How safe is your food?

Twenty dead. A company under siege. The unsettling truth about the Maple Leaf outbreak.

The Clark family reunion was scheduled for the third weekend of July. Uncles and cousins and in-laws, from as far away as Utah and Florida, had marked the date on their calendars months before. Plane tickets were booked, motel rooms were reserved, and the venue was set: Madoc, Ont., (population 2,044).

Frances Clark was at the centre of the plans. The matriarch of the clan baked some of her famous raspberry pies, froze a few pans of homemade lasagna, and spent weeks making sure the house—built by her late husband’s grandfather in 1915—looked just right for the big party. “At breakfast one morning she said: ‘You know, I’m going to go into the other room and work on that window,’ ” recalls her daughter Karen. “There was some caulking on the frame, and she wanted to dig it out so the window would go up and down for the reunion. You wouldn’t believe that woman with a hammer. Mike Holmes had nothing on her.” Frances Clark was 89 years old.

Later that morning, as she walked to her bedroom for a brief nap, Frances tripped and dislocated her left shoulder. When Karen came home from work, she found her mom lying on the floor, flat on her back. “I said: ‘Mother, what are you doing down there?’ She said: ‘Well, I thought I’d lie here and count the ceiling tiles.’ She was hilarious. Count the ceiling tiles? How many women, at 89, would say that?”

Frances spent the next four weeks at Belleville General Hospital. Although she missed the reunion, her out-of-town relatives made sure to stop by for a visit. After they left, Frances made a difficult decision: maybe it’s time I move into a nursing home. “She was transferred to Stirling Manor on Aug. 7,” her daughter says. “She was making new friends, talking everyone’s ear off. This was not some old lady who had lost her marbles and was tied up somewhere, drooling. Far from it.”

Two weeks later, Clark was back in a hospital bed, barely conscious and gasping for air. Her temperature skyrocketed, her eyes glazed over, and the antibiotics proved no match for the bacteria coursing through her body. “A couple of times she tried to say a word or two, but it was incomprehensible,” says her son Tim. On Aug. 25, at 5:15 in the morning, Frances passed away. The official cause of death was listeriosis, a flu-like infection that attacks the central nervous system.

Today, Clark’s family knows this much for sure: during her initial stay at the Belleville hospital, she was served Maple Leaf “Sure Slice” ham (lot #21440) three separate times. After moving to the nursing home, she ate even more Maple Leaf lunch meat, including turkey and roast beef. “A double dose,” Karen says. Those products, of course, were later recalled after lab tests revealed dangerously high levels of Listeria monocytogenes, the powerful pathogen that causes listeriosis. But for Clark—and 19 others killed by Maple Leaf meat—the recall came too late. She was in the last hours of her life, surrounded by children and grandchildren, when Michael McCain, the company president, went on television to apologize to the victims.

“To Canadians who are ill and to the families who have lost loved ones, I offer my deepest sympathies,” said a sombre McCain. “Words cannot begin to express our sadness for your pain. Our best efforts failed, and we are deeply sorry.”

Class-action lawsuits have already been launched against Maple Leaf Foods Inc., demanding hundreds of millions of dollars for the company’s alleged “negligence.” Tim and Karen Clark are among the plaintiffs. “I’m sure Michael feels really bad about it, and I think his apology was very sincere,” says Tim, who watched McCain’s two paid commercials. “How this happened, who knows? But prevention is the key now. We want to make sure this doesn’t happen again.”

So do the biologists, bureaucrats and every other food safety expert who has spent the past 30 years in a scientific wrestling match with this nasty little bug. But unfortunately, after countless case studies and many high-tech breakthroughs, the experts have reached one uncomfortable conclusion: listeria is unstoppable. As much as the Clarks—and all Canadians—would like a foolproof fix, there isn’t one. No matter what companies do, no matter how many safeguards they adopt, there will always be that chance, however slim, that your ham sandwich or your bagged lettuce or your brick of cheese contains a deadly helping of listeria. “You’re going to see another outbreak just like this one,” says microbiologist F. Ann Draughon, co-director of the University of Tennessee’s Food Safety Center of Excellence. “Listeriosis cases are decreasing, but there will be more outbreaks. It is inevitable.”

