Search

As risk-based approaches are increasingly recognized and used to manage food safety hazards, their implementation requires a recognition and appreciation of residual risk. We define residual risk as the one that remains even after a fully compliant food safety system has been implemented. As true ‘zero risk’ is essentially unattainable, understanding and assessing the residual risks for different products is essential for the different actors involved in the food production system. Understanding residual risk is particularly critical as improved surveillance systems (e.g. facilitated by whole genome sequencing) can detect small outbreaks and potentially link cases to a product, even when they are consequences of residual risk rather than a non-compliant food safety system. Future work on assessing residual risk for different pathogen-food combinations are essential at both the company and governmental level to further fine tune food safety systems with the definition of an acceptable residual risk.

Microgreens have gained increasing popularity as food ingredients in recent years because of their high nutritional value and diverse sensorial characteristics. Microgreens are edible seedlings including vegetables and herbs, which have been used, primarily in the restaurant industry, to embellish cuisine since 1996. The rapidly growing microgreen industry faces many challenges. Microgreens share many characteristics with sprouts, and while they have not been associated with any foodborne illness outbreaks, they have recently been the subject of seven recalls. Thus, the potential to carry foodborne pathogens is there, and steps can and should be taken during production to reduce the likelihood of such incidents. One major limitation to the growth of the microgreen industry is the rapid quality deterioration that occurs soon after harvest, which keeps prices high and restricts commerce to local sales. Once harvested, microgreens easily dehydrate, wilt, decay and rapidly lose certain nutrients. Research has explored preharvest and postharvest interventions, such as calcium treatments, modified atmopsphere packaging, temperature control, and light, to maintain quality, augment nutritional value, and extend shelf life. However, more work is needed to optimize both production and storage conditions to improve the safety, quality, and shelf life of microgreens, thereby expanding potential markets.

In previous studies, parabens in model systems enhanced the thermal inactivation of foodborne pathogens, including Cronobacter sakazakii, Salmonella enterica serotype Typhimurium, Escherichia coli O157:H7, and Listeria monocytogenes. However, few studies have been conducted to evaluate this phenomenon in actual food systems. In the present study, the potential enhancement of thermal inactivation of C. sakazakii by butyl para-hydroxybenzoate (BPB) was evaluated in powdered infant formula (PIF) and nonfat dry milk (NFDM) in dry and rehydrated forms. When PIF was rehydrated with water at designated temperatures (65 to 80°C) in baby bottles, BPB did not enhance thermal inactivation. When rehydrated NFDM and lactose solutions with BPB were inoculated and heated at 58°C, BPB enhancement of thermal inactivation of C. sakazakii was negatively associated with the concentration of NFDM solutions in a dose-dependent manner, whereas thermal inactivation was enhanced in the presence of lactose regardless of its concentration, suggesting an interaction between proteins and BPB. Fluorescence testing further indicated an interaction between BPB and the proteins in PIF and NFDM. In inoculated dry NFDM with and without BPB stored at 24 and 55°C for 14 days, BPB did not substantially enhance bacterial inactivation. This study suggests that BPB is not likely to enhance mild thermal bacterial inactivation treatments in foods that have appreciable amounts of protein.

After studies with powdered infant formula indicated that the enhancement of thermal inactivation of Cronobacter sakazakii by butyl para-hydroxybenzoate (BPB) was blocked by high protein concentrations, we hypothesized that BPB would retain its synergistic activity in foods with limited protein and lipid concentrations. This hypothesis was tested by examining the ability of BPB to enhance the thermal inactivation of C. sakazakii 607 at 58°C in commercial apple juice, including examining the effects of pH and possible synergistic effects with malic acid. Apple juice was adjusted to designated pH values of 3.2 to 9.0, supplemented with selected concentrations of BPB (≤125 ppm), inoculated with early-stationary-phase C. sakazakii 607, and thermally treated (58°C) for 15 min with a submerged coil apparatus. The same methods were used to study the enhancement of thermal inactivation by malic acid. Samples were plated on tryptic soy agar for recovery and enumeration. Survival curves were plotted, and D-values were calculated by linear regression and compared using the Tukey honestly significant difference test. BPB significantly enhanced thermal inactivation in a concentration dependent manner, with D-values of a few seconds at the original pH (3.8). The enhancement of thermal inactivation was pH dependent over the pH range of 3.4 to 9.0. Malic acid enhanced thermal inactivation; the pH was decreased from 3.8 to 3.2. These results support the hypothesis that BPB can enhance the thermal inactivation of C. sakazakii in low-protein and low-lipid foods.

Recent outbreaks of foodborne disease associated with leafy greens have led to increased pre-harvest testing for pathogens and indicator microorganisms. However, the scientific and statistical rationale and the performance attributes for pre-harvest sampling methods are not well understood. The performance of three pre-harvest sampling methods, random, stratified random, and Z-pattern sampling, was evaluated by consideration of their mathematical derivations, computer simulations and field validation. Consideration of the probabilistic basis of the sampling methods indicated that the mean detection rates were similar. However, use of simulation modeling to assess the uncertainty associated with the three sampling methods indicated that the inherent variability of the Z-pattern sampling method was substantially greater than the other two sampling methods. A simulation tool was developed in Matlab that allowed the evaluation of the effectiveness of the three sampling methods. A limited validation study also observed that Z-pattern sampling had higher variability than the other two sampling methods. This study indicates that while the mean detection probabilities for the three sampling methods are similar, the random or stratified random sampling are less variable, particularly when the number of contamination sites or number of samples analyzed are small.

