Antimicrobial resistance surveillance in Canada: Retail food highlights, 2003-2012

Posted on November 8, 2014 by Douglas Powell

The Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) is a collaborative, integrated program designed to track antimicrobial resistance (AMR) among enteric bacteria isolated from various livestock commodities along the food-producing continuum (“farm to fork”) and in humans.

Objective: To provide a summary of the prevalence and trends in AMR among select bacteria isolated from raw, fresh chicken, pork, and beef in 2012 at the retail food level and to link these data with other findings from CIPARS.

Methods: Meat samples were collected from randomly selected geographic areas across Canada weighted by population for subsequent isolation of bacteria and interpretation of the associated AMR profiles. Salmonella, Campylobacter and generic Escherichia coli (E. coli) were tested in chicken, and E. coli was tested in beef and pork. Data were analyzed for 2012 and temporal and regional trends were examined between 2003 and 2012 by province/region.

Results: Overall, resistance levels to Salmonella in retail chicken varied widely by region and year. For example, ceftiofur resistance to Salmonella in retail chicken was significantly lower in 2012 than in 2004 in Ontario and in Québec; however among all regions sampled, resistance to Campylobacter in retail chicken was relatively low in 2012 (<16%) with the exception of tetracycline resistance. In 2012, ciprofloxacin resistance to Campylobacter in chicken declined in British Columbia but significantly increased in Ontario, compared to 2011. In 2012, β-lactam resistance to E. coli in retail beef remained low (≤1%) and was also relatively low comparable to previous years in pork.

Conclusion: In Canada, as is the case worldwide, there is evidence of resistance to medically important antimicrobials among bacteria from retail meats. Resistance among organisms isolated from poultry, beef, and pork at the retail food level is characterized by wide variation over time and across different regions.

Public Health Agency of Canada, CCDR, Volume 40 S-2

B. Avery, E. Parmley, R. Reid-Smith, D. Daignault, R. Finley, R. Irwin

http://www.phac-aspc.gc.ca/publicat/ccdr-rmtc/14vol40/dr-rm40s-2/dr-rm40s-2-surv-4-eng.php

So what should be done about it? Listeria at deli

Posted on November 4, 2014 by Douglas Powell

Postprocessing contamination in processing plants has historically been a significant source of Listeria monocytogenes in ready-to-eat delicatessen meats, and therefore a major cause of human listeriosis cases and outbreaks. Recent risk assessments suggest that a majority of human listeriosis cases linked to consumption of contaminated deli meats may be due to L. monocytogenes contamination that occurs at the retail level. To better understand the ecology and transmission of Listeria spp. in retail delicatessens, food and nonfood contact surfaces were tested for L. monocytogenes and other Listeria spp. in a longitudinal study conducted in 30 retail delis in three U.S. states. In phase I of the study, seven sponge samples were collected monthly for 3 months in 15 delis (5 delis per state) prior to start of daily operation; in phase II, 28 food contact and nonfood contact sites were sampled in each of 30 delis during daily operation for 6 months. Among the 314 samples collected during phase I, 6.8% were positive forL. monocytogenes. Among 4,503 samples collected during phase II, 9.5% were positive for L. monocytogenes; 9 of 30 delis showed low L. monocytogenes prevalence (<1%) for all surfaces. A total of 245 Listeria spp. isolates, including 184 Listeria innocua, 48 Listeria seeligeri, and 13 Listeria welshimeri were characterized. Pulsed-field gel electrophoresis (PFGE) was used to characterize 446 L. monocytogenes isolates. PFGE showed that for 12 of 30 delis, one or more PFGE types were isolated on at least three separate occasions, providing evidence for persistence of a given L. monocytogenes subtype in the delis. For some delis, PFGE patterns for isolates from nonfood contact surfaces were distinct from patterns for occasional food contact surface isolates, suggesting limited cross-contamination between these sites in some delis. This study provides longitudinal data on L. monocytogenes contamination patterns in retail delis, which should facilitate further development of control strategies in retail delis.

Listeria monocytogenes and Listeria spp. contamination patterns in retail delicatessen establishments in three U.S. states

Simmons, Courtenay; Stasiewicz, Matthew J.; Wright, Emily; Warchocki, Steven; Roof, Sherry; Kause, Janell R.; Bauer, Nathan; Ibrahim, Salam; Wiedmann, Martin; Oliver, Haley F.

