Assessing the host disease status of wildlife and the implications for disease control:Mycobacterium bovisinfection in feral ferrets

Abstract

SummaryEstimating the basic reproductive rate (R0) of disease and/or the related threshold population density (KT) for disease establishment is fundamental to determining the host status of wildlife for disease, and thus the effective management of pathogens in free‐ranging wildlife. The strength of inference of host status is largely dependent on the precision with which eitherR0orKTis estimated, yet only a small proportion of studies of disease in wildlife have estimated the precision of these parameters.We used a combination of observations, field experiments and modelling to estimate the basic reproductive rate ofMycobacterium bovisinfection in ferretsMustela furoin New Zealand. Estimates ofR0ranged from 0·18 at the lowest standardized annual density (0·5 ferrets km−2) recorded, to 1·2 at the highest standardized annual density (3·4 ferrets km−2) recorded. The estimates ofR0were moderately imprecise, with a coefficient of variation of 37%.The estimated threshold standardized annual density (K̂T) for disease establishment was 2·9 ferrets km−2(95% confidence interval 1·7–10·6 ferrets km−2); however, only in limited geographical areas of New Zealand do ferret population densities exceed this. In these areas, ferrets would be considered maintenance hosts (R0 > 1) for the disease, and active management (e.g. density reduction or vaccination) of ferrets would be required to eradicateM. bovisfrom ferret populations, in addition to the elimination of sources of interspecific transmission, notably brushtail possumsTrichosurus vulpecula.Synthesis and applications. The results have considerable implications for the management ofM. bovisin ferret populations in New Zealand and elsewhere in the world. In areas of high ferret density, ferrets are probably a maintenance host of bovine tuberculosis, so active management is required to reduce disease incidence. More widely, the results demonstrate clearly how modelling and field observations may be combined to better inform managers of a wildlife disease.