Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: DATA ACQUISITION AND MODELING FOR POULTRY FOOD SAFETY

Location: Residue Chemistry and Predictive Microbiology

Title: Extrapolation of a predictive model for growth of a low inoculum size of Salmonella typhimurium DT104 on chicken skin to higher inoculum sizes

Author
item Oscar, Thomas

Submitted to: Journal of Food Protection
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 4, 2011
Publication Date: October 1, 2011
Citation: Oscar, T.P. 2011. Extrapolation of a predictive model for growth of a low inoculum size of Salmonella typhimurium DT104 on chicken skin to higher inoculum sizes. Journal of Food Protection. 74(10)1630-1638.

Interpretive Summary: Computer models that forecast how levels of human disease causing microorganisms in food change as a function of food handling and storage conditions are valuable tools for helping improve food safety. Development of such models in food is time consuming and expensive. Thus, if it can be demonstrated that a computer model can predict pathogen behavior in food as a function of new conditions that were not used in model development considerable time and money could be saved by identifying situations for which new models are not needed. In the current study, a computer model was developed and validated for predicting growth of a low level (10 cells) of Salmonella on chicken stored at 68 to 113F for up to 8 h and then ability of the model to predict growth from higher levels (100, 1,000 or 10,000 cells) of Salmonella contamination was evaluated. Salmonella are a leading cause of foodborne illness and poultry foods are often implicated as sources of salmonellosis in humans. The model was found to provide good predictions of growth from higher levels of Salmonella for most times and temperatures of storage investigated. However, at 86 to 113F, Salmonella grew faster from the two highest levels and thus, the model provided bad predictions in these situations. It was concluded that including initial level of Salmonella in the model would improve its ability to predict food safety.

Technical Abstract: Validation of model predictions for independent variables not included in model development can save time and money by identifying conditions for which new models are not needed. A single strain of Salmonella Typhimurium DT104 was used to develop a general regression neural network model for growth of a low inoculum size (0.9 log) on chicken skin as a function of time (0 to 8 h) and temperature (20 to 45C). Ability of the model to predict growth of higher inoculum sizes (2, 3 or 4.1 log) was evaluated. When the proportion of residuals in an acceptable prediction zone (pAPZ) from -1 (fail-safe) to 0.5 (fail-dangerous) log was greater than or equal to 0.7, the model was classified as providing acceptable predictions of the test data. The pAPZ for dependent data was 0.93 and for independent data for interpolation was 0.88. The pAPZ for extrapolation to higher inoculum sizes of 2, 3 or 4.1 log were 0.92, 0.73 and 0.77, respectively. However, residual plots indicated local prediction problems with pAPZ less than 0.7 for an inoculum size of 3 log at 30, 35 and 40C and for an inoculum size of 4.1 log at 35 and 40C where predictions were fail-dangerous indicating faster growth at higher inoculum sizes. The model provided valid predictions of S. Typhimurium DT104 growth on chicken skin from inoculum sizes of 0.9 and 2 log at all temperatures investigated and from inoculum sizes of 3 and 4.1 log at some but not all temperatures investigated. Thus, the model can be improved by including inoculum size as an independent variable.

Last Modified: 9/21/2014
Footer Content Back to Top of Page