Press Briefing Transcript

CDC Briefing on the Burden of Foodborne Disease Illness Papers

Wednesday, December 15, 2010 – 11:00am ET

• Audio recording (MP3, 9.44MB)

Operator: Good morning everyone. During the question and answer portion of our all, you will be prompted to press star 1 on your phone. Be sure to record your name and affiliation so that you may be announced to ask your question. The conference is also being recorded. If you have objections you may disconnect at this time. I will turn the conference over to Miss Lola Russell.

Lola Russell: Thank you so much. Good morning, today we will be providing a review regarding the release of two reports. The first being the "Foodborne Illness Acquired in the United States—Major Pathogens" and "Foodborne Illness Acquired in the United States—Unspecified Agents." These reports have been released in today's online edition of CDC's Emerging Infectious Diseases journal. Both versions are now available. This is the report of burden of foodborne diseases in the United States. The report provides new and more precise estimates on illnesses, hospitalizations, and deaths from foodborne illness acquired in this country.

Our speaker today will be Dr. Chris Braden, that B-R-A-D-E-N, director of CDC's Division of Foodborne, Waterborne and Environmental diseases. Elaine Scallen, assistance clinical professor at the University of Colorado-Boulder and the former lead for CDC's FoodNet team is the lead author of these reports. She is also available to discuss any methodology questions that may arise. After the briefing by Dr. Braden we will open the lines for one question and one follow up from the reporters on the line. We have some excellent resources available online which we expect will be very helpful for you. They may be found at www.cdc.gov/foodborneburden. If you have any additional questions after this briefing, please call the CDC press office at 404-639-3286 and we'll be happy to assist you. The transcript from this briefing in both manuscript and audio files will be posted to the CDC Division of Media Relations website within a few hours. I will now turn it over to Dr. Braden.

Chris Braden: Good morning and thank you for being with us today to learn about the CDC's new foodborne illness estimates. I'm delighted to be joined by CDC's lead author for this report, Dr. Elaine Scallen, and some other CDC foodborne experts who may be best suited to answer specific questions. Today we are publishing the new estimates, the number of illnesses, hospitalizations and deaths caused by food each year in the United States. These are the first comprehensive estimates of foodborne illness in the United States since estimates published in 1999. And the first national estimates that focused solely on illnesses caused by foods eaten here in the United States.

I'd like to start with an overview of the research findings and what they mean, and then I will go back into more detail. First, we want to emphasize that these estimates indicate that one in six Americans or 48 million people get sick each year, 128,000 are hospitalized and about 3,000 die each year from foodborne illnesses in the United States. These illnesses are associated with billions in health care costs and also have a substantial human cost in severe illnesses, and in some cases long-term health effects that linger after the initial illness subsides. With this many illnesses affecting Americans each year, prevention and interventions to address these diseases are vitally important. These are preventable diseases. For many of these diseases, we know what interventions worked to prevent them and we need to do more to lower the impact of these diseases in the United States.

Let me go into a little more detail about what is causing the illnesses and how we develop these estimates. During the past decade, we've made great strides in understanding foodborne illness and in using this knowledge to better prevent and rapidly detect and respond to outbreaks. We know, we now have more and in some cases better data on foodborne diseases than we had in 1999 and we have refined our methods to try to develop the most accurate estimates we can. This research is very important because it provides a window into which of the 31 known foodborne pathogens are causing most harm to the public and which cause the most severe illness. By knowing this, CDC and our regulatory partners can better target those pathogens so we can have the better help to protect the nation's food supply. The reports today can help us understand what we still need to learn about what is causing illness by providing estimates about other causes of foodborne illness, which we are calling unspecified agents. These are agents that have yet to be discovered, those that are not yet recognized as causing foodborne illness, and agents for which there is not have enough data to estimate the amount of illnesses that they cause.

