Use this URL to cite or link to this record in EThOS:
Title: The design and analysis of post-epidemic foot-and-mouth disease surveillance programmes
Author: Handel, Ian G.
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Post-epidemic surveillance is a necessary component in the process by which areas having experienced epidemic or endemic animal disease gain a 'disease-free' status. Whilst it is useful to rapidly and surely regain this status, surveillance uses expensive resources and imposes a time delay on regaining disease-freedom. It is beneficial to design and analyse surveillance to maximise efficiency and use data effectively. This thesis addresses issues of surveillance design and analysis in the context of demonstration of disease freedom after foot-and-mouth disease (FMD) epidemics or outbreaks. International regulations require the process of demonstrating disease freedom to include a serological survey of livestock after the epidemic has apparently ended. Currently, animal holdings are sampled from randomly to achieve a defined probability of detecting disease, if present in the region. Using a risk model applied to demographic and epidemic data from the Devon UK 2001 FMD outbreak, I estimate the efficiency gains of risk based sampling compared to random sampling. With this technique farms at a high risk of harbouring undiscovered infection are selectively targeted for sampling. This approach is robust to model errors and reduces the number of farms to be sampled from 1083 to 225 to achieve a 95% level of confidence that the area is disease free. Additionally, using the risk model to order the sampling of farms will reduce delays to declaration of disease-freedom by approximately 11 days on average. The Thrace region of Turkey is a buffer zone between the rest of Europe, which is generally FMD free, and Anatolian Turkey, which has endemic FMD. A EU/FAO funded project vaccinates and serologically tests Thrace's livestock. This provides a data set for the exploration of survey analysis techniques. The testing is for the purpose of demonstrating disease freedom so Turkey may join the European Union. Using a stochastic model simulating different disease scenarios and diagnostic test performances I evaluate the current sampling strategy. It generally satisfies the sensitivity requirement to detect disease in the area. However the initial specificity is low, resulting in a high rate of false positive classifications of villages, requiring the use of confirmatory tests. I describe the results from three surveys in 2005-2006 and adjust for an apparent shift in test result datum over the period. The results are then used to parameterise a multilevel, mixture model of FMD in Thrace. Estimates from this model suggest a high village-level prevalence of disease with high variability of within-village prevalence. The temporal changes are compatible with the reported FMD outbreaks though the absolute estimates are much higher than expected from outbreak data. I suggest that there may be relatively few truly exposed villages with the large number of apparently exposed villages the consequence of imperfect diagnostic test specificity in the vaccinated population. Post-epidemic surveillance relies on serological and clinical surveillance. I develop a simulation model based on a within-group stochastic model and models of stockholder observation and diagnostic test response to estimate the time varying results of clinical surveillance and the additional benefits of serological surveillance. This quantifies current belief that serological surveillance should be focused on more extensive production systems. The additional benefits of serological surveillance are greatest in poorly observed, sparse outbreaks allowing, in the simulated scenarios, disease freedom to be declared in animal groups 15-33 days sooner than by clinical observation alone.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available