DeLauro Urges USDA IG to Investigate FSIS Beef Testing Protocol
Asks IG to investigate scientific merits and potential shortcomings of industry testing method adopted by FSIS
Washington, DC – Congresswoman Rosa L. DeLauro (D-CT) today, in light of the recent E. coli outbreak involving ground beef, sent a letter to the USDA Office of the Inspector General (IG) calling for an investigation into the scientific merits and potential shortcoming of the N-60 testing protocol, which is a beef testing protocol promoted by the meat industry and relied upon by USDA. The N-60 testing protocol is used by most of the largest beef trim manufacturers to verify that product is free from E. coli.
"To protect the public health, verification testing must be robust and ensure that adulterated product is not sold to the American consumers," DeLauro wrote. "I am troubled by the shortcomings of the N-60 test and the associated food safety implications, and that is why I am requesting an investigation into the scientific merits of this beef testing protocol," DeLauro stated.
The full text of the letter is as follows:
November 12, 2009
Phyllis K. Fong
Inspector General
United States Department of Agriculture
Room 117-W Jamie Whitten Bldg
1400 Independence Avenue SW
Washington, DC 20250
Dear Ms. Fong:
I am writing to express concern about the efficacy of E. coli O157:H7 testing of beef trim products that is conducted by the industry as well as by the U.S. Department of Agriculture (USDA). The N-60 testing protocol is used by most of the largest beef trim manufacturers to verify that product is free from E. coli O157:H7. Every day, USDA Food Safety and Inspection System (FSIS) relies on these industry tests and applies its seal of approval to any lot of product in which industry tests found no contamination. FSIS also has adopted the N-60 protocol for its own sampling of domestic and imported beef.
To protect the public health, verification testing must be robust and ensure that adulterated product is not sold to the American consumers. Therefore, I strongly urge the Office of the Inspector General to investigate the scientific merits and potential shortcomings of the N-60 test and the application of N-60 test results to beef trim products.
As you know, a current food-borne illness outbreak involving E. coli contamination in ground beef has resulted in at least two deaths and over two dozen illnesses in 11 states. This outbreak has been traced to a meat processing facility that employs a ‘test-and-hold policy,' which is the practice of maintaining control over a production lot until microbiological test results are obtained. In addition, this particular facility samples product every 10 to 20 minutes to check for contamination, which likely is more frequent than the industry average. However, despite these precautions, it was not enough to prevent contamination.
Beef manufacturing trimmings may be processed into raw ground beef that is sold to and consumed by the general public. It is crucial that these beef trimmings be free of harmful bacteria, like E. coli O157:H7, to ensure the safety of our food supply. FSIS currently relies on industry N-60 testing to verify that these products are free of E. coli O157:H7.
While it is important to strengthen our food safety testing methods, I am troubled by the shortcomings of the N-60 test and the associated food safety implications. The agency has made several fundamental changes to its position on related issues, including the definition of a "sampled lot." It also has shifted from emphasizing the ability of E. coli O157:H7 to cross-contaminate product to the belief that E. coli O157:H7 is a point-source contamination event. FSIS provided no documentary support for this shift in several public meetings during which the N-60 protocol was discussed.
These decisions and others have affected the agency's policy for verification testing of beef trim. My concerns about the N-60 testing protocol may be categorized as follows: the statistical validity of the test; the sample collection and analysis; and the application of test results.
In the course of the investigation into the N-60 testing protocol, I urge you to answer the following questions based on the categories outlined above.
Statistical Validity of N-60 Testing
· What is the prevalence of E. coli O157:H7 in domestic beef trim used for raw ground beef production at processing facilities in the United States? When and how was this estimate reached? How frequently will it be re-assessed?
· What is the confidence level of the N-60 sampling method as currently performed? Is that level appropriate for ground beef?
· What is the definition of a product "lot" used in the N-60 testing and how was this definition determined? How is this definition applied to individual processing establishments? Is this an appropriate application?
· How does this definition affect the statistical viability of this test? How confident are we that lots testing negative are truly negative if the definition of "lot" varies among establishments?
Sample Collection and Analysis
· Are all samples collected, stored, shipped, and analyzed by trained FSIS employees?
· Are laboratories instructed to reject samples that do not meet minimum quality standards? If so, what are those minimum quality standards? How is it assured each sample meets them?
· Are the minimum standards associated with sample collection, storage, shipment, and analysis adequate? What type of testing standards must be met by laboratories analyzing the samples?
· Because E. coli O157:H7 is not necessarily distributed homogenously in a product, how did the agency decide on its current collection methodology? Are samples collected truly representative of the entire lot?
Application of N-60 Test Results
· Has FSIS established a protocol for reassessing an establishment's HACCP plan based upon N-60 test results? What actions are taken at an establishment after a positive N-60 test result?
· What is the implication of a positive N-60 test result on the individual lot tested? On other lots from the same establishment produced on the same date?
· How are the FSIS N-60 test results and the establishment's N-60 test results correlated? How do N-60 testing protocols differ between FSIS and the industry?
· What are the implications of the USDA label associated with a negative N-60 test?
Thank you very much for considering conducting this investigation. I look forward to your response.
Sincerely,
ROSA L. DeLAURO Chairwoman, House Appropriations Subcommittee on Agriculture Rural Development, Food and Drug Administration, and Related Agencies