The Christmas Eve Recall of E. coli-tainted steaks by National Steak and Poultry, that have sickened people in several states in yet as unnamed restaurants, made me think a bit more about the problems the meat industry and the FSIS has had over the last sixteen years (post JITB) in dealing with E. coli O157:H7 on and in whole muscle meat – specifically, needle-tenderized steaks.  It does have a bit of history as these slides from Minnesota show:

Since then there have been several additional E. coli outbreaks linked to needle-tenderized steaks.  Here was a recall of over 400,000 pounds.  It is likely to have sickened dozens at restaurants in at least one state.  Here was a recall in 2007.  And, another recall, with illnesses, from 2008.  Please, do not forget the outbreak and recall linked to Sizzler in 2000.

So, what is FSIS’s official position?  You will like this from 2006 from the FSIS website:

Q. Does FSIS test the mechanically tenderized products for E. coli O157:H7?

A. FSIS does not test the mechanically tenderized beef for E. coli O157:H7 at this time. Such raw products are non-intact beef products, however, and thus would be adulterated if contaminated with E. coli O157:H7. In Q&A #1 in attachment 1 to Directive 10,101.1, states that FSIS intends to sample non-intact beef products other than ground beef in the future.

In FSIS Notice 51.09, issued in July 31, 2009, the USDA announced its intent to start testing bench trim, i.e., the trim generated post-slaughter in cutting the carcasses into primals and subprimals. Before then, testing of trimmings was focused on trimmings generated from cutting the carcass prior to the final rail. This change was prompted by the Nebraska Beef and other recalls involving intact cuts from which trimmings had been generated and that ended up in ground beef, thus adulterating. FSIS Notice is here:

Here is the specific mention in the Notice at p. 2:

NOTE: If an establishment produces bench trim from steaks or roasts that are non-intact or are to be made non-intact (e.g., they have been or will be needle tenderized), a finding that the bench trim is positive for E. coli O157:H7 would evidence that the steaks or roasts are also positive and thus adulterated. However, the establishment may have a supportable basis for distinguishing the steak or roast from the bench trim. For example, if the establishment applies an antimicrobial treatment to the steaks or roasts before tenderization, but not to the bench trim, the establishment may be able to support that the positive applies only to the bench trim.

To its credit, this was the first time that the agency ever started to connect (duh!) that if trimmings tested positive, the “intact” cuts from which the trimmings came might be positive too.

As for the policy statement that first distinguished between intact and non-intact, and identified needle-tenderized meat as NON-intact (and thus subject to zero-tolerance), that can be found here:

Our friends in the meat industry lobbied to have needle-tenderized steaks excluded from the E. coli O157:H7 policy, based on this research.  Here is interesting and ongoing research on the issue of heat-inactivation of O157 in needle-tenderized steaks, and the risk that such steaks are perceived as being intact and thus won’t be cooked to 165 degrees.

A blog post my law partner, Denis Stearns did about the JBS recall discusses the intact vs. non-intact issue pretty thoroughly, with citation and quote from the Model Food Code that first made the distinction.

So, the solution?  Here is a portion of a speech I gave in Washington DC a few years ago:

This leads me to the final, and likely most controversial issue facing this room today: Should primal cuts and boxed beef contaminated with E. coli O157:H7 (or for that matter any pathogen) be considered adulterated?

This is both a complicated and simple issue. One that we all have had first-hand experience from the now infamous Kreifall v. Sizzler and Excel case. In part that case was fought over the desire of the meat industry to hide from liability on behalf of the victims of their contaminated product. But more to the point, an appellate court decided that an intact cut of meat is in fact adulterated, if it is contaminated with E. coli O157:H7 and the seller knew it was to be further processed, in that instance "needle tenderized" in the restaurant.

A little more history. The intact/non-intact distinction was first announced in a July 19, 1999 FSIS policy statement that the meat industry had long pushed as a way of deeming O157 an "adulterant" only in ground beef and other non-intact meat not "further processed" in a federally-inspected facility. This was only a "policy statement" but has since been treated as if it was a rule. This was further elaborated on by FSIS in an October 7, 2002 policy statement issued in response to the ConAgra outbreak and recall. Notably, what was never clarified was whether this "rule" was really meant to apply to meat that was intact when it left a meat plant, or instead, meat that only reached a consumer as intact. As such, the meaning of "further processing" was never explicitly said to apply to processing that happened at retail. Meanwhile, the meat industry has consistently operated under the assumption that O157 can be on anything that is intact when it leaves a meat plant. Indeed, in the last several years, meat processors have started using disclaimer statements to introduce meat that is known to be possibly contaminated with O157 into commerce.

There should either be a zero tolerance policy or not. But as it currently exists, the non-intact meat rule is the exception that swallowed the rule; it is a loophole you could drive a caravan of trucks through. This issue needs to be resolved. Producers, retailers and consumers need to know the score.

So, zero tolerance or not?  Merry Christmas.