Demonstrating freedom from disease using multiple complex data sources 2: case study--classical swine fever in Denmark

Vaccination of poultry against highly pathogenic avian influenza - Part 2. Surveillance and mitigation measures

Selecting appropriate diagnostic methods that take account of the type of vaccine used is important when implementing a vaccination programme against highly pathogenic avian influenza (HPAI). If vaccination is effective, a decreased viral load is expected in the samples used for diagnosis, making molecular methods with high sensitivity the best choice. Although serological methods can be reasonably sensitive, they may produce results that are difficult to interpret. In addition to routine molecular monitoring, it is recommended to conduct viral isolation, genetic sequencing and phenotypic characterisation of any HPAI virus detected in vaccinated flocks to detect escape mutants early. Following emergency vaccination, various surveillance options based on virological testing of dead birds ('bucket sampling') at defined intervals were assessed to be effective for early detection of HPAIV and prove disease freedom in vaccinated populations. For ducks, virological or serological testing of live birds was assessed as an effective strategy. This surveillance could be also applied in the peri-vaccination zone on vaccinated establishments, while maintaining passive surveillance in unvaccinated chicken layers and turkeys, and weekly bucket sampling in unvaccinated ducks. To demonstrate disease freedom with > 99% confidence and to detect HPAI virus sufficiently early following preventive vaccination, monthly virological testing of all dead birds up to 15 per flock, coupled with passive surveillance in both vaccinated and unvaccinated flocks, is recommended. Reducing the sampling intervals increases the sensitivity of early detection up to 100%. To enable the safe movement of vaccinated poultry during emergency vaccination, laboratory examinations in the 72 h prior to the movement can be considered as a risk mitigation measure, in addition to clinical inspection; sampling results from existing surveillance activities carried out in these 72 h could be used. In this Opinion, several schemes are recommended to enable the safe movement of vaccinated poultry following preventive vaccination.

Utility of surveillance data for planning for dengue elimination in Yogyakarta, Indonesia: a scenario-tree modelling approach

INTRODUCTION

Field trials and modelling studies suggest that elimination of dengue transmission may be possible through widespread release of Aedes aegypti mosquitoes infected with the insect bacterium Wolbachia pipientis (wMel strain), in conjunction with routine dengue control activities. This study aimed to develop a modelling framework to guide planning for the potential elimination of locally acquired dengue in Yogyakarta, a city of almost 400 000 people in Java, Indonesia.

METHODS

A scenario-tree modelling approach was used to estimate the sensitivity of the dengue surveillance system (including routine hospital-based reporting and primary-care-based enhanced surveillance), and time required to demonstrate elimination of locally acquired dengue in Yogyakarta city, assuming the detected incidence of dengue decreases to zero in the future. Age and gender were included as risk factors for dengue, and detection nodes included the probability of seeking care, probability of sample collection and testing, diagnostic test sensitivity and probability of case notification. Parameter distributions were derived from health system data or estimated by expert opinion. Alternative simulations were defined based on changes to key parameter values, separately and in combination.

RESULTS

For the default simulation, median surveillance system sensitivity was 0.131 (95% PI 0.111 to 0.152) per month. Median confidence in dengue elimination reached 80% after a minimum of 13 months of zero detected dengue cases and 90% confidence after 25 months, across different scenarios. The alternative simulations investigated produced relatively small changes in median system sensitivity and time to elimination.

CONCLUSION

This study suggests that with a combination of hospital-based surveillance and enhanced clinic-based surveillance for dengue, an acceptable level of confidence (80% probability) in the elimination of locally acquired dengue can be reached within 2 years. Increasing the surveillance system sensitivity could shorten the time to first ascertainment of elimination of dengue and increase the level of confidence in elimination.

Scenario tree modelling to inform surveillance design for maintaining freedom from Coxiella burnetii infection in Australian commercial dairy goat herds

We used scenario tree methods to determine how different disease detection methods might be used to provide quantitative evidence that Australian dairy goat herds are free of coxiellosis. The aim of our proposed C. burnetii surveillance programme is to find evidence of the absence of antigen as well as evidence of an absence of an immune response to C. burnetii infection in individual dairy goat herds. We defined a C. burnetii infected dairy goat herd as a herd in which at least one doe was showing evidence of either active infection or past C. burnetii exposure using four candidate surveillance system components (SSCs): (1) testing of individual doe whole blood using the C. burnetii com1 PCR; (2) testing of individual doe whole blood using the C. burnetii ELISA; (3) testing bulk tank milk (BTM) using the com1 PCR and the C. burnetii ELISA; and (4) investigations of abortions and stillborn kids submitted to a diagnostic laboratory for testing. Of eight candidate surveillance strategies (combinations of the SSCs listed above) individual doe ELISAs every six months combined with monthly BTM PCR and ELISA testing returned the highest surveillance system sensitivity of 0.963 (95% probability interval [PI] 0.911-0.982) for the lowest cost, at AUD 28.94 (95% PI 28.38-30.59) over a 12-month period, for every one percent increase in surveillance system sensitivity. Assuming a probability of disease freedom of 0.10 at the start of the surveillance program and a probability of C. burnetii introduction per month of 0.01 we estimate that 95% confidence that C. burnetii was absent from a herd could be achieved after a single round of individual doe ELISAs followed by period of 6 consecutive monthly BTM PCR and ELISA tests. The results of this study show that selection of the most efficient combination of surveillance system components requires a good understanding of initial herd C. burnetii status and the probability of introduction of infection and how this may change over time. Scenario tree analyses results have provided insight into the key determinants of C. burnetii detection ability.

Use of scenario tree modelling to plan freedom from infection surveillance: Mycoplasma bovis in New Zealand

Since mid-2018, the New Zealand (NZ) Ministry for Primary Industries (MPI) has been operating an eradication program for an incursion of Mycoplasma bovis. Although NZ is still delimiting the outbreak, consideration is being given to how freedom from M. bovis will be demonstrated. Rapid demonstration of freedom will minimise the length of the program, significantly reducing its financial burden. This collaborative research was undertaken to help inform planning of surveillance to demonstrate freedom after M. bovis is believed eradicated. Scenario tree modelling (STM) involves assimilating multiple surveillance system components to determine whether disease is absent. STM has infrequently been used to plan appropriate surveillance but this was the approach used here. A stochastic simulation model was implemented in R. The model represented the NZ commercial dairy and non-dairy cattle industries and the current surveillance components that are also planned to be used to gather evidence of absence of M. bovis once it is eradicated. Different surveillance intensities and risk based versus random surveillance were simulated and compared for probability of freedom, financial cost of sampling and testing and the time to demonstrate freedom. The results indicate that the current surveillance components will enable demonstration of freedom. Surveillance components included bulk tank milk testing, herd testing and testing at meat processing plants, predominantly using an imperfect ELISA. Several combinations of surveillance components appeared most efficient achieving >95 % confidence of freedom over 2-4 years, whilst sampling 4-7 % of the non-dairy herds and less than 25 % of dairy herds annually. The results indicate that surveillance intensity can be lower than is currently occurring to support the delimiting phase, thereby saving significant resources in the post eradication phase (proof of freedom phases). Further consideration is required to enable the assumption of 100 % herd specificity made in the model to be achieved. The ELISA used is very specific, but will yield some false positives that must be resolved to their true status. This may occur for example through modified diagnostic test interpretation (e.g. cut point optimisation at individual and herd level) or resolution of putative false positive herds with epidemiological investigation. In conclusion this research demonstrates the utility of STM for planning surveillance programs, and in this instance has highlighted efficient and effective surveillance components for demonstrating freedom from M. bovis in NZ. It also highlights the need to achieve 100 % specificity for M. bovis in herds tested during the proof of freedom phases.