That certainly doesn’t exonerate Maple Leaf Foods. Their products killed 20 people (maybe more), sickened dozens of others, and made everyone else think twice about eating those cold cuts sitting in the fridge. But was the company negligent? That answer may be impossible to pin down. When it comes to listeria control, the definition of “best efforts” is more a matter of scientific interpretation than undisputed fact.

No one—not Maple Leaf, and not the federal government—can test every ounce of lunch meat before it leaves the plant. There would be nothing left to eat. The alternative, then, is an imperfect compromise that attempts to decrease the danger as much as possible while still ensuring that our supermarket shelves are sufficiently stocked. Keeping the food chain free of listeria is a complicated, ever-evolving recipe of common sense, third-party oversight, targeted testing—and a dash of calculated risk. No two companies, or countries, follow the same playbook, and the debate over how best to battle this bacteria is still very much unsettled.

Can more be done to lower the odds of another Frances Clark dying? Absolutely. Are the experts unanimous on the best way to do that? Absolutely not.

Listeria is a microscopic, rod-shaped bacteria, and although it isn’t visible to the naked eye, it is all around us. Soil, water, raw meat, your shoes. For the most part, the organism is harmless, except for that one particular species: Listeria monocytogenes. Ironically enough, L. mono was first identified as a food-borne pathogen in Canada, when a bad batch of coleslaw killed 17 people in 1981.

In the three decades since, much has been gleaned about the bug. It is strong, stubborn and sturdy enough to grow in cold temperatures. The bright side? Like salmonella or E. coli, it can be killed with heat. Cook your prime rib properly, and you’ll be just fine.

However, ready-to-eat foods—the deli meats, fresh produce and boxed cereals that go straight from the grocery cart to your mouth—present a whole different challenge. They are designed to be eaten with no fuss and no preparation, so if they leave the factory laced with L. mono, those cells are going to end up in your stomach (unless you’re among the small minority who fries his bologna).

Most of the time, though, a single serving won’t require an ambulance. In fact, researchers believe the average person ingests L. mono every three or four days—that’s 100 times a year—without ever realizing it. The reasons vary. For one, most people are naturally healthy enough to escape its wrath. L. mono also comes in dozens of different strains, and some are much more virulent than others. And sickness is a matter of dosage, not existence; most listeriosis victims ate foods that contain extremely high levels—well over 1,000 “colony forming units” (cfu) per gram.

“There is no such thing as a 100 per cent safe food product,” says microbiologist Elliot T. Ryser, a food-borne diseases expert at Michigan State University. “But people are consuming listeria on a regular basis, and we aren’t dropping over like flies.” Indeed. South of the border, the listeriosis rate is 2.7 cases per million people. In 2003 (the latest figures available) the number of confirmed cases in Canada was 59. Salmonella, on the other hand, infects up to 12,000 Canadians a year.

Still, those lopsided stats tell another, much more dire, story. Yes, listeriosis is responsible for only 0.02 per cent of all food-borne illnesses, but the fatality rate is extremely high, accounting for 28 per cent of all deaths from food-borne illness. This bitsy bacterium also attacks the most vulnerable: children, the elderly, and those with weak immune systems. Pregnant women are 20 times more likely to contract listeriosis, with side effects ranging from miscarriage to stillbirth. “My wife is seven months pregnant and she doesn’t touch deli meats,” says food scientist Doug Powell, director of the International Food Safety Network at Kansas State University. “But not everyone has a Ph.D. in food science.”

Adding to the complexity of listeria control is the fact that most contaminated products don’t start out that way. The cow or chicken that becomes your lunch meat, for example, is cleaned and cooked after slaughter, destroying most bacteria. Yet in the final stages of production, when the meat is sliced and packaged, it has the potential to become re-contaminated by listeria cells lurking in the factory. For ready-to-eat food companies, it is a never-ending game of seek and destroy. “This is an insidious organism that is very hardy and survives very nicely,” says John Cerveny, now retired as manager of microbiology at Oscar Mayer Foods, the U.S. firm famous for its hot dogs. “We’re doing everything we can to minimize the problem, but we’ll never eliminate it. Not in my lifetime.”

