Analysis of CDC Database On Foodbourne Illness: Most Outbreaks Not Associated With Produce; Foodservice/At-Home Mishandling Is Chief Cause Of Produce-Related Outbreaks
Jim Prevor’s Perishable Pundit, March 29, 2010
Last fall, when the Center for Science in the Public Interest came out with its list of the “Ten Riskiest Foods,” we objected to the assertions being made with a Special Edition:
An Opportunity Missed: ‘Ten Riskiest Foods’ List Highly Deceptive, Worse Than Useless to Consumers — CSPI’s Quest For The Headlines Means America Misses Out On a Rational Discussion About Risk.
The piece was widely praised, linked to and reprinted in many outlets, and we published a follow-up piece that included a baker’s dozen of letters:
Pundit’s Mailbag — Letters Pour In On CSPI’s Highly Deceptive Riskiest Foods List
Now a very different organization, the Alliance for Food & Farming, is publishing a study of its own. This one is titled Analysis of Produce Related Foodborne Illness Outbreaks.
This study is different from the one done by CSPI in several ways: First, it is not limited to FDA-regulated foods, as was the CSPI effort. Second, this looks at produce versus other foods as a source for foodborne illness, not breaking out individual items — be they produce or non-produce — and, third and most importantly, this study tries to tease out to what extent illnesses attributed to produce are due to problems at the farm or at the processing plant as opposed to actions later in the supply chain, say at a restaurant or home kitchen.
It seemed like an intriguing approach, so we asked Pundit Investigator and Special Projects Editor Mira Slott to find out more:
Alliance for Food & Farming
Q: What are the key points you want to get across?
A: The first thing is that the Alliance for Food and Farming commissioned an independent scientist to analyze the Center for Disease Control’s data base on foodborne illness outbreaks. And our report is unique because what it does, which is not readily available in CDC information, is it identifies the sources of the outbreaks.
We categorized the sources into different areas. One category is associated with produce outbreaks period. What we found was from all the foodborne outbreaks, covering the period from 1990 through 2007, which was the most recent data available, 12.3 percent were associated with produce and the remaining 87.7 percent of all foodborne outbreaks are not even associated with produce at all.
Ten percentage points of that 12.3 percent associated with produce represent outbreaks that occurred after the produce was handled by either a foodservice establishment or in someone’s home. The largest portion of that 10 percentage points is at foodservice establishments themselves. So 65 percent of that 10 percent is in a foodservice operation, 14 percent of that is at community events, and 13 percent of that actually happens in the consumer kitchen.
The other 2.2 percentage points of that 12.3 percent is actually produce outbreaks that are associated with growing, packing, shipping or processing. In other words, 2.2 percent of all foodborne illness outbreaks can be traced back to the farm, packing house, processing plant, or even in transportation.
We do categorize this information by outbreaks and also by illnesses. There are differences, and the numbers vary a bit.
Our point really is there’s a lot of media coverage and a lot of talk about the propensity of produce to cause foodborne illness, and we just really wanted to put that in perspective. It’s clear that produce really is not responsible for most of the outbreaks. That being said, we really want to be careful we make sure everyone understands that the produce industry is responsible for its portion. It can’t just say, hey, this is not our problem. That is really important. One of the main talking points from our report is that 2.2 percent coming from farms is still too high.
Q: To that point, have you examined earlier data within this time period for context and to track progress? In recent years, for example, the produce industry has been linked, fairly or not, to several major outbreaks…
A: We did the same analysis two years ago, and the results were exactly the same, with the same analyst and everything, and that information was through 2004 because that was what was available at the time. This is an update of that information. The numbers have not changed that much at all, which is good, but we’d like to see some progress. There have been a few very high profile outbreaks that have occurred during that time.
Q: The data in this report doesn’t go beyond the 2007 time period. The spinach crisis spurred and expedited aggressive food safety initiatives. Yet, the true impact of those industry measures wouldn’t be accounted for in this analysis… Couldn’t these more stringent practices decrease the incidents?
A: That’s an important point. We should note that many of the programs that are in place now to reduce outbreaks, like the Leafy Greens Marketing Agreement, and the tomato industry’s programs, were not even in place at all for the 2007 data.
