Mechanisms For Trace-Back
And Trace-Forward Needed
Jim Prevor’s Perishable Pundit, April 17, 2007
With two recent pieces, Center For Produce Safety Established: An Act Of Faith In The Future and Fresh Express Research Grant Is Allocated To Scientists, we chronicled industry efforts to enhance our understanding of E. coli 0157:H7 and other pathogens that affect fresh produce.
It is hoped that enhanced understanding will lead to more effective methods of preventing foodborne illness outbreaks. Perhaps we will learn how to prevent these pathogens from getting on the crops to begin with or more effective methods to clean and process the products.
This stage of the recovery from the spinach crisis, in which the industry makes a very public offering at the altar of science was absolutely crucial to rebuilding regulatory and consumer confidence. It says to the world in a very public and concrete way that the industry does not find having its products used as a carrier of foodborne illness to be at all acceptable and will work diligently to find a solution.
However, the very nature of the problem — outdoor field-grown crops consumed raw without a kill step for pathogens — means that we have every reason to expect more outbreaks. The research will enhance our knowledge, but it is long-term and it can take a very long time to translate better understanding of something into a useful preventative — just ask the pharmaceutical industry.
Because identified foodborne illness outbreaks on fresh produce as a percentage of servings produced are statistically rare, even effective food safety techniques, such as better Good Agricultural Practices documents, have almost no effect on the reports of foodborne illness. They are almost random events.
In other words, the odds of you winning the lotto are so small that changes in the lotto rules making it ten percent harder to win have virtually no effect on your chances of winning the lotto. Equally the fact that the industry is doing a better job on the farm can provide no assurance against such a rare event.
So , despite our best efforts, the expectation has to be that there will be outbreaks in the future.
This said, and acknowledging that we are working to reduce the likelihood of such outbreaks, the role of industry leadership has to be to develop a response to the highly predictable incidence of future outbreaks that will minimize the damage to the industry.
How do we do this?
The logical way to reduce the impact of any future foodborne illness outbreak is to limit its scope. This is what traceability is all about. It is such an important issue that we have run three Guest Pundits on the subject written by Gary Fleming, Vice President, Industry Technology and Standards at the Produce Marketing Association.
The pieces, entitled Guest Pundit — Traceability And the Need For A Common Language, Guest Pundit — Pairing The Global Language With Technology, and Guest Pundit: Traceability — A Forgotten Piece Of Food Safety, all dealt with important aspects of traceability, and it is impossible to overstate the importance of making sure the executive level in each organization has made clear to everyone that nothing less than world-class traceability standards are acceptable.
Some of Gary’s writing draws on the Traceability Best Practices document that PMA and CPMA jointly developed. It is a terrific effort, and it has a great summary of the ten best practices:
BEST PRACTICE #1 (Page 16)
Add the lot number to fixed-weight consumer packs containing a supplier ID.
BEST PRACTICE #2 (Page 17)
On the retail / foodservice level, mark cases with human readable data including supplier name, product description and lot number.
BEST PRACTICE #3 (Page 18)
On the distribution center level, encode GTIN and lot number in a UCC.EAN-128 barcode.
BEST PRACTICE #4 (Page 18)
On the distribution center level, use human-readable supplier name, product description, and lot number.
BEST PRACTICE #5 (Page 19)
During the slotting process, scan the supplier case and link to the internal pallet number.
BEST PRACTICE #6 (Page 20)
During receiving on the distribution center level, use supplier pallet tags by encoding the company prefix and serial number in a UCC.EAN-128 barcode.
BEST PRACTICE #7 (Page 20)
On the distribution center level, receive the EDI ASN (Advance Ship Notice).
BEST PRACTICE #8 (Page 20)
On the distribution center level, scan supplier pallet data during the receiving process and match to EDI ASN data.
BEST PRACTICE #9 (Page 22)
On the supplier level, use supplier case coding by encoding GTIN and lot number in a UCC.EAN-128 barcode, as well as human-readable supplier name, product description, and lot number.
BEST PRACTICE #10 (Page 22)
On the supplier level, use supplier pallet tags by encoding company prefix and serial number in a UCC.EAN-128 barcode.
The document goes into great detail about what this all means and how to do it.
This is all vital. So much so that as much as we support PMA’s contribution, along with that of Taylor Farms, to get the Center for Produce Safety off the ground, we hope the next big food safety investment PMA makes is in the traceability arena.
One of the lessons of the spinach crisis, though, is that traceability as a concept has to be broadened beyond the issues addressed in the Traceability Best Practices document. The model on which that document is based is really the notion that we find a product at retail and need to trace it back to the producer, the field, etc.
This is vital, of course; only by having this available can the regulators feel comfortable that they can quickly discover the cause of a problem. We want the situation to be that a regulator could pick up a bag and contact the processor, and that the processor — without speaking to another human being — should be able to identify where and when that bag was packed and what product from where was used in it.
But two things happened in the spinach crisis that wouldn’t be solved by this kind of trace-back information, and PMA, with its interest in marketing and expertise in this area, is perfectly situated to help the industry by addressing them.
First, there was enormous confusion at the FDA on who produced what. Dole, which did not process spinach at all, was asked to participate in the first conference call with the FDA because the FDA didn’t know that Natural Selection Foods packed all the Dole brand spinach.
In fact, confusion over all the different brands and who packed what led to a feeling of lack of control over the situation at FDA and may have contributed to a “pox on all their houses” attitude.
In light of this experience, we can expect that the very first question the FDA will ask in a future outbreak is: Who packs what labels?
PMA should agree to act as a simple, confidential, data repository for processors. All will agree to utilize an EDI system to send a notification every time they pack under a given label. We will thus have an industry registry available for regulators of who packed for whom and when. PMA will maintain the data in a secure place providing 24/7 access to appropriate health authorities.
So next time, if the FDA gets word of different brands being implicated, it can instantaneously see how many processors are actually involved.
The other lesson of the spinach crisis is that although the FDA warned consumers not to eat bagged fresh spinach on September 14, 2006, and Natural Selection Foods issued its recall on September 15, 2006, we had recalls announced almost every day through September 22, 2006, by firms that had used some Natural Selection Foods product in spring mix or other products. That must never happen again.
Even more important to regulators than trace-back is trace-forward capabilities, especially if they might be used in processing.
Bagged, labeled product can be recalled, even bulk cartons of product shipped under a trade brand can be recalled, but product, sometimes sold in bulk, sometimes in large industrial or foodservice packages used in further processing, loses its identity when it is used in another company’s spring mix or salad dressing.
This means consumers are in danger. This means regulators will feel a need to act.
It also makes the industry look sleazy. When a processor announces a recall a week after the fact, the inclination is to think that it was laying low and hoping to avoid being noticed. One suspects that the FDA called after a bag was implicated or through working through purchase records and asked about the disposition of a purchase. So the buyer acts stunned and issues a recall.
We need to put people in a situation where they have no choice but to do the right thing. So traceability needs to include a trace-forward mechanism identifying all processing customers.
Basically, there are going to be new things that we won’t anticipate, but let us at least make sure we don’t fall victim to the same problems we experienced on the spinach crisis.
As part of its leadership on traceability issues, PMA should develop a mechanism for these two things:
The FDA and other regulators can never again be confused by who produces what brands.
We can never have recalls dribbling in for days because we don’t know instantly where our product went.
The Center for Produce Safety will, one day, solve the industry’s food safety problems. But it is PMA’s actions on traceability that are likely to determine the impact of the next outbreak.