Activities, Achievements, and Lessons Learned during the First 10 Years of the Foodborne Diseases Active Surveillance Network:

Transcription

1 FOOD SAFETY Frederick J. Angulo, Section Editor INVITED ARTICLE Activities, Achievements, and Lessons Learned during the First 10 Years of the Foodborne Diseases Active Surveillance Network: Elaine Scallan Lead FoodNet Unit, Enteric Diseases Epidemiology Branch, Division of Foodborne, Bacterial, and Mycotic Diseases, National Center for Zoonotic, Vectorborne, and Enteric Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia Since the establishment of the Foodborne Diseases Active Surveillance Network (FoodNet) in 1996, it has been an essential resource for the surveillance and investigation of foodborne disease in the United States. FoodNet has had a major impact on food safety because it conducts population-based, active surveillance for laboratory-confirmed infections from 9 pathogens commonly transmitted through food. Each year, FoodNet publishes the National Report Card on Food Safety, which is used by regulatory agencies, industry and consumer groups, and public health personnel to prioritize and evaluate food safety interventions and monitor progress toward national health objectives. FoodNet also determines the human-health impact of foodborne illness by conducting related epidemiological studies that contribute to the estimates of the overall burden of foodborne illness, attribute the burden of foodborne illness to specific foods and settings, and address important foodborne disease-related issues, such as antimicrobial resistance and sequelae from foodborne infections. This article summarizes the activities, achievements, and lessons learned during the first 10 years of FoodNet. Foodborne infections are an important public health problem that caused an estimated 76 million illnesses in the United States in 1999 [1]. Although advances in technology, regulation, and education have dramatically improved the safety of the food supply, the mass production, distribution, and importation of food has brought new challenges. Moreover, new and emerging foodborne pathogens and novel food vehicles continue to be described. Surveillance to determine the human health burden of disease and to measure improvements of food safety, therefore, continues to be a challenging, but critical, public health priority. In the United States, the Foodborne Diseases Active Surveillance Network (FoodNet) has been producing more precise and accurate estimates of the burden and sources of specific foodborne diseases since its establishment in 1996 [2, 3]. This Received 17 October 2006; accepted 20 November 2006; electronically published 24 January Reprints or correspondence: Dr. Elaine Scallan, Lead FoodNet Unit, Enteric Diseases Epidemiology Branch, Div. of Foodborne, Bacterial, and Mycotic Diseases, National Center for Zoonotic, Vectorborne, and Enteric Diseases, Centers for Disease Control and Prevention, 1600 Clifton Rd. MSD63, Atlanta, GA (escallan@cdc.gov). Clinical Infectious Diseases 2007; 44: This article is in the public domain, and no copyright is claimed /2007/ DOI: / article summarizes the activities, achievements, and lessons learned during the first 10 years of FoodNet. FOODNET ORGANIZATION FoodNet is the principal foodborne disease component of the Centers for Disease Control and Prevention s (CDC) Emerging Infections Program and is a collaborative project among the CDC, the US Department of Agriculture s Food Safety and Inspection Service, the US Food and Drug Administration, and the state health departments of Emerging Infections Program sites. When FoodNet was established in 1996, there were 5 FoodNet sites (California, Connecticut, Georgia, Minnesota, and Oregon). Since then, the number of sites has doubled, with the addition of New York and Maryland in 1998, Tennessee in 2000, Colorado in 2001, and New Mexico in 2004 (figure 1). The population that is surveilled by FoodNet is large and accounted for 15% of the US population (44.9 million persons) in Although the population that is surveilled by FoodNet was not chosen to be representative of the United States, a demographic comparison of the 2005 FoodNet and US census populations suggests that differences are limited; the only no- 718 CID 2007:44 (1 March) FOOD SAFETY

