Characterising Biosecurity Initiatives Globally to Support the Development of a Progressive Management Pathway for Terrestrial Animals: A Scoping Review.

While biosecurity is of increasing importance globally, there is still limited evidence of the factors or elements that support the progressive and sustainable scaling up of biosecurity along the value chains from the local to the global level. To gain insight into the current body of literature on biosecurity, a mixed-methods approach was used based on a scoping literature review and an online survey with subject matter experts. Six databases were searched for published literature, and textual information from titles and abstracts of all included records (n = 266) were analysed through inductive content analysis to build biosecurity-relevant categories and identify strengths, weaknesses, opportunities, and threats (SWOT) of existing biosecurity systems or initiatives (such as projects or programs). Most records focused on initiatives in high-income countries, traditional livestock species (pigs, poultry, and large ruminants), and the production stage and had a disease-specific focus. No records described a comprehensive or global framework to progressively scale up biosecurity. Overall, the findings highlight the need for initiatives such as the FAO Progressive Management Pathway for Terrestrial Animal Biosecurity (FAO-PMP-TAB), which is a stepwise approach for strengthening biosecurity management along value chains to enhance the health, resilience, and sustainability of animal sectors. The findings highlight important elements and provide recommendations useful for developing approaches or a global framework to progressively improve biosecurity management.

Development of a transboundary model of livestock disease in Europe.

Epidemiological models of notifiable livestock disease are typically framed at a national level and targeted for specific diseases. There are inherent difficulties in extending models beyond national borders as details of the livestock population, production systems and marketing systems of neighbouring countries are not always readily available. It can also be a challenge to capture heterogeneities in production systems, control policies, and response resourcing across multiple countries, in a single transboundary model. In this paper, we describe EuFMDiS, a continental-scale modelling framework for transboundary animal disease, specifically designed to support emergency animal disease planning in Europe. EuFMDiS simulates the spread of livestock disease within and between countries and allows control policies to be enacted and resourced on a per-country basis. It provides a sophisticated decision support tool that can be used to look at the risk of disease introduction, establishment and spread; control approaches in terms of effectiveness and costs; resource management; and post-outbreak management issues.

Assessment of Evaluation Tools for Integrated Surveillance of Antimicrobial Use and Resistance Based on Selected Case Studies.

Regular evaluation of integrated surveillance for antimicrobial use (AMU) and resistance (AMR) in animals, humans, and the environment is needed to ensure system effectiveness, but the question is how. In this study, six different evaluation tools were assessed after being applied to AMU and AMR surveillance in eight countries: (1) ATLASS: the Assessment Tool for Laboratories and AMR Surveillance Systems developed by the Food and Agriculture Organization (FAO) of the United Nations, (2) ECoSur: Evaluation of Collaboration for Surveillance tool, (3) ISSEP: Integrated Surveillance System Evaluation Project, (4) NEOH: developed by the EU COST Action "Network for Evaluation of One Health," (5) PMP-AMR: The Progressive Management Pathway tool on AMR developed by the FAO, and (6) SURVTOOLS: developed in the FP7-EU project "RISKSUR." Each tool was scored using (i) 11 pre-defined functional aspects (e.g., workability concerning the need for data, time, and people); (ii) a strengths, weaknesses, opportunities, and threats (SWOT)-like approach of user experiences (e.g., things that I liked or that the tool covered well); and (iii) eight predefined content themes related to scope (e.g., development purpose and collaboration). PMP-AMR, ATLASS, ECoSur, and NEOH are evaluation tools that provide a scoring system to obtain semi-quantitative results, whereas ISSEP and SURVTOOLS will result in a plan for how to conduct evaluation(s). ISSEP, ECoSur, NEOH, and SURVTOOLS allow for in-depth analyses and therefore require more complex data, information, and specific training of evaluator(s). PMP-AMR, ATLASS, and ISSEP were developed specifically for AMR-related activities-only ISSEP included production of a direct measure for "integration" and "impact on decision making." NEOH and ISSEP were perceived as the best tools for evaluation of One Health (OH) aspects, and ECoSur as best for evaluation of the quality of collaboration. PMP-AMR and ATLASS seemed to be the most user-friendly tools, particularly designed for risk managers. ATLASS was the only tool focusing specifically on laboratory activities. Our experience is that adequate resources are needed to perform evaluation(s). In most cases, evaluation would require involvement of several assessors and/or stakeholders, taking from weeks to months to complete. This study can help direct future evaluators of integrated AMU and AMR surveillance toward the most adequate tool for their specific evaluation purpose.