As I mentioned, these are the first new comprehensive estimates since 1999 and they do show fewer illnesses, hospitalizations, and deaths than estimated at that time. However, the lower numbers are largely the result of more and in some cases better data than we had in 1999 and also innovative, new methods that we have developed to calculate the current estimates. For this reason, we really can't compare the two estimates to measure trends, as tempting as that may be. There is evidence that some foodborne illnesses have decreased. In a separate analysis not included in the papers being released today, CDC tracks trends in some important foodborne pathogens every year in the FoodNet surveillance system. We have seen a decrease of about 20% in illnesses due to the major pathogens under surveillance in that system during the past ten years. We are proud of the success and I don't want, in any way, to minimize this accomplishment. However, the pathogens that FoodNet tracks are only a small fraction of the pathogens that are included in the new estimates. So as I mentioned, the vast majority between 1999 and the current estimates can be explained by improved data and methods.

Either way, this is encouraging news for Americans and for the public health. It means that fewer people are getting sick from the foods they eat and that we know now more than ever about the pathogens that are causing illnesses so we can focus our efforts on those. These strides would not have been possible without an effective use of resources and the strong partnership between the CDC and the nation's food regulators – the FDA and the USDA. I want to be clear that the 1999 estimate was the best available at the time. It used the best methods and the best data that was available. But since then, we have more and better data to work from and other studies that have increased our understanding of someone foodborne pathogens, which has also influenced the numbers. We use this new knowledge to develop the estimates we're releasing today, which we believe to be the most accurate to date.

Now let me talk a little bit about the specific pathogens that are causing the most illnesses and these are among known pathogens of foodborne illnesses. This is not including the unspecified agents. So among the known pathogens, salmonella is the leading cause of hospitalizations and deaths. It was responsible for about 28% of deaths and 35% of hospitalizations. We estimate that about 90% of all illnesses due to known pathogens and the associated hospitalizations and deaths are caused by the following seven pathogens: Salmonella, norovirus, Campylobacter, Toxoplasma, E. Coli O157, Listeria and Clostridium perfringens. In terms of overall illnesses, we estimate that norovirus caused nearly 60% of total illnesses due to known pathogens. We will have more details of the impact of each of these pathogens including the percent of illnesses, hospitalizations and deaths available after this call on CDC's website, www.cdc.gov/foodborneburden. This is the same website mentioned before.

Though we have seen clear progress in the past ten years, these numbers illustrate a very real impact of foodborne illnesses in the United States. It affects tens of millions of Americans each year. And I repeat, that translates into one in six Americans ill from foodborne illnesses each year. It remains a substational public health problem. A lot more work still needs to be done, for our new estimates will serve as a foundation for the future to help in the development in food safety policies and in targeting the known causes of illnesses and deaths and in guiding CDC as we work to uncover the causes of foodborne illnesses yet to be discovered. If we could reduce foodborne illnesses by just 1%, we could keep about half a million people each year from getting sick from foods that they eat. CDC remains committed to our role in helping to safeguard the American food supply and prevent foodborne illnesses no matter what the cause. The full report is available on the website of CDC's Emerging Infectious Diseases journal which is published it online today. Their website is www.cdc.gov/eid. We'll have much more detailed information on the numbers themselves and the methods available at again, www.cdc.gov/foodborneburden. To learn more about foot safety visit www.foodsafety.gov. Thank you again for being with us today. We can now take question.

Operator: Thank you. At this time if you would like to ask say question please president star one on your touch tone phone. Record your name and affiliation and we ask that you limit your question to one question and one follow up. Once again, please record your name and affiliation at this time. We have a question from Miriam Falco from CNN.

Miriam Falco: Good morning. One quick note, 2011 estimates even though we haven't finished 2010 yet? These are the projections of how many people will get sick?

Chris Braden: So, actually we called them the 2011 estimates because the actual publication date for the journal is 2011. It's the January 2011 edition. Even though they are being published today, but the data that is used to estimate the illnesses actually spanned a number of years centering around 2006.

Miriam Falco: Okay, um, so my real question is then, you looked at the seven most common sources of foodborne illness, but in how many cases and I don't know if you can tell us, people assume because the peanut butter I got was messed up, how many of these cases were because of improper food handling? People using the same cutting board to cut their salad that they cut the chicken on? Or leaving food out too long?

Chris Braden: So what is included in the estimates that we are releasing today is not a breakdown of or by how the foods became contaminated. Or, for that matter, which foods are responsible for the most illnesses. These estimates are just giving us the total number of illnesses out there. What these numbers are important for, however, is to be able to then go forward with these analyses to take these total number of illnesses and attribute them to specific food commodities. All right, say fruits and vegetables or poultry or beef. That is an analysis that is underway and we hope to have more information to be more specific about what CDC, FDA and USDA can do to protect the food supply in the future.