Journal of Food Protection®, Number 11, November 2014, pp. 1844-2003, pp. 1929-1939(11); DOI: http://dx.doi.org/10.4315/0362-028X.JFP-14-183

http://www.ingentaconnect.com/content/iafp/jfp/2014/00000077/00000011/art00012

Listeria species have been isolated from diverse environments, often at considerable prevalence, and are known to persist in food processing facilities. The presence of Listeria spp. has been suggested to be a marker for Listeria monocytogenes contamination. Therefore, a study was conducted to (i) determine the prevalence and diversity ofListeria spp. in produce production and natural environments and (ii) identify geographical and/or meteorological factors that affect the isolation of Listeria spp. in these environments. These data were also used to evaluate Listeriaspp. as index organisms for L. monocytogenes in produce production environments. Environmental samples collected from produce production (n = 588) and natural (n = 734) environments in New York State were microbiologically analyzed to detect and isolate Listeria spp. The prevalence of Listeria spp. was approximately 33 and 34% for samples obtained from natural environments and produce production, respectively. Coisolation of L. monocytogenes and at least one other species of Listeria in a given sample was recorded for 3 and 9% of samples from natural environments and produce production, respectively. Soil moisture and proximity to water and pastures were highly associated with isolation of Listeria spp. in produce production environments, while elevation, study site, and proximity to pastures were highly associated with isolation of Listeria spp. in natural environments, as determined by randomForest models. These data show that Listeria spp. were prevalent in both agricultural and nonagricultural environments and that geographical and meteorological factors associated with isolation of Listeria spp. were considerably different between the two environments.

Geographical and meteorological factors associated with isolation of Listeria species in New York state produce production and natural environments.

Chapin, Travis K.; Nightingale, Kendra K.; Worobo, Randy W.; Wiedmann, Martin; Strawn, Andlaura K.

Journal of Food Protection®, Number 11, November 2014, pp. 1844-2003, pp. 1919-1928(10); DOI: http://dx.doi.org/10.4315/0362-028X.JFP-14-132

http://www.ingentaconnect.com/content/iafp/jfp/2014/00000077/00000011/art00011

FoodNet Canada 2013 short report

Posted on September 22, 2014 by Douglas Powell

FoodNet Canada tracks illnesses of the gut, commonly known as food-poisoning, in Canadians and traces them back to their sources, such as food, water and animals. These data are analyzed to help determine which sources are causing the most illness among Canadians and helps us track illnesses and their causes over time.

In the 2013 surveillance year, FoodNet Canada was active in two areas: the Region of Waterloo Public Health, and the Fraser Health Authority of lower mainland British Columbia (BC). In each location, or “sentinel site”, enhanced human disease surveillance is performed in parallel with active surveillance of specific bacteria, viruses and parasites and the possible sources to which the ill may have been exposed.

The purpose of this report is to present the preliminary findings from the 2013 surveillance year in both sentinel sites. This report will be followed by a comprehensive annual report which will include more extensive analyses of temporal trends and subtyping information for an integrated perspective on enteric disease from exposure to illness. With eight years of data from two different sentinel sites, FoodNet Canada continues to provide important information on enteric disease in Canada. This information is essential to develop robust food and water safety policies in Canada.

In general, the incidence rates of reportable enteric diseases have decreased over the past seven years. In 2013, Campylobacter and Salmonella remain the most common causes of human enteric illness in the sentinel sites, and across Canada. Information gained from the exposure surveillance within FoodNet Canada (retail, farm, and water) provide insight into the potential sources and routes of exposure for both of these pathogens.
Campylobacter, Salmonella, and Listeria monocytogenes continue to be commonly found on skinless chicken breasts sold at retail in both sentinel sites, as well as on processed chicken products such as ground chicken and frozen chicken nuggets. Listeria monocytogenes has also consistently been found on ground beef, although at lower levels than in the retail chicken products.
Interestingly, all of the parasites and viruses that have been tested for were detected on leafy greens sold at retail in both sentinel sites. This information is shared with food safety partners in industry, Health Canada and the Canadian Food Inspection Agency in an ongoing effort to inform food safety policy. Because these pathogens were detected by molecular approaches, their ability to cause infection is unknown. Further research in this area would be helpful to estimate the risks to humans.
At the farm level, Campylobacter remains the most frequently detected pathogen in cattle manure and also appears to be common in turkeys. In broiler chickens, Salmonella is the most commonly detected enteric pathogen.
Campylobacter, Salmonella, and VTEC continue to be found in untreated surface water in both rural and urban areas, at freshwater beaches, larger and small reaches of the Grand River, and in irrigation canals and ditches in the two watersheds in the BC site.
Exposure to retail meat products remains a potential source of infection for human enteric illness. Other exposure sources, however, such as the farm environment and water, are also possible. Continued monitoring of human illness and the potential exposures is important to ensure the continued health and safety of Canadians.