Redesigning the serological surveillance program for notifiable avian influenza in Belgian professional poultry holdings.

This study was aimed at redesigning the Belgian active surveillance program for domestic birds in professional poultry holdings based on a risk analysis approach. A stochastic quantitative analysis, combining all data sources, was run to obtain sensitivity estimates for the detection of an infected bird in the different risk groups identified. An optimal number of holdings for each risk group was then estimated on the basis of the different sensitivities obtained. This study proved to be a useful tool for decision makers, providing insight on how to reallocate the total amount of samples to be taken in the coming year(s) in Belgium, thus optimizing the field resources and improving efficiency of disease surveillance such as required by the international standards.

Evaluating integrated surveillance of antimicrobial resistance: experiences from use of three evaluation tools.

BACKGROUND: Integrated antimicrobial resistance (AMR) surveillance programmes require regular evaluation to ensure they are fit for purpose and that all actors understand their responsibilities. This will strengthen their relevance for the clinical setting, which depends heavily on continued access to effective treatment options. Several evaluation tools addressing different surveillance aspects are available. OBJECTIVES: The aim was to understand the strengths and weaknesses of three evaluation tools, and to improve guidance on how to choose a fit-for-purpose tool. SOURCES: Three tools were assessed: (a) AMR-PMP-the Progressive Management Pathway tool on AMR developed by the Food and Agriculture Organization (FAO) of the United Nations, (b) NEOH developed by the EU COST Action 'Network for Evaluation of One Health' and (c) SURVTOOLS developed in an FP7-EU project 'RISKSUR'. Each tool was assessed with regard to contents, required evaluation processes including stakeholder engagement and resource demands, integration coverage across relevant sectors and applicability. They were compared using a predefined scoring scheme and a strengths-weaknesses-opportunities-threats (SWOT)-like format for commenting. CONTENT: All three tools address multiple decision-making levels and aspects of stakeholder engagement. NEOH focuses on system features, learning, sharing, leadership and infrastructure, and requires a description of the underlying system in which AMR develops. AMR-PMP focuses on four areas: awareness, evidence, governance and practices and assesses the implementation degree of pre-chosen aspects within these areas. This requires less of the evaluator, but warrants participation of multiple stakeholders. SURVTOOL provides information and references on how to evaluate effectiveness, process and comprehensiveness of surveillance programmes. All three tools require veterinary epidemiology expertise and varying levels of evaluation methodology training to use appropriately. IMPLICATIONS: The tools covered AMR surveillance and One Health aspects to varying degrees. This study provides guidance on aspects to consider when choosing between available tools and embarking on an evaluation of integrated surveillance.

Development of a quantitative microbial risk assessment for human salmonellosis through household consumption of fresh minced pork meat in Belgium.

A quantitative microbial risk assessment (QMRA) according to the Codex Alimentarius Principles is conducted to evaluate the risk of human salmonellosis through household consumption of fresh minced pork meat in Belgium. The quantitative exposure assessment is carried out by building a modular risk model, called the METZOON-model, which covers the pork production from farm to fork. In the METZOON-model, the food production pathway is split up in six consecutive modules: (1) primary production, (2) transport and lairage, (3) slaughterhouse, (4) postprocessing, (5) distribution and storage, and (6) preparation and consumption. All the modules are developed to resemble as closely as possible the Belgian situation, making use of the available national data. Several statistical refinements and improved modeling techniques are proposed. The model produces highly realistic results. The baseline predicted number of annual salmonellosis cases is 20,513 (SD 9061.45). The risk is estimated higher for the susceptible population (estimate 4.713 x 10(-5); SD 1.466 x 10(-5)) compared to the normal population (estimate 7.704 x 10(-6); SD 5.414 x 10(-6)) and is mainly due to undercooking and to a smaller extent to cross-contamination in the kitchen via cook's hands.

NUSAP method for evaluating the data quality in a quantitative microbial risk assessment model for Salmonella in the pork production chain.

