Bayesian global regression model relating product characteristics of intermediate moisture food products to heat inactivation parameters for Salmonella Napoli and Eurotium herbariorum mould spores.

Thermal inactivation of pathogenic and spoilage organisms in low and intermediate moisture foods is of critical importance for guaranteeing microbiological safety and stability of these products. Producers tendentially reduce salt in low and intermediate moisture foods because of nutritional health considerations, but it is unclear how this affects microbial inactivation rates during pasteurization. In this study we predict the time to achieve a pre-defined 6-log reduction for Salmonella enterica subsp. enterica serovar Napoli (hereafter: S. Napoli) and Eurotium herbariorum mould spores (hereafter: E. herbariorum spores) and the relationship with product characteristics. We tested 31 design products for heat inactivation of S. Napoli and 29 design products for heat inactivation of E. herbariorum spores. We used a global Bayesian regression combining primary Weibull models with a secondary regression model to relate pasteurization temperature and product characteristics (water activity (aw), pH, and fractions of sodium chloride, sucrose and oil on product) to microbial counts. With this model, we predict the time to 6-log reduction. Thermal inactivation rates were much higher for vegetative S. Napoli than for E. herbariorum spores. Also, inactivation curves were non-linear for many experiments. There were significant associations between the Weibull model parameters and temperature, and pH and aw for S. Napoli and E. herbariorum spores, respectively. We parameterized an inactivation model for S. Napoli and E. herbariorum spores using design products with a broad range of characteristics and showed how the simplified approach of using D-values does not accurately describe the non-linearity of thermal inactivation for both types of organism. Results of our model can be used to produce accurate heat inactivation predictions as input for the pasteurization process in factories where intermediate moisture foods are manufactured.

Quantifying the sources of Salmonella on dressed carcasses of pigs based on serovar distribution.

Salmonella serotyping data, qualitatively described by van Hoek et al. (2012), were used to quantify potential sources of Salmonella in a Dutch pig slaughterhouse. Statistical tests to compare per-day Salmonella prevalence and serotyping data from multiple points in the chain were used to find transmission pathways. A statistical model based on serotyping data was developed to attribute Salmonella on dressed carcasses to the most likely source. Approximately two-third of dressed carcasses carrying Salmonella on the medial surface had been contaminated by house flora. For carcasses carrying Salmonella on the distal surface, transient Salmonella from incoming pigs was a more important source. The relevance of the different sources of Salmonella varied within and between sampling days. Results were compared to those of another modeling approach, in which Salmonella concentration data from the same samples were used (Smid et al., 2012). They mostly agreed. The approach chosen by an individual slaughterhouse depends on the data that are collected.

Campylobacteriosis in returning travellers and potential secondary transmission of exotic strains.

SUMMARY: Multilocus sequence types (STs) were determined for 232 and 737 Campylobacter jejuni/coli isolates from Dutch travellers and domestically acquired cases, respectively. Putative risk factors for travel-related campylobacteriosis, and for domestically acquired campylobacteriosis caused by exotic STs (putatively carried by returning travellers), were investigated. Travelling to Asia, Africa, Latin America and the Caribbean, and Southern Europe significantly increased the risk of acquiring campylobacteriosis compared to travelling within Western Europe. Besides eating chicken, using antacids, and having chronic enteropathies, we identified eating vegetable salad outside Europe, drinking bottled water in high-risk destinations, and handling/eating undercooked pork as possible risk factors for travel-related campylobacteriosis. Factors associated with domestically acquired campylobacteriosis caused by exotic STs involved predominantly person-to-person contacts around popular holiday periods. We concluded that putative determinants of travel-related campylobacteriosis differ from those of domestically acquired infections and that returning travellers may carry several exotic strains that might subsequently spread to domestic populations even through limited person-to-person transmission.

Attribution of human Salmonella infections to animal and food sources in Italy (2002-2010): adaptations of the Dutch and modified Hald source attribution models.

The Dutch and modified Hald source attribution models were adapted to Italian Salmonella data to attribute human infections caused by the top 30 serotypes between 2002 and 2010 to four putative sources (Gallus gallus, turkeys, pigs, ruminants), at the points of animal reservoir (farm), exposure (food), and both combined. Attribution estimates were thus compared between different models, time periods and sampling points. All models identified pigs as the main source of human salmonellosis in Italy, accounting for 43-60% of infections, followed by G. gallus (18-34%). Attributions to turkeys and ruminants were minor. An increasing temporal trend in attributions to pigs and a decreasing one in those to G. gallus was also observed. Although the outcomes of the two models applied at farm and food levels essentially agree, they can be refined once more information becomes available, providing valuable insights about potential targets along the production chain.

