Arie H. Havelaar

The Beta Poisson dose-response model is not a single-hit model.

The choice of a dose-response model is decisive for the outcome of quantitative risk assessment. Single-hit models have played a prominent role in dose-response assessment for pathogenic microorganisms, since their introduction. Hit theory models are based on a few simple concepts that are attractive for their clarity and plausibility. These models, in particular the Beta Poisson model, are used for extrapolation of experimental dose-response data to low doses, as are often present in drinking water or food products. Unfortunately, the Beta Poisson model, as it is used throughout the microbial risk literature, is an approximation whose validity is not widely known. The exact functional relation is numerically complex, especially for use in optimization or uncertainty analysis. Here it is shown that although the discrepancy between the Beta Poisson formula and the exact function is not very large for many data sets, the differences are greatest at low doses--the region of interest for many risk applications. Errors may become very large, however, in the results of uncertainty analysis, or when the data contain little low-dose information. One striking property of the exact single-hit model is that it has a maximum risk curve, limiting the upper confidence level of the dose-response relation. This is due to the fact that the risk cannot exceed the probability of exposure, a property that is not retained in the Beta Poisson approximation. This maximum possible response curve is important for uncertainty analysis, and for risk assessment of pathogens with unknown properties.

World Health Organization Estimates of the Global and Regional Disease Burden of 22 Foodborne Bacterial, Protozoal, and Viral Diseases, 2010: A Data Synthesis

Background Foodborne diseases are important worldwide, resulting in considerable morbidity and mortality. To our knowledge, we present the first global and regional estimates of the disease burden of the most important foodborne bacterial, protozoal, and viral diseases. Methods and Findings We synthesized data on the number of foodborne illnesses, sequelae, deaths, and Disability Adjusted Life Years (DALYs), for all diseases with sufficient data to support global and regional estimates, by age and region. The data sources included varied by pathogen and included systematic reviews, cohort studies, surveillance studies and other burden of disease assessments. We sought relevant data circa 2010, and included sources from 1990–2012. The number of studies per pathogen ranged from as few as 5 studies for bacterial intoxications through to 494 studies for diarrheal pathogens. To estimate mortality for Mycobacterium bovis infections and morbidity and mortality for invasive non-typhoidal Salmonella enterica infections, we excluded cases attributed to HIV infection. We excluded stillbirths in our estimates. We estimate that the 22 diseases included in our study resulted in two billion (95% uncertainty interval [UI] 1.5–2.9 billion) cases, over one million (95% UI 0.89–1.4 million) deaths, and 78.7 million (95% UI 65.0–97.7 million) DALYs in 2010. To estimate the burden due to contaminated food, we then applied proportions of infections that were estimated to be foodborne from a global expert elicitation. Waterborne transmission of disease was not included. We estimate that 29% (95% UI 23–36%) of cases caused by diseases in our study, or 582 million (95% UI 401–922 million), were transmitted by contaminated food, resulting in 25.2 million (95% UI 17.5–37.0 million) DALYs. Norovirus was the leading cause of foodborne illness causing 125 million (95% UI 70–251 million) cases, while Campylobacter spp. caused 96 million (95% UI 52–177 million) foodborne illnesses. Of all foodborne diseases, diarrheal and invasive infections due to non-typhoidal S. enterica infections resulted in the highest burden, causing 4.07 million (95% UI 2.49–6.27 million) DALYs. Regionally, DALYs per 100,000 population were highest in the African region followed by the South East Asian region. Considerable burden of foodborne disease is borne by children less than five years of age. Major limitations of our study include data gaps, particularly in middle- and high-mortality countries, and uncertainty around the proportion of diseases that were foodborne. Conclusions Foodborne diseases result in a large disease burden, particularly in children. Although it is known that diarrheal diseases are a major burden in children, we have demonstrated for the first time the importance of contaminated food as a cause. There is a need to focus food safety interventions on preventing foodborne diseases, particularly in low- and middle-income settings.

