The ecology of wildlife disease control : simple models revisited

1996 
Simple disease/host models are used to identify the ecological criteria governing the likely relative effectiveness of culling, sterilization and vaccination as methods of wildlife disease control. The models include non-linear host death or birth rate/density relationships, and non-linear disease contact rate/density relationships. The following conclusions are based on these simple homogeneous-mixing models, and may be modified by certain spatial effects such as territoriality and density-dependent settlement of dispersing hosts. The threshold density of susceptible hosts for elimination of a disease, relative to the disease-free carrying capacity, is largely determined by characteristics of the disease. For any given threshold, relative to the carrying capacity, the degree of control required to maintain the host density at this threshold is solely determined by ecological characteristics of the host. The sustained rate of control required to eliminate disease can be expressed, exactly or to a good approximation, in terms of the basic reproductive rate of the disease, R 0 the host density-independent birth and death rates, and the parameters describing non-linearities in the host density dependence and disease contact rate/density functions. In terms of the sustained rates of control (e.g. percentage per year) necessary to eliminate disease, sterilization is as effective as culling (i.e. it requires the same level of control effort). Culling is likely to be more effective than vaccination (i.e. requires a lower rate of control) when R 0 is high (> 3) or host density dependence acts on mortality rather than recruitment. When R 0 is low (< 3) and density dependence in the host acts on recruitment rather than overall mortality, then either culling or vaccination can be more effective depending on the ratio of the maximum host death rate to its intrinsic rate of increase, and the non-linearity in the host density dependence and/or the disease contact rate/density relationship; vaccination is favoured by a low death rate relative to r, the host intrinsic rate of increase, and by non-linear density relationships. For a combined strategy of culling and vaccination, the total rate of control required to eliminate disease exceeds that for at least one of the strategies in isolation; to this extent the combined control is less effective. Using the above criterion for comparison, for all six published case studies considered as examples, culling is theoretically the most effective control strategy given the authors' assumptions in the original models. More recent data for possum density dependence suggests that vaccination could more closely match culling in effectiveness. The extent to which vaccination appears more effective in practice, in cases like fox rabies in Europe, probably reflects the spatial behaviour of the host not considered in these simple models. Comparing strategies on the basis of the relative rates at which they reduce endemic disease or the density of susceptible hosts, gives different results. In particular, sterilization is less effective than vaccination, and very much less effective than culling. Non-linearities in the density relationships greatly increase the effectiveness of vaccination relative to cutting ana sterilization, but the distinction between host density-dependent mortality or recruitment is no longer important. The biological reasons for these conclusions are discussed.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    20
    References
    106
    Citations
    NaN
    KQI
    []