Towards Risk-Based Control of Campylobacter: Developing the Evidence Base using Epidemiological and Bacteriological Approaches

The work proposed in this project is a comprehensive package of multi-disciplinary research aimed at supplying data on which to base practical policy decisions on control of Campylobacter in broiler production.
<P>

In order to identify or quantify a problem it is necessary to use suitable methodology, which does not have to be perfect, but its limitations should be known so that suitable correction factors for underestimation of colonisation and contamination can be applied. Accurate data is essential when considering issues of infectious dose for humans and birds via various routes and for assessing risk and the likely degree of success of various control options through risk modelling.
<P>
Objective One aims to validate the sampling methodology used for the EU survey in terms of potential for false positives caused by external contamination of caeca, the sensitivity of the pooled sample for detecting a low prevalence of positive samples or a low number of organisms and the influence of the various plating technique options proposed. Hence this work will deliver data on sensitivity and specificity of the survey methodology and enhance the comparability of data from the various member states. <P>
In addition to this there is a need to determine the most suitable methods for identification and quantification of Campylobacter from faecal and environmental samples in order to carry out meaningful epidemiological studies. It is likely that much environmental contamination is currently being missed and quantification in the environment will facilitate a realistic appraisal of risk. The work carried out in this project, together with up to date review data from published and unpublished studies will fill this large data gap and provide suitable tools for further epidemiological studies, both in the current project and in future work. <P>
The strong recommendation from the EU to set up a national monitoring programme has been endorsed and accepted by the UK. The basic survey will provide a minimal data set sufficient to determine with reasonable accuracy the rate of infected slaughter batches, which is the information desired by the EU. The extended survey proposed in Objective Two offers the opportunity to gather much more useful data in relation to seasonal effects and the effect of sequential depopulation of flocks (thinning). The interval between slaughter batches is also crucial in terms of a build up of infection after introduction at the first thin and may provide data which can be used to limit this infection in future. In addition it will be possible to determine within-batch prevalence and the numbers of organisms in individual birds using methodology validated in Objective One. These data will help to fill gaps in risk assessment models. The isolates obtained from the survey will also be a valuable statistically derived sample resource for molecular epidemiological studies, particularly comparison with human strains by MLST or other suitable methods. <P>
Objective Three focuses on more detailed epidemiological investigations of the thinning process. This will take advantage of study situations generated by both the EU Salmonella survey and the Campylobacter survey so that the whole chain of events from farm through thinning and the slaughter process can be followed in detail by quantitative bacteriology and molecular tracking of strains through to the final carcase. In addition this work will link with an develop on an existing University of Bristol FSA project, in which VLA is already involved on a consultancy basis. This will provide substantial added value by virtue of provision of strains from Bristol studies of the thinning process for molecular tracking to elucidate the most important sources and this study will as a pilot for later VLA studies from thin to carcase.
It is also proposed within the current project to conduct exposure experiments in which samples from the thinning process will be directly exposed to birds to ascertain the relative risk of infection from environmental strains introduced from the outside of the house during and after thinning compared with isolates from previous flocks processed by the thinning team found to be contaminating equipment and personnel. This large objective should therefore result in several articles and publications which will be of great benefit to the poultry industry in its efforts to comply with the FSA strategy and retailer requirements to further reduce campylobacter levels in flocks and at slaughter.
<P>
All of the above work will provide good quality data with which to expand the VLA risk assessment model (Objective 4). This model, which could also be linked with other models as part of a consolidation initiative under the Med-Vet-Net programme, will act as a valuable tool for determining the relative risk of Campylobacter entering broiler farms at various stages of production and by different routes. Various potential control measures can then be modelled - e.g.. reducing the proportion of positive farms by various degrees by biosecurity, reducing the pre-slaughter load of Campylobacter by chemical intervention such as chlorate administration or biological treatments such as phage, competitor organisms or bacteriocins. This can then be compared with options for modifying the thinning process, slaughter methods or post-slaughter interventions. This information derived from studies relevant to the UK broiler industry can then be used to direct policy decisions on future control options and intervention studies.