Modelling in veterinary epidemiology: Why model building is important

Some consider modelling to be very important for (veterinary) epidemiology, others severely criticise the use of modelling. Before joining this heated debate it is worthwhile to reflect on the role of mathematical modelling. Mathematical modelling is useful for the study of complex phenomena, like the population dynamics of infectious agents, because models show how separate measurements can be seen as manifestation of the same underlying processes. To build models that can act as connecting theories, careful model building is very important. It is shown how modelling helped to understand how transmission depends on underlying factors. Through a process of careful model building and comparisons of different model assumptions and model predictions with data one hypothesis was falsified and therewith the plausibility of another strengthened. In conclusion, the gain of the modelling was not the resulting model, but instead the insight into the population dynamics of infectious agents that was obtained in the process of model building and model analysis on the one hand, and interpreting experimental and observational data on the other.