Existence and Quality of Data on Control Programs for EU Non-regulated Cattle Diseases: Consequences for Estimation and Comparison of the Probability of Freedom From Infection

Some European countries have successfully implemented country-specific control programs (CPs) for infectious cattle diseases that are not regulated or are regulated only to a limited extent at the European Union (EU) level. Examples of such diseases include bovine viral diarrhea (BVD), infectious bovine rhinotracheitis (IBR), and Johne's disease (JD). The CPs vary between countries in the design and quality of collected data as well as methods used to detect infection and estimate prevalence or probability of freedom from infection. Differences in disease status between countries and non-standardized approaches to assess freedom from infection pose a risk for countries with CPs for non-regulated diseases as infected animals may influence the progress of the disease control or eradication program. The implementation of output-based standards allows estimation and comparison of the probability of freedom for non-regulated cattle diseases in European countries. The aim of the current study was to assess the existence and quality of data that could be used for estimating freedom from infection in European countries. The online data collection tool was sent to 32 countries participating in the SOUND control COST Action and was completed by 24 countries. Data on cattle demographics and data from CPs of IBR and BVD exist in more than 50% of the response countries. However, data describing risk factors and CP of JD was reported as existing in <25% of the countries. The overall quality of data in the sections on demographics and CPs of IBR and BVD were evaluated as "good", but risk factors and JD data were mostly evaluated as "fair." Data quality was considered less good mainly due to two quality criteria: accessibility and accuracy. The results of this study show that the quantity and quality of data about cattle populations and CPs are relatively similar in many surveyed countries. The outcome of this work provides an overview of the current situation in the European countries regarding data on EU non-regulated cattle diseases and will further assist in the development and implementation of output-based standards.

Evaluation of the sensitivity of the classical swine fever surveillance system in two free zones in Colombia

Infection with the classical swine fever virus (CSFV) causes a disease in pigs that ranges from a hyperacute form in which animals die in a few hours to subclinical disease. Due to this wide range of virulence, several complementary surveillance strategies should be implemented for the early detection of the disease. The objective of the present study was to determine the sensitivity of the surveillance system to detect CSFV outbreaks in a free zone (Zone 1) and in a zone undergoing an eradication process (Zone 2) in Colombia. Stochastic scenario tree models were used to describe the population and surveillance structures and to determine the probability of CSFV detection. The total sensitivity of the surveillance system in the case of a single infected farm in Zone 1 was 31.4% (CI 95%: 7.2-54.1) and in the case of 5 infected farms was 85.2% (CI 95%: 67.3-93.7), while in Zone 2 the sensitivities were 27.8% (CI 95%: 6.4-55.1) and 82.5% (CI 95%: 65-92.9), respectively. The on-farm passive surveillance shows the highest sensitivity for detection of a single CSFV infected farm in both zones (22.8% in Zone 1 and 22.5% in Zone 2). The probability of detection was higher in a family / backyard premise than on a commercial farm in both zones. The passive surveillance at slaughterhouse had a sensitivity of 5.3% and 4.5% for the detection of a single infected farm in Zone 1 and 2, respectively. Active surveillance presented a range of sensitivity between 2.2% and 4.5%. In conclusion, the sensitivity of the surveillance in the two studied zones was quite high, one of reasons for this good sensitivity being the sentinel network based on the voluntary participation of 5,500 collaborators that were trained for the identification and notification of diseases of national interest.

Sample Size Estimation in Veterinary Epidemiologic Research

In the design of intervention and observational epidemiological studies sample size calculations are used to provide estimates of the minimum number of observations that need to be made to ensure that the stated objectives of a study are met. Justification of the number of subjects enrolled into a study and details of the assumptions and methodologies used to derive sample size estimates are now a mandatory component of grant application processes by funding agencies. Studies with insufficient numbers of study subjects run the risk of failing to identify differences among treatment or exposure groups when differences do, in fact, exist. Selection of a number of study subjects greater than that actually required results in a wastage of time and resources. In contrast to human epidemiological research, individual study subjects in a veterinary setting are almost always aggregated into hierarchical groups and, for this reason, sample size estimates calculated using formulae that assume data independence are not appropriate. This paper provides an overview of the reasons researchers might need to calculate an appropriate sample size in veterinary epidemiology and a summary of sample size calculation methods. Two approaches are presented for dealing with lack of data independence when calculating sample sizes: (1) inflation of crude sample size estimates using a design effect; and (2) simulation-based methods. The advantage of simulation methods is that appropriate sample sizes can be estimated for complex study designs for which formula-based methods are not available. A description of the methodological approach for simulation is described and a worked example provided.

Why lockdown? Why national unity? Why global solidarity? Simplified arithmetic tools for decision-makers, health professionals, journalists and the general public to explore containment options for the 2019 novel coronavirus

As every country in the world struggles with the ongoing COVID-19 pandemic, it is essential that as many people as possible understand the epidemic containment, elimination and exclusion strategies required to tackle it. Simplified arithmetic models of COVID-19 transmission, control and elimination are presented in user-friendly Shiny and Excel formats that allow non-specialists to explore, query, critique and understand the containment decisions facing their country and the world at large. Although the predictive model is broadly applicable, the simulations presented are based on parameter values representative of the United Republic of Tanzania, which is still early enough in its epidemic cycle and response to avert a national catastrophe. The predictions of these models illustrate (1) why ambitious lock-down interventions to crush the curve represent the only realistic way for individual countries to contain their national-level epidemics before they turn into outright catastrophes, (2) why these need to be implemented so early, so stringently and for such extended periods, (3) why high prevalence of other pathogens causing similar symptoms to mild COVID-19 precludes the use of contact tracing as a substitute for lock down interventions to contain and eliminate epidemics, (4) why partial containment strategies intended to merely flatten the curve, by maintaining epidemics at manageably low levels, are grossly unrealistic, and (5) why local elimination may only be sustained after lock down ends if imported cases are comprehensively excluded, so international co-operation to conditionally re-open trade and travel between countries certified as free of COVID-19 represents the best strategy for motivating progress towards pandemic eradication at global level. The three sequential goals that every country needs to emphatically embrace are contain, eliminate and exclude. As recently emphasized by the World Health Organization, success will require widespread genuine national unity and unprecedented global solidarity.

Pushing past the tipping points in containment trajectories of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) epidemics: A simple arithmetic rationale for crushing the curve instead of merely flattening it

Countries with ambitious national strategies to crush the curve of their Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) epidemic trajectories include China, Korea, Japan, Taiwan, New Zealand and Australia. However, the United States and many hard-hit European countries, like Ireland, Italy, Spain, France and the United Kingdom, currently appear content to merely flatten the curve of their epidemic trajectories so that transmission persists at rates their critical care services can cope with. Here I present a simple set of arithmetic modelling analyses that are accessible to non-specialists and explain why preferable crush the curve strategies, to eliminate transmission within months, would require only a modest amount of additional containment effort relative to the tipping point targeted by flatten the curve strategies, which allow epidemics to persist at supposedly steady, manageable levels for years, decades or even indefinitely.

