The Decibel - How Canada’s food inspectors missed a deadly listeria outbreak

Episode Date: December 18, 2024

In July, there was a recall on two specific brands of plant-based milks, Silk and Great Value, after a listeria outbreak that led to at least 20 illnesses and three deaths. Public health officials determined the same strain of listeria had been making people sick for almost a year. When Globe reporters began looking into what happened, they found a surprising fact: the facility that the bacteria was traced to had not been inspected for listeria in years. So how did this happen?They also learned that in 2019 the Canadian Food Inspection Agency introduced a new system that relies on an algorithm to prioritize sites for inspectors to visit. Investigative reporters Grant Robertson and Kathryn Blaze Baum talk about why this new system of tracking was created, and what went wrong.Questions? Comments? Ideas? E-mail us at [e-mail]

Muriel Sampson was a 76-year-old grandmother living in Toronto. She was someone who had been diagnosed with cancer recently and had adopted a healthy lifestyle and was really, you know, changing her eating habits and everything to live a healthier life. And she had really taken a turn for the better. That's Globe investigative reporter Grant Robertson. You know, she had more energy than ever. Her family was amazed at how well she was doing. And in June of this year, she suddenly took a turn for the worse. And within a week, her health deteriorated. She was in the hospital. Suddenly she's in intensive care. And days later, she's on life support and passes away. The family was shocked to find out that when Muriel was in hospital, the doctors figured out

she had listeriosis and that she had gotten sick from something that she ate. Muriel is one of three people who died in a Listeria outbreak this summer. When Grant and fellow investigative reporter Catherine Blaisbaum started looking into the deadly outbreak, they found serious gaps in oversight from the country's food inspection agency. Those affected by the outbreak are left with questions about a system they say failed Canadians. Her family, especially her son, Cale, looking at what happened to Muriel, really think the system in Canada failed them in terms of, you know, our food safety system that's supposed

to look out for, you know, Canadians and the products we consume, everything in our fridge and in our pantries. They feel that system really let them down. So today, Grant and Catherine are here to explain how the system that oversees what we eat missed a deadly outbreak. I'm Mainika Raman-Wilms, and this is The Decibel from The Globe and Mail. Grant, Catherine, thank you so much for being here. Thank you. Thanks for having us. So Catherine, let's start here. Can you tell us about this Listeria outbreak and how serious was it? Serious. We can confidently say that this is one of the country's most serious and

prolonged Listeria outbreaks on record involving Canadian products. So what that looks like in terms of official numbers is 20 people who were sickened and three people who died. We know that there were 15 hospitalizations. We also know that the true numbers are higher. Not all of those who became ill were actually counted. We also know that the official numbers don't include miscarried pregnancies. So it's safe to say that many more people were affected. And in terms of what products were involved in the outbreak, the problem was linked by federal public health authorities to a brand of plant-based milk sold under the name Silk.

And that prompted a recall on July 8th of almond milk, coconut milk, and other varieties that were produced on the same production line at a facility in Pickering and sold under the Silk and Great Value brands. Common products, things that you see all the time, things that are often found in people's fridges. Yeah. So the issue was a listeria outbreak then in these plant-based milk products. Catherine, can you just go through like what exactly is listeria and how does it spread? So I knew that listeria was serious. I knew from my own pregnancies that you should really avoid getting it. But until this reporting, I didn't fully appreciate just how serious listeria is. So listeria monocytogenes is the bacteria that cause listeriosis in humans.

It's naturally occurring and it tends to thrive in cool and damp environments. But what's important here is that it can't survive pasteurization, which is known in sort of the food safety industry as a kill step. And so the thing about listeria is that once it sets up shop in a production facility, it can be very hard to get rid of it. And once it gets into a product, it can grow extremely fast and rise to the level of a dangerous dose. And if it is consumed by a human, it can have pretty grave consequences. Just to be clear, there is a spectrum of illness here. And so, yes, there are instances where people may be able to manage the illness at home, but it is a dangerous bacteria, specifically for the elderly, for children and those who are pregnant or have immune deficiencies. And we know that it can cause serious blood infections, meningitis, miscarriages, and death. If caught in time, it can be treated

with antibiotics typically administered through IV. Okay. So it sounds like there were a real range of things, though, if it is left untreated, it can be very serious for a lot of people. That's correct. And I think the range also is part of why so many cases end up underreported and underrecorded because of that range. So if you're able to manage it at home, you're not presenting at a hospital, you're not being tested for it, and public health authorities aren't alerted to the cases. Okay. Grant, I want to turn to you here. We talked a little bit about what these tainted products were. What do we know, though, about where they originated from? So immediately after people start showing up in

