Language selection


The food safety puzzle: piecing together an outbreak investigation

Behind every outbreak of food poisoning, a team of detectives works around the clock to pinpoint the source. This podcast episode, 4 experts from the Canadian Food Inspection Agency, Health Canada and the Public Health Agency of Canada, explain how they connect the dots to save lives. Subscribe to food recalls and alerts.

The food safety puzzle: piecing together an outbreak investigation – Audio transcript

Michelle Strong (co-host): If you thought you had food poisoning, would you report it? Would you get tested?

When it comes to foodborne illness outbreaks, there are teams of detectives working tirelessly around-the-clock to find the culprits and save lives.

Greg Rogers (co-host): This episode, we'll be speaking with 4 of these detectives, who've investigated many cases across Canada. Alright Michelle, let's introduce our guests.

Michelle: Hi, Courtney. Hi, Tanis. Do you want to introduce yourselves?

Tanis Kershaw (guest): Sure. Thank you so much for having us. So I'm Tanis Kershaw. I'm currently an acting manager with the Outbreak Management Division at the Public Health Agency of Canada.

Courtney Smith (guest): Yes, my name is Courtney Smith. I'm a senior epidemiologist and I work right alongside Tanis in the Outbreak Management Division.

Courtney: By the book, you might see an epidemiologist defined as someone who investigates, you know, patterns and causes of disease. More informally, we often explain the role of an epidemiologist as sort of a disease detective. And I'm doing air quotes for those who can't see me who are listening in.

Tanis: I do, I do think of us like detectives. I'm not sure if real detectives would appreciate the analogy (laughs), but the disease detective that Courtney mentioned, we really do feel like we are looking for clues and pieces of evidence and trying to put that all together.

Michelle: Next we want to introduce you to Fred, a food inspector at the Canadian Food Inspection Agency.

Fred Jamieson (guest): Hello, I'm Fred Jamieson. I'm the Food Safety Recall Specialist with the Office of Food Safety Recall in the Food Safety Investigation Recall Unit.

I'm a little bit of Sherlock Holmes, and a little bit of CSI because you kind of use the deductive reasoning to create a hypothesis of the potential source or root.

However, you know, unlike the TV show, it takes longer than 45 minutes plus commercials to conduct a food safety investigation.

There's no machine that you can put a sample in and get a result in 2 minutes. And finally, we don't get to carry guns, which is probably good because I'd probably shoot my toe off or something like that.

Michelle: And finally, we have our fourth detective, Angela.

Angela Catford (guest): So my name is Angela Catford and I am the Section Head of the Health Risk Assessment Section at Health Canada.

I've actually been in the food safety and in the risk assessment world for over 13 years now. It's an expertise you kind of build up over a long time.

Greg: Have you ever gotten sick yourself from food?

Michelle: What a question.

Angela: I know, right? Okay. It's our greatest fear to hold a work potluck and have an outbreak associated from our team. (laughs)

Greg: Hi, I'm Greg.

Michelle: And I'm Michelle. And this is Inspect and Protect, the Canadian Food Inspection Agency's official podcast.

Greg: All about food safety, animal health and plant health.

So how do you go about investigating a foodborne illness outbreak? Like can you walk us through kind of a step-by-step how it usually goes?

Tanis: So in terms of how we detect the outbreak, first people have to be exposed to the contaminated food and become sick and then seek care. So they need to go to the hospital or to a physician to have a stool specimen tested. So if it comes back positive for something like E. coli or salmonella, then what will happen is it will be reported to the local public health unit where that person lives. Someone will call them up from the local public health unit and they're going to ask them questions about the foods they ate before they got sick, animal contact they might have had, travel, and water exposures. At the same time, it's also going for further laboratory testing.

Fred: When there's an identification that there's a foodborne illness, and in particular, where then by definition, if there's 2 or more, then it's an outbreak. So the first step is identifying the trigger. So this was the being notified that there's an outbreak. That's the trigger. The next step is investigating.

Angela: And that's where the Public Health Agency can kick in. And they're trying to say, okay, I have to start solving this mystery.

