By: James Peacock
Towards the beginning of each year, the CDC publishes the data that it has collected through the FoodNet system over the previous year. While this data is only preliminary, it still allows the previous year to be compared to other years. Through this process, many conclusions can be drawn regarding the effectiveness of plans and procedures put in place to help deter the occurrence of foodborne illnesses. When the CDC publishes this report each year, it also releases the final data from 2 years prior. Therefore, when the CDC report was published earlier this year, we received the preliminary data for 2016, and the final data for 2015. Both offer insights to the foodborne illnesses taking place in the United States, so let’s take a look at some of the data included in the report.
The FoodNet System
The FoodNet System was established in July 1995 by the CDC, and represents a collaborative effort between the Centers for Disease Control and Prevention (CDC), the United States Department of Agriculture’s Food Safety and Inspection Service (USDA-FSIS, or FSIS), and the Food and Drug Administration (FDA). FoodNet also gets data from ten state health departments, including Connecticut, Georgia, Maryland, Minnesota, New Mexico, Oregon, Tennessee, New York, Colorado, and California. Most of the states listed pool data from all counties in the state, but California, Colorado, and New York only report data from select counties. With the combined resources of all of these organizations, the FoodNet System is able to boast a surveillance area that covers more than 48 million people in the United States. FoodNet originally tracked only Campylobacter, Listeria, Salmonella, Shigella, Vibrio, Yersinia, and STEC O157 infections. In 1997, Cryptosporidium and Cyclospora were added. STEC non-O157 infections were added in 2000. At first, only cases of illness confirmed by a bacterial culture were added to the data, but due to the recent increase in CIDT diagnosis techniques, those tests have been included in FoodNet data.
2015 Final Data
Out of the 48 million people covered in the FoodNet System, the final data from 2015 reveals that 20,098 laboratory-confirmed infections took place. These infections led to 4,598 hospitalizations and 77 deaths. Out of the 20,000 infections, only 925 were connected to outbreaks. The nine different pathogens that are tracked are depicted in the chart below.
|Pathogen||Number of Infections||Incidence Rate|
These incidence rates are also calculated for various age groups, and these numbers show that children under five have the highest incidence rates for all of the pathogens except for Vibrio, Listeria, and Cyclospora. The occurrence of hemolytic uremic syndrome (HUS) is tracked for children under the age of 18. The reported numbers are for one year prior to the data recorded. In the 2015 final report, it was reported that 61 cases of HUS took place in 2014. More than half of those cases took place in children under the age of 5. Infections are also tracked by the month that they take place in. Campylobacter, Cryptosporidium, Cyclospora, Salmonella, STEC O157, Vibrio, and STEC non-O157 all peaked in the summer, while Listeria, Shigella, and Yersinia infections remained fairly constant throughout the year. Some pathogens, like Salmonella and E. coli, are made up of several different serotypes that cause virtually the same illness but differ slightly in structure. Salmonella has over 2000 known serotypes, many of which are exceedingly rare. The ten most common Salmonella serotypes are depicted in the table below.
|Salmonella Serotype||Number of Confirmed Infections|
|I 4,, 12:i:||497|
The CDC will also usually compare the infection numbers and incidence rates with previous FoodNet data years, in order to ascertain whether a particular pathogen is being effectively combated. In the 2015 data, the numbers were compared to three date ranges: 2012-2014, 2006-2008, and 1996-1998. All pathogens were occurring 30% less than in the 1996-1998 period. The comparison with 2006-2008 yielded more varied results, but still saw decreases for Campylobacter (26%), Listeria (45%), Shigella (48%), STEC O157 (44%), and Yersinia (59%). Vibrio (114%) and Cryptosporidium (68%) were the only ones to see an increase. Salmonella rates remained mostly constant. When compared to numbers from 2012-2014, an increase in incidence was seen in STEC non-O157 (41%) and Cryptosporidium (60%). Yersinia, Vibrio, STEC O157, Shigella, Salmonella, Listeria, and Campylobacter all saw little to no change in incidence.