For Cerveny and his fellow scientists, the most frustrating fact about listeria is that it multiplies over time, even in a refrigerator. If sliced turkey contains just a trace, that bacteria will inevitably reproduce, especially in an ideal breeding ground like moist, uncured cold cuts. Leave that same meat on a counter in room temperature, and the growth can be even more rapid—doubling in size every 15 minutes. Think about that. A one-gram piece that begins with 10 listeria cells (an amount widely believed to be harmless) has the potential to reach 20 in just 15 minutes. In less than three hours, it can grow to 20,000—the same fatal dose discovered in some Maple Leaf products tested after the outbreak.

It is no surprise, then, that Health Canada and the Canadian Food Inspection Agency (CFIA) focus their listeria-fighting efforts on the highest-risk products. According to the current guidelines, foods that do not support the growth of listeria and/or have a shelf life of less than 10 days (cereal, for example, or bagged salads) can contain up to 100 cfu per gram. However, those products that do support listeria growth—lunch meat, hot dogs, soft cheeses and ready-made sandwiches—must be completely free of the pathogen. Even a smidgen will trigger a recall.

Those guidelines, however, are just that: guidelines. Despite its tough, zero-tolerance stance, even Health Canada concedes that up to 10 per cent of all ready-to-eat foods for sale right now contain some levels of L. mono. Reaching zero is simply not possible. It would be like trying to nab every last motorist who drives one kilometre over the speed limit.

An investigation ordered by Stephen Harper is in the works, but the prime suspect in the Maple Leaf outbreak has already been identified: two industrial-sized slicers at the company’s Bartor Road factory in Toronto. Experts who inspected the plant after it shut down believe the bacteria was hiding “deep inside” the machines, thriving in hard-to-reach crevasses that weren’t cleaned during routine sanitation. The company suspects the cells originated from one of four possible sources, including a drain or an elevator station, before slinking their way into the slicer.

The CFIA has since ordered every ready-to-eat meat company to conduct “a systematic and thorough cleaning” of all similar equipment. However, the much larger question still looms: how could so many cold cuts be so heavily contaminated and still get out the door without anybody noticing?

In the old days, inspectors focused on finding the bacteria where it mattered most: in the final product. Morsels from high-risk lots were tested for L. monocytogenes, and if found, the entire load was scrapped. But that method is far from perfect. Again, it isn’t feasible to test every slice. The only alternative—random sampling—isn’t fail-safe either, because listeria lands sporadically, a few cells here, a few cells there. If you have a skid full of corned beef, for example, and only one per cent contains L. mono, you would need to test 299 samples to ensure a 95 per cent probability of discovering it—and even then, there is still that five per cent chance you won’t find it.

Guided by those unsettling statistics, the modern-day protocol has evolved into an industry-wide system of Good Manufacturing Practices (GMPs) verified by regular environmental sampling. Simply put, companies now swab their machines and their factories for signs of listeria, with much less emphasis on the food itself. “The focus used to be: ‘Let’s test the dickens out of the finished product,’ ” says Jeffrey Kornacki, a Wisconsin microbiologist who has advised dozens of food companies. “But people have now realized that almost all the contamination of ready-to-eat foods is coming from the environment of the facility, so we have put an extreme emphasis on testing that.”

The devil, as always, is in the details.

Although L. monocytogenes is the enemy, food companies actually conduct a generic test for all listeria species, not the dangerous one in particular. The scientists support this approach (it is quicker and cheaper to look for all types), but the technique also provides industry with a convenient escape clause that the government is willing to live with. If companies swab their equipment specifically for L. mono, a positive hit will almost always prompt a costly, embarrassing recall of any meat that touched the line. By looking for listeria—which may or not be L. mono—companies avoid the smoking gun. “They will look at that listeria result and say: ‘Okay, we don’t have evidence that our product might be contaminated with L. monocytogenes, but we do have evidence that we need to improve our sanitation,’ ” Kornacki explains. “So it kind of gets them out of a sticky situation.”

