Case studies and generalisation

In WP3, the objective is to validate and optimise STOC-FREE DATA and STOC-FREE MODEL that are developed in WP1 and WP2 in case studies that are conducted by all partners.

Case example: BVD

Several European countries have implemented national or regional surveillance, control, or eradication programmes for endemic infections of cattle which are currently not regulated at EU level. Programmes aiming to control and eradicate these infections may bring tangible benefits to farmers, processors, the wider public and national economies. Nevertheless, such programmes also create difficulties for trade, particularly of live animals, between member states where differences exist in either the presence or absence of such programmes. Additionally,  the statuses attained by the programmes (control/eradication ongoing or freedom achieved) create difficulties for trade due to the risks of impairing progress with control/eradication or re-introducing infection where freedom has been achieved.

Bovine viral diarrhoea (BVD) caused by BVD virus, is a good example of such a disease and has therefore been chosen as an example disease for the development of the STOC free framework in this project.

BVDV is recognised to be an economically important pathogen of cattle and to be endemic in countries without systematic control and eradication efforts. The infection is primarily maintained and spread by persistently infected (PI) calves, which are generated as a result of in utero infection following exposure of their dams. These PI animals shed high levels of virus throughout their lifetime, and their movement between herds is recognised to be the principal means by which infection can be transferred. Trade of PI animals between member states also poses particular risks to ongoing, or completed eradication programmes.

A number of member states have initiated or completed national or regional eradication programmes for BVDV. While these have a common goal, they follow a variety of pathways to achieve this. For example, some focus primarily on testing and assigning the subsequent BVDV free status at herd level, while individual animals are the unit of interest in others. Some BVDV programmes have already achieved freedom, other programmes are still underway but have made good progress, and a third group of BVDV programmes are in their early stages. There are control programmes that are implemented on a voluntary basis and others that are compulsory and supported by legislation. The risks associated with movement of cattle between areas with different programmes and statuses is therefore an ongoing concern for both farmers and programme managers and for this reason BVD has been chosen as the model disease for this project.

The countries in our consortium reflect the variation in both BVDV programme design and prevalence, with a variety of programmes in place and disease statuses ranging from endemic to free, which have generated considerable amounts of data that are available to develop a framework to determine the confidence of freedom of infection at different levels of aggregation for BVD.

When completed, this framework will also be applicable to other infectious cattle diseases.

WP3: Highlights of year 1

The work foreseen in WP3 will be conducted from 2019 on. During the annual meeting in year 2, the case studies that will be carried out by each of the partner countries will be discussed.

WP3: Highlights of year 2

The case studies are foreseen for the third year. The developed method eventually has to be generic enough to be able to transfer to other diseases as well. The model should include generic epidemiological features which are common for multiple diseases. Aspects to take into account for generalizability that might differ between diseases include the preferred time steps, the definitions of freedom and infection, the usability of serology to determine the infectious state, and the relevant target population that should be included in the model. The work that is conducted within the COST Action SOUND control, which is closely linked to STOC free, will provide additional input on how to generalise STOC free model to other species and other diseases. Additionally, within SOUND control aspects such as economics and sociology will also be taken into consideration.

WP3: Highlights of year 3

In year 3 the case studies were started and the results on the exercise in which the equivalence between the BVDV CP in the partner countries were reported to EFSA. In the participating six countries, roughly three different situations with regard to BVDV control can be distinguished: 1. Endemic situation on country level, CP on herd level (Netherlands, France, Scotland), 2. Endemic situation on herd level, CP on animal level (Germany and Ireland) and 3. BVDV free situation (Sweden). When a country is completely disease free, the model did not converge given that it needs some infections to estimate the latent class (the true BVDV infection status). In this situation, a scenario tree model is better suited, which is tested for Sweden. For the countries that have a CP on animal level, an extra animal level module was developed that enables inclusion of raw animal level input data in the model. During the case studies several meetings between the members working on the case studies and the team that developed the model were held to discuss the challenges an possible solutions. Based on the feedback of the people working on the case studies, the model was updated several times to be able to incorporate the needs of the different countries. In year four it is foreseen to finish the case studies and to organise a workshop to gather input for the generalisability of the STOC free framework.

The work that is conducted within the COST Action SOUND control, which is closely linked to STOC free, will provide additional input on how to generalise STOC free model to other diseases. Additionally, within SOUND control aspects such as economics and sociology will also be taken into consideration.

WP3: Highlights of year 4

In year 4, the first round of case studies was finished. All countries were able to run the model, except Sweden where BVD is absent. For this reason, Sweden developed a scenario tree model to estimate the confidence of freedom from BVDV. The case studies resulted in several recommendations for further development of the STOC free model. First, the model and input formats needed further refinement to avoid disproportionate influence of priors on model outputs. Also, some countries had problems with convergence of the models and some results did not seem to reflect the true situation. The results of the initially conducted case studies were reported to EFSA. Different solutions have been evaluated to tackle the problems that were encountered. Also a guiding document was written that explained in more detail how the priors should be calculated and how the model had to be run. After some adjustments to the model new case studies were performed with data of CPs based on BVDV testing in ear notch samples of newborn calves.  The results of the second round of case studies seem very promising and are currently written down in a scientific paper.

To investigate the generalisability and needs for further development of the STOC free model, a workshop with invited modellers was held. The workshop included an introduction to the project and guided use of the model. Participants were asked for improvements in the data input format and suggested adding functions such as simultaneous modelling of sampling at different levels and inclusion of follow-up testing. Conclusions from the workshop, combined with results from work in SOUND control, suggest that there is a strong potential for using the model as a tool to assess the output from control programmes targeting other endemic diseases and livestock species.

Deliverables can be found here