Increased risk for Campylobacter jejuni and C. coli infection of pet origin in dog owners and evidence for genetic association between strains causing infection in humans and their pets.

We compared Campylobacter jejuni/coli multilocus sequence types (STs) from pets (dogs/cats) and their owners and investigated risk factors for pet-associated human campylobacteriosis using a combined source-attribution and case-control analysis. In total, 132/687 pet stools were Campylobacter-positive, resulting in 499 strains isolated (320 C. upsaliensis/helveticus, 100 C. jejuni, 33 C. hyointestinalis/fetus, 10 C. lari, 4 C. coli, 32 unidentified). There were 737 human and 104 pet C. jejuni/coli strains assigned to 154 and 49 STs, respectively. Dog, particularly puppy, owners were at increased risk of infection with pet-associated STs. In 2/68 cases vs. 0.134/68 expected by chance, a pet and its owner were infected with an identical ST (ST45, ST658). Although common sources of infection and directionality of transmission between pets and humans were unknown, dog ownership significantly increased the risk for pet-associated human C. jejuni/coli infection and isolation of identical strains in humans and their pets occurred significantly more often than expected.

A biotracing model of Salmonella in the pork production chain.

In biotracing systems, downstream chain information and model-based approaches are used to trace the sources of microbial contamination in a food chain. This article includes the results of a biotracing model for Salmonella in the pork slaughter process chain. A Bayesian belief network model was used in which information on the Salmonella level at different locations in the slaughterhouse were used in combination with prior knowledge about the dynamics of Salmonella throughout the slaughter line. Data collected in a Dutch slaughterhouse were used to specify prior beliefs about the model inputs and to iteratively refine the distributions of the parameters in the model to obtain an optimal description of that specific slaughterhouse. The primary purpose of the model is to trace the sources of contamination for individual Salmonella-positive carcasses at the end of the slaughter line. The model results indicated that house flora on or in the carcass splitter was the source of contamination for many carcasses, especially for those that carried contamination on the cutting side. The results also indicated that the parameter values of the model may be subject to temporal variation and can be used as a tool to provide estimates of such trends. This model illustrates the concept of biotracing, gives insight into the dynamics of Salmonella in the slaughter line, and indicates the sites in the line where data collection is most effective for biotracing. This biotracing model was implemented as an interactive computer application, which is a step in the process toward an operational biotracing system by which a stakeholder can initiate immediate responses to Salmonella contamination and other hazards in the pork slaughterhouse.

A practical framework for the construction of a biotracing model: application to Salmonella in the pork slaughter chain.

A novel purpose of the use of mathematical models in quantitative microbial risk assessment (QMRA) is to identify the sources of microbial contamination in a food chain (i.e., biotracing). In this article we propose a framework for the construction of a biotracing model, eventually to be used in industrial food production chains where discrete numbers of products are processed that may be contaminated by a multitude of sources. The framework consists of steps in which a Monte Carlo model, simulating sequential events in the chain following a modular process risk modeling (MPRM) approach, is converted to a Bayesian belief network (BBN). The resulting model provides a probabilistic quantification of concentrations of a pathogen throughout a production chain. A BBN allows for updating the parameters of the model based on observational data, and global parameter sensitivity analysis is readily performed in a BBN. Moreover, a BBN enables "backward reasoning" when downstream data are available and is therefore a natural framework for answering biotracing questions. The proposed framework is illustrated with a biotracing model of Salmonella in the pork slaughter chain, based on a recently published Monte Carlo simulation model. This model, implemented as a BBN, describes the dynamics of Salmonella in a Dutch slaughterhouse and enables finding the source of contamination of specific carcasses at the end of the chain.

Strengths and weaknesses of Monte Carlo simulation models and Bayesian belief networks in microbial risk assessment.

We discuss different aspects of farm-to-fork risk assessment from a modelling perspective. Stochastic simulation models as they are presented today represent a mathematical representation of nature. In food safety risk assessment, a common modelling approach consists of a logic chain beginning at the source of the hazard and ending with the unwanted consequences of interest. This 'farm-to-fork' approach usually begins with the hazard on the farm, sometimes with different compartments presenting different parts of the production chain, and ends with the 'dose' received by the consumer or in case a dose response model is available the number of cases of illness. These models are typically implemented as Monte Carlo simulations, which are unidirectional in nature, and the link between statistics and simulation model is not interactive. A possible solution could be the use of Bayesian belief networks (BBNs) and this paper tries to discuss in an intuitive way the possibilities of using BBNs as an alternative for Monte Carlo modelling. An inventory is made of the strengths and weaknesses of both approaches, and an example is given showing an additional use of BBNs in biotracing problems.