Maple Leaf subscribed to that philosophy. Before the outbreak, staff swabbed the Bartor Road plant for listeria 3,000 times a year. If a positive was discovered on a food-contact surface, such as a conveyor belt, employees did what countless other companies do—and what Health Canada advises. They scrubbed the line and tested it again. If the next three samples were negative, the problem was considered solved—even though the meat, long gone from the factory, was never examined.

As shocking as that may sound, the science has long supported that strategy. Study after study has concluded that a lone positive on a food-contact surface is insufficient proof of a widespread problem, and that scouring the machine—not checking the meat—is the prudent response. Even Health Canada recommends end-product testing only after a second straight environmental positive, which typically suggests a much bigger problem.

But isn’t that practice risky, regardless of what the scientific literature says?

“Inevitably, some of the product may become contaminated,” Kornacki says. “But is it going to be contaminated at a level that human beings are going to get sick? To be honest with you, I’m always shocked when I see the fatalities because I deal with so many companies that struggle with this organism, and people aren’t dying.” Consider, for a moment, all the factors that must conspire against a company like Maple Leaf: the listeria species must be L. monocytogenes, the strain must be virulent, it must grow to massive levels, and it must be eaten by a vulnerable consumer. Someone like Frances Clark.

Maple Leaf refuses to reveal the results of its pre-outbreak sampling, saying only that the data is available to the CFIA. However, the company has confirmed that “numerous” positives were collected from the infamous slicing machines in the months before the recall. After cleanup, though, there was never a subsequent hit, which means the meat was declared edible and loaded onto a truck.

We now know, of course, that those slicers were covered with listeria, despite what the company tests were showing. The key question, then—and the one that continues to divide the scientists—is whether it’s time to rethink the standard reaction to a positive environmental sample. Should the first sign of listeria trigger a closer look at the meat? Product testing may not be infallible, but is it not, at the very least, extra insurance?

“It’s a wise decision to do that because if not, you are taking a slight risk that the meat could have levels of L. mono that could cause someone to be ill,” says Draughon, the Tennessee professor. “If you find that positive listeria you don’t know if it’s L. monocytogenes. But you don’t want to ship product before you know it’s free of it.”

Other experts agree. “The concept of environmental sampling is not sufficient, by itself, to address this food-borne disease problem,” says Michael Doyle, a leading listeria researcher and food microbiology professor at the University of Georgia. “If they’re relying only on environmental testing, and then we have this big outbreak, that’s the proof in the pudding that this does not work. There has to be some robust end-product testing program in place to verify that contaminated product is not being released.”

Neither Canada nor the U.S. has a specific law dictating how to react to a first positive, and the ready-to-eat food industry is anxious to keep it that way. Randy Huffman, another well-known listeria expert, is president of the American Meat Institute Foundation, the research arm of the industry’s largest trade organization (Maple Leaf Foods is a member). “In isolation, a single positive on a food-contact surface is simply not enough information to make a decision about whether or not you move to full product testing,” he says. “Is this positive within normal, expected, sporadic positives that you’re going to find from time to time, or does it indicate a trend where we might expect that the sanitary conditions of the facility are out of control? Those are the decisions that have to be made.”

The goal, Huffman says, is to encourage companies to search for positives and attack the source. A regulatory system that is too proscriptive may actually promote the opposite. A company, after all, can easily create a sampling regime that never finds anything—tricking themselves, and the feds, into a false sense of security. “If you trigger product testing on every food-contact positive, you might as well hand over the keys to the regulators,” Huffman says. “It just doesn’t work, and frankly, it’s not good for public health.”

The regulators don’t want the keys, anyway. Three years ago, the CFIA stopped conducting its own environmental testing and passed that responsibility to industry. Today, the agency spends more time auditing company paperwork than walking the floor. After the outbreak, the CFIA did announce a temporary rule that compels all companies, Maple Leaf included, to alert local inspectors about a first positive environmental sample. But the pre-outbreak action plan—clean the equipment and retest it—hasn’t changed.

