Statistical Evaluation of the Achievements by Member States of the EU Salmonella Reduction Targets in Animal Populations

EU Member States are required to collect, evaluate and report data on zoonoses, zoonotic agents, antimicrobial resistance and food-borne outbreaks to the European Commission each year. The European Food Safety Authority is responsible for examining, analyzing and summarizing these data, and for publishing the results in the European Union Summary Report. The identification of trends in the occurrence of the zoonotic agents and the sources of human infections, in order to study the likelihood of Member States to achieve the European Union reduction targets, is one of the key analyses in the Summary Report. In this report, particular interest lies on the Salmonella European Union reduction targets in animal populations, particularly in flocks of breeding and laying hens of Gallus gallus. The main objective of this investigation was to explore and assess appropriate statistical methodologies enabling to evaluate the achievement by Member States of the Salmonella European Union reduction targets in animal populations. Salmonella flock prevalence data in breeding and laying hens of Gallus gallus at two levels – aggregated country-level monitoring data for all Member States, as well as nonaggregated, detailed, sample-level data for a number of Member States – were used for the investigation. For the aggregated-level data the extremely short time sequence available of minimum four annual time points implied that only simplistic models could be considered. It was concluded that reliable trend analyses could not be established based on such very limited amount of information. In contrast, for the sample-level data, a number of modelling approaches proved meaningful and stable enough to provide insight into the progress made by Member States towards the achievement of the Salmonella reduction targets in breeding and laying hens. For more reliable and informative trends analyses based on aggregated-level data it is recommended that Member States would provide quarterly periods or monthly prevalence data, rather than yearly values.