Quantifying sources of error in Salmonella thermal inactivation models for meats and low-moisture foods.

摘要

Despite an incomplete understanding of the impact of experimental methodologies on resulting Salmonella thermal inactivation models, thermal resistance data and parameters continue to be generated, reported, and presumably applied to food safety analyses. Therefore, the objectives were to: (1) evaluate the impacts of varied experimental methods on the observed thermal resistance of Salmonella, (2) investigate the effects of substantively similar thermal inactivation methods on the quantification of Salmonella thermal resistance, and (3) investigate the effects of regression methods on the estimation of Salmonella thermal resistance parameters and their associated errors. These objectives were accomplished with two cross-laboratory comparison studies. The first study evaluated the effects of two Salmonella inactivation methods in ground beef on the resultant inactivation kinetics, based on data generated by two different laboratories. The two methods used yielded characteristically different Salmonella inactivation kinetics, regardless of the laboratory. The second study investigated the effects of five different inoculation methods on the subsequent stability and thermal resistance of Salmonella in wheat flour, and the repeatability of those results, based on data generated by two different laboratories. These methods yielded significantly different Salmonella thermal resistances, and only two yielded repeatable initial Salmonella populations and subsequent thermal resistances. Overall, thermal inactivation methods significantly impacted Salmonella thermal resistance in both meat and low-moisture food matrices.