2 Figure 1. The Foodborne Diseases Active Surveillance Network 2005 table difference is the underrepresentation of the Hispanic population at FoodNet sites (9% vs. 14%). FOODNET OBJECTIVES The key objectives of FoodNet are to determine the burden of foodborne disease, monitor trends of the burden of specific foodborne diseases over time, attribute the burden of foodborne illness to specific foods and settings, and develop and assess interventions to reduce the burden of foodborne illness. To meet these objectives, FoodNet conducts active surveillance and epidemiologic studies [2, 3]. DETERMINING THE BURDEN OF FOODBORNE DISEASE FoodNet has had a major impact on food safety in the United States because it conducts population-based, active surveillance for laboratory-confirmed infections from pathogens commonly transmitted through food. Surveillance for Campylobacter species, Listeria monocytogenes, Salmonella species, Shigella species, Shiga toxin producing Escherichia coli O157, Vibrio species, and Yersinia enterocolitica has been conducted since Surveillance for Cryptosporidium and Cyclospora species began in 1997, and surveillance for non-o157 Shiga toxin producing E. coli began in Although some laboratory-confirmed infections are passively reported to the CDC, there were no reliable data on the burden of Campylobacter, E. coli O157, L. monocytogenes, Vibrio, or Y. enterocolitica infections before the establishment of FoodNet. In contrast to passive laboratory-based surveillance systems, which rely upon reports of foodborne diseases from clinical laboratories to state and local health departments and then to the CDC, FoodNet personnel actively contact clinical laboratories (either weekly or monthly, depending on the size of the laboratory) to ascertain laboratory-confirmed cases occurring within the surveillance area. Clinical laboratories are identified using state licensing lists and physician surveys [4]. All clinical laboratories serving the FoodNet population are contacted, including larger reference laboratories that receive specimens from but are geographically located outside the 10 FoodNet sites. Each clinical laboratory is audited at least twice yearly to ensure that all cases of disease under surveillance are ascertained and that changes in incidence are not a result of surveillance artifacts. In 2006, there were 1600 clinical laboratories serving the FoodNet sites. Those located outside FoodNet sites contribute approximately one-fifth of the cases. Frequent audits demonstrate 195% ascertainment during prospective active surveillance. Several surveillance steps are necessary for a case to be ascertained by laboratory-based surveillance: the ill person must seek medical care, a stool specimen must be submitted, and the laboratory must test for the pathogen and report the case to a public health agency (figure 2). Therefore, although active surveillance and routine audits ensure that all laboratory-confirmed cases occurring in the FoodNet surveillance area are ascertained, cases may not be ascertained because a person seeks medical care and diagnostic practices are performed. Before the inception of FoodNet, there were no precise estimates of the overall human health burden of foodborne illness in the United States. Policy-makers and regulatory agencies, when faced with allocating limited resources, had difficulty as- FOOD SAFETY CID 2007:44 (1 March) 719

3 Figure 2. Surveillance steps that must occur for a laboratory-confirmed case to be ascertained through active surveillance. sessing the burden of foodborne illness. FoodNet advanced the science of this aspect by adopting a paradigm known as the burden-of-illness pyramid. By estimating the frequency of cases of foodborne disease that go undetected at each surveillance step (seeking of medical care, stool specimen submission, and laboratory testing), FoodNet can account for surveillance gaps and, therefore, allow an extrapolation from laboratoryconfirmed cases (top of the burden-of-illness pyramid) to estimate the overall burden of disease in the community (bottom of the burden-of-illness pyramid). FoodNet estimates the frequency of cases of foodborne disease that go undetected because of laboratory testing practices by conducting surveys of clinical diagnostic laboratories that serve the FoodNet population [5, 6]. Repeated cycles of the FoodNet Laboratory Survey also allow FoodNet to detect temporal or geographical differences of clinical laboratory practices that may contribute to variations of pathogen isolation rates. For example, the percentage of stool samples cultured