1-in-8, sorta like the US; foodborne illness in Canada

Posted on August 19, 2014 by Douglas Powell

The Canadians were busy today, when most of them are off at the cottage.

The Public Health Agency of Canada has joined with the U.S. and now estimates that each year about 1 in 8 Canadians (4 million people) get sick from the food they eat (used to be 1-in-3, or 1-in4).

Four pathogens cause about 90% of the 1.6 million illnesses caused by known pathogens: Norovirus (1 million cases), Clostridium perfringens (177,000 cases), Campylobacter (145,000 cases) and nontyphoidal Salmonella (88,000 cases). These estimates are based on multiple complementary disease surveillance systems and the peer-reviewed literature.

Understanding the burden of foodborne illness is useful for decision-makers, supporting the development of food safety and public health interventions, for research and for consumer education. Future efforts will focus on estimating the number of foodborne hospitalizations and deaths, the economic cost of food-borne illness and the burden of water-borne illness in order to provide crucial information to support research, policy and action.

A guidance document, Weight of Evidence: Factors to Consider for Appropriate and Timely Action in a Foodborne Illness Outbreak Investigation was developed to assist federal government decision-makers weigh the scientific evidence collected during a foodborne illness outbreak investigation in order to inform risk mitigation actions.

The objective of the document is to provide guidance on how to weigh evidence collected during epidemiologic, laboratory and food safety investigations in a food-borne illness outbreak investigation, as part of an overall health risk assessment process carried out by Health Canada. This is a short summary of the document.

And, to highlight the Public Health Agency of Canada’s Foodborne Illness Outbreak Response Protocol (FIORP), the primary guidance document for investigations of multi-jurisdictional food-borne illness outbreaks in Canada.

Approach: The current version of the FIORP was developed in 2010 by the Public Health Agency of Canada following consultation with Health Canada, the Canadian Food Inspection Agency, and provincial and territorial stakeholders.

Results: The FIORP outlines guiding principles and operating procedures to enhance collaboration and coordination among multiple investigative partners in response to multi-jurisdictional food-borne illness outbreaks. It has provided guidance for the conduct of 22 such investigations led by the Public Health Agency of Canada’s Centre for Food-borne, Environmental and Zoonotic Infectious Diseases between 2011 and 2013. Furthermore, it has also served as a guide for the development of provincial protocols.

Conclusion: The timely and effective investigation of and response to multi-jurisdictional food-borne illness outbreaks in Canada is facilitated and enhanced by the FIORP.

Of course, none of these documents were peer-reviewed and published in scientific journals, so it’s just a lot of back-slapping.

Is it really foodborne illness? Is it? Factors contributing to decline in foodborne disease outbreak reports, United States

Posted on August 5, 2014 by Douglas Powell

The number of foodborne disease outbreaks reported in the United States declined substantially in 2009, when the surveillance system transitioned from reporting only foodborne disease outbreaks to reporting all enteric disease outbreaks. A 2013 survey found that some outbreaks that would have been previously reported as foodborne are now reported as having other transmission modes.

Since 1973, the Centers for Disease Control and Prevention (CDC) has collected data on foodborne disease outbreaks submitted by all states, the District of Columbia, and US territories through the Foodborne Disease Outbreak Surveillance System. In 2009, existing foodborne and waterborne disease outbreak surveillance systems were transitioned to an enhanced reporting platform, the National Outbreak Reporting System (NORS), which also collects reports of enteric disease outbreaks transmitted through person-to-person contact, contact with animals, environmental contamination, and indeterminate means (1). A new electronic reporting form and data entry interface were also introduced. In 2009, the number of reported foodborne disease outbreaks declined 32% compared with the mean of the preceding 5 years (2); the number also remained below the pre-2009 average during 2010–2012 (2,3). The decline was largely observed among outbreaks attributed to norovirus, which can be transmitted through many routes: in comparison, the number of outbreaks attributed to Salmonella spp., which is usually transmitted through food, remained relatively constant (1,2).

