Interrelationships between Multiple Climatic Factors and Incidence of Foodborne Diseases.

Climatic factors can affect the incidence of foodborne diseases (FBDs). Moreover, microbial network inference is useful for predicting the interrelationships between the incidence of FBDs and climatic factors. However, the interrelationships between FBD pathogens and most climatic factors are unknown. Using principal component analysis (PCA) and partial correlation coefficient matrices (PCCMs), we determined the intra-ecosystem interrelationship network of the multiple combined effects of 5 climatic factors (temperature, relative humidity, rainfall, insolation, and cloudiness) and the monthly incidences of 12 bacterial FBDs. Many FBD pathogens are interrelated with multiple combined factors. Salmonellosis has strong positive interrelationships with Vibrio parahaemolyticus and enterohemorrhagic Escherichia coli, and the interrelationships between Staphylococcus aureus/enteropathogenic E. coli/enterotoxigenic E. coli exhibits a typical triangular pattern with the combined effects of all 5 climatic factors. Meanwhile, campylobacteriosis and Clostridium perfringens infections are negatively interrelated with insolation and cloudiness. Enteroinvasive E. coli, Bacillus cereus, Listeria spp., and Yersinia enterocolitica are significantly interrelated with any climatic factor combination. The interrelationships or higher-order interrelationships among these climatic factors play an important role in the incidence of FBDs, although the underlying mechanisms remain unclear. Our results will serve as a foundation for more sophisticated models of future FBD patterns with regard to climate change.

Combined influence of multiple climatic factors on the incidence of bacterial foodborne diseases.

Information regarding the relationship between the incidence of foodborne diseases (FBD) and climatic factors is useful in designing preventive strategies for FBD based on anticipated future climate change. To better predict the effect of climate change on foodborne pathogens, the present study investigated the combined influence of multiple climatic factors on bacterial FBD incidence in South Korea. During 2011-2015, the relationships between 8 climatic factors and the incidences of 13 bacterial FBD, were determined based on inpatient stays, on a monthly basis using the Pearson correlation analyses, multicollinearity tests, principal component analysis (PCA), and the seasonal autoregressive integrated moving average (SARIMA) modeling. Of the 8 climatic variables, the combination of temperature, relative humidity, precipitation, insolation, and cloudiness was significantly associated with salmonellosis (P<0.01), vibriosis (P<0.05), and enterohemorrhagic Escherichia coli O157:H7 infection (P<0.01). The combined effects of snowfall, wind speed, duration of sunshine, and cloudiness were not significant for these 3 FBD. Other FBD, including campylobacteriosis, were not significantly associated with any combination of climatic factors. These findings indicate that the relationships between multiple climatic factors and bacterial FBD incidence can be valuable for the development of prediction models for future patterns of diseases in response to changes in climate.

Use of Internet Search Queries to Enhance Surveillance of Foodborne Illness.

As a supplement to or extension of methods used to determine trends in foodborne illness over time, we propose the use of Internet search metrics. We compared Internet query data for foodborne illness syndrome-related search terms from the most popular 5 Korean search engines using Health Insurance Review and Assessment Service inpatient stay data for 26 International Classification of Diseases, Tenth Revision, codes for foodborne illness in South Korea during 2010-2012. We used time-series analysis with Seasonal Autoregressive Integrated Moving Average (SARIMA) models. Internet search queries for "food poisoning" correlated most strongly with foodborne illness data (r=0.70, p<0.001); furthermore, "food poisoning" queries correlated most strongly with the total number of inpatient stays related to foodborne illness during the next month (ß=0.069, SE 0.017, p<0.001). This approach, using the SARIMA model, could be used to effectively measure trends over time to enhance surveillance of foodborne illness in South Korea.

Bacterial contamination in Saeng-go-gi, a ready-to-eat fresh raw beef dish sold in restaurants in South Korea.

This study investigated the bacterial contamination levels in ready-to-eat fresh raw beef, Saeng-go-gi in Korean, sold in restaurants. A total of 462 samples were analyzed by performing an aerobic bacterial plate count, a coliform count, and an Escherichia coli O157:H7 count. Aerobic bacterial plate counts of fresh raw beef obtained from Seoul, Cheonan, Daegu, Gunsan, and Gwangju retail store restaurants were 6.46, 6.89, 6.39, 6.58, and 6.67 log CFU/g, respectively, and coliforms were 4.05, 4.97, 4.76, 3.62, and 3.32 log CFU/g, respectively. Among the 462 assessed samples, suspected E. coli O157:H7 colonies were found in 32, 24, 20, 22, and 16 samples obtained from Seoul, Cheonan, Daegu, Gunsan, and Gwangju, respectively. The identity of these isolated colonies was further assessed by using a latex agglutination kit. The agglutination assay data showed that the isolates were not E. coli O157:H7. The data from this study could be used to design better food handling practices for reducing foodborne illnesses linked to fresh raw beef consumption.

Estimating the burden of foodborne disease, South Korea, 2008-2012.

Estimating the actual occurrence of foodborne illness is challenging because only a small proportion of foodborne illnesses are confirmed and reported. Many studies have attempted to accurately estimate the overall number of cases of foodborne illness, but none have attempted to estimate the burden of foodborne disease in South Korea. This study used data from the Health Insurance Review and Assessment Service (HIRA), a public health surveillance system in South Korea, to calculate the number of cases and hospitalizations due to 18 specific pathogens and unspecified agents commonly transmitted through contaminated food between 2008 and 2012 in South Korea while accounting for uncertainty in the estimate. The estimated annual occurrences of foodborne illness were 336,138 (90% credible interval [CrI]: 258,379-430,740), with inpatient stays (hospitalizations), outpatient visits (foodborne disease infections), and patients' experiences (without visiting physicians) accounting for 2.3% (n=7809 [90% CrI: 7016-8616]), 14.4% (n=48,267 [90% CrI: 45,883-50,695]) and 83.3% (n=280,062 [90% CrI: 201,795-374,091]), respectively. Escherichia coli, including enterohemorrhagic E. coli, caused most illnesses, followed by nontyphoidal Salmonella spp., Staphylococcus aureus, hepatitis A virus, and norovirus. These results will be useful to food safety policymakers for the prevention and control of foodborne pathogens in South Korea.

