Beyond the D Value - Modeling Curvature in Microbial Death Data. J. D. Legan and M. Peleg, Kraft Foods Technology Center, University of Massachusetts
The ‘traditional’ approach to analysis of microbial death data is based on the dogma that microbes die following a first order process, and hence that plots of log survival ratios vs. time are linear. This approach is mathematically convenient since the time needed for a one-log decrease in survival (the ‘D value’) is simply the reciprocal of the slope, which can be easily determined. Food processes are set to give some desired ‘safe’ result, e.g., 12 log reduction in Clostridium botulinum or 6-log reduction in Salmonella. Examination of isothermal microbial survival curves, however, frequently shows that they have a certain degree of curvature, often being described in terms such as ‘shoulders’ or ‘tails’. In such cases passing a straight line through the data should be discouraged since it would introduce a large degree of uncertainty into the required process conditions. A problem with the conventional method to calculate the efficacy of thermal processes, in terms of an ‘F value’, is the reliance on an arbitrary reference temperature. Both problems can be avoided if the isothermal survival curves are not considered to be necessarily linear. Instead, it can be assumed that the momentary inactivation rate only depends on the actual mortality patterns of the organism under isothermal conditions in the given medium, the momentary temperature and the momentary survival ratio. The resulting differential equation can then be easily solved numerically using currently available commercial software to produce the survival curve under almost any given temperature profile. The advantages are considerable for process setting and for any risk assessment that includes microbial death during processing as part of exposure assessment. The concept is demonstrated with simulated and actual survival data of food pathogens.
Go to . . .
2001 SRA Annual Meeting Table of Contents
2001 SRA Annual Meeting Author Index
Main Abstracts Menu Page
RiskWorld Home Page