Tanis: So just to set the stage, Courtney and I work at the Outbreak Management Division, which is in the Centre for Foodborne Environmental and Zoonotic Infectious Diseases, part of the Public Health Agency of Canada. So we coordinate the overall investigation and we also conduct the epidemiologic part of the investigation.

Courtney: While investigating our outbreaks, we do have like a lot of different tools in our tool belt, but the main one is definitely case interviewing.

And the focus is gathering those very specific details on what the cases ate before they were sick. So essentially, as part of that re-interview process, we're calling each case in an outbreak to ask them what they might have eaten in a certain number of days before they were sick. And our goal ultimately is to find, you know, the commonality across all of the cases and the source of the outbreak. But, of course, like anything, there's a catch. It can often take, you know, 4 to 6 weeks between when a person is sick and between when we're giving them a call for this re-interview. But luckily, we have a lot of expert interviewers and they have a lot of different strategies to help our cases remember what they might have eaten one or even two months ago, sometimes even longer.

Tanis: So we're going to be looking at is there anything in common among these people in terms of the foods they ate? Are they all reporting? Are they all healthy eaters reporting a really, you know, fruit and vegetable type diet, or are they all reporting a very specific type of one food? We're also looking at age. Are they all different ages or are we seeing a lot of young children in a cluster which might give us a clue? What would they be eating that might be different from other age groups?

Michelle: This is detective work.

Tanis: Yeah, we think so. And so we're also looking at where people live. So urban versus rural in particular can be a helpful clue because if you live in a small town, there might only be one grocery store that you get your food from and that might just limit the availability of different food items.

Courtney: We might be asking even more questions about, for example, you know, if they can look into their bank statements or their credit card transactions to see if they went to a restaurant that day or went to the grocery store or a cafe, we might ask if they have receipts from any grocery store purchases for that time period. A lot of people these days are taking photos of food on their phone so we can ask them to look back at their phone to see if they have any pictures of particularly aesthetically pleasing meals they might have photographed during the time that we're interested in. And we can even ask our cases about their loyalty card information. So for example, in a lot of grocery stores these days, you go in, you have a card and you're collecting points. With the case's consent, we can take their loyalty card number, follow up with that grocery store and then get a detailed record of everything they purchased in a very specific time frame.

Greg: Wow.

Tanis: So we might have learned you had a salad before, but now we want to know every ingredient that was in the salad, the dressing that was used, did you make it yourself? Where did you purchase it from? And then we also are looking for food safety information as well. So that's where we work very closely with the Canadian Food Inspection Agency. So looking at trace back, one piece that's very helpful for us is supplier information. Sometimes we might see people reporting…everyone's reporting yogurt, for example, but it's all different brands. Well, is it possible that these brands could be made at the same place? That's something that CFIA would be able to look and see if that's a possibility, helps us rule in or out different items as potential sources.

Fred: So what we'll do is that based on the information we get, we kind of do the tracing of the foods from the consumer back through the retail and then possibly in the case of hotel, restaurant institution issues, back through distribution to production, to the processing facilities to pinpoint the suspect source of the problem. So it's like much like in a criminal investigation that everything seems to point at a particular food if we're lucky.

If they gave up a particular product with a particular brand and a code, then what we would look at is we would reach out to the company to start getting information on that particular product, the volume, where it was shipped and sent, if there's a shelf life. And then at the same time, we're looking at if it's within Canada. So we're trying to get like production records, the volumes. Was there any quality assurance issues? Was there any sampling? Based on the nature of that, say it's a pathogen, which is going to cause a foodborne illness: do they do testing for it? Have they had a problem? Is there any indication there? So you're trying to build that case to build documented evidence.

Now what you have is how many different products were involved, because normally it could be one product, but sometimes you could have several. Sometimes it gives us almost like a laundry list or a grocery list of products.

And so we've had that. A lot of it is that the initial information, can we confirm the suspect? You have to kind of like, almost like an onion, you do the first one, then you go back and you do the same investigation until you literally go and find it. It can go very quickly if it's singular. But in some cases, they will bring us a list of a lot of products and we have to run each down. And that can take weeks or even longer until we can get the confirmation. And you have to literally go through each one and take them off the list of suspects.

