The industry is racing to develop better, faster, more effective tests to detect microbial pathogens. But even real-time testing may not solve the problem.
by Dan Murphy, contributing editor
It's 6:15 a.m. and Rick Mason (not his real name) has less than 10 minutes to find out if a killer is on the loose.
Rick manages the sanitation program at a sprawling Midwestern packing plant, and he's armed with nothing more than a small plastic device containing chemicals that trigger a distinct color shift in the presence of certain bacteria.
Rick's mission: detect pathogenic E. coli 0157:H7 in beef trimmings headed for the grinder. As the clock ticks toward start-up, he carefully inserts two drops of liquefied meat from a small sample he earlier removed from the combo sitting on the receiving scale.
At 6:18 a.m., he loads the thin cylinder containing the liquid sample onto a chemical reaction canister. As a colored panel slowly illuminates a bright blue, he slides the canister into an automated scanner linked to a nearby computer. A series of numbers flash onto the screen.
At 6:23 a.m.-two minutes ahead of schedule-he logs onto the production program. "Batch #2395 is cleared," he types. "Data uploaded."
Rick smiles. Another day, another disaster avoided.
Future world?
Science fiction? Yes, with some observers no doubt suggesting an emphasis on "fiction." But don't dismiss the scenario as one that will never happen.
In fact, if officials at USDA were granted a magic wish, a scene something like the one just mentioned would be replicated thousands of times a day as meat packers and processors utilize advanced testing technology to detect the harmful microbes lurking in the nation's meat supply.
Under fire from consumer activists and a suddenly sensitized Congress, USDA has dug in deeply on the issue of microbial testing: There will be microbial standards for pathogenic bacteria, industry has been informed; and the industry will be required to test for the presence-or absence-of those pathogens.
In fact, the only remaining question among USDA strategists is how soon can rapid microbial tests be developed so that the industry can be forced to test product?
"Under the proposed Pathogen Reduction strategy, we will probably require rapid [microbial] tests for in-plant use that will compare with standard lab procedures," explains Jerry Carosella, acting chief for the FSIS Research Coordination Branch. Carosella heads up a group screening new technology under the auspices of the agency's newly formed Technology Assessment and Research Division.
"Eventually, these tests would be used under actual production conditions," he says.
Anything but rapid
The development of reliable rapid testing technology has been anything but rapid. It has been more than three years since USDA issued its first "cry for help," asking laboratories and microbiological supply houses to put development of rapid testing methods onto the fast track. The department's October 1992 Federal Register notice called for the development of technology to provide processing plants with "real time capability" for testing meat products for the presence of microbial pathogens.
Along with faster, more reliable results, USDA specified that these rapid tests must require only minimal hands-on technical competence and physical resources. In other words, the proverbial magic bullet: fast, easy and effective.
It sounds promising, but real-time results have been as slow in coming as the conventional microbial analyses the department is seeking to replace.
"Unfortunately, with all the microbial tests currently available, it is still necessary to perform an enrichment step," Carosella notes. "The actual tests themselves take only five to 15 minutes. But it requires from 12 to 24 hours to complete the enrichment process before you can obtain results."
To date, USDA has not received a single protocol that meets its criteria for a rapid test. But it's not for a lack of urgency among the principals.
In August 1994, USDA was quick to tout a new rapid test for measuring total bacterial counts in fresh meat. Saying it would give meat inspectors "micro vision," then-USDA Secretary Mike Espy hailed the five-minute test-which measures only "generic" bacterial levels-as a "significant step by USDA as we enter a new age of meat inspection."
Based on the technology of ATP bioluminescence, which provides an estimate of total microbial levels, the five-minute test was supposed to target unacceptable levels of fecal contamination on beef and pork carcasses.
Subsequent reaction from industry scientists, however, pointed out that while ATP bioluminescence is scientifically sound, the test did nothing to identify the pathogens causing food safety problems, such as listeria, salmonella or E. coli 0157:H7.
Shortly afterward, the authoritative Council for Agricultural Science and Technology (CAST) organized a special task force to investigate the whole question of food-borne pathogens. (Editor's note: The CAST task force included Peggy Foegeding, chair of the North Carolina University Department of Food Science; Michael Doyle, head of the Center for Food Safety at the University of Georgia; Carol Tucker Foreman, former USDA assistant secretary and currently head of the Safe Food Coalition; Joseph Madden, deputy chief of the FDA's Division of Microbiology; James L. Smith, director of USDA's Eastern Regional Research Unit; and 10 other prominent researchers and scientists.)
The task force prioritized 15 goals for industry and government.
No. 4 on the list-after basing policy goals on risk assessment, providing more funding for R&D, and expanding the database on food-related illnesses-called for "development of new, rapid, reliable, sensitive and economical methods to allow fast, accurate detection of hazardous organisms."
At the same time, a Blue Ribbon Task Force organized by the National Live Stock and Meat Board, took on the task of neutralizing the threat of E. coli 0157:H7, particularly in ground beef, the primary source not only of bad meat but bad press.