We considered 3 possible reasons for the decline in the number of reported foodborne disease outbreaks: 1) classification of some outbreaks that previously would have been reported as foodborne as caused by another modeof transmission; 2) technical issues associated with the introduction of the new system; and 3) staffing and resource limitations related to the influenza A(H1N1)pdm09 virus pandemic. Clarification of how these factors might have affected reporting would provide accurate conclusions about trends in foodborne disease outbreaks. In 2013, we conducted a survey to identify possible reasons for the decline in the number of foodborne disease outbreak reports that started in 2009.

In January 2013, we conducted a voluntary, anonymous, internet-based survey of public health officials who are responsible for entering foodborne disease outbreak data into NORS at US state and territory health departments. The survey contained 33 questions in multiple choice, rating scale, or text formats. The questions asked about reporting procedures that might influence data quality and completeness, challenges and practices when determining the mode of transmission for each outbreak, the usability of the online reporting interface, and resource limitations.

Of the 133 public health officials in 56 jurisdictions who received the link to the survey, 50 (38%) from 39 (70%) jurisdictions completed the survey in whole or in part. The denominator varied for different questions because of nonresponse and exclusion of responses when “don’t know” was selected. Also, some respondents had not used the previous reporting system. Most respondents (36/43, 84%) assigned a high priority to entering foodborne disease outbreak data, rating outbreak reporting as 4 or 5 on a scale of 1 to 5, where 1 indicated low priority and 5 high priority. Similarly, 38/47 (81%) of respondents reported that 90%–100% of foodborne disease outbreaks investigated in their health departments were entered into NORS; 5/47 (11%) reported entering <50% of outbreaks.

The survey included 1 question to determine whether respondents had experienced difficulties identifying the primary mode of transmission for some outbreaks and 1 question to understand which modes of transmission they found difficult to distinguish from foodborne transmission. Many respondents (35/47, 74%) reported sometimes having difficulties in determining an outbreak’s primary mode of transmission. More than half (26/47, 55%) of respondents reported that, since 2009, they had used the newly established category of indeterminate/other/unknown to report an outbreak for which the mode of transmission was not clear. Twenty (80%) of 25 respondents reported that they had experienced difficulty distinguishing between foodborne and person-to-person transmission. In comparison, determining whether an outbreak was caused by transmission of the infectious agent through food or by animal contact was a problem for only 6 of 21 (29%) respondents. Respondents who reported difficulty distinguishing between foodborne and another mode of transmission were asked if they would have reported the outbreak as foodborne to the previous reporting system (before 2009). Most respondents indicated that an outbreak was very likely or likely to have been reported to the previous system as a foodborne disease outbreak if there was a problem determining whether an outbreak was caused by foodborne or person-to-person transmission (15/20 respondents, 75%); by foodborne or environmental contamination (8/11 respondents, 73%); or if a specific mode of transmission could not be determined with confidence (13/19 respondents, 68%).

Regarding usability of the NORS reporting interface, most respondents (36/37, 97%) reported that usability of the new interface was the same as or better than that of the previous system. Most respondents (26/31, 84%), reported that technical issues with the NORS system did not prevent them from entering outbreak reports; only 2 respondents reported that >10% of outbreaks were not entered because of technical issues. When asked if their health departments experienced decreased resources available to work on foodborne disease outbreaks in 2009 while dealing with influenza A(H1N1)pdm09 virus, 19 (57%) of 35 respondents reported decreased resources for foodborne disease outbreak investigations; 14 (44%) of 32 respondents reported decreased resources for outbreak detection (e.g., laboratory capacity); and 14 (38%) of 37 respondents reported decreased resources for outbreak data entry and reporting.

Clarifying the factors that affect foodborne disease outbreak surveillance enables accurate interpretion of observed changes over time. The findings of this survey suggest that the large decline in the number of foodborne disease outbreaks reported in 2009 was likely a combined result of the following: 1) a surveillance artifact, in that some outbreaks previously reported as foodborne are now attributed to other modes of transmission; and 2) limited availability of resources to detect, investigate, and report foodborne disease outbreaks during the influenza A(H1N1)pdm09 pandemic. The total number of outbreaks reported increased after 2009 but remained below pre-2009 numbers, which suggests that the effect of the surveillance artifact is persistent and that outbreaks are now being more accurately categorized by mode of transmission. Although we hypothesized that technical issues with the new reporting interface might have affected reporting, this explanation appears less likely.

