Predictive microbiology: towards the interface and beyond.

Abstract

This review considers the concept and history of predictive microbiology and explores aspects of the modelling process including kinetic and probability modelling approaches. The "journey" traces the route from reproducible responses observed under close to optimal conditions for growth, through recognition and description of the increased variability in responses as conditions become progressively less favourable for growth, to defining combinations of factors at which growth ceases (the growth/no growth interface). Death kinetics patterns are presented which form a basis on which to begin the development of nonthermal death models. This will require incorporation of phenotypic, adaptive responses and may be influenced by factors such as the sequence in which environmental constraints are applied. A recurrent theme is that probability (stochastic) approaches are required to complement or replace kinetic models as the growth/no growth interface is approached and microorganisms adopt a survival rather than growth mode. Attention is also drawn to the interfaces of predictive microbiology with microbial physiology, information technology and food safety initiatives such as HACCP and risk assessment.