After an exhaustive review of the entire processing chain, the task force concluded that although development of rapid testing should proceed without haste, the results of such tests should not be considered a definitive solution to the problem.
"Since many pathogen levels on meat are extremely low, very large samples would be required to detect pathogens," the report concluded. "Test-and-hold policies can lead to a false sense of security, and should be avoided."
Industry experts agree.
"Microbial testing should be used to verify that pathogen control programs are being managed," says Dave Theno, vice president of Quality Assurance and Food Safety for San Diego-based Foodmaker Inc., parent company of the Jack in the Box fast-food chain. "You use the information from total plate counts and measurement of indicator organisms, such as generic E. coli, to determine if there has been a breakdown in procedures, or if there has been a lapse in the sanitation program."
No matter how rapid, microbial testing has to be woven into the entire Hazard Analysis and Critical Control Points program, says another industry authority.
"Testing needs to be part and parcel of a HACCP plan that is used to measure the effectiveness of a company's sanitation program," says Dell Allen, vice president of quality and training for Wichita, Kan.-based Excel Corp. "Testing end products for the presence of E. coli 0157:H7 is not what we should be doing."
Hard to spot
E. coli 0157:H7 is an unpredictable organism, Allen explains. It's hard to isolate, and it's present in extremely low levels in product.
"Too many companies are using testing as a 'feel-good' program," he says. "They're doing it so they feel good about their efforts. But it doesn't solve the problem."
Allen says the problem of bacterial pathogens is analogous to being trapped in a thunderstorm. At some point, lightning is going to strike. But when or where?
"What do you do?" he asks. "You put up lightning rods. The lightning rods in a packing plant are the tests for coliform and generic E. coli; the total plate counts taken off the kill floor, at pre-op, and on the final product. Those are your lightning rods."
There's an additional consideration, as well. The value of microbial testing is as closely tied to public perception as it is to data collection.
"Testing doesn't really provide any additional knowledge about the problem [of microbial pathogens]," says Ann Hollingsworth, vice president of food safety for Keystone Foods. "No amount of microbial testing provides proof positive that you don't have a problem.
"Routine sampling allows us to say: 'Our products are as safe as we can make them in today's environment,' " she adds.
Leaders are aggressive
Moreover, Theno points out that the companies doing the best in terms of managing the problem of pathogens are the companies testing aggressively for their presence.
"On a total plate count, the commercial 'standard' for ground beef is about 1 million organisms per gram," he points out. At Foodmaker, we average about 20,000 per gram. For generic E. coli, the standard is about 10 organisms per gram.
"At Foodmaker, we'll average less than five per gram," Theno adds. "Why? Because we have control of our process-and we verify that control through testing."
Theno admits that rapid testing technology-which he insists is coming-will be a significant benefit.
"With a rapid test, you could integrate information about pathogens into the decision-making process," he says. "I'd much rather be doing the testing on the front end of the process, so I could take immediate action [if there was a problem]."
Real-time or not, however, the consensus on rapid testing remains reserved-and its critics skeptical.
"I have no problem with microbial testing," Excel's Allen says. "As long as people realize that as far as solving the problem of pathogens, it's basically a waste of time."
To Test É or Not to Test?
Maybe that is not he question
by Dan Murphy, contributing editor
As food safety concerns continue to push USDA's inspection reform efforts toward mandatory Hazard Analysis and Critical Control Point programs, testing and verification procedures for potential microbial pathogens are emerging as a key component.
Eventually, what everyone-from USDA bureaucrats to academic researchers to lab supply company marketers-would love to develop are microbial ID tests that could be done on-line during production ... with real-time results.
Positive or negative confirmation would be obtained almost instantly, and problems could be dealt with before product enters distribution.
Recent advances in both enrichment techniques and in testing methodology itself have brought microbial testing much closer to that goal. But if and when such technology becomes available to industry, it may still not provide the "ultimate" answer everyone is seeking.
Heart of the problem
The problem is not the testing. It's the sampling, according to one industry expert.
"No matter how fast and how sophisticated testing technology becomes, we will never truly have a 'litmus test' to reliably determine whether or not meat is contaminated," says James L. Marsden, regent's distinguished professor of meat science at Kansas State University. "A processor would need to sample virtually everything to detect the low levels of such pathogens as E. coli 0157:H7-and you can't sample everything."
The alternative is process control. Marsden likes to point out that controlling microorganisms in a HACCP-type system has actually been around for many years.
"It's called pasteurization," he says. "Once it came into place after World War II, the era of outbreaks from raw milk ended. Pasteurization is nothing more than using time and temperature as a critical control point to destroy pathogens in the product."
That's what the meat industry needs to consider, Marsden argues.
"Microbial testing-even real-time testing-is faulty," he says. "You have to control pathogens at the source."
In ground beef, for example, E. coli 0157:H7 needs to be controlled in the raw material, Marsden says. Then, microbial testing becomes a way of validating that the critical control point is working.
"The inspection system is being built on HACCP, we're headed for a train wreck if we're going to rely on testing final product," he warns. "We've got to eliminate the pathogens. Testing is simply not the answer."