Some Practical and Statistical Aspects of the Comparative Evaluation of Microbiological Culture Media
Quality assessment of culture media for use in a laboratory is an essential requirement of laboratory accreditation procedures. Over some years a number of procedures have been developed to permit comparison of the performance of batches of both liquid and solid culture media, whether made in the laboratory or purchased from outside suppliers. These procedures include estimation of the productivity ratio (PR), selectivity factor (SF) and other parameters for which various acceptance criteria have been recommended. However, such criteria have not been set in a systematic manner that recognises the need for statistical appraisal of the data. This has resulted, for instance, in confusion as to whether or not the counts of colony forming units (cfu) should be compared before or after mathematical transformation. This chapter discusses statistical aspects of the tests, their shortcomings and benefits and illustrates simple ways in which data can be compared.