The numeral unit spread assessment pedigree (NUSAP) system was implemented to evaluate the quality of input parameters in a quantitative microbial risk assessment (QMRA) model for Salmonella spp. in minced pork meat. The input parameters were grouped according to four successive exposure pathways: (1) primary production (2) transport, holding, and slaughterhouse, (3) postprocessing, distribution, and storage, and (4) preparation and consumption. An inventory of 101 potential input parameters was used for building the QMRA model. The characteristics of each parameter were defined using a standardized procedure to assess (1) the source of information, (2) the sampling methodology and sample size, and (3) the distributional properties of the estimate. Each parameter was scored by a panel of experts using a pedigree matrix containing four criteria (proxy, empirical basis, method, and validation) to assess the quality, and this was graphically represented by means of kite diagrams. The parameters obtained significantly lower scores for the validation criterion as compared with the other criteria. Overall strengths of parameters related to the primary production module were significantly stronger compared to the other modules (the transport, holding, and slaughterhouse module, the processing, distribution, and storage module, and the preparation and consumption module). The pedigree assessment contributed to select 20 parameters, which were subsequently introduced in the QMRA model. The NUSAP methodology and kite diagrams are objective tools to discuss and visualize the quality of the parameters in a structured way. These two tools can be used in the selection procedure of input parameters for a QMRA, and can lead to a more transparent quality assurance in the QMRA.

Evaluation of antibody-ELISA and real-time RT-PCR for the diagnosis and profiling of bluetongue virus serotype 8 during the epidemic in Belgium in 2006.

In 2006 bluetongue (BT) emerged for the first time in North-Western Europe. Reliable diagnostic tools are essential in controlling BT but data on the diagnostic sensitivity (Se) and specificity (Sp) are often missing. This paper aims to describe and analyse the results obtained with the diagnostics used in Belgium during the 2006 BT crisis. The diagnosis was based on a combination of antibody detection (competitive ELISA, cELISA) and viral RNA detection by real-time RT-PCR (RT-qPCR). The performance of the cELISA as a diagnostic tool was assessed on field results obtained during the epidemic and previous surveillance campaigns. As the infectious status of the animals is unknown during an epidemic, a Bayesian analysis was performed. Both assays were found to be equally specific (RT-qPCR: 98.5%; cELISA: 98.2%) while the diagnostic sensitivity of the RT-qPCR (99.5%) was superior to that of the cELISA (87.8%). The assumption of RT-qPCR as standard of comparison during the bluetongue virus (BTV) epidemic proved valid based on the results of the Bayesian analysis. A ROC analysis of the cELISA, using RT-qPCR as standard of comparison, showed that the cut-off point with the highest accuracy occurred at a percentage negativity of 66, which is markedly higher than the cut-off proposed by the manufacturer. The analysis of the results was further extended to serological and molecular profiling and the possible use of profiling as a rapid epidemiological marker of the BTV in-field situation was assessed. A comparison of the serological profiles obtained before, during and at the end of the Belgian epidemic clearly showed the existence of an intermediate zone which appears soon after BTV (re)enters the population. The appearance or disappearance of this intermediate zone is correlated with virus circulation and provides valuable information, which would be entirely overlooked if only positive and negative results were considered.

Salmonella in Belgian laying hens: an identification of risk factors.

Since the 1980s, the prevalence of Salmonella in Belgian poultry layers and broilers has greatly fluctuated with a rise observed in 2003 and a significant decrease in 2005. In order to alleviate the risk at egg consumer level, it is crucial to understand the factors which influence the contamination and the spread of Salmonella in laying hens. To study such determinants we explored the Belgian data from the 2005 baseline study on the prevalence of Salmonella in laying flocks of Gallus gallus in the European Union. The response variables corresponded to presence or absence of Salmonella from dust and faecal samples taken from the environment of a Belgian layer flock. The explanatory variables included: region of Belgium, sampling time (month the flock was sampled), production type (cage or barn and free range), Salmonella vaccination status, flock age and flock size. Analyses of these data were performed using a bivariate logistic regression model assuming independence between the two responses and bivariate generalized estimating equations model, which incorporates the correlation between the two responses on the same flock. The main risk factor that was identified was rearing flocks in cages compared to barns and free-range systems. The results also showed a significant higher risk for Salmonella for a 1 week increase in flocks' age as well as with a unit increase in the size of the flock.