In a written statement, Health Canada says it “applies the best science available in its guidelines, which rely on evidence-based decisions. A correlation between environmental contamination and final product contamination has not been demonstrated convincingly to date.” Paul Mayers, a CFIA spokesman, echoed that remark, but he said the ongoing investigation will examine all possible remedies. “Anything we’ve learned that we can use in terms of enhancing the overall food safety in the system, we’re going to act on.”

Michael H. McCain has not spent the past two months hiding in his office. For a CEO in full damage-control mode, he has been surprisingly honest, accessible and genuinely remorseful. His message is clear—“The buck stops here”—and in between press conferences and television commercials, he has done his very best to learn everything there is to know about Listeria monocytogenes. When McCain recently sat down for a series of interviews with Maclean’s, he did not bring along a company scientist or a team of lawyers. “It’s been a tragedy for everyone in the Maple Leaf organization and, certainly most importantly, a tragedy for the people affected by this,” he says, sitting in the main boardroom of his Toronto headquarters. “We had a breach, and we took accountability.”

It is easy to respect McCain, and not just because he did the moral thing and apologized to the victims. What makes him most credible is that he isn’t afraid to engage in a detailed discussion about the complexities of listeria control at a time when consumer confidence is shattered and lawsuits are pending. And as much as he wants to reassure rattled shoppers, he is realistic. No magic answer. No lofty promises. “In microbiological hazards, it’s about risk reduction,” he says. “It’s not about risk elimination. You can have the very best systems in the world, and reduce the risk to its absolute lowest possible level, and still have outcomes like this.”

That may sound like spin-doctoring. Listeria is everywhere. Oh well. But his point is valid, and any debate about the way forward must be framed by some troubling truths. Listeria is in every ready-to-eat food factory, and always will be. You’ve eaten it before and you will again. And—as heartless as this may sound—the risk of death truly is astronomically low. You are much more likely to choke on your lunch than catch listeriosis from it.

“Canadians need to understand that the food supply is safer than ever, that it’s continuously improving, that the risks are infinitesimally low—but they’re not zero,” McCain says. “I think it’s a travesty when I see this in the press get characterized as a broken system. This is about taking a system that is very strong and making it better.”

Last year, Maple Leaf sold $5.2 billion worth of groceries. It is one of the largest food processors in Canada, if not the world, and when it comes to deli meat, half of the cold cuts in the country are stamped with the Maple Leaf logo. McCain is adamant that his company always far exceeded the safety standards set by both the government and any of his competitors. “We had a program that was in the penthouse, relative to our understanding and belief at the time,” he says. “But looking at yesterday’s behaviour, you can always find things. ‘Gee, I might have done this different, or I might have done that different.’ ”

Today, much has already changed at Maple Leaf. Staff is studying new ways to clean those hard-to-reach zones deep within the slicers (some companies wheel the machines into the smokehouse and “cook” the entire thing). It also plans to expand environmental testing protocols, which includes doubling the amount of swabs (to 6,000) collected at the Bartor Road plant. At the same time, a new database will be installed to analyze the results, pinpointing dangerous trends and potential harborage points much quicker than before. Finished product testing also will be expanded, although McCain is quick to point out what the scientists have said for years: product testing is nowhere close to a guarantee.

But what if a food-contact surface tests positive for listeria species? Will the company go beyond Health Canada’s recommendation and sample the product right away?


Maple Leaf meat will now be quarantined if a food-contact shows listeria. If a second environmental sample comes back positive after the machine is washed, then the product will be tested. The difference? Under the old system, the meat was shipped before the environmental samples were collected. Today, it is still in the plant—and out of the reach of customers—if a line tests positive.

However, the basic premise hasn’t changed. Quarantined or not, the meat still won’t be examined after the first positive on a line. If the machine is scrubbed and all seems well, that package will be sold without ever being tested for L. monocytogenes, just like before.

When asked if the new protocol would have helped prevent the outbreak, McCain answers this way: “I genuinely don’t know. That’s an unanswerable question. This is about risk reduction, not about risk elimination, so because you’ve reduced the risk it doesn’t mean you’ve eliminated it.

