Estimates of foodborne illness can be used to direct food safety policy and interventions. We used data from active and passive surveillance and other sources to estimate that each year 31 major pathogens acquired in the United States caused 9.4 million episodes of foodborne illness (90% credible interval [CrI] 6.6–12.7 million), 55,961 hospitalizations (90% CrI 39,534–75,741), and 1,351 deaths (90% CrI 712–2,268). Most (58%) illnesses were caused by norovirus, followed by nontyphoidal Salmonella spp. (11%), Clostridium perfringens (10%), and Campylobacter spp. (9%). Leading causes of hospitalization were nontyphoidal Salmonella spp. (35%), norovirus (26%), Campylobacter spp. (15%), and Toxoplasma gondii (8%). Leading causes of death were nontyphoidal Salmonella spp. (28%), T. gondii (24%), Listeria monocytogenes (19%), and norovirus (11%). These estimates cannot be compared with prior (1999) estimates to assess trends because different methods were used. Additional data and more refined methods can improve future estimates.

/pdf/foodborne-illness-acquired-in-the-united-states-major-rhlh3i3qv5.pdf

Foodborne Illness Acquired in the United States—Major Pathogens

https://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736(13)60844-2.pdf

Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case-control study.

The severe acute respiratory syndrome (SARS) is responsible for the first pandemic of the 21st century. Within months after its emergence in Guangdong Province in mainland China, it had affected more than 8000 patients and caused 774 deaths in 26 countries on five continents. It illustrated dramatically the potential of air travel and globalization for the dissemination of an emerging infectious disease and highlighted the need for a coordinated global response to contain such disease threats. We review the cause, epidemiology, and clinical features of the disease.

/pdf/the-severe-acute-respiratory-syndrome-29m26drr2u.pdf

The severe acute respiratory syndrome.

To the Editor: The updated estimates of foodborne illness in the United States reported by Scallan et al. probably overestimate the occurrence of illness caused by unspecified agents because they did not account for the apparent sensitivity of the population survey to the occurrence of norovirus (1,2). The number of illnesses attributed to unspecified agents was derived from the simultaneous processes of extrapolation and subtraction: extrapolation from the population survey to create a base of diarrheal illnesses and subtraction of known agents from this base. Scallan et al. averaged illness rates from 3 successive population surveys to come up with a rate of 0.6 episodes of acute gastroenteritis per person per year. However, the individual rates were 0.49 (2000–2001), 0.54 (2002–2003), and 0.73 (2006–2007). The 2006–2007 survey was conducted at the time of widespread norovirus activity. The estimated rate of population illness was strongly correlated with the number of confirmed and suspected norovirus outbreaks reported to the Centers for Disease Control and Prevention Foodborne Disease Outbreak Surveillance System during each of the survey periods (300, 371, and 491, respectively; R2 = 0.97, p<0.0001). No other known agents were correlated with the population survey rates, and the total numbers of outbreaks were inversely correlated with the population survey data.

The strength of the correlation between norovirus outbreaks and survey results suggests that the population survey is sensitive to norovirus activity and that norovirus may account for much of what is considered to be unspecified. The fact that the highest observed population rate was ≈50% greater than the lowest rate suggests that annual variation in norovirus activity may account for a considerable proportion of what otherwise seems to be unspecified. More thorough and timely investigation and reporting of outbreaks could facilitate the development of models to evaluate the number of illnesses and update them annually.

/pdf/foodborne-illness-acquired-in-the-united-states-7ynabtz2k3.pdf

Foodborne Illness Acquired in the United States

Studies published between 1986 and 1999 indicated that rotavirus causes ≈22% (range 17%–28%) of childhood diarrhea hospitalizations. From 2000 to 2004, this proportion increased to 39% (range 29%–45%). Application of this proportion to the recent World Health Organization estimates of diarrhea-related childhood deaths gave an estimated 611,000 (range 454,000–705,000) rotavirus-related deaths.