A wind density model to quantify the airborne spread of Culicoides species during north-western Europe bluetongue epidemic, 2006.

Increased transport and trade as well as climate shifts play an important role in the introduction, establishment and spread of new pathogens. Arguably, the introduction of bluetongue virus (BTV) serotype 8 in Benelux, Germany and France in 2006 is such an example. After its establishment in receptive local vector and host populations the continued spread of such a disease in a suitable environment will mainly depend on movement of infected vectors and animals. In this paper we explore how wind models can contribute to explain the spread of BTV in a temperate eco-climatic setting. Based on previous work in Greece and Bulgaria filtered wind density maps were computed using data from the European Centre for Medium-Range Weather Forecasts (ECMWF). Six hourly forward wind trajectories were computed at pressure levels of 850 hPa for each infected farm as from the recorded onset of symptoms. The trajectories were filtered to remove wind events that do not contribute to possible spread of the vector. The suitable wind events were rastered and aggregated on a weekly basis to obtain weekly wind density maps. Next to this, cumulated wind density maps were also calculated to assess the overall impact of wind dispersal of vectors. A strong positive correlation was established between wind density data and the horizontal asymmetrical spread pattern of the 2006 BTV8 epidemic. It was shown that short (<5 km), medium (5-31 km) and long (>31 km) distance spread had a different impact on disease spread. Computed wind densities were linked to the medium/long-distance spread whilst short range spread was mainly driven by active Culicoides flight. Whilst previous work in the Mediterranean basin showed that wind driven spread of Culicoides over sea occurred over distances of up to 700 km, this phenomenon was not observed over land. Long-distance spread over land followed a hopping pattern, i.e. with intermediary stops and establishment of local virus circulation clusters at distances of 35-85 km. Despite suitable wind densities, no long range spread was recorded over distances of 300-400 km. Factors preventing spread Eastwards to the UK and Northwards to Denmark during the 2006 epidemic are discussed. Towards the east both elevation and terrain roughness, causing air turbulences and drop down of Culicoides, were major factors restricting spread. It is concluded that the proposed approach opens new avenues for understanding the spread of vector-borne viruses in Europe. Future developments should take into consideration both physical and biological factors affecting spread.

Field observations during the Bluetongue serotype 8 epidemic in 2006. II. Morbidity and mortality rate, case fatality and clinical recovery in sheep and cattle in the Netherlands.

Data collected in the Netherlands during the Bluetongue serotype 8 (BTV-8) epidemic indicated that in outbreak cattle herds, predominantly dairy and nursing cows were clinically affected and not young stock, beef cattle, beef calves, or breeding animals. In outbreak sheep flocks, mainly ewes and--if present--rams, were clinically affected and not the lambs. Median morbidity rate in outbreak herds was 1.85 per 100 sheep-month at risk and 0.32 per 100 cattle-month at risk for sheep and cattle, respectively. The mean proportion of BT-affected animals in outbreak herds that recovered from clinical disease was approximately eight times higher for cattle compared to sheep in the Netherlands. Median mortality rate in outbreak herds was 0.5 per 100 sheep-month at risk of dying and 0 per 100 cattle-month at risk of dying for sheep and cattle, respectively. Median recovery time of both sheep and cattle that recovered from clinical disease in outbreak herds was 14 days. Median case fatality was 50% in sheep outbreak flocks and 0% in outbreak cattle herds. It is concluded that morbidity and mortality in outbreak cattle herds was very limited during the BTV-8 epidemic in the Netherlands in 2006. In outbreak sheep flocks, morbidity was limited, with exceptions for a few flocks. However, almost 50% of the clinically sick sheep died in outbreak sheep herds.

Field observations during the bluetongue serotype 8 epidemic in 2006. I. Detection of first outbreaks and clinical signs in sheep and cattle in Belgium, France and the Netherlands.