“Am I comfortable with it?” he continues. “Here’s what I’m comfortable with. Our program is among the most conservative we’re aware of today in practice, anywhere in North America, if not the world. Secondly, I’m comfortable with the fact that if the industry, through review and scientific discourse, comes to a conclusion that there is one more step we should take, we’ll be the first one there.”

In the U.S., some companies are already taking an extra step (and in some cases, two) with a little push from Washington.

Although the experts agree that listeria can never be eradicated, food processors are now aiming for the next best thing: stunting its growth. Many have started adding so-called “inhibitors”—an antimicrobial mix of lactate and sodium diacetate—to their lunch meat recipes. The blend won’t destroy listeria, but it ensures that small doses remain small (i.e., harmless). Some processors also employ a final “kill step,” such as pasteurization, which will heat and destroy most bacteria that happens to sneak into the final package.

The United States Department of Agriculture is so enchanted by the latest safety procedures that it designed its current enforcement strategy around them. Any company that employs both (growth inhibitors and a post-packaging kill step) will be bothered the least by federal inspectors. Use neither and you’ll be at the top of the scrutiny chart. “They are not silver bullets, but both will greatly decrease the likelihood of an outbreak,” Ryser says. “Look at the track record in the U.S. now. After we moved to post-package pasteurization and antimicrobials, we have been basically free of outbreaks since 2002—knock on wood. That doesn’t mean we won’t have an outbreak tomorrow, but for the last six years we’ve been in good shape.”

In Canada, neither growth inhibitors nor in-package pasteurization has been widely adopted by the ready-to-eat food industry. It wasn’t until after the Maple Leaf outbreak that Health Canada even approved the use of sodium diacetate. “This was sort of an accident waiting to happen,” Ryser says.

McCain says he will consider using sodium diacetate now that the feds have approved the preservative. But he also says something that most CEOs, especially one reeling from a public relations nightmare, may not have the guts to say out loud. “We have to be sensitive to customers’ interests here,” he says. “There are many customers out there who will say: ‘You know what? I’m not at risk. Don’t laden my products with chemicals to reduce a risk that I don’t have.’ This is a risk that is four in one million. If you stack up all the risks in life that you and I take every day, this would be well down the list. I’m not trying to be insensitive to the four in one million, and the effort that we need to take to protect against that, but we have to consider the 999,996 as well—and how we respond to what is an infinitesimally low, low, low risk.”

As a society, we must also consider whose job it is to manage that risk. Is it Maple Leaf’s responsibility to warn pregnant women to avoid lunch meat? Should cold cuts be banned from retirement home dining rooms? And what about butchers? Companies can do their darndest to keep listeria out of their meat, but those slicers at the deli can easily recontaminate what was once a clean product. One recent study found that lunch meat purchased at your local deli counter is seven times more likely to contain listeria than a sealed package straight from the factory.

And then there is this question, which most people rarely consider: if you leave your ham sandwich on the kitchen table all afternoon, and the listeria cells double every 15 minutes, is that your fault or Michael McCain’s?

“Educating consumers about these products is the most important thing,” Draughon says. “As far as I’m concerned, these products should not be served in a nursing home, period. They are not appropriate for anybody who has impaired health, and I will argue with anyone all day long about that. I would not give it to a pregnant woman, I would not give it to small children, and I would not give it to the elderly. It’s just too high risk a food—and I love my luncheon meats.”

McCain says warning labels are “a debate worth having,” but for now, his company plans to distribute information brochures to all nursing homes and hospitals that serve Maple Leaf goods. “The most important asset we have is the trust of our consumers, and they put their trust in us to perform,” he says. “It’s our obligation to respect that trust.”

For Frances Clark’s daughter, that trust has been forever shaken, regardless of how many brochures the company prints. Karen Clark says she will never again buy Maple Leaf products. In fact, since the day her mom died, she can’t look at a grocery store shelf without wondering the worst. “You never questioned the safety of your food,” she says. “You just assumed it was good.”

Looking for more?

Get the Best of Maclean's sent straight to your inbox. Sign up for news, commentary and analysis.