Limitations of the survey included the length of time between the transition to NORS and the survey, which meant that some survey participants who are current NORS users had not used the previous reporting system or worked on foodborne disease outbreaks in 2009. This limitation explains the low number of responses to survey questions that required knowledge of practices before 2009. Also, the overall response rate for the survey was low. Possible explanations include staffing and resource limitations, but the survey was voluntary and anonymous, and no follow-up efforts were made to determine reasons for the low response rate. In addition, because some jurisdictions have >1 reporting administrator, personnel in some jurisdictions may have compiled a single response. On the other hand, the experiences of some health departments that did not compile responses might have been overrepresented. Further, the survey was not designed to measure the proportion of reported outbreaks affected by the introduction of NORS. Other potential reasons for the decline in the number of reported foodborne disease outbreaks, such as resource limitations and loss of public health positions resulting from budget cuts during the recession (4), were not explored.

In summary, the results of this survey provide unique insights into the decline in the number of foodborne outbreak reports submitted in 2009 and thereafter. NORS provides more comprehensive surveillance of outbreaks and a better understanding of the epidemiology of pathogens with multiple transmission pathways (1). These findings may be useful to improve guidance and training for outbreak reporting, particularly in reporting of the mode of transmission when multiple pathways exist for a pathogen.

At the time of the survey, Dr Imanishi was an Epidemic Intelligence Service Officer with the Division of Foodborne, Waterborne, and Environmental Diseases of the National Center for Emerging and Zoonotic Infectious Diseases, CDC. Her research interests include epidemiology and control of foodborne diseases and zoonoses.

Centers for Disease Control and Prevention, Emerging Infectious Diseases, Volume 20, Number 9, September 2014, DOI: 10.3201/eid2009.140044

Maho Imanishi, Karunya Manikonda, Bhavini P. Murthy, and L. Hannah Gould

Prevalence, distribution, and diversity of Escherichia coli, Staphylococcus aureus, and Salmonella in kiwifruit orchards and processing plants

Posted on July 24, 2014 by Douglas Powell

The kid loves her kiwi (right, not exactly as shown).

I’ve started putting one in her daily lunch, and my wife eats them with skin on, and I just like them.

But are there microbial risks?

Feng et al. report in Foodborne Pathogens and Disease that the aim of the study was to investigate the prevalence, distribution, and diversity of three foodborne bacteria in kiwifruit orchards and processing plants. Fourteen kiwifruit orchards and two processing plants in Shaanxi province were visited for sampling in 2012. Fruit samples and environmental samples in orchards and plants were taken for isolation of Escherichia coli, Staphylococcus aureus, and Salmonella. All isolates were characterized by antimicrobial susceptibility testing and detection of virulence genes. Selected isolates were further examined by pulsed-field gel electrophoresis (PFGE) analysis. In total, 160 E. coli isolates and 14 S. aureus isolates were recovered from 407 samples from orchards and plants, while no Salmonella was recovered. E. coli isolates displayed resistance most frequently to streptomycin (65.6%), and S. aureus isolates displayed resistance most frequently to erythromycin (21.4%). Three E. coli isolates (1.9%) were positive for stx2 and two S. aureusisolates (14.3%) were positive for both seb and seh. Seventy-seven E. coli isolates and 14 S. aureus isolates were analyzed by PFGE. PFGE results showed that both E. coli and S. aureus isolates were diverse, and blades for slicing during the processing could be an important contamination source. This study could provide useful information for kiwifruit growers and industry to establish proper management practices that help minimize the chance of microbial contamination from farm to table.

Food-poisoning pathogen detection speeds up dramatically

Posted on July 18, 2014 by Douglas Powell

New York is on the front lines of detecting foodborne pathogen outbreaks, thanks to a partnership between public health scientists and Cornell researchers.

Members of the Cornell Food Safety Lab, led by food science professor Martin Wiedmann and research associate Henk den Bakker, are helping the New York State Department of Health (NYSDOH) harness the capabilities and cost efficiencies of next-generation DNA sequencing techniques to rapidly identify strains of salmonella and read the results in a way that could quicken responses to potential outbreaks.

Traditional methods of assessing bacteria samples submitted to public health laboratories, based on pulsed-field gel electrophoresis (PFGE), often do not deliver the level of precision needed to pinpoint specific strains of pathogens, their relationships to each other and whether they share a common origin – vital information when trying to trace the source of illness outbreaks.