Estimate of the prevalence and burden of food poisoning by natural toxic compounds in South Korea.

Many studies have attempted to accurately estimate the overall number of cases of foodborne illness, but there have not been many attempts to estimate the burden of foodborne disease caused by natural toxic compounds. This study estimated the number of cases due to specific natural toxins (seafood toxins, plant toxins, and mycotoxins) during 2008-2012 in South Korea, using data from the Health Insurance Review and Assessment Service (HIRA), while accounting for uncertainty in the estimate. The estimated annual occurrences of foodborne illness from natural toxic agents were 1088 (90% credible interval [CrI]: 883-1315), which suggests there are 21 times more cases than are reported, with 45.6% (n=496 [388-614]) and 54.4% (n=592 [423-790]), accounting for inpatient stays and outpatient visits, respectively. Among toxins, mushroom and plant toxins caused the highest illnesses, followed by toxic agents in seafood and mycotoxins. The 55-59year olds had the highest proportion of illnesses and those over the age of 40 accounted for 70.6% of all cases. The cases caused by mushroom poison, poisonous plants, and seafood toxins showed clear seasonal and regional differences. These results will be useful to food safety policymakers for the prevention and control of natural food poisons in South Korea.

Analysis of microbiological contamination in mixed pressed ham and cooked sausage in Korea.

The objective of this study was to investigate the microbial contamination levels (aerobic bacteria plate count [APC], coliforms, Escherichia coli, Staphylococcus aureus, and Listeria monocytogenes) in mixed pressed ham and cooked sausage. A total of 180 samples were collected from factories with and without hazard analysis critical control point (HACCP) systems at four steps: after chopping (AC), after mixing (AM), cooling after the first heating process, and cooling after the second heating process. For ham, APCs and coliform and E. coli counts increased when ingredients were added to the meat at the AC step. Final product APC was 1.63 to 1.85 log CFU/g, and coliforms and E. coli were not detected. S. aureus and L. monocytogenes were found in nine (15.0%) and six (10.0%) samples, respectively, but only at the AC and AM steps and not in the final product. Sausage results were similar to those for ham. The final product APC was 1.52 to 3.85 log CFU/g, and coliforms and E. coli were not detected. S. aureus and L. monocytogenes were found in 29 (24.2%) and 25 (20.8%) samples at the AC and AM steps, respectively, but not in the final product. These results indicate that the temperature and time of the first and second heating are of extreme importance to ensure the microbiological safety of the final product regardless of whether a HACCP system is in place. Microorganism contamination must be monitored regularly and regulations regarding sanitization during processing should be improved. Education regarding employee personal hygiene, environmental hygiene, prevention of cross-contamination, ingredient control, and step-by-step process control is needed to reduce the risk of food poisoning.

Rapid detection of viable Escherichia coli O157 by coupling propidium monoazide with loop-mediated isothermal amplification.

Conventional molecular detection methods cannot distinguish between viable and dead Escherichia coli O157 cells. In this study, the loop-mediated isothermal amplification (LAMP) method combined with propidium monoazide (PMA) treatment was developed to selectively detect viable E. coli O157 cells. Four primers, including outer primers and inner primers, were specially designed for the recognition of six distinct sequences on the serogroups (O157) of the specific rfbE gene of the E. coli O157 genome. PMA selectively penetrated through the compromised cell membranes and intercalated into DNA. Amplification of DNA from dead cells was completely inhibited by 3.0 µg/ml PMA, whereas the DNA derived from viable cells was amplified remarkably within 1 h by PMA-LAMP. Exhibiting high sensitivity and specificity, PMA-LAMP is a suitable method for evaluating the inactivation efficacy of slightly acidic electrolyzed water in broth. PMA-LAMP can selectively detect viable E. coli O157 cells. This study offers a novel molecular detection method to distinguish between viable and dead E. coli O157 cells.

A probability model for enterotoxin production of Bacillus cereus as a function of pH and temperature.

Bacillus cereus is frequently isolated from a variety of foods, including vegetables, dairy products, meats, and other raw and processed foods. The bacterium is capable of producing an enterotoxin and emetic toxin that can cause severe nausea, vomiting, and diarrhea. The objectives of this study were to assess and model the probability of enterotoxin production of B. cereus in a broth model as affected by the broth pH and storage temperature. A three-strain mixture of B. cereus was inoculated in tryptic soy broth adjusted to pH 5.0, 6.0, 7.2, 8.0, and 8.5, and the samples were stored at 15, 20, 25, 30, and 35°C for 24 h. A total of 25 combinations of pH and temperature, each with 10 samples, were tested. The presence of enterotoxin in broth was assayed using a commercial test kit. The probabilities of positive enterotoxin production in 25 treatments were fitted with a logistic regression to develop a probability model to describe the probability of toxin production as a function of pH and temperature. The resulting model showed that the probabilities of enterotoxin production of B. cereus in broth increased as the temperature increased and/or as the broth pH approached 7.0. The model described the experimental data satisfactorily and identified the boundary of pH and temperature for the production of enterotoxin. The model could provide information for assessing the food poisoning risk associated with enterotoxins of B. cereus and for the selection of product pH and storage temperature for foods to reduce the hazards associated with B. cereus.