Rigorous surveillance is necessary for high confidence in end-of-outbreak declarations for Ebola and other infectious diseases

The World Health Organization considers an Ebola outbreak to have ended once 42 days have passed since the last possible exposure to a confirmed case. Benefits of a quick end-of-outbreak declaration, such as reductions in trade/travel restrictions, must be balanced against the chance of flare-ups from undetected residual cases. We show how epidemiological modelling can be used to estimate the surveillance level required for decision-makers to be confident that an outbreak is over. Results from a simple model characterizing an Ebola outbreak suggest that a surveillance sensitivity (i.e. case reporting percentage) of 79% is necessary for 95% confidence that an outbreak is over after 42 days without symptomatic cases. With weaker surveillance, unrecognized transmission may still occur: if the surveillance sensitivity is only 40%, then 62 days must be waited for 95% certainty. By quantifying the certainty in end-of-outbreak declarations, public health decision-makers can plan and communicate more effectively. This article is part of the theme issue 'Modelling infectious disease outbreaks in humans, animals and plants: epidemic forecasting and control'. This issue is linked with the earlier theme issue 'Modelling infectious disease outbreaks in humans, animals and plants: approaches and important themes'.

Base protocol for real time active random surveillance of coronavirus disease (COVID-19) - Adapting veterinary methodology to public health

The pandemic of new coronavirus disease COVID-19 is threatening our health, economy and life style. Collaborations across countries and sectors as a One Health World could be a milestone.

We propose a general protocol, for setting timely active random surveillance of COVID-19, at the human community level, with systematic repeated detection efforts. Strengths and limitations are discussed.

If considered applicable by public health, the protocol could evaluate the status of COVID-19 epidemics consistently and objectively.

•

A base protocol is proposed for setting active random surveillance of COVID-19.

•

The protocol is based on widely known veterinary surveillance methodologies.

•

By applying the protocol, COVID-19 epidemics could be consistently assessed.

•

The protocol could be a milestone in the battle against the pandemic.

•

Authorities could consider if the protocol is applicable for public health.

Evaluation of the English bovine brucellosis surveillance system considering probability of disease introduction and non-random sampling

The English surveillance system for bovine brucellosis was evaluated. The confidence in detecting at least one infected herd in the local population (surveillance system sensitivity or SSe), and the confidence in freedom from disease (PFree) adjusted (PFreeAdj) for the probability of disease introduction from abroad by imported animals (PIntro), were estimated for quarterly surveillance periods of 2016; because dairy herds were tested quarterly on bulk tank milk (BTM) with an antibody indirect ELISA. A stochastic model was developed and six surveillance components (representing also the local population strata), were evaluated. All English herds and their relative risk (RRs) of infection within each stratum were considered. The importance of each component was assessed using actual national data, which reflected non-random sampling. The contribution of the abortions testing was assessed with particular focus, because a decline in statutory submissions was observed in recent years. Beef herds without submissions (B-NoTest herds) at the laboratories were still considered as a population stratum, where infected cattle could be imported. Additionally, we evaluated the importance of different hypothetical design between-herds prevalence (Ph) values, at which the country could be classified as "infected". The potential negative effect on SSe due to the dilution of antibodies when individual samples are pooled within BTM and tested by the milk iELISA, was also investigated. The quarterly median SSe and PFreeAdj were both ≥ 95 % if at least four (0.008 %) herds were infected in the country due to independent import events. The system appeared able to substantiate Official Brucellosis Free (OBF) status frequently (on quarterly basis) using Ph=0.2 % (EU legislation). The component based only on BTM testing (M herds) showed the highest sensitivity; while the surveillance components based on abortions or post import calving (PIC) testing, had very low sensitivity at the (considered) Ph values lower than 0.2 %. In contrast, at Ph = 0.2 %, components based on abortion testing had median sensitivity between 91.3 % and 99.9 %, and the dilution effect on BTM testing did not change remarkably the SSe and PFreeAdj. When Ph was set to 1-2 infected herds (0.002-0.004 %), these were usually allocated by the model within the B-NoTest stratum (the largest stratum) and SSe reduced. Thus, if policy considers necessary increasing the SSe for low Phs (system's optimization as an early warning system); the cost efficiency of active risk based surveillance in beef herds (considering imports) could be investigated in the future.

Strategies for the Global Eradication of Peste des Petits Ruminants: An Argument for the Use of Guerrilla Rather Than Trench Warfare

Many historical disease eradication campaigns have been characterized by large-scale mobilization and long-term campaigns of mass vaccination. As the duration of a program increases, the total cost also increases, but the effectiveness and sustainability decrease, sometimes resulting in premature loss of stakeholder support, field team fatigue, and failure or major set-backs. In contrast to this trench warfare approach, this paper proposes an eradication strategy modeled on guerrilla tactics: use exceptionally good, locally relevant and timely intelligence; strike rapidly and effectively in small areas; achieve your goals; and keep moving. For peste des petits ruminants eradication, this means a shift away from long-term mass vaccination, focusing instead on addressing some of the challenges that have plagued previous eradication programs: ineffective surveillance and movement management. Recent developments in surveillance have shown that it is now feasible to capture information about almost all cases of disease, all movements and all control activities, from the entire population in real time. Developing powerful, effective and sustainable surveillance systems is an essential prerequisite for rapid, affordable PPR eradication. PPR can be rapidly eliminated from small populations by achieving very high levels of vaccination coverage for only a short period. The key challenge is then to prevent the re-introduction of disease as immunity wanes, and to respond rapidly and effectively in the case of further local outbreaks. A comprehensive understanding of movement patterns and their drivers will allow rapid progressive eradication to be implemented. The population can be divided into manageably small units, targeted sequentially for high-coverage short-duration vaccination, then moving to the next unit based on the distribution of disease and the direction of animal flow. This approach optimizes the use of available resources, and minimizes the challenge and disruption of managing retrograde movement from infected to uninfected areas. High levels of community engagement are required to achieve the quality of surveillance, movement management and rapid response necessary for success. Traditionally, long-term vaccination has been used to first eliminate the virus from a population, and then to protect it against re-introduction of the disease. Under the guerrilla strategy, continuous real-time information, not long-term vaccination, is the main tool for disease eradication.

Classical swine fever in Victorian domestic pigs: evidence of disease freedom

OBJECTIVE

Australia is currently regarded as free of classical swine fever (CSF), a highly contagious disease of pigs caused by a pestivirus. This study aimed to provide additional evidence that the Victorian domestic pig population is free of CSF.

DESIGN

A structured representative sero-prevalence survey of Victorian domestic pigs at slaughter.

METHOD

Three-hundred and ninety-one pigs from 23 holdings were sampled at the time of slaughter between March 2016 and October 2017.

RESULTS

All samples were negative for CSF virus Ab on ELISA. Because of uncertainty in the sensitivity of the CSF Ab ELISA, estimates of the true prevalence of CSF were calculated using Bayesian methods. The median and upper bound of the 95% credible intervals for the true prevalence of CSF was zero when the diagnostic sensitivity of the CSF Ab ELISA was assumed to range from 0.75 to 0.95.

CONCLUSION

These results provide evidence that the population of domestic pigs in Victoria in 2016-2017 was free of CSF.