for E. coli O157 in laboratories that cultured all bloody stool samples for E. coli O157 varied markedly by site (from 58% in Georgia to 96% in Connecticut) [7]. The frequency at which persons with an acute diarrheal illness seek medical care and submit a stool culture has been estimated by cross-sectional telephone surveys of the general population [8 10]. The FoodNet Population Survey also provides an estimate of the prevalence of acute diarrheal illness, and repeated cycles of the survey help validate that changes of disease incidence are not because of changes of care seeking or specimen submission [11, 12]. This platform has also been used to collect baseline information of the consumption of risky foods; these data have been useful for hypotheses generation during outbreak investigations [13]. The attitudes and practices of a range of foodborne illness-related topics, including antibiotics [14], food irradiation [15], food-handling [16], and sources of gastroenteritis [17], have also been described. Originally, FoodNet considered that physician surveys might provide information about the frequency of stool sample submissions [4]. However, several biases were considered to have lead to an overestimation of stool sample submission rates as a result of this method, and patient reports from the FoodNet Population Survey have been used instead. FoodNet Physician Surveys have, however, been useful for the description of physician practices and determination of the current role of physicians as food-safety educators [4, 18]. By augmenting surveillance with information from FoodNet Laboratory Surveys [5, 6] and the FoodNet Population Survey [8, 9], FoodNet has been able to estimate the overall burden of illness due to specific pathogens. For example, FoodNet estimates that, during 2004, for each laboratory-confirmed case of Salmonella infection, there were 38 cases in the community [19]. Data from FoodNet surveillance and related epidemiological studies were vital to the CDC s estimate of foodborne illness in the United States [1], and these data have contributed to estimates of the economic cost of foodborne illness [20, 21]. For determination of the overall burden of foodborne illness, FoodNet also considers the impact of complications from foodborne infections. FoodNet began surveillance for hemolytic uremic syndrome, a serious complication of E. coli O157 infection, in persons aged!18 years during 1997, using prospective reporting by pediatric nephrologists. Three FoodNet sites have pilot-tested the use of hospital discharge data to determine the burden of Guillain-Barré syndrome, an important sequelae of Campylobacter infection. In addition to these syndromes, 2 sites have examined the impact of reactive arthritis following a laboratory-confirmed bacterial infection, and 1 site conducted a survey to determine the frequency of selfreported complications from enteric infections [23]. MONITOR TRENDS OF THE BURDEN OF SPECIFIC FOODBORNE DISEASES The expansion of the FoodNet sites presented a challenge with regard to monitoring the change of incidence over time, because site-to-site variation of the incidence of laboratory-confirmed infection can cause the collective crude incidence to change, even if no actual change of incidence actually occurred. FoodNet reports of the description of trends of the incidence of laboratory-confirmed infections over time from before 2001 limited their assessment to the 5 original sites [7, 24 27]. In 2001, FoodNet overcame this limitation by adopting the use of a negative binomial model to account for variations intro- 720 CID 2007:44 (1 March) FOOD SAFETY

4 duced by changes of population [28]. This model uses all available data to estimate the effect of time on disease incidence; the model treats time as a categorical variable and estimates the relative change of incidence in comparison with a baseline period. During April of each year, FoodNet publishes a report in the CDC s Mortality and Morbidity Weekly Report that describes preliminary surveillance data for the preceding calendar year and compares these data with a baseline period. Preliminary data from 2005 are shown in table 1 and figure 3. This report has become known as the National Report Card on Food Safety and is used by regulatory agencies, industry and consumer groups, and public health personnel to prioritize and evaluate food safety interventions and monitor progress toward national health objectives. For