Studies to determine the appropriateness of the use of populations of indicator bacteria on poultry carcasses for process verification were undertaken in commercial slaughterhouses. Samples were collected from neck skin by excision or from whole carcass rinses and were examined for a range of presumptive process hygiene indicator bacteria. Coefficients of variation were calculated for each bacterial indicator and were significantly lower in excised samples, indicating more reproducible bacterial recovery by this sampling method. Total viable counts of aerobic bacteria, Enterobacteriaceae, and Pseudomonas in samples collected by excision had the lowest coefficients of variation when compared with other indicators and were therefore used for further study. The uncertainties associated with the quantification of each bacterial indicator were calculated and were lowest overall for total viable counts of aerobic bacteria. In general, uncertainty was higher for lower bacterial numbers. Results of microbiological testing on pooled excised neck skin samples were not significantly different from the mean of individually analyzed samples. Bacterial numbers increased by 1 log unit when cultures were stored under chilled conditions typical of those used for transporting samples to external laboratories, but the increases were not significant for Pseudomonas and aerobic bacteria when storage time was less than 17 h. Weak relationships were identified between bacterial indicator numbers and duration of processing, although cleanliness of the processing environment diminished visibly during this time. In the plants visited for this study, there was a poor relationship between presumptive bacterial indicator numbers and process hygiene. Consequently, bacterial analyses for process verification purposes may be of limited value.
Evidence of exactness of the mean-field theory in the nonextensive regime of long-range classical spin models