Heat-resistant foodborne pathogens have been a concern in low-moisture foods and ingredients (LMFs). Due to low thermal conductivity of low moisture materials, thermal treatment is not efficient and may cause nutritional loss. This study investigated the enhancement of thermal treatment of meat and bone meal (MBM) at low water activity (aw ) by inclusion of butylparaben (BP) as a model antimicrobial compound. Stationary phase Escherichia coli O157:H7 (Shiga toxin-negative) or Salmonella enterica serotype Typhimurium was inoculated into MBM containing 0-2000 ppm BP and incubated at 55 or 60°C for up to 5 hr. A biphasic inactivation pattern was observed for both pathogens, indicating existence of potentially thermal resistant subpopulations. Addition of 1000 ppm BP to MBM (aw = 0.4) significantly lowered the D-value at 55°C for E. coli O157:H7 (2.6 ± 0.5 hr) compared to thermal treatment alone (5.1 ± 0.6 h) during the treatment after the first 1 hr (p

Although high-temperature heat treatments can efficiently reduce pathogen levels, they also affect the quality and nutritional profile of foods and increase the cost of processing. The food additive butyl para-hydroxybenzoate (BPB) was investigated for its potential to synergistically enhance thermal microbial inactivation at mild heating temperatures (54 to 58°C). Four foodborne pathogenic bacteria, Cronobacter sakazakii, Salmonella enterica Typhimurium, attenuated Escherichia coli O157:H7, and Listeria monocytogenes, were cultured to early stationary phase and then subjected to mild heating at 58, 55, 57, and 54°C, respectively, in a model food matrix (brain heart infusion [BHI]) containing low concentrations of BPB (≤125 ppm). The temperature used with each bacterium was selected based on the temperature that would yield an approximately 1- to 3-log reduction over 15 min of heating in BHI without BPB in a submerged coil system. The inclusion of BPB at ≤125 ppm resulted in significant enhancement of thermal inactivation, achieving 5- to >6-log reductions of the gram-negative strains with D-values of

Topalcengiz, Zeynal/0000-0002-2113-7319 Domestic and wild animal intrusions are identified as a food safety risk during fresh produce production. The purpose of this study was to evaluate the survival of Shiga toxin-producing Escherichia coli (STEC) in cattle, feral pig, waterfowl, deer, and raccoon feces from sources in California, Delaware, Florida, and Ohio. Fecal samples were inoculated with a cocktail of rifampin-resistant STEC serotypes (O103, O104, O111, O145, and O157) (10(4) to 10(6) CFU/g of feces). Inoculated feces were held at ambient temperature. Populations of surviving cells were monitored throughout 1 year (364 days), with viable populations being enumerated by spread plating and enrichment when the bacteria were no longer detected by plating. Representative colonies were collected at various time intervals based on availability from different locations to determine the persistence of surviving STEC serotypes. Over the 364-day storage period, similar survival trends were observed for each type of animal feces from all states except for cattle and deer feces from Ohio. STEC populations remained the highest in cattle and deer feces from all states between days 28 and 364, except for those from Ohio. Feral pig, waterfowl, and raccoon feces had populations of STEC of

Topalcengiz, Zeynal/0000-0002-2113-7319 Heightened concerns about wildlife on produce farms and possible introduction of pathogens to the food supply have resulted in required actions following intrusion events. The purpose of this study was to evaluate the survival of Salmonella in feces from cattle and various wild animals (feral pigs, waterfowl, deer, and raccoons) in California, Delaware, Florida, and Ohio. Feces were inoculated with rifampin-resistant Salmonella enterica cocktails that included six serotypes: Typhimurium, Montevideo, Anatum, Javiana, Braenderup, and Newport (10(4) to 10(6) CFU/g). Fecal samples were stored at ambient temperature. Populations were enumerated for up to 1 year (364 days) by spread plating onto tryptic soy agar supplemented with rifampin. When no colonies were detected, samples were enriched. Colonies were banked on various sampling days based on availability of serotyping in each state. During the 364-day storage period, Salmonella populations decreased to

Improved control measures starting in the 1990s have greatly reduced the prevalence of L. monocytogenes in many food categories, particularly in meats and meat products. However, the rate of listeriosis has remained constant during the last decade and the more severe, systemic (invasive) form of listeriosis is now recognized as occurring more frequently in small outbreaks than previously recognized. This review addresses the recent advances in epidemiology and virulence, in growth and modelling, and insights from the risk assessments. Recognition of recent outbreaks from food vehicles not traditionally associated with L. monocytogenes (celery, cantaloupe, mung bean sprouts, stone fruits, caramel apples and ice cream) was facilitated by PFGE and, increasingly, whole genome sequencing. The Key Events framework, an understanding of the key individual biochemical steps from ingestion to infection, provides a structure for relating new knowledge on strain variability, mutations, and host susceptibility to the probability of illness. Guidance for determination of the growth/no growth potential of a food has been issued by several regulatory authorities and the risk assessments indicate that prevention of growth remains a principle control element. The recognition of biofilm formation and the possible existence of dormant, non-dividing persister cells will require additional attention. The recent outbreaks underscored the individual characteristics of specific foods (melons vs all fruit; microenvironments in the caramel apples) and raised questions about the current understanding of infectivity of lower doses and the susceptibility of specific individuals. Advances have been made in these areas, but further research is clearly necessary to control this pathogen.