example, since 2003, FoodNet has reported a significant decrease of the incidence of infection due to E. coli O157 [29]. This decrease is likely a result of a major initiative by the Food Safety and Inspection Service and the meat industry to reduce E. coli O157 in ground beef. Conversely, the lack of a decrease of the incidence of Salmonella infection in recent years supports the need for initiatives to reduce the presence of Salmonella species in raw meat and poultry products. ATTRIBUTING FOODBORNE ILLNESS TO SPECIFIC FOODS AND CONTEXTS The knowledge of which foods cause the most human illness is essential for the development and prioritization of food safety interventions. In the United States, the approach to attribution has been multifaceted, involving a variety of public health, regulatory, and academic organizations [30]. FoodNet has been able to make a unique contribution to attribution by using its surveillance platform to conduct studies to determine the risk factors for sporadic infections [31 43]. FoodNet case-control studies have highlighted known risk factors for specific foodborne infections, including ground beef and E. coli O157 infection [37] and chicken and Campylobacter infection [31]. FoodNet studies have also identified food vehicles previously unrecognized as risk factors in the United States, including chicken as a risk factor for Salmonella enterica infection [33, 40] and hummus and melon as risk factors for Listeria monocytogenes infection [41]. A recent study identified riding in a shopping cart next to raw meat or poultry as a novel risk factor for Salmonella and Campylobacter infection in infants [42, 43]. In 2006, FoodNet will launch a case-control study involving Table 1. Incidence of bacterial and parasitic infection and postdiarrheal hemolytic uremic syndrome that were surveilled by the Foodborne Diseases Active Surveillance Network (FoodNet; United States, 2005), by site, compared with national health objectives for Pathogen FoodNet Site CA CO CT GA MD MN NM NY OR TN Overall for 2005 National health objective a Bacterium Campylobacter species Listeria species Salmonella species Shigella species NA b Shiga toxin producing Escherichia coli O Non O157 Shiga toxin producing E. coli NA Vibrio species NA Yersinia species NA Parasites Cryptosporidium species NA Cyclospora species NA Hemolytic uremic syndrome c c 0.9 Population in surveillance, millions NOTE. Data are incidence of infection per 100,000 persons, unless otherwise indicated. CA, California; CO, Colorado; CT, Connecticut; GA, Georgia; MD, Maryland; MN, Minnesota; NM, New Mexico; NY, New York; OR, Oregon; TN, Tennessee. a Objectives are for incidence of Campylobacter, Salmonella, and Shiga-toxin producing E. coli O157 infections for 2010 and for incidence of Listeria infection for b Not applicable because there is no national health objective. c 2004 data reported for hemolytic uremic syndrome incidence. Incidence rate of postdiarrheal hemolytic uremic syndrome in children!5 years old; rate calculation based on population!5 years old in surveillance. FOOD SAFETY CID 2007:44 (1 March) 721

5 Figure 3. Relative rates, compared with baseline period of laboratory-diagnosed cases of Campylobacter, Shiga-toxin producing Escherichia coli O157, Listeria, Salmonella, and Vibrio infection, by year, Foodborne Diseases Active Surveillance Network, United States, emerging Salmonella serotypes with limited information for attribution, S. serotype Javiana and S. serotype I 4,[5],12:i:-. This study will include validation substudies aimed at better understanding of the impact of selection and misclassification bias. To additionally advance the methodology that is employed for sporadic case-control studies of enteric infections, FoodNet is collaborating with the European Med-Vet-Net for a study to assess the impact of population immunity and writing an international review of methods with other members of the International Collaboration on Enteric Disease Burden of Illness Studies. Data from FoodNet surveillance contributes to attribution in other ways. Since 2004, FoodNet has collected information on international travel history from patients with Salmonella and E. coli O157 infections, which allows an estimate of the burden of domestically acquired illness. FoodNet can also enhance surveillance to collect additional exposure information. For example, during 2005, all patients with laboratory-confirmed Shigella infection