Author
 |
Tamplin, Mark |
|
Submitted to: Meeting Abstract
Publication Type: Abstract Only Publication Acceptance Date: 10/15/2002 Publication Date: 12/1/2002 Citation: Tamplin, M.L. 2002. Improving microbial models to estimate the behavior of bacterial pathogens in food procesing environments. U.S.-Japan Natural Resources Panel. Abstract p. C1-5. Interpretive Summary: Technical Abstract: Microbial models are useful tools to determine the behavior of bacteria under various environmental conditions. To achieve high accuracy and utility, models must be based on experimental data that have been collected over a range of relevant experimental conditions. The resulting models include response surfaces that describe microbial behavior, as well as software interfaces to facilitate model application and interpretation. Importantly, models must be validated under both experimental and non-experimental conditions to define model accuracy and bias. In this regard, research in the Microbial Food Safety Research Unit at the USDA-ARS Eastern Regional Research Center (ERRC) is focused on developing and validating models for foodborne bacterial pathogens in food processing environments. The resulting information is relevant to the development of quantitative microbial risk assessments and in the design and implementation of Hazard Analysis and Critical Control Points (HACCP) systems. The ARS Pathogen Modeling Program (PMP) is a group of models that are highly utilized by industry, government, and academia to access the behavior of bacterial pathogens in food. In this regard, the USDA Food Safety & Inspection Service encourages the food industry to use microbial models to develop HACCP plans and to determine appropriate corrective actions when deviations occur. However, because the majority of the PMP models are based on data derived from bacteriological broth studies, the end-user must ensure that the model has been validated for each specific food processing application. Recent research shows that predictions of broth-based models can markedly vary from observed outcomes in foods. Such variations can result from direct effects of the natural food matrix, food additives, microbial competition, strain heterogeneity, and pathogen density and physiology. These variations can occur for lag and exponential bacterial growth phases, resulting in under- and over-estimations of contamination levels. On-going research at the ERRC will result in more accurate models which can be used by the food industry to define specific controls for bacterial pathogens. The newly formed ERRC Center of Excellence in Microbial Modeling and Informatics is also meeting this need through a network of international scientists that contribute information to the field of Predictive Microbiology. |