Toxoplasma gondii is one of the leading foodborne pathogens in the United States. The Centers for Disease Control and Prevention (CDC) reported that T. gondii accounts for 24% of deaths due to foodborne illness in the United States. Consumption of undercooked pork products in which T. gondii has encysted has been identified as an important route of human exposure. However, little quantitative evaluation of risk due to different pork products as a function of microbial quality at the abattoir, during the production process, and due to consumer handling practices is available to inform risk management actions. The goal of this study was to develop a farm-to-table quantitative microbial risk assessment (QMRA) model to predict the public health risk associated with consumption of fresh pork in the United States. T. gondii prevalence in pigs was derived through a meta-analysis of existing data, and the concentration of the infectious life stage (bradyzoites) was calculated for each pork cut from an infected pig. Logistic regression and log-linear regression models were developed to predict the reduction of T. gondii during further processing and consumer preparation, respectively. A mouse-derived exponential dose-response model was used to predict infection risk in humans. The estimated mean probability of infection per serving of fresh pork products ranges from 3.2 × 10 −7 to 9.5 × 10 −6 , corresponding to a predicted approximately 94,600 new infections annually in the U.S. population due to fresh pork ingestion. Approximately 957 new infections per year were estimated to occur in pregnant women, corresponding to 277 cases of congenital toxoplasmosis per year due to fresh pork ingestion. In the context of available data, sensitivity analysis suggested that cooking is the most important parameter impacting human health risk. This study provides a scientific basis for risk management and also could serve as a baseline model to quantify infection risk from T. gondii and other parasites associated with meat products.

Prevention of microbial cross-contamination during postharvest handling is an important step to minimize microbial food safety hazards. Dump tanks and flume systems are widely used in states like Florida to transfer/wash tomatoes, and are one of the most critical points where cross-contamination may occur. Some processors in states such as New Jersey, New York and California utilize dry dump systems, with or without overhead spray bars, to process tomatoes, while others states such as Maryland field-pack tomatoes. This study was conducted in 2013 and 2014, from five growing regions in Florida and New Jersey each and from four growing regions in Maryland. A total of 1600 and 1597 composite samples were analyzed for aerobic plate count (APC), and total coliforms (TC) and generic E. coli (EC), respectively, from both pre- and post-processed tomatoes. Seventeen samples for APC and 72 for TC had counts outside the countable range and failed to provide any valid result, and were not included in the final data sets. The least square mean (LSM) value of APC for all samples (both pre- and post-processed) was 6.8 log10 CFU/tomato (n = 1583), whereas the LSM for TC counts was 4.9 log10 CFU/tomato (n = 1438). Ninety out of 1528 (5.9%) and 1498 out of 1597 (93.8%) samples had TC and EC counts below the detection limit of 1.3 log10 CFU/tomato, respectively. APC and TC counts in post-processed samples were significantly lower (p

Leafy green vegetables, including lettuce, are recognized as potential vehicles for foodborne pathogens such as Escherichia coli O157:H7. Fresh-cut lettuce is potentially at high risk of causing foodborne illnesses, as it is generally consumed without cooking. Quantitative microbial risk assessments (QMRAs) are gaining more attention as an effective tool to assess and control potential risks associated with foodborne pathogens. This study developed a QMRA model for E. coli O157:H7 in fresh-cut lettuce and evaluated the effects of different potential intervention strategies on the reduction of public health risks. The fresh-cut lettuce production and supply chain was modeled from field production, with both irrigation water and soil as initial contamination sources, to consumption at home. The baseline model (with no interventions) predicted a mean probability of 1 illness per 10 million servings and a mean of 2,160 illness cases per year in the United States. All intervention strategies evaluated (chlorine, ultrasound and organic acid, irradiation, bacteriophage, and consumer washing) significantly reduced the estimated mean number of illness cases when compared with the baseline model prediction (from 11.4- to 17.9-fold reduction). Sensitivity analyses indicated that retail and home storage temperature were the most important factors affecting the predicted number of illness cases. The developed QMRA model provided a framework for estimating risk associated with consumption of E. coli O157:H7-contaminated fresh-cut lettuce and can guide the evaluation and development of intervention strategies aimed at reducing such risk.

A recent study by the Centers for Disease Control and Prevention reported that between 1998 and 2008, leafy greens outbreaks accounted for 22.3% of foodborne outbreaks in the United States. Several studies on the growth of bacteria at different temperatures have been conducted; however, there is a need for the prediction of bacterial growth when leafy greens are transported without temperature control. Food products, when taken out of refrigeration, undergo a temperature change, with the rate of temperature change being proportional to the difference in the temperature of food and its surroundings. The objective of this study was to estimate the growth of Escherichia coli O157:H7, Salmonella enterica , and L. monocytogenes in leafy greens during transportation from retail to home at ambient temperatures ranging from 10 to 40°C for up to 10 h. Experiments were conducted to monitor the temperature increase in fresh spinach taken from refrigeration temperature to ambient temperature. The growth of pathogens was predicted using these changing temperature profiles with the three-phase linear model as a primary model and the square root model as the secondary model. The levels of E. coli O157:H7, S. enterica , and L. monocytogenes increased by 3.12, 2.43, and 3.42 log CFU at 40°C for the 10-h period, respectively, when no lag phase was assumed. If leafy greens are not kept out of refrigeration for more than 3 h, when the air temperature is 40°C or more, pathogen growth should be less than 1 log CFU. These results would assist in developing recommendations for food transportation without refrigeration.