were interviewed about risk factors that may have been prevalent during the week before illness onset. Although a comparison group was not interviewed, this provided useful information on the proportion of patients with risk factors for shigellosis, and the proportion of patients who did not report any of the typical nonfood exposures provided a conservative estimate of the foodborne transmission. FoodNet has also been working with the National Antimicrobial Resistance Monitoring System and the Food and Drug Administration to conduct a Retail Food Study to monitor the prevalence and antimicrobial resistance of Campylobacter and Salmonella organisms that are isolated from random sampled meat and poultry products from grocery stores at FoodNet sites [44]. A FoodNet Grocery Store Survey was launched at 2 FoodNet sites during 2006 to quantify Campylobacter levels in retail chicken products and to test the hypothesis that regional differences result from differences in Campylobacter dose on retail poultry products. OTHER FOODNET CONTRIBUTIONS FoodNet s greatest asset is the network of public health professionals that it brings together to work on foodborne issues of national importance. This infrastructure has allowed FoodNet sites to initiate rapid surveillance for new and emerging foodborne pathogens that are not currently under surveillance. During past years, FoodNet has conducted surveillance for variant Creutzfeldt-Jakob disease and Enterobacter sakazakii [3]. More recently, FoodNet has implemented surveillance for community-acquired Clostridium difficile because of nationwide discussions about the changing epidemiology of this pathogen. Other surveillance systems managed by the CDC collect information about enteric pathogens; therefore, epidemiological and laboratory data from 1 person may be contained in 11 system. Linking data between systems increases the utility of surveillance data, and FoodNet is working with others to achieve this goal. For example, retrospective linking of FoodNet and National Antimicrobial Resistance Monitoring System data 722 CID 2007:44 (1 March) FOOD SAFETY

6 demonstrated the human health impact of antimicrobial-resistant infections result [47 49], which was used to support the Food and Drug Administration ban on use of fluoroquinolones during poultry production [49]. When case-control study data are available, linked data allows for the identification of risk factors associated with specific resistance patterns, phage types, or molecular subtypes. For example, prior antimicrobial agent use has been associated with an increased risk of multidrugresistant S. enterica serotype Typhimurium infection [35], international travel has been associated with an increased risk of fluoroquinolone-resistant Campylobacter infection [37], and Newport multidrug-resistant AmpC infections have been shown to be acquired through the US food supply most likely from bovine and, perhaps, poultry sources and particularly among persons already taking antimicrobial agents [50]. A recent case-control study was able to discriminate between chicken and eggs as risk factors for S. Enteriditis infection with PT8 and 13, respectively [40]. FoodNet sites serve as a model for improving surveillance nationwide. For example, FoodNet sites pilot-tested an enhanced surveillance scheme for listeriosis, which was subsequently rolled out nationwide during FoodNet has also been working with the CDC s Outbreak Network for Foodborne Disease Surveillance and Response and the Environmental Health Specialists Network to better characterize the factors associated with determining an etiology and to report contributing factors (e.g., cross-contamination) by collecting supplemental information about outbreaks of infection that occur at FoodNet sites [45]. In an effort to improve stool culturing rates and etiology determination, 1 FoodNet site pilot-tested a program for the delivery stool kits during outbreak investigations [46]. FoodNet has also had an impact on the surveillance of foodborne disease internationally. FoodNet s approach to estimating the impact of foodborne disease using the burden of illness pyramid has been adopted in other countries, and FoodNet is part of an international working group, the International Collaboration on Enteric Disease Burden of Illness Studies, which aims to further improve the methods used to estimate the burden of foodborne illness [10, 22]. FoodNet is also an active member of