Establishing post-outbreak freedom from African horse sickness virus in South Africa's surveillance zone

An African horse sickness (AHS) outbreak occurred in South Africa's AHS controlled area in autumn 2016. A freedom from disease survey was performed to establish the likelihood of ongoing circulation of the associated virus during the same period the following year. A single-stage surveillance strategy was employed with a population-level design prevalence of 1% to establish a survey population sensitivity of 95% (probability that one or more positive horses would be detected if AHS was present at a prevalence greater than or equal to the design prevalence). In March 2017, a total of 262 randomly selected horses from 51 herds were sampled from the 2016 outbreak containment zone. Three within-herd and herd-level design prevalence scenarios were used in evaluating the post-survey probability of freedom. Depending on the underlying design prevalence scenarios, effectively ranging between 0.8% and 6.4%, and the use of informed or uninformed priors, the probability of freedom derived from this surveillance ranged between 73.1% and 99.9% (uninformed prior) and between 96.6% and 100% (informed prior). Based on the results, the authors conclude that it is unlikely that the 2016 AHS virus was still circulating in the autumn of 2017 in the 2016 outbreak containment zone. The ability to perform freedom from disease surveys, and also to include risk-based methods, in the AHS controlled area of South Africa is influenced by the changing underlying population at risk and the high level of vaccination coverage in the horse population. Ongoing census post-outbreak must be undertaken to maintain a valid sampling frame for future surveillance activity. The seasonality of AHS, the restricted AHS vaccination period and the inability to easily differentiate infected from vaccinated animals by laboratory testing impact the ability to perform a freedom from disease survey for AHS in the 12 months following an outbreak in the controlled area.

Cost-effectiveness assessment of three components of the bovine tuberculosis surveillance system by intradermal tuberculin testing in French cattle farms by a scenario tree approach

In most officially bovine tuberculosis (bTB)-free countries, bTB has not been fully eradicated. Costly and time-consuming surveillance and control measures are therefore still in place to control this infection. An officially bTB-free status, both at the national and at the herd level, influences whether and when animals can be sold. Thus, this infection is still an economic issue, justifying measures towards its eradication. An evaluation of the cost-effectiveness of such measures would be highly useful, especially to optimise the costs of control measures and their adaptation to a local epidemiological context. We evaluated the cost-effectiveness of three mandatory surveillance protocols currently used in France by herd type (type of production, size, and turnover of the herd) under French field conditions. The first protocol ("strict") implies the direct slaughter and post-mortem analyses of any intradermal cervical tuberculin test (ICT) reactor, and negative results to a second intradermal cervical comparative tuberculin test (ICCT) to regain bTB-free status of the herd. In the second protocol ("compliant quick-path") bTB-free status can be regain if post-mortem analyses of reactors to the first ICT are negative. In the third protocol ("compliant slow-path"), ICCT-reactive animals are tested using the interferon gamma assay; the results of this test influence the path of further investigation. We built scenario trees for each of these protocols at the animal level. They allowed us to estimate herd sensitivity and the total cost of each protocol by herd type. The protocols could be ordered by decreasing herd sensitivity and cost, regardless of the herd type, as follows: strict protocol, compliant quick-path protocol, and compliant slow-path protocol. We calculated a cost-effectiveness index to evaluate the cost-effectiveness of each protocol. The strict protocol was never the most cost-effective, regardless of herd type, due to higher costs relative to the other protocols, despite better herd sensitivity. We found the compliant quick-path to be the most cost-effective protocol for big beef, big dairy, and mixed herds. The compliant slow-path was the most cost-effective for small-scale beef and dairy herds. All differences were significant. This comparison of the cost-effectiveness of the protocols by herd type could help authorities to choose the most suitable protocol in the investigation of suspected cases, depending on the herd type, but could be improved by accounting for important sociological data, such as the acceptability of the protocols.

Sampling Methodology to Maximize the Efficient Use of National Abattoir Surveillance: Using Archived Sera to Substantiate Freedom From Bluetongue Virus Infection in Ireland

In recent years, there has been increasing recognition of the value of multiple data sources available to fulfill surveillance objectives, and the use of these has been applied to address many questions relating to animal health surveillance. In Ireland, we face a slightly different problem, namely, best use of an existing surveillance resource (serological samples collected over many years from cull cows at slaughter), which has been used to substantiate freedom from Brucella abortus following its successful eradication in 2009. In this study, we evaluate a sampling methodology to use this resource to substantiate freedom from bluetongue virus (BTV) infection. An examination of the degree to which cull cows were resident in the same herd throughout the midge biting season showed that, of 50,640 samples collected between 17 October and 23 December 2016, 80.2% were from animals resident in the same herd between 01 April 2016 and 2 months prior to their slaughter date, 74.1% for 1 month prior, 70.1% for 2 weeks prior, 66.4% for 1 week prior, and 56.4% up to 1 day prior to slaughter. An examination was made of the degree to which individual samples within the same 88-well frozen storage block came from geographically clustered herds, whether from a concentration of animals from the same herd in a single block, or from clustering around the slaughterhouse where the samples were taken. On the basis of these analyses, a sampling strategy was derived aimed at minimizing the number of storage blocks which needed to be thawed, whilst ensuring a large enough and representative sample, geographically stratified according to the bovine population of 51 squares, each 45 × 45 km, covering the entirety of Ireland. None of the 503 samples tested were positive for BTV, providing reassurance of national BTV freedom. More broadly, the study demonstrates the use of abattoir-based serological samples collected for one large scale surveillance programme in surveillance for other bovine infections.

Evaluation of herd-level sampling strategies for control of Salmonella in Swedish cattle

Based on Swedish legislation, all herds where Salmonella of any serotype is detected are put under restrictions, and measures aiming at eradication are required. Costs for sampling and control have increased in recent years and the aim of this study was to investigate the efficiency of different sampling strategies. We also compiled test results from recent surveillance activities and used these to complement and compare with calculated results. Sensitivities and specificities at group and herd level were calculated for different test strategies. A scenario-tree modeling approach was used to account for the hierarchy of animals within herds, and different relative risk of salmonella in different age groups. Negative and positive predictive values (NPV and PPV), and probability of freedom from Salmonella were calculated to compare the added value of different sampling strategies. Results showed that more fecal samples than serological samples per group are needed to reach a group sensitivity >0.50. This also means that serological testing leads to a higher NPV. For example, with 10 negative test-results from a group of 25 animals in a herd with a suspicion of Salmonella, the NPV based on serology was 0.75 and based on culture was 0.56. For the PPV, testing based on culture from fecal sampling was superior, as specificity of such testing was close to perfect. By changing the threshold for considering a group positive, from 1 test-positive animal to 2, the PPV of serological results could be increased without substantial loss in NPV. The herd sensitivity based on (1) bulk milk sampling, (2) fecal sampling of all animals, and (3) bulk milk sampling and individual sera from 20 animals within each age group was 0.53, 0.88, and 0.95, respectively. In low-prevalence regions, this sensitivity was enough to verify a high probability of freedom (>0.99), as the probability of infection in such Swedish regions has been shown to be 0.01. For herds with a higher prior probability of infection, repeated sampling (2-9 sampling occasions) was needed to reach the same level of confidence. Analysis of surveillance data indicated that boot swabs can be used to replace the standard fecal sampling presently used in Sweden. It was also confirmed that the individual specificity of the tests used for serological testing of Swedish calves is high (0.99). The results can form a basis for fit-for-purpose testing strategies (e.g., surveillance or prepurchase testing).