Leafy greens contaminated with Salmonella enterica have been linked to large number of illnesses in many countries in recent years. Listeria monocytogenes is also a pathogen of concern for leafy greens because of its prevalence in the growing and processing environment and its ability to grow at refrigeration temperatures. Experimental data for the growth and survival of S. enterica and L. monocytogenes under different conditions and storage temperatures were retrieved from published studies. Predictive models were developed using the three-phase linear model as a primary growth model and square-root model to calculate specific growth rate (ln CFU g−1 h−1) at different temperatures (°C). The square-root model for S. enterica was calculated as μ = (0.020(Temperature+0.57))2. The square-root model for L. monocytogenes was fitted as μ = (0.023(Temperature-0.60))2. The growth-survival model for S. enterica and growth model for L. monocytogenes were validated using several dynamic time-temperature profiles during the production and supply chain of leafy greens. The models from this study will be useful for future microbial risk assessments and predictions of behavior of S. enterica and L. monocytogenes in the leafy greens production and supply chain.

This randomized controlled trial characterized the transfer of E. coli from animal feces and/or furrow water onto adjacent heads of lettuce during foliar irrigation, and the subsequent survival of bacteria on the adaxial surface of lettuce leaves. Two experiments were conducted in Salinas Valley, California: (1) to quantify the transfer of indicator E. coli from chicken and rabbit fecal deposits placed in furrows to surrounding lettuce heads on raised beds, and (2) to quantify the survival of inoculated E. coli on Romaine lettuce over 10 days. E. coli was recovered from 97% (174/180) of lettuce heads to a maximal distance of 162.56 cm (5.33 ft) from feces. Distance from sprinklers to feces, cumulative foliar irrigation, and lettuce being located downwind of the fecal deposit were positively associated, while distance from fecal deposit to lettuce was negatively associated with E. coli transference. E. coli exhibited decimal reduction times of 2.2 and 2.5 days when applied on the adaxial surface of leaves within a chicken or rabbit fecal slurry, respectively. Foliar irrigation can transfer E. coli from feces located in a furrow onto adjacent heads of lettuce, likely due to the kinetic energy of irrigation droplets impacting the fecal surface and/or impacting furrow water contaminated with feces, with the magnitude of E. coli enumerated per head of lettuce influenced by the distance between lettuce and the fecal deposit, cumulative application of foliar irrigation, wind aspect of lettuce relative to feces, and time since final irrigation. Extending the time period between foliar irrigation and harvest, along with a 152.4 cm (5 ft) no-harvest buffer zone when animal fecal material is present, may substantially reduce the level of bacterial contamination on harvested lettuce.

The Risk Assessment Subcommittee of the National Advisory Committee on Microbiological Criteria in Foods has prepared a generic document on the principles of risk assessment as applied to biological agents that can cause human foodborne disease. Typical biological agents include bacteria, viruses, fungi, helminths, protozoa, algae, parasites, and the toxic products that these agents may produce. Basic principles elaborated to characterize food pathogen risks include the four broadly accepted components of risk assessment. The role of surveillance and investigational activities to link biological agents and their food sources to consumer illness is described as is the role of predictive modeling for food pathogens.

The National Advisory Committee on Microbiological Criteria for Foods has been developing a concise document that provides key definitions and principles related to the application of quantitative risk assessment techniques to illnesses caused by foodborne biological agents. Key components of the document are outlined, and its status is reported.

(Received for publication October 5, 1992) The effects of pH and lactic acid or acetic acid concentration on Listeria monocytogenes inactivation were studied in brain heart infusion broth using a three strain mixture. Combinations of lactic acid/sodium lactate and acetic acid/sodium acetate were used to achieve concentrations of 0.1, 0.5, 1.0, and 2.0 M in conjunction with pH values of 4.0,5.0,6.0, and 7.0. Cultures adjusted with HCI to pH 3.0 to 7.0 in 0.5 pH unit intervals were used.as 0.0 M controls. Each pH/concentration combination was inoculated to a level of 108 CFU/ml and incubated at 28°C for up to 60 d. Bacterial populations were determined periodically by plate counts. Inactivation was exponential after an initial lag period. Survivor curves (log# versus time) were fitted using a linear model that incorporated a lag period. The model was subsequently used to calculate D values and "time to a 4-D (99.99%) inactivation" (t4-o);t4-0values were directly related to pH and inversely related to acid concentration. At acid/pH combinations that supported growth, the level of the organism increased slightly (2- to l O-fold) before declining. In the HCI-adjusted controls with pH's :0;5.5, the rate of inactivation was linearly related to pH. In the presence of the monocarboxylic acids, the duration of the lag period and the rate of inactivation were dependent on the pH, as well as the identity and concentration of acid. 4-D inactivation times were related to the level of undissociated lactic and acetic acids. That relationship was described by the equations, t4-0 = exp(-0.1773*LN·5 + 7.3482) and t4-0 = exp (-0.1468*AN·5 + 7.3905) for lactic and acetic acids, respectively, where LA and AA are mM of undissociated acid. These relationships were used in conjunction with the Henderson-Hasselback equation to develop a model for predicting the rate of inactivation as a function of pH and total organic acid concentration.

