AOAC ERP MICRO AUGUST 2018

Statistical Review of Enumeration of Enterobacteriaceae in Select Foods and Environmental Surfaces Peel Plate EB

Complements on the extensive validations and on the manuscript detailed description of the method as well as the studies. As the raw data are not included in the report, statistical calculations cannot be checked thoroughly. Units are missing in the manuscript, even in the abstract. Log is not a unit, but part of a numerical value. The units are cfu/test portion, so for the mean values, the standard deviations, the mean difference and the upper and lower confidence level thereof, “log 10 cfu/ g(mL)” should be included behind the figures. Line 19-21: Please reformulate according to the suggestion of Stat Reviewer 1, that the results were log 10 transformed before statistical analysis, and that the alternative method and reference method are shown equivalent by the performance requirement of all 95% confidence intervals on mean difference falling between -0.5 and +0.5 log 10 cfu/g(mL). Please do not use “mean log difference”,- it is like “mean 100 difference” and when writing that the “confidence intervals greater than |0.5 log|” it would indicate that the interval is outside [-0.5;0.5] log 10 cfu/g (mL) not falling between -0.5 and +0.5 log 10 cfu/g(mL) (double mistake). Line 22-24: Include units for levels, mean difference, repeatability and reproducibility Line 58: “….within the ranges of 95% confidence intervals (CI) log (-0.5, 05) (1, 2, 3, 4)” – Please reformulate as above and remove reference No. 3 as that ISO standard is regarding definitions only. Line 67-69: “ Log 10 Mean Difference between Candidate and Reference Methods. …” Please consider reformulation according to above, and delete the first word “Log 10 ” in the headline. Line 85-86: “ The method limit of detection is 1 or greater colony forming units per milliliter or gram of test sample. The accurate quantitative range for Enterobacteriaceae is 1 to 150 CFU per test.” Any documentations, e.g. have you determined the LOD ?

Line 209: Should the number be 1-150?

Line 349-350: ..”mean log averages..” delete “log averages”

Line 351: “..mean log differences..” delete “log”

Line 462 & 503 ++: see comments above on the acceptability levels for the difference of the means

Line 466: Reference 17 seems missing

Line 484-493: Include units (log 10 cfu/g(mL)) after the parentheses Table 3: Column “achieved concentration levels” - cfu/mL for whole milk is OK. Should it be cfu/g for the solid samples?