Starting August 2006, a major epidemic of bluetongue (BT) was identified in North-West Europe, affecting The Netherlands, Belgium, Germany, Luxembourg and the North of France. It was caused by BT virus serotype 8 (BTV-8), a serotype previously unknown to the European Union (EU). In this outbreak, the virus caused clinical disease in a few individual animals within cattle herds, whereas overt clinical disease was usually restricted to sheep. Investigations in Belgium suggested that the first clinical signs of BTV-8 appeared mid July 2006 in a cattle herd, while the first suspicion of a BT-outbreak in Belgium was reported on 17 August 2006. In the first 10 BTV-8 outbreaks in the Netherlands, the owners indicated that the first clinical signs started approximately 12-17 days before a suspicion was reported to the veterinary authorities via a veterinary practitioner. In BTV-8 affected sheep flocks, erosions of the oral mucosa, fever, salivation, facial and mandibular oedema, apathy and tiredness, mortality, oedema of the lips, lameness, and dysphagia were among the most frequent clinical signs recorded. The most prominent clinical signs in BTV-8 affected cattle herds were: crusts/lesions of the nasal mucosa, erosions of lips/crusts in or around the nostrils, erosions of the oral mucosa, salivation, fever, conjunctivitis, coronitis, muscle necrosis, and stiffness of the limbs. Crusts/lesions of nasal mucosa, conjunctivitis, hyperaemic/purple coloration and lesions of the teats, and redness/hypersensitivity of the skin were relatively more seen on outbreak farms with cattle compared to sheep. Mortality, oedema of the head and ears, coronitis, redness of the oral mucosa, erosions/ulceration of tongue mucosa, purple coloration of the tongue and tongue protrusion and dyspneu were relatively more seen on outbreak farms with sheep compared to cattle.

Human salmonellosis: estimation of dose-illness from outbreak data.

The quantification of the relationship between the amount of microbial organisms ingested and a specific outcome such as infection, illness, or mortality is a key aspect of quantitative risk assessment. A main problem in determining such dose-response models is the availability of appropriate data. Human feeding trials have been criticized because only young healthy volunteers are selected to participate and low doses, as often occurring in real life, are typically not considered. Epidemiological outbreak data are considered to be more valuable, but are more subject to data uncertainty. In this article, we model the dose-illness relationship based on data of 20 Salmonella outbreaks, as discussed by the World Health Organization. In particular, we model the dose-illness relationship using generalized linear mixed models and fractional polynomials of dose. The fractional polynomial models are modified to satisfy the properties of different types of dose-illness models as proposed by Teunis et al. Within these models, differences in host susceptibility (susceptible versus normal population) are modeled as fixed effects whereas differences in serovar type and food matrix are modeled as random effects. In addition, two bootstrap procedures are presented. A first procedure accounts for stochastic variability whereas a second procedure accounts for both stochastic variability and data uncertainty. The analyses indicate that the susceptible population has a higher probability of illness at low dose levels when the combination pathogen-food matrix is extremely virulent and at high dose levels when the combination is less virulent. Furthermore, the analyses suggest that immunity exists in the normal population but not in the susceptible population.

The dog and cat population on Maio Island, Cape Verde: characterisation and prediction based on household survey and remotely sensed imagery.

The objective was to estimate and characterise the dog and cat population on Maio Island, Cape Verde. Remotely sensed imagery was used to document the number of houses across the island and a household survey was carried out in six administrative areas recording the location of each animal using a global positioning system instrument. Linear statistical models were applied to predict the dog and cat populations based on the number of houses found and according to various levels of data aggregation. In the surveyed localities, a total of 457 dogs and 306 cats were found. The majority of animals had owners and only a few had free access to outdoor activities. The estimated population size was 531 dogs [95% confidence interval (CI): 453-609] and 354 cats (95% CI: 275-431). Stray animals were not a concern on the island in contrast to the rest of the country.

Analytical performance of several classical swine fever laboratory diagnostic techniques on live animals for detection of infection.

The diagnostic properties of several assays on live animals were studied using data from different experiments. These experiments involved 128 classical swine fever virus (CSFV) infected pigs (weaner pigs, fatteners and sows). Since all pigs in the study were infected with CSFV, only the proportion of test positive results and the time until a test positive result is obtained were evaluated. The RT-nPCR detected the highest proportion of infected pigs (98.9%), whereas the Antigen ELISA gave the worst detection results (74.7%). Within the group of test positive animals, infection was detected earliest using the leukocyte count and latest using Antigen ELISA. Using the virus neutralisation test, antibodies against CSFV were detectable on average 7.6 days after the onset of viraemia in virus isolation in whole blood. Using survival analysis, the time until the first positive diagnosis and the proportion of detected animals were combined in one test. Results showed that RT-nPCR performed significantly better than either virus isolation in different blood fractions or antigen ELISA. It is concluded that the RT-nPCR technique is the best diagnostic tool available for early detection of a classical swine fever infection.

Indicators for tracking European vulnerabilities to the risks of infectious disease transmission due to climate change.

A wide range of infectious diseases may change their geographic range, seasonality and incidence due to climate change, but there is limited research exploring health vulnerabilities to climate change. In order to address this gap, pan-European vulnerability indices were developed for 2035 and 2055, based upon the definition vulnerability = impact/adaptive capacity. Future impacts were projected based upon changes in temperature and precipitation patterns, whilst adaptive capacity was developed from the results of a previous pan-European study. The results were plotted via ArcGISTM to EU regional (NUTS2) levels for 2035 and 2055 and ranked according to quintiles. The models demonstrate regional variations with respect to projected climate-related infectious disease challenges that they will face, and with respect to projected vulnerabilities after accounting for regional adaptive capacities. Regions with higher adaptive capacities, such as in Scandinavia and central Europe, will likely be better able to offset any climate change impacts and are thus generally less vulnerable than areas with lower adaptive capacities. The indices developed here provide public health planners with information to guide prioritisation of activities aimed at strengthening regional preparedness for the health impacts of climate change. There are, however, many limitations and uncertainties when modeling health vulnerabilities. To further advance the field, the importance of variables such as coping capacity and governance should be better accounted for, and there is the need to systematically collect and analyse the interlinkages between the numerous and ever-expanding environmental, socioeconomic, demographic and epidemiologic datasets so as to promote the public health capacity to detect, forecast, and prepare for the health threats due to climate change.

Bluetongue surveillance system in Belgium: a stochastic evaluation of its risk-based approach effectiveness.

BACKGROUND: The aim of this study was to assess the sensitivity of the four major bluetongue surveillance components implemented in Belgium in 2007 for farmed animals and prescribed by the European Union regulation; winter serological screening, sentinel system, passive clinical surveillance, export testing. Scenario tree methodology was used to evaluate the relative sensitivity of detection and targeted approach of each component in terms of early detection and freedom of infection substantiation. Field data collected from the previous year's outbreaks in Belgium were used to determine the risk groups to be considered. RESULTS: The best sensitivities at herd level, taking into account the diagnostic test sensitivity, design prevalence and the number of animals tested within a herd were obtained with the winter screening and sentinel component. The sensitivities at risk group level, taking into account the obtained herd sensitivity, effective probabilities of infection and number of herds tested were high in all components, except for the export component. Component sensitivities ranged between 0.77 and 1 for all components except for the export component with a mean value of 0.22 (0.17-0.26). In terms of early detection, the probability of detection was best using the passive clinical component or the sentinel component. Sensitivity analysis showed that the passive clinical component sensitivity was mostly affected by the diagnostic process and the number of herds sampled. The sentinel and export components sensitivity were mainly affected by the relative risk estimates whereas the winter screening component was mainly affected by the assumptions about the design prevalence. CONCLUSIONS: This study revealed interesting features regarding the sensitivity of detection and early detection of infection in the different surveillance components and their risk based approach as requested by the international standards.

Factors affecting Bluetongue serotype 8 spread in Northern Europe in 2006: the geographical epidemiology.