Herd level estimation of probability of disease freedom applied on the Norwegian control program for bovine respiratory syncytial virus and bovine coronavirus

A national control program against bovine respiratory syncytial virus (BRSV) and bovine coronavirus (BCV) was launched in Norway in 2016. A key strategy in the program is to test for presence of antibodies and protect test-negative herds from infection. Because these viruses are endemic, the rate of re-introduction can be high, and a disease-free status will become more uncertain as time from testing elapses. The aim of this study was to estimate the probability of freedom (PostPFree) from BRSV and BCV antibodies over time by use of bulk tank milk (BTM) antibody-testing, geographic information and animal movement data, and to validate the herd-level estimates against subsequent BTM testing.

BTM samples were collected from 1148 study herds in West Norway in 2013 and 2016, and these were analyzed for BRSV and BCV antibodies. PostPFree was calculated for herds that were negative in 2013/2014, and updated periodically with new probabilities every three months. Input variables were test sensitivity, the probability of introduction through animal purchase and local transmission. Probability of introduction through animal purchase was calculated by using real animal movement data and herd prevalence in the region of the source herd. The PostPFree from the final three months in 2015 was compared to BTM test results from March 2016 using a Wilcoxon Rank Sum Test.

The probability of freedom was generally high for test-negative herds immediately after testing, reflecting the high sensitivity of the tests. It did however, decrease with time since testing, and was greatly affected by purchase of livestock. When comparing the median PostPFree for the final three months to the test results in 2016, it was significantly lower (p < 0.01) for test positive herds. Furthermore, there was a large difference in the proportion of test positive herds between the first and fourth quartile of PostPFree. The results show that PostPFree provides a better estimate of herd-level BTM status for both BRSV and BCV than what can be achieved by relying solely on the previous test-result.

Assessing Biosecurity Risks for the Introduction and Spread of Diseases Among Commercial Sheep Properties in New South Wales, Australia, Using Foot-and-Mouth Disease as a Case Study

Sheep production systems are a major industry in Australia, with a gross value of roughly $4.66 billion; 87.3% of which is attributable to export markets. Exotic diseases such as foot-and-mouth disease (FMD) are a potential threat to the viability of Australia’s export market. Previous outbreaks of FMD in developed countries, and challenges in the management of onshore biosecurity, signify the importance of on-farm biosecurity in controlling disease transmission. This study aims to investigate the risk of disease introduction and spread among New South Wales (NSW) sheep properties using FMD as a case study and draw recommendation for the industry. Exposure and partial consequence assessments, using scenario trees and Monte Carlo stochastic modeling, were conducted to identify pathways of introduction and spread and calculate the probabilities of these pathways occurring. Input parameters were estimated from the data obtained during qualitative interviews with producers and scientific literature. According to the reported practices of sheep producers and assuming each pathway was carrying the FMD virus, the exposure assessment estimates the median (5–95%) probability of FMD exposure of sheep on a naive property to be 0.619 (0.541–0.698), 0.151 (0.085–0.239), 0.235 (0.153–0.324), and 0.710 (0.619–0.791) for introduction through new stock, wildlife, carriers (humans, dogs, and vehicles), and neighbors, respectively. The spread assessment estimated the median probability of FMD spreading from an infected sheep property to neighboring enterprises to be 0.603 (0.504–0.698). A similar probability was estimated for spread via wildlife (0.523; 0.404–0.638); and a lower spread probability was estimated for carriers (0.315; 0.171–0.527), sheep movement (0.285; 0.161–0.462), and dead stock (0.168; 0.070–0.312). The sensitivity analysis revealed that the introduction of an FMD-infected sheep was more influential for exposure via new stock than isolation practices. Sharing adjacent boundaries was found to be the most influential factor for exposure and spread between neighboring enterprises, and to a lesser extent, hygiene practices were found to have the most influence on exposure and spread through carriers. To minimize the potential risk of FMD introduction and spread between sheep properties, maintenance of boundary fences, identification of infected animals before introduction to the property, and hygiene and disinfection practices should be improved.

Low- and High-Pathogenic Avian Influenza H5 and H7 Spread Risk Assessment Within and Between Australian Commercial Chicken Farms

This study quantified and compared the probability of avian influenza (AI) spread within and between Australian commercial chicken farms via specified spread pathways using scenario tree mathematical modeling. Input values for the models were sourced from scientific literature, expert opinion, and a farm survey conducted during 2015 and 2016 on Australian commercial chicken farms located in New South Wales (NSW) and Queensland. Outputs from the models indicate that the probability of no establishment of infection in a shed is the most likely end-point after exposure and infection of low-pathogenic avian influenza (LPAI) in one chicken for all farm types (non-free range meat chicken, free range meat chicken, cage layer, barn layer, and free range layer farms). If LPAI infection is established in a shed, LPAI is more likely to spread to other sheds and beyond the index farm due to a relatively low probability of detection and reporting during LPAI infection compared to high-pathogenic avian influenza (HPAI) infection. Among farm types, the median probability for HPAI spread between sheds and between farms is higher for layer farms (0.0019, 0.0016, and 0.0031 for cage, barn, and free range layer, respectively) than meat chicken farms (0.00025 and 0.00043 for barn and free range meat chicken, respectively) due to a higher probability of mutation in layer birds, which relates to their longer production cycle. The pathway of LPAI spread between sheds with the highest average median probability was spread via equipment (0.015; 5-95%, 0.0058-0.036) and for HPAI spread between farms, the pathway with the highest average median probability was spread via egg trays (3.70 × 10^-5; 5-95%, 1.47 × 10^-6-0.00034). As the spread model did not explicitly consider volume and frequency of the spread pathways, these results provide a comparison of spread probabilities per pathway. These findings highlight the importance of performing biosecurity practices to limit spread of the AI virus. The models can be updated as new information on the mechanisms of the AI virus and on the volume and frequency of movements shed-to-shed and of movements between commercial chicken farms becomes available.

The effect of genetic susceptibility and targeting of sampling on the sensitivity of the surveillance system and certainty-of-freedom for classical scrapie in Finland in 2008-2014

We applied scenario tree modeling to study how the genetic distribution of the sheep population in Finland and the focusing on fallen stock would influence the surveillance sensitivity of scrapie. To incorporate the unevenly distributed susceptibility into the estimation we used data from GB where the genetic distribution and scrapie occurrence have been documented in both normally slaughtered and deceased animals. Finland's sheep population is more susceptible to scrapie than the sheep population in GB and surveillance is concentrated on fallen stock. As a result, there is high systemic sensitivity in Finland even with the moderate number of studied animals. The certainty of the freedom-of-disease status is clearly elevated by the low probability of previous disease occurrence and low probability of introduction. The results highlight the need to change the concept from surveillance system sensitivity to freedom-of-disease status and to also consider the risk of introduction and the cumulative nature of the disease prevalence information due repeated surveillance efforts.

