Microsoft Word - Candidates for 2018 ERP of the Year

276

Cons/Weaknesses of the Manuscript: ER 1 None. ER 2

The manuscript is lengthy.

NA

ER 3 ER 4

Need to elaborate on the issues and possible causes of those issues of laboratories whose data were not used in the study. • Correction on table 2 page 19 for cottage cheese matrix, AOAC OMA 993.12, lab 6 is excluded in the analysis and should have a superscript b. • It is unclear as to why the reference method chosen is not the BAM reference method which would be in this reviewer’s opinion a better and more comprehensive method to use. While the reference method used is an OMA method, the use of the BAM method would have given more confidence to regulatory bodies reviewing this method. • The uses of positive and negative controls are not well defined. • There may be some value in stating the ISO status of the testing labs to add more confidence on the results. • The reasons for the exclusion of labs 6 and 13 for the cottage cheese sample set and lab 8 and 10 for the deli turkey sample set may need more detail. In all cases the data set suggests that these labs were simply excluded since they detected false negative results which I am sure is not the case; however, no detail scientific explanation is provided. • While many of the acronyms used in the data and statistical tables are well known and are described in the Appendix J of the AOAC method validation guidelines, it may be beneficial to include a section on the explanation of these terms. Since this is my first time reviewing such a manuscript, I am not sure if these were included in the past but they may help some understand the tables. There are some mistakes in the report especially regarding the total number of positive samples for the 2 methods and that creates some controversy between text and results. For cottage cheese, page 13, from line 27, candidate method is claimed at 65/132 presumptive positive for low level when 66 shall be accounted (64 confirmed & 2 false positive, according to table 3). The report claimed 63 were confirmed when 64 were confirmed from the presumptive positive (from table 3) + 1 false negative (lab 8).The same applies to the reference method for which it's reported 73/132 positive for low level, but table 3 gave only 67 (the same number as for the MDA Listeria spp study which was using the same samples and therefore the same reference method and so results). These numbers will modify all the calculations reported in table 2014.1A. For deli turkey, page 14, from line 16, the false positive and false negative results obtained at low level for the candidate method shall be clearly disclosed: 66 presumptive positive results of which 64 were confirmed and 2 were not confirmed and so were false positive results. Additionally, 3 other samples were found negative by MDA Listeria monocytogenes but confirmed positive (3 false negative: one from lab 12 and 2 from lab 15). Did lab 12 participate in the ring trial (it has a note "b" in table 4 stipulating that results were not used ... If so the text shall be corrected accordingly; page 20 from line 8). Table 2014.2A, page 26, for lab 1, X = 5 for CP column: from table 3 page 19, I conclude that X = 6 as table 3 reports 5 MDA confirmed positive and 1 false positive (noted "c" & footnote claims false positive results). Same remark applies to lab 3 for which X = 7 when it shall be 8.

ER 5

ER 6

NO

ER 7

03/12/2018