Quantitative analysis of intervention technologies on reduction of Listeria monocytogenes in poultry products

To control Listeria contamination in ready-to-eat (RTE) poultry products, attentions have been focused on the post-package pasteurization process. Thermal inactivation of L. innocua in fully-cooked chicken breast strips during post-package hot water pasteurization was investigated at 60, 70, 80, and 90°C. Four different primary models were used to fit survivor curves and modified Gompertz and Weibull model were selected as the best-fitted models based on the evaluation of goodness-of-fit and parameter estimates. The performance of primary models was validated against independent data from pilot plant tests. Secondary models based on the modified Arrhenius model were successfully applied to predict the effect of temperature on the kinetic parameters obtained from the modified Gompertz model. The root mean square error (RMSE) and 95% prediction interval indicated a good fit of the secondary models.;A quantitative risk assessment model was developed to estimate Listeria monocytogenes concentration in RTE chicken breast products from in-plant cross-contamination to retail storage. The relative importance of intervention strategies on the reduction of final L. monocytogenes concentration were evaluated through risk simulation. Scenarios analysis was used to evaluate the effect of different intervention scenarios on the model output. The result of scenario analysis showed that different intervention strategies could result in 0.03% to 63% of final bacterial concentration as compared with the baseline model. Increasing the maximum irradiation dose to 2.5 KGy was the most effective intervention scenario, followed by increasing the maximum post-package heating time to 5 min and median heating temperature to 75°C. The developed risk assessment model will provide useful information for risk managers to determine appropriate measures for reduction of L. monocytogenes contamination in RTE poultry products.;Thermal inactivation of L. monocytogenes in chicken drumettes during post-package hot water pasteurization was also studied. Fully-cooked chicken drumettes were treated for combinations of different times and temperatures. Weibull model was found to provide a good fit for the bacterial survival curves based on the RMSE, residual plots, and plots of predicted versus observed bacterial survivors. The Weibull model was validated by predicting a new data set generated in the pilot plant. The bias factors and accuracy factors ranged from 0.97 to 1.17 and from 1.05 to 1.15, respectively, indicating an optimal performance of the model. The predictive models could be used to design thermal processing schedules and integrated into risk assessment model for evaluation of intervention technologies.