/pdf/rotavirus-and-severe-childhood-diarrhea-1l7s8d8s0d.pdf

Rotavirus and Severe Childhood Diarrhea

The incidence of calicivirus infection in Ghana and many other African countries is not known. Thirteen (15.9%) of the 82 diarrhoeic stool samples tested for caliciviruses were positive for noroviruses (NoVs). NoVs were present in all age groups and were detected only during the diarrhoea peak that coincided with the peak rotavirus season. Ten (76.9%) of the NoV detected were genogroup II (GII) NoVs and the remaining three (23.1%) genogroup I (GI) NoVs. The predominant GII detected was GII-4 (60%, 6/10). Three of the GII NoVs were determined to be recombinants of GII-8/GII-14 as deduced from the sequencing of the region spanning the Orf1/2 junction. The GII genotypes formed four clusters with published GII sequences. The data shown enhances understanding of NoV diversity in Ghanaian children and demonstrate the global spread of distinct common genotypes to African countries. J. Med. Virol. 78:1480–1485, 2006. © 2006 Wiley-Liss, Inc.

Characterisation of norovirus strains in rural Ghanaian children with acute diarrhoea.

Background: Infectious intestinal disease (IID), usually presenting as diarrhoea and vomiting, is frequently preventable. Though often mild and self-limiting, its commonness makes IID an important public health problem. In the mid 1990s around 1 in 5 people in England suffered from IID a year, costing around £0.75 billion. No routine information source describes the UK's current community burden of IID. We present here the methods for a study to determine rates and aetiology of IID in the community, presenting to primary care and recorded in national surveillance statistics. We will also outline methods to determine whether or not incidence has declined since the mid-1990s. Methods/design: The Second Study of Infectious Intestinal Disease in the Community (IID2 Study) comprises several separate but related studies. We use two methods to describe IID burden in the community a retrospective telephone survey of self-reported illness and a prospective, all-age, population-based cohort study with weekly follow-up over a calendar year. Results from the two methods will be compared. To determine IID burden presenting to primary care we perform a prospective study of people presenting to their General Practitioner with symptoms of IID, in which we intervene in clinical and laboratory practice, and an audit of routine clinical and laboratory practice in primary care. We determine aetiology of IID using molecular methods for a wide range of gastrointestinal pathogens, in addition to conventional diagnostic microbiological techniques, and characterise isolates further through reference typing. Finally, we combine all our results to calibrate national surveillance data. Discussion: Researchers disagree about the best method(s) to ascertain disease burden. Our study will allow an evaluation of methods to determine the community burden of IID by comparing the different approaches to estimate IID incidence in its linked components. Background Infectious intestinal disease (IID), usually presenting as diarrhoea and vomiting, is frequently preventable. Its commonness makes it an important public health problem. In the mid 1990s around 1 in 5 people in England were found to suffer from IID, with around 35,000 hospital admissions and 300 deaths annually, an estimated annual cost to the nation of £0.75 billion [1]. Although IID is very common in the community, not all cases present to the healthcare system, and not all cases that present are reported to national surveillance. Many IID cases are therefore not captured by routine data sources, a pattern described as the surveillance pyramid (Figure 1). Routinely available surveillance data in the United Kingdom, and elsewhere, thus underestimate the total IID burden, and that degree of underestimation needs to be calibrated [1]. * Correspondence: sarah.o'brien@manchester.ac.uk 1 School of Translational Medicine, University of Manchester, Manchester Academic Health Science Centre, Stopford Building, Oxford Road, Manchester, M13 9PL, UK Full list of author information is available at the end of the article BioMed Central © 2010 O'Brien et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. O'Brien et al. BMC Medical Research Methodology 2010, 10:39 http://www.biomedcentral.com/1471-2288/10/39 Page 2 of 13 Most studies for estimating community disease burden in developed countries are either prospective cohort studies or retrospective cross-sectional surveys. The prospective cohort design consists of recruiting volunteers and asking them to record relevant symptoms, often in some form of diary. The retrospective study involves contacting people, usually by telephone and asking about symptoms in the recent past. A major advantage of population-based, prospective cohort studies is the ability to request stool specimens from people who report illness so that the range of gastrointestinal pathogens causing symptoms can be determined. Retrospective study designs provide no information on the microbiological causes of illness (a major weakness) but are significantly quicker and less costly to complete. Our present study builds on a previous study of IID conducted in England during the early 1990s, which underlined the public health impact of gastrointestinal infection [1]. At that time, approximately 20% of the population was found to suffer from IID annually. The only other notable prospective study to date, based on active detection of IID, was conducted in the Netherlands. Whilst in England there were an estimated 194 cases of infectious intestinal disease (IID) per 1,000 person years (equivalent to 0.2 per person per year) [1], in the Netherlands there were 283 cases per 1000 person years (equivalent to 0.3 per person per year) [2]. Although the two studies shared some common methods, the case definitions differed. A major contribution of the IID study in England was calibration of the national surveillance system, i.e. estimating the factor by which the reported number of infections with specified pathogens needed to be multiplied in order to establish the actual number of infections in the community. For every case of IID reported to national surveillance, 136 cases had occurred in the community. For campylobacters it was estimated that the ratio of cases in the community to those reported to national surveillance was 8 to 1. For salmonellas and noroviruses the corresponding ratios were 3 to 1 and over 1,500 to 1 respectively [1]. By contrast there are many more retrospective studies reported in the literature, mainly from North America [3]. These retrospective studies have been used in both outbreak and non-outbreak settings [4-8]. The studies have varied in their estimation of illness rates from 0.6 to 3.2 episodes per person year, though case definitions often differed from one study to another. For example, in the US, the overall prevalence of self-reported acute diarrhoea in the four weeks before telephone interview in randomly selected residents from defined catchment areas was 11% or 1.4 episodes of diarrhoea per person per year [6]. In Australia, in a cross-sectional telephone survey of randomly selected individuals, 11.2% of respondents reported gastroenteritis with an overall weighted incidence of 0.92 (95%CI 0.77-1.6) cases per person per year [9]. The problem is that estimates of disease burden in the community differ substantially between retrospective and prospective study designs even when using identical case definitions. This was highlighted in original the IID Study where the retrospective element of the study estimated disease incidence at 0.55 episodes per person per year and the prospective element at 0.19 [10]. At the time this discrepancy was attributed to "telescoping" whereby people remember disease episodes as being more recent than they actually are, thereby inflating the incidence over the recall period. Further concerns about the validity of retrospective surveys of diarrhoeal incidence were raised in a study centred on a waterborne outbreak [11]. However, a number of explanations, in addition to telescoping, might explain the discrepancy in incidence rates derived from prospective and retrospective studies. Potential issues with active reporting in population-based prospective cohort studies include loss to follow-up, sensitisation and reporting fatigue. In the IID Study in England 39% of the original cohort of 9,296 were lost to follow-up over 6 months and it is not clear whether loss to follow-up might be correlated with risk of illness. Sensitisation occurs when respondents become more aware of their health because they are participating in a healthrelated study [12], and may perceive more symptoms during early follow-up than before enrollment. If they subsequently become fatigued with completing a health diary, or returning data via postcard or e-mail, the number of Figure 1 The surveillance pyramid illustrating attrition of data at each level. Data appear in national statistics Positive laboratory test results People who provide faeces sample requested by GP People who go to their GP with IID and are asked to provide a sample People in the community with IID O'Brien et al. BMC Medical Research Methodology 2010, 10:39 http://www.biomedcentral.com/1471-2288/10/39 Page 3 of 13 symptoms reported over time may decline [12,13]. This pattern of sensitisation-fatigue, where illness reporting is highest during the early weeks of follow-up and subsequently decreases, is characteristic of much longitudinal data [12,14,15]. Fatigue might also have played a part in the decline in data quality observed in some studies [14,16-18]. Sensitisation and fatigue are potential problems because estimated incidence rates may vary as a function of the length of the follow-up period if either of these time-dependent conditioning effects is present [12]. For retrospective studies the main problem is accuracy of retrospective reporting. This includes not only inaccuracy in the respondent remembering whether or not they suffered the symptoms requested but whether they recall other important information such as the date of onset of symptoms, duration and severity. The potential issue of telescoping, whereby people report illness that occurred just outside the recall period as having occurred during the recall period, so increasing the estimated disease incidence, has already been raised. Conversely people might be more likely to remember events that occurred close to the interview, so reducing the estimated incidence. A further issue with both study designs is the issue of the representativeness of participants. Prospective studies can be designed to invite a representative sample of the population according to whatever criteria 