For Salmonella enterica serovar Enteritidis, for instance, 85 percent of isolates can be classified into just five PFGE types, and 40 percent belong to one subtype in particular.

“There’s so little variation in the genome, and when there’s an outbreak, it’s almost impossible to differentiate using that method,” den Bakker said.

By sequencing all 4.5 million base pairs of the bacteria’s DNA using single nucleotide polymorphisms (SNPS) in a technique known as rapid whole-genome sequencing, scientists are able to get much more nuanced information.

“This kind of detailed information improves our ability to tell whether outbreaks are isolated, sporadic or part of a cluster, which allows for more thorough epidemiologic investigations,” said NYSDOH collaborator William J. Wolfgang.

The introduction of small, affordable, bench-top, whole-genome sequencing equipment has made it possible for clinical and public health labs to consider adding the technology to their arsenal. The NYSDOH’s Wadsworth Center in Albany, New York, was one of the first public health labs to make the investment – Cornell was able to provide the bioinformatics expertise to enable the lab to make sense of the data it would be collecting and to analyze it quickly.

Their proof of concept was published July 16 in the Centers of Disease Control journal Emerging Infectious Diseases in a paper that uses a case study of a salmonella outbreak in a long-term care facility to demonstrate how the technique could benefit public health labs.

In a regional collaboration, samples collected by the Connecticut Department of Public Health were sequenced by NYSDOH and analyzed by Cornell, and researchers discovered the outbreak was even larger than suspected. In addition to the seven residents identified in 2010 as being sickened in the outbreak, nine additional samples from patients in surrounding communities matched the main strain.

“This suggests a common contaminated source outside the long-term care facility. Knowledge of these cases at the time of the outbreak might have improved the chances of finding the outbreak source, which was never identified,” the researchers wrote in the paper.

The technique is already gaining traction in several other states through the GenomeTrakr initiative sponsored by the U.S. Food and Drug Administration (FDA), which is providing sequencing equipment, reagents, funds for personnel, and training on the equipment to seven State Public Health Laboratories including New York. In return, the public health labs upload raw sequence data to a centralized site for analysis. Any clusters that appear are reported back to local labs and epidemiologists, allowing for a quick, coordinated response.

The study was partly funded by a grant from the USDA Agriculture and Food Research Initiative, the Center for Food Safety and Applied Nutrition at the FDA and the NYSDOH Wadsworth Center).

I didn’t poop in public but camera says I did: acute gastroenteritis and video camera surveillance on a cruise ship

Posted on July 8, 2014 by Douglas Powell

Researchers in Norway report that a fecal accident was discovered in front of a passenger cabin of a cruise ship. After proper cleaning of the area the passenger was approached, but denied having any gastrointestinal symptoms. However, when confronted with surveillance camera evidence, she admitted having the accident and even bringing the towel stained with diarrhea back to the pool towels bin. She was isolated until the next port where she was disembarked.

Acute gastroenteritis (AGE) caused by Norovirus is very contagious and easily transmitted from person to person on cruise ships. The main purpose of isolation is to avoid public vomiting and fecal accidents. To quickly identify and isolate contagious passengers and crew and ensure their compliance are key elements in outbreak prevention and control, but this is difficult if ill persons deny symptoms. All passenger ships visiting U.S. ports now have surveillance video cameras, which under certain circumstances can assist in finding potential index cases for AGE outbreaks.

IBM, others to help public health officials improve food safety

Posted on July 5, 2014 by Douglas Powell

I normally don’t run company press releases because they are long on possibilities and short on actualities.

But this one may have some public health benefit. And was published in a journal.

Using novel algorithms, visualization, and statistical techniques, a new tool developed by IBM can use information on the date and location of billions of supermarket food items sold each week to quickly identify with high probability a set of potentially “guilty” products with in as few as 10 outbreak case reports. This research was published today in the peer-reviewed journal PLOS Computational Biology together with collaborators from Johns Hopkins University, Purdue University and the German Federal Institute for Risk Assessment (BfR).

Foodborne disease outbreaks of recent years demonstrate that due to increasingly interconnected supply chains food-related crisis situations have the potential to affect thousands of people, leading to significant healthcare costs, loss of revenue for food companies, and –in the worst cases– death. In the United States alone, one in six people are affected by food-borne diseases each year, resulting in 128,000 hospitalizations, 3,000 deaths, and a nearly $80B economic burden.