Angela:  So Public Health [Agency of Canada] starts the ball rolling when we start suspecting different food sources, the [Canadian] Food Inspection Agency hops in and starts doing food safety investigation. They're going to start looking for what's going on at different places where food is made, manufactured. Maybe they're going to go take some food samples and do some testing for the different bacteria or parasites or viruses. And all the time that this is happening, we're all continually meeting all the different folks – Public Health [Agency of Canada], [Canadian] Food Inspection [Agency] and Health Canada, and we're on the phone talking to each other.

And then when we start to have a clearer picture and we say, okay, we know what's causing this from the hazard point of view, the bacteria or the virus, and we have a suspicion about what food is involved, then they're going to pass the ball to our court and we're going to do our risk assessment at Health Canada.

Greg: Right.

Angela: And then finally we say, okay, we have an understanding. How bad would it be? You know, what's the hazard? How sick could it make someone? Is it possible and how did it get there? And how much might have been there when someone ate the food and then we'll make a determination about the risk. So we put it together and we identify a level. And the level could be either a health risk 1, which is the most serious, health risk 2, or health risk 3. And it's possible that we can take a look at everything and say there's no health risk too, or that the whole situation doesn't result in a health issue really for Canadians.

Once we do the risk assessment, it goes back to the [Canadian] Food Inspection Agency.

Fred: So then a decision is made by the Agency, and they'll do a recall. And in the case of an outbreak and a pathogen in a product, it would go to a public release. So we'll notify the company that we're requesting a recall. The company would then either say yes or no, whether doing the recall.

Greg: Does anyone ever refuse to recall the product?

Fred: Since 1999, there have been approximately about 10,000 recalls. And in that time, we've only had 7 what we call mandatories. And this is where, if a company is not available or not willing to conduct the recall voluntarily, the Minister of Health under the Canadian Food Inspection Agency Act can order a company to recall a product where the minister believes that it poses a risk to the public, animals and public health. So when you think only 7 mandatories out of over, close to 10,000 recalls … and the first one was in 1999, just after the Agency was created and the last one was actually in April 2004. I think those were learning moments for the industry.

Initially, they waited for the investigation and the decision. But now, many companies are now conducting their own assessments and they will come to us and say based on our information, we want to take this product off. So they do it. And more and more because after 23, 24 years, the industry has learned to get ahead of it and they've always been very good at doing the right thing at the right time once the information comes up.

Greg: If you aren't receiving food recall alerts, we suggest you sign up using our link in the description. You'll be notified of products that are being recalled for all kinds of reasons, including allergens and food safety.

Michelle: Tanis, you mentioned laboratory testing. Can you tell us more about it?

Tanis: One of the main methods that we use is called whole genome sequencing. And this is a laboratory method that is essentially DNA fingerprinting for a bacteria. So for something like E. Coli, the bacteria have individual DNA fingerprints, genetic fingerprints, just as people have genetic fingerprints as well. So then what we do is these go into a national database at the National Microbiology Laboratory. The ones that, the bacteria that have a similar strain, they are more likely to be coming from a common source. And this is the main way that we learn about new clusters of illnesses that are all part of this same genetic profile.

Michelle: So how quickly does all that happen?

Courtney: Not quickly.

Michelle: Like I'm just picturing waiting in the emergency room there, like you're fine, Michelle, go home or…

Tanis: Yeah it can definitely take some time. Once you get your specimen collected, it can take a few days for testing. It has to go through the whole chain from the time the person actually is tested, from the time we learn about it can be 4 to 6 weeks. So it's definitely a big time lapse there. You likely will know that you have E. coli or have salmonella within a couple of weeks or a shorter period of time. But to do that, further laboratory testing does take a little bit more time.

Greg: So we talked about the successful investigations, but are there cold case files, like the ones that got away that you just never. Never… I'm thinking like the seasoned detective, you know, had this one criminal that was just, they could never track down.

Michelle: This is true detective.

Courtney: For sure. I mean, we don't have a 100% success rate. We give it our all for every outbreak. We have a motto to leave no stone unturned. But certainly we're not always successful for a variety of reasons.

Fred: Some of the challenges that we have is the inability to find the root source or cause. And so this happens often with outbreaks associated with the consumption of short shelf life vegetables. A good example, like leafy greens, romaine lettuce. So when you look at it that by the time the product is harvested, shipped, put on retail, you purchase it, you eat it. That could be a couple of days later. Now you're not feeling well. You go to the doctor, you have to give a specimen that has to be analyzed and then it has to go off and have this isolate or the genetic fingerprint done.

Weeks go by and on a shelf life. So then what happens is that then you have a challenge where by the time it comes to us and the information that there's an outbreak, the products that are suspect, there's no samples available. So we can't confirm it. And then given the nature of some of the fresh greens that are harvested, they do them by different states and fields as they move around. So it's sometimes very difficult to identify.

Tanis: If we don't find the source it means the outbreak has ended. So we would never stop investigating if new cases are continuing to be reported and we haven't identified what that source is. We're always going to keep investigating until it either ends or we've identified the source.

Fred: We had an issue with the leafy greens, romaine lettuce, where it actually took 3 years of recalls, and every year we got a little closer where, I think it was the third year, they actually identified the source. And again too, when you look at some of the greens, they're mixing and matching and adding a variety of different lettuces from different sources. If you go to the grocery store and you find one of these mixed greens packs, which source is it? So this is where it becomes challenging, where you got a variety of different ingredients on short level. So that's one of the biggest issue with challenges.

Michelle: You've got me so curious. So what was like the most notable foodborne illness outbreak you've ever worked on?

Tanis: Well, I'll go first.

Courtney: There's so many. Tanis, you go first with yours.

Tanis: So for me, I've worked on quite a few over the years, but one of the most interesting ones for me is still one of the first ones I worked on, which was an E. coli outbreak that was found to be linked to flour. And flour was not a known source really at the time. So it wasn't something that was already on our questionnaire. And it was the first time that this was identified in Canada. So it was quite a challenge because, as you can imagine, flour is not a food people eat. If I asked you about what you ate in the last 10 days, I'm thinking almost no one to no one is going to say I ate flour because it's an ingredient. Yes. So people are using it in baking, but they're also using it to make sauces, gravies, homemade Play-Doh, even in some beauty products.

But another challenge with flour is that a lot of people transfer the flour into a separate container in their house. So there's no lot code or best-before date. There's no packaging. Even if you didn't remember the brand you bought, you don't have that bag anymore.

Michelle: Labels are important, people.

Tanis: (laughs) It is important. Yeah. Eventually, we were able to crack the case. We did get a positive sample and we were able to link it all together. It was a great learning because we did make some changes from that. So flour is definitely something we ask all cases now about for any E. coli investigations, but also it's changed public messaging, too. So I think before people were a bit aware of the raw dough being a risk because of the potential for raw eggs. But what this investigation taught us is that raw flour is also a risk. It's a raw agricultural product. It does need to be cooked. And so even on the bags, the industry started putting that warning right on the bag that you do not eat raw flour or raw dough. They were updating websites, our own websites were updated, social media. We do know that saying not to eat raw dough is quite controversial.

Michelle: Yup. Raw cookie dough!

Tanis: We did get a lot of feedback on social media – a lot of people saying that they would never, ever give that up. And for us, you know what? That's a choice, right. So our job is to make sure that they have the right information to make that informed decision about their own health.

Michelle: And on that particular example, how long did it take to identify the source, that it was flour?

Tanis: Yeah. So I vividly recall we identified this cluster at Christmas time in 2016 because almost everyone was off. And I remember going: "Oh, I think we have an outbreak." And it took us from there about 2 months, about 2 and a half months before we were able to confirm the source of that outbreak. So this was a longer one.

Michelle: What was your example?

Courtney: The most notable outbreak when was kind of thinking of all the ones I've worked on, I think the most notable for me was an outbreak of E. coli in kimchi. This outbreak was more or less the exact opposite of the flour outbreak that Tanis was describing because we solved it in less than 48 hours. So kimchi is, as you can imagine, a more unique exposure. The majority of the population isn't eating kimchi on a regular basis.

