Assessing the sensitivity of European surveillance for detecting BSE in cattle according to international standards

Abstract

European surveillance for Bovine Spongiform Encephalopathy (BSE), initiated in 2001, has shown a steady exponential decline in the number of infected cattle, demonstrating that control measures have been effective. In 2016 23 European countries out of 28 demonstrated negligible risk status for the disease. The international standard setting body, the World Organization for Animal Health (OIE), prescribes that for countries where there is a non-negligible BSE risk, surveillance should allow the detection of one case per 100,000 in the adult cattle population with 95% confidence (Type A surveillance). The Cattle TSE Monitoring Model (C-TSEMM) was developed to estimate the sensitivity of surveillance systems to detect BSE in cattle. The model includes a cohort-based back calculation model to estimate the number and age of infected animals, the subset of those that are detectable by the diagnostic test, and the stream by which infected animals exit the standing population, that is, healthy slaughter, emergency slaughter, fallen stock and clinical suspects. Data collected by the European Commission (EC) and European Food Safety Authority (EFSA), containing the details of over 91 million cattle tested within the European Member States (EU25) surveillance scheme from 2002 to 2011, was used to populate the model.When considering the EU25 surveillance scheme as a single epidemiological unit, the model estimated that the surveillance scheme in place exceeded the OIE required threshold for Type A surveillance, and that a revised monitoring system excluding all healthy slaughter animals would also exceed this threshold.Results indicated a variation in individual country sensitivity of surveillance with seven countries (Germany, Spain, France, Ireland, Italy, Poland and the UK) exceeding the required threshold individually. Key assumptions included the fitting of an exponential distribution for all countries to describe the decline in prevalence over time and, for those members with no detected cases of BSE, an estimated area prevalence was used as a proxy. These results were part of the evidence base that lifted the testing requirement for healthy slaughter animals in European countries in 2013. The model has been maintained and updated by EFSA each year with the addition of new member states, and updated surveillance data.