Interesting article from the University of Minnesota Center for Infectious Disease Research and Policy about the US Government improving methods for sifting through data to estimate which foods are contributing to outbreaks. They focused the report on the four main outbreaks: Salmonella, Escherichia coli O157, Listeria monocytogenes, and Campylobacter.
It’s great that the government is focusing on these issues and trying to use data to draw correlations to try to minimize these outbreaks, but what I found most intriguing were the breakdowns of the various food groups by outbreak. Here’s what they found:
- Salmonella: seeded vegetables (18%), eggs (12%), fruits (12%), chicken (10%), sprouts (8%), beef (9%), and pork (8%)
- E coli O157: beef, 46%; vegetable row crops, 36%
- Campylobacter: dairy foods, 66%; chicken, 8%
- Listeria: fruits, 50%; dairy, 31%
Salmonella spans quite a variety of food groups and dramatically more than the others. The full article is below:
US government agencies today reported on what they billed as an improved method for sifting foodborne disease outbreak data to estimate the contributions of different foods to outbreaks sparked by four common types of foodborne bacteria.
The report, focusing on outbreaks involvingSalmonella, Escherichia coli O157, Listeria monocytogenes, and Campylobacter, estimates the percentages of such outbreaks that were related to various foods from 1998 to 2012.
It was developed by the Interagency Food Safety Analytics Collaboration (IFSAC), a partnership of the Centers for Disease Control and Prevention (CDC), the Food and Drug Administration, and the Department of Agriculture’s Food Safety and Inspection Service, according to a CDC statement today.
In general, the analysis found that Salmonella outbreaks were caused by a wide range of food categories, with no particular one predominant, whereas just two food categories were dominant contributors to outbreaks of each of the other three pathogens.
“The new estimates, combined with other data, may shape agency priorities and support the development of regulations and performance standards and measures, among other activities,” the CDC statement said. “The recently developed method employs new food categories that align with categories used to regulate food products and emphasizes more recent outbreak data.”
Four leading pathogens
The CDC estimates that the four pathogens cause 1.9 million cases of foodborne illness each year.
The 12-page report says the four pathogens were blamed for 2,655 foodborne outbreaks between 1998 and 2012, but the study focused only on 952 outbreaks for which the implicated food or foods could be assigned to a single food category. Of the 952 outbreaks, 597 were caused by Salmonella, 170 by E coli O157, 161 by Campylobacter, and 24 by Listeria.
The report describes various statistical methods used to refine the estimates, including steps to smooth variations in outbreak size and reduce the influence of outliers. In the interest of timeliness, the model gives greater weight to data from 2008 through 2012 than data from the earlier years. The agencies divided foods into 17 categories.
Among principal findings, the authors found that seven food categories accounted for 77% ofSalmonella cases: seeded vegetables (18%), eggs (12%), fruits (12%), chicken (10%), sprouts (8%), beef (9%), and pork (8%).
In contrast, for each of the other three pathogens, just two food categories accounted for the majority of cases, as follows:
- E coli O157: beef, 46%; vegetable row crops, 36%
- Campylobacter: dairy foods, 66%; chicken, 8%
- Listeria: fruits, 50%; dairy, 31%
The CDC cautions, however, that the Listeria data were sparse, leading to considerable statistical uncertainty (wide confidence intervals), and the 50% estimate for fruit reflects the impact of a large cantaloupe-related outbreak in 2011.
Toward greater consistency
The report acknowledges a number of limitations, including that it covers only foodborne disease outbreak cases, not sporadic cases.
It states, however, “Our novel approach produces better estimated attribution percentages than those based solely on the observed numbers of outbreaks and outbreak illnesses, and can be used to produce new estimates when outbreak data are updated.” In addition, it says that having consensus on one analytic approach may make for greater consistency in interpretation of estimates across different federal agencies.
The CDC said IFSAC was scheduled to describe its methods at a public meeting in Washington, DC, today, as part of federal efforts to improve foodborne illness source attribution.