Michelle: I love Kimchi.

Courtney: Oh, really? Okay, so you're one of these ones maybe eating it on a regular basis. But when we had a few cases reporting eating kimchi within the 10 days before they became sick, we knew we were on to something pretty early and it went very, very quickly from there.

So we worked closely with the Canadian Food Inspection Agency, and there was a recall of a very specific kimchi product within the same week that we identified the outbreak and it was removed from the market and nobody else got sick, which was very rewarding for this outbreak in particular because it was a specific strain of E. coli that's known to be very severe. It was super rewarding that we were able to find the source so quickly. And it was also really interesting because I think many people, including ourselves during this outbreak, thought that an acidic food like kimchi wouldn't be conducive to bacteria growth.

Michelle: Right.

Greg: Right.

Courtney: But we learned that that's not always the case. And that's definitely a key takeaway from working in outbreaks for several years is that you cannot assume anything. So we're always, always learning as we go.

Michelle: Are there any other types of challenges with this job?

Fred: Some of the challenges, as the person doing the investigation, like emotionally challenging is when there's illnesses that are hospitalization, deaths. I've dealt with many investigations where there were people that passed away. That's tough because it actually is that manifestation of the food safety risk that we're trying to protect. So that's tough.

Angela: In the last 2 years, I've already got several examples where we are up at like 2 am to do our risk assessment at Health Canada. It's all very well-defined. We have 8 hours turnaround time, so it has to be done basically in one working day. Although when I say one working day, we often receive our requests at the end of the day and our 8 hours starts then.

Michelle: Obviously, we don't want these outbreaks to happen. But when they do, it seems the lessons learned are valuable and they make positive impacts.

Tanis: So every time we have an investigation, we're always looking at: what did we learn from this? Is there something new that we should be letting people in Canada know about? And one kind of real life example of that was in 2021, we investigated an outbreak of salmonella that was found to be linked to frozen corn. And this was the first investigation of salmonella to be found to be linked to any frozen vegetables. So before this, it wasn't really considered to be a huge risk. And so we have updated the messaging about this now on the website so that people are aware that frozen vegetables do need to be cooked according to the package instructions. They're not a ready-to-eat product.

Angela: In the long term, we all want to work together to make sure it doesn't happen again. So we try and stop the outbreak at the time that it's happening. But then we also want to stop future similar outbreaks. And that's where the partners can also work together and we can implement food policies that are different. We can do some more research. And there's definitely work that happens afterwards, but that's probably about 2 or 3 other podcasts' worth of information.

Fred: Each one of them bring a different part of the expertise. So if you want to think baseball or football, each one is a key player on the team and each one is important. And if they don't do their job like they always do very well, we don't succeed.

Things do happen. It takes time. And I understand when people are getting sick, there's a lack of patience. But you do really want to get it right. You don't want to make a mistake. You don't want to recall the wrong thing because it'll continue to happen. I would put our National Food Safety Emergency Response team up against any in the world.

Greg: Thank you so much for joining us today. I learned a lot. I just still have this detective kind of picture in my mind, with these cold case files of unsolved mysteries.

Michelle: Thanks for sharing your knowledge with us on this.

Courtney: Thank you so much for having us.

Tanis: Thank you.

Fred: I appreciate it.

Angela: It's super exciting. I think all of us are really passionate about our job and everybody needs to eat and nobody wants to get sick from the food they eat.

Michelle: Right

Greg: For sure.

Michelle: Foodborne outbreak investigations are a team effort across Canada. Local, provincial, territorial and federal public health units work together to identify the issue and recall all related products.

Greg: If you aren't currently receiving food recalls and alerts, you should really sign up for them. We've put the link in our description or in the Learn More section on our web page.

Michelle: This was Inspect and Protect, the Canadian Food Inspection Agency's official podcast.

Greg: Thanks for listening!

[End of recording]

Listen on Apple Podcasts

Listen on Google Podcasts

Listen on Spotify

Learn more

Get more Inspect and Protect

Date modified: