In this webinar, we learn how food professionals can access consumer data from consumers about foodborne illness to be warned of new outbreaks earlier than through traditional channels.
I’m joined by Patrick Quade, the Founder and CEO of Dinesafe and IWasPoisoned.com, to tell us how his organisation collects and aggregates foodborne illness data and how brands can use this knowledge to reduce the size of recalls and limit brand damage.
We also did an impromptu dive into some possible food fraud-related consumer reports.
Transcript
KC:
Okay, hello, good morning, good afternoon, good evening. Hello everyone, welcome to another live event with The Rotten Apple Newsletter. I'm Karen Constable, it's fantastic to have you here.
This webinar is called Early Warnings with Consumer Data and in it we're going to learn how food professionals can access information from consumers about foodborne illness to be warned of new outbreaks earlier than through traditional reporting channels.
I've invited Patrick Quade, the founder and CEO of DineSafe and I Was Poisoned, to tell us how his organisation collects and aggregates foodborne illness data and how brands can use this knowledge to reduce the size of recalls and limit brand damage.
At DineSafe and I Was Poisoned, Patrick leads the development of crowdsourced surveillance systems to detect and identify food safety issues in real time.
He leverages over a decade of expertise in public health informatics and data analytics, skills honed during his previous career in finance on Wall Street. His work in advancing public health outcomes has been featured in the New York Times, the Wall Street Journal, the Washington Post and he's been published in peer-reviewed journals including Open Forum Infectious Diseases and the JMIR Public Health and Surveillance Journal.
We're going to have time for questions at the end so feel free to drop them in the chat.
I will also be asking questions that might be relevant to you who are watching the recorded version and we can get on with it right now.
So welcome very much Patrick, it's fantastic to have you here.
PQ:
Great to be here, thanks so much Karen.
KC:
Excellent, so you've got a presentation for us this morning Patrick, did you want to just jump straight into the screen share, present your screen and we will, I will interrupt if I've got any, if we've got any questions as we go.
PQ:
Yeah please do, that sounds great. I'll jump straight into it so hopefully you can see the whole screen.
KC:
Yeah that looks great.
PQ:
Okay so yeah delighted to be here and a big fan of the Rotten Apple so yeah it's especially cool I've only discovered recently shockingly so I'm really delighted I have and met with Karen.
So why are we talking about crowdsourcing and food safety and social media? It's pretty obvious to most but the bottom line is it's massive, it's everywhere, it's what people do. Eight in 10 use social media and in the US it's estimated 6.5 hours is the average time spent on the internet.
People are accustomed to doing it, to sharing, to interacting with the world so naturally food safety becomes a part of that and right up front there's benefits to that and there are challenges so the benefits are real-time monitoring, rapid detection, larger volumes of data, broad accessibility and reach and it includes non-reportable diseases and novel causation and it's not bounded by borders, it's not like a single state or a single county, it has an international and national regional view all wrapped into it.
The challenge is formal diagnosis is less common, a definitive food attribution is often lacking and it's unstructured data, the information is inconsistent so they're the pros and cons. Why am I speaking about it?
I am the founder as Karen said of IWasPoisoned.com, I was inspired to food safety through a personal experience and started this website quite some time ago. I was not launching a business or a company, I just thought something like this should exist and it grew organically through just organic growth basically and it's a website that allows people to share their experiences online with respect to food safety and it has social aspects like sharing likes and comments and the thing that differentiates it from other social media is that it's custom-built for food safety and as much as the data is unstructured generally on social media and for sure the way that people describe their situation is unique each time, we have a lot of structure around the data to facilitate the metrics that we are able to create and the anomaly detection.
So what do people report? I'm gonna just go through pretty quickly a few examples. So it can be extremely detailed, it can include lab work and a very detailed description of what happened and specifically relate to foodborne illness. There’s a campy [Campylobacter] report here [on screen].
It could be physical contamination, so this is shards of glass in a product, a Kraft product, could be allergens related, allergic reactions either where someone knows what an allergic reaction is, they're eating something that shouldn't have the allergen and they experience that and sometimes it's people who are describing an allergic reaction they actually don't know they're having one, it has all the hallmarks of it.
…. Preparation issues, poor cooking, so the person isn't yet sick but they've observed something that may make them sick or they think may be concerning for others. It could be specifically calling out products, it's not just restaurants and we'll talk a bit more about this later but there can be indications of fraud, food fraud as well.
So we've got a platform, we've got people reporting a whole variety of different types of food safety topics, who cares? Who cares about it? So industry cares and public health cares.
We partner with both and we've developed that over the years and we're really proud of that. From a US public health perspective, we've got over 500 agencies in the US and that includes environmental health, epidemiology and internationally there's seven countries involved.
If anyone's listening from another country we'd welcome hearing from you and we operate natively in seven languages and we've had consumers participation from over 170 countries and we're also really proud of giving, helping give people a voice everywhere.
And the news media cares. So news media follow our site and originate stories and with typically in sync with public health or environmental health but they'll also just originate data based on their own investigations and their own work and originating with data or signals from our site.
KC:
I've got a question for you Patrick, back on that last slide. So for example with Food Safety News, is there a way for them to sort of look into your system and go “Oh hey I think there might be a problem with say the restaurant Pasha in Texas?” or do you sort of go to them or how does that exactly work?
PQ:
It depends, it's case by case. I think I don't know that…, well I'll take the Dominican Republic as an example. That was a huge amount of reporting, a massive amount of reporting.
We were trying to work with the public health agency and the brand in that case but we couldn't get any traction so we'd actually alerted, I think we alerted news media just out of a pure public health duty because that was such a big issue there.
The FDA…. we were partnering directly with the FDA at that time. [Moving on to] Tamales Bay, I don't recall how, sometimes they follow us and will ask us for data and we'll give them data and there are times if it's a public health risk, a clear public health risk….
And of course it's up to the news media to do their own work and make the calls to the public health agency, the environmental health agency, to the brand and do their own work as news agencies to satisfy whatever it needs to make a story if they see fit to do that. So it's a mixed bag.
KC:
Okay so if I was at the New York Times or Food Safety News, I could actually, would you, like if you saw a sort of signal that looked interesting, would you post that somewhere and then Food Safety News will be like, “Oh hey, this might be a story?”
PQ:
It really depends, yeah. Sometimes they'll come to us, some of them watch us, some of them will episodically get their own signals, look at our data, they don't even come to us, they'll just say, “Hey we've just seen the 10th report for XYZ and we have our own intel, like this is a story. We called the environmental health agency, here's their comment” or “They gave no comment” if they chose not to comment.
So it's one of the reasons we really recommend brands and public health get involved because we don't want anyone getting sideswiped. But if no one's watching then you will get sideswiped because the news media does [that].
And that's why we're happy we've got so many environmental health and public health agencies involved so that they're getting a look at this stuff in real time and we're also growing industry participation for the same reason. We're constantly working on that to make sure that people know that, hey, we're here as a partner, we're not out to get anyone.
I think because of the adversarial name, iwaspoisoned.com, it's not like a friendly sounding name. That actually gets back to what I said at the start, I wasn't launching a company or it wasn't like a business being launched, I was just like, I was very sick and I had a poor outcome from the experience that I had and I was like, ”there should be a website for this”.
I created it [the website] and frankly, there's so many squatters… even back then on domain names, everything was either website names were either used or just had squatters who bought anything that was like any whole range of things.
So I had to kind of get creative and looking back, that probably was not, maybe it wasn't the best name for the website because I now have to sort of say, hey, we're here to partner with anyone in the food sector from any side who's interested in getting Intel. And actually, that's part of why Dinesafe exists, which is more like our B2B service, but yeah. Yeah.
KC:
I think, as you get a critical mass of users, and particularly when you’ve got the buy-in from public health providers and public health policy makers and enforcement agencies, and you'll have that sort of critical mass of credibility, I think that's really helping.
And I've been aware of I Was Poisoned for many years, but it's only now that I'm really wanting to learn more about it because I can see the credibility and particularly the baseline reporting, which you'll get onto shortly. So yeah, I'll let you get on with it.
PQ:
Yeah. Super, super. Yeah.
So why are these, to your point, Karen, why would these news agencies or why would we have followers? [For] early detection. So that was actually on the other side, the Pasha, San Antonio, it was 250 people sick.
That was Chipotle in Palo Alto, we had 470 reports. That was Clostridium. This was a super interesting one.
I think it's a great example. This was a particular type of oyster. And we were seeing reports for Hog Island Oysters pop up around the country.
And it's like, “What the heck, what's this?” I don't even know where Hog Island is or why would these oysters, having never seen a report, start popping up. And we contacted the, there's a shellfish regulator on the West Coast that's responsible for that area and the environmental health division or agency in that area that's responsible.
And it turned out to be a norovirus outbreak, the farm was shut down. The key point is until they were able to get cleaned up and get passed to open again and start shipping their oysters, the key point is that no one agency would have figured that out.
They just wouldn't have [realised there was an outbreak] because a single report in a county, you're only seeing what's in your county. And there's 3,000 counties in the US. Some of it's consolidated at the state level, but not all. It's really when you look across [jurisdictions], you can add them along. So this is more than what should exist with a history of never existing.
And that gets back to the Lucky Charms story a bit. So we think that's a pretty cool feature. This was another one where you're asking, I think, or maybe we're chit-chatting about what can people report.
This ended up being a well at a zipline park in Tennessee. We had over 500 reports. It turned out to be E. coli in [the] well water.
So the other thing that's really important and unique that we do is that we've had a lot of coverage about how we detected the Chipotle outbreaks. The first and most of the subsequent [outbreaks] in real time [the first hints] originated on our site. But the other thing that is also something we can do, we're able to benchmark.
So we can say, “What’s the average?”
We have a benchmark set of brands that we use. That's 10 national brands at around 70,000 locations spread across the US. What's the average amount of foodborne illnesses that you might expect?
And then you can look at a single brand and say, “Oh, is that in line with the average or is that lower and better than the average or high? Is it trending up? Is it trending down?” … That type of thing.
In the case of Chipotle, they were 10 to 20 times higher – not 10% to 20%, 10 to 20 times higher – than that benchmark – [the benchmark] for over two years prior to that first outbreak. So that's for brands that don’t have… they simply can't do that.
We have that data. I'll come back to it later.
Only around a third of the reports we receive on I Was Poisoned actually show on the website. This is another reason that we encourage people to subscribe and get involved because by just looking at the website, you're not getting the full picture.
That's a consumer choice. We allow consumers to report. They can opt into outbreak detection [thereby] alerting public health, but do they want their report to show on the website or not? So it's why we do that.
Because we encourage brands, you want to know where you sit on this [as a brand] because to us, prior to that first outbreak, there was a systemic problem. We didn't know what it was because there's a clear break between what we can do and when it passes over to public health.
But from our perspective, crystal clear. And it turned out that there were a range of systemic failures. They [Chipotle] revamped their food safety program and have done a lot of work since then. But yeah, we think that's a pretty cool feature.
And this is a more recent look at some brands and just a sampling of brands to say, hey, where are you trending? Where do you sit?
This was for a singular year. But we think looking at it over time is just as important as looking at a snapshot.
KC:
Yeah, I agree: having the data to look at the trends. And so as a sceptical food safety professional, as some who are watching this might be, when I first came across I Was Poisoned, I was like, “Well, of course a really popular chain like Chipotle is going to have more reports” because as you know, consumers will typically blame the last meal that they had on an illness.
So if a lot of people are dining at Chipotle and they're getting sick from other sources, a lot of them are just going to naturally blame their last meal. So a more popular chain is going to get more reports. [But] what you're doing is seeing a trend where people are going to be always just doing that and that's not a problem. And then when you see this huge trend go up, that's where you're seeing the problem.
And I think when I talked about having that critical mass of data, yeah, that's the slide I'm talking to. If there's hundreds back in 2012, that's not necessarily indicative of any food safety problem. It's a reporting factor.
And then you see the trend and then you go, “Okay, now there's something”. And this is where I think the value of I Was Poisoned is, is because you know the baseline and you can see when things change. And that's just amazing.
PQ:
Yeah. And well, and also, (thanks Karen). Also, this is all normalized for store count.
And so our take [is that] last meal bias exists. But our opinion is it's not skewed towards one brand more than another brand. That's our take.
People don't have last meal bias only for Chipotle. It's a factor that's evenly distributed across brands.
KC:
So you're actually normalizing it like statistically against the number of stores?
PQ:
Yeah, which is critical.
KC:
Oh, perfect. Okay. I wasn't aware of that. So that even takes away that problem.
PQ:
Yeah. The normalization is critical because without the normalization, and that's what this is, this is normalizing, taking that large cohort, normalizing it, and then normalizing other brands because you can't compare a 100 chain store against a 10,000 chain store, but you can [do it] on reports per hundred stores – [that’s] how we do it.
And yes, you've got to do a haircut across all of it for last meal bias, but it's like we said, spread evenly. So when you're 20 times higher, if you're 10 to 20 times multiple of your peers, that's a red flag.
Even before we get to these spikes, even before you get to these spikes, that's how we see it. And this is for supermarkets as well.
KC:
And so the supermarket data is also normalized per hundred stores or whatever?
PQ:
Yeah, normalized location count. Yeah. So that you could compare. I mean, we're not showing the brand names here, but yeah, if you're in a supermarket and you want to know, we encourage people [brand owners] to know that if you're that red bar, that's something worth knowing.
And we also think it's worth knowing if you're the green bar, because if you're spending money on food safety protocols, you want to know that it's working. Well, bravo. Like yes, it is.
KC:
Brand 1's doing great!
PQ:
And you want to know if also if you're trending up, trending down, stuff like that. So we think these are things that are worth knowing.
So well, from there, I was actually going to jump across and I have some food fraud things that I was going to talk about, but I also, I actually didn't put it in the deck, but I thought we'd walk through seeing as you mentioned it the uh… (The Cheerios?)
The Lucky Charms.
KC:
Yeah. Lucky Charms, yes.
PQ:
Yeah. So this is a LinkedIn post, sorry, not a presentation, but I can just go through a couple of the slides here. Yeah.
So this is a case where we got some criticism for like, hey, you know, oh, like, because we had 8,000 reports citing Lucky Charms, an extraordinary number of reports citing Lucky Charms.
KC:
And I'm just going to interrupt here. So for those who aren't familiar with Lucky Charms as a brand, because I actually hadn't heard of it back when this, this story broke (so back in 2021)…
Lucky Charms is a dried breakfast cereal. And so when food poisoning reports started to come for a dry breakfast cereal, a lot of food safety professionals were like, “Huh, what, like, this is weird, like what's going on?” And so that's, that's sort of the background.
This is a really interesting case of poisoning or poisoning reports. And Patrick's got this really unique perspective on what happened there.
I was writing about it in the Rotten Apple and the brand owner never, and the FDA never sort of pinned any sort of smoking gun onto anything particularly.
So yeah, it's a really interesting case. And yeah, Patrick's just sharing Lucky Charms for those of us who aren't from North America. It’s um, it's a unique sort of a food, a sugary breakfast cereal. Sorry to interrupt you there, Patrick.
PQ:
[It’s] Ultra - what are we calling it – an ultra processed food, I think that's, it's in that bucket. So yeah, highly processed breakfast cereal, or maybe they all are, I don't consider ultra processed, I'm not sure.
So the key thing here is this report really focuses on what we saw initially. So we had, you know, a baseline of nearly, you know, nearly zero reports a year for Lucky Charms. And we're thinking (I should make sure I'm on the right one [slide]).
And then Q3 of 2021, you know, this was the, this was a signal that pops up on our dashboard.
So to us, that's, you know, as any of our critics will say, oh, last meal bias, you know, people, people don't know what they're getting sick from. That, you know, all of the very fair criticism, those things are all true. [But here is the data from] 2019 through 2021 – is our take on this.
So for no reason, no press, you know, no kind of, you know, no reason this appears, this bar appears. So to us, that's a worth investigating signal, that that's a signal worth investigating.
So, you know, statistically significant deviation mathematically speaking, you know, it's less than a 0.1% chance of, you know, probability that this was by chance.
Okay. So it stayed high. So, you know, that trend stayed high.
You know, we continued to watch it. And the, we had, I believe we kind of notified, you know, the FDA who we were working with that generally at that time. And they were starting to, you know, take a look at it, and at least look at our data.
And they were, I believe they started to get some reports as well through FAERS, which is [the FDA’s] adverse event reporting system.
A backdrop, which we thought was interesting, there's nothing to prove here, but on the [General Mills] earnings call in March of 2022, they said “The biggest issue we're seeing is really around material selection. So ingredients coming into our products. And some of our products we've reformulated over 20 times a year to date. Every time you make an ingredient change, you have to change the formulation, obviously a lot of work.”
So we thought that was an interesting comment that came out in the earnings call. We can't tie it to this, but we thought that was notable.
We do optical character recognition. It's one of the technologies that we've been doing for some time. When people send a photo in, we scan it and save the text. So it's searchable and reportable.
And these codes indicate that 95% of the reports trace back to the Buffalo plant and the Cedar Rapids plant. Now, also, you know, we don't know what percentage of Lucky Charms are produced [in those plants] - maybe that's mostly where they're produced, it’s unclear, but that's just a data point that we collected.
And so after the media coverage started on that, we went from… because the numbers here are not massive, you know, they're a massive signal to us to go from zero to like a sustained high level citing this same [problem], and it was all from around the same use-by date, which means the production date a year earlier, like a very, very similar time.
It exploded. Around 8,000 people came forward. And I think some folks focused on [said], well, [these reports are just due to the] media coverage.
No, none of this was media coverage. This was all just people independently coming to the site.
And then we did… like this here is a quick study of other household name products that got immense media coverage. And how much of this stuff is consumed per day? So if this last meal bias theory is [in play, then] anytime, if someone's eating a lot of something, they saw, “Oh, this is making people sick. I ate this thing. Therefore, I'm sick from that thing.”
That's like a criticism.
So for McDonald's, the E. coli outbreak, 17 million servings a day, massive media coverage. We got less than a hundred uptick in reports. The Jif peanut butter, we estimate 21 million servings a day. So everyone's eating this. It was all over the news.
And both of these were like real. This (McDonalds] was E. coli and this [Jif] was Salmonella. We got upticks, but they were less than, they were orderly upticks. What you maybe might expect in something that there's maybe, it's not in every product.
Lucky Charms, we estimate 3 million servings per day are eaten, [and the reports were] 8,000. So the message here as part of this was to say, “Hey, it doesn't work like this.” This theory that, oh, well, if it's a household name product, it's in the news, now everyone's going to report. It doesn't play out evenly. Did I make that aspect clear there?
KC:
Yeah. Yeah. Being able to compare with these other [cases], number of servings per day and how the increase changed.
Yeah. That's really, really helpful. I mean, 8,000 is a lot more than < 100.
PQ:
Yeah. So, yeah. And that was, I mean, the FDA were unable to identify causative agent, which isn't uncommon, but that happens. They indicated that their belief was that it was not microbial.
KC:
Yeah. There were very nondescript symptoms, if I recall? I think there was just sort of like a bit of nausea and a bit of something and…. Is that correct?
PQ:
There was, it was a mixed bag. Like there was, the thing that stuck out was there was an enormous amount of green diarrhea, an enormous amount of that. I think it was green diarrhea, like colored; something that was colored like one of the things in the Lucky Charms, which...
KC:
Yeah, they have little green pieces.
PQ:
Yeah, I think it was green diarrhoea. So that was like a super common thing. I think the onset was...
Yeah, yeah, that was super common. Now, having...
KC:
Sorry, what did you say the onset was?
PQ:
I feel like there might have been like a slightly faster onset, if I recall correctly. But I have a different study on that. That wasn't the kind of the focus of this one.
And I don't know where that is. Otherwise, I'd put my hand on it…
KC:
Oh, that's okay. Just for some background for anyone who's watching.
PQ:
But there were a few notable things. It wasn't just like all over the place. There was definitely some very specific things about that. So…
KC:
Good, I couldn't remember. I thought it was a bit non-specific, but that was very specific.
PQ:
Because for sure, out of the 8,000, there's some false positives and last meal bias. There's no doubt about that. But like you say, with these other ones, it's a very different number.
So it's a hugely different number.
But yeah, that was just a write-up to kind of help people understand the original signal. And for sure, we think that after news media, that signal was also important, even though for sure there's false positives in there.
And again, all of this is just to say that: is this something as a brand you want to look at? And we say, yeah, that signal is worth taking a look at. It's worth trying to understand it and see what's going on and signals like that.
KC:
For sure. And I like to think that even though General Meals didn't respond publicly to you or to the media, that they privately went scrambling and looking at some of their recently reformulated ingredients. That would be my hope. So even though we don't publicly learn about those changes, the illnesses did stop.
PQ:
Yeah. Yeah. So that's, yeah, exactly.
So yeah. And then, yeah, I mean, [this webinar] is The Rotten Apple. So I pulled out a few, some sample reports with possible food fraud indications on them.
So I don't know if you want to, if it makes sense to take a look at some of these and just see, I don't know, I'd love to get your thoughts on them. I haven't shown you these before, so this is like a...
KC:
No, I'm up for it. Let's go through one by one and see what it says.
PQ:
So putting Karen on the spot, she's never seen these before. And being not an expert in fraud, like it's, I don't… You know, I'm definitely like more of a newbie on a lot of this stuff. So the first one's premium white grapes sold in zip top plastic bag from Peru.
[Symptoms were] unusual lower back ache, general joint aches and pains, severe tinnitus, no relief from taking extra strength acetaminophen. Didn't see the label saying to wash them until eating some more today then researched and found cause for concern.
I don't know anything about grapes and fraud. I don't know if that's a big one or coming from Peru or no idea.
KC:
So the consumer says they researched [the issue] and found cause for concern. I assume they saw the label saying wash before eating and decided that there was too much pesticide on the grape. That's just my guess, but I don't really know.
And then they probably associated their symptoms with consuming too much pesticide. Possible. We probably don't have enough information there from just one report.
And then in terms of, in terms of whether that's fraud or not? So for food fraud there, there needs to be… So [listener] Yvette’s just jumped in [to the chat] and says, “Grapes are low risk. Peru has a great food safety record.” So, so she's sort of standing up for the, for the grape industry in Peru and saying, probably not a pesticide risk.
Honestly, the amount of pesticide that a grower would have to put on a table grape to cause noticeable symptoms in a consumer would actually make the process very expensive for the grower. So unless there's some weird accidental overdose thing, it's probably (oh and Yvette says she works in the produce industry as well), it’s more likely to be some sort of accidental thing than a deliberate fraud thing.
PQ:
Okay.
KC:
Yeah.
PQ:
Okay. Perfect. All right.
Next one. [Reading] “Bought some Nanac Desi Ghee. I do not know what yellow substance was in this product, but what was sold was not ghee.
Immediately after eating a few teaspoons on my lentils, I had intense nausea and dizziness. Not to mention the ghee smelled rancid and probably was, regardless of the manufacturer's date.”
(By the way we can stop this, if this, if this game is not fun, we can stop.)
KC:
Oh no, I love talking about food fraud. I can do it all day!
So there is [ghee fraud] in India, there is… ghee is a very highly valued product.
Ghee is clarified butter. So it's basically butter where it's been sort of melted down and all the watery bits have been taken away from it. And so a pure ghee from a purist perspective is just 100% animal fat.
And in India, there's this huge fraud problem with ghee at the moment where, because butterfat is very expensive, dodgy operators are making fake ghee. And so they're, they're making it with vegetable sort of margarine and chemicals and colors and stuff like that. And then they're selling it in the packaging that mimics the famous ghee brands.
So there's a double fraud in that they're sort of, they're pretending it's ghee when it's really not. And then they're also counterfeiting the genuine brand owners as well. So there is a big problem with ghee in India.
[But I] haven't heard about that in the US. And rancid ghee will, if it's rancid enough, it'll contain these nasty sort of free radical breakdown chemicals from rancidity. And that could potentially make you sick.
But again, I'm not sure that you'd count that as food fraud if the ghee had just been degraded and sort of gone, gone rancid.
PQ:
I wish I had a stamp. I could put a cross on this one. A maybe on this one?
KC:
Well, I mean, who knows? Food fraud is a very difficult concept to get hold of. You know, I mean, grapes from Peru could have been knocked back for import to Canada because of a pesticide problem and then transhipped into the US.
And that would be fraudulent because the importer might know that they're dodgy and try to get them through customs anyway. I don't think that would have happened in this case, but that would make it fraud instead of an accident. But it's very hard to pin that sort of thing down.
PQ:
Yeah. Yeah. We're not going to get any yeses here, but do we want to see if there's any others?
KC:
Well, salmon is something that is pretty vulnerable to food fraud. So he got sick from eating them. That's again…
So the thing about food fraud is that if you were a conscientious food fraud perpetrator, you don't want people to get sick. When people get sick, you get caught. And so it is less common for food fraud to result in illnesses.
Not to say that it doesn't happen, but particularly the illnesses that we typically see coming out of food fraud can often be sort of more chronic long-term illnesses. Like lead poisoning is the big one from the lead-based colorants that are used illegally [in spices].
PQ:
Yeah.
KC:
And the melamine poisoning, which was a long-term buildup of kidney problems in babies who had melamine contaminated formula. So microbial illnesses from food fraud can happen, but it's much harder to pin that down – [to know if] it was an accidental contamination or some, the result of some fraudulent activity.
Yeah. Olive oil going off? Again, you could you could repackage an expired olive oil and pretend that it's fresh. That would be fraud.
And then someone eats rancid olive oil and it makes them nauseous that, you know, that could be the result of fraud or not. [It’s] hard to tell. But I mean, these are all really interesting cases as well.
PQ:
Let's see. Let me zoom out. Okay.
So [reading] “I bought packaged salmon from H-E-B, one of my favorite foods. Very painful trapezius muscle pain and headache. The label is that it is Chilean salmon. I usually eat salmon that is North Atlantic or Northern Pacific. For me, tilapia is an extreme trigger food. This is the first time I've had a migraine after ingesting salmon.”
KC:
Okay. So that's a really interesting one.
So tilapia is a freshwater fish and it's raised mostly in, (oh gosh, somebody watching this is going to be way more of a seafood expert than me),
I think in countries like Vietnam and I think aquaculture. And so as a species and as a type of flesh, it's probably pretty different than salmon. And it is used as a replacement fish in fraudulent substitution.
So if you have a white fish fillet and a white fish fillet tilapia is a white fish fillet, you can swap out a more expensive fillet for tilapia, which is cheaper. [This is a] known food fraud – [it] definitely happens. But salmon is not, doesn't look like tilapia.
Tilapia is like white, salmon's more pinky. So I, if I was a fraudster, I don't think I'd swap out my Chilean salmon for tilapia. Um, because I think it would be kind of obvious.
Um, but the other thing is that it could be that Chilean salmon (and again, I'm not a seafood expert), um, might be a different species, um, not Atlantic salmon. It might be a Pacific salmon or something like that. Um, it's from further South. So maybe it's a trigger food that this migraine sufferer hasn't crossed before, but it seems unlikely that the salmon would have been swapped for tilapia.
PQ:
Okay. I actually thought they, I thought they dyed salmon. Like I thought salmon didn't actually come out pink. I thought that was one of the secret, the kind of hidden things about salmon. I'm sure I watched a YouTube, which means it was probably wrong.
KC:
No. So if you, you know, if I go down to the beach near my house and catch a, um, an Australian salmon, it will have pinky colored flesh naturally. But if you raise it in a, um, in a fish pen in aquaculture, um, they don't necessarily consume enough of their, um, vitamins that give them the pink flesh.
And so I think they give them feed that does have some of the, I don't know, the carotenoids or whatever it is. I'm sure I'm getting this wrong, but they, they will put, um, pinky colored stuff in the food and the flesh will be more pink than if they didn't, I believe. I don't think that's wrong, but I don't think it's like they're dying it. They just give [the salmon food], it's like when flamingos eat algae that's pink, the [flamingoes] get pink. I think it’s the same sort of deal.
PQ:
Uh, cool. So, uh, infant formula [reading] “The bottom of the can came clumpy and gummy. My five month old daughter, she wasn't finishing her bottles and kept having colics, ER, switched to formula.
“Are we getting a different quality of formula? It's a new label look. They say it's the same. Um, as soon as I opened the first can with new look, this happened.” (I'm not sure what that means). “It was a horrible experience for my baby.”
Now this is probably not, I'm not sure why I put this one in there. It would be hard to know.
KC:
You know, if, if this consumer was in China, we know there is a huge market for, um, counterfeit versions of infant formula. There's just been an absolutely massive scandal in Vietnam where, so people are making formula and they're putting it in the tins that look like premium brands. Um, and the formula is not obviously not necessarily made to the same quality and safety standards, um, as the genuine brands are.
There is nothing to say that this couldn't happen in the US. Um, I doubt that Walmart’s supply chain is porous enough to allow this to happen. Um, but as an example, I'm not sure that this [report] would be a case of that, but it is certainly something that can and has happened, um, where people have bought a brand of infant formula thinking it's a genuine brand and they've got something that isn't.
So I think it was definitely worth putting that in.
PQ:
Okay. All right. Um, so yeah, uh, lots of maybes here.
Last one, Waiakea Water. [Reading]: “[The] case [was] on clearance. Uh, didn't think to check it, but it was still good or not. Why wouldn't it be? Drank it, [it] felt gritty. I thought I was getting sick, cramped up. My wife notices clouds in the next bottle I was about to open.
“Most of the dates stamped on the bottom of a bottle are partially there because they're placed right on the bottom. But one might say 2020. Today is 2024 and every other bottle in that case has a small cloud. It looks like growing from a small black dot seed fungus.
I don't know.”
That's just gross.
KC:
It's really gross. And that also, I mean, it's always a question mark, but that also could be food fraud. It sounds like the water is too old [and] that it's got some sort of, um, fungal spoilage in it.
And, um, there are lots of, um, documented cases of food fraud where, so for example, um, Waiakea Waters’ distributor might've gone, “Oh, this is too old. We can't sell it. We want to protect our brand.”
They'll send it off to be disposed of. And there are, um, certified waste disposal companies that will do a certified secure destruction of the, of the food so that it can't be sold. (Um, yeah, Yvette’s saying she also thinks it may be date code fraud.)
Um, so they'll send it off with their, their destruction company. And there was this huge racket going on in certain parts of the U.S. and I'm sure it happens in other countries as well, where the waste disposal company would, um, pretend that they properly disposed of the food so that it couldn't reenter the human supply chain.
They would issue a certificate - a certificate of secure disposal - that was fraudulent. And then they would divert the food back into supply chains.
Um, there was a massive ring (I think) that was uncovered last year in parts of Eastern Europe, where this was happening. And so the lower quality expired food was going back into human supply chains in, um, less wealthy countries. Um, and sometimes with the, the correct expiry date overstickered with a new expiry date and sometimes not.
Um, and this [Waiakea Waters complaint] actually, this has all the symptoms of that. Um, so, you know, if they were expiring in 2020, um, Sam's Club might've been approached by some distributor saying, “Hey, we've got like this super, super cheap water. You can make a really good buck.”
And maybe, it's not Sam's Club’s corporate chain. Maybe it's just one store. Someone rocks up with a van full of really cheap water, [saying] “Hey, you can make a motza [a lot of money] if you know, you buy, buy them for 20 cents a bottle.”
Um, that, that is 100% something that can and has happened. Um, and yeah, it sounds like the person got sick because of the, um, maybe some sort of mold growth because the water was old.
But yeah, that to me, that sounds quite like food fraud. And [in the chat] it sounds like Yvette's agreeing with me there.
PQ:
Okay. Awesome. So I, that was amazing.
Thank you for going through that. Um, that was like, uh, great to hear those things, but also just, I was giving, partially doing this to give an example that we do just get this really broad range of things that people talk about on the site that, um, that can wade maybe into the, into the fraud…
At least if you're a brand, then, you know, do you want to be taking a look at this stuff to see if something looks worth taking a look into, you know.
KC:
If I was Waiakea Water, I would be having a really, I would be really upset with that because that's, you know, that's damaging my brand. And it's probably due to the actions of something that's a little bit outside that brand owner's control, but it's important [for them] to know about it.
PQ:
Yeah. Yeah. So, um, my super, yeah.
So, you know, the takeaways like, you know, that for, if anyone who gets to watch this or is, is, um, interested in it, like, uh, you know…
We think it's a great idea to, to monitor, um, you know, and take seriously signals coming from, you know, social platforms. Like obviously we're biased towards our own and we think we do the best at it where, and do a great job at it, but there are, you know, it's not just us, you know, there are other platforms.
And we kind of say like, and I think, um, we've spoken before, Karen, you've pointed out that like, um, you know, it might be someone else in your division or in your company that's taking care of this. So it may be away from you, but you know, this is, it's, it's something to keep in mind if, if this stuff comes to you from your social, you know, um, media crew or whoever's doing it.
And if there isn't someone doing it, we think it's worth doing it. Um, and we think no matter what, like the chances that as a brand that you'll escape something, um, like is low. Like viral moments, uh, um, what's it called? The Coldplay kiss, you know, like that's, uh, I, um, not, not food related, but like where something like can blow up and really, you know, be impactful. Like they're just happening more and more in different ways and, you know, are becoming, you know, more impactful.
So we say [to food brands] at least understand, you know, the types of, you know, what's out there, what are the platforms that people speak up on, um, like understand the nuances of the platforms. Like in our case, I mentioned earlier, only around a third of the reports are actually showing on the [IWasPoisoned] website. So you, you would have to subscribe or, you know, get engaged to get the full picture.
You know, you can look at things like Facebook and look at Facebook and think, “Oh, I looked on Facebook and I didn't see anything”. Well, a lot of activity, you know, activity happens inside groups. Are you in those groups?
You know, in Reddit, do you know how to navigate Reddit and know the places that, you know…. Some of these platforms, uh, you know, are, are not the things… you don't want to be trying to figure out in the fog of war. You want to be figuring it out before that happens.
And, you know, like LinkedIn is a social platform. You know, this is like a general hint. Um, it's a social platform, but the people aren't talking about how they got sick. [LinkedIn] is typically not accumulating reports of foodborne illness. It's just not what people do [on LinkedIn]. They're talking about their career or some event they're attending or some personal achievement. It's not...
So, but it's just working your way through that and saying, you know… If you're in Yelp, is that helpful? In the U.S. yes. I think outside of the U.S. maybe not. [And for] manufactured brands? No, I don't think so.
It really, it depends on your use case. Um, and then, you know, looking at, looking at your own data, um, and, uh, you know, as another way to stay abreast of this and, um, yeah, the bottom line, you know, is, uh, you know, these channels: are they the unique, you know, they should be used to compliment existing risk management protocols.
Are they perfect? No. But they absolutely contain super valuable information. They should be used to compliment existing risk management protocols and, and industry and public health organizations are already doing that. Um, so, um, you know, if you have a responsibility for a product or a market sector holistically, then, you know, be prepared.
If you're interested in seeing what we can do to help, we'd love to hear from you. My email’s right there, or you can find me on LinkedIn and, um, we'll explain what we can do. There's a lot we can do. Our specialty is like carving out, you know, you know, any type of view, whether it's like a certain type of product, that's only distributed to these stores in these markets…
We can do any type of surveillance: that's, that's our bread and butter. Um, so we can be very specific, so you're not getting inundated with like, uh, you know, uh, like a deluge of things that don't matter.
And, um, we're the only organization that can do benchmark[ing] because we're the only ones that have data across regions internationally across all brands or products. Um, and if, if you're responsible for third parties or suppliers, I think there's an argument to say, “Hey, make, make sure your suppliers are looking at this stuff!”
You know, make sure that, um, your third parties distribute, you know, the third party companies that are bringing product into your store or, or whatever the case might be, you know, it's a team effort, you know, so they're going to spot things that you won't spot.
So we'd love to hear from them as well.
So, yeah, that's, I think that's all I was…
KC:
Yeah, adding to that: we often talk in food safety about, um, it's, it's a tough job to be in because when you're doing the job right, there's often feels like there's nothing to celebrate: “Oh, nothing [bad] happened this week. Hooray. We, we met our KPIs.”
Um, and so if we're looking for, um, KPIs in food safety [and] things to celebrate, um, that benchmarking, um, information that you had for supermarket chains and for quick service restaurants, um, they are points of data to celebrate. When you're doing really well compared to your peers in industry [the data] is literal, um, evidence that your food safety systems are working really well.
So I think there's another really, really, really good benefit to, um, I Was Poisoned and DineSafe is the ability to, to show your shareholders and your upper management that the food safety programs that you're doing are working and that you're doing better than your peers or as well as your peers in food safety as well.
PQ:
Yeah, totally agree. And, you know, on that count, we have, um, um, we allow for custom peer groups. So you can kind of say, well, I want to be compared to the people who do what I do, because there's a slightly different profile for everything, so we, we take that into account.
And yeah, we think it's super powerful, like if you wind up [like] your peers…. We deidentify peers, so we don't share, but we'll, we'll group them and, you know, uh, deidentify and show where you sit.
Like some teams are doing incredible, you know, incredible work and just simply, we just do not hear from them. We don't do not hear reports about them and bravo, you know,
KC:
And that's something they should be able to celebrate, in an independently verified way, um, for their shareholders and senior management. So I think that’s absolutely amazing.
PQ:
Yeah. So, um, I think that's all we had to go over. So I didn't, unless there were any questions that came in or any final questions.
KC:
No, I'm just checking. Uh, no final questions have come into the chat. Um, and I didn't have any further questions either.
But what I learned today, um… We, we all know in food safety about the last meal bias, but I hadn't heard it given that name before, so I'm going to tick that off as something I learned today, um, as, as a, as a phrase [to describe that thing].
PQ:
I think it's recency bias. As in the kind of like cohort of cognitive biases of which there are 150 or so (there's some large number), I think it's called recency – [that] is actually the actual name for it.
KC:
Yeah, but in food safety to, to have a label for that last meal thing where you blame the last meal for being sick, I love that. I've put that on my “What I learned today” [list]. Um, and also that, uh, I Was Poisoned is effectively a social media platform [and] that you can, um, actually go to the website and see some data, or you can become a subscriber and see more data.
Um, and so it's something that as a food professional, you could access some of the information for free, just by having a look at the website. Um, and so in that context, it's quite easy to follow and stay in touch with it.
And I will certainly be doing that in the future for The Rotten Apple subscribers: if I see anything interesting popping up, [then] for sure, I will be able to pop it in the newsletter. Um, and…
Yeah and then I also learned that this is a new tool for celebrating the wins in food safety as well for, for larger brands. So, um, that's fantastic too.
Excellent. All right. Well, I might wrap that up then.
PQ:
Wonderful. Thanks so much. Really appreciate it, Karen.
KC:
Thank you so much for joining us, Patrick. And if anyone who's watching, wants to get in touch with Patrick, it's Patrick at DineSafe.org. (I think?)
PQ:
Yes.
KC:
Yep. Excellent. Um, and I'll be publishing a recording of this soon and Patrick, I will send you a link.
Thank you, everyone for coming and for everyone else who is watching the recording, um, I'll be seeing you soon. Bye.
PQ:
Bye-bye.
