Assessing the equivalence of Vibrio parahaemolyticus MPN and PCR quantification methods in oyster samples: A...

Data collected from FDA proficiency tests (PT) during 2012-2018 was used to evaluate the performance of most probable number (MPN) and polymerase chain reaction (PCR) methods used to enumerate Vibrio parahaemolyticus in oyster samples. The primary aim was to establish whether the MPN and PCR methods can be considered equivalent. The following criterion for equivalence was applied: the absolute value of mean bias and between-sample standard deviation must both be less than 0.1 (log10). Final calculations showed mean bias and between-sample standard deviation (SD) were 0.031 and 0.117 (log10) respectively. The between-sample SD criterion was slightly relaxed because with close to 700 results, the data set was large and overall mean bias was low. It was concluded that the two methods can be considered equivalent. The use of PT data for the assessment of method rather than laboratory performance is a secondary topic addressed in this paper. Important requirements for this use of PT data include availability of sufficient results for both methods and use of real food matrices. Ultimately, the results presented here provide an example of how PT data can be used to monitor method performance across many laboratories and samples as well as to assess method equivalence.