AOAC RI ERP Final Action Recommendations eBook

Collaborative Study Approved Protocol Expert Review Panel Use Only

1 2 3 4 5 6 7 8 9

g and 375 g) and chicken carcass rinse (USDA-FSIS MLG); cream cheese, fresh bagged lettuce, and stainless steel (FDA BAM); and raw ground beef, dry pet food (375 g), and stainless steel (HC CAM). Inclusivity testing demonstrated that the BAX method was reactive with 317 Salmonella isolates, representing over 100 different serotypes. The BAX method did not detect 37 different

non- Salmonella strains tested.

Following the completion of the PTM study, an additional 17 matrices (15 foods/beverages and 2 environmental surfaces) were evaluated in a pre-collaborative study. Orange juice, frozen peas, peanut butter, shell eggs, shrimp, white pepper, and infant formula were compared to the FDA BAM method; ice cream, nonfat dry milk, and cocoa were compared to both the FDA BAM and HC CAM methods; hot dogs, beef trim, ground beef with soy, and ground turkey were compared to the USDA-FSIS MLG method and HC CAM methods. Environmental surfaces (ceramic tile and plastic) were compared to both the FDA BAM and the USDA-FSIS MLG methods. Method comparison data is presented in Appendix 10.1. Results obtained using the BAX® System Real Time Salmonella test kit indicate no statistical difference when compared to the corresponding reference method results. A full pre-collaborative study manuscript is

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

currently being drafted.

1.3

Study Directors

Morgan Wallace, Ph.D.

DuPont Qualicon ESL Building 400

PO Box 80400

Route 141 & Henry Clay Road Wilmington, Delaware 19880

Phone: 302.695.5473 Fax: 302.695.5877

Email: Morgan.Wallace@usa.dupont.com

Patrick Bird

Q Laboratories, Inc. 1400 Harrison Avenue Cincinnati, Ohio 45214 513-471-1300 ext. 259

Email: pbird@qlaboratories.com

2.0 39 40 At least 13 laboratories will be solicited to participate in this study. A minimum of 10 valid laboratory 41 data sets per matrix are required to successfully complete the study. Each collaborator will receive 42 Collaborators

4