/pdf/the-second-study-of-infectious-intestinal-disease-in-the-2i1h6b4pon.pdf

The Second Study of Infectious Intestinal Disease in the Community

Of a total of 508 stool specimens from children with acute diarrhoea, mostly under the age of 5 years, collected in nine cities in the western and southeastern regions of Turkey between May 2000 and October 2002, 119 (23.4%) were found positive for rotaviruses (RV) by ELISA. Positive samples were characterized by electropherotyping and G and P genotyping. A subset of G and P types were confirmed by nucleic acid sequencing. The most prevalent types found in this collection included G4P[8], accounting for 27/64 (42.2%) of the fully characterized strains. G1P[8], G2P[4] and G3P[8] were found in 17 (26.6%), 2 (3.1%) and one (1.5%) samples respectively. Less common strains such as G9P[8] were found in two (3.2%) samples and G2P[8], G1P[6], G2P[6] and G4P[6], possible reassortant viruses, were found in five (7.8%), 2 (3.1%), one (1.5%) and four (6.3%) samples respectively. Mixed infections were found in six (7.3%) samples and were associated with combinations of G1 + G2, G1 + G4, G1 + G9 and G4 + G9 strains. This is the first molecular epidemiology study of its kind to be carried out in Turkey and suggests a significant diversity of co-circulating rotavirus strains.

Molecular characterization of rotaviruses circulating in the population in Turkey.

Evaluation of the Loopamp® (loop-mediated isothermal amplification) kit for detecting Norovirus RNA in faecal samples

The aim of this study was to apply sequence analysis of a hyper variable region of the norovirus (NoV) genome in order to identify point source outbreaks associated with suspect food or water. The hyper-variable region of the gene encoding the P2 domain was chosen as small differences in sequence are likely to indicate virus from different sources whereas identical sequence may reveal transmission routes and the source of contamination. Strains with 100% similarity were considered as originating from a common source, whereas, strains with one or more mutations in the hyper variable region sequenced were regarded as representing unrelated transmission events. This study was able to identify a point source outbreak of a dominant strain, GII-4, on a cruise ship but also of a less common strain, GII-2, between two schools. Also identical GII-3 strains were demonstrated in food handlers amongst the same outbreak; however epidemiologically related outbreaks showed different GII-3 strains indicating multiple sources of contamination.

Tracking the transmission routes of genogroup II noroviruses in suspected food-borne or environmental outbreaks of gastroenteritis through sequence analysis of the P2 domain.

Jim Gray

Papers

Characterisation of norovirus strains in rural Ghanaian children with acute diarrhoea.

The Second Study of Infectious Intestinal Disease in the Community

Molecular characterization of rotaviruses circulating in the population in Turkey.

Evaluation of the Loopamp® (loop-mediated isothermal amplification) kit for detecting Norovirus RNA in faecal samples

Tracking the transmission routes of genogroup II noroviruses in suspected food-borne or environmental outbreaks of gastroenteritis through sequence analysis of the P2 domain.