the World Health Organization Global Salm-Surv ( which is a global network of laboratories and public health professionals working to strengthen laboratory-based surveillance of foodborne diseases worldwide. DEVELOP AND ASSESS INTERVENTIONS TO REDUCE THE BURDEN OF FOODBORNE ILLNESS FoodNet is an important resource that provides vital information to aid national decision-making when allocating resources and directs prevention strategies aimed at improving the safety of the food supply. By describing trends in laboratoryconfirmed infections, FoodNet has been able to describe the burden of foodborne illness, identify populations at risk, and provide the data for prioritizing and evaluating the success of interventions. Related epidemiological studies have supported the development and assessment of interventions by improving our understanding of the risk factors for infection, generating new ideas on how foodborne illnesses are transmitted, and better defining the overall burden and human-health impact. FoodNet highlights important public health messages and works closely with regulatory and other partners to ensure that the information gained through this network is used to inform food safety policy. FoodNet trends and findings from FoodNet studies are widely disseminated using a variety of mechanisms, including annual summaries (Morbidity and Mortality Weekly Report and annual reports), journal and newsletter publications, oral and poster presentations at national and international meetings, presentations to industry and consumer groups, and the FoodNet Web site ( FUTURE DIRECTION Since 1996, FoodNet has enhanced the surveillance and investigation of foodborne infection. FoodNet is an excellent example of partnership between federal and states agencies has contributed to the improvement of food safety in the United States in numerous ways. As FoodNet moves into its second decade, it will continue to publish annually the National Report Card on Food Safety, as well as strive to improve the examination of temporal, geographical, and demographical trends. Growing interest in the burden of foodborne disease studies internationally has presented an exciting opportunity for FoodNet to expand its science base and contribute to the global estimate of the burden of foodborne disease. FoodNet will continue to use its infrastructure to respond to newly emerging public health threats and to address important foodborne disease related issues. Antimicrobial resistance and sequelae from foodborne infections remain important subjects of focus. During 2006, a joint FoodNet National Antimicrobial Resistance Monitoring System cohort study was launched to prospectively compare clinical outcomes from multidrug-resistant and pan-susceptible Salmonella infections. A FoodNet cohort study of persons with laboratory-confirmed E. coli O157 infection was launched during 2006 to determine the effect of antibiotics on the development of hemolytic uremic syndrome. This study aims to inform patient treatment practices and provide the data needed to form a consensus with regard to the relative importance of this putative risk factor. A maturation of FoodNet methods for determining and monitoring disease burden has allowed a shift in focus from the first 2 FoodNet objectives to the third objective, which is FOOD SAFETY CID 2007:44 (1 March) 723

7 the attribution of the burden of foodborne disease to specific foods and contexts. Although FoodNet case-control studies have contributed and will continue to contribute to the attribution, FoodNet is working on a variety of other approaches that will also help meet this objective. The attribution of cases of foodborne disease to specific food vehicles and other sources will be an invaluable tool for policymakers who are developing strategies aimed at reducing the impact of foodborne disease. Acknowledgments FoodNet Working Group Members at the Centers for Disease Control and Prevention, the United States Department of Agriculture s Food Safety and Inspection Service, the United States Food and Drug Administration, and the 10 FoodNet sites (California, Colorado, Connecticut, Georgia, Maryland, Minnesota, New Mexico, New York, Oregon, and Tennessee). Financial support. Centers for Disease Control and Prevention, US Department of Agriculture s Food Safety and Inspection Service, and the US Food and Drug Administration. Potential conflicts of interest. All authors: no conflicts. References 1. Mead