Predictive modeling of lag time and growth phase of Listeria monocytogenes in ready-to-eat meat products.

Ready-to-eat (RTE) meat products remain a significant food safety concern due to the persistence of Listeria monocytogenes. Predictive microbiology offers an efficient alternative to traditional experimental approaches for estimating microbial growth and supporting shelf-life determination. Generalized broth-based models overestimate growth, causing over-processing while food-specific models provide more realistic predictions for shelf-life determination. This study developed and validated gamma-based predictive models for L. monocytogenes growth in RTE beef, pork, and poultry products using temperature, pH, water activity, nitrite, and four organic acids as environmental factors. Meat-specific models were compared with a combined all-meat and poultry model using 978 growth curves from literature and ComBase. Growth rates were estimated using logistic-with-delay and Baranyi primary models (R2 > 0.9), and secondary gamma models were developed using an 80:20 training-validation approach. During development, the gamma logistic-with-delay approach showed higher R2 (0.72-0.89) and lower RMSE (0.054-0.075 h-1) than the gamma Baranyi model. During validation, it provided more accurate predictions, while the Baranyi model underestimated growth. Overall, the logistic-with-delay gamma-based all-meat and poultry combined model reduced prediction error across meat types. Lag time was modeled using the Relative Lag Time (RLT) approach and the Baranyi h₀ method. The RLT approach showed higher correlation (R2 = 0.58-0.84), lower RMSE (0.779-1.190 ln(h)), and lower prediction error, whereas the h₀ method resulted in a higher proportion of fail-dangerous predictions. Overall, the gamma logistic-with-delay framework combined with the RLT method provides a robust and precautionary strategy for predicting L. monocytogenes growth and lag behavior in RTE meat products, supporting improved risk assessment and shelf-life management.