Disease burden of foodborne pathogens in the Netherlands, 2009

To inform risk management decisions on control, prevention and surveillance of foodborne disease, the disease burden of foodborne pathogens is estimated using Disability Adjusted Life Years as a summary metric of public health. Fourteen pathogens that can be transmitted by food are included in the study (four infectious bacteria, three toxin-producing bacteria, four viruses and three protozoa). Data represent the burden in the Netherlands in 2009. The incidence of community-acquired non-consulting cases, patients consulting their general practitioner, those admitted to hospital, as well as the incidence of sequelae and fatal cases is estimated using surveillance data, cohort studies and published data. Disease burden includes estimates of duration and disability weights for non-fatal cases and loss of statistical life expectancy for fatal cases. Results at pathogen level are combined with data from an expert survey to assess the fraction of cases attributable to food, and the main food groups contributing to transmission. Among 1.8 million cases of disease (approx. 10,600 per 100,000) and 233 deaths (1.4 per 100,000) by these fourteen pathogens, approximately one-third (680,000 cases; 4100 per 100,000) and 78 deaths (0.5 per 100,000) are attributable to foodborne transmission. The total burden is 13,500 DALY (82 DALY per 100,000). On a population level, Toxoplasma gondii, thermophilic Campylobacter spp., rotaviruses, noroviruses and Salmonella spp. cause the highest disease burden. The burden per case is highest for perinatal listeriosis and congenital toxoplasmosis. Approximately 45% of the total burden is attributed to food. T. gondii and Campylobacter spp. appear to be key targets for additional intervention efforts, with a focus on food and environmental pathways. The ranking of foodborne pathogens based on burden is very different compared to when only incidence is considered. The burden of acute disease is a relatively small part of the total burden. In the Netherlands, the burden of foodborne pathogens is similar to the burden of upper respiratory and urinary tract infections.

World Health Organization Global Estimates and Regional Comparisons of the Burden of Foodborne Disease in 2010

Illness and death from diseases caused by contaminated food are a constant threat to public health and a significant impediment to socio-economic development worldwide. To measure the global and regional burden of foodborne disease (FBD), the World Health Organization (WHO) established the Foodborne Disease Burden Epidemiology Reference Group (FERG), which here reports their first estimates of the incidence, mortality, and disease burden due to 31 foodborne hazards. We find that the global burden of FBD is comparable to those of the major infectious diseases, HIV/AIDS, malaria and tuberculosis. The most frequent causes of foodborne illness were diarrheal disease agents, particularly norovirus and Campylobacter spp. Diarrheal disease agents, especially non-typhoidal Salmonella enterica, were also responsible for the majority of deaths due to FBD. Other major causes of FBD deaths were Salmonella Typhi, Taenia solium and hepatitis A virus. The global burden of FBD caused by the 31 hazards in 2010 was 33 million Disability Adjusted Life Years (DALYs); children under five years old bore 40% of this burden. The 14 subregions, defined on the basis of child and adult mortality, had considerably different burdens of FBD, with the greatest falling on the subregions in Africa, followed by the subregions in South-East Asia and the Eastern Mediterranean D subregion. Some hazards, such as non-typhoidal S. enterica, were important causes of FBD in all regions of the world, whereas others, such as certain parasitic helminths, were highly localised. Thus, the burden of FBD is borne particularly by children under five years old–although they represent only 9% of the global population–and people living in low-income regions of the world. These estimates are conservative, i.e., underestimates rather than overestimates; further studies are needed to address the data gaps and limitations of the study. Nevertheless, all stakeholders can contribute to improvements in food safety throughout the food chain by incorporating these estimates into policy development at national and international levels.

Assessment of the risk of infection by Cryptosporidium or Giardia in drinking water from a surface water source

The significance of the presence in drinking water of the protozoan microparasites Cryptosporidium parvum and Giardia lamblia for public health may be analyzed by means of risk assessment. This requires quantitative knowledge of all the contributing factors, from the concentration of these organisms in the source water to the dose-response relation for the probability of infection or disease in a human host. The major contributing factors are: the concentration of cysts or occysts in raw water, the recovery of the detection method, the viability of recovered cysts or oocysts, the removal of organisms in the treatment process, and the daily consumption of unboiled tap water. To enable analysis of the uncertainty in the calculated risk of infection, each of these factors is treated as a stochastic variable, for which a suitable distribution is proposed. A frequency distribution for the probability of infection is then constructed with standard sampling techniques. This first evaluation of the calculation of the risk of infection due to exposure to Cryptosporidium oocysts and Giardia cysts via drinking water, shows that the uncertainty in the estimated removal efficiency of the treatment process dominates over uncertainities in other contributing factors.

