I have updated an article on food sampling and microbial testing following a book chapter that I have been writing on raw drinking milk (RDM). Recent changes in legislation in the UK require critical pathogens to be absent in 25 ml of RDM and in the absence of a Critical Control Point particular emphasis is being placed on microbiological testing to ensure safety. This article consider the distribution of microorganisms including clusters in foods and the limitations of testing are discussed. The economic and public health significance of clustering is illustrated by the meticulous work of Jongenburger et al. (2011). These workers investigated why a consignment of powdered infant milk formula (PIM) produced in Europe (the batch of approximately 22,000 kg was produced in 3 shifts) was rejected because of contamination with Cronobacter spp. by the importing Chinese authorities. Jongenburger et al. (2011) sampled 31 bags close in production time to the bag that tested positive for Cronobacter spp. in China using multiple samples (2290) and found 8 positive samples from 3-bags. The article also considers the question: assuming that all the samples taken for testing are negative, what might the probable prevalence of a pathogen or fault-causing organism in a food be? Interactive calculators to determine the statistical significance of sampling schemes and the effect of sample numbers and mass on prevalence are provided. The article is available at https://www.dairyscience.info/index.php/food-model/275-sampling.html .
Microbial testing is still important but it is critical to understand its limitations in assuring food safety.