Artificial sausage casings were used as a model for studying bacterial attachment to meat connective tissue. Sausage casings of known mass were exposed to suspensions of Salmonella typhimurium in 0.15 M NaCl under various time, temperature, and inoculum level regimes, then washed to remove unattached bacteria. Attached bacterial cells were enumerated using both plate counts and scanning electron microscopy. Bacterial cells attached to sausage casing surfaces within 1 min of incubation. Numbers of attached cells increased with increasing temperature and inoculum levels and with time. Rates of attachment of S. typhimurium to sausage casings were comparable with those reported for attachment to meat surfaces. Sausage casings appear to be a convenient model for examining mechanisms of bacterial attachment to meats.

The combined effects of temperature (5 to 42°C), NaCl (0.5 to 4.5%), pH (5.3 to 7.3), and NaN02 (0 to 200 μg/ml) on the aerobic growth of Aeromonas hydrophila K144 were studied in brain heart infusion (BHI) broth using a modified central composite design. Variable combinations were tested in triplicate aerobic flasks; viable cell counts were made at intervals during incubation by surface plating on tryptic soy agar. Growth curves were generated using the Gompertz equation in conjunction with a nonlinear iterative regression analysis. Values for the four Gompertz parameters (A, C, B, and M) were obtained for the variable combinations tested. Using response surface techniques, quadratic and cubic equations containing the four variables of temperature, pH, NaCl, and NaN02 were developed to yield predictive values for the B and M Gompertz values. Goodness of fit evaluation of the models was by R2 values. Comparison of predicted and observed values of B and M and evaluation of predicted lag times and generation times indicated that the quadratic model gave a better fit. Overall, the variable combinations interacted to decrease the generation time and increase the lag time. The results indicate that pH, salt, and nitrite can decrease the growth of A. hydrophila when combined with low temperature incubation.

The effects and interactions of temperature (5, 19, 28, 37°C), initial pH (6.0 and 7.5), atmosphere (aerobic and anaerobic), sodium chloride content (0.5 and 4.5%), and sodium nitrite concentration (0, 50, 100, 200, 1000 μg/ml) on the growth of Listeria monocytogenes Scott A were determined using Tryptose Phosphate Broth. Growth data were analyzed by regression analysis to generate "best-fit" Gompertz equations, which were used subsequently to calculate lag phase duration, exponential growth rate, generation time, and maximum population density values. The data indicated that the growth kinetics of L. monocytogenes was dependent on the interaction of the five variables, particularly in regard to exponential growth rates and lag phase durations. The data suggest that sodium nitrite can have significant bacteriostatic activity against L. monocytogenes and may provide cured meats with a degree of protection against this microorganism, particularly if employed in conjunction with a combination of acidic pH, vacuum packaging, high salt concentrations, and adequate refrigeration.

Comminuted bacon, processed to contain target levels of 40 μg NaNO2/g and 0, 0.25 or 0.75% sucrose or 0.75% glucose, was inoculated with a mixture of spores of 20 strains of Clostridium botulinum (400 spores per g) and was canned under vacuum. Portions were irradiated using 137Cs at doses of 0, 0.19, 0.38, 0.75, 1.12 and 1.5 Mrad. Cans were incubated for 1, 2, 4 or 8 wk at 30°C. Some cans of nonirradiated bacon without or with 0.25% sucrose became toxic in 2 wk; with 0.75% sucrose, toxin production was delayed to 8 wk. Bacon irradiated at 0.75 Mrad, made with or without sucrose, became toxic in 2 to 4 wk, whereas most cans of bacon irradiated at 1.5 Mrad remained toxin-free for the 8-wk incubation period. A comparison of bacon made with 0.75% sucrose or glucose showed no difference between the sugars in the rates of toxin production by C. botulinum in irradiated cans of bacon. Irradiation at 0.19 Mrad increased the rate of toxin formation over nonirradiated bacon in sugar-containing (0.75%) bacon, but had...

Pre-harvest testing is increasingly used to enhance the microbial safety of fresh produce. Traditional sampling assumes that sample collectors have no information on potential contamination sources. Knowledge of such factors could potentially increase the effectiveness of pre-harvest sampling programs. Simulation modeling and field validation trials were used to evaluate a hybrid "Samples of Opportunity" (SOO) sampling method that included a portion of the samples based on the sampler's knowledge of risk factors in pre-harvest produce fields. Relative effectiveness of SOO sampling was compared with three traditional sampling methods. These evaluations were based on three non-random contamination scenarios. The mean detection probability of SOO is 96% higher than traditional sampling methods (p 0.001). However, if the site of actual contamination is offset from assumed area of contamination, the detection probability of SOO sampling drops, and becomes similar or even worse than that achieved by the other sampling methods. Preliminary field validation trials indicated indeed that SOO performed better than the other three sampling methods. This study provides a mathematical approach for evaluating the effectiveness of four pre-harvest sampling methods, and suggests that having a priori knowledge of the contamination source in the field would improve effectiveness of sampling, particularly if done using a standardized protocol.