Lola Russell: Do you have a follow up?

Miriam Falco: No, I think I used up my two questions, thank you.

Lola Russell: Thank you, next question please.

Operator: The next is from Lyndsey Layton with the Washington Post. Your line is open.

Lyndsey Layton: Thank you for taking my call. When we look at the 1999 numbers from the Mead report and the referenced the mortality figure like 5,000 a year and that has been what we often reference when we write about foodborne illness, this new set of data refers to 3,000 deaths a year. So, it would be erroneous for us to say that the number of deaths — or the fatality rate has decreased. We can't compare the 3,000 to the old figure of 5,000? You are saying it would be wrong to assume that the fatality rate has decreased?

Chris Braden: That is exactly right. The differences between the estimates in 1999 and the estimates in 2011, are mainly due to the data that we now have that is better than before and the methods that we have used that are improved from before. Again, they are not comparable for — in order to measure trends. What we would say is that these numbers are the now the most precise that we have of estimates of illnesses, hospitalizations, and deaths in the United States. And with the emphasis that these are, anyway you cut it, a lot of illnesses, hospitalizations, and deaths, and there is more that we need to do.

Lyndsey Layton: Thank you.

Lola Russell: Do you have a follow up?

Lyndsey Layton: No, thank you.

Lola Russell: Thank you, next question please.

Operator: The next question is from Jane Norman from Congressional Quarterly.

Jane Norman: Thank you for taking my question. If you could explain to me why there hasn't been new data since 1999 and you mentioned going forward you will be diving deeper into these numbers is there going to be more updating of these overall estimates?

Chris Braden: So, just a little explanation of these estimates. You know the data that were generated in order for us to be able to accomplish these analysis were actually developed and collected over time, over years. Since the publication of the 1999 estimates, we have worked to improve our surveillance, improve our laboratory methods for instance in order to detect norovirus, improve our methods and the data that was used to generate these estimates were actually data that were collected over a number of years as I explained before. They center around 2006. But this is not something that we can just generate year by year. We will continue to improve our methods, improve our laboratory surveillance, improve our case surveillance in order to provide more precise estimates in the future, but it is not something that we can do year by year.

Jane Norman: So we shouldn't expect to see estimates each year on a regular timeline?

Chris Braden: So what we do plan to do is take these estimates and, again as I said before, provide additional analyses that can focus priorities on certain food commodities or at certain points in the process of foods.

Jane Norman: Thank you.

Lola Russell: Okay next question please?

Operator: Our next question comes from Betsy Mckay with the Wall Street Journal.

Betsy McKay: Thank you. I was wondering if we could talk about specific pathogens for a second. Are there differences from 1999 and 2011 in numbers of prevalence of particular pathogens or a change that surprised you? I know we can't compare the two, but is there a pathogen out there that you have identified as being more frequent than another?

Chris Braden: Yes, thanks for the question. So, there are differences in the proportion of illnesses, hospitalizations, and deaths, you know, on a pathogen-by-pathogen basis when you look at this estimate versus the Mead estimate. But again, I emphasize a lot of that difference is because of the data that we have been able to acquire that would give us more precise estimates by pathogen. But in general, when we talk about the seven pathogens that are responsible for the most illnesses, hospitalizations and deaths, those are the pathogens that we saw also in Mead being major pathogens.

Betsy McKay: Okay. Can I ask a quick follow up?

Lola Russell: Sure go ahead.

Betsy McKay: This is based on the earlier question about the number of deaths. Um, I wondered if you could explain the difference to us in a little more detail I mean, I always thought it clear when someone dies, what they die of. If there is not a decline, why have we arrived at a different number?

Chris Braden: I'm going to refer your question to the first author of the paper who may be best able to explain that, Dr. Scallen.

Elaine Scallen: There seems like two parts of your question and I'll address the one about the data sources. We used slightly different data sources for the current estimates than were used in 1999. This was based on discussions with experts plus ended up with a lower estimated number of acute gastroenteritis hospitalization, or deaths, excuse me. And you also asked about the question about when someone dies we know what they die of. Would you like to address that one?

