September 27, 1999


Tom Billy, Administrator

Food Safety & Inspection Service

United States Department of Agriculture

Washington, DC 20250


Dear Tom:

During the FSIS-industry teleconference on Friday, I raised the issue of the Salmonella performance standard, and you said that you would be willing to look at this issue before we look elsewhere for a resolution.

The major issue is the science base of the agency’s Salmonella performance standard for raw meat and poultry, and the bias in that standard against ground beef derived from cows. The enclosed letters from NMA’s Ken Mastracchio to Associate Administrator Margaret Glavin dated May 4, 1999 and her response dated July 14, 1999 provide background correspondence.

Certain strains of Salmonella are known to be pathogenic to humans. Thus, it is highly desirable that raw meat and poultry be handled in such a manner as to reduce the incidence of this pathogen. This is basic to the Pathogen Reduction Final Rule published July 25, 1996, and we acknowledge the importance of reduction efforts even in light of new data released by USDA’s Economic Research Service on lower estimates of the disease by the Centers for Disease Control and Prevention. There is no data to suggest a scientific rationale for different performance standards by species of livestock and poultry. The same number of pathogenic Salmonella if directly ingested by a person will cause illness, regardless of whether the transmission vehicle was meat, poultry, or any other food for that matter. Thus, the fundamental flaw in the agency’s performance standard in terms of science is differentiation of the performance standard by class of product.

This is not the first time, nor likely to be the last, that the agency has bent the science to fit its regulatory intentions. To some extent, the industry has learned to accept such diversions and to live with them. In the instant case, the Salmonella performance standard, the agency has further conflicted itself as it developed the performance standards for different species by setting a standard for ground beef without recognizing the inherent differences between ground beef derived from cows/bulls compared to that derived from steers/heifers. The base line data, published in the Nationwide Beef Microbiological Baseline Data Collection Program: Cows and Bulls, December 1993 – November 1994 for these two classes of bovine species is significantly different from the baseline data published in the Nationwide Beef Microbiological Baseline Data Collection Program: Steers and Heifers, October 1992 – September 1993. A reasonable person would conclude that ground beef derived from the different species would recognize the significant difference.

Ms. Glavin states categorically: "We disagree." This is the wrong conclusion. She goes on to suggest that because of some new data that the industry has not yet seen, our concerns "will disappear." This is encouraging, but does not deal with the immediate situation in the industry. Further, our requests for the raw data remain unanswered.

When the final rule was published, there was broad understanding that standards and requirements would be subjected to constant revision based on new information and data. A performance standard is an action-forcing measure. In the current instance, the action is likely to force smaller firms in the industry that slaughter cows and bulls right out of business since the standard doesn’t recognize the inherent difference between the two major sub-classes of the bovine species. Rather the agency drew a line at the mid-point of the volume data for all classes of the bovine species, and those below the mid-point fail!

Now we can go on to pick differences with what we’ve been told, such as Ms. Glavin says that FSIS included 563 sample sets from establishments producing approximately 94 million pounds of raw ground beef per week, but her source material, the Nationwide Federal Plant Raw Ground Beef Microbiological Survey, August 1993 – March 1994, states that the 94 million pound number is attributed to 1590 plants of which only 661 were sampled and their data used for the final statistical summaries. This disparity is unexplained. It should also be noted that this survey was completed six months before the previously referenced survey on cows and bulls was completed, once again indicating that the data on this latter class of bovine species was not a consideration by policy makers at the agency in developing the performance standard. Further, this survey represents only an eight-month period and is not representative of the full of a full cycle of seasonal fluctuations as the two earlier surveys do. Another difference is the inherent bias of a substantial number of the sample sets from steer/heifer setting an average. It translates into the fact that about half of the companies, mostly those using cow/bull materials and frequently smaller firms, will fail the standard. Further, there are inequities with the enforcement provisions. Finally, Salmonella doesn’t start in packing houses, it arrives with the live animal, and the agency has not designed a pre-harvest QA effort to ensure elimination on the farm.

We have reviewed copies of the Decision Memoranda submitted to you by Phil Derfler on January 7, 1999 entitled: Decision on Implementation of Suspension Actions Following Three Consecutive Series of Salmonella Tests and Decision on Agency Procedures Following Suspension Actions for Failure to Meet Pathogen Reduction Performance Standards. Clearly, there were problems but there is no clarification as to why there were problems. Again, in working with our members, many of whom are producing ground beef exclusively from cow/bull meat, they are on the losing side of a line that is drawn at the mid-point of the data. This is an impossible standard because volume of the large packers is on the passing side of the line, assuring that the smaller volume of the smaller packers is on the losing side of the line.

Fairness and equity are at issue. If the industry is to accept different performance standards by species, which is not good science but a regulatory goal, then the agency needs to recognize the sub-sets within the bovine industry and set performance standards accordingly. We ask that you immediately suspend action on this performance standard against firms that are producing ground beef from cows/bulls, that you set a new standard that is statistically and species justified.

This is a very serious matter with National Meat Association. We look forward to your early response.


  Rosemary Mucklow

Executive Director