Due to multiple outbreaks and large-scale product recalls, Salmonella has emerged as a priority pathogen in dry pet food and treats. However, little data are available to quantify risks posed by these classes of products to both pets and their owners. Specifically, the kinetics of Salmonella survival on complex pet food matrices are not available. This study measured the long-term kinetics of Salmonella survival on a dry pet food under storage conditions commonly encountered during production, retail, and in households (aw

Toxoplasma gondii is a prevalent protozoan parasite worldwide. Human toxoplasmosis is responsible for considerable morbidity and mortality in the United States, and meat products have been identified as an important source of T. gondii infections in humans. The goal of this study was to develop a farm-to-table quantitative microbial risk assessment model to predict the public health burden in the United States associated with consumption of U.S. domestically produced lamb. T. gondii prevalence in market lambs was pooled from the 2011 National Animal Health Monitoring System survey, and the concentration of the infectious life stage (bradyzoites) was calculated in the developed model. A log-linear regression and an exponential doseresponse model were used to model the reduction of T. gondii during home cooking and to predict the probability of infection, respectively. The mean probability of infection per serving of lamb was estimated to be 1.5 cases per 100,000 servings, corresponding to ∼6,300 new infections per year in the U.S.Based on the sensitivity analysis, we identified cooking as the most effective method to influence human health risk. This study provided a quantitative microbial risk assessment framework for T. gondii infection through consumption of lamb and quantified the infection risk and public health burden associated with lamb consumption.

The second edition of Microorganisms in Foods 7: Microbiological Testing in Food Safety Management updates and expands on information on the role of microbiological testing in modern food safety management systems. After helping the reader understand the often confusing statistical concepts underlying microbiological sampling, the second edition explores how risk assessment and risk management can be used to establish goals such as a “tolerable levels of risk,” Appropriate Levels of Protection, Food Safety Objectives or Performance Objectives for use in controlling foodborne illness. Guidelines for establishing effective management systems for control of specific hazards in foods are also addressed, including new examples for pathogens and indicator organisms in powdered infant formula, Listeria monocytogenes in deli-meats, enterohemorrhagic Escherichia coli in leafy green vegetables, viruses in oysters and Campylobacter in poultry. In addition, a new chapter on application of sampling concept to microbiological methods, expanded chapters covering statistical process control, investigational sampling, environmental sampling, and alternative sampling schemes.

Toxoplasma gondii is a protozoan parasite that is responsible for approximately 24% of deaths attributed to foodborne pathogens in the United States. It is thought that a substantial portion of human T. gondii infections is acquired through the consumption of meats. The dose-response relationship for human exposures to T. gondii-infected meat is unknown because no human data are available. The goal of this study was to develop and validate dose-response models based on animal studies, and to compute scaling factors so that animal-derived models can predict T. gondii infection in humans. Relevant studies in literature were collected and appropriate studies were selected based on animal species, stage, genotype of T. gondii, and route of infection. Data were pooled and fitted to four sigmoidal-shaped mathematical models, and model parameters were estimated using maximum likelihood estimation. Data from a mouse study were selected to develop the dose-response relationship. Exponential and beta-Poisson models, which predicted similar responses, were selected as reasonable dose-response models based on their simplicity, biological plausibility, and goodness fit. A confidence interval of the parameter was determined by constructing 10,000 bootstrap samples. Scaling factors were computed by matching the predicted infection cases with the epidemiological data. Mouse-derived models were validated against data for the dose-infection relationship in rats. A human dose-response model was developed as P (d) = 1-exp (-0.0015 × 0.005 × d) or P (d) = 1-(1 + d × 0.003 / 582.414)(-1.479) . Both models predict the human response after consuming T. gondii-infected meats, and provide an enhanced risk characterization in a quantitative microbial risk assessment model for this pathogen.

Consumption of locally, organically grown produce is increasing in popularity. Organic farms typically produce on a small scale, have limited resources, and adopt low technology harvest and postharvest handling practices. Data on the food safety risk associated with hand harvesting, field packing, and packing-house handling with minimal treatment, at this production scale, are lacking. We followed produce from small organic farms from the field through postharvest handling and packing. Pre- and postharvest produce (177 samples) and water (29 samples) were collected and analyzed quantitatively for Escherichia coli, total coliforms (TC), aerobic bacteria (APC), yeasts, molds (M), and enteric pathogens. No pathogens were recovered. E. coli was detected in 3 (3.6%) of 83 preharvest produce samples, 2 (6.3%) of 32 unwashed and 0 of 42 washed postharvest produce samples, and 10 (34.5%) of 29 water samples. No correlation was found between bacterial levels in irrigation water and those on produce. Postharvest handling without washing was a factor for APC and M counts on tomatoes, with lower frequencies postharvest. Postharvest handling with washing was a factor for leafy greens for TC counts, with higher frequencies postharvest. APC (P = 0.03) and yeast (P = 0.05) counts were higher in preharvest than in unwashed postharvest tomatoes. Washed postharvest leafy greens had higher M counts (P = 0.03) and other washed produce had higher TC counts (P = 0.01) than did their preharvest counterparts. Barriers were found to the use of sanitizer in wash water for leafy greens among small farms using organic practices. Hand harvesting and dry handling did not appear to be associated with a significant food safety risk, but washed leafy greens carried higher levels of some microbial indicators, possibly because of the lack of sanitizer in the wash water. The development of resources and materials customized for this sector of growers could enhance dissemination of information on best practices for handling of leafy greens.

