Of Tomatoes, Traceability, and Terrorism
It hasn't been a good summer for the food industry in North America.
In April there was a major outbreak of Salmonellosis caused by the bacterium Salmonella Saintpaul. Over 1414 cases were reported which, given the typical ratio of actual to reported cases, suggests 50,000 people may have been affected by this one outbreak. (By comparison it is reckoned that there are normally about 50,000 cases of food poisoning in total in America each year).
Until August it looked like fresh tomatoes were responsible. For a while, your Big Mac came without tomatoes as supermarkets and restaurants withdrew tomatoes from sale. Reports suggested that two farms in Mexico might be to blame.
Finally in August the outbreak was traced to jalapeno and serrano peppers from two different Mexican farms and processed by a common processing plant. Tomatoes were finally in the clear, and tomato eaters could breathe a sigh of relief.
Then in August an outbreak of the deadly disease listeriosis, caused by the bacteria Listeria monocytogenes, started killing people in Canada. Twelve people are known to have died so far in the outbreak. A production facility of Maple Leaf Foods was shut down, and the 220 products produced by the plant recalled. The precise source of the outbreak was never found, although a production line and a production date range was implicated.1
(Maple Leaf foods, incidentally, handled the aftermath of the disaster well - saying and doing the right things. The video statement by their president, Michael McCain, is worth watching on YouTube.)
Of interest in both these cases was the difficulty in tracing a problem with a product (food) back to its origins. Listeriosis, incidentally, is particularly difficult to trace due to the long time between infection and the appearance of symptoms - twenty-one to seventy days. What did you eat three weeks ago and where did you buy it from? What about ten weeks ago? Salmonellosis is a little easier with typically a 12 to 72 hour gap. The problem here is that tomatoes and peppers appear in garnishes and salsas: they don't appear on the restaurant receipt and you may not remember eating them.
I'm not in the food industry... Why is this relevant?
Most of us aren't in the food industry, but there are still plenty of things we can learn and questions we should ask ourselves when we look at these events.
Do we have a plan for communicating with customers and the media if a problem is detected in our product or service?
By all accounts Maple Leaf Foods did an excellent job in communicating with the public in this case. You can watch a heartfelt video by their president and CEO Michael McCain on YouTube, and the press has covered in detail the steps Maple Leaf takes to ensure that the food they supply is generally safe.
If there is a problem with your product, can you (and your customers) easily identify whether they are affected?
The Maple Leaf plant had a provincial inspection label on their food (they were "Establishment 97B"), but the recall affected some 200 products sold under a myriad of premium and discount brands. The recall alone may have cost $20 million, but the damage to customers' faith in the brands and in the distinction between premium and discount meat brands may be irreparable.
If there is a problem in your product, can you identify where it originated, which parts were used, and whether your processing or a particular batch of a supplier's product is to blame?
Could you limit your recall to a limited number of customers or to production in a particular time period, or would all production be suspect?
If one of my suppliers discovered a problem in a batch of goods they supplied to us, could we identify which of our products were affected, and could we confine recalls to only those products?
In these cases the initial recall levels were "all tomatoes in North America" and "everything produced by "Establishment 97B" which hasn't been eaten yet. Contrast this with laptop recalls or automobile recalls, where recalls can be narrowed down to a serial number range or an individual customer.
Can you distinguish your products from those of your competitors?
Can customers distinguish between a tainted tomato and your pristine tomato? If they can't, then your competitor's bad practices can affect sales of your product. It wasn't just the sales of the farms associated with the outbreak that were affected.
It was a case study for terrorists too...
We have just received a lesson in the difficulty the food industry has in tracing contamination by two common pathogenic bacteria. Terrorists, extortionists, Hollywood script writers, and other bad guys read and think about the news too. While politicians may applaud the success of early warning schemes for biological terrorism and SARS-like illnesses in detecting the problems, in reality they did not identify the source of the contamination until weeks after the damage had been done.
If the common bacteria Salmonella Saintpaul and Listeria monocytogenes can affect so many people when introduced into the food supply in a single plant or farm, who needs weaponized anthrax?
 I actually find this lack of result rather reassuring: the food production line was tested regularly for Listeria monocytogenes (and the batch would have been rejected if the bacteria was found). If the equipment hadn't been properly cleaned between the detection of the outbreak and the time when it was traced to the plant, it would be really worrying.
 The quality of this response has been somewhat dented by press reports on September 10 which seemed to be trying to shift the blame for the problem onto the manufacture of Maple Leaf's equipment. If Maple Leaf really felt they couldn't keep the equipment clean, they shouldn't have been using it in a production environment, should they?
 No I'm not giving anybody ideas. It's already been tried at least once. See http://en.wikipedia.org/wiki/1984_Rajneeshee_bioterror_attack