Nobody wants a recall.  Nobody wants to make people sick.  Generally, food manufacturers do have safeguards in place–some more than others–to avoid outbreaks of foodborne disease and the recalls that typically follow.  But outbreaks and recalls happen.  Bacteria and viruses are microscopically small and can be hard to detect, and there will always be an inordinate number of companies who simply are not doing everything they can to control pathogenic hazards on food.  It would be inspiring to report to the food consuming public that it has a safety net in the United States Department of Agriculture (USDA) and the Food and Drug Administration (FDA), but this is not the case.  Ultimately, the web cast by these agencies is not, and cannot be, a reliable alternative to good food safety practices.  As a result, the onus for food safety, and avoiding recall or outbreak situations, lies squarely on the shoulders of food manufacturers.  

In the food context, “recall” is a less than apt description for a process that, a lot of the time, is really geared toward just stopping the bleeding.  The USDA and FDA define a Class I recall[1] as “a situation in which there is a reasonable probability that the use of or exposure to a violative product will cause serious adverse health consequences or death.”  The definition implicitly contemplates the ability to physically secure the violative product, or at least ensure that it is not distributed as originally intended.  

But food is different from many products in that it often has an extremely finite shelf life and is thus not reachable by even well-intentioned recalling companies.  Frequently, particularly in the context of lettuce and other fresh vegetables, shelf lives are measured in days, and getting product back from distributors and retailers, much less the ultimate consumer, can be an exercise in futility.  

In its 2004 report to Congress on the success of food recalls generally,[2] the United States Government Accountability Office (GAO) discussed four recalls that it had tracked, finding that only a recall of canned soup occurred “well before” the end of the soup’s shelf-life.  Packaged turkey sandwiches were recalled “near the end” of the products’ shelf-lives, and recalls of ground beef and fresh-cut, bagged lettuce occurred “well after” the recommended shelf-life had expired.  A more recent example is the spinach STEC[3] outbreak in September 2006.  The FDA announced this large outbreak on September 14, 2006, by which point most of the affected product had been consumed, and most outbreak cases were already sick.     

Food manufacturers need not waste time looking to the FDA or USDA for help in most recall situations.  For one thing, the agencies lack virtually any legal teeth in the food context.  In stark contrast to the authority of certain regulatory bodies over other products, Congress has not vested the USDA or FDA with the authority to seize and detain food products, even when the agencies well know that the products at issue are contaminated.  And to complicate matters, it is less than certain whether the USDA and FDA are equipped to effectively assist at all.  They certainly cannot lengthen the shelf lives of perishable products to make effective recalls more realistic.  And, as the GAO notes in its 2004 report:

USDA and FDA do not know how promptly and completely companies are carrying out recalls.  Neither agency’s guidance provides time frames for companies on how quickly to initiate and carry out recalls.  Consequently, companies may have less impetus to notify downstream customers and remove potentially unsafe food from the marketplace.  
***
USDA and FDA do not promptly verify that recalls have reached all segments of the distribution chain, yet monitoring the effectiveness of a company’s recall actions is the agencies’ primary role in a food recall.[4]  

Aside from a sometimes less than efficient effort from the government, and the problems posed by extremely perishable product, yet another obstacle exists for a food manufacturer trying to recall contaminated food:  the dissemination of information to consumers.  Currently, recall listings in the form of basic news releases are distributed to the media and posted on the FDA or USDA Website, as well as on the recalling company’s Website.  Realistically, however, if a person does not learn about a recall in the print or broadcast media, the only audience for online releases is interested people in the food industry or government–not a good cross-section of the food-consuming public.  My experience is that the food-consuming public generally will not become aware of a recall unless it’s been issued as a result of an outbreak and has therefore garnered media attention–in other words, we don’t become aware of a recall unless people get sick.  And if there are outbreak victims, it’s probably too late to do much more than stop the bleeding.  

I’m the first to point out that food manufacturers face a daunting uphill battle in trying to successfully recall a contaminated food product.  The simple fact that these products are typically highly perishable renders recall efforts, a lot of the time, public relations exercises where there is little expectation that much, if any, affected product will be secured.  As a result, food manufacturers cannot afford to think retrospectively about food safety.  Avoiding the brand injury and liability costs associated with sick people means avoiding contamination in the first place.  

1.  A Class II or Class III recall is not appropriate when recalling products due to probable, or even potential, microbial contamination.  The presence of pathogenic microbes on food presents an immediate, undeniably severe health risk to all who consume the contaminated food.  

2.  The GAO’s analysis was based data gathered from the FDA and USDA.

3.  “STEC” stands for Shiga-toxin producing Escherichia coli, which includes the most famous serogroup E. coli O157:H7, as well as many other strains.  The phrase “STEC” is used here because the spinach outbreak involved several different STEC strains, including E. coli O157:H7.

4.  The GAO found that, for 10 USDA monitored recalls in 2003, USDA staff took an average of 38 days to complete verification checks with companies that had received recalled product.