Q: Are there complexities or challenges segmenting the CDC numbers, and why doesn’t CDC break out the data for more clarity?
A: The other point we’d like to make with this report is that it was somewhat difficult to do this analysis because of the way CDC tracks foodborne illnesses. We think it is very important we look at this so we can measure our progress. CDC doesn’t go into this level of detail, and in their normal recording they don’t go into the specifics as we did, and we think that is really important.
Q: In many of these outbreaks, they never determine the source, or at least not definitively. Some of these investigations have been fraught with problems. Doesn’t that complicate matters?
A: That is true. They don’t always officially determine the source. That is something our analyst went back and looked at. In some cases, she had to go to other sources, like legal documents she was able to access from attorney’s websites, when there was a lawsuit involved, for example.
Some of this information is really clear. The people who got sick in the outbreak were all in one community, or they were all members of the same church. So even if they couldn’t determine the source, it’s clear that it wasn’t something that was from the farm. I should also mention, if the analyst couldn’t determine where the source was, she used the farm as the default category. If she couldn’t prove it wasn’t from the farm, she said it was, because we wanted to be very conservative in this analysis.
Q: Could you tell us more about the independent scientist conducting the analysis? Are you concerned that skeptics might view the report as biased at the onset?
A: This report was commissioned by the Alliance for Food and Farming. Analysis was conducted by Marilyn L. Duman,
M.S. Duman holds a Masters of Science degree from the University of Hawaii and is a private consultant who specializes in biochemistry and biostatistics. We’ve used her expertise for other studies we’ve done. The CDC data base is public information, and she examined that data. We are of the mind, and what we’d really prefer, is to have CDC do this analysis. We would love to have CDC check our math and officially validate its accuracy. What we want is for the government to be the third party.
It’s really important to emphasize why we did the report. The produce industry should know this information as a way to measure our progress in reducing foodborne illnesses. We think that’s true for restaurants and consumers as well. You can’t have a farmer make all these changes in their practices and have any impact on a problem that happens in the restaurant kitchen or in someone’s home.
Q: Wouldn’t this also put into perspective where we should be investing resources?
A: Exactly. But once again, that is not to say that 2.2 percent is OK. When we get to Zero, then I guess we can say, look somewhere else.
Q: Is it realistic or even possible to get to 0 percent? Mother Nature does create some problems outside of our control…
A: It’s a goal though. I think it would help measure the effectiveness of some of the programs the government is trying to implement.
Q: Could you bullet point the public’s key misconceptions?
A: The general misconception, which treads on our main premise that we want to take responsibility, is that there just seems to be a general feeling that whenever there’s an outbreak associated with produce, that it happens on the farm.
Q: So it starts with an assumption, guilty until proven innocent?
A: Exactly. A lot of times there have been implicated products that have never had a foodborne illness associated with a farm, but they’re listed by these activists as one of the riskiest commodities. A great example is potatoes. Usually, it’s a problem with potato salad.
Q: In that case, it’s probably from the mayonnaise going bad after the salad has been out in the hot sun for hours at a family picnic, or contaminated in the preparation stages…
A: Potatoes can be a vehicle for foodborne illness but they’re not the source. Growing them is definitely not a source of contamination. Potatoes are cooked virtually 100 percent of the time. That’s an example of a misconception that is out there, that all the problems with foodborne illness in produce is how it is grown. It is true that it is much more likely for produce to get contaminated in the kitchen.
Q: Cross-contamination would be a major culprit… or a mother leaves the groceries in her car trunk in the summer while running errands, dropping the kids off at soccer camp, etc.
A: That underscores more of our responsibility to help people understand that if they are going to eat raw produce, you need to be careful. We have to put a greater emphasis on education.
Q: How will you capitalize on this report? Will you be reaching out to the mainstream media?
A: We’re sending out a press release on Monday, and there’s a public workshop FDA is holding on Tuesday about measuring progress, so we’re going to be submitting this report to the FDA. Again, our recommendation there is, please help us with food tracking and monitoring for us to have really easy access to this kind of information.
Q: Are you also proposing that CDC change the way it reports data?
A: Yes. We’d like the information to go more in-depth to provide context and a better perspective of reality.
