Workshops

Inge Santman-Berends, Royal GD
Anouk Veldhuis, Royal GD
Mirjam Nielen, Utrecht University
Wouter Swolfs, White Rabbit strategy
Contact: ingesantman@gmail.com

Workshop objectives:
Researchers want to do research that has impact. However, they often do not consider from the start of their project how they can actually maximize their impact. Within the DECIDE project we have learned that thinking about business models for your research already early in your project can help to get focus and improve the connection with the end users of your research. In this workshop we will inform the participants what we have learned and how they can use this information to improve their own research. This workshop will provide a business model that helps researchers to maximize the impact of their research.

In the workshop we will create awareness about the importance of considering aspects such as:

• Who will use your results?
• How does your research bring value to your user?
• Through which channels can you reach the user?
• Do you need funds to create impact?

By considering these aspects, preferably at the start of a research project, the research will be more focused on meeting needs of stakeholders, maximizing the chance of implementation of results in practice. In addition, during the lifetime of a project insights may change and good business models adapt to changing environments.

The expected outcomes include an enhanced understanding of the importance of considering value propositions and business models in research. This will result in better focus, an increased likelihood of finding financers for your research and greater chances that your research will have real impact by being implemented in practice.

Assumed knowledge of participants:
No prior knowledge required.

Esben Østergaard Eriksen & Ingrid Toftaker
Norwegian University of Life Sciences, Department of Production Animal Clinical Sciences.
Contact: Esben.ostergaard.eriksen@nmbu.no

As veterinary epidemiologists, we often aim to establish how an exposure affects an outcome. Even when veterinary epidemiologists don’t state causal aims explicitly, they often use causal language when interpreting and presenting their results, revealing that they are actually interested in causal inference¹. Unfortunately, approaches supporting causal inference are not yet fully adopted in our community; for instance, it is still common to use statistical criteria to select variables for multivariable models². In contrast, in the field of human epidemiology, the counterfactual causal inference framework has gained acceptance in recent years². It is time for the vet-epi community to embrace causal inference as well². In brief, we must learn to explicitly state one or a few quantities that represent the casual effect of interest; consider whether key assumptions of the counterfactual framework are met; and use subject-matter knowledge to specify assumptions about the studied causal system in a directed acyclic graph (DAG) and let this guide the statistical model building.

Workshop objectives:
This workshop will help the participants to begin applying a structured methodological framework for causal inference. The workshop gives a rapid but pedagogical introduction to the fundamental theory and participants should gain an overview of the field, thus empowering them to apply the principles in their own work.

After the workshop, the participants should;
1. Be able to define what a cause is.
2. Understand the types of aims we have in veterinary epidemiology, and the implication for the analysis.
3. Be able to state a causal aim or research question.
4. Be familiar with the fundamental assumptions that must be met before we claim that a statistical association between a factor and a health outcome represents a causal effect.
5. Write up a causal diagram in a DAG.
6. Determine the minimal set of adjustment variables based on a DAG.
7. Know that there are several techniques to achieve exchangeability (confounder adjustment) in non-randomized studies including inverse probability of treatment weighting (IPTW), adjustment by regression, matching, stratification etc.
8. Have a basic understand of IPTW.

Assumed knowledge of participants:
Participants are assumed to have basic understanding of epidemiology including the main features of common observational study designs (cohort, case-control, cross-sectional) so most participants at the SVEPM should be able to follow this workshop.

Pranav Pandit & Pranav Kulkarni
UC Davies
Contact: pspandit@ucdavis.edu

Workshop objectives:
By the end of this workshop, participants will be able to:

• Understand the role of machine learning and statistical inference in veterinary epidemiology.
• Apply hands-on methods for tokenizing, preprocessing, and analyzing epidemiological datasets.
• Build and interpret predictive models for disease risk and surveillance (including anomaly detection and inference).
• Gain practical experience with open-source tools (Python, Jupyter, selected libraries) for epidemiological applications.
• Critically evaluate model outputs for decision support in veterinary public health.

Assumed knowledge of participants:
Basic epidemiology and statistics. Some familiarity with data analysis (R, Python, or equivalent). No advanced programming experience required (code templates provided).

Lis Alban, Chief Scientist, Danish Agriculture & Food Council
Marnie Brennan, Director of the Centre for Evidence-based Veterinary Medicine, University of Nottingham
Koen Mintiens, International Animal Health and Welfare Consultant, Ven40 Consulting
Contact: koen.mintiens@ven40.com

Workshop objectives:

• To provide insight from policy, industry, and academia as to the impact veterinary epidemiology is or isn’t making in society.
• To identify opportunities and threats to the value of veterinary epidemiology in modern times for animal and human health and welfare.

Expected outcomes:

• Potential strategic options to explore and incorporate into future veterinary epidemiology research and collaborations
• Guiding actions that can help veterinary epidemiology to stay ahead of current technical, societal, and policy-related trends

Assumed knowledge of participants:
Experience in veterinary epidemiology research. Experience with presenting research results and providing expert advice to stakeholders may be an asset. Curiosity in these subjects would be a requirement in case the experience is low.

Pablo Alarcon & Barbara Haesler
Royal Veterinary College, London, UK
Contact: palarcon@rvc.ac.uk

Workshop objectives:
By the end of the workshop, participants will be able to:

• Understand what is meant by Economics of One Health
• Conceptualize the One Health value proposition spanning the health of people, animals, and the environment
• Describe methods used in One Health Economics to conduct economic evaluations
• Discuss trade-offs that arise when taking a systems view to issues affecting the health of people, animals, and the environment.

Expected outcomes:

• Participants will get a broad understanding of how economics can support One Health projects.
• Have basic knowledge to start thinking about economic aspects of One Health in their own work.
• Be able to read and understand simple economic studies related to One Health.
• Know where to look for more resources and opportunities to learn about One Health economics.

Assumed knowledge of participants:
The workshop is directed to people who are interested in discovering economics and its application in One Health.

Gaël Beaunée
Oniris, INRAE, BIOEPAR, 44300, Nantes, France
Contact: gael.beaunee@inrae.fr

Workshop objectives:
This half-day workshop aims to strengthen participants’ ability to implement mathematical models that are clear, testable, and reproducible, and to document them to an agreed standard. By the end of the session, participants will understand how to structure model code with explicit configuration and separation of concerns; how to apply lightweight verification and validation through unit and property tests; and how to capture the full computational environment and the other elements needed for reproducibility (including version control, dependency management, seeds, run logs, etc.). They will also learn to profile before optimizing, selecting simple, evidence-based improvements such as vectorization or preallocation where appropriate.

Participants are expected to leave with tangible outputs they can reuse immediately: a refactored version of a small, intentionally flawed SIR model example; a minimal but complete run workflow (ready-to-run script, configuration file, and reproducible environment); and a filled documentation skeleton aligned with recognized reporting frameworks. In addition to the hands-on outputs, the workshop strengthens participants’ ability to select suitable documentation standards and how to communicate modelling assumptions and limitations transparently.

Assumed knowledge of participants:
Introductory familiarity with epidemic modelling concepts (e.g., SIR/SEIR, compartments, parameters, and time steps). Minimal experience with git (cloning, committing, pushing) is helpful but not required, brief reminders are included.