When a foodborne disease outbreak is detected, identifying the contaminated food quickly is vital to minimize the spread of illness and limit economic losses. However, the time required to detect it may range from days to weeks, creating extensive strain on the public health system.

Perhaps surprisingly, the petabytes of retail sales data have never before been used to accelerate the identification of contaminated food. In fact, this data already exists as part of the inventory systems used by retailers and distributors today, which manage up to 30,000 food items at any given time with nearly 3,000 of them being perishable.

Recognizing this issue, IBM scientists built a system that automatically identifies, contextualizes and displays data from multiple sources to help reduce the time to identify the mostly likely contaminated sources by a factor of days or weeks. It integrates pre-computed retail data with geocoded public health data to allow investigators to see the distribution of suspect foods and, selecting an area of the map, view public health case reports and lab reports from clinical encounters. The algorithm effectively learns from every new report and re-calculates the probability of each food that might be causing the illness.

“Predictive analytics based on location, content, and context are driving our ability to quickly discover hidden patterns and relationships from diverse public health and retail data,” said James Kaufman, Manager of Public Health Research for IBM Research, “We are working with our public health clients and with retailers in the U.S. to scale this research prototype and begin focusing on the 1.7B supermarket items sold each week in the United States.”

To demonstrate the system’s effectiveness, IBM scientists worked with the Department of Biological Safety of the German Federal Institute for Risk Assessment. In this demonstration, the scientists simulated 60,000 outbreaks of foodborne disease across 600 products using real-world food sales data from Germany.

Unfortunately, in real life cases of foodborne disease do not show up all at once as outbreaks are reported over a period of time. Depending on the circumstances, it takes public health officials weeks or months to identify the real cause.

“The success of an outbreak investigation often depends on the willingness of private sector stakeholders to collaborate pro-actively with public health officials. This research illustrates an approach to create significant improvements without the need for any regulatory changes. This can be achieved by combining innovative software technology with already existing data and the willingness to share this information in crisis situations between private and public sector organizations,” said Dr. Bernd Appel, Head of the Department Biological Safety, BfR.

Non-O157 shiga toxin–producing Escherichia coli in U. S. retail ground beef

Posted on July 3, 2014 by Douglas Powell

Shiga toxin–producing Escherichia coli (STEC) serotype O157:H7 and serogroups O26, O45, O103, O111, O121, and O145 are the leading cause of STEC-associated infections in humans in the United States. In the United States, these organisms are considered adulterants in raw nonintact beef products and in intact beef destined to be made into or used in nonintact raw beef products.

The objective of this study was to provide an estimate of the burden of the six serogroups of non-O157 STEC in ground beef obtained from retail stores across the United States. A convenience sample of commercial ground beef products (n = 1,129) were purchased from retail stores in 24 states from October 2011 to May 2012. The samples had various lean/fat proportions, muscle group of origin (chuck, round, sirloin, or not specified), and packaging types. For each ground beef sample, 25 g was inoculated in 225 ml of modified tryptic soy broth, stomached for 1 min, and then incubated at 41°C for 18 ± 2 h. These enrichment cultures were then screened for stx, eae, and O group genes using a commercially available, closed-platform PCR-based method. The potential positive samples were subjected to immunomagnetic separation and plated on modified Rainbow agar. Morphologically typical colonies were subjected to latex agglutination and PCR determination of stx and eae genes. Nine (0.8%) of the ground beef samples were potentially positive for at least one STEC serogroup after PCR screening. The serogroups detected by PCR assay were O26 (four samples), O103 (four samples), O145 (three samples), O45 (two samples), and O121 (one sample).

No STEC isolates belonging to these serogroups were recovered from the sample cultures. The current research provides updated surveillance data for non-O157 STEC isolates among commercial ground beef products and information regarding the potential sources of contamination from different parts of beef trims destined for ground beef production.

Journal of Food Protection®, Number 7, July 2014, pp. 1052-1240, pp. 1188-1192(5)

Liao, Yen-Te¹; Miller, Markus F.¹; Loneragan, Guy H.¹; Brooks, J. Chance¹; Echeverry, Alejandro¹; Brashears, Mindy M.²

http://www.ingentaconnect.com/content/iafp/jfp/2014/00000077/00000007/art00019

barfblog

safe food from farm to fork

Tag Archives: Surveillance