In 2006, Bluetongue serotype 8 was notified for the first time in north-western Europe, more specifically in Belgium, the Netherlands, Luxemburg, Germany and France. The disease spread very rapidly, affecting mainly cattle and sheep farms. In this paper, we examined risk factors affecting the spatial incidence of reported Bluetongue events during the first outbreak in 2006. Previous studies suggested that the Bluetongue incidence was enhanced by environmental factors, such as temperature and wind speed and direction, as well as by human interventions, such as the transport of animals. In contrast to the previous studies, which were based on univariable analyses, a multivariable epidemiological analysis describing the spatial relationship between Bluetongue incidence and possible risk factors is proposed in this paper. This disentangles the complex interplay between different risk factors. Our model shows that wind is the most important factor affecting the incidence of the disease. In addition, areas with high precipitation are slightly more sensitive to the spread of the infection via the wind. Another important risk factor is the land cover; high-risk areas for infection being characterized by a fragmentation of the land cover, especially the combination of forests and urban areas. Precipitation and temperature are also significant risk factors. High precipitation in areas with a large coverage of forests and/or pasture increases the risk whereas high temperature increases the risk considerably in municipalities covered mainly with pasture. Local spread via the vector is strongest in areas with a large coverage of forests and smallest in highly urbanized areas. Finally, the transport of animals from infected areas is a risk factor.

Improving the accuracy of livestock distribution estimates through spatial interpolation.

Animal distribution maps serve many purposes such as estimating transmission risk of zoonotic pathogens to both animals and humans. The reliability and usability of such maps is highly dependent on the quality of the input data. However, decisions on how to perform livestock surveys are often based on previous work without considering possible consequences. A better understanding of the impact of using different sample designs and processing steps on the accuracy of livestock distribution estimates was acquired through iterative experiments using detailed survey. The importance of sample size, sample design and aggregation is demonstrated and spatial interpolation is presented as a potential way to improve cattle number estimates. As expected, results show that an increasing sample size increased the precision of cattle number estimates but these improvements were mainly seen when the initial sample size was relatively low (e.g. a median relative error decrease of 0.04% per sampled parish for sample sizes below 500 parishes). For higher sample sizes, the added value of further increasing the number of samples declined rapidly (e.g. a median relative error decrease of 0.01% per sampled parish for sample sizes above 500 parishes. When a two-stage stratified sample design was applied to yield more evenly distributed samples, accuracy levels were higher for low sample densities and stabilised at lower sample sizes compared to one-stage stratified sampling. Aggregating the resulting cattle number estimates yielded significantly more accurate results because of averaging under- and over-estimates (e.g. when aggregating cattle number estimates from subcounty to district level, P <0.009 based on a sample of 2,077 parishes using one-stage stratified samples). During aggregation, area-weighted mean values were assigned to higher administrative unit levels. However, when this step is preceded by a spatial interpolation to fill in missing values in non-sampled areas, accuracy is improved remarkably. This counts especially for low sample sizes and spatially even distributed samples (e.g. P <0.001 for a sample of 170 parishes using one-stage stratified sampling and aggregation on district level). Whether the same observations apply on a lower spatial scale should be further investigated.

Expert judgement in a risk assessment model for Salmonella spp. in pork: the performance of different weighting schemes.

A structured expert judgement study was carried out in order to obtain input parameters for a quantitative microbial risk assessment (QMRA) model. This model aimed to estimate the risk of human Salmonella infections associated with the consumption of minced pork meat. Judgements of 11 experts were used to derive subjective probability density functions (PDFs) to quantify the uncertainty on the model input parameters. The performance of experts as probability assessors was measured by the experts' ability to correctly and precisely provide estimates for a set of seed variables (=variables from the experts' area of expertise for which the true values were known to the analyst). Subsequently different weighting schemes or "decision makers" (DMs) were applied using Cooke's classical model in order to obtain combined PDFs as a weighted linear combination of the expert's individual PDFs. The aim of this study was to compare the performance of four DMs namely the equal weight DM (each expert's opinion received equal weight), the user weight DM (weights are determined by the expert's self-perceived level of expertise) and two performance-based DMs: the global weight DM and the item weight DM. Weights in the performance-based DMs were calculated based on the expert's calibration and information performance as measured on the set of seed variables. The item weight DM obtained the highest performance with a calibration score of 0.62 and an information score of 0.52, as compared to the other DMs. The weights of the performance-based DMs outperformed those of the best expert in the panel. The correlation between the scores for self-rating of expertise and the weights based on the experts' performance on the calibration variables was low and not significant (r=0.37, p=0.13). The applied classical model provided a rational basis to use the combined distributions obtained by the item weight DM as input in the QMRA model since this DM yielded generally more informative distributions for the variables of interest than those obtained by the equal weight and user weight DM. Attention should be paid to find adequate and relevant seed variables, since this is important for the validation of the results of the weighting scheme.