emergency rooms in various places in Canada, public health officials start trying to cross reference everything that they've eaten, which is sort of how you do it, and trying to figure out if there's a product that this is linked to. And eventually they link it to the plant-based milks through people's grocery receipts, through product people had in their fridge still that they were able to test, things like that. And they determined that it was a product made at a plant in Pickering that was operated by a company named Jiriki. And Jiriki was contracted to make the plant-based milk for Danone Canada, which is a subsidiary of the French dairy giant Danone, which makes the silk brand. So they trace it back to that one plant and a production line at that plant. They have not been able to determine the exact source of where the Listeria came from. What we know is that, like regular milk,

plant-based milks are subject to pasteurization. So that would take out any pathogens after it's made, before packaging. So most likely, the contamination in this case would have occurred between the pasteurization step and before the cartons were sealed up. But they haven't determined the actual source of it yet. What's interesting is, and this is the real moment that this went from a typical outbreak to a much bigger problem, is public health officials did genome sequencing on the listeria

strain. And what they found was that people who had turned up with illnesses in August 2023 were also the result of this strain or a similar strain that were showing up in June 2024. So that same strain had been infecting people for almost an entire year then. Correct. So what that told them was the listeria problem had been in the plant for 11 months and had gone unnoticed, undetected, you know, what was going on in that 11-month period is unclear. So can I ask, though, like, if this outbreak was going on for 11 months, like, were people getting sick from this the entire time? Well, it's difficult because, as Catherine said, this is an underreported type of condition.

So people might get sick, but they may not think it's listeriosis. They might think they just have a really bad flu or a gastro problem. They may not be going to hospital. So what they know is that there were cases in August 2023 and September 2023 that were the same strain as the ones that started emerging in much greater numbers in June 2024. And so we started asking questions about that. And I think over the course of this investigation, we asked more than 90 questions of the CFIA about this outbreak. And as we started asking more questions, the answer started to get more and more concerning. Catherine, maybe this is for you, like some of the individuals that you talked to over the course

of this investigation. Do we know where people impacted by this outbreak over the course of that year? So one of the most challenging and emotionally difficult interviews I've actually done in my career was for this story. I interviewed a woman named Sonia Jabeen, and she's a 32-year-old PhD student at the University of Toronto. And she became violently ill in December of last year, so just about a year ago this month. And at the time, she was 18 weeks pregnant with her first child due to be a baby boy. And she had been, you know, nauseous during her first trimester and tired and, you know, having all of the weird and wonderful pregnancy symptoms and sort of chalked

it up to that, but said she forced down smoothies made with silk oat milk because she was determined to make sure the baby was getting enough nutrients. Her symptoms escalated and really took a turn. She started getting intense headaches and vomiting, and she ended up in the hospital. And initially, at the ER, she was told, you know, probably a pregnancy complication, baby looked fine on ultrasound. But the next day, she was hit with intense stomach pains. And her worst nightmare came true that day. Alone in her bathroom, she went into preterm labor and delivered the baby. And she remembers and described to me quite vividly that experience and just how tiny and fully formed her baby boy's lifeless body was at the time. And so as it turned out, she did herself test positive for listeriosis.

So did the placenta, the amniotic fluid and the fetus. And it wasn't until July of this year, more than seven months later, that she became aware of the recall of the milk she was drinking. And so she was contacted by public health officials after she was diagnosed, but she doesn't know if she was included in the outbreak statistics. And one thing that I think will stay with me as the reporter on this is the reality that her husband, Hassan, dug the grave for the child that they lost. And they got a death certificate from their mosque, which shows that they named him Walid, which is Arabic for newborn. Wow. So that's, I mean, a really heartbreaking example of how this can affect someone so seriously. I know you also spoke to another man named Tyler in his 20s. Can you tell me about his story?

Yeah, this story was quite significant because, you know, we often think of this as, you know, the risk is for, you know, the elderly, the immunocompromised, and as Catherine mentioned, for pregnant mothers. But we spoke with Tyler McMahon, a 27-year-old healthy guy, you know, who, you know, worked out, you know, several times a week. He says he was putting the almond milk in his protein shakes, and he got seriously ill within a matter of days and ended up in hospital. It escalated from what he thought were flu-like symptoms to, you know, basically being put on life support and was on intense antibiotic treatment and antifungals and everything. They couldn't figure out what it was and was basically in hospital

getting these treatments for two months. And when he finally came to, he had lost the use of his arms and his legs, and he was essentially had a form of paralysis. His arms came back, but he's really spent the past year in physiotherapy, learning how to walk again and, you know, trying to get his life back. Now, Tyler was never officially diagnosed with listeriosis. He was diagnosed with bacterial meningitis, which is caused by listeria. And he suspects that came from the tainted product. Okay, so I think the big question here is then, if this outbreak was going on for nearly a year and people were getting sick, why wasn't this listeria outbreak caught earlier?

