Validation of the Predictive Models for the Behaviour of Salmonella Enteritidis during Storage of Yoghurt at Different Temperatures

Introduction: Yoghurt is a fermented dairy product usually considered healthy and safe. However, a predictive model recently developed for the temperature range of 4-25ºC showed that Salmonella may survive for days in yoghurt during storage.

Purpose: Predictive models for the behaviour of Salmonella Enteritidis in yoghurt were validated.

Methods:  Salmonella survival data in yoghurt were obtained at 16^oC by plate counting method and then a fit-for-use, survival model with shoulder and tailing (Geeraerd et all., 2000) using the package nlsMicrobio in R software (version 3.2.2). The parameters, survival rate (k_max) and shoulder (S_l), were compared with the estimates given by the previously, developed predictive models. The validation procedure was performed based on the use of the bias factor (Bf) and the accuracy factor (Af).

Results: The kinetic parameters for the survival curves obtained at 16ºC were found to be 0.17 h^-1 (k_max) and 28.64 h (S_l). The Bf and Af values of the secondary models for k_max - temperature were calculated as 0.82 and 1.21, respectively. The S_l, Bf and Af values were 0.80 and 1.24, respectively. According to these indices, the model predictive capacity can be considered acceptable, although it sub-estimated both survival rate and shoulder length, resulting in fail-safe predictions.

Significance: Validation of models is essential to evaluate the construction and robustness of models. The results confirm that the developed models can be used to predict the behaviour of Salmonella Enteritidis during storage of yoghurt at different temperatures.