Disability weights for the Global Burden of Disease 2013 study

BACKGROUND The Global Burden of Disease (GBD) study assesses health losses from diseases, injuries, and risk factors using disability-adjusted life-years, which need a set of disability weights to quantify health levels associated with non-fatal outcomes. The objective of this study was to estimate disability weights for the GBD 2013 study. METHODS We analysed data from new web-based surveys of participants aged 18-65 years, completed in four European countries (Hungary, Italy, the Netherlands, and Sweden) between Sept 23, 2013, and Nov 11, 2013, combined with data previously collected in the GBD 2010 disability weights measurement study. Surveys used paired comparison questions for which respondents considered two hypothetical individuals with different health states and specified which person they deemed healthier than the other. These surveys covered 183 health states pertinent to GBD 2013; of these states, 30 were presented with descriptions revised from previous versions and 18 were new to GBD 2013. We analysed paired comparison data using probit regression analysis and rescaled results to disability weight units between 0 (no loss of health) and 1 (loss equivalent to death). We compared results with previous estimates, and an additional analysis examined sensitivity of paired comparison responses to duration of hypothetical health states. FINDINGS The total analysis sample consisted of 30 230 respondents from the GBD 2010 surveys and 30 660 from the new European surveys. For health states common to GBD 2010 and GBD 2013, results were highly correlated overall (Pearsons r 0·992 [95% uncertainty interval 0·989-0·994]). For health state descriptions that were revised for this study, resulting disability weights were substantially different for a subset of these weights, including those related to hearing loss (eg, complete hearing loss: GBD 2010 0·033 [0·020-0·052]; GBD 2013 0·215 [0·144-0·307]) and treated spinal cord lesions (below the neck: GBD 2010 0·047 [0·028-0·072]; GBD 2013 0·296 [0·198-0·414]; neck level: GBD 2010 0·369 [0·243-0·513]; GBD 2013 0·589 [0·415-0·748]). Survey responses to paired comparison questions were insensitive to whether the comparisons were framed in terms of temporary or chronic outcomes (Pearsons r 0·981 [0·973-0·987]). INTERPRETATION This study substantially expands the empirical basis for assessment of non-fatal outcomes in the GBD study. Findings from this study substantiate the notion that disability weights are sensitive to particular details in descriptions of health states, but robust to duration of outcomes. FUNDING European Centre for Disease Prevention and Control, Bill and Melinda Gates Foundation.

Effect of accelerated natural lactic fermentation of infant food ingredients on some pathogenic microorganisms.

Accelerated natural lactic fermentation in mixtures of water and ground sorghum, millet and pigeon pea was obtained by gradual selection of lactic acid bacteria, through inoculum recycling. Weaning food prepared from ingredients fermented this way, contained approx. 0.7% lactic and 0.05% acetic acids and had a pH of about 3.8. In porridges, a pH of less than or equal to 4.0 was required to cause death of Salmonella typhimurium and Staphylococcus aureus. Several intestinal pathogenic bacteria were inoculated into sour porridge. The most resistant Salmonella sp. died at a rate of 1.2 log cycle/h; the most resistant Shigella sp. at 0.9 log cycle/h; and the most resistant Escherichia coli strain at 0.6 log cycle/h. A yeast, Candida albicans, could grow well in the sour product, whereas a bacteriophage (MS-2) was inactivated at a rate of 0.1 log cycle/h. In the acid-sensitive bacterial cultures, no gradual adaptation to acid environments could be observed. The survival studies were carried out at 30 degrees C.

A comparison of risk assessments on Campylobacter in broiler meat.

In recent years, several quantitative risk assessments for Campylobacter in broiler meat have been developed to support risk managers in controlling this pathogen. The models encompass some or all of the consecutive stages in the broiler meat production chain: primary production, industrial processing, consumer food preparation, and the dose-response relationship. The modelling approaches vary between the models, and this has supported the progress of risk assessment as a research discipline. The risk assessments are not only used to assess the human incidence of campylobacteriosis due to contaminated broiler meat, but more importantly for analyses of the effects of control measures at different stages in the broiler meat production chain. This review paper provides a comparative overview of models developed in the United Kingdom, Denmark, the Netherlands and Germany, and aims to identify differences and similarities of these existing models. Risk assessments developed for FAO/WHO and in New Zealand are also briefly discussed. Although the dynamics of the existing models may differ substantially, there are some similar conclusions shared between all models. The continuous introduction of Campylobacter in flocks implies that monitoring for Campylobacter at the farm up to one week before slaughter may result in flocks that are falsely tested negative: once Campylobacter is established at the farm, the within-flock prevalence increases dramatically within a week. Consequently, at the point of slaughter, the prevalence is most likely to be either very low (<5%) or very high (>95%). In evaluating control strategies, all models find a negligible effect of logistic slaughter, the separate processing of positive and negative flocks. Also, all risk assessments conclude that the most effective intervention measures aim at reducing the Campylobacter concentration, rather than reducing the prevalence. During the stage where the consumer handles the food, cross-contamination is generally considered to be more relevant than undercooking. An important finding, shared by all, is that the tails of the distributions describing the variability in Campylobacter concentrations between meat products and meals determine the risks, not the mean values of those distributions. Although a unified model for risk assessment of Campylobacter in the broiler meat production would be desirable in order to promote a European harmonized approach, it is neither feasible nor desirable to merge the different models into one generic risk assessment model. The purpose of such a generic model has yet to be defined at a European level and the large variety in practices between countries, especially related to consumer food preparation and consumption, complicates a unified approach.

Future challenges to microbial food safety

Despite significant efforts by all parties involved, there is still a considerable burden of foodborne illness, in which micro-organisms play a prominent role. Microbes can enter the food chain at different steps, are highly versatile and can adapt to the environment allowing survival, growth and production of toxic compounds. This sets them apart from chemical agents and thus their study from food toxicology. We summarize the discussions of a conference organized by the Dutch Food and Consumer Products Safety Authority and the European Food Safety Authority. The goal of the conference was to discuss new challenges to food safety that are caused by micro-organisms as well as strategies and methodologies to counter these. Management of food safety is based on generally accepted principles of Hazard Analysis Critical Control Points and of Good Manufacturing Practices. However, a more pro-active, science-based approach is required, starting with the ability to predict where problems might arise by applying the risk analysis framework. Developments that may influence food safety in the future occur on different scales (from global to molecular) and in different time frames (from decades to less than a minute). This necessitates development of new risk assessment approaches, taking the impact of different drivers of change into account. We provide an overview of drivers that may affect food safety and their potential impact on foodborne pathogens and human disease risks. We conclude that many drivers may result in increased food safety risks, requiring active governmental policy setting and anticipation by food industries whereas other drivers may decrease food safety risks. Monitoring of contamination in the food chain, combined with surveillance of human illness and epidemiological investigations of outbreaks and sporadic cases continue to be important sources of information. New approaches in human illness surveillance include the use of molecular markers for improved outbreak detection and source attribution, sero-epidemiology and disease burden estimation. Current developments in molecular techniques make it possible to rapidly assemble information on the genome of various isolates of microbial species of concern. Such information can be used to develop new tracking and tracing methods, and to investigate the behavior of micro-organisms under environmentally relevant stress conditions. These novel tools and insight need to be applied to objectives for food safety strategies, as well as to models that predict microbial behavior. In addition, the increasing complexity of the global food systems necessitates improved communication between all parties involved: scientists, risk assessors and risk managers, as well as consumers.

Assessment of the dose-response relationship of Campylobacter jejuni

Mathematical relations describing the risk of infection after exposure to enteropathogens are important tools for the evaluation of the potential health risk from exposure via food and water. A quantitative description of the dose-response relation for Campylobacter jejuni with the Beta-Poisson model was fitted to experimental data of infection with Campylobacter jejuni (as determined by shedding of C. jejuni) obtained in human feeding studies performed by Black et al. (1988). The maximum likelihood estimates for the Beta-Poisson model parameters based on these data are: alpha = 0.145 and beta = 7.59. The fit of the model on the experimental data was good: the difference between the likelihood obtained with the Beta-Poisson model and the maximum possible likelihood was not significant. The occurrence of symptoms of intestinal illness did not follow a similar dose-related trend. Overall, 22% of the infected volunteers developed symptoms (diarrhea, fever). The highest illness-to-infection ratio was found at an intermediate dose (9 x 10(4)). The dose-response relation and the illness-to-infection ratio appeared to differ between different C. jejuni isolates. The dose-response relation derived from feeding studies with a single isolate should therefore be considered indicative. The absence of experimental data in the low dose range resulted in a relatively large confidence interval at low doses. However, in cases where the dose-response relation has been applied so far to estimate the health risk of exposure to C. jejuni in water, the uncertainty in the dose-response relation was insignificant compared to the uncertainty in the exposure estimate.