Small- and medium-size farms in the mid-Atlantic region of the United States use varied agricultural practices to produce leafy greens during spring and fall, but the impact of preharvest practices on food safety risk remains unclear. To assess farm-level risk factors, bacterial indicators, Salmonella enterica , and Shiga toxin-producing Escherichia coli (STEC) from 32 organic and conventional farms were analyzed. A total of 577 leafy greens, irrigation water, compost, field soil, and pond sediment samples were collected. Salmonella was recovered from 2.2% of leafy greens ( n = 369) and 7.7% of sediment ( n = 13) samples. There was an association between Salmonella recovery and growing season (fall versus spring) ( P = 0.006) but not farming system (organic or conventional) ( P = 0.920) or region ( P = 0.991). No STEC was isolated. In all, 10% of samples were positive for E. coli : 6% of leafy greens, 18% of irrigation water, 10% of soil, 38% of sediment, and 27% of compost samples. Farming system was not a significant factor for levels of E. coli or aerobic mesophiles on leafy greens but was a significant factor for total coliforms (TC) ( P < 0.001), with higher counts from organic farm samples. Growing season was a factor for aerobic mesophiles on leafy greens ( P = 0.004), with higher levels in fall than in spring. Water source was a factor for all indicator bacteria ( P < 0.001), and end-of-line groundwater had marginally higher TC counts than source samples ( P = 0.059). Overall, the data suggest that seasonal events, weather conditions, and proximity of compost piles might be important factors contributing to microbial contamination on farms growing leafy greens.

The preharvest and preslaughter steps of food production constitute a first stage at which food can become contaminated with foodborne and toxigenic pathogens. Contamination at this early stage of food production can lead to amplification as food travels through the production and supply chain, accentuating the crucial need to address hazards and establish science-based metrics that are feasible to implement. This article discusses the preharvest food safety regulatory landscape in the United States, with a specific emphasis on fresh produce crops. Best practices, certification, audit schemes and challenges due to market channels, economies of scales, and grower behavior are considered in relation to the Food Safety Modernization Act. An outlook on the needs to facilitate implementation of the new law, develop educational programs for growers and stakeholders, and continue to better align food safety with environmental goals are presented.

The need and desirability of being able to perform quantitative microbial risk assessments for food processing and preparation operations has been discussed extensively for the past several years. However, there has been little application of this approach in large part due to the need to account for the changes in bacterial populations as a result of food environments and processing. The use of predictive food microbiology models has the potential for overcoming these limitations. Through integration of predictive modeling with risk assessment, it is possible to estimate how changes in unit operations are likely to effect the overall safety of a food. Hypothetical examples of how these techniques could be applied to both single-step and multiple-step food-processing and preparation operations are provided.

Pathogens in soil are readily mobilized by infiltrating water to travel downward through the soil. However, limited data are available on the horizontal movement of pathogens across a field. This study used a model system to evaluate the influence of soil type, initial soil moisture content, and field slope on the movement of Salmonella enterica serovar Newport across a horizontal plane of soil under flooding conditions. Three soil types of varying clay content were moistened to 40, 60, or 80% of their maximum water-holding capacities and flooded with water containing 6 log CFU/ml Salmonella Newport and Citrobacter freundii , the latter being evaluated as a potential surrogate for S. enterica in future field trials. A two-phase linear regression was used to analyze the microbial populations recovered from soil with increasing distance from the flood. This model reflected the presence of lag distances followed by a quantifiable linear decrease in the population of bacteria as a function of the distance from the site of flooding. The magnitude of the lag distance was significantly affected by the soil type, but this was not attributable to the soil clay content. The rate of the linear decline with distance from the flood zone was affected by soil type, initial soil moisture content, and soil incline. As the initial soil moisture content increased, the rate of decline in recovery decreased, indicating greater bacterial transport through soils. When flooding was simulated at the bottom of the soil incline, the rate of decline in recovery was much greater than when flooding was simulated at the top of the incline. There was no significant difference in recovery between Salmonella Newport and C. freundii , indicating that C. freundii may be a suitable surrogate for Salmonella Newport in future field studies.

Low-moisture foods have been responsible for a number of salmonellosis outbreaks worldwide over the last few decades, with cross contamination from contaminated equipment being the most predominant source. To date, actions have been focused on stringent hygienic practices prior to production, namely periodical sanitization of the processing equipment and lines. Not only does optimum sanitization require in-depth knowledge on the type and source of contaminants, but also the heat resistance of microorganisms is unique and often dependent on the heat transfer characteristics of the low-moisture foods. Rheological properties, including viscosity, degree of turbulence, and flow characteristics (for example, Newtonian or non-Newtonian) of both liquid and semisolid foods are critical factors impacting the flow behavior that consequently interferes heat transfer and related control elements. The demand for progressively more accurate prediction of complex fluid phenomena has called for the employment of computational fluid dynamics (CFD) to model mass and heat transfer during processing of various food products, ranging from drying to baking. With the aim of improving the quality and safety of low-moisture foods, this article critically reviewed the published literature concerning microbial survival in semisolid low-moisture foods, including chocolate, honey, and peanut butter. Critical rheological properties and state-of-the-art CFD application relevant to quality production of those products were also addressed. It is anticipated that adequate prediction of specific transport properties during optimum sanitization through CFD could be used to solve current and future food safety challenges.