Chris Braden: What happens with death certificates where a lot of this data is coming from is that people take, um, a best guess at the cause of illness and they may have several. And so some of this the issues around using death certificate data can be quite complicated.

Betsy McKay: Okay.

Lola Russell: Next question please.

Operator: Our next question is from Lynne Terry with the Oregonian, your line is open.

Lynne Terry: Hi, thank you for taking my call. Um, question for you, you said that the data between 1999 and the present data is not comparable. But we've seen a 20% drop in pathogens in the past ten years could you talk about that?

Chris Braden: Sure. I want to emphasize when we are talking about trends and the percent decrease, that is not something that is part of these reports. What we're saying is we can't use the two reports to determine trends. But what can we use to determine trends? There is a whole different analysis that uses surveillance from what we call the FoodNet surveillance system. And if people are not familiar with that, that is the surveillance system in ten sites where they actively collect the identification of certain foodborne pathogens in all the clinical laboratories in their sites. And there are sites in ten states around the United States. It covers 15% of the U.S. Population. And we have done that kind of proactive surveillance since 1996. And that surveillance system was established in order to be able to track trends over time. Each April, in the MMWR publication, we come out with the preliminary FoodNet data for the preceding year. And each April, we have a graphic in there that shows on a line graph by pathogen, that is under FoodNet surveillance what the trend is. If we put that information kind of together, for the major pathogens under surveillance in FoodNet we see about a 20% decline over the past decade. Now, most of that decline occurred early in the past ten years and has been more stable in the last several years.

Lynne Terry: But these are tracking known pathogens correct? So, the overall figures could be quite different?

Chris Braden: Yes, so those pathogens are basically nine pathogens tracked in FoodNet, of course, of the known pathogen in these reports, we look at 31. And so, of course the FoodNet data would feed into the overall estimates for the nine pathogens, but other surveillance systems and other sources of data were used for the other pathogens in the 31 known pathogens included in the report. Is that clear?

Lynne Terry: Yeah, what are those nine pathogens are that's the seven that you listed?

Chris Braden: No, they are actually, there is an overlap there. But they are not exactly the same.

Lynne Terry: Okay, I don't want to get too tedious for everyone else.

Chris Braden: Okay all right. So, obviously what I would refer you to is MMWR, April edition, you can go to the CDC.gov website for that report. It is called FoodNet Preliminary Data. It gives you all the trend data and information on FoodNet.

Lynne Terry: Thank you.

Lola Russell: Operator do we have anyone else in cue?

Operator: We have no one else at this time. If you would like to ask a question please press star one and record your name and affiliation at this time. We do have a question from Brenda Goodman from WebMD, your line is open.

Brenda Goodman: I wanted to ask how confident you might be that you accounted for underreporting for foodborne illnesses. But that is something that we have always thought about if the numbers are likely to be low.

Chris Braden: Yes, you are correct in that the estimates do include, you know, a correction for under diagnosis and under reporting. As you can well imagine, not everybody who becomes ill sees their doctor and even when they do, not everyone has tests done to determine what is causing the illness. So that is taken into account in the estimates. However, one could look at the methods for these estimates and determine that our methods were quite conservative and in reality for a number of these pathogens, it could be more than we've estimated. And that has a couple of consequences. Number one, it would be more illnesses due to known pathogens if we were actually overly conservative in our estimates, but number two, we're looking at a total pie of all gastrointestinal illnesses out there when we look at that unspecified agents category. And so if the number of known pathogen illnesses were to go up that unspecified agent category number would come down. There is a balance there too.

Brenda Goodman: Thank you.

Operator: Our next question comes from Lisa Schnirring with CIDRAP News, your line is open.

Lisa Schnirring: Thank you for taking our questions. On the unspecified agent report, could you tell me which the last time it was that the CDC looked at that and also the follow up question was is there anything in that report that jumped out at you or surprised you? Thank you.

