Infectious diseases among animals : combining models with data

Abstract

To eradicate or control the spread of infectious diseases, knowledge on the spread of the infection between (groups of) animals is necessary. Models can include such information and can subsequently be used to observe the efficacy of various control measures in fighting the infection. However, the availability of information and data to build and quantify these models is essential for applying such models in real life. In this thesis, models on the spread of infectious diseases in animals are always combined with data concerning the host, the infectious agent, their interactions and often also case data from epidemic or endemic disease situations. To do so, various infections that are present in the Netherlands are discussed. We show that for Phocine Distemper Virus (PDV) the data of the epidemic in 1988 show that clustering of animals has a strong influence on the transmission and survival of the animals. A previously applied model to analyse the seal situation did not fit the data very well. A model that incorporates the clustering of the animals on sand banks gave a better fit. Due to their clustering, the death rate was higher than could be expected from the first model. To control and eradicate Infectious Bovine Rhinotracheitis (IBR) it is known that the persistence of the infection in previously infected animals may cause a delay in eradication. We have quantified the probability for such a virus to reactivate in the field, and combined that with a model that calculates the expected time to extinction. Thus, control measures in the eradication process can easily be compared for efficacy and a time frame can be defined. The control of BSE was an important issue in the last decade, but due to limited data, exact advice was difficult to find. By quantifying the transmission parameters, the various control measures can be compared for efficacy. Now that more information is available concerning risks for humans and cattle, optimisation of the surveillance and control can be introduced based on such models. The age distribution of BSE cases offers information on the efficacy of BSE control in the past, but also concerning the prevalence of the infection in the future. Risk assessment and modelling are especially important for countries that would like to prove the absence of BSE in their country. Monitoring of an animal disease situation may have many purposes, for instance to prove freedom from infection for a certain farm or country. In the last chapter of the thesis, we give an opening for extended modelling to quantify the risk of missing a new outbreak in a country that has the Free from disease status. An integrated approach of transmission models that specifically includes the frequency of sampling over time enables us to calculate the probability that an epidemic escapes from detection