Animal & Veterinary
Human Health Impact from Food-borne Disease
Dr. Fred Angulo
DR. ANGULO: As most of you saw in the risk assessments, much of the data that was provided in the risk assessment is through a new project that has been established at CDC which is called the Food-borne Disease Active Surveillance Network or FoodNet.
FoodNet is the primary food-borne disease component of CDC's emerging infections program. It was established in 1995 within the EIP sites. And it is a collaborative effort between the participating state health departments, U.S. Department of Agriculture Food Safety Inspection Service and FDA.
Oh, if you would like more information including the annual summaries and descriptions about FoodNet, there is a website that is available and we are happy to -- please let us know and we could provide the website to you at the end of the talk, also.
In 1999, the FoodNet population catchment area is 28 million. We are happy to announce that we are adding a ninth site. The ninth site will be in the west. And the actual site will be announced tomorrow and will bring our population up to over 30 million persons within the population catchment area.
The primary objectives of FoodNet are to determine more precisely and to monitor better the burden of food-borne diseases and to determine -- secondarily, determine the proportion of food-borne diseases which are attributable to specific foods. We, therefore, see and are pleased to play a role in risk assessments because we see the data generated within FoodNet as being data essential to doing precise risk assessments.
Equally important, we see FoodNet as a platform to monitor the reduction of food-borne illness that might occur when interventions have been put in place. We work very closely with the USDA FSIS to monitor the pathogen -- monitor success through the pathogen reduction and HACCP plan.
FoodNet conducts active surveillance on seven bacterial pathogens, one of which is Campylobacter. This active surveillance for Campylobacter is conducted by visiting at least monthly, but in most cases, weekly each of the clinical laboratories within the population catchment area.
Presently, there is about 350 clinical laboratories. These laboratories receive a stool sample from a person who is ill enough to seek care, a physician concerned enough to gather a stool sample and send it to the clinical laboratory, laboratory test, and then we ascertain the cases actively from those clinical laboratories.
This just shows the type of data that is available. This is from the annual report which is on the web. And it shows the seasonal distribution of culture-confirmed cases for the foremost commonly identified bacterial pathogens, Campylobacter being the most commonly identified culture-confirmed illness each month of the year. And you also see the marked seasonal distribution of Campylobacter which has been discussed.
Since FoodNet has been in place since 1996, we can begin to assess trends in food-borne illness. And this is some of the exciting data that we published in March of this year in the MMWR, and also we published in collaboration with FSIS and report to Congress.
And it is very subtle. You see the Salmonella in the second bar there declined a very small proportion from 1996 through 1998. But because we have serotype-specific data, we can explore within specific serotypes declines. And to show you the amount of precision that is within FoodNet, that small decline of Salmonella which we detected in the first three years of the project we believe is -- and particularly when considering the reduction in Salmonella that is present in the slaughter sampling through the pathogen reduction plan, they correlate -- those declines of Salmonella correlate so closely that we believe this decline in Salmonella is attributed to the -- in large part to improved safety of meat and poultry.
Equally exciting is a remarkable decline in Campylobacter and poultry-confirmed illness, most prevalent in California. This points out that had the risk assessment been done based on 1997 data, there would have been 25 percent more illness. It also suggests that in 1999, because this trend is appearing to continue into 1999, the primary report of that trend will be published in the March 2000 MMWR.
But the trend appears to be continuing in 1999. And had the risk assessment been done on 1999 data, there would have been probably an order of that magnitude decline in the outcome identified in the risk assessment.
Besides ascertaining culture-confirmed cases, we ascertain -- the FoodNet personnel ascertain outcomes of those patients which include whether the patients were hospitalized or not. And there has been some misstatements at the meeting that Campylobacter does not frequently result in hospitalization.
In fact about ten percent -- there is actually 12 percent of persons with culture-confirmed Campylobacter infections are hospitalized. So relatively a large burden of illness. We also ascertain deaths.
But all that -- this active case finding within FoodNet is -- although giving enough precision to monitor trends over time which is quite exciting, the enhancements to the FoodNet are really what are novel. And these enhancements are the recognition that the burden of illness caused by food-borne diseases, that the numbers of people that are sick in the community, illness in the general community is a reflection.
When we do surveillance only based upon culture-confirmed cases at the top of the pyramid, we miss all the people who may be seeking care, but don't get a culture collected or they get a culture collected, but it is not tested for the pathogen that caused their illness, etcetera. Any break in the chain of these events will cause the person to not be culture-confirmed.