Ability of bacteria to develop cross-protection to various stresses is well-known. However, exploration into cross-protection between conventional physiological stresses and emerging antimicrobial treatments is limited. Previously, we developed two new antimicrobial treatments based on a combined UV-A light and gallic acid (UVA + GA) treatment and enhanced antimicrobial action of gallic acid by UV-C exposure (UVC-GA) against Escherichia coli O157:H7. In this study, we investigated (a) whether prior exposure to sub-lethal stresses (heat, acid, osmotic and oxidation) can increase the resistance of E. coli O157:H7 towards these newly developed treatments, and (b) whether repeated exposure to these two treatments selects for E. coli O157:H7 sub-population resistant towards heat, acid, and oxidative challenge. Prior exposure to heat or acid stress increased (p 0.05) on the sensitivity of the bacteria to either of the treatments. Repeated exposure to UVA + GA or UVC-GA treatment selected for a sub-population that demonstrated higher (p

Peanuts and peanut-containing products have been linked to at least seven salmonellosis outbreaks worldwide in the past two decades. In response, the Technical Committee on Food Microbiology of the North American Branch of the International Life Sciences Institute collaborated with the American Peanut Council to convene a workshop to develop a framework for managing risk in low-moisture food commodities where large data sets are unavailable (using peanuts as the example). Workshop attendees were charged with answering questions regarding the appropriate statistical and scientific methods for setting log reduction targets with limited pathogen prevalence and concentration data, suitable quantities of data needed for determining appropriate log reduction targets, whether the requirement of a 5-log reduction in the absence of data to establish a target log reduction is appropriate, and what targeted log reduction would protect public health. This report concludes that the judgment about sufficient data is not solely scientific, but is instead a science-informed policy decision that must weigh additional societal issues. The participants noted that modeling efforts should proceed with sampling efforts, allowing one to compare various assumptions about prevalence and concentration and how they are combined. The discussions made clear that data and risk models developed for other low-moisture foods like almonds and pistachios may be applicable to peanuts. Workshop participants were comfortable with the use of a 5-log reduction for controlling risk in products like peanuts when the level of contamination of the raw ingredients is low (1 CFU/g) and the process well controlled, even when limited data are available. The relevant stakeholders from the food safety community may eventually conclude that as additional data, assumptions, and models are developed, alternatives to a 5-log reduction might also result in the desired level of protection for peanuts and peanut products.

The majority of foodborne outbreaks in the United States associated with the consumption of leafy greens contaminated with Escherichia coli O157:H7 have been reported during the period of July to November. A dynamic system model consisting of subsystems and inputs to the system (soil, irrigation, cattle, wild pig, and rainfall) simulating a hypothetical farm was developed. The model assumed two crops of lettuce in a year and simulated planting, irrigation, harvesting, ground preparation for the new crop, contamination of soil and plants, and survival of E. coli O157:H7. As predicted by the baseline model for crops harvested in different months from conventional fields, an estimated 13 out of 257 (5.05%) first crops harvested in July would have at least one plant with at least 1 CFU of E. coli O157:H7. Predictions indicate that no first crops would be contaminated with at least 1 CFU of E. coli O157:H7 for other months (April to June). The maximum E. coli O157:H7 concentration in a plant was higher in the second crop (27.10 CFU) than in the first crop (9.82 CFU). For the second crop, the probabilities of having at least one plant with at least 1 CFU of E. coli O157:H7 in a crop were predicted as 15/228 (6.6%), 5/333 (1.5%), 14/324 (4.3%), and 6/115 (5.2%) in August, September, October, and November, respectively. For organic fields, the probabilities of having at least one plant with ≥1 CFU of E. coli O157:H7 in a crop (3.45%) were predicted to be higher than those for the conventional fields (2.15%). IMPORTANCE This study is the first attempt toward developing a mathematical system model to understand the pathway of E. coli O157:H7 in the production of leafy greens. Results of the presented system model indicate that the seasonality of outbreaks of E. coli O157:H7-associated contamination of leafy greens was in good agreement with the prevalence of this pathogen in cattle and wild pig feces in a major leafy greens-producing region in California. On the basis of comparisons among the results of different scenarios, it can be recommended that the concentration of E. coli O157:H7 in leafy greens can be reduced considerably if contamination of soil with wild pig and cattle feces is mitigated.

Toxoplasma gondii is a widely distributed protozoan parasite. The Centers for Disease Control and Prevention reported that T. gondii is one of three pathogens (along with Salmonella and Listeria), that together account for >70% of all deaths due to foodborne illness in the United States. Food animals are reservoirs for T. gondii and act as one of the sources for parasite transmission to humans. Based on limited population-based data, the Food and Agriculture Organization/World Health Organization estimated that approximately 22% of human T. gondii infections are meatborne. The objective of the current study was to conduct a systematic meta-analysis to provide a precise estimation of T. gondii infection prevalence in food animals produced in the United States. Four databases were searched to collect eligible studies. Prevalence was estimated in six animal categories (confinement-raised market pigs, confinement-raised sows, non-confinement-raised pigs, lamb, goats, and non-confinement-raised chickens) by a quality-effects model. A wide variation in prevalence was observed in each animal category. Animals raised outdoors or that have outdoor access had a higher prevalence as compared with animals raised indoors. T. gondii prevalence in non-confinement-raised pigs ranked the highest (31.0%) followed by goats (30.7%), non-confinement-raised chickens (24.1%), lambs (22.0%), confinement-raised sows (16.7%), and confinement-raised market pigs (5.6%). These results indicate that T. gondii-infected animals are a food safety concern. The computed prevalence can be used as an important input in quantitative microbial risk assessment models to further predict public health burden.