Well, that's a major question in all of this. And so we started asking around, started developing sources within the CFIA and with the Public Health Agency of Canada. The CFIA is Canadian Food Inspection Agency, yeah. That's right. And a picture started to form that this plant had not been inspected for Listeria or Listeria protocols. And I remember the moment where I reached one person who works with inspectors. And I said, look, why was this plant never inspected in that time? Plausibly, the Listeria problem could have been detected. And the answer, and I'll still remember the moment to this day, and for a long time, he said, we didn't check that plant because the algorithm told us not to go in. The algorithm? That's what I said. The algorithm? We'll be right back.

So why is the CFIA relying on a system, I guess, that relies on an algorithm? So then the investigation took a turn, you know, towards what is this algorithm? How does it work? And what's it doing, essentially? What's it deciding? So Canada's worst Listeria outbreak was in 2008, and that was the Maple Leaf Foods Listeria outbreak, which involved lunch meats. And that resulted in 24 people dying and many more getting sick and hospitalized. And that situation really led to the system we have now, because after Maple Leaf Foods,

there was an examination of the food safety system. And one of the problems they had identified was a lack of inspector resources and an inability to spend enough time examining that plant, which was a meat plant. And so what they decided was over the next several years, they would go to a system where calculations would be done on which plants were the most riskiest and divert attention towards those plants. And that's what the algorithm is. They take information supplied by the companies, and then they take information from within the CFIA, and they form a picture of all the various plants. And the algorithm will spit out a priority level for each plant. And that dictates where the inspectors go, but it also dictates where they

don't go. And in the case of this plant, the algorithm didn't flag it for inspection, not just for one year, but for several years. When we started paying attention to the algorithm, we realized that is the whole crux of this entire outbreak. And it's why this problem was allowed to get as bad as it did. So let's talk about this facility specifically then, Jareki. Catherine, why did this system fail specifically then at this facility? Well, I think put simply, the Canadian Food Inspection Agency failed to properly calculate the risk of the plant. And as a result, as Grant said, it didn't flag the plant for inspection. And what that means, practically

speaking, is in the course of our reporting, you know, I know Grant and I can remember going back and forth with the CFIA over the course of a few months. Initially, we wanted to know how many times has it been in the plant to look for listeria control practices and to test for listeria. In the 11-month outbreak window, we were told not at all. And then we thought, okay, well, when was the last time you were in the plant to look for listeria and check on listeria protocols? And on that one, they said, well, they had been in the plant as far back as 2019, but that was actually on an unrelated matter that had to do with consumer complaints related to possible mold. So that didn't involve checking for listeria either.

And then we went back and we were like, okay, well, when was the last time you were in there to check for listeria and check on their listeria protocols? And they came back and basically said, we don't have a date for you. Wow. And the implication of that is a very long time ago or never. What we were hearing from current and former inspectors is that inspectors were sort of no longer using their intuition to guide their work. Instead, they were relying on an algorithm that in this case, they agreed missed some crucial red flags. So just very quickly running through them. One is

so much of the information that gets inputted into the algorithm is provided by companies themselves. And what we found first through sources and later confirmed by the CFIA was that the company hadn't updated a questionnaire that sort of feeds into the algorithm. So they weren't accurately reflecting the volumes of the product being produced at the facility based on some new categories that the CFIA had introduced. When we asked Riki about inaccurate information on the questionnaire, the company said, quote, we are not aware of this. So the system is relying on the company to self report this information, but the company wasn't doing it. And it seemed like nobody noticed that then really. Twelve of the 16 inputs into the algorithm are company supplied. Another issue is just related to consumer complaints.

Now, complaints falls into the bucket that the CFIA is supposed to be monitoring for. Those are among the four things that the CFIA inputs into the algorithm. That's our understanding. And we know that despite the fact that there were consumer complaints related to possible mold at the plant, those concerns weren't enough to kick the plant up the priority list high enough to actually see an inspector. And the third key point, I think, it was easy to sort of get caught up in this being a story about plant-based milks. What's wrong with plant-based milks? We thought plant-based milk should be fine. And it turns out that the government misjudged the risk of plant-based milks too. And so our reporting showed that the CFIA's position on the product is not only contradictory, but when it comes to prioritizing plants that make plant-based

milks, it was wrong. So the upshot is that the algorithm appears to have minimized the risk of the product itself. And so I guess all of this being said, specifically when you're looking at that facility, what did those who run the Jiriki facility, what did they say in response to your reporting here? So Jiriki, which is the third-party manufacturer that was producing the plant-based milks at its Pickering facility, disputes the CFIA's assertions entirely, specifically that it had not been conducting any testing for Listeria in finished products or on any food contact surfaces. In a statement to The Globe, the company said it had a monitoring program for Listeria in place, which included swabbing the

