Skip to main content
ARS Home » Research » Publications at this Location » Publication #170784
Title: QUANTIFYING THE ROBUSTNESS OF A BROTH-BASED, E. COLI O157:H7 GROWTH MODEL IN GROUND BEEF

Author
item CAMPOS, DANIEL - MICH. STATE UNIV.
item MARKS, BRADLEY - MICH. STATE UNIV.
item POWELL, MARK - USDA, ORACBA
item Tamplin, Mark

Submitted to: Journal of Food Protection
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/10/2005
Publication Date: 10/1/2005
Citation: Campos, D., Marks, B., Powell, M., Tamplin, M.L. 2005. Quantifying the robustness of a broth-based, e. coli o157:h7 growth model in ground beef. Journal of Food Protection. 68:2301-2309

Interpretive Summary: Mathematical models that describe the behavior of bacterial pathogens in foods are useful tools for designing effective food safety systems and in the development of microbial risk assessments. The value of such models is increased when the user knows the accuracy of the model predictions for different food matrices. In these studies, a new measure of model performance, a robustness index (RI), was developed. The RI is the ratio of the error of a model's prediction for microbial behavior in a new food matrix, compared to the error of the model against the original data used to develop the model. To illustrate the application of the RI, we evaluated a model developed for Escherichia coli O157:H7 in microbiological broth against data of how this microbial pathogen grows in raw sterile ground beef at different storage temperatures. The results showed that at temperatures between 15 to 40°C, the RI was close to and smaller than 1, indicating that the growth model was relatively robust in that temperature range. However, at lower storage temperatures, the model was less robust, showing RI values approaching 3. The RI will prove to be a useful tool for comparing the accuracy of models for different types of foods potentially contaminated with microbial pathogens.

Technical Abstract: Assessing the robustness of a microbial growth model is essential before applying it to additional food matrices. Therefore, a methodology for quantifying robustness was developed. A robustness index (RI) was computed as the ratio of the standard error of prediction (SEP) to the standard error of calibration (SEC) for a given model, where the SEC was defined as the root mean square error of the growth model against the data (log10CFU/g vs. time) used to parameterize the model, and the SEP was defined as the root mean square error of the model against an independent data set. The use of this technique was illustrated by evaluating the robustness of a broth-based model for aerobic growth of Escherichia coli O157:H7 in predicting growth in ground beef at different storage temperatures. Comparison against previously published data (132 data sets = 1,178 time-log10 cfu data points) from experiments in ground beef at various experimental conditions (4.8-45°C, 5.5-5.9 pH) yielded RI values ranging from 0.11 to 2.99. The estimated overall RI was 1.13. At temperatures between 15 to 40°C, the RI was close to and smaller than 1, indicating that the growth model is relatively robust in that temperature range. By quantifying the predictive accuracy, relative to the expected accuracy, the RI could be a useful tool for comparing various models under different conditions.