Dennis Wenger, the Pennsylvania dairy farmer I wrote about in my previous post, recalled in his talk Saturday an incident that came up as part of his experience using a private laboratory to do split samples of his farm’s raw milk. I didn’t think it was important enough to include in an already-long report, but now that the peanut scare has taken center stage in the food-borne-illness arena, I realize it has a great deal of relevance.
When the state was re-testing his milk (in concert with a private lab Wenger had engaged) after the original listeria finding, and the results were ready, Wenger recalled receiving a phone call from the private lab. The Pennsylvania Department of Agriculture had called the lab seeking the test results. The lab officials decided they should check with Wenger first as to whether they could release the results to the state. “Absolutely not,” Wenger told the lab. Very understandable—after all, he had commissioned the private lab test and didn’t want state regulators to possibly be influenced in what their results showed.
As it turned out, both tests were negative (though significant differences remained in standard plate counts), and Wenger was cleared to re-open for business.
But I thought about this little exchange as I read some of the reports about the peanut butter outbreak, in which eight people have died and more than 500 been sickened. Peanut Butter Corp. of America had commissioned tests that showed salmonella on a dozen occasions over the previous two years.
Washington lawmakers are beginning to clamor for legislation that would require private labs to report their results directly to the U.S. Food and Drug Administration. You can then bet that state agriculture agencies will do the same. Indeed. Wenger and others reported at the CARE session Saturday hearing of moves afoot for the PDA to seek the power to require private labs to report their results to the PDA first.
There are obviously all kinds of problems in the peanut illnesses, including apparently total incompetence at both the Georgia Department of Agriculture and the FDA. Think about how long a raw dairy would have been able to get away with even a tiny fraction of the problems at Peanut Corp. of America—and peanut butter is a big item with all the little children the FDA and Centers for Disease Control are always trying to get us to weep about in restricting raw milk. More than 100 of the peanut butter victims are children under five.
This threatens to turn into a classic case of using an isolated incident involving massive bungling and dishonest corporate behavior to restrict larger freedoms. The regulators would love nothing more than to gain access, preferably exclusive access, to private lab reports, reducing the ability of farmers to challenge the authorities.
There is definitely a valid ethical question deserving consideration about the responsibility of farmers and food producers to warn consumers and possibly recall product if private lab tests show pathogens. Lykke, in comments following my previous post, provides a number of examples and viable approaches for constructive use of private lab results.
Raw dairy farmers have very little leverage as it is to challenge regulators. Take away the independent influence afforded by the split sample, private lab alternative, and there really isn’t much left.
Does anyone know if PCA had in-house testing? Or was all the testing done by outside private labs? How many different labs were each of these 12 salmonella-positive samples sent to before the sample tested negative? How accurate and reliable is the salmonella testing? If the sample first tests positive, but negative at the next lab, how reasonable is it to think the first testing was in error? Was PCA truly malevolently "lab-shopping" until they got a negative result, or are there enough uncertainties in the testing that it was reasonable to retest? If the peanuts were contaminated with salmonella, how come these 12 samples were able to finally test negative?
Lykke, can you shed any light on the testing process?
And how does salmonella get into peanut products? Does it start in the peanuts, or is it from unsanitary processing, or is it transmitted from the workers?
Those are excellent questions and relate to the need for a food producer to consider carefully their lab testing scheme, and how to respond to positive or uncertain results. As you point out, "screening" is not the same as lab shopping. A producer or their lab shouldn’t be penalized for using a screening test followed by confirmation with a more rigorous testing method (provided the protocol is within standard of practice – keep in mind, many of the test kits on the market have been studied in only a few foods; something like raw milk is a very different matrix compared to ground beef or fruit juice). It will be interesting to see all the details from the peanut investigation and type of tests conducted (if they are released to the public).
1. J Leek Associates Inc (Pert Laboratories)
145 Peanut Dr, Edenton, NC 27932-9604 http://www.manta.com/coms2/dnbcompany_g54zc0
They link to EMSL Analytical, which says this about their Salmonella assay:
http://www.emsl.com/index.cfm?nav=Services&action=show&ServiceId=410
"This method detects the Presence or Absence of the bacterium. A confirmation charge may be incurred in cases of Presumptive Positive screens. 50g minimum sample is recommended. Environmental surface samples can also be analyzed."
EMSL Test Code: M225
Sample retention time – 1 week
2. Deibel Laboratories
http://www.deibellabs.com/services_FoodMicroBiology.html
This lab appears to do more sophisticated testing (serotyping), and was perhaps used for confirmation of Salmonella findings?
Two excerpts from this Associated Press article below:
http://www.google.com/hostednews/ap/article/ALeqM5iH_omeaDBaB1UKxFZ4QGmvPKuuQQD961SUHG0
The FDA reported this week that federal inspectors who visited the plant since the salmonella outbreak found roaches, mold, signs of a leaking roof and numerous other sanitation problems….
The FDA, citing internal company documents, said Peanut Corp. had hired a lab that conducted at least 12 positive tests for salmonella between 2007 and 2008 at its Georgia processing plant. The FDA said the company then used a different lab to retest the products, and those tests came back negative and the product was shipped to customers.
and from here, someone who has worked in industrial microbiology for 16 years explains how lab tests can get manipulated.
http://consumerist.com/5141924/how-unscrupulous-food-manufacturers-manipulate-lab-tests
I hope the upcoming investigations will shed light on testing methodologies and limitations.