Q: What else do you hope to achieve?
A: I really want to make sure people understand that the produce industry is really concerned about food safety. Farmers have the most to lose in many ways when there is a foodborne illness outbreak.
Even if it does happen in a restaurant, people do associate that product with illness, and then are reluctant to buy it and eat it. So the farmers have the most to lose, other than those that get sick of course. They do have a lot at stake to make sure they have a safe product, so they are very motivated to implement strong food safety practices. This is a little bit of a misconception too — that they don’t want to be bothered by it, that it’s expensive — and I don’t think that’s true.
Q: Even if not for the obvious moral reasons, a grower’s whole livelihood can be destroyed, a company could go bankrupt if an outbreak occurs. And many of these businesses have been nurtured for generations.
A: That’s right. It doesn’t make good business sense, not to mention they’re feeding the same produce to their kids.
Marilyn is one of the most effective representatives for the industry and the Alliance does important work, including this study.
One also has to praise the Alliance for keeping the focus on safety, not PR for the produce industry. The study indicates that most foodborne illnesses are not related to produce and that most that are related to produce are really related to incorrect handling later in the supply chain, not a problem at the farm or processing plant. Despite this fact, the study does not gloat and emphasizes time and again the importance of reducing pathogens at all levels of the supply chain.
The challenge is that the nature of this kind of research is problematic.
1) Although we appreciate the desire to be conservative — and it appears the study was super conservative in the sense that if the researcher couldn’t determine a cause, she assumed it was “the growing, packing, shipping or processing of produce” — we actually would prefer more transparency. If they do the study again, we would suggest a category be set up for those illnesses whose cause is not reasonably certain, rather than lumping them into another category.
2) The study relies on outbreaks of confirmed etiology. This makes perfect sense as how else would one know how to categorize the illnesses and outbreaks? However, it is problematic because the vast, vast majority of what CDC says are foodborne illnesses in the United States never have a confirmed etiology. As we discussed in our piece, At the Corner of Food Safety and Media Bias, which ran in Pajamas Media, though the Produce Safety Project Study at Georgetown University claimed there were nearly 82 million cases of foodborne illness in the US, fully 67 million of those supposed cases are caused by unknown agents.
3) We also just don’t have much data on how pathogens get into a kitchen to begin with. In other words, a watermelon is a very safe item — the thick rind protects the fruit from most contamination. But if you take a knife, cut up a raw chicken and then, without proper sanitation, cut up the watermelon, you can have cross contamination. It does seem to us that the food industry, as a collective, won’t get off so easily just by showing that an individual product is not the cause. The presence of pathogens in the kitchen still implies that food was typically the source. Even if a cook didn’t practice proper personal hygiene, the question is where he got the pathogen to pass on?
Still, a journey of a thousand miles begins with a single step and this study serves the useful purpose of getting the rest of the supply chain to realize it has responsibilities for food safety, and it is not something that can simply be dumped on growers and processors. We have tried to highlight the way some foodservice operators and retailers deal with these responsibilities with pieces such as this one highlighting the efforts of Cheesecake Factory and this one which detailed how Steritech Group, a company often hired by retailers, can help retailers ensure they are doing their part for food safety.
If the consumer media will give the report some play, it can also give consumers the useful lesson that they too have a role to play and that simply expecting everything to be safe won’t work because even if a cooked food is perfectly safe, improper handling in the kitchen could lead to cross-contamination before the item is cooked.
The study is an attempt to think hard about food safety and aspects not often spoken of. For doing that hard work, the Alliance for Food and Farming deserves much praise.
You can see the report here.
Learn more about the Alliance for Food and Farming here.
Find out here about a webinar taking place today, Monday, March 29, 2010, at 1:00 PM Pacific Time at which time the report will be explained and the issues discussed by: Marilyn Dolan, Executive Director, Alliance for Food and Farming; Teresa Thorne, Alliance for Food and Farming, staff project leader; Ed Beckman, President, California Tomato Farmers and Board Project Leader for the Alliance for Food and Farming Management Board and Matt McInerney, Executive Vice President, Western Growers and Alliance for Food and Farming Management Board Chairman.
Matt McInerney, Teresa Thorne and Ed Beckman