Ocorrência de brucelose e tuberculose bovinas no Rio Grande do Sul com base em dados secundários

RESUMO: Avaliaram-se os resultados de testes para diagnóstico da brucelose e tuberculose bovinas objetivando analisar a distribuição de positividade e características de performance do programa de controle. A análise utilizou como dados secundários todos os resultados de testes para diagnóstico realizados em bovinos no ano de 2008 no estado do Rio Grande do Sul (RS) por médicos veterinários habilitados para atuação no âmbito do Programa Nacional de Controle e Erradicação de Brucelose e Tuberculose Animal. Os dados foram estatisticamente ajustados para minimizar os efeitos de valores extremos e de dados faltantes. Os testes alcançaram 66,80% dos 497 municípios do RS e incluíram a participação de 165 médicos veterinários habilitados. 40,21% dos testes foram realizados nos meses de abril, setembro e outubro, sendo que em 73,90% dos estabelecimentos foram realizados testes para ambas as enfermidades. No caso da brucelose verificou-se que foram testados 35.289 animais, com uma frequência de positivos de 0,25%, enquanto a frequência de estabelecimentos positivos entre os 3.406 testados foi 0,94%. Fêmeas apresentaram frequência de positividade mais elevada (0,29%, de 26.724 testadas) do que machos (0,13%, de 5.235 testados). Animais entre 48-60 meses de idade apresentaram chance de positividade superior às demais faixas etárias (Razão de Chances (RC) =2,63; IC 95%=1,63-4,26). Animais de aptidão leiteira representaram 62,66% dos animais testados, e apresentaram maior chance de positividade do que animais de corte (RC=2,32; IC 95%=1,38-3,90). Adicionalmente, 73,74% dos estabelecimentos testados eram dedicados à produção de leite. Já no caso da tuberculose foram testados 62.149 animais distribuídos em 5.151 propriedades, com frequência de positivos de 0,87% e 3,13%, respectivamente. A chance de positividade aumentou com a idade, sendo que animais com idade acima de 48 meses de idades apresentaram valores superiores aqueles verificados em animais mais jovens (RC=2,07; IC 95%=1,73-2,48). 59,74% dos animais testados eram de aptidão leiteira, os quais apresentaram mais chance de serem positivos do que aqueles de corte (RC=5,03; IC 95%=4,09-6,94). Propriedades leiteiras representaram 78,50% da totalidade dos testes para tuberculose. A análise comparativa do presente trabalho com estudos precedentes de prevalência da brucelose sugere que as ações de controle em curso têm sido efetivas na redução da prevalência da enfermidade no RS. Por outro lado, as informações obtidas no âmbito da tuberculose podem ser indicadoras da condição desta enfermidade, especialmente pela inexistência de estudos de prevalência realizados há menos de 30 anos. Adicionalmente, conclui-se que a utilização de dados secundários, desde que devidamente ajustados, pode ser uma ferramenta eficaz na gestão de programas de saúde animal e em sistemas de monitoramento e vigilância.

Animal Health Surveillance in Scotland in 2030: Using Scenario Planning to Develop Strategies in the Context of "Brexit"

Animal health surveillance is necessary to protect human and animal health, rural economies, and the environment from the consequences of large-scale disease outbreaks. In Scotland, since the Kinnaird review in 2011, efforts have been made to engage with stakeholders to ensure that the strategic goals of surveillance are better aligned with the needs of the end-users and other beneficiaries. The aims of this study were to engage with Scottish surveillance stakeholders and multidisciplinary experts to inform the future long-term strategy for animal health surveillance in Scotland. In this paper, we describe the use of scenario planning as an effective tool for the creation and exploration of five plausible long-term futures; we describe prioritization of critical drivers of change (i.e., international trade policy, data-sharing philosophies, and public versus private resourcing of surveillance capacity) that will unpredictably influence the future implementation of animal health surveillance activities. We present 10 participant-developed strategies to support 3 long-term visions to improve future resilience of animal health surveillance and contingency planning for animal and zoonotic disease outbreaks in Scotland. In the absence of any certainty about the nature of post-Brexit trade agreements for agriculture, participants considered the best investments for long-term resilience to include data collection strategies to improve animal health benchmarking, user-benefit strategies to improve digital literacy in farming communities, and investment strategies to increase veterinary and scientific research capacity in rural areas. This is the first scenario planning study to explore stakeholder beliefs and perceptions about important environmental, technological, societal, political, and legal drivers (in addition to epidemiological "risk factors") and effective strategies to manage future uncertainties for both the Scottish livestock industry and animal health surveillance after Brexit. This insight from stakeholders is important to improve uptake and implementation of animal heath surveillance activities and the future resilience of the livestock industry. The conclusions drawn from this study are applicable not only to Scotland but to other countries and international organizations involved in global animal health surveillance activities.

A method of determining where to target surveillance efforts in heterogeneous epidemiological systems

The spread of pathogens into new environments poses a considerable threat to human, animal, and plant health, and by extension, human and animal wellbeing, ecosystem function, and agricultural productivity, worldwide. Early detection through effective surveillance is a key strategy to reduce the risk of their establishment. Whilst it is well established that statistical and economic considerations are of vital importance when planning surveillance efforts, it is also important to consider epidemiological characteristics of the pathogen in question-including heterogeneities within the epidemiological system itself. One of the most pronounced realisations of this heterogeneity is seen in the case of vector-borne pathogens, which spread between 'hosts' and 'vectors'-with each group possessing distinct epidemiological characteristics. As a result, an important question when planning surveillance for emerging vector-borne pathogens is where to place sampling resources in order to detect the pathogen as early as possible. We answer this question by developing a statistical function which describes the probability distributions of the prevalences of infection at first detection in both hosts and vectors. We also show how this method can be adapted in order to maximise the probability of early detection of an emerging pathogen within imposed sample size and/or cost constraints, and demonstrate its application using two simple models of vector-borne citrus pathogens. Under the assumption of a linear cost function, we find that sampling costs are generally minimised when either hosts or vectors, but not both, are sampled.

OptisampleTM: Open web-based application to optimize sampling strategies for active surveillance activities at the herd level illustrated using Porcine Respiratory Reproductive Syndrome (PRRS)

Porcine reproductive and respiratory syndrome virus (PRRSv) infection causes a devastating economic impact to the swine industry. Active surveillance is routinely conducted in many swine herds to demonstrate freedom from PRRSv infection. The design of efficient active surveillance sampling schemes is challenging because optimum surveillance strategies may differ depending on infection status, herd structure, management, or resources for conducting sampling. Here, we present an open web-based application, named 'OptisampleTM', designed to optimize herd sampling strategies to substantiate freedom of infection considering also costs of testing. In addition to herd size, expected prevalence, test sensitivity, and desired level of confidence, the model takes into account the presumed risk of pathogen introduction between samples, the structure of the herd, and the process to select the samples over time. We illustrate the functionality and capacity of 'OptisampleTM' through its application to active surveillance of PRRSv in hypothetical swine herds under disparate epidemiological situations. Diverse sampling schemes were simulated and compared for each herd to identify effective strategies at low costs. The model results show that to demonstrate freedom from disease, it is important to consider both the epidemiological situation of the herd and the sample selected. The approach illustrated here for PRRSv may be easily extended to other animal disease surveillance systems using the web-based application available at http://stemma.ahc.umn.edu/optisample.