Chris Braden: In this comprehensive analysis, the one other time that we looked at unspecified agents was in the 1999 estimates that we called the Mead estimates for the first author, Dr. Mead. So, since then, there have been, you know, again, better surveillance, better laboratory methods and so forth for the known pathogens and actually, you know, we have looked for, you know, these unknown pathogens. That would be part of that unspecified agent category and there have been some discoveries made since the last report but not discoveries that we would say would lead to a huge proportion of this unspecified illnesses. So we do know somewhat more about that category. There is a lot more that we really need to learn.

Lisa Schnirring: Thank you.

Operator: Our next question is from Lea Winerman with PBS News Hour, your line is open.

Lea Winerman: Thank you for taking our questions. You have mentioned a few times that the methods this time were more precise, and I was wondering if you could just expand on that a little bit and explain a little bit about how the methods changed between the Mead report and this one.

Chris Braden: Sure. I'm going to give you two examples and these two actually could account for a lot of the differences that we are seeing. So, if you recall I mentioned before, about the fact that you know a lot of this is based upon how many people get acute gastrointestinal illnesses in the United States in any one year, and when we try to assess that number, what we do is again, through FoodNet sites, do surveys of people. You could have been one of them and could be one of them in the future where we call and ask a number of questions about, you know, have you had these symptoms in the past month? And then we define what constitutes a gastrointestinal illness. When the Mead paper was published, we had one set, basically, of survey data like that. It was about 9,000 interviews. And we used that data to come up with the total number of gastrointestinal illnesses out there at the time. And that was the basis for some of the estimates in the Mead report. This time around we have about five times more than that of that kind of data. More than 48,000 interviews have been done. And so when we look at that data, we actually see that the number of people out there with acute gastrointestinal illnesses is less than what we thought before and estimated before. For instance, in the Mead paper, they estimated that about eight in ten people became ill with acute gastroenteritis each year. In this new estimate it is about six in ten. That is a big difference when you are looking at tens of millions of illnesses and just that difference alone would account for some of the, quite a bit of the difference we are seeing between Mead and the 2011. Did you follow me on those? That example?

Lea Winerman: Yes.

Chris Braden: Okay, so the other example has to do with norovirus. And norovirus is a difficult one to estimate because clinical laboratories can't diagnose a norovirus infection. There is no commercial test for norovirus. However, our public health laboratories have become much better at identifying norovirus, especially in outbreaks. And this is very important because the big question is how much of norovirus is foodborne? A lot of novovirus is due to person-to-person transmission. You hear about cruise ship outbreaks and you hear about nursing home outbreaks. Most of that is person-to-person transmission. Well, when we did the estimates in 1999, we estimated the portion of norovirus that was foodborne and at that point we thought it was 40%. When we looked at better data now with better diagnostics that we have at our public health laboratories, that comes down to 26%. That is again, a big difference that we have between Mead and the current estimates and I will stop there.

Lola Russell: Can we have one last question please?

Operator: It is a follow up with Lyndsey Layton of the Washington Post. Your line is reopened.

Lyndsey Layton: I want to ask the one question that I think most lay people would want to know from this new data and also from your knowledge of the Mead paper. And that is, is our food supply any safer today than it was 11 years ago? Can you take a stab at that?

Chris Braden: There has been a lot of advancements in food safety over these years. You know, we have seen outbreaks and we have seen our surveillance numbers having to do with for instance Listeria, E. Coli O157 and other pathogens and we can see where interventions have been put in place that have made a difference. For instance, we set a goal in 2000 to decrease E. Coli O157 illnesses in the United States by 50%. Last year's data in FoodNet said we reached that goal as a HealthPeople 2010 objective. So yes, there's been a number of interventions that have made a big difference. However, I would like to stress, still that one in six Americans each year get ill each year with foodborne illnesses in the United States, 125 – greater than 125,000 are hospitalized, the cost that is associated with these illnesses is in the billions of dollars, 3,000 died. There is more that we need to do.

Lola Russell: Thank you Lyndsey. As I said earlier, we have excellent resources with this data that is available on our website. That will provide you will a list of charts and details that will explain the differences that will explain the trends. So there is an excellent resources there at that foodborne burden link as well. We will also be posting a website so expect to receive a copy of that as well. Thank you so much for participating today.

Operator: That does conclude today's conference call. You may now disconnect and we thank you all for participating.

###
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES