1 National Food Institute, Technical University of Denmark2 Division of Epidemiology and Microbial Genomics, National Food Institute, Technical University of Denmark
Microbiological criteria (MCs) offer a practical tool for food safety control and they are currently under discussion internationally. To meet the present scientific standards, there is an increasing demand for so- called “risk based” microbiological criteria that are based on risk assessment. In this project we studied the potentials for setting risk based microbiological criteria on Campylobacter in chicken meat by studying the potential impact that specific microbiological criteria would have in different Nordic countries. This is done on the basis of different data sets that have been collected in these countries in the past, and for the 2008 EU baseline survey data. The approach used is similar to that applied for the EFSA opinion of Campylobacter control (EFSA 2011, Nauta, Sanaa and Havelaar 2012), but in this study additional data sets are analysed. Next, as an alternative approach for setting risk based microbiological criteria, the “case-by-case” risk assessment methodology is used (Christensen et al 2013) and its impact is analysed on the basis of the same data sets. In both approaches the same risk assessment model for Campylobacter in broiler meat is used. The difference between the approaches is that for the microbiological criterion the rule for compliance is based on the traditional definition, defined by the number of samples containing more than a critical concentration of bacteria (Van Schothorst et al 2009), whereas for the “case-by-case” it is based on a critical risk estimate. The study confirms that the risk of campylobacteriosis from broiler meat produced in the Nordic countries (and especially Norway and Finland) is low compared to most other European countries. When using different data sets from the same country, the results differ between them, but the ranking of countries is unaltered. It is for example found that microbiological criterion based on n= 5 samples, with a threshold concentration of m=1000 cfu/g that may be exceeded in c=1 out of 5 samples, gives between 0 and 10% non-complying batches of poultry meat in the Nordic countries. The risk reduction obtained by implementation of this MC varies greatly, and is, in general, larger when more non-complying batches are no longer accepted on the fresh meat market. Detailed results per country can be obtained from the report. The analyses in the report focus on the evaluation of one specific microbiological criterion, which was previously selected as an example scenario by EFSA (2011), and one comparable “case-by-case” criterion. Results suggest that the efficiencies, in terms of potential risk reduction versus the percentage of non-complying batches, are similar for both methods. However, when studying the uncertainties, the uncertainty attending the “case by case” approach seems to be a little smaller. This preliminary result suggests that the “case by case” approach may be a more reliable method. One way to study this further will be to proceed with Bayesian data analysis as presented in this report. We have shown that risk based microbiological criteria can be established, given the availability of a risk assessment model that links the bacterial concentration measured at the point where the criterion is set to the public health risk. The methods described in this report can offer a risk management tool where the choice for the optimum criterion can be based on combination of the potential risk reduction and the percentage of noncomplying batches that, in some way, require sanctioning. It is, however, unsure to what extent the anticipated effects on public health risk will be achieved. Not only is every microbiological risk assessment attended by considerable uncertainty, also the effects of setting criteria in terms of decreased concentrations of bacterial pathogens on the meat as an effect of (potential) sanctioning, are difficult to predict. A software tool will be developed to facilitate the use of the methods described in this report by interested parties, such as industrial and governmental food safety managers in the international community.
Main Research Area:
National Food Institute, Technical University of Denmark, 2013