Evaluation of Australian surveillance for freedom from bovine tuberculosis.

Abstract

Australia declared freedom from bovine tuberculosis (TB) in accordance with international guidelines in 1997 and has since maintained ongoing surveillance for this disease, primarily through abattoir surveillance of cattle carcases (meat inspection) for TB-like granulomas. The objectives of this study were to estimate the sensitivity of Australia's surveillance system and quantify the probability that Australia is free from bovine TB at the specified design prevalence. The analysis included approximately 80 million records of individual cattle slaughtered and meat inspected at Australian abattoirs between 2005 and 2015 calendar years. Animals were identified and aggregated by property of birth within year. Herd- and population-level sensitivities and probability of freedom were estimated on an annual basis using a simulation model to account for uncertainty about the unit sensitivity of the meat inspection process. The estimated median population sensitivity of Australia's TB surveillance system varied from a high of 80%, in years when traditional meat inspection was used, to as low as 50% after the introduction of visual-only meat inspection, for a design prevalence of 0.01% (19 herds) of Australian breeding cattle herds and 0.5% of animals within infected herds. The level of confidence in Australia's freedom from bovine TB was >95% after the first year of the analysis and >99.5% from 2007 through to the end of the analysis period in 2015. Reducing the animal-level or herd-level design prevalence, or the estimated sensitivity of meat inspection, reduced system sensitivity but confidence of disease freedom remained high. The results demonstrated a very high level of confidence that Australia's cattle herd is truly free from bovine TB. Meat inspection for granulomas in the head and thorax of slaughtered cattle underpins this result by providing surveillance data on an extremely large number of animals each year.