2016 Preliminary Data
In the FoodNet system’s preliminary data from 2016, the CDC was able to track 24,029 cases of foodborne illness. These illnesses were connected to 5,512 hospitalizations and 98 deaths. More so in this year than in other years, the CDC needed to include data not only gleaned from bacterial cultures but also data from CIDTs. Culture-Independent Diagnostic Tests (CIDTs) offer a different way of testing for foodborne pathogen. This new method can lead to a much faster diagnosis, but because no culture is generated, only more limited data is gathered. This can hinder the investigative process and make it more difficult to establish patterns and incidence rates. The CDC’s FoodNet Fast system estimates that only 5% of reported illnesses were connected to outbreaks. The nine different pathogen tracked by the FoodNet system are depicted in the table below.
|Pathogen||Number of Infections||Incidence Rate|
The prevalence of CIDT usage increased for Campylobacter, Yersinia, STEC, Shigella, Vibrio, and Salmonella. Across these pathogens, the overall increase in CIDT usage was 144% when compared to 2013-2015. Out of the 2016 data, most pathogens, except for Shigella and Yersinia, showed some sort of peak rate of incidence during the summer months. Cyclospora, the pathogen with the greatest month-to-month change, caused 40% of its 2016 cases in the month of July alone. The FoodNet system also tracks the occurrence of hemolytic uremic syndrome (HUS) in children under the age of 18. Just like the 2015 data, the HUS data is for the previous year. The 2016 report states that 62 cases of HUS were reported in 2015. More than half of these cases occurred in children under five. There was no significant difference in incidence when compared to 2012-2014 data.
|Salmonella Serotype||Number of Confirmed Infections|
|I 4,, 12:i:||432|
The 2016 preliminary data only offers comparisons for incidence rates in two date ranges: 2006-2008 and 2013-2015. When compared to incidence rates of culture confirmed cases of illness between 2006 and 2008, the 2016 rate of infections showed no significant change for Campylobacter, Listeria, Salmonella, Shigella, STEC O157, and Yersinia. Vibrio showed about a 50% increase, while Cryptosporidium had an increase of upwards of 90%. Interestingly, when CIDT –positive infections are included, a comparison over the same time period will produce very different results. Campylobacter, instead of having no significant change, shows about a 40% increase in incidence when CIDT is included. The increase in Vibrio cases is about 20% higher when CIDT is included. Yersinia cases shift from no significant change to more than a 50% increase.
When compared to infection rates between 2013 and 2015, only Campylobacter, Listeria, Shigella, Salmonella, and Vibrio do not show some level of significant change. Cryptosporidium (~50%), Yersinia (~30%), and STEC (~20%) all showed an increase in incidence. Again, these numbers change when CIDTs are included. Shigella, instead of hardly any change, shows an increase of about 20%. Vibrio also moves from no change to some change, increasing by about 15% when CIDT is included. STEC nearly doubles in incidence because of CIDT, from about 20% to nearly a 50% increase. Yersinia also sees a huge difference, as the CIDT incidence change is more than double (~80%) than what occurs simply from culture confirmation (~30%).
Limitations and Conclusions
This data allows health professionals to take a look at trends in the foodborne illnesses that affect society. By looking at the rate at which they occur, they can draw conclusions about whether or not implemented methods of illness reduction are having the desired effects. There are, though a couple of limitations that are attached to these data sets. Highlighted in both the 2015 and 2016 reports, the first of these limitations has to do with changes in “diagnostic landscape”. Perhaps best seen in the rise of CIDT, the way that doctors diagnose and report illnesses has been known to change over time, which can lead to issues with tracking trends. The other problems that the rise of CIDTs presents, mainly the lack of detailed subtyping and pathogen information, also contribute to difficulty establishing trends in foodborne illnesses. The other major limitation that the FoodNet data is subject to is the number of people who seek health care for their illnesses. For instance, the number of Salmonella infections confirmed and reported by FoodNet in 2016 totaled only 7554, while CDC estimates place the number of annual Salmonella infections at over 1 million. This huge disparity is due to two reasons. First, the entire population is not represented in the FoodNet data. If you recall from earlier, the system only covers about 48 million people. For the purposes of collecting and analyzing data to spot trends, this is plenty of people, but it can lead to some variance in the number of illnesses reported. The second, and more important reason, is that not everyone seeks and receives care for their foodborne illnesses. In keeping with Salmonella as our example, out of the more than one million cases each year, the CDC states that in most of them, the person affected will recover on their own within a week. Because it is impossible to predict just how many people will seek medical attention for their illnesses, disparities can appear in the data. Even with these limitations, the FoodNet system is able to compile more than enough data to look at trends in foodborne illnesses. It remains the most detailed look at foodborne illnesses in the United States, and has helped health officials make smarter choices about combating illnesses. With the help of systems like this, health officials have been able to reduce foodborne illnesses by 30% across the board. The FoodNet system provides the basis for forming new food safety policies, and the vast majority of outbreak investigations rely on it to provide insight and point towards spikes in the occurrence of illnesses.