Epidemiological performance and subsequent costs of different surveillance strategies to control bovine herpesvirus type 1 in dairy farms

This study aimed at comparing the surveillance program of bovine herpesvirus type 1 (BHV1) as laid down by EU Decision 2004/558/EC and 2007/584/EC ('conventional design') with an alternative design. The alternative design was based on monthly bulk-milk testing, clinical surveillance and a risk-based component that involves testing of animals that are purchased from non-free cattle herds. Scenario-tree analyses were carried out to determine sensitivities of the surveillance system (and its components) and the monthly confidence of freedom on herd-level. Also, the expected costs per surveillance design and components thereof were calculated. Results showed that the conventional (EU) and alternative surveillance designs to obtain a BHV1-free status performed equally well in terms of sensitivity. However, total costs per cattle herd to obtain a free status were highest in the conventional design. In an endemic situation and with a within-herd design prevalence of 10%, the conventional design led to a varying probability of freedom ranging from 99.6% to 100% per month. With the alternative design, in this situation, a constant probability of freedom of >99.9% per month was found. In a disease-free situation, both designs performed equally well (probability of freedom >99.9% per month). The yearly costs per farm for monitoring the disease-free status decreased by approximately 25% in the alternative design. The alternative strategy based on monthly bulk-milk monitoring therefore was deemed most cost-effective. This study showed that the surveillance regime to attain and maintain a BHV1-free status as described by EU-legislation can be improved to reduce the monitoring costs without reduction of the system's sensitivity, given a within-herd design prevalence of 10%. The assessment of various surveillance designs could be highly useful to support decision-making towards a more risk-based approach of animal health surveillance.

Active animal health surveillance in European Union Member States: gaps and opportunities

Animal health surveillance enables the detection and control of animal diseases including zoonoses. Under the EU-FP7 project RISKSUR, a survey was conducted in 11 EU Member States and Switzerland to describe active surveillance components in 2011 managed by the public or private sector and identify gaps and opportunities. Information was collected about hazard, target population, geographical focus, legal obligation, management, surveillance design, risk-based sampling, and multi-hazard surveillance. Two countries were excluded due to incompleteness of data. Most of the 664 components targeted cattle (26·7%), pigs (17·5%) or poultry (16·0%). The most common surveillance objectives were demonstrating freedom from disease (43·8%) and case detection (26·8%). Over half of components applied risk-based sampling (57·1%), but mainly focused on a single population stratum (targeted risk-based) rather than differentiating between risk levels of different strata (stratified risk-based). About a third of components were multi-hazard (37·3%). Both risk-based sampling and multi-hazard surveillance were used more frequently in privately funded components. The study identified several gaps (e.g. lack of systematic documentation, inconsistent application of terminology) and opportunities (e.g. stratified risk-based sampling). The greater flexibility provided by the new EU Animal Health Law means that systematic evaluation of surveillance alternatives will be required to optimize cost-effectiveness.

Evaluation of temporal surveillance system sensitivity and freedom from bovine viral diarrhea in Danish dairy herds using scenario tree modelling

BACKGROUND

The temporal sensitivity of the surveillance system (TemSSe) for Bovine Viral Diarrhea (BVD) in Danish dairy herds was evaluated. Currently, the Danish antibody blocking ELISA is used to test quarterly bulk tank milk (BTM). To optimize the surveillance system as an early warning system, we considered the possibility of using the SVANOVIR ELISA, as this test has been shown to detect BVD-positive herds earlier than the blocking ELISA in BTM tests. Information from data (2010) and outputs from two published stochastic models were fed into a stochastic scenario tree to estimate the TemSSe. For that purpose we considered: the risk of BVD introduction into the dairy population, the ELISA used and the high risk period (HRP) from BVD introduction to testing (at 90 or 365 days). The effect of introducing one persistently infected (PI) calf or one transiently infected (TI) milking cow into 1 (or 8) dairy herd(s) was investigated. Additionally we estimated the confidence in low (PLow) herd prevalence (<8/4109 infected herds) and the confidence in complete freedom (PFree) from BVD (< 1/4109).

RESULTS

The TemSSe, the PLow, and the PFree were higher, when tests were performed 365 days after BVD introduction, than after 90 days. Estimates were usually higher for the SVANOVIR than for the blocking ELISA, and when a PI rather than a TI was introduced into the herd(s). For instance, with the current system, the median TemSSe was 64.5 %, 90 days after a PI calf was introduced into eight dairy herds. The related median PLow was 72.5 %. When a PI calf was introduced into one herd the median TemSSe was 12.1 %, while the related PFree was 51.6 %. With the SVANOVIR ELISA these estimates were 99.0 %; 98.9 %, 43.7 % and 62.4 %, respectively.

CONCLUSIONS

The replacement of the blocking ELISA with the SVANOVIR could increase the TemSSe, the PLow and PFree remarkably. Those results could be used to optimize the Danish BVD surveillance system. Furthermore, the approach proposed in this study, for including the effect of the HRP within the scenario tree methodology, could be applied to optimize early warning surveillance systems of different animal diseases.

Scenario tree model for animal disease freedom framed in the OIE context using the example of a generic swine model for Aujeszky's disease in commercial swine in Canada

"Freedom from animal disease" is an ambiguous concept that may have a different meaning in trade and science. For trade alone, there are different levels of freedom from OIE listed diseases. A country can: be recognized by OIE to be "officially free"; self-declare freedom, with no official recognition by the OIE; or report animal disease as absent (no occurrence) in six-monthly reports. In science, we apply scenario tree models to calculate the probability of a population being free from disease at a given prevalence to provide evidence of freedom from animal disease. Here, we link science with application by describing how a scenario tree model may contribute to a country's claim of freedom from animal disease. We combine the idea of a standardized presentation of scenario tree models for disease freedom and having a similar model for two different animal diseases to suggest that a simple generic model may help veterinary authorities to build and evaluate scenario tree models for disease freedom. Here, we aim to develop a generic scenario tree model for disease freedom that is: animal species specific, population specific, and has a simple structure. The specific objectives were: to explore the levels of freedom described in the OIE Terrestrial Animal Health Code; to describe how scenario tree models may contribute to a country's claim of freedom from animal disease; and to present a generic swine scenario tree model for disease freedom in Canada's domestic (commercial) swine applied to Aujeszky's disease (AD). In particular, to explore how historical survey data, and data mining may affect the probability of freedom and to explore different sampling strategies. Finally, to frame the generic scenario tree model in the context of Canada's claim of freedom from AD. We found that scenario tree models are useful to support a country's claim of freedom either as "recognized officially free" or as part of a self-declaration but the models should not stand alone in a claim. The generic AD scenario tree model demonstrated the benefit of combining three sources of surveillance data and helped to design the surveillance for the next year. The generic AD scenario model is one piece in Canada's self-declaration of freedom from AD. The model is strongly supported by the fact that AD has never been detected in Canada.

Sensitivity of Bovine Tuberculosis Surveillance in Wildlife in France: A Scenario Tree Approach

Bovine tuberculosis (bTB) is a common disease in cattle and wildlife, with an impact on animal and human health, and economic implications. Infected wild animals have been detected in some European countries, and bTB reservoirs in wildlife have been identified, potentially hindering the eradication of bTB from cattle populations. However, the surveillance of bTB in wildlife involves several practical difficulties and is not currently covered by EU legislation. We report here the first assessment of the sensitivity of the bTB surveillance system for free-ranging wildlife launched in France in 2011 (the Sylvatub system), based on scenario tree modelling. Three surveillance system components were identified: (i) passive scanning surveillance for hunted wild boar, red deer and roe deer, based on carcass examination, (ii) passive surveillance on animals found dead, moribund or with abnormal behaviour, for wild boar, red deer, roe deer and badger and (iii) active surveillance for wild boar and badger. The application of these three surveillance system components depends on the geographic risk of bTB infection in wildlife, which in turn depends on the prevalence of bTB in cattle. We estimated the effectiveness of the three components of the Sylvatub surveillance system quantitatively, for each species separately. Active surveillance and passive scanning surveillance by carcass examination were the approaches most likely to detect at least one infected animal in a population with a given design prevalence, regardless of the local risk level and species considered. The awareness of hunters, which depends on their training and the geographic risk, was found to affect surveillance sensitivity. The results obtained are relevant for hunters and veterinary authorities wishing to determine the actual efficacy of wildlife bTB surveillance as a function of geographic area and species, and could provide support for decision-making processes concerning the enhancement of surveillance strategies.

