (pext = 0.09) of the simulations, respectively, with an average of more than 50% of all females infected in the latter case (27 + 32 out of 46 + 15 individuals [mean + SE]). When animal releases were continued, varying degrees of control were achieved with vaccination interventions. If all individuals released after the first two years were vaccinated, the probability of disease extinction increased marginally—an effect roughly equivalent to the stop-release policy (for β = 0.5, pext = 0.28—Fig. C1; for β = 1, pext = 0.06). More substantial impacts on disease spread in the colonizing population were possible when wild-born young were vaccinated in addition to released individuals. Vaccinating all wild-born young in a buffer zone around the release site proved superior on average to vaccinating 70% of wild-born young in the entire population (for β = 0.5, pext = 0.67 vs. pext = 0.56, respectively—Fig. C1; for β = 1, pext = 0.37 vs. pext = 0.25, respectively), though variation in the number infected at year 20 was greater for the former policy due to some epidemics that escaped the buffer zone (see Fig. C2 for β = 0.5). A complete-coverage intervention (i.e., vaccinating 100% of all wild-born young) led to a notable increase in the effect of the intervention but not to eradication because of maternal transmission (for β = 0.5, pext = 0.74, Fig. C1 and Fig. C2; for β = 1, pext = 0.56). Reducing maternal transmission to zero, in addition to the complete-coverage intervention, stopped all transmission and caused disease extinction by year 13 (for β = 0.5, Fig. C1 and Fig. C2).
LITERATURE CITED
de Lisle, G. W., R. G. Bengis, S. M. Schmitt, and D. J. O'Brien. 2002. Tuberculosis in free-ranging wildlife: detection, diagnosis and management. Revue Scientifique Et Technique De L Office International Des Epizooties 21:317–334.
Griffin, J. F. T., and C. G. Mackintosh. 2000. Tuberculosis in deer: Perceptions, problems and progress. Veterinary Journal 160:202–219.