PS, Slutsker L, Dietz V, et al. Food-related illness and death in the United States. Emerg Infect Dis 1999; 5: Hardnett FP, Hoekstra RM, Kennedy M, Charles L, Angulo FJ. Epidemiologic issues in study design and data analysis related to FoodNet activities. Clin Infect Dis 2004; 38(Suppl 3):S Allos BM, Moore MR, Griffin PM, Tauxe RV. Surveillance for sporadic foodborne disease in the 21st century: the FoodNet perspective. Clin Infect Dis 2004; 38(Suppl 3):S Hennessy TW, Marcus R, Deneen V, et al. Survey of physician diagnostic practices for patients with acute diarrhea: clinical and public health implications. Clin Infect Dis 2004; 38(Suppl 3):S Voetsch AC, Angulo FJ, Rabatsky-Ehr T, et al. Laboratory practices for stool-specimen culture for bacterial pathogens, including Escherichia coli O157:H7, in the FoodNet sites, Clin Infect Dis 2004; 38(Suppl 3):S Jones JL, Lopez A, Wahlquist SP, Nadle J, Wilson M. Survey of clinical laboratory practices for parasitic diseases. Clin Infect Dis 2004; 38(Suppl 3):S Bender JB, Smith KE, McNees AA, et al. Factors affecting surveillance data on Escherichia coli O157 infections collected from FoodNet sites, Clin Infect Dis 2004; 38(Suppl 3):S Herikstad H, Yang S, Van Gilder TJ, et al. A population-based estimate of the burden of diarrhoeal illness in the United States: FoodNet, Epidemiol Infect 2002; 129: Imhoff B, Morse D, Shiferaw B, et al. Burden of self-reported acute diarrheal illness in FoodNet surveillance areas, Clin Infect Dis 2004; 38(Suppl 3):S Scallan E, Majowicz SE, Hall G, et al. Prevalence of diarrhoea in the Community in Australia, Canada, Ireland, and the United States. Int J Epidemiol 2005;34: Roy SL, Scallan E, Beach MJ. The rate of acute gastrointestinal illness in developed countries. J Water Health 2006; 4(Suppl 2): Jones TF, McMillan MB, Scallan E, et al. A population-based estimate of the substantial burden of diarrheal disease in the United States; FoodNet, Epidemiol Infect (in press). 13. Centers for Disease Control and Prevention. Foodborne Diseases Active Surveillance Network (FoodNet): Population Survey Atlas of Exposures, Atlanta: Centers for Disease Control and Prevention, Vanden Eng J, Marcus R, Hadler JL, et al. Consumer attitudes and use of antibiotics. Emerg Infect Dis 2003; 9: Frenzen PD, DeBess EE, Hechemy KE, et al. Consumer acceptance of irradiated meat and poultry in the United States. J Food Prot 2001; 64: Green L, Selman C, Banerjee A, et al. Food service workers selfreported food preparation practices: an EHS-Net study. Int J Hyg Environ Health 2005; 208: Green LR, Selman C, Scallan E, Jones TF, Marcus R. Beliefs about meals eaten outside the home as sources of gastrointestinal illness. J Food Prot 2005; 68: Wong S, Marcus R, Hawkins M, et al. Physicians as food-safety educators: a practices and perceptions survey. Clin Infect Dis 2004; 38(Suppl 3):S Voetsch AC, Van Gilder TJ, Angulo FJ, et al. FoodNet estimate of the burden of illness caused by nontyphoidal Salmonella infections in the United States. Clin Infect Dis 2004; 38(Suppl 3):S Frenzen P, Riggs T, Buzby J, et al. Salmonella cost estimate update using FoodNet data. Food Review 1999; 22: Frenzen PD, Drake A, Angulo FJ. Economic cost of illness due to Escherichia coli O157 infections in the United States. J Food Prot 2005; 68: Flint JA, Van Duynhoven YT, Angulo FJ, et al. Estimating the burden of acute gastroenteritis, foodborne disease, and pathogens commonly transmitted by food: an international review. Clin Infect Dis 2005; 41: Rees JR, Pannier MA, McNees A, Shallow S, Angulo FJ, Vugia DJ. Persistent diarrhea, arthritis, and other complications of enteric infections: a pilot survey based on California FoodNet surveillance, Clin Infect Dis 2004; 38(Suppl 3):S Marcus R, Rabatsky-Ehr T, Mohle-Boetani JC, et al. Dramatic decrease in the incidence of Salmonella serotype Enteritidis infections in 5 FoodNet sites: Clin Infect Dis 2004; 38(Suppl 3):S Samuel MC, Vugia DJ, Shallow S, et al. Epidemiology of sporadic Campylobacter infection in the United States and declining trend in incidence, FoodNet Clin Infect Dis 2004; 38(Suppl 3): S Shiferaw B, Shallow S, Marcus R, et al. Trends in population-based active surveillance for shigellosis and demographic variability in FoodNet sites, Clin Infect Dis 2004; 38(Suppl 3):S Ray SM, Ahuja SD, Blake PA, et al. Population-based surveillance for Yersinia enterocolitica infections in FoodNet sites, : higher risk of disease in infants and minority populations. Clin Infect Dis 2004; 38(Suppl 3):S Centers for Disease Control and Prevention. Preliminary FoodNet data on the incidence of foodborne illnesses: selected sites, United States, Morb Mortal Wkly Rep 2002; 51: Centers for Disease Control and Prevention. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food: 10 States, United States, Morb Mortal Wkly Rep 2006; 55: Batz MB, Doyle MP, Morris G JR, et al. Attributing illness to food. Emerg Infect Dis 2005; 11: Friedman CR, Hoekstra RM, Samuel M, et al. Risk factors for sporadic Campylobacter infection in the United States: A case-control study in FoodNet sites. Clin Infect Dis 2004; 38(Suppl 3):S Hennessy TW, Cheng LH, Kassenborg H, et al. Egg consumption is the principal risk factor for sporadic Salmonella serotype Heidelberg infections: a case-control study in FoodNet sites. Clin Infect Dis 2004; 38(Suppl 3):S Kimura AC, Reddy V, Marcus R, et al. Chicken consumption is a newly identified risk factor for sporadic Salmonella enterica serotype Enteritidis infections in the United States: a case-control study in FoodNet sites. Clin Infect Dis 2004; 38(Suppl 3):S Mermin J, Hutwagner L, Vugia D, et al. Reptiles, amphibians, and human Salmonella infection: a population-based, case-control study. Clin Infect Dis 2004; 38(Suppl 3):S Glynn MK, Reddy V, Hutwagner, et al. Prior antimicrobial agent use increases the risk of sporadic infections with multidrug-resistant Sal- 724 CID 2007:44 (1 March) FOOD SAFETY

8 monella enterica serotype Typhimurium: a FoodNet case-control study, Clin Infect Dis 2004; 38(Suppl 3):S Rowe SY, Rocourt JR, Shiferaw B, et al. Breast-feeding decreases the risk of sporadic salmonellosis among infants in FoodNet sites. Clin Infect Dis 2004; 38(Suppl 3):S Kassenborg HD, Smith KE, Vugia DJ, et al. Fluoroquinolone-resistant Campylobacter infections: eating poultry outside of the home and foreign travel are risk factors. Clin Infect Dis 2004; 38(Suppl 3):S Kassenborg HD, Hedberg CW, Hoekstra M, et al. Farm visits and undercooked hamburgers as major risk factors for sporadic Escherichia coli O157:H7 infection: data from a case-control study in 5 FoodNet sites. Clin Infect Dis 2004; 38(Suppl 3):S Roy SL, DeLong SM, Stenzel SA, et al. Risk factors for sporadic cryptosporidiosis among immunocompetent persons in the United States from 1999 to J Clin Microbiol 2004; 42: Marcus R, Varma JK, Medus C, et al. Re-assessment of risk factors for sporadic Salmonella serotype Enteritidis infections: a case-control study in five FoodNet Sites, Epidemiol Infect 2006; Varma JK, Samuel MC, Marcus, R, et al. Listeria monocytogenes infection from foods prepared in a commercial establishment: a case-control study of potential sources for sporadic illness in the United States. Clin Infect Dis (in press). 42. Jones TF, Ingram LA, Fullerton KE, et al. A case control study of the epidemiology of sporadic Salmonella infection in infants. Pediatrics 2006; 118: Fullerton KE, Ingram LA, Jones TF, et al. Sporadic Campylobacter infection in infants: a population-based surveillance case-control study in eight FoodNet sites. Pediatr Infect Dis J 2006; 118: Gupta A, Nelson JM, Barrett TJ, et al. Antimicrobial resistance among Campylobacter strains, United States, Emerg Infect Dis 2004; 10: Jones TF, Imhoff B, Samuel M, et al. Limitations to successful investigation and reporting of foodborne outbreaks: an analysis of foodborne disease outbreaks in FoodNet catchment areas, Clin Infect Dis 2004; 38(Suppl 3):S Jones TF, Bulens SN, Gettner S, et al. Use of stool collection kits delivered to patients can improve confirmation of etiology in foodborne disease outbreaks. Clin Infect Dis 2004; 39: Varma JK, Molbak K, Barrett TJ, et al. Antimicrobial-resistant nontyphoidal Salmonella is associated with excess bloodstream infections and hospitalizations. J Infect Dis 2005; 191: Devasia RA, Varma JK, Whichard J, et al. Antimicrobial use and outcomes in patients with multidrug-resistant and pansusceptible Salmonella Newport infections, Microb Drug Resist 2005; 11: Nelson JM, Smith KE, Vugia DJ, et al. Prolonged diarrhea due to ciprofloxacin-resistant campylobacter infection. J Infect Dis 2004; 190: Varma JK, Marcus R, Stenzel SA, et al. Highly resistant Salmonella Newport-MDRAmpC transmitted through the domestic US food supply: a FoodNet case-control study of sporadic Salmonella Newport infections, J Infect Dis 2006; 194: FOOD SAFETY CID 2007:44 (1 March) 725