Assessment of the probability of introduction of bovine tuberculosis to Danish cattle farms via imports of live cattle from abroad and immigrant workers

Denmark has been recognized as officially free (OTF) from bovine tuberculosis (bTB) since 1980. In this study, we estimated the annual probability (PIntro) of introducing Mycobacterium bovis into the Danish cattle population, through (a) imports of cattle and (b) foreign personnel working in Danish cattle herds. Data from 2000 to 2013 with date, number and origin of imported live cattle were obtained from the Danish Cattle Federation. Information on immigrants working in Danish cattle herds was obtained through a questionnaire sent by email to a sample of Danish cattle farmers (N=460). Inputs obtained from data analysis, expert opinion, the questionnaire and literature were fed into three stochastic scenario tree models used to simulate the effect of import trade patterns, and contact between immigrant workers and cattle. We also investigated the opportunity of testing animals imported from OTF countries by tuberculin skin test and animals from non-OTF countries by interferon-γ test (IFN-γ), exemplified by using year 2009 where the number of imported animals was higher than usual. Results showed that PIntro is driven mainly by importation of live cattle. The combined median annual probability of introducing M. bovis into the Danish cattle population by either imported live cattle or infectious immigrant workers, ranged from 0.3% (90% prediction interval (P.I.): 0.04%:1.4%) in 2001 to 4.9% (90% P.I.: 0.6%; 19.2%) in 2009. The median of the median PIntro estimates from the 14 years was 0.7% (median of 90% P.I.: 0.08%; 3.5%). Hence, on average, at least one introduction each 143 years could be expected, if the annual number of imported animals does not change remarkably in the future. If the number of imported animals increases, compared to the years we analyzed, additional testing of imported cattle might be considered. For example, in 2009, PIntro would have been reduced from 4.9% to 0.8% (90% P.I.: 0.1%; 4.7%) if animals from OTF countries had been tested with the tuberculin skin test and animals from non-OTF countries had been tested with the IFN-γ test. The presented model could be used easily in other countries with similar bTB status to Denmark, where wildlife represents a negligible probability of infection for domestic cattle and where the imported live cattle represent the main pathway of bTB introduction into the local cattle population.

Comparison of output-based approaches used to substantiate bovine tuberculosis free status in Danish cattle herds

We compared two published studies based on different output-based surveillance models, which were used for evaluating the performance of two meat inspection systems in cattle and to substantiate freedom from bovine tuberculosis (bTB) in Denmark. The systems were the current meat inspection methods (CMI) vs. the visual-only inspection (VOI). In one study, the surveillance system sensitivity (SSe) was estimated to substantiate the bTB free status. The other study used SSe in the estimation of the probability of freedom (PFree), based on the epidemiological concept of negative predictive value to substantiate the bTB free status. Both studies found that changing from CMI to VOI would markedly decrease the SSe. However, the two studies reported diverging conclusions regarding the effect on the substantiation of Denmark as a bTB free country, if VOI were to be introduced. The objectives of this work were: (a) to investigate the reasons why conclusions based on the two models differed, and (b) to create a hybrid model based on elements from both studies to evaluate the impact of a change from CMI to VOI. The hybrid model was based on the PFree approach to substantiate freedom from bTB and was parametrized with inputs according to the newest available information. The PFree was updated on an annual basis for each of 42 years of test-negative surveillance data (1995-2037), while assuming a low (<1%) annual probability of introduction of bTB into Danish cattle herds. The most important reasons for the difference between the study conclusions were: the approach chosen to substantiate the bTB free status (SSe vs. PFree) and the number of years of surveillance data considered. With the hybrid model, the PFree reached a level >95% after the first year of surveillance and remained ≥96% with both the CMI and VOI systems until the end of the analyzed period. It is appropriate to use the PFree of the surveillance system to substantiate confidence in bTB free status, when test-negative surveillance results can be documented over an extended period of time, while maintaining a low probability of introduction of bTB into the cattle population. For Denmark, the probability of introduction of bTB should be kept <1% on an annual basis to sustain the high confidence in freedom over time. The results could be considered when deciding if the CMI can be replaced by VOI in cattle abattoirs of countries for which bTB freedom can be demonstrated.

A rationale to unify measurements of effectiveness for animal health surveillance

Surveillance systems produce data which, once analysed and interpreted, support decisions regarding disease management. While several performance measures for surveillance are in use, no theoretical framework has been proposed yet with a rationale for defining and estimating effectiveness measures of surveillance systems in a generic way. An effective surveillance system is a system whose data collection, analysis and interpretation processes lead to decisions that are appropriate given the true disease status of the target population. Accordingly, we developed a framework accounting for sampling, testing and data interpretation processes, to depict in a probabilistic way the direction and magnitude of the discrepancy between "decisions that would be made if the true state of a population was known" and the "decisions that are actually made upon the analysis and interpretation of surveillance data". The proposed framework provides a theoretical basis for standardised quantitative evaluation of the effectiveness of surveillance systems. We illustrate such approaches using hypothetical surveillance systems aimed at monitoring the prevalence of an endemic disease and at detecting an emerging disease as early as possible and with an empirical case study on a passive surveillance system aiming at detecting cases of Highly Pathogenic Avian Influenza cases in Vietnamese poultry.

Epidemiology and Economics Support Decisions about Freedom from Aquatic Animal Disease

In this study, we review the application of epidemiology and economics to decision-making about freedom from aquatic animal disease, at national and regional level, and recent examples from Europe. Epidemiological data (e.g. pathogen prevalence and distribution) determine the technical feasibility and cost of eradication. The eradication of pathogens which exist in wild populations, or in a latent state, is technically difficult, uncertain and expensive. Notably, the eradication of diseases of molluscs is rarely attempted because host populations (farmed and wild) cannot be completely removed from open water systems. Doubt about the success of eradication translates into uncertain ex-ante cost estimates. Additionally, the benefits of an official disease-free status cannot be estimated with any accuracy. For example, in Europe, official freedom from epizootic ulcerative syndrome and white spot syndrome virus has not been pursued, arguably because the evidence does not exist for the benefits (reduced risk of disease in wild populations) to be estimated and thus weighed against the costs of maintaining disease freedom (e.g. restriction on imports). Economic analysis must assess not only whether the benefits of disease freedom outweigh costs, but whether it is the economically optimal disease control option. Government may also want to compare investment in aquatic animal health with other opportunities. As resources become scarce, governments have sought to share costs of disease control with industry, and thus to ensure equity, the distribution benefits must be known so costs can be borne by those who benefit. The economic principles to support decisions about disease freedom are well established, but their application is constrained by lack of epidemiological data, which may explain the lack of economic analysis in support of aquatic animal management in Europe. The integration of epidemiology and economics in disease control planning will identify research aimed at improving the underpinning evidence base.