production line and conducting finished product testing. And it specifically asserted that its program, quote, exceeds CFIA requirements. We asked them to see, you know, details of that testing, you know, what dates did you do the tests and what did those tests find? And the company did not share that information with us. They also made the point in their statement to us that the CFIA never raised any concerns about its program prior to the outbreak and said that it has only seen government data on one of the outbreak cases. So I know your investigation really did focus on the situation at this facility, but also looked broadly to it how this system then is working. So Grant, if we look at this kind of a big picture level, all these things that you're talking about, this system that's based around this algorithm, is this working, I guess, the way that it is supposed to? I think the evidence shows the algorithm-based system is not working anywhere close to how it was intended to work because you have plants falling through the system

that clearly need to be checked and aren't being checked. And not only that, the CFI isn't making its own stated targets of inspecting plants like this. And so the algorithm process, which was designed to put focus on high-risk plants, has really let down Canadians in terms of inspecting other plants. And that's what we see here. So if we look at the system as a whole and the number of plants that are operating, do we have a sense of, like, how many are getting inspected? How many are supplying this information? Do we know? There's about 8,000 federally licensed food production facilities in Canada. And a number that really surprised us that the CFA has told us is that there are 1,443

of those that at this point don't even have a formal ranking for risk because they haven't submitted their information to the agency that would be plugged into the algorithm. And so what the CFIA told us is that they've been given essentially a default rating of Priority 6. What that means is there's a seven-level system that they use. So Priority 6 is the second lowest risk rating. So they haven't supplied the information, so they're being ranked low, so they're probably not really going to be inspected very frequently. Absolutely.

And what it also means is that there's 1,443 plants that haven't been checked yet that are operating, and they've been given this default low priority rating because they haven't submitted their information, which shows you a real weakness in this system. It's company self-reporting, and the information may or may not be validated a lot of times, inspectors tell us, and in this case may not even be supplied. And so potentially, the upshot of that is you've got other plants falling through the system. So what has the CFIA, the Canadian Food Inspection Agency, what have they said about all of this? So publicly, the Canadian Food Inspection Agency has issued a statement that dates back to the end of October, expressing their sympathies to the families and also stating that they had concluded their food safety investigation related to the contaminated products at the Pickering facility. And, you know, they said that they weren't able to identify the specific source at that facility. They also said that its inspector general would begin an initial review of the circumstances surrounding this recall, as is kind of standard practice. So I think it behooves us to press them on what exactly they will be

doing in the face of this outbreak to understand what went wrong here and to address it. I guess I wonder then kind of the next level up when we look at the federal government, what their response has been. So the federal health minister is Mark Holland. Has he commented on the situation? We're waiting. What we know is that the inspector general process that is ongoing, they're not looking specifically at the algorithm, which is really at the heart of all of this. And so that's the big question. Is the minister going to expand their probe into looking at the root cause, which is,

you know, this algorithm system that allows plants to fall through the cracks? So as of today, is this algorithm still in use in this system? Absolutely. I want to come back to the people that we talked about earlier who've really been quite affected by this outbreak. I guess, where does all of this leave them? They're picking up the pieces, and many victims, including some of the people we interviewed, are planning to be part of proposed class action lawsuits against the companies. But those could take years to be resolved. I think Grant and I both received messages from some of the victims after the

story came out. They knew a lot of the reporting that was going to be in there before it was published. But I think seeing it all put together was quite startling to them. For example, Natalie, one of the women that I spoke with whose seven-year-old daughter became violently ill and ended up in the hospital for a couple of weeks. She is entirely paranoid to put anything on her dinner table for her kids to eat. So it's this fear that she lives in of not only what is she feeding her kids, but is her daughter okay? Because she fears that there will be lasting symptoms. And in the meantime, yeah, the victims are left to pick up the pieces. One sort of good news story, I suppose, is that Sonia, last time I spoke with her, she was actually pregnant with a girl and was due any minute. So I actually texted her yesterday. I would love to hear if she had given birth. But I do know that that pregnancy was incredibly stressful for her

for obvious reasons. She didn't trust her body. She didn't trust the food. So that's one of the consequences of this whole thing. And Kale Sampson's the same way. You know, he was really reluctant to speak publicly for this story. Him and I talked for a few weeks on, you know, he was really reluctant to speak publicly for this story. Him and I talked for a few weeks on, you know, why he thought it was important because his mother Muriel was a very private person and he wanted to respect that. But he also knew she was somebody who, you know, believed in her convictions and he didn't think she should have died this way. And so he believes this system needs to change. And he's hoping that is his mother's legacy. Grant, Catherine, thank you so much for being here today. Thank you.

Thank you. That's it for today. I'm Mainika Raman-Wellms. Kevin Sexton produced this episode. Our producers are Madeline White Michal Stein and Allie Graham David Crosby edits the show Adrian Chung is our senior producer

and Matt Frainer is our managing editor Thanks so much for listening and I'll talk to you tomorrow