Well, the beauty of FoodNet is that we are doing surveys and studies in all of these chains of events to identify what the loss in reporting is of each of the steps and that these surveys are very robust relatively. In terms of, for instance -- it's on the next slide.
For instance, we are doing a population survey. The population survey is in its third cycle. In each of the cycles, there has been almost 10,000 persons interviewed. We are interviewing 150 people per month in each of the sites and with nine sites coming on-line. Over 1,000 people are interviewed a month.
Those people are interviewed and asked had they had diarrhea in the last week -- excuse me, in the last month. If they had diarrhea, they are asked if they submitted a stool sample, etcetera. So we begin to get information about the prevalence of diarrhea in the population and people seeking care, etcetera, and to begin to understand what is happening at the bottom of the pyramid.
Equally robust is a survey of physicians that we did. We surveyed 5,000 physicians in the FoodNet sites which was close to one-third of all physicians in private -- that handled patients that see -- that see patients with a diarrheal illness. And although the response rate from the physicians survey was only 67 percent, it is a remarkably high response rate for a physician survey. And we have information from the physician survey about how frequently physicians culture patients who seek care.
And we also survey on an annual basis, but in detail, every two years each of the laboratories within the FoodNet sites to see whether their culture practices are changing from year-to-year.
The exciting piece of this, besides the FoodNet being used as a platform to monitor -- actively monitor in a consistent and comprehensive manner culture-confirmed illness, we can estimate what is happening at the bottom of the pyramid. And this was published in September of 1999 -- the first author is Paul Meade -- in CDC's Emerging Infectious Disease Journal which is available on-line and copies of which of this article I have at the table at the back.
These are the new estimates and we believe the most precise, to-date estimates of food-borne illness in the United States. We believe that there are 76 million infections each year in food-borne illnesses. These are infections due to contaminated foods.
And so previous statements of a one in a ten risk of food-borne illness appeared to be -- well, we don't -- we perceive a greater risk than had previously stated. And it also points out the numbers of hospitalizations. This is not all mild illness, although much of it is self-limiting illness, and the numbers of deaths that we attribute to food.
And these estimates actually demonstrate that the risk of cases is somewhat higher than previous risks, but the number of deaths are lower than previous risks -- previous estimates.
This is itemized in the paper that I mentioned that is available at the back. But these are the numbers of -- these are the most common food-borne illnesses with a known etiology. So the estimate of 76 million includes even an estimate for, we believe, food-borne illness that we have -- public health has not even identified the pathogen yet. So about two-thirds of the 76 million infections are actually unidentified pathogens.
But then amongst the known pathogens, these are the ten -- these are the most common known pathogens just to point out that in terms of illness amongst the known pathogens, Campylobacter causes 14 percent of the food-borne illness amongst the known pathogens. Salmonella accounts for less, ten percent.
But then as you look at the number of deaths to point out the -- to reiterate the severity of Salmonella infections, Salmonella accounts for 30 percent of the deaths associated with food-borne diseases. And Campylobacter, although not an insignificant number -- 100 deaths are attributed to Campylobacter each year, 99 deaths. That is only five percent of the total deaths.
Also interesting, just an aside, is these are the most commonly identified food-borne pathogens. And so germane to our discussion here is you can begin to say, well, which of these pathogens can carry resistant determinants through the food supply. And, therefore, it points out the need to focus on Campylobacter and Salmonella in particular, and also perhaps some other pathogens.
But Salmonella and Campylobacter are clearly the ones to monitor closely for the transmission of resistant determinants through the food supply because we believe that Campylobacter and Salmonella is seldom transmitted person to person and is largely transmitted through the food supply.
So if anybody would like additional information about FoodNet, the web page is available. And please take a moment, if you like, to pick up the article published in the Emerging Infectious Diseases which provides the estimates of food-borne illness in the United States.
DR. STERNER: Are there questions for Dr. Angulo Thank you, Fred. Our next presenter is going to deal with food-borne resistant pathogens. Dr. Glenn Morris graduated from Rice University in Houston, Texas with a bachelor of arts in 1973. He received his M.D. degree, magna cum laude in 1997.
And from 1989 until the present, he has been employed at the University of Maryland Medical School where he currently serves as the Chief of the Infectious Diseases Service and is the head of the Department of Epidemiology and Medicine. Dr. Morris.