The USDA news release dated April 5, 2011, and reported in Food Safety News April 6 by Helena Bottemiller, had this for a headline:


              USDA Announces Proposed Test and Hold Requirement

                              for Meat and Poultry Products

          Will enhance existing procedures, help reduce foodborne illnesses

There has been a lot of press and follow up discussions since that press conference about the need or lack thereof for this new requirement. Much emphasis has also been placed on the possible effect it will have on the very small processors who grind for local customers based on their order for the day. 

In Ms. Bottemiller’s story I was quoted as saying the requirement might be “onerous” for some of the smallest grinders. Doc Mudd tried to take me to task in the discussion following the story, but I will point out right now I did not say whether the required test and hold policy was a good one or not, simply that it might put some very small grinders out of business.


All the numbers you need for a discussion are right here in the Federal Register. These numbers need to come to light for further discussion about the accuracy of the headline that says this “will … help reduce foodborne illness.”


This discussion is especially important if one reads the transcript from the press conference where this proposed new policy was announced. During the Q and A, Secretary Vilsack is quoted as saying “our best estimate is this is going to help prevent 25,000 illnesses.” You can see this for yourself and not have to take my word for it.


If only that were true. If only 25,000 illnesses could be prevented then I might have embraced the AMI petition when they sent it to me as the Undersecretary for Food Safety in June of 2008. If it were true, I would not be worried about one or two very small grinders having to close if they cannot comply with the new test and hold proposal.

But it is not true. A three-year time frame, 2007-2009, was used by the USDA for their statement that 44 recalls happened because product was tested and shipped into commerce before a positive result came back. Of those recalls, 22 were for ground beef with an E coli O157:H7 positive, and 22 were for ready-to-eat products with positives for L. monocytogenes.

Of those 44 recalls, 43 had this statement in the recall notice: “no reports of illnesses associated with consumption of this product.” One recall did have two illnesses associated with it. So the headline that this action will “help reduce foodborne illnesses” is technically correct. But 25,000?

The proposed regulations in the Federal Register, written by FSIS staff, include this statement: “The FSIS model estimated the upper 95% confidence bound of averted E. coli O157:H7 illnesses to be approximately 2.6 for a three year period.”

It goes on to state that the estimated number of L. monocytogenes illnesses that would be averted over the same three year period of time would be 0.18.


These numbers are a long way from the 25,000 stated by the Secretary. That number impressed many in the media and they touted this as a great public health move in the right direction.

Furthermore, FSIS estimates the fiscal savings by avoiding these illnesses would be $7,464 annually, and I seriously question whether that minimal savings warrants the risk of losing some small grinders in rural America. I even question whether an annual savings of 7K warrants the amount of time, energy and money that went into writing the proposed regs in the Federal Register.


And I definitely question whether reducing foodborne illnesses by one case per year, when the CDC estimates there are 63,000 foodborne illnesses per year, warrants a press conference and the media hype.

I think the consensus of the responses I have been seeing echoes these thoughts. Test and hold is a good thing, and the great majority of the grinders already do it. The impact is not going to be a significant reduction of illness, but rather a reduction of negative messages being sent to consumers every time there is a recall. These messages remind them that raw ground beef cannot be considered pathogen free.

The AMI petitioned the Undersecretary back in 2008 because they wanted to improve public relations and all of their members already practiced test and hold, and had been doing so for years. I think another motive for the petition might have been that every time a small grinder needed to do a recall, people like John Munsell were quick to point out that the contamination probably did not occur at the very small grinder, but at the large packing house that they bought ground beef components from.


And the industry as a whole did not embrace mandatory test and hold at that time; the Federal Register points that out very clearly. To say it does now is another inaccuracy.

Looking for total transparency, here are a few more numbers that come from the USDA and FSIS web pages and might shed more light on the true impact of this action.

The total pounds of ground beef recalled as a result of FSIS testing in 2009 were 52,667. Ground beef produced and consumed in the United States, as estimated by the USDA, in 2009 was over 20 billion pounds. The recalled ground beef amounted to 0.00026% of the total produced in 2009.

Total samples tested for ground beef in 2009 amounted to 11,686, of which only 41 were positive, or 0.35% of samples. Of these, 33 were held and then destroyed or cooked and eight, as stated earlier, were out in commerce.

Remember, that is eight lots out of 11,686 lots that we are talking about here.

Lastly, these very small grinders are only subjected to testing for E coli O157:H7 approximately once every three months. That means they can grind and sell product that is not tested every other day of the quarter. How much of that product would test positive?

I don’t know, but I know test and hold is not going to make the ground beef I buy at the local store significantly safer.