| The global spread, economic impact and public health implications of H5N1 highly pathogenic avian influenza virus (HPAIV) have attracted the attention of the international health and scientific communities to the importance of avian influenza virus (AIV) research and surveillance. Guidelines for AIV surveillance, intended to improve early detection of the disease and enable the efficient use of control and prevention resources have been established worldwide. Successful and efficient early detection of the disease in a population largely depends on the quality of collected data, as well as methods of collection, data sharing, and communication. The overall goal of the dissertation is to explore, develop, and evaluate sets of analytical techniques to improve AIV surveillance in Europe, and to advance our understanding of AIV epidemiology and to incorporate that information into the decision-making process in the event of an epidemic. The findings of this thesis addressed important aspects of AIV epidemiology and surveillance in Europe, including: 1) a critical review of current global animal disease surveillance, common analytical methods applied to surveillance data, and the nature and limitations of surveillance data, 2) an exploration of alternative scan-based approaches to estimate space-time distribution of AIV cases and to quantify the associated risk factors in a regional setting, using 2006 H5N1 HPAIV surveillance data from Sweden and Denmark as a case study, 3) presentation of a case study---based on 2005-2007 H5N1 HPAIV sequence data from Romania---to model the association between selected epidemiological factors and sequence variation in the HA, NA, and NS genes of AIV, 4) an exploration of the use of spatial ecological niche modeling to evaluate AIV surveillance in Europe, using Low Pathogenic AIV (LPAIV) surveillance data in Europe as a case study, 5) an exploration of Bayesian phylodynamic analytical pipeline for molecular-based surveillance of AIV using 2005-2010 H5N1 HPAIV sequence data from Europe as a case study. Results of this work provide novel insights on the epidemiology of current and future emerging AIVs, advance our understanding of how surveillance data can be used to make optimal policy decisions during disease control and prevention campaigns